Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Neuronal hyperactivity and hyperexcitability of the cerebral cortex and hippocampal region is an increasingly observed phenomenon in preclinical Alzheimer’s disease ( AD ) . In later stages , oscillatory slowing and loss of functional connectivity are ubiquitous . Recent evidence suggests that neuronal dynamics have a prominent role in AD pathophysiology , making it a potentially interesting therapeutic target . However , although neuronal activity can be manipulated by various ( non- ) pharmacological means , intervening in a highly integrated system that depends on complex dynamics can produce counterintuitive and adverse effects . Computational dynamic network modeling may serve as a virtual test ground for developing effective interventions . To explore this approach , a previously introduced large-scale neural mass network with human brain topology was used to simulate the temporal evolution of AD-like , activity-dependent network degeneration . In addition , six defense strategies that either enhanced or diminished neuronal excitability were tested against the degeneration process , targeting excitatory and inhibitory neurons combined or separately . Outcome measures described oscillatory , connectivity and topological features of the damaged networks . Over time , the various interventions produced diverse large-scale network effects . Contrary to our hypothesis , the most successful strategy was a selective stimulation of all excitatory neurons in the network; it substantially prolonged the preservation of network integrity . The results of this study imply that functional network damage due to pathological neuronal activity can be opposed by targeted adjustment of neuronal excitability levels . The present approach may help to explore therapeutic effects aimed at preserving or restoring neuronal network integrity and contribute to better-informed intervention choices in future clinical trials in AD .
Electrophysiologically , Alzheimer’s disease ( AD ) is commonly characterized by ‘negative’ findings: a gradual , diffuse slowing of brain activity ( notably the posterior dominant rhythm ) , decreases in functional connectivity , and a loss of network structure and complexity [1–3] . In recent years however , various studies have reported observations that challenge this notion: in the preclinical AD , Mild Cognitive Impairment ( MCI ) and early AD stages , neuronal hyperactivity and increased functional connectivity has been observed at various scales [4–11] . While these increases were first interpreted as a compensation mechanism for synaptic dysfunction in AD , the evidence now clearly points in a different direction; neuronal hyperactivity as a part of the pathophysiological cascade in AD . For instance , it has been shown that damage induced by the typical amyloid deposits in AD leads to neuronal hyperexcitability and disinhibition [12 , 13] . In turn , neuronal hyperactivity itself has also been demonstrated to drive amyloid deposition rates [14 , 15] , a very intriguing finding that has led some to hypothesize that neuronal dynamics may play a causal role in AD pathophysiology , possibly as part of a positive feedback loop involving the neurotoxic amyloid deposits [16–18] . Additional , indirect support for this view is provided by the very early phase in which neuronal dynamics and connectivity are disrupted in AD , the independent associations of AD risk factors ( age and ApoE4 status ) with neuronal activity levels , and the increased incidence of epilepsy and epileptiform neuronal activity in this population [6 , 9 , 19–27] . Regardless of the exact role of disrupted neuronal dynamics in the pathological cascade of AD , the observations described above may have therapeutic relevance , since neuronal or synaptic function can be targeted effectively by various ( non- ) pharmacological means . In fact , the current medicinal ( symptomatic ) treatment for AD aims to improve neuronal communication by enhancing cholinergic or glutamatergic neurotransmission [28–30] . Unfortunately , the large majority of the current clinical trials are aimed at decreasing amyloid load , while investigations on improving neuronal function are relatively limited . Nevertheless , a few recent studies have confirmed that targeting neuronal or synaptic behavior may be beneficial in AD . For example , experimental studies in rats and in humans with mild cognitive impairment ( MCI ) indicated that the anti-epileptic drug levetiracetam diminishes hippocampal hyperactivity while improving cognitive performance [31] . Also , enhancement of synapse formation and function by medical nutrition was reported to have a positive effect on memory function in AD patients [32 , 33] . Non-pharmacological therapeutic studies with the specific aim to modify neuronal activity ( e . g . deep brain stimulation or transcranial magnetic stimulation ) have not yet produced significant clinical benefits in AD , but are under development [34–36] . With these possibilities in mind , a key question is how to predict and optimize the effect of therapy in the complex , highly interconnected and highly dynamic human brain . While therapeutic strategies usually target specific brain areas or structures , it is naïve to expect that local interference will not influence surrounding or connected parts of the brain . As our awareness of the brain as a complex , distributed system is growing , we should appreciate the consequences: even subtle changes can have wide-ranging and paradoxical effects [37 , 38] . Variations in efficacy and unforeseen or adverse effects are frequently encountered in neurological therapy , and perhaps could be partly explained by our lack of insight in this regard . How can we get a grip on this complex behavior , in order to guide our hypotheses and experiments more reliably ? In recent years , studies of the complexity of the brain have entered a new era , due to rapid advances in data acquisition technology and the powerful application of theoretical concepts of complex network analysis [39 , 40] . In clinically oriented studies , this method has mainly been employed to interpret patient data in order to gain a better understanding of disease mechanisms or to find diagnostic and prognostic markers of disease [41 , 42] . However , top-down patient-driven research is not its sole application: the combination of network analysis and computational modeling can offer an interesting complementary bottom-up approach [43–45] . For example , several studies have combined neural mass modeling with network analysis to study the effect of lesions on the brain [38 , 46] . With regard to AD , a previous study from our group indicated that neuronal hyperactivity may play a substantial role in the disease mechanism by demonstrating that an activity-dependent degeneration regime yielded remarkably similar results to studies in AD and mild cognitive impairment ( MCI ) patients , including selective hub vulnerability and initial neuronal hyperactivity followed by slowing , disconnection , and loss of network topology [18 , 47] . Despite the theoretical advances , complex network analysis of the brain has not yet led to any improvement for AD patients . Although abstract by nature , in our opinion network modeling might facilitate a transition towards clinical applications: in addition to simulating degenerative or lesion effects , models may also be employed to explore therapeutic intervention strategies , in order to make dynamic network changes more predictable and easier to understand in terms of their underlying mechanisms . Therefore , we used our computational AD degeneration model for the present study , but now with the addition of various ‘therapeutic’ network interventions . In the model , neuronal excitability levels of both excitatory and inhibitory neurons can be adjusted selectively ( roughly simulating medication effects or nonpharmacological stimulation/inhibition techniques ) . Since neuronal hyperactivity was the main driver of network degeneration in this model , we hypothesized that the strategy employing global neuronal inhibition would be most effective in countering this process . Fig 1 provides an overview of the overall workflow .
Relative power in the lower alpha band ( 8–10 Hz ) inevitably decreased during the degeneration process , but the rate at which this happens differed substantially between the various strategies ( see Fig 3A ) . Without intervention ( black line ) , lower alpha power breaks down around T = 20 , decreasing to a level between 0 . 1 and 0 . 2 . Global inhibition , inhibition of inhibitory neurons , and especially stimulation of excitatory neurons was able to prolong a normal level of relative power in the lower alpha band ( around 0 . 8 ) and postpone the subsequent collapse . The gradual decrease of the ( posterior ) alpha peak value is a robust finding in AD , and levels lower than 8 Hz are generally regarded as pathologic [48] . Without intervention , the alpha peak is stable for a short while , but starts to decrease steadily around T = 25 . The same pattern was observed during all intervention strategies , although the peak level stayed above 8 Hz for a substantially longer period . This seems to be mainly due to a higher initial peak value ( Fig 3B ) . Overall functional connectivity in the network showed a critical transition period around T = 20 with a breakdown even more abrupt than in the lower alpha power results ( see Fig 4 ) . In the ‘no intervention’ condition , PLI values were much higher than in the control network at the start of therapy due to the ADD process , but it rapidly fell to low levels . The ‘stimulation of excitatory neurons’ regime was able to maintain normal/high PLI values most successfully , but suddenly dropped down to the same level as the other scenarios around T = 30 . AD network degradation as assessed by EEG/MEG has previously been characterized by a decrease of the ( normalized ) clustering coefficient [49 , 50] , contributing to the loss of presumably efficient small-world network topology . Here , a pattern similar to the PLI results developed , and with the global inhibition , inhibition of inhibitory neurons and stimulation of excitatory neurons as most beneficial regimes ( Fig 5A ) . Macroscopic brain connectivity networks have been shown to possess a ( hierarchical ) modular structure , a feature that is presumed to promote network efficiency and robustness . In AD , modularity gradually weakens [51] . ‘Stimulation of excitatory neurons’ seemed to retain normal modular structure relatively well ( Fig 5B ) . Algebraic Connectivity is a graph spectral measure of overall network robustness , higher values signifying a network that is harder to tear apart [52 , 53] . Strategies that were relatively successful in upholding the robustness of the network were ‘global inhibition’ , ‘inhibition of inhibitory neurons’ and ‘stimulation of excitatory neurons’ ( Fig 5C ) . It has been reported previously that network hub structure is lost in AD [54] . In this simulation similar findings became apparent , with a later decrease of the MST Leaf number in the most successful scenarios , ‘inhibition of excitatory neurons’ , ‘global inhibition’ and ‘stimulation of excitatory neurons’ . Structural connectivity decreases inevitably over time in the model , but this process can be delayed by interventions . Fig 6 compares the ‘no intervention’ condition ( blue lines ) with the relatively successful stimulation of excitatory neurons ( red lines ) . For different time points normalized node strength is plotted against original structural degree . This measure is the ratio of present structural degree over its original degree , so a value smaller than 1 indicates a loss of structural connectivity . The loss was not equal for all nodes; hubs tend to decrease more , confirming the previously reported hub vulnerability ( 18 , 49 , 55 ) . The result of the intervention was that initially damage is counteracted better ( see T = 20 ) , but ultimately the network apparently suffered more than in the control state ( see T = 40 ) . In the intervention condition hub vulnerability is less pronounced; the slope of the lines is less steep . Remarkably , at T = 30 , overall structural connectivity was fairly equal between the conditions , while functional measures differed considerably ( see results above ) . To assess the effect of the specific delay with which the intervention is started , we compared functional connectivity results ( PLI ) for three different intervention starting points ( T = 0 , T = 10 and T = 20 ) , using the relatively successful ‘stimulation of excitatory neurons’ strategy . As can be judged from Fig 7 , the response of the PLI to the differently timed interventions was similar , and their ability to uphold PLI values near or above the control network levels seems mainly dependent on the moment in the degeneration process at which the intervention was initiated . Overall performance of the four most successful strategies was assessed by combining results into a total sum per category for the four most successful interventions ( Fig 8 , see caption or method section for score definition ) . This enables a comparison at a glance between the various strategies regarding different qualities , and show that all these strategies bring an overall improvement compared to the ‘no intervention’ state , with individual differences . The most successful strategy was the ‘stimulation of excitatory neurons’ intervention , but also inhibitory strategies performed fairly well .
The observation that the ‘stimulation of excitatory neurons’ strategy is most successful in retaining network organization in a situation where neuronal activity leads to damage seems contradictory at least , and is in stark contrast with our initial hypothesis . Intuitively , one would be inclined to choose a strategy to slow down the degeneration process by either inhibiting excitation or by stimulating inhibition , and , judging from our results , the inhibitory strategies perform quite well . Several pharmacological studies aimed at countering aberrant neuronal hyperactivity provide circumstantial support for this view [31 , 55] . However , as can be judged from the present experiments , the effects of the different types of stimulation and inhibition are quite unpredictable: in this particular topology , the increased excitability in excitatory neurons apparently leads to a condition where desirable network topology is retained longer . Although this result begs for an mechanistic explanation , this was not the primary goal of our current experiment . However , we did perform further analyses to better understand the outcome ( see also S3 Text ) : we hypothesized that one reason for a beneficial effect might be that the net effect of stimulation of excitatory neurons on the entire neural mass is inhibitory , due to the interplay with the relatively influential inhibitory neuron groups . However , analysis showed that this is not the case . Since all neural masses are equal except for their connectivity pattern , the explanation may lie in the network topology itself . For example , the successful scenarios seem to suppress functional hub strength of the network ( see Fig 5D ) , and while hubs may generally improve network efficiency , they can also facilitate the spread of pathology[56] . Different topologies should be tested with the same degeneration algorithm in future experiments to explore this idea . An alternative explanation , is that the philosophy behind a strategy should perhaps not be based on trying to oppose the degeneration mechanism , but on keeping the network ‘alive’ for as long as possible in general by stimulating the remaining neurons in a decaying network; compensatory activation . However , inhibitory strategies performed quite well in our experiments , and current literature clearly points to the pathological aspect of the hyperactivity in AD [57–59] . With the present degeneration algorithm , intervention success is relative , since regardless of the chosen strategy , network breakdown was inevitable . This may be explained by specific features of the algorithm , as described in the study limitations paragraph . Another issue is our selection of strategies . As a first exploration we opted for global , continuous interventions , but many more scenarios are possible , and might lead to different outcomes: selective stimulation or inhibition of specific regions ( for example hub regions ) , different delay or timing of interventions , interventions based on individual connectivity patterns , adaptive interventions , structural interventions , and so on . A present-day major objective in the field of neurodegenerative disease research is the aim for early intervention: there is a growing consensus that , to be successful , therapy should be commenced before structural pathology is widespread [60] . We examined the influence of three different intervention starting points ( see Fig 7 ) on functional connectivity ( PLI ) , an outcome measure that is the basis of all graph theoretical analysis , and a measure that produced clear differences among the various strategies . Here however , no clear benefit of early intervention initiation is found; in fact , the response of the network to the ‘stimulation of excitatory neurons’ intervention seems independent of the moment in de degeneration process . This is surprising , since underlying structural connectivity and topology is inevitably weaker at later time points , raising the question whether some kind of adaptive therapy could be effective , i . e . starting stimulation at low PLI levels , and stopping at high levels . Another intriguing finding was the sharp peak in activity and connectivity just prior to sudden decrease in value in several of the scenarios , including the most successful ones , reminiscent of a non-linear system near a critical transition ( close to breakdown in this case ) . Although we did not elaborate on this , this finding may also point to the potential merit of an adaptive strategy that can avoid the increases and thereby prevent network collapse . Also , the success of a therapy depends on how ‘success’ is defined . In this exploratory study , we adhered to the general idea of keeping a system in or close to its original functional state over time , which seems straightforward and in line with goals in medical practice in general . Of course , in our model , the virtual time base does not permit statements about actual timelines or tempo of deterioration , but when a strategy is able to uphold certain desired characteristics approximately twice as long ( see for example Fig 4 ) as in the ‘no-intervention’ state , this can reasonably be thought of as successful; real-life therapy that delays degenerative damage effects in AD this long would be considered a spectacular improvement . Although a translation of our present findings to clinical experiments and pharmacological recommendations is too ambitious at this point , an appealing thought is that the current medication for AD leads to stimulation of excitatory neurons , the most successful strategy in our experiments as well . Ultimately , intervention modeling studies will require validation through real interventions targeting brain activity , to evaluate their true value . Various arbitrary choices that were made in this exploratory study may have had an influence on its outcome . For example , one might wonder whether the model is sufficiently detailed to realistically simulate cerebral dynamics . Neural masses were identical ( except for their connectivity characteristics ) , while there are distinct regional differences in the brain . Also , the number of masses ( n = 78 ) could be augmented based on more recent and detailed structural connectome datasets . Our present DTI-based structural connectivity data in the model is based on deterministic tractography , which has been demonstrated to be inadequate for detecting crossing white matter fibers , and newer , more powerful methods are available [61 , 62] . Repeating the current approach using alternative connectome data is an important way to validate model findings , and should be performed in future studies . Another interesting recent approach is to replace the structural network by one that is based on a generative model itself [63–65] . Also , one could argue that connectome data taken from an older control group would have led to different , more representative results . Furthermore , the neural masses themselves model a neuronal region macroscopically , but not on a microscopic level , while detailed neurophysiological models at ( sub ) cellular level are certainly available . Could more detail have led to a different outcome ? While that certainly is possible , we have several motives for our current approach . First , the global macroscopic level of analysis is of particular interest , since clinical measurements ( e . g . EEG , ( f ) MRI ) are at this level; detailed in vivo measurements on a microscopical level in memory clinic patients are not ( yet ) feasible . Second , the observation that , even in this relatively basic model of human connectivity , the outcomes of fairly straightforward interventions defy an easy mechanistic explanation underscore the complexity of network effects , and the need for modeling . Third , data analysis time and computing power are finite , and added detail can increase demands placed on these aspects exponentially . Other limitations stem from specific choices that are made in the activity-dependent degeneration ( ADD ) algorithm that was used in this study . The algorithm was chosen because it is based on the singular assumption that neuronal hyperactivity leads to dysfunction and damage , and because of its strong resemblance with Alzheimer-related network degeneration , including features like early-stage disinhibition and hub vulnerability [18] . However , in the ADD algorithm the network is damaged regardless of the absolute level of activity; even regions with a normal level of activity will be damaged , although exponentially less ( see ‘loss function’ in S1 Text ) . A degeneration regime that spares regions that exhibit a normal range of neuronal activity might be more plausible ( metabolically ) , and could result in strategies that will be able to maintain network integrity for a longer , perhaps even indefinite , period of time . Also , neuronal plasticity as a defense mechanism to degeneration was not incorporated in this model , but may yield more realistic results and enhance network survival . For example , mechanisms such as synchronization-dependent or growth-dependent plasticity could be implemented in de model [66] . Our rather ‘unforgiving’ degeneration mechanism may have underestimated the impact of the various therapies , and alternative algorithm choices may lead to different intervention outcomes . However , adding more assumptions ( and complexity ) to the model or its degeneration algorithm will probably enlarge the unpredictability of the interventions . Finally , the chosen Vd stepsize of 0 . 5 in our analysis was a pragmatic choice , and although our Vd analysis in the Supporting Information ( S2 Text ) suggests that Vd values in between would not produce very different results , a smaller stepsize may have been more insightful . With the growing interest in the human connectome in general , and more specifically in connectivity as a biomarker of neurodegenerative disease [67 , 68] , intervention modeling is a logical step forward . Computer simulations serving as a virtual test ground for intervening in complex systems is common practice in related fields , such as economy , meteorology or systems biology . With the advent of enhanced acquisition techniques such as DTI and MEG , enabling increasingly accurate human connectivity datasets , and of increasing computational power , analysis will become more elaborate , faster and hopefully more user-friendly and available to clinical researchers . Two long-term aims seem meaningful to pursue: the further development of dynamic connectome ( “dynome” ) intervention models , and the validation of these models with clinical experiments , perhaps involving individual structural connectome data to achieve personalized results . This way , the modeling approach may become a viable intermediate step in the bidirectional interpretation of both etiological hypotheses and clinical experiments in connectivity-related research in general [56] . The observations in this computational modeling study surprisingly suggest that AD-like network degeneration due to neuronal hyperactivity can be countered most effectively by global stimulation of excitatory neurons . The wide-ranging and unpredictable impact of intervention strategies in this straightforward dynamic human connectome model , with a limited number of identical neural masses and macro scale cerebral topology , illustrates the fact that interventions in complex systems can lead to counterintuitive results . In general , network intervention analysis may help to explore and explain therapeutic effects aimed at preserving network integrity , and thereby potentially guide the design and hypothesis selection of clinical intervention trials .
For this study we simulated neurophysiologic activity of cortical regions embedded in a realistic structural network topology to evaluate hypotheses about the relation between neuronal activity and ( structural and functional ) connectivity . The output of this model provides information about neuronal activity in the form of average voltage and spike density per region , and generates EEG-like data that can be subjected to further analysis . Furthermore , hypotheses about brain pathophysiology can be tested by artificially changing structural or dynamical properties of the model . The general workflow of our analysis can be described as follows ( see also Fig 1 for a graphical overview ) : the dynamic network model is run with the degeneration algorithm and , simultaneously , one of the interventions ( or no intervention ) . This means that the network is damaged according to local neuronal activity levels , but at the same time , by changing neuronal excitability levels due to varied threshold potential ( Vd ) settings ( see below for details ) , a counterstrategy is employed to diminish the effect of the damage and maintain network topology close to the original state . The resulting oscillatory , connectivity and network topology changes are then described using the selected quantitative outcome measures ( see below ) to evaluate the effect of the different interventions over time , and in the final step these are compared statistically to obtain an impression of the most successful strategy . We used a model of interconnected neural masses , where each neural mass represents a large population of excitatory and inhibitory neurons generating an EEG ( or MEG ) like signal . The model was recently employed in two other graph theoretical studies [18 , 69] . The basic unit of the model is a neural mass ( NM ) of the alpha rhythm [70–72] . This model considers the average activity in relatively large groups of interacting excitatory and inhibitory neurons . Spatial effects ( i . e . distance ) are ignored in this model; brain topology is introduced by coupling multiple NMs together . The average membrane potential and spike density of the excitatory neurons of each of the NMs separately were the multichannel output that was subject to further analysis . A diffusion tensor imaging ( DTI ) based study by Gong et al . published in 2009 that focused on large-scale structural connectivity of the human cortex resulted in a connectivity matrix of 78 cortical regions [73 , 74] . The connectivity matrix was implemented in our model software , and used as topological framework for the 78 coupled NMs . Coupling between two NMs , if present , was always reciprocal , and excitatory . Note that at the start of the simulation , the coupling strength between all NM pairs was identical , and the only difference between the cortical regions ( or NMs ) was their degree of connectivity to other NMs ( cortical regions ) . Please refer to the supporting information for full details . The neural mass model described above was extended to be able to deal with activity dependent evolution of connection strength between multiple coupled NMs . Activity dependent degeneration ( ADD ) was realized by lowering the ‘synaptic’ coupling strength as a function of the spike density of the main excitatory neurons ( all neural mass model parameters and functions are summarized and explained in S1 Text ) , in a similar way as previously described ( 18 ) . The effects of ADD were measured by changes in ‘total power’ ( local average membrane potential ) and spike density , and these two measures were used as representations of neuronal activity in further analyses . The computational model is incorporated in our custom developed analysis software ( ‘BrainWave’ , v0 . 9 . 151 . 5 ) , written by C . J . Stam ( available for download at http://home . kpn . nl/stam7883/brainwave . html ) . For the present study , we introduced counterstrategies against ADD that involve altering the neuronal excitability , either at a global level or of excitatory or inhibitory neural masses selectively . In the neural mass model , a transfer function determines the translation of membrane potential to spike density and vice versa ( see S1 Text ) . Vm is the average membrane potential , and Vd is the threshold potential ( Vd1 for excitatory , Vd2 for inhibitory neurons ) . Altering the level of Vd results in a sigmoid function describing the resulting spike density ( see S1C Fig ) ; a lower threshold leads to a higher spike density and vice versa . This way , the excitability of a neural mass can be changed either for excitatory or inhibitory groups selectively , or for both simultaneously . In this model , extremely low or high neuronal excitability levels ( Vd1 or Vd2 lower than 4 or higher than 10 , ) cause the system to quickly reach non-functional states , either shutting down functional connectivity or generating cascades of uncontrolled activity , respectively ( reminiscent of epileptic seizure activity ) . Therefore , these were excluded from further analysis . Within the biologically plausible range , we tested various excitability levels and their effect on network dynamics ( see also Fig 2 ) . For clarity purposes , we limited the number of strategies to six distinct types: global ( both excitatory and inhibitory ) stimulation ( Vd1 = 6 , Vd2 = 6 ) , global inhibition ( Vd1 = 8 , Vd2 = 8 ) , selective stimulation of excitatory neurons ( Vd1 = 5 , Vd2 = 7 ) , selective stimulation of inhibitory neurons ( Vd1 = 7 , Vd2 = 5 ) , selective inhibition of excitatory neurons ( Vd1 = 8 , Vd2 = 7 ) , and selective inhibition of inhibitory neurons ( Vd1 = 7 , Vd2 = 6 . 5 ) . The strategies consisted of maintaining constant neuronal excitability levels; the initial settings of a strategy did not change over time during the degeneration period . Since for each of these interventions the threshold potential adjustment is arbitrary , we conducted simulations with six different values within each category ( for example , for global stimulation six Vd1 settings between 4 and 6 . 5 were used , with a 0 . 5 increment ) , and compared the findings to pick a representative value . See S2 Text for an illustration of this analysis . All intervention strategies were initiated after ten degeneration cycles without any therapy , at T = 11 , to simulate the manifestation of disease and subsequent therapy initiation with some delay . Model output is generated for >50 cycles of the degeneration algorithm , to simulate a progressive neurodegenerative process over time . Therefore , the X-axis variable has been defined as ‘virtual time’ . Note that there is no defined relation to real time; this parameter should not be interpreted as hours , days or otherwise . In this study , we compared relative differences within the model .
|
Alzheimer’s disease ( AD ) is a growing burden on society , without a cure in sight . Pathological high neuronal activity and excitability is an increasingly observed phenomenon in early stage AD . Its exact role in the disease process is unclear , but it may form an interesting therapeutic target . However , although brain dynamics can be influenced in many ways , the highly complex nature of the brain makes it difficult to predict what approach will be most effective . To test our hypothesis that neuronal hyperactivity can be countered effectively by altering neuronal excitability levels , we examined various strategies aimed at preserving brain network integrity in a computational AD model of the human brain . Of these strategies , a scenario involving stimulation of excitatory neurons extends the period with normal network function most successfully . The results of this ‘virtual trial’ suggest that network effects of pathological neuronal activity can be opposed by selective altering of neuronal excitability levels . In general , this approach can explore therapeutic effects aimed at preserving or restoring brain network integrity , and thereby contribute to selecting promising interventions for future clinical trials in AD .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"infographics",
"cognitive",
"neurology",
"medicine",
"and",
"health",
"sciences",
"neural",
"networks",
"neurodegenerative",
"diseases",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"cognitive",
"neuroscience",
"mathematics",
"algebra",
"network",
"analysis",
"alzheimer's",
"disease",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"cognitive",
"impairment",
"dementia",
"mental",
"health",
"and",
"psychiatry",
"cellular",
"neuroscience",
"cell",
"biology",
"data",
"visualization",
"neurology",
"neurons",
"physiology",
"graphs",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"algebraic",
"topology",
"topology",
"cognitive",
"science"
] |
2017
|
Altering neuronal excitability to preserve network connectivity in a computational model of Alzheimer's disease
|
Merkel cell polyomavirus ( MCPyV ) is considered the etiological agent of Merkel cell carcinoma and persists asymptomatically in the majority of its healthy hosts . Largely due to the lack of appropriate model systems , the mechanisms of viral replication and MCPyV persistence remain poorly understood . Using a semi-permissive replication system , we here report a comprehensive analysis of the role of the MCPyV-encoded microRNA ( miRNA ) mcv-miR-M1 during short and long-term replication of authentic MCPyV episomes . We demonstrate that cells harboring intact episomes express high levels of the viral miRNA , and that expression of mcv-miR-M1 limits DNA replication . Furthermore , we present RACE , RNA-seq and ChIP-seq studies which allow insight in the viral transcription program and mechanisms of miRNA expression . While our data suggest that mcv-miR-M1 can be expressed from canonical late strand transcripts , we also present evidence for the existence of an independent miRNA promoter that is embedded within early strand coding sequences . We also report that MCPyV genomes can establish episomal persistence in a small number of cells for several months , a time period during which viral DNA as well as LT-Ag and viral miRNA expression can be detected via western blotting , FISH , qPCR and southern blot analyses . Strikingly , despite enhanced replication in short term DNA replication assays , a mutant unable to express the viral miRNA was severely limited in its ability to establish long-term persistence . Our data suggest that MCPyV may have evolved strategies to enter a non- or low level vegetative stage of infection which could aid the virus in establishing and maintaining a lifelong persistence .
Merkel cell polyomavirus ( MCPyV ) , first identified in 2008 in tissue from Merkel cell carcinoma ( MCC ) [1] , is the only human polyomavirus considered to be the etiological agent of tumors arising in its natural host . Several lines of evidence , including frequent detection of monoclonally integrated sequences bearing hallmark mutations and constitutive expression of T antigens in tumor tissues suggest that the virus is causally linked to MCC pathogenesis ( reviewed in [2] ) . Epidemiological studies suggest that MCPyV infection occurs in childhood and persists for life in the majority of the adult healthy population [3–7] . Hence , the occurrence of MCC is an extremely rare complication of MCPyV infection . Like all polyomaviruses , MCPyV encodes the early large and small T antigens ( LT- and sT-Ag ) , as well as the late structural antigens VP1 and VP2 [1 , 8] . Whether MCPyV also expresses a functional VP3 antigen remains a matter of debate [9] . Polyomavirus T antigens are produced via alternative splicing from a single gene cassette that is transcribed early during infection . In MCPyV , alternative splicing of early transcripts additionally produces a 57K T antigen of hitherto unknown function [8] . A recent study furthermore revealed the existence of an alternative open reading frame ( ALTO ) which can be produced by leaky scanning of T-Ag encoding transcripts [10] . Although ALTO shares certain sequence features with the middle T antigens ( mT-Ag ) of other polyomaviruses , its precise functions remain unknown . Experimental evidence suggests that presence or absence of ALTO does not affect viral DNA replication [10] . In addition to above protein products , MCPyV has been found to encode a single microRNA ( miRNA ) precursor which can produce two mature miRNAs , termed mcv-miR-M1-5p and -3p [11] . miRNAs are small ( ~22 nt . ) , non-coding RNAs that can be produced from primary transcripts via sequential processing by the nucleases Dicer and Drosha [12] . After incorporation into the RNA-induced silencing complex ( RISC ) , mature miRNA can negatively regulate the expression of transcripts that are recognized via sequence complementarity . In animals , target site recognition is primarily guided by perfect Watson-Crick pairing of the so-called seed sequence ( nucleotides 2–8 ) of the miRNA , whereas the distal sequences typically only exhibit poor sequence complementarity [13] . This partial pairing leads to translational inhibition of the mRNA ( although frequently a modest reduction in overall transcript levels can also be observed ) . Although rarely seen for animal miRNAs , plant miRNAs as well as miRNAs encoded by some animal viruses can also bind to their targets with perfect complementarity , resulting in RISC-mediated endonucleolytic cleavage of the mRNA . Recent studies have shown that a number of human and animal polyomaviruses encode miRNAs [11 , 14–19] . Although their precise genomic location varies , all known polyomavirus miRNA are expressed from sequences that are located in antisense orientation to the early T antigen encoding transcripts . Consequently , mature miRNA species expressed from these loci exhibit perfect complementarity to early transcripts , and a number of studies suggest that all hitherto identified polyomavirus miRNAs share the ability to negatively regulate expression of early gene products [11 , 15–17 , 19 , 20] . For most of the known PyV miRNAs ( including MCPyV ) , experimental evidence for the above is limited to ectopic heterologous reporter systems . However , miRNA-knockout viruses have been generated for SV40 , murine PyV and BKPyV , and in vitro studies using such viruses demonstrated that the viral miRNAs are indeed able to efficiently limit LT-Ag expression as well as DNA replication in the context of authentic episomes [15 , 16 , 20] . So far , experimental in vivo infections with miRNA-deficient viruses have only been performed for SV40 and murine PyV [15 , 21] . Indeed , miRNA-deficient SV40 mutants produce consistently higher viral DNA loads in both liver and kidney of infected syrian golden hamsters when compared to wt viruses . However , both wt and mutant viruses were able to establish persistent infections , and thus far only limited evidence for increased clearance of miRNA-mutants has been observed [21] . In the case of murine PyV , the kinetics of both infection establishment as well as subsequent viral clearance in experimentally inoculated mice were comparable between wt and mutant viruses , indicating that ( at least under the experimental conditions used ) murine PyV miRNA expression is not essential for the infection of mice [15] . The above therefore suggests that the role of PyV miRNAs during natural infection may involve aspects of acquisition , spread or persistence which are not properly recapitulated by the experimental in vivo systems used . Hence , while evolutionary conservation suggests important function for miRNA-mediated autoregulation of LT-Ag expression and DNA replication , the precise selectional advantage conferred by this regulatory mechanism remains unclear [22–25] . The molecular mechanisms that lead to polyomavirus miRNA expression thus far have not been studied in much detail . Circumstantial evidence , however , suggests that at least in some polyomaviruses transcriptional read-through beyond weak late strand polyadenylation signals can generate primary RNA molecules that traverse the miRNA precursor sequences [15–17] . In such a model , miRNA expression is coupled to expression of coding transcripts that originate from the late promoter in the non-coding control region ( NCCR ) . Indeed , a recent study of BK polyomavirus ( BKPyV ) has demonstrated that NCCR rearrangements which naturally arise in patients suffering from BKPyV-associated disease result in decreased late strand transcription and miRNA expression [20] . In contrast , archetype viruses express robust levels of the viral miRNA , which in turn dampens T antigen expression and viral replication . As the archetype virus is thought to be responsible for establishment of persistent urinary tract infections , these findings suggest that , similar to herpesviruses , polyomaviruses may employ miRNAs to facilitate chronic infection of their host [20 , 26] . Whether similar mechanisms as the above may dictate viral miRNA expression in MCPyV , a virus that is only distantly related to BKPyV , has thus far not been elucidated . Given its association with human tumors , experimental research on MCPyV thus far has been largely focused on growth promoting and transforming functions of early viral gene products . In contrast , there is a profound lack of knowledge regarding the natural life cycle of the virus . In large part , this is due to the fact that all currently available in vitro systems produce only very low titers of viral progeny [27–30] . Although recent evidence suggests that MCPyV may persist in the hematopoietic compartment [31–33] , it is unknown which type of cell may support viral replication and/or serve as a reservoir for persistent infection in vivo . It is therefore also unclear whether the low transmissibility observed in vitro reflects an inherent property of the virus ( e . g . , similar to what is observed for archetype BKPyV ) or simply results from the lack of appropriate cell culture systems . In addition to ( and partially as a result of ) the above deficits , there is only very limited knowledge regarding the MCPyV transcription program . Thus far , experimental studies addressing this subject have mainly employed subgenomic MCPyV fragments under the control of heterologous promoters to study expression and processing of the viral miRNA , or to explore the structure and coding potential of early region transcripts [8 , 11] . Additionally , endogenous expression of early gene products and the viral miRNA has been investigated in MCC-derived cell lines ( MCCL ) or MCC tissues [11 , 19 , 34 , 35] . These studies have shown that the defective viral genomes integrated in MCC constitutively express proteins encoded by the early region , but only produce the viral miRNA at low levels . Thus , it remains unknown whether intact episomal MCPyV genomes express the miRNA at levels which permit efficient autoregulation of LT-Ag expression and viral DNA replication . We have previously established a semi-permissive replication system which is based on synthetic MCPyV genomes ( MCVSyn ) that are 100% identical to prototypical field strain sequences [27] . After transfection , viral genomes undergo active DNA replication , express early and late antigens , and produce infectious progeny ( albeit at very low titers ) . Here , we have used the above system to study the viral transcription program and elucidate expression mechanisms and functions of the viral miRNA during short and long term culture of cells harboring actively replicating episomes .
Ectopic expression of a select number of computationally predicted pre-miRNA candidates previously identified a single pre-miRNA hairpin ( termed mcv-miR-M1 ) encoded by MCPyV [11] , located in an antisense orientation to the early coding region at genomic coordinates 1168 to 1251 ( Fig 1A ) . While low-level expression of mature miRNAs from this genomic locus has been confirmed in primary MCC tissues [34] , an unbiased investigation of small RNAs produced from intact and replicating viral episomes had previously not been performed . We therefore sought to determine i ) whether the previously identified miRNA mcv-miR-M1 is expressed at significant levels by replicating episomes , ii ) whether mcv-miR-M1 is the only miRNA expressed by authentic MCPyV genomes and iii ) whether mature mcv-miR-M1 moieties may undergo differential processing in MCC-derived cell lines ( MCCL ) . The latter point was of particular interest given that a recent study reported MCC-derived mature 5p miRNAs that differed from those described by Seo et al . in a 2 nt . shift , resulting in an altered seed sequence and a therefore a differential set of predicted cellular target transcripts [11 , 34] . To investigate above questions , we transfected the neuroectodermal tumor cell line PFSK-1 cells with MCVSyn , a viral genome that is 100% identical to prototypical field strain sequences [27] . Small RNA moieties were harvested after 4 days of transfection and subjected to high throughput sequencing using the NEBNext library preparation protocol . As shown in Fig 1A , the great majority ( 99 . 5% ) of all MCVSyn-derived small RNAs mapped to the previously identified mcv-miR-M1 locus . The remaining reads were randomly scattered across the viral genome ( not visible at the scale shown in Fig 1A; see S1 Dataset for complete coverage data ) , suggesting they represent random mRNA breakdown products . Hence , the previously identified mcv-miR-M1 is the only miRNA expressed by actively replicating MCPyV genomes . Comparison of viral and human miRNA read counts suggests that mcv-miR-M1-derived miRNAs are highly expressed in MCVSyn-transfected PFSK-1 cells , accounting for approximately 3% of the total of 19 . 3 million mature miRNA reads ( Table 1 ) . Mature miRNAs derived from the 5'-arm of the pre-miRNA hairpin were approximately twofold more abundant than those derived from the 3'-arm ( mcv-miRs-M1-5p and -3p , respectively , in Table 1 ) . Even though transfection efficiencies achieved with re-circularized genomes were generally below 5% , mcv-miRs-M1-5p and -3p were the 7th and 14th most highly expressed miRNAs , respectively , amongst all mature miRNAs detected in PFSK-1:MCVSyn cultures ( Table 2 and S2 Dataset ) . As shown in Fig 1B , in perfect accord with the original findings by Seo et al . we find that the majority of mcv-miR-M1-5p reads are derived from nucleotides 16–37 of the pre-miRNA hairpin . The seed sequence ( GGAAGAA ) in this miRNA extends from nucleotide 17–23 of the pre-miRNA ( underlined in Fig 1B ) . Overall , mature miRNAs with this seed ( referred to as 5p17-23 species in the following ) accounted for greater than 94% of all 5p reads ( left panel in Fig 1C , green bars , and S3 Dataset ) . We additionally detected alternatively processed mature 5p species ( isomiRs ) of lower abundance . Although the mature miRNA species identified by Lee and colleagues ( seed sequence CUGGAAG , termed 5p15-21 in the following ) in MCC tissues was the most prominent among these , its relative abundance accounted for only 5 . 5% of all 5p reads . To investigate whether these miRNA species may be more abundant in MCC-derived cells , we performed additional small RNA sequencing from the MCPyV-positive MCCLs WaGa and MKL-1 . In both cell lines , the frequency of mcv-miR-M1-derived miRNAs was more than 3 orders of magnitude lower than in MCVSyn-transfected PFSK-1 cells ( approx . 0 . 001% of all mature miRNA reads , see Table 1 ) . However , the relative distribution of seed sequences was very similar to that seen in PFSK-1 MCVSyn cells , with the seed sequence observed by Lee being only marginally abundant at 10 to 14% ( red and blue bars in the left panel of Fig 1C ) . We hypothesized that a potential bias during library preparation might have been responsible for the discrepancies between our or Seo et al . 's results and those observed by Lee and colleagues . It is well documented that biases especially during the ligation step can result in a gross underrepresentation of individual miRNA species between different library preparation methods [36–40] . To formally investigate this possibility , we re-sequenced the same small RNA material using the standard Illumina TruSeq small RNA library preparation kit . Indeed , as shown in the right panel of Fig 1C this analysis primarily recovered mature miRNAs of type 5p15-21 . Importantly , however , as in the first set of experiments , the observed seed distribution was similar between PFSK-1 MCVSyn , WaGa and MKL-1 cells , demonstrating that MCPyV miRNAs do not undergo differential processing in MCC-derived cell lines . Generally , normalized read counts for 5p15-21 miRNAs were comparable between the two library preparation methods while those for 5p17-23 species were about 100fold less abundant in the TruSeq experiments ( S3 Dataset ) , suggesting that the observed differences in relative seed distributions were likely due to failure to retrieve 5p17-23 species during TruSeq library preparation . As for mature 5p miRNAs , 3p miRNA species also showed a differential seed sequence distribution depending on whether libraries were prepared with NEBnext or TruSeq protocols ( Fig 1D ) . Again , however , we observed no major differences between miRNA processing in MCCL or PFSK-1 MCVSyn cells . The most abundant 3p species in the NEB dataset mapped to nts . 51–72 of the mvc-mir-M1 hairpin ( seed sequence 52–58: UGCUGGA , see Fig 1B and 1D ) , whereas in the TruSeq data , reads were more evenly distributed between this miRNA species and an isomiR offset by -1 nucleotide . Regardless of their exact seed sequence , all mature miRNA species are perfectly complementary to transcripts originating from the opposite strand of the MCPyV genome . Consequently , provided they are efficiently incorporated into RISC , all species should be able to negatively regulate early transcripts . Indeed , a number of previous studies have suggested that the ability to autoregulate LT-Ag expression may represent an evolutionary conserved function of polyomavirus miRNAs [11 , 15–17 , 19 , 20 , 22–25] . In support of this notion , Seo and colleagues have formally demonstrated that ectopically expressed mcv-miR-M1 can negatively regulate luciferase expression of a chimeric reporter construct containing the mcv-mir-M1 complementary region and flanking sequences [11] . To determine whether mcv-mir-M1 also can suppress LT-Ag expression in the context of transcription from intact episomes , we generated a mutant MCPyV genome unable to express the viral miRNA . To this end , we introduced a total of 14 mutations designed to disrupt the mcv-mir-M1 pre-miRNA hairpin structure ( Fig 2A ) which is required for the processing of pre- and mature miRNAs by Drosha and Dicer , respectively . All nucleotide substitutions were designed such that the coding capacity of LT-Ag encoded on the opposite strand remained unaltered ( S1 Fig ) . The resulting viral genome ( referred to as MCVSyn-hpko in the following ) or the parental MCVSyn genome was transfected into PFSK-1 cells . As shown in Fig 2B , parental MCVSyn genomes expressed viral miRNA moieties that were readily detectable by northern blotting at 4 days post infection . As expected , no mcv-mir-M1 pre- or mature miRNAs were produced in cells transfected with MCVSyn-hpko mutants . As shown in Fig 2C , the absence of viral miRNA expression resulted in substantially higher expression of LT-Ag on the protein level . To investigate whether elevated LT-Ag expression also affected the efficacy of viral DNA replication , we performed a DpnI resistance assay on HIRT extracts . As shown in the Southern Blots of Fig 2D , MCVSyn-hpko genomes indeed replicated to appreciably higher levels when compared to the wt genome . Collectively , the above data thus suggest that i ) complementary target sites in full length LT-Ag transcripts are accessible to mcv-mir-M1 binding , ii ) levels of mcv-mir-M1 expressed by replicating MCPyV genomes are sufficient to induce substantial downregulation of LT-Ag expression and iii ) the extent of LT-Ag downregulation mediated by mcv-mir-M1 is sufficient to limit the replication of transfected MCPyV genomes . The use of our semi-permissive replication system extended the possibility to perform an in depth analysis of transcripts expressed by intact viral episomes . In addition to providing valuable information about the structure of coding transcripts , we expected that such analyses would also provide clues with regard to the mechanisms that control viral miRNA expression . As MCPyV transcripts have thus far only been evaluated by Northern Blotting in cells transfected with early region expression cassettes driven by a heterologous CMV promoter [8] , the location of transcriptional initiation and polyadenylation sites remains unknown . Since such sites are often difficult to capture in standard RNA-seq protocols due to the usually poor coverage of accurate 5'- and 3' transcript ends , we performed 5'- and 3'-RACE on RNA isolated from MCVSyn transfected PFSK-1 cells after 4 days of transfection . Polyadenylation sites were determined with a conventional 3’RACE protocol , using gene specific 5'-primers for the distal coding regions of early and late transcripts together with anchored oligo dT 3'-primers ( Fig 3A ) . Amplification products were subcloned in bulk , and between 16 and 26 ( for early and late transcripts , respectively ) randomly picked clones were subjected to Sanger sequencing . As shown in Fig 3A and 3B , 100% ( 16 of 16 ) clones derived from early transcripts terminated at position 3094 , 14 nucleotides downstream of a canonical polyadenylation signal ( AAUAAA ) which overlaps with the T-Ag stop codon , and 9 nucleotides upstream of a GU-rich element ( Fig 3B ) . The 3’-RACE products from late transcripts were more diverse: As shown in Fig 3A and 3C , 14 ( 53% ) of the 26 clones terminated at position 2842 ( pA site L1 ) , 317 nucleotides downstream of the VP1 stop codon . Another 8 clones ( 31% ) terminated in a distance of 451 from the VP1 stop codon at position 2708 ( pA site L2 ) . Canonical polyadenylation signals were observed immediately upstream of both cleavage sites ( Fig 3C ) . Although U-rich regions are present 13 or 31 nucleotides downstream of pA sites L1 and L2 , respectively , neither site exhibits a clearly discernible GU-rich element . The four remaining clones from late transcripts were predominantly found at A-rich regions of the viral genome , suggesting they had resulted from mispriming of the oligo dT primers to internal regions of viral transcripts extending into the early region . Together , these results suggested highly efficient termination of early transcripts , but relatively weak late polyadenylation signals that allow at least some transcriptional read-through . For the determination of transcriptional initiation sites we employed Cap-dependent 5'-RACE , a protocol which greatly decreases the rate of false positives that result from degradation products and/or premature termination of reverse transcription . Gene-specific RT-PCR anchor primer sites for late transcripts were situated approximately 400 nucleotides downstream of the VP2 start codon ( Fig 4A ) . To allow the detection of putative transcripts that may initiate near the recently identified ALTO reading frame [10] , primers for early transcripts were designed to bind to a region in the second exon of the LT-Ag . As we expected that initiation sites may be more heterogeneous than polyadenylation sites , amplification products were analyzed by high throughput sequencing ( HTS ) instead of Sanger sequencing . After mapping of reads to the MCPyV genome , we counted the number of reads that initiated at a given nucleotide position . Only nucleotides which received at least 1% of the total reads were considered as potential transcriptional initiation sites . As shown in Fig 4B and 4C , the great majority ( 93% ) of the ~55 . 000 analyzed reads from early transcripts initiated between nucleotides 147–150 ( TI-E1 in Fig 4B and 4C ) with a marked peak at position 149 ( 56 . 7% of all reads ) . As shown in Fig 4C , a canonical TATA Box is present 26 nucleotides upstream of the major initiation site . A second , much weaker accumulation of reads ( 4% ) was observed between nucleotides 112 and 120 , with a peak at nucleotide position 115 ( marked with an asterisk in Fig 4C ) , suggesting that a minority of early transcripts initiates upstream of the TATA box . Of note , approximately 10% of the corresponding reads exhibited a splice junction which fused nucleotide 141 to the previously identified splice acceptor of the second LT-Ag exon . This splice event generates a transcript in which the first AUG triplet is the start codon of the ALTO open reading frame , 49 nucleotides downstream of the transcript's 5'-end . While we have formally confirmed the existence of the junction by RT-PCR primers ( S2 Fig , lane 3 ) , whether or not such rare transcripts contribute to the production of ALTO remains to be established . The remainder of reads was randomly scattered across the viral genome , indicating they were derived from breakdown or premature RT termination products . Consistent with the fact that the region between the origin of replication and VP2 lacks a canonical TATA box , we observed that late transcripts were derived from a broader initiation zone ( termed TI-L1 in the following ) located between nucleotides 5264 and 5222 , with a total of 15 nucleotide positions accumulating at least 1% of the ~159 . 000 total reads ( Fig 4B and upper panel in Fig 4D , S4 Dataset ) . The bulk of initiation sites ( ~72% ) mapped to a C/T-rich region between nucleotides 5241 and 5250 , with the major initiation site ( 35% of all reads ) being located at position 5245 , 127 basepairs upstream of the VP2 start codon . Interestingly , another 4 . 452 reads ( 2 . 8% ) mapped to nucleotide position 1367 , well outside of the NCCR ( TI-L2 in Fig 4B and lower panel in Fig 4D ) . The observed initiation site is located 116 nucleotides upstream of the mcv-miR-M1 locus , suggesting the existence of miRNA-encoding transcripts that originate outside of the NCCR . Given the observation of transcripts initiating upstream of the viral miRNA we sought to investigate whether mcv-miR-M1 could be expressed independently of NCCR-initiated transcription . For this purpose , we sub-cloned the entire early T-Ag coding region in either sense or antisense orientation downstream of an heterologous CMV promoter ( pCMV:ER-S and –AS , respectively; see Fig 5A ) . As expected , forced transcription of the early region antisense strand gave rise to readily detectable pre- and mature miRNA moieties of mcv-miR-M1 ( Fig 5B , left panel ) . However , similar levels of miRNA expression were observed when the CMV-promoter initiated transcription traversed the early region in the sense ( i . e . T-Ag coding ) orientation ( Fig 5B , center panel ) . Indeed , a promoterless construct harboring the entire early region ( pER ) was likewise able to express the viral miRNA ( Fig 5B , right panel ) , albeit at considerably ( approx . 10 fold ) lower levels than either CMV promoter-driven construct ( see GAPDH-normalized stem-loop RT-qPCR data in Fig 5C; note that the Northern Blot in the right panel Fig 5B was exposed for longer time period than those shown for the pCMV constructs ) . While we presently cannot explain the seemingly disparate observation that strong miRNA expression was observed independent of the CMV promoter’s orientation relative to mcv-miR-M1 , we suspect that CMV-promoter driven transcription through the locus may activate an intrinsic promoter . The fact that we had observed a transcriptional initiation site approx . 100 nt . upstream of the miRNA using a 5’-CAP dependent RACE protocol suggested that such transcripts are likely produced by RNA polymerase II . To investigate this assumption , we treated pER-transfected PFSK-1 cells with α-amanitin , a potent inhibitor of RNA-polymerase ( RNA-pol ) II , and investigated mcv-miR-M1 expression 24 hours later . As controls for RNA pol II and III transcribed RNAs , we additionally measured levels of GAPDH mRNA and tRNA-meth , respectively . As shown in Fig 5D , α-amanitin treatment strongly reduced expression of GAPDH and mcv-miR-M1 , but not that of tRNA-meth . Hence , an intrinsic promoter activity within in the early region of MCPyV can lead to RNA pol II-dependent transcription of mcv-miR-M1 . If the region upstream of the mcv-miR-M1 locus exhibits promoter activity , then intact episome should exhibit an open chromatin conformation that permits transcriptional initiation at this position . To investigate this notion , we performed ChIP-seq experiments to evaluate patterns of the postranslational histone modification H3K4me3 , a mark which is strongly enriched at transcriptional start sites . Additionally , we performed ChIP-seq to elucidate binding patterns of LT-Ag binding across the viral episome . For this purpose , we transfected PFSK-1 cells with MCVSyn and , 48 hours later , performed chromatin immunoprecipitation with antibodies specific for H3K4me3 , or with the LT-Ag antibody CM2B4 . An immunoprecipitation with IgG served as a negative control . The complete coverage data is given in S5 Dataset . As shown in the top panel of Fig 6B , the CM2B4 antibody produced a marked peak centered at the core origin of replication , consistent with the previously observed binding of LT-Ag to an array of GRGCC pentamers located in this region [41 , 42] . No additional peaks were observed , indicating that , at least under the conditions used here , LT-Ag appears not to stringently bind to other loci on the viral episome . As expected , the NCCR of MCPyV was also highly enriched in the activation-associated histone mark H3K4me3 ( Fig 6B , center panel ) . The H3K4me3 profile in this region presented as a broad peak which extended from the late to the early transcription initiation sites mapped during our 5’-RACE analysis . Indeed , consistent with our previous experiments that had suggested promoter activity of the region upstream of the viral miRNA locus , a second prominent H3K4me3-enriched zone was located within the T-Ag coding region . The summit of this peak mapped precisely to the transcriptional start site upstream of the viral miRNA locus identified during our 5’-RACE analysis . We next sought to determine whether mutation of upstream sequences would negatively affect miRNA expression . Given that the H3K4me3 enriched region is located in the early coding region , deletion of the putative promoter would also disrupt LT-Ag expression and thus abrogate the replication ability of MCVSyn episomes . However , we hypothesized that introduction of synonymous triplet mutations which preserve the LT-Ag coding capacity may be sufficient to ablate or reduce promoter activity . Accordingly , we generated a MCVSyn mutant ( termed MCVSyn-pmt in the following ) with a total of 68 triplet mutations in a ~200 bp region located 28 nt upstream of the miRNA locus ( Fig 6C ) . Introduction of the same set of mutations in the context of early region-only construct pER ( see Fig 5A ) and subsequent transfection of the resulting plasmid pER-pmt into PFSK-1 cells confirmed that the mutations led to a significant reduction of NCCR-independent miRNA expression ( S3 Fig ) . To investigate the effect of the mutations on miRNA expression by full length genomes , MCVSyn-pmt or the parental MCVSyn construct were transfected into PFSK-1 cells , and levels of mature mcv-miR-M1-5p were evaluated 48h later by quantitative stem-loop PCR . The hairpin knockout mutant MCVSyn-hpko served as a negative control . As shown in Fig 6D , MCVSyn-pmt indeed expressed mcv-miR-M1-5p at significantly lower levels compared to the wildtype genome , albeit the residual expression levels ( approx . 60% ) were higher than those observed with the early region construct pER-pmt ( approx . 25%; see S3 Fig ) . To investigate the effect of the introduced mutations on the chromatin level , we additionally performed ChIP-seq analysis of PFSK-1 cells transfected with MCVSyn-pmt . As expected , neither LT-Ag binding to the viral origin nor H3K4me3 accumulation at the NCCR was affected in MCVSyn-pmt . However , in accord with the observed decrease in mcv-miR-M1-5p expression levels , the mutations resulted in the almost complete elimination of the H3K4me3 peak upstream of the miRNA locus . Collectively , the above data thus suggest that the genomic region upstream of mcv-miR-M1 exhibits promoter activity and can contribute to NCCR-independent expression of the viral miRNA in the context of replicating episomes . However , given that elimination of the H3K4m3 peak had reduced but not abrogated miRNA expression , we suspected that a considerable fraction of the viral miRNA may also be generated from late strand transcripts that originate from the NCCR and ignore the apparently weak late strand polyadenylation sites . To investigate this possibility , and to furthermore evaluate the overall structure of viral transcripts and the influence of the viral miRNA on transcript abundance , we proceeded to perform strand-specific mRNA-seq experiments . To evaluate viral transcription patterns we transfected PFSK-1 cells with MCVSyn or the miRNA knockout MCVSyn-hpko and harvested mRNA after 4 days of transfection . Two independent rounds of transfection and sequencing were carried out for each the parental and mutant genomes . In Fig 7B and 7C , we present coverage plots which represent the accumulated/mean data of the replicates . S4 Fig shows the individual data plots for each of the replicates and demonstrates that both experiments produced near-identical results . Full coverage data are provided in S6 Dataset . Consistent with the observation of a highly efficient early polyadenylation site we observed that the coverage of early transcripts exhibited a sharp decline towards the 3'-end of T-Ag coding sequences . As shown in the upper plot in Fig 7B and Table 3 , in MCVSyn transfected cells greater than 99% of all reads from the early strand mapped to the region flanked by the transcriptional initiation and polyadenylation sites identified in our RACE analysis . In contrast , the coverage data for transcripts originating from the late strand were indicative of profound read-through beyond polyadenylation sites pA-L1 and-L2 . While approx . 60% of late strand reads mapped to the region delineated by the initiation site TS-L1 and the late polyadenylation sites , the remaining 40% were derived from the antisense strand of the early region and the NCCR . As shown in Fig 7C , the overall late strand coverage profiles in MCVSyn-hpko transfected profiles were near identical to those observed in cells transfected with the parental genome . However , consistent with the observation that mcv-miR-M1 negatively regulates LT-Ag expression , the relative fraction of reads derived from early transcripts was considerably increased in MCVSyn-hpko transfected cells ( 90% in MCVSyn-hpko vs . 61% in MCVSyn transfected cells; see Table 3 ) . Moreover , early strand coverage profiles in PFSK-1 MCVSyn-hpko cells showed increased coverage immediately downstream of the region antisense to mcv-miR-M1 . This is consistent with the fact that the RNA-seq protocol captures polyadenylated RNAs and therefore selects for 3’-fragments of miRNA-cleaved transcripts . In contrast , 5’-cleavage products lack a polyA tail and thus are lost prior to library preparation , resulting in a relative decrease in read counts upstream of the cleavage site . Together , the above data thus suggest that viral MCPyV miRNAs negatively regulate T-Ag expression via cleavage of early strand transcripts . Thus far , elucidation of viral splice patterns has been limited to evaluation of the ectopically expressed early region [8] . To investigate splice patterns of early and late transcripts expressed from full-length genomes , we analyzed the structure and frequency of spliced reads from our RNA-seq experiments . We considered such reads as evidence of an authentic splice event if i ) the event was supported by at least two independent observations among the individual experiments , ii ) the junction was consistently observed in both replicates of MCVSyn or MCVSyn-hpko transfections and iii ) the splice sites exhibited the sequence features commonly observed at donor and acceptor sites . For each of the identified sites , we additionally calculated the number of unspliced reads to evaluate the efficiency with which the given site underwent splicing . In Table 4 , we present the identified donor sites and junctions along with the read numbers and splice frequencies as calculated from the accumulated data from both datasets of MCVSyn or MCVSyn-hpko transfected cells . The structures of known or novel early and late transcripts are shown in Fig 7D and 7E , respectively . For transcripts mapping to the major early and late transcription cassettes , an estimation of relative abundance is shown after each transcript . S5 Fig shows the sequence context of known and the novel splice sites observed in this study . The great majority of splice events among early transcript mapped to the junctions previously identified by Shuda and colleagues [8] . Together , these transcripts ( T1 to T4 in Fig 7D ) are estimated to account for approximately 95% of all early strand mRNAs . Additional splice junctions were detected only at low frequency . The corresponding putative mRNAs include transcripts ( tentatively named T’5 through T’7 ) , which are predicted to encode T antigens with estimated molecular weights of 9 , 21 or 64 kDa . All of these protein products contain the first 93 amino acids shared by LT- and sT-Ag , but entirely or partially lack the sequences encoded by the second exon of LT-Ag . Similar low early region transcripts of low abundance have been previously observed in other polyomaviruses , but in most cases it is not clear whether their protein products are of biological significance [43–49] . While splice events that originated outside of the major early transcription cassette were infrequent , the majority of such events consisted of the junction already observed during our 5’-RACE analysis in the putative ALTO-encoding transcript ( d141-a861 in Table 4 , transcript T’8 in Fig 7E ) . Additionally , we observed a very rare splice event , which uses the same acceptor at position 861 , but a donor upstream of the origin ( d5355 ) . Such transcripts may either be produced by upstream initiation events which were too infrequent to be picked up by our RACE analysis , or by occasional read through beyond the early region polyadenylation site . RT-PCR formally confirmed the existence of this rare splice event ( S2 Fig , lane 1 ) . Among the late strand transcripts we detected a total of 5 splice junctions involving 2 donor and 3 acceptor sites ( Table 4 and Fig 7E ) . Abundance estimation predicts that the majority of late messages are unspliced transcripts which encode VP2 ( L1 in Fig 7E ) . Splicing from a donor at position 5145 to an acceptor 503 nt downstream ( a5119 ) in approximately 10–15% of late strand transcripts generates a message in which the first AUG codon initiates the VP1 ORF ( transcript L2 ) . The same donor is joined to an alternative acceptor at position 5119 in another 2–4% of late transcripts ( L3 in Fig 7E ) . This event leads to removal of an immediately upstream of the VP2 start codon , and the resulting L3 transcripts are thus predicted to code for VP2 . We did not detect transcripts which splice to the start codon of the predicted VP3 ORF . Interestingly , 4–5% of all splice events observed for the donor at position 5145 ( Table 4 ) connect to an acceptor ( a5308 ) which is located upstream of the late transcriptional start sites identified in our 5’-RACE analysis ( LL in Fig 7E ) . This splice event consequently requires primary transcripts which traverse the entire episome , similar to the leader-to-leader splice observed in other polyomaviruses [50–55] . RT-PCR analysis with junction spanning primers indicates that multiple copies of the leader can be present at the 5’-end of late transcripts ( S2 Fig , lane 7 ) , indicating that the RNA polymerase can complete several rounds of transcription along the viral episome . As the leader sequence does not contain putative AUG start codons , its presence is not expected to alter the coding capacity of transcripts L1 , L2 or L3 . We additionally detected another splice event which extended over the NCCR , joining a donor at downstream of the viral miRNA ( d1142 ) to the acceptor at position 5308 . While the existence of this splice was confirmed by RT-PCR ( S2 Fig , lane 5 ) , only ~5–13% of all reads traversing the donor are spliced ( Table 4 ) , and transcripts containing this splice ( tentatively named L’4 in Fig 7E ) are therefore expected to be rare . As such transcripts may originate from transcriptional read-through beyond late polyadenylation sites or from transcripts which are initiated upstream of the viral miRNA , their coding potential remains unknown . The results described thus far demonstrate that mcv-miR-M1 efficiently suppresses early gene expression and viral DNA replication between two and four days post transfection . These time points were chosen because they guarantee robust genome amplification and allow readily detectable expression of viral genes . Interestingly , however , in an independent set of experiments we had repeatedly observed that , after transfection in a number of cell lines , MCPyV genomes remained detectable for several weeks or even months by Southern Blotting and qPCR . To confirm these findings in PFSK-1 cells , and to furthermore investigate influence of the viral miRNA expression on long term persistence of MCPyV genomes , we transfected PFSK-1 cells with MCVSyn or MCVSyn-hpko and monitored the resulting cultures for a period of at least 3 months . At regular intervals , we collected total DNA , small RNA and protein to evaluate relative MCPyV genome copy numbers as well as expression of mcv-miR-M1 and LT-Ag . Fig 8A shows relative genome copy numbers and viral miRNA expression of the wt MCVSyn episome as determined by qPCR or quantitative stem-loop RT-PCR , respectively . All values were normalized for genomic GAPDH locus copy numbers and are shown relative to the earliest sampled time point at d2 post transfection ( set to 1 ) . Consistent with the previous results from our short-term DNA replication assays , MCVSyn genomes exhibited an initial increase of relative copy numbers within the first ~10 days , which was followed by a steep decline over more than two orders of magnitude in the following two weeks . After this loss phase , however , MCVSyn copy numbers did not further decline , suggesting that viral genomes had entered a state of near-stable long term maintenance . Furthermore , temporal changes of viral miRNA expression levels closely mirrored changes in relative genome copy numbers , indicating the per-genome expression levels of mcv-miR-M1 remained stable over the course of the experiment . Interestingly , and contrary to what might have been expected based on increased DNA replication in short term assays ( Fig 2D ) , MCVSyn-hpko genomes were unable to reach a state of stable maintenance . Whereas copy numbers of parental genomes remained stable even beyond a 6 month time point , the hairpin mutant was progressively lost from the cultures such that it became undetectable by day 105 ( blue and red symbols , respectively , in Fig 8B ) . The qPCR results were furthermore confirmed by Southern Blotting of DpnI-resistant DNA ( Fig 8C; note that owing to the lower sensitivity of these assays MCVSyn-hpko genomes become already undetectable at day 70 ) . Western Blot analysis of the bulk cultures confirmed the absence of LT antigen in MCVSyn-hpko transfected cells after the loss of genomic DNA ( Fig 9A , lane 12 ) . In contrast , LT antigen expression could be readily observed in MCVSyn transfected cultures even after more than 160 days ( lane 11 ) . To also analyze LT-Ag expression on the single cell level , we performed immunofluorescence analyses using the CM2B4 antibody . Fig 9B shows representative images from an early ( 4d ) and several late time points of MCVSyn transfected cells . LT-Ag staining presented as distinct , strictly nuclear dots that are likely to represent foci of viral DNA replication . At 4 days post-transfection , we estimated the percentage of LT-Ag positive cells in both MCVSyn and MCVSyn-hpko transfected cultures to be approximately 2–3% . In accord with our qPCR and southern blot experiments , LT-Ag positive cells became approximately 100fold less frequent , but cells with multiple foci ( albeit smaller than those observed after 4 days ) remained clearly detectable for several months in MCVSyn cultures . Apart from the fact that LT-Ag positive cells were absent from late time points of MCVSyn-hpko-transfected cultures , we did not detect fundamental differences in the LT-Ag staining patterns between MCVSyn or MCVSyn-hpko transfected cells . In Fig 8D and 8E , we show two independent repeats of our long term maintenance assays . Although parental MCVSyn genomes were less efficiently maintained in these experiments , the miRNA-deficient mutant consistently demonstrated an accelerated rate of loss and became undetectable at least four weeks earlier than the wt genome . Although PFSK-1 cells produce only very low levels of infectious virus particles and do not allow efficient serial transmission [27] , we considered it formally possible that the decreased long term persistence of MCVSyn-hpko genomes may reflect alterations in particle production . To directly investigate this scenario , we inspected viral genome copy numbers in total genomic DNA and DNaseI treated supernatants from freshly transfected PFSK1 cells ( 4d p . t . ) . In accord with our previous results , total copy numbers of viral genomes were higher in MCVSyn-hpko transfected cells ( S6A Fig ) . However , there was no appreciable difference between DNaseI-resistant genome copy numbers in the supernatants of MCVSyn or MCVSyn-hpko transfected cells , indicating comparable levels of virion production . We additionally used freeze-thaw lysates from such cultures to inoculate fresh PFSK-1 cultures and measure the amount of nuclear viral DNA recovered after 4 or 8 days post-inoculation ( S6B Fig ) . As expected , the overall amount of DNA recovered from infected cells was strongly reduced compared to levels observed in transfected input cultures . Again , however , we did not detect significant differences between cultures inoculated with lysates from MCVSyn or MCVSyn-hpko-transfected cells , suggesting that the differences observed in long-term maintenance assays are likely to be independent of potentially altered particle production levels . Given the very low level of infectious viral particles produced in PFSK-1 cells , we hypothesized that long term persistence of MCVSyn genomes could either result from efficient episomal maintenance , or ( similar to MCC-derived cell lines ) reflect stable transmission of integrated genomes . To investigate these possibilities , we first established a FISH assay for MCPyV . As a control for the sensitivity and specificity of the assay , we analyzed the two MCPyV positive MCC cell lines MKL-1 and WaGa cells . As shown in Fig 10A , we detected a single distinct signal per cell in MKL-1 and two foci in WaGa cells , indicative of one or two integration events , respectively . In contrast , FISH analysis of the long-term PFSK:MCVSyn cultures shown in Figs 8A–8C and 9 detected a considerably larger number of foci per cell nucleus ( Fig 10B; see S7 Fig for exemplary images taken at a lower magnification ) . In accord with our LT-Ag immunofluorescence assays , while approximately 2 . 5% of all cells were positive for MCVSyn at 4 days post transfection , the number of positive cells dropped over time and reached a steady state of ~0 . 01% at late time points . In agreement with the qPCR and Southern Blotting results , we were able to detect MCVSyn positive cells by FISH for more than 160 days ( lower panel in Fig 10B ) . While the observation of multiple nuclear foci of viral DNA argues against rare integration events being responsible for long-term maintenance , FISH analysis cannot provide definite proof of episomal persistence . To more directly address this issue , we therefore performed rolling circle amplification ( RCA ) for MCPyV DNA , a protocol which selectively amplifies circular templates and produces large concatameric DNA molecules [56] . In Fig 11 , we present an RCA analysis of DNA isolated from PFSK-1: MCVSyn cultures ( the same cultures as shown in Figs 8A–8C , 9 and 10B ) at 4 or 136 days post transfection ( lanes 3–4 and 9–10 , respectively ) . Genomic DNA from the MCC-derived cell lines WaGa and MKL-1 ( lanes 5–6 and 7–8 , respectively ) served as a control for cells harboring integrated viral genomes . Indeed , while no RCA products were observed in mock-transfected PFSK-1 cells or the two MCCL cultures , the material from early and late PFSK-1:MCVSyn cultures yielded efficiently amplified viral DNA . Collectively , the data presented in Figs 8–11 thus suggest that MCPyV genomes are able to persist as extrachromosomal episomes for several months after transfection into PFSK-1 cells , and furthermore that a miRNA knockout mutant is considerably impaired in its ability to establish long term episomal maintenance .
In this study , we report an in-depth analysis of the MCPyV-encoded miRNA miR-M1 and its functions during short and long-term replication of intact viral episomes . Besides of confirming prior studies which had suggested that mcv-miR-M1 autoregulates T-Ag expression , we demonstrate that mcv-miR-M1 can be expressed independently of NCCR-initiated transcription and uncover an unexpected role for the viral miRNA in episomal persistence . Our small RNA sequencing data suggest that replicating MCPyV genomes express the viral miRNA to very high levels . In our analysis of MCVSyn-transfected PFSK-1 bulk cultures , mature mcv-miR-M1 species ranked among the top 15 of all miRNAs . Owing to generally low transfection efficiencies achieved with re-circularized genomes , only ~2–5% of cells in such cultures carry the viral genome . Hence , it is likely that mcv-miR-M1 species dominate the spectrum of expressed miRNAs in MCVSyn-positive cells . In contrast , we find that mcv-miR-M1 is expressed at only very low levels in MCC-derived cell lines . The observed frequencies of viral miRNA reads ( ~0 . 001% ) are in very good accord with those calculated from a recent metastudy of mcv-miR-M1 expression in primary tumor material ( 0 . 002% ) [19] . Overall , when taking into account transfection efficiencies , mcv-miR-M1 expression levels are estimated to be more than four orders of magnitude higher in cells harboring replicating episomes . Our cross-comparison of mcv-miR-M1 expression levels in PFSK1:MCVSyn and MCCL thus strongly supports the previous notion that the viral miRNA is unlikely to contribute to the progression of MCC via the continuous downregulation of cellular target transcripts [19] . We also find no evidence for the hypothesis that MCC cells may preferentially express an isomiR variant that could target host immune response genes [34] . Our side-by-side comparison of two different library preparation methods rather shows that the relative distribution of miRNA seeds is very similar between MCCL and PFSK1:MCVSyn cells . While the isomiR identified by Lee et al . was indeed the most prominent variant when using one of the two investigated library preparation methods , absolute and relative read counts ( Table 1 and S3 Dataset ) strongly suggest that the seeming dominance of this isomiR was due to failure to capture the 5p17-23 species . The observed discrepancies likely resulted from biases during small RNA library preparation , most notably the influence of small RNA and adapter sequence combinations on the efficiency of 3’ adapter ligation ( 27–31 ) . Given the difficulties in determining the accurate seed sequences of viral miRNAs , our results thus once more underline the notion that identification of potential host targetomes must be based on unbiased experimental screens with authentic precursors rather than computational prediction alone . Non-withstanding above considerations , our data show that mcv-miR-M1 expression negatively regulates expression of early messages transcribed from the opposite strand . Similar to SV40 , murine PyV and BKPyV [15 , 16 , 20] , a miRNA knockout mutant exhibited appreciably higher levels of LT-Ag expression and DNA replication , but did not produce significantly altered levels of viral progeny ( S6 Fig ) . However , given that the currently available MCPyV replication system generally produces only very low titers of infectious progeny , it remains possible that mcv-miR-M1 may behave differently in a fully permissive system . Comparison of relative transcript abundance between MCVSyn or MCVSyn-hpko transfected cells suggests that the majority of early transcripts are negatively regulated by miR-M1 , as they become more abundant in cells harboring the miRNA knockout ( Fig 7 ) . A notable exception is transcript T’5 , which is approximately 30fold more abundant in MCVSyn-transfected cells . This observation is consistent with the fact that the d420-a2778 splice removes the sequences complementary to mcv-miR-M1 . As the miRNA knockout is expected to selectively destabilize those transcripts which contain target sites , it is to be expected that T’5 accounts for a lower relative fraction of early viral transcripts in MCVSyn-hpko cells . It is also interesting to note that , among the remaining transcripts , those that encode LT-Ag and 57K-Ag ( T1 and T4 , respectively ) appear to be most strongly upregulated in PFSK-1:MCVSyn-hpko cells . This may suggest that they are more efficiently targeted by mcv-miR-M1 , e . g . due to secondary structures that facilitate binding of mature miRNAs to their target sites . Certainly , however , further investigation will be required to establish whether this is indeed the case . Our RNA-seq analysis also identified early strand splice events which originate outside of the major transcriptional cassette defined by 5’- and 3’-RACE analyses . These products are of interest as they may produce dedicated ALTO-encoding messages . While we have formally confirmed the existence of these junctions , RNA-seq coverage also indicates that they are of very low abundance . It is thus unlikely that such transcripts significantly contribute to ALTO production in our system , considering that the protein can be efficiently produced from canonical early region transcripts via leaky scanning [10] . However , it remains possible that increased usage of alternative upstream initiation sites ( e . g . that observed at position 115 ) could elevate transcript abundance . Previous studies from other polyomaviruses such as mouse PyV , SV40 and JCPyV indicate that upstream initiation sites are increasingly used during late stages of the infection cycle [57–59] . Once a fully permissive MCPyV replication system becomes available , it thus will be interesting to study whether similar mechanisms could lead to increased expression of dedicated ALTO-messages during late stages of a productive infection . We detected two splice junctions that map within the major late strand transcription cassette . While the resulting transcripts are predicted to produce VP1 and VP2 , we do not observe splice events which remove the VP2 start codon to produce a dedicated transcript for the putative minor capsid protein VP3 . As the VP2 start codon is in a strong Kozak context [9] , translation of VP3 from above transcripts via leaky scanning is predicted to be inefficient . Our data are thus in support of a recent study that has concluded MCPyV is unlikely to express a functional VP3 [9] . Interestingly , we find that MCPyV also expresses late strand transcripts with multiple copies of 5’-structures that are reminiscent of the leader-to-leader splice events observed in mouse polyomavirus [50–55] . Similar to the early stages of mouse PyV infection [60] , we find that the major transcriptional start site is located downstream of the leader-to-leader splice acceptor . As productive mouse PyV infection proceeds , transcriptional initiation occurs with increasing frequency at alternatives sites which are located upstream of the leader-to-leader splice acceptor [60] . As for early messages , it is thus possible that a similar shift in transcriptional initiation site usage may occur in a fully permissive infection system . Production of leader-to-leader transcripts from primary transcripts which ignore late strand polyadenylation sites has been shown to contribute to the accumulation of late strand transcripts during productive mouse polyomavirus infection [51 , 53 , 55] . It was suggested that the presence of nuclear antisense RNAs produced from the intron of the late leader-to-leader splice leads to abundant A-to-I editing and subsequent destabilization of early transcripts . Since the processed late mRNA does not contain the complementary intron sequences , it is not subject to editing and thus the ratio of late versus early messages increases [54] . To investigate whether similar mechanisms may occur in our system we scrutinized our RNA-seq data for evidence of A-to-I editing events . Even when allowing 20 mismatches during the alignment step , the rate of A-to-I transitions was below 0 . 1% . Hence , at least under the conditions and in the semi-permissive system used here , negative regulation of early genes by late strand transcription products appears to proceed predominantly via expression of the viral miRNA . Notably , this does not rule out a role for leader-to-leader splicing , as these events would allow processing of the miRNA from intronic sequences while still preserving the integrity of late strand coding mRNAs . Given that mouse PyV encodes a miRNA at a similar genomic location [15] it appears possible that intron-derived miRNAs could also contribute to accumulation of late mouse PyV mRNAs . In addition to miRNA expression from canonical late strand transcripts , we also provide evidence for NCCR-independent expression of mcv-miR-M1 . Evidence for this conclusion includes ( i ) identification of a transcriptional initiation site ( TI-L2 ) upstream of the genomic miRNA locus which ( ii ) exhibits profound enrichment of the histone modification H3K4me3 , ( iii ) autonomous expression of the viral miRNA from subgenomic fragments containing the early coding region and ( iv ) an approximately 2fold reduction in miRNA expression concomitant with ( v ) elimination of H3K4me3 peaks upon the introduction of triplet mutations upstream of the genomic mcv-miR-M1 locus . At first , the 2fold reduction in miRNA expression by MCVSyn-pmt may appear unexpected when considering the relatively low frequency of 5’-RACE products observed at TI-L2 ( approx . 3% ) . However , as exonic miRNA processing destroys the precursor transcript , the 5’-RACE protocol can only capture those transcripts which have escaped processing by Drosha . The primary rate of transcriptional initiation events at this site may thus be higher . It should also be noted that , while our data suggest that transcription occurs via RNA polymerase II , we were unable to drive expression of luciferase via the early strand coding fragments upstream of the viral miRNA locus . Hence , it is possible that sequences downstream of the initiation site may be required for efficient expression , perhaps similar to the as of yet undefined genomic pre-miRNA sequence features that can mediate autonomous transcription of some human miRNA loci [61] . The fact that distally initiated transcription through the mcv-miR-M1 locus seems ( Fig 5 ) seems to increase intrinsic promoter activity is also of interest , as such a mechanism could potentially provide a negative feedback for early gene expression . Interestingly , expression of a viral miRNA from an internal promoter has been previously reported for bandicoot papillomatosis carcinomatosis virus type 1 and 2 ( BPCV1 and -2 , respectively ) , two viruses which shares distinct features of both the polyomavirus and papillomavirus families [62–64] . In BPCV1 and -2 , the non-coding region 2 ( NCR2 ) contains the genomic template as well as promoter for miRNA expression [64] . It was noted that the NCR2 region also contains a predicted LT-Ag consensus binding site ( GRGGC ) , yet whether BPCV LT-Ag indeed binds to this site remains unknown [62–64] . Indeed , the region flanking initiation site TI-L2 contains a cluster of 6 GRGGC pentamers , including two overlapping sites in a head-to-head orientation just 6 nucleotides upstream of the transcriptional start position ( S8 Fig ) . All sites are furthermore perfectly conserved in Gorilla gorilla gorilla polyomavirus 1 ( GggPyV1 ) and Pan troglodytes verus polyomavirus 2 ( PtvPyV2a ) , two close relatives of MCPyV that have been shown to encode orthologues of mcv-miR-M1 [64 , 65] . While it is intriguing that , aside from the viral origin of replication , no other locus in the MCPyV , GggPyV1 or PtvPyV2a genomes exhibits a similar accumulation of GRGGC pentamers , we clearly did not observe LT-Ag peaks upstream of the viral miRNA in our ChIP-seq experiments ( Fig 6 ) . Hence , if LT-Ag indeed binds to these sites under the conditions used here , it must do so transiently or with significantly reduced affinity compared to the viral origin of replication . Considering all of the above , additional experiments will undoubtedly be required to fully characterize factors and features which regulate NCCR-independent expression of mcv-miR-M1 . If mcv-miR-M1 expression can occur independent of late gene expression , then why do integrated genomes in MCC fail to efficiently express the viral miRNA ? While we presently can only speculate , given that high level early antigen expression is required for survival of MCC cells [66] we would predict that tumor progression selects for silencing of the viral miRNA promoter . We are currently evaluating the chromatin status of integrated genomes to investigate this scenario . Perhaps the most intriguing finding of our study is that , in the absence of any selection pressure , wild type MCPyV genomes can persist in continuously growing cultures for more than 6 months after the initial transfection . In contrast , the mcv-miR-M1 knockout mutant was consistently lost at an accelerated rate . The fact that MCVSyn genome copy numbers reached a stable plateau phase in only one out of the three independently performed experiments , however , also suggests that long-term maintenance is affected by as of yet unknown stochastic events . While the results presented in Fig 11 clearly argue against chromosomal integration constituting such an event , another possibility would be the accumulation of adaptive mutations or genomic rearrangements in long term cultures . The fact that long term maintenance did not require selection would seem to argue against such a possibility . More importantly , however , we have subjected RCA amplification products from long-term cultures to high-throughput sequencing and found their sequences to be 100% identical to those of the input genomes . It is interesting to note that the observations made here bear some resemblance to the latency establishment phase of the gammaherpesviruses EBV and KSHV . While incoming genomes rapidly adopt latent gene expression profiles , EBV as well as KSHV episomes exhibit an accelerated loss rate during the first few weeks of infection until rare epigenetic events of hitherto unknown nature lead to stabilization of episomes and subsequent long term maintenance [67–69] . Considering the transient increase in copy numbers observed at day 84 of the experiment shown in Fig 8A–8C , it is tempting to speculate that a stochastic event occurring at this time point may have allowed subsequent stabilization of MCPyV episomes . Unfortunately , due to the very low frequency of MCPyV-positive cells at this or later time points we were unable to investigate the chromatin state of stable viral episomes in these cultures . We are currently striving to enrich cells harboring stable MCPyV episomes to allow a detailed investigation of their epigenetic landscape and gene expression profile . It should be pointed out that we currently cannot exclude the possibility that MCPyV maintenance may also involve continuous shedding of infectious virions . Thus , long term maintenance may result from latency-like episomal persistence , continuous low level infection of initially untransfected cells ( or cells having lost the viral genome ) , or a combination of such processes . However , in our view the fact that PFSK-1 cells do not allow efficient particle production and serial transmission together with the observation that mcv-miR-M1 deficiency does not affect the levels of viral progeny ( S6 Fig ) argues against continuous virion shedding as the primary maintenance mechanism . We are currently performing long term studies with additional MCPyV mutants to investigate whether persistence requires virion production . Why is the miRNA-knockout mutant MCVSyn-hpko impaired in long-term persistence ? Several possibilities come to mind . Firstly , given that small RNAs can induce local chromatin changes via hitherto poorly understood mechanisms [70–73] , viral miRNA expression may negatively affect a potentially stabilizing event as discussed above . While formally possible , we do not consider this possibility to be very likely . We instead favor a second and much simpler possibility: That continued mcv-miR-M1 expression prevents accumulation of LT-Ag protein to levels which impair cell growth . It has previously been shown that LT-Ag can have growth promoting activities , but via carboxyterminal sequences may also inhibit proliferation if expressed at high levels [74] . We have formally confirmed that PFSK-1 cells respond to LT-Ag expression with a dose dependent decrease in proliferation rates ( S9 Fig ) . Hence , we suspect that lack of the viral miRNA results in aberrantly high LT-Ag expression , leading to accelerated loss of MCPyV-positive cells and a decreased probability that genome stabilization may occur . This model provides a convenient explanation for the fact that MCVSyn-hpko genomes exhibit increased DNA replication at early time points , but nevertheless exhibit accelerated loss in long-term cultures . Lastly , a third ( and not mutually exclusive ) possibility is that , similar to some herpesvirus miRNAs [22 , 23 , 75–77] , mcv-miR-M1 may downregulate host transcripts to create a cellular environment that is supportive of long-term episomal maintenance . Although attractive , experimental identification of candidate host targets ( preferably via unbiased screens ) will be required to substantiate such a scenario . What is the biological significance of the observed ability of MCPyV episomes to persist over extended periods of time ? It is tantalizing to speculate that MCPyV may have evolved similar mechanisms as papillomaviruses to persist in a non-vegetative state of infection . However , it must be pointed out that it is currently unclear to what extend the cell system used here adequately reflects the behavior of MCPyV-infected cells in vivo . While PFSK-1 cells ( like all other cell lines or-types tested thus far ) do not support lytic growth of MCPyV , the precise cells types in which the virus establishes productive and/or persistent infections in vivo remain unknown . Since , obviously , long term persistence would not provide a significant benefit in fully permissive infected cells ( which would likely die within a few days after initial infection ) , one would thus have to postulate that there may be semi-permissive host cell types ( or differentiation stages ) in which the virus establishes latent or quasi-latent infections . Although we find this to be a very intriguing possibility , verification of such models will ultimately have to await identification of the authentic in vivo target cells in healthy carriers . Finally , are the findings reported here of any relevance for the pathogenesis of MCC ? At present , the limited amount of available data does not allow us to draw such a conclusion . Within the limits of the caveats discussed above , we propose that , similar to BKPyV [20] , the physiological function of the MCPyV miRNA may be to augment viral persistence which , given the existence of an autonomous promoter , may proceed in a non-vegetative state of infection . If so , one may also speculate that prolonged episomal persistence could increase the overall chance of integration events that are expected to be extremely rare during natural infection , but which are a hallmark of all MCPyV-positive MCC . Thus , while spurious expression of the viral miRNA is likely to be inconsequential once integration has occurred , in the above scenario mcv-miR-M1 may very well make an indirect contribution by supporting long-term persistence of viral episomes . Of note , LT-Ag has been demonstrated to directly interact with the bromodomain protein BRD4 , a factor which is targeted by several papillomavirus E2 to mediate episomal tethering during persistent infection [78–83] . Although highly speculative , given recent findings suggesting a role for the E2/BRD4 interaction during papillomavirus integration [84] , one may envision that the interaction between BRD4 and LT-Ag could also more directly contribute to MCPyV integration events . Indeed , our own preliminary studies indicate that LT-Ag binds to a large number of host chromosome loci in a non-random manner . We are currently investigating whether such loci may also represent preferred sites of chromosomal integration in MCC . In the meantime , the findings reported here open new lines of investigation that are expected to significantly improve our understanding of the MCPyV lifecycle .
PFSK-1 cells ( ATCC , CRL-2060 ) and MCC cell lines ( MKL-1 [85] and WaGa [8] ) were maintained in RPMI medium supplemented with 10% fetal calf serum ( FCS ) and 5% penicillin/streptomycin . HEK293 cells [86] were cultured in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% FCS and 5% penicillin/streptomycin . All cells were grown in a 5% CO2 humidified atmosphere at 37°C . Construction of the synthetic consensus clone pMK-MCVSyn used for production of re-circularized MCPyV genomes has been described previously [27] . To generate the hairpin mutant MCVSyn-hpko , a 384 bp fragment spanning nucleotides 1118–1501 of the MCVSyn genome and containing 14 nucleotide substitutions in the region encoding mcv-miR-M1 ( S1 Fig ) was synthetically generated ( GeneART , Regensburg ) . A fragment spanning the same genomic region , but containing 68 nucleotide exchanges ( S1 Fig ) in the suspected promoter region upstream of the mcv-miR-M1 locus was synthesized to generate the mutant genome MCVSyn-pmt . Both fragments were inserted into pMK-MCVSyn using BamHI and SanDI restriction sites . Plasmid pER was produced by amplification of the MCPyV early coding region from MCVSyn using primers MCPyV EcoRV F/MCPyV XhoI R and subsequent cloning into the pCR2 . 1 plasmid ( life technologies ) . To generate plasmid pER-pmt , the region upstream of the viral miRNA locus in pER was substituted by the mutated region from MCVSyn-pmt . Plasmids pCMV:ER-AS and pCMV:ER-S were generated by excision of the MCPyV early coding region from plasmid pER and directional cloning into pCDNA3 . 1 . All primer- and oligonucleotide-sequences used in this study are listed in S1 Table . Re-circularization and transfection were carried out as described previously [27] . Briefly , the bacterial backbone of pMK-MCVSyn , pMK-MCVSyn-hpko or pMK-MCVSyn-pmt plasmids was excised by SacI digestion ( FastDigest , Thermo Scientific ) . After gel purification of the linearized viral genomes , intramolecular ligation was performed using T4 DNA Ligase ( Thermo Scientific ) , followed by spin column purification of the recircularized DNA . 3x105 PFSK-1 cells were seeded in 6-well dishes one day prior to transfection . 200 ng of re-circularized viral DNA together with 500 ng carrier DNA ( pUC18 ) were transfected using X-tremeGene HP DNA Transfection Reagent ( Roche ) following the manufacturer’s instructions . Mock treated cells were transfected with equivalent amounts of carrier DNA only . Cells were harvested for analysis at the indicated time points . DpnI resistance/replication assays were performed after HIRT extraction of low molecular weight DNA ( HIRT DNA ) as previously described [87] . Extracted DNA was subsequently digested with EcoRI and DpnI ( FastDigest , Thermo Scientific ) for 60 min at 37°C . 1 μg of the resulting DNA was separated on a 0 . 8% agarose gel and transferred to a nylon membrane ( Zeta-Probe GT Membrane , Bio-Rad ) . For detection of MCVSyn DNA , a genomic fragment was amplified from the early viral region using primers MCPyVLT fw and MCPyV LT rev and labeled with 32P dCTP ( Rediprime II DNA Labeling System , GE Healthcare ) . Blots were hybridized with the labeled probe for 16h at 42°C in ULTRAhyb buffer ( Ambion ) . Blots were washed 2x 20 min with 1% SSC , 0 . 1% SDS and 2x20 min with 0 . 1x SSC , 0 . 1% SDS at 50°C . After at least 24 h of exposure , blots were scanned with the Fuji phosphorimager FLA7000 and analyzed with Multigauge software . For the determination of MCPyV genome copy numbers , genomic DNA ( gDNA ) was extracted from isolated nuclei of transfected cells . Nuclei were prepared by adding 2x lysis buffer ( 0 . 65 M Sucrose , 20 mM Tris-HCl pH 7 . 8 , 10 mM MgCl2 , 2% Triton-X 100 ) to a final concentration of 1x to trypsinized cells resuspended in PBS . After 5 min incubation on ice , nuclei were centrifuged and resuspended in 50 μL PBS . 300 μL gDNA lysis buffer ( 100 mM NaCl , 10 mM Tris-HCl pH 8 , 25 mM EDTA pH 8 , 0 . 5% SDS ) supplemented with 200 μg proteinase K ( Peqlab ) were added , followed by incubation at 54°C for 16 h . The DNA was purified by phenol/chloroform extraction and isopropanol precipitation . After treatment with RNase A ( Peqlab ) for 30 min at 37°C gDNA was digested with EcoRI and DpnI ( FastDigest , Thermo Scientific ) for 60 min at 37°C . 25 ng gDNA were used as input for quantitative PCR . Viral genomes were quantified with primers binding to the late region ( MCPyV VP1 fw and MCPyV VP1 rev ) , spanning three DpnI restriction sites . SacI digested , linear MCVSyn DNA was used to generate a standard curve with a defined number of viral genomes . The number of viral genomes was normalized to GAPDH locus copy numbers ( primers: GAPDH DNA fw and GAPDH rev ) , which were determined using a standard curve with a defined number of GAPDH locus copies . Rolling circle amplification was performed with the TempliPhi 100 amplification kit ( Amersham Biosciences ) according to the manufacturer's instructions with additional 450 μM dNTP as described in Rector et al . 2004 [88] . RCA products were digested with restriction enzymes , separated on a 0 . 8% agarose gel and analyzed by ethidium bromide staining . Libraries of RCA products for HTS analysis were generated with the NEBNext Ultra DNA Library Prep Kit for Illumina and sequenced on an Illumina HiSeq2500 instrument . Reads were subjected to de novo assembly using the Trinity package ( v r2013-02-25 ) [89] and resulting full-length MCPyV genomes ( average coverage >200 ) were compared to input sequences by blast analysis ( BLAST Plus Package v 2 . 2 . 28 ) . For detection of small RNAs by northern blotting , total RNA was harvested using RNABee ( AMS Biotechnology ) according to the manufacturer's instructions . 14 μg of total RNA were separated on a denaturing 15% polyacrylamide urea gel and transferred to Zeta-Probe GT membranes ( Bio-Rad ) by electro blotting . Blots were hybridized to a 32P dATP labeled antisense oligonucleotide probe ( mcv-miR-M1 probe , S1 Table ) in ExpressHyb ( BD Biosciences Clontech ) hybridization buffer for 16h at 37°C . Membranes were washed twice in 2x SSC , 0 . 1% SDS at room temperature and subsequently subjected to autoradiography . Blots were scanned on the BAS-Reader and analyzed with AIDA Software . For quantitative stem-loop RT-PCR , reverse transcription ( RT ) was performed as described by Varkonyi-Gasic et al . [90] using 1 μg of total RNA as input for each sample and a mcv-miR-M1 specific stem-loop primer ( SL mcv-miR-M1 ) . For normalization , a reverse primer for GAPDH ( GAPDH rev ) was included in the RT reaction . 1 . 5 μL cDNA per sample were analyzed by real-time PCR on the Rotor-Gene Q ( Qiagen ) using the Rotor-Gene Multiplex PCR Mastermix ( Qiagen ) and the following primer pairs: mcv-miR-M1: mcv-miR-M1 fw/universal rev; GAPDH: GAPDH BSP fw/GAPDH rev . Differently labeled TaqMan probes ( Taqman probe mcv-miR-M1 , Taqman probe GAPDH ) were used to quantify the expression of mcv-miR-M1 and GAPDH in the same real-time PCR reaction . Analysis of real-time PCR experiments was carried out with Rotor-Gene Q software ( Qiagen ) . Chromatin Immunoprecipitation ( ChIP ) was performed as previously described [91 , 92] . In brief , 4d post transfection with MCVSyn wt or MCVSyn mutants , chromatin of 1×106 cells was crosslinked by incubation with 1% formaldehyde . The reaction was stopped by the addition of glycine . Chromatin was extracted from isolated nuclei and fragmented by sonication ( Bioruptor , Diagenode ) to an average length of 200–500 bp . A fraction of the total chromatin sample was set aside for the preparation of input control . The remaining material was pre-cleared with BSA blocked protein-G sepharose beads ( GE Healthcare ) to reduce non-specific background binding . For immunoprecipitation , 2 μg of antibodies specific for the histone modification H3K4-me3 ( Millipore , #04–745 ) or for MCPyV LT-Ag ( CM2B4 , Santa Cruz Biotechnology , sc-136172 ) or IgG anti-rabbit ( Millipore , #12–370 ) antibody were added to the chromatin and incubated for 16 h at 4°C . Chromatin-immunocomplexes were precipitated by the addition of protein-G sepharose beads , washed with increasing salt concentrations , eluted and de-crosslinked for 16 h at 65°C . DNA was purified by phenol-chloroform extraction and ethanol precipitation . For HTS analysis , libraries were prepared from ChIP samples using the NextFlex ChIP-Seq kit ( Bioo Scientific ) and sequenced on the Illumina HiSeq2500 . Reads were mapped to MCVSyn or MCVSyn-pmt genome sequences using Bowtie ( v 0 . 12 . 9 ) . For Western Blotting , MCVSyn transfected cells were harvested at the indicated time points and resuspended in lysis buffer ( 50 mM Tris pH 8 . 0 , 150 mM NaCl , 1% NP40 , 0 . 5% Na-Deoxycholat , 5 mM EDTA , 0 . 1% SDS , proteinase inhibitor cocktail , Roche ) . 25 μg of protein were separated by SDS-PAGE ( 10% gels ) and electroblotted on a PVDF-membrane . Blots were incubated with MCPyV LT-Ag antibody CM2B4 ( Santa Cruz , sc-136172 ) . For immunofluorescence analyses , PFSK-1 cells grown on coverslips were fixed with methanol at room temperature for 30 min and rinsed in PBS for 10 min . Fixed cells were blocked with 4% BSA in PBS for 30 min and then incubated with a 1:50 dilution of the LT-Ag antibody CM2B4 in 4% BSA in PBS/0 . 05% Tween-20 for 2 h . Coverslips were washed three times in PBS for 10 min each , followed by incubation with a 1:1000 dilution of goat anti-mouse Alexa Fluor 555-conjugated secondary antibody ( Life Technologies , A 21422 ) in 4% BSA in PBS/0 . 05% Tween-20 for 2 h . Coverslips were washed three times in PBS for 10 min each , counterstained and mounted with vectashield mounting medium with DAPI ( Vector , H-1200 ) . Images were acquired with a confocal laser-scanning microscope ( Nikon C2+ ) . 5x104 cells were cytospun for 5 min at 900 rpm on Superfrost/Plus slides ( Fisher ) and fixed in methanol , followed by digestion ( 0 . 01% pepsin , 0 . 01 N HCl ) for 5 min at 37°C and RNase A incubation ( 100 μg/ml ) in 2x SSC for 1 h at 37°C . After washing in PBS and refixation ( 3% formaldehyde/PBS , 50mM MgCl2 ) , slides were passed through a dehydration series of 70% , 85% , and 100% ethanol for 5 min each and air dried . For denaturation , slides were incubated in 70% formamide in 2x SSC for 5 min at 73°C and then promptly placed in ice-cold 70% ethanol for 5 min , and dehydrated again as described above . 1 μg of MCVSyn DNA was labeled with Dig-Nick Translation Mix ( Roche , 11745816910 ) according to the manufacturer’s instructions . Labeled DNA was ethanol precipitated in the presence of excess sonicated salmon sperm DNA ( Life Technologies ) . The final product was resuspended in hybridization buffer ( 50% formamide and 10% dextran sulfate in 2x SSC ) to a final concentration of 10 ng/μl and stored at -20°C . 5 μl ( 50ng ) of the probe were heat-denatured for 5 min at 73°C and placed under a coverslip on the appropriate area of the slide . The coverslip was fixed with fixogum ( Marabu , 290110000 ) . Slides were hybridized in a humid chamber overnight at 37°C . After hybridization , slides were washed three times ( 2x SSC , 0 . 2% Tween ) for 2 min each , twice at 20°C and in between at 70°C , followed by blocking with 4% BSA/PBS for 30 min at 37°C and incubation with sheep-anti-Digoxigenin-FITC-antibody ( Roche , 11207741910 ) , diluted 1:50 in 4% BSA/PBS with 0 . 2% Tween , for 2hr at 37°C in the dark . Slides were washed with PBS/0 . 2% Tween three times for 10 min each at 20°C in the dark , counterstained and mounted with vectashield mounting Medium with DAPI ( Vector , H-1200 ) . Images were acquired with a confocal laser-scanning microscope ( Nikon C2+ ) . 3’-RACE analysis was performed according to the protocol of Scotto-Lavino et al . [93] with minor modifications . In brief , 5 μg of total RNA of MCVSyn transfected PFSK-1 cells 4d post transfection were subjected to cDNA synthesis using Superscript III ( Invitrogen ) and an anchored oligo-dT primer ( QT ) . The input RNA was digested by addition of 1 . 5 U RNAse H ( NEB ) and incubation at 37°C for 20 min . For the first round of amplification , a gene specific primer ( LTo/VP1o ) and a primer specific for the sequence of the QT primer were used . A second round of amplification with nested gene specific reverse primers ( LTi/VP1i ) and the forward primer Qi was used to increase specificity and add restriction sites to the ends of the PCR product . After digestion with the respective restriction enzymes ( Fast Digest , Thermo scientific ) , RACE products were cloned into pCR2 . 1 plasmid . After transfection into bacteria , individual clones were subjected to Sanger sequencing . 5’ RACE analysis of MCVSyn transfected PFSK-1 cells was performed with the GeneRacer Kit protocol ( Invitrogen ) according to the manufacturer’s instructions . Briefly , 5 μg of total RNA were dephosphorylated , decapped and then ligated to a 5’ RACE RNA adapter . cDNA was synthesized with gene specific primers ( early/late region rev , S1 Table ) using Superscript III according to the manufacturer’s instructions . After touchdown PCR with gene specific primers , nested PCR was performed by which restriction sites were added at both ends of the amplification products ( primers: early/late region rev nested ) . All RACE PCR amplifications were performed with Pfu Ultra II ( Agilent technologies ) according to the manufacturer’s protocol . 5’ RACE products were analyzed by HTS after library preparation with the NEBNext Ultra DNA Library prep Kit for Illumina . Reads were mapped to the MCVSyn genome with TopHat2 [94] . For sequencing of small RNA moieties , RNA from MCVSyn transfected cells and MCC cell lines was subjected to library preparation using the TruSeq Small RNA Sample Preparation Kit ( Illumina ) or the NEBNext Small RNA Library Prep Set for Illumina . Small RNA libraries were sequenced on the Illumina HiSeq platform . After adapter trimming , mapping of reads to the MCVSyn genome and quantification of mature miRNA deposited in the miRNA registry ( miRBase ) release 21 [95] were performed using CLC Genomics Workbench v7 . 5 . 1 ( Quiagen ) , allowing an offset of 5 nucleotides of mature miRNAs along the precursor to ensure detection of isomiRs . Library preparation for strand specific RNA sequencing was carried out using the NEXTflex Directional RNA-Seq Kit ( Bioo Scientific ) according to the manufacturer’s instructions . Libraries were sequenced on the Illumina HiSeq 2500 platform . To allow detection of splice events that extend over the origin , reads were mapped to two concatenated copies of the MCVSyn or MCVSyn-hpko genomes using TopHat2 v 2 . 0 . 13 [94] . The positions of mapped reads and junctions were subsequently collapsed back on unit-length genomes . From the resulting SAM files , we counted the number of unspliced reads that extended over splice sites of junctions detected by TopHat to determine splice site efficacy and frequency of individual junctions . To estimate transcript abundance , for each combination of splice junctions that mapped within either the major early or late transcription cassettes we calculated a relative strand-specific combinatorial frequency value by multiplying observed frequency values for individual donor sites . The relative ratio of late to early transcripts was subsequently estimated by calculation of normalized RPKM ( reads per kilobase per million mapped reads ) for each of the transcripts . Four days after transfection with re-ligated MCVSyn , cell culture supernatants were collected and sterile filtered . 1 ml of supernatant was supplemented with 10x DNaseI reaction buffer and DNA was digested with 10 μl DNase I ( amplification grade , Invitrogen ) for 1 h at 25°C . DNase I was heat inactivated for 10 min at 65°C in the presence of 2 . 5 mM EDTA . Proteins were degraded by addition of 5 μl Proteinase K and 1% SDS at 50°C for 16h . DNA was retrieved by phenol-chloroform extraction and precipitation . For re-infection assays , PFSK-1 cells were transfected with MCVSyn or MCVSyn-hpko . 8 days post transfection; cells were lysed by three freeze-thaw-cycles . Cell debris was removed by centrifugation and lysates were passed through a 0 . 22 μm filter . Lysates prepared from a 10 cm dish were used to inoculate one 6-well of freshly seeded PFSK-1 cells . 24h post infection , medium was changed and cells were incubated for additional 3–7 days as indicated prior to DNA isolation . PFSK-1 cells in 10 cm dishes were transfected with the indicated amounts of a pCDNA3 . 1 LT-antigen expression plasmid added up to 10 μg of total plasmid DNA with an empty pCDNA3 . 1 plasmid . At 24h post transfection , cells were seeded in 96-well plates and grown for another 24h . For measurement of proliferation , cells were incubated with 10 μl MTT reagent ( Chemicon ) per well for 4h . Afterwards , cell culture medium was removed and the formazan crystals were resuspended in 200 μL DMSO . Absorbance was measured at 540 nm with a reference wavelength of 690 nm . All MCPyV sequences and genome coordinates in this study refer to the MCVSyn genome , which is 100% identical to the prototypical MCPyV field strain R17b ( genbank accession numbers JN707599 and NC_010277 , respectively ) . The accession numbers for Gorilla gorilla polyomavirus 1 ( GggPyV1 ) and Pan troglodytes verus polyomavirus 2 ( PtvPyV2a ) sequences as shown in S8 Fig are HQ385752 . 1 and HQ385748 . 1 , respectively . Primary read and mapping data of all small RNA-seq , RACE-seq , ChIP-seq and mRNA-seq experiments performed in this study are publicly available at the European Nucleotide Archive ( ENA , http://www . ebi . ac . uk/ena ) under accession numbers PRJEB9667 ( small RNA-seq data ) , PRJEB9666 ( RACE-seq data ) , PRJEB9670 ( ChIP-seq data ) and PRJEB9669 ( mRNA-seq data ) .
|
MCPyV is a recently discovered human polyomavirus that is likely to cause the majority of cases of Merkel cell carcinoma ( MCC ) , a rare but highly aggressive skin cancer . While much research has been focused on understanding transforming functions of MCPyV gene products , owing to the lack of fully permissive replication systems , the natural lifecycle of the virus is poorly understood . Using high-throughput analyses , here we have interrogated a semi-permissive replication system to study the viral transcription program and elucidate the functions of the viral microRNA ( miRNA ) mcv-miR-M1 . We find that , similar to other polyomavirus miRNAs , mcv-miR-M1 has the ability to negatively regulate expression of viral gene products required for viral DNA replication . Unexpectedly , however , we also observe that mcv-miR-M1 augments long-term episomal persistence of MCPyV genomes . Given that MCPyV establishes persistent infections in the majority of healthy human adults , our observations shed new light on the mechanisms that may be employed by this tumor virus to mount a lifelong chronic infection of its host .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
A Comprehensive Analysis of Replicating Merkel Cell Polyomavirus Genomes Delineates the Viral Transcription Program and Suggests a Role for mcv-miR-M1 in Episomal Persistence
|
Urinary Schistosomiasis infection , a common cause of morbidity especially among children in less developed countries , is measured by the number of eggs per urine . Typically a large proportion of individuals are non-egg excretors , leading to a large number of zeros . Control strategies require better understanding of its epidemiology , hence appropriate methods to model infection prevalence and intensity are crucial , particularly if such methods add value to targeted implementation of interventions . We consider data that were collected in a cluster randomized study in 2004 in Chikhwawa district , Malawi , where eighteen ( 18 ) villages were selected and randomised to intervention and control arms . We developed a two-part model , with one part for analysis of infection prevalence and the other to model infection intensity . In both parts of the model we adjusted for age , sex , education level , treatment arm , occupation , and poly-parasitism . We also assessed for spatial correlation in the model residual using variogram analysis and mapped the spatial variation in risk . The model was fitted using maximum likelihood estimation . The study had a total of 1642 participants with mean age of 32 . 4 ( Standard deviation: 22 . 8 ) , of which 55 . 4 % were female . Schistosomiasis prevalence was 14 . 2 % , with a large proportion of individuals ( 85 . 8 % ) being non-egg excretors , hence zero-inflated data . Our findings showed that S . haematobium was highly localized even after adjusting for risk factors . Prevalence of infection was low in males as compared to females across all the age ranges . S . haematobium infection increased with presence of co-infection with other parasite infection . Infection intensity was highly associated with age; with highest intensity in school-aged children ( 6 to 15 years ) . Fishing and working in gardens along the Shire River were potential risk factors for S . haematobium infection intensity . Intervention reduced both infection intensity and prevalence in the intervention arm as compared to control arm . Farmers had high infection intensity as compared to non farmers , despite the fact that being a farmer did not show any significant association with probability of infection . These results evidently indicate that infection prevalence and intensity are associated with risk factors differently , suggesting a non-singular epidemiological setting . The dominance of agricultural , socio-economic and demographic factors in determining S . haematobium infection and intensity suggest that disease transmission and control strategies should continue centring on improving socio-economic status , environmental modifications to control S . haematobium intermediate host snails and mass drug administration , which may be more promising approaches to disease control in high intensity and prevalence settings .
According to [1] , Schistosomiasis infections affect an estimated 779 million people , with consequences in health nutritional and educational development of infected individuals [2] . The disease causes an annual loss of 4 . 5 million disability-adjusted-lifeyears ( DALYs ) [3] . In SSA alone , 207 million individuals are estimated to be infected with Schistosomiasis: S . haematobium and S . mansoni [1] . S . haematobium is reported to be endemic in 53 countries in the Middle east and most of the African continent including islands of Madagascar and Mauritius [4] , whereas S . mansoni is mostly endemic in sub-Saharan Africa [4] . Schistosomiasis can be effectively treated with single dose oral therapies of praziquantel that are safe , inexpensive and required at periodic intervals [5] . Treatment is typically implemented through mass chemotherapy whereby the entire at-risk population is treated , as part of either school or community- based campaigns , referred to as mass drug administration ( MDA ) . The transmission intensity of Schistosomiasis is a function of parasitic worm load within a group of individuals , which can indirectly be quantified by the number of eggs that are excreted . Host heterogeneities in exposure and susceptibility to infection may lead to an aggregated distribution of worm burden across individuals [6] . For this reason , a few individuals would harbour large numbers of worms , whilst the majority of individuals are uninfected or only carry a low worm burden [6] . In addition , widely used diagnostic approaches for Schistosomiasis like the Kato-Katz technique for S . mansoni diagnosis fail to detect some infected individuals , particularly when only a single stool sample is examined and infection intensities are light [7] . Due to these two issues , often a large proportion of individuals are considered as “zero egg excretor” [6] . The standard Poisson distribution , which assumes equal mean and variance , commonly employed to model such count data , is inappropriate to fit observed egg counts since the variance of the counts is much larger than their mean , a case known as over-dispersion [8] . The use of negative binomial ( NB ) distribution has been proposed to model the extra-Poisson variation [9] , and applications of NB in analysing helminth egg counts are many [8] , [10] . Although NB models may be ideal for over-dispersion , they may not be suitable when data is zero-inflated . Other distributions like the hurdle models or zero inflated ( ZI ) or zero augmented models that may be more appropriate for modeling data with such excess zeros are reported [8] . These models can have more than one mode , including a mode at zero . ZI models attempt to account for excess zeros , i . e . , zero inflation arises when one mechanism generates only zeros and the other process generates both zero and nonzero counts hence they can be expressed as a two-component mixture model where one component has a degenerate distribution at zero and the other is a count model [11] . ZI models estimate two equations , one for the count model and one for the excess zeros . ZI models assume that a proportion of individuals have no chance to be infected , as they are not exposed . In other words , there is a process which determines whether an individual is likely to be infected at all and a second process determining the number of excreted eggs among those who are at risk of infection . Zero inflated Poisson ( ZIP ) models assume that the number of excreted eggs follows a Poisson distribution . Zero-inflated negative binomial ( ZINB ) models assume that the number of worms among those who are at risk of infection has a negative binomial distribution [6] . ZI count data are common in a number of applications . Examples of data with too many zeros from various disciplines include agriculture , econometrics , patent applications , species abundance , medicine , and use of recreational facilities [8] . The zero-inflated Poisson ( ZIP ) regression models with an application to defects in manufacturing is described in [12] , while zero-inflated binomial ( ZIB ) regression model with random effects into ZIP and ZIB models are defined in [13] . The idea for a hurdle model , a modified count model in which the two processes generating the zeros and the positives are not constrained to be the same , was developed in [14] . The two processes are modeled using a mixture of two models ( i . e , two part or a hurdle model ) . The first part is a binary outcome model , and the second part is a truncated count model . Such a partition permits the interpretation that positive observations arise from crossing the zero hurdle or the zero threshold . The first part models the probability that the threshold is crossed , in our case thatan infection occurred . In principle , the threshold need not be at zero; it could be any value , and it need not be treated as known . The zero value has special appeal because in many situations it partitions the population into subpopulations in a meaningful way , one on infection status and the other for those infected it captures intensity . In contrast to the zero-inflated model , the zero and non-zero counts are separated in the hurdle model [15] which makes them very useful in inferential studies . Hurdle models are sometimes referred to as zero-altered models [16] . Zero-altered Poisson and negative binomial models are thus referred to , respectively , as ZAP and ZANB . They have also been termed overlapping models [17] . The application of hurdle or two part models in epidemiology has not been common so far . Use of ZI models have been reported . One such an application was in Cote d'Ivore , in which a ZINB model within a model-based geostatistics ( MBG ) framework for S . mansoni infection was applied [6] . This study showed that geostatistical ZI models produce more accurate maps of helminth infection intensity than the spatial negative binomial counterparts . However , to our knowledge , no hurdle or two part model has been applied in Schistosomiasis or geohelminth epidemiology . This paper demonstrates the applications of hurdle models to helminth epidemiology ( S . haematobium ) and encourage its wider application in helminth disease control programmes . Its advantage is that it allows joint modeling of infection status and intensity . Although , a multinomial model maybe used [18]–[20] , its limitation is that it involves stratifying egg counts , leading to a loss of information , whereas the negative binomial hurdle model approach makes full use of intensity data on a continuous scale , therefore , ideal to model latent infection intensity . In addition , hurdle models are robust when over-dispersion is present . In [8] , it was concluded that the ZIP models were inadequate for the data as there was still evidence of over-dispersion . Moreover , the negative binomial hurdle model , which allows for over-dispersion and accommodates the presence of excess zeros through a two-part model has a natural epidemiological interpretation within the case study considered here .
The data which motivated this work were collected in 2004 in Chikhwawa district , in the Lower Shire Valley-southern Malawi . This is a rural area whose population is mainly engaged in subsistence farming . This area lies between 100 and 300 m above sea level . The rainy season extends from December to March . Temperatures can rise up to in months preceding rainy season . Malaria is known to be holoendemic [21] . Data were collected in eighteen villages , purposively selected from the control and intervention arms of a cluster randomized study design . There was only one round of treatment following community based and house to house approaches for mass drug administration ( MDA ) . Over 90 percent of the eligible population were treated . All infected participants in non-intervention arm received appropriate treatment . After the follow-up assessment , both arms had mass treatment . In the study , polyparasitism was considered basing on the number of species an individual was hosting . The focus was on Hookworm , S . mansoni , S . haematobium and Ascaris . Polyparasitism is the epidemiology of multiple species parasite infections . Ten percent of the households were randomly selected from the villages for baseline survey using random number tables [22] . Subjects for geo-helminth survey were selected using a two stage-design . Briefly , at first stage , villages were selected , then at second stage , sample of households was listed and chosen . In the selected households all members aged one year and above were invited to participate . Consenting individuals had their demographic details completed and were given full body clinical examinations ( except genitals for females ) for chronic manifestations of human helminths . In addition they had anthropometric measurements taken and were asked to provide a single fresh stool and urine sample . All individuals ( aged>1 year ) were requested to provide a finger prick blood sample [22] . Further details are provided in [22] . Fresh stool samples were transported in a cooler box to the laboratory and processed within four hours of collection . A single Kato-Katz thick smear was prepared from each sample and immediately examined under a light microscope for parasite eggs ( within 15–20 minutes ) . Standardized and quality controlled procedures were followed . Briefly 41 . 7 mg of sieved stool was placed on a microscope slide through a punched plastic template . Ova for each parasite observed were counted and expressed as eggs per gram ( epg ) of stool . Five percent of the slides were randomly selected for re-examination for quality control purposes [22] . Urine samples were processed on the day of collection . A measured volume ( maximum 10 ml ) was centrifuged at 300 rpm for five minutes . The sediment was then examined under a light microscope . The eggs seen were counted and the intensity of infection per 10 ml of urine accordingly determined . All those infected were treated with praziquantel at 40 mg/kg [22] . The study that collected data from Chikhwawa , Malawi received ethical clearance from Malawi's College of Medicine Research Ethics Committee ( COMREC ) [22] . Individual informed consent was orally obtained from each participant or ( if they were aged<16 ) from one of their parents or a legal guardian . COMREC approved oral informed consent because the study was determined to be of minimal risk . The consent process was a four stage process . First stage , oral informed consent was obtained at the traditional authority ( TA ) level . Second stage , at village head level and third stage at the household level from the head of the household and fourth at individual level from each individual in the household ( if applicable ) else from parent/guardian if an individual was aged<16 . Registers were kept for documentation whereby , for each individual in the selected household , a column was kept to indicate whether an individual had orally consented to participate in the study or not . Various statistical models have been developed to model helminths disease burden as reviewed in the introduction . For purposes of this paper , we assumed a negative binomial logit hurdle ( NBLH ) model for joint analysis of infection prevalence and intensity of Schistosomiasis hematobium in Malawi . Following on [11] , a NBLH model can be written as: ( 1 ) ( 2 ) where are observed counts taking values for each individual . The probability of infection is , such that indicates there are no zero counts and the model reduces to a truncated Negative Binomial distribution ( TNegBinom ) ; while means there are no infections . The observed counts are modelled by assuming two processes: ( 3 ) The first is assumed to model the infection prevalence ( first hurdle ) and the other the intensity of infection ( second hurdle ) . The first hurdle assumes a binary outcome defining whether an individual is infected or not . This is modeled as a logit regression for a given set of risk factors . After determining infection status we are interested in analyzing the number of eggs - as a measure of intensity of infection , which is defined by the second hurdle . We model the second hurdle as a negative binomial regression model for a given set of risk factors . The NB model is suited for count data with over-dispersion . In many cases , the same risk factors are used in the logit and count regression models , i . e . . The two regression models , incorporating the risk factors , are given by: ( 4 ) ( 5 ) The model parameters and are estimated using maximum likelihood estimation in which the likelihoods ( or log-likelihoods ) are maximized separately . The covariates included in the model are given in Table 1 . Age and polyparasitsm were fitted as continuous variables , while sex , education levels , village type , fishing , gardening and occpuation were entered in the model as categorical variables , with the first category of each variable selected as the reference group . For both parts of the model we used the same set of covariates . We also fitted a number of count models , with the Poisson as the null model , for comparison and evaluated the number of zeros each model correctly predicts . We also compared model fit using AIC and zero capturing . A difference of 10 indicates the model with the smallest AIC is superior to others . Furthermore , deviance residuals were assessed for spatial correlation using variogram and were subsequently mapped using kriging to depict spatial variation in risk . Statistical model fitting was carried out using Political Science Computational Laboratory ( PSCL ) package [23] in R statistical software ( The R Foundation for Statistical Computing , Version 2 . 14 . 0 ) . Variogram analysis and kriging were implemented in geoR [24] .
Table 1 gives summary statistics for study participants . The study had 1642 participants of which 55 . 4 % were female . The mean age ( years ) of 32 . 4 ( standard deviation: 22 . 8 ) . Of these , 324 had hookworm representing 19 . 7 % of sample population , 71 of these had S . mansoni representing 4 . 3 % and 233 had S . haematobium representing a prevalence of 14 . 2 % . Figure 1 shows that a large proportion of individuals i . e . 85 . 8 % were “zero egg excretors” hence the data were inflated with zeros . The likelihood ratio test for overdispersion between Poisson and negative binomial at = 0 . 05 showed a critical value test statistic = 2 . 7 with a test statistic = 10606 . 5 , p-value<0 . 001 . Indeed , there was overwhelming evidence of overdispersion . This was confirmed by the presence of excess zeros ( Figure 1 ) . Using the AIC and zero capturing , the predicted counts using the NBLH indicate a closer fit with the observed values . In Table 2 , AIC results show that the NBLH offers a better fit compared to using Poisson Logit Hurdle ( PLH ) or a negative binomial ( AIC = 3 , 482 for NBLH; AIC = 6 , 854 for PLH and AIC = 3 , 576 for NB respectively ) . The AIC further showed a difference of 10 , 700 for the NBLH compared to the Poisson and a difference of 19 comparing NBLH with ZINB , thus NBLH is superior among all competing models . With regards to zero capturing , the Poisson model was again not appropriate as it could only capture 515 of the zeros whereas the NB-Zero adjusted based models were much better in capturing the zero counts . The NBLH model captured 971 zeros which were equal to the observed ( Table 3 ) . Since NB logit hurdle model offered the best fit to zero inflated helminth data in terms of the AIC ( minimum value for all the models fitted ) as well as true zero count capturing , it therefore became a natural choice for fitting a final model to model helminth infection intensity and determination of factors that foster infections . Table 4 provides estimates for the fixed effects . The probability of infection was found to be associated with age ( Odds Ratio [OR] = 0 . 97 , 95 % Confidence Interval [CI]: 0 . 96–0 . 99 ) , the risk of infection was decreasing with age . This assumed a linear relationship with age; 6 years being the baseline age . The risk of infection was low in males than in females ( OR = 0 . 61 , 95 % CI: 0 . 41–0 . 89 ) . The association between risk of infection with education at both primary level ( OR = 1 . 18 , 95 % CI: 0 . 81–1 . 71 ) and secondary level ( OR = 1 . 37 , 95 % CI: 0 . 41–4 . 60 ) relative to those with no education was not significant ( p-value = 0 . 62 ) . Infection probability was found to be associated with village type; whether one was in the intervention area or control area ( OR = 0 . 38 , 95 % CI: 0 . 26–0 . 54 , p-value<0 . 001 ) . Those in the intervention area were at a reduced chance of infection relative to those in control area . We observed a negative association between infection probability and fishing ( OR = 0 . 73 , 95 % CI: 0 . 44–1 . 20 ) though not significant; contrary to the expectation . Working in the garden was observed not to be significant albeit it was positive ( OR = 1 . 34 , 95 % CI: 0 . 90–1 . 99 ) . Again , occupation ( farmer/other ) showed a negative association with infection probability though with marginal significance ( OR = 0 . 61 , 95 % CI: 0 . 35–1 . 06 ) with a p-value = 0 . 17 . We also noted that chances of infection were increasing with number of parasite species an individual was hosting ( Table 4 ) ( OR = 7 . 30 , 95 % CI:5 . 56–9 . 59 ) . From Table 4 , it was observed that infection intensity reduced with an increase in age ( Relative Risk [RR] = 0 . 96 , 95 % CI: 0 . 95–0 . 98 ) . Similar to infection prevalence , a linear relationship was assumed between infection intensity and age . There was no difference of infection intensity between males and females ( RR = 1 . 03 , 95 % CI: 0 . 72–1 . 47 ) . Primary school children showed a high infection intensity relative to those that are in pre-school level ( RR = 1 . 54 , 95 % CI: 1 . 08–2 . 19 ) whereas those in secondary level showed a reduced infection intensity ( RR = 0 . 34 , 95 % CI: 0 . 11–1 . 06 ) though not significant . There was a reduced risk for those in intervention area relative to those in the control area , though , not significant ( RR = 0 . 81 , 95 % CI: 0 . 58–1 . 13 ) . A positive association was also observed between those who did fishing in Shire river relative to those who did not fish ( Table 4 ) . We observed an increased infection intensity in those working in the gardens relative to those who did not ( RR = 1 . 21 , 95 % CI: 0 . 82–1 . 81 ) , albeit not significant and also increased infection intensity for farmers compared to non-farmers ( RR = 1 . 83 , 95 % CI: 1 . 16–2 . 91 ) . Estimating the continuous surface using variogram analysis and kriging , spatial patterns in the residuals were observed and subsequently mapped . There was some degree of spatial dependence in residuals distribution across the study area , as evidenced by the spherical model ( Figure 2 ) . The magnitude of spatial correlation decreased with separation distance until at distance of 10 km . The predicted spatial surface , in Figure 3 , showed a relatively increased risk of infection in the northern part of the study area compared to other areas . Low risk areas were in the southern parts , more especially in the south-eastern part of the study region ( Figure 3 ) .
The current study found a prevalence of 14 . 2 % for S . haematobium in Chikhwawa district . This prevalence was well below national estimates , which a previous study in Malawi indicated to be between 40 and 50 % [25] . The finding serves to highlight the fact that Schistosomiasis infections are highly localised and that nationwide surveys tend to overlook the focus of heterogeneity of infection . Indeed , in a study conducted in the northern lakeshore area [26] , school children from four schools screened for Schistosomiasis reported a wide range of prevalence: 5 %–57 % of S . haematobium infection . A national survey , representative of all school children in the country , and undertaken just before the rainy season , showed far lower levels of 7 % for S . haematobium [25] . We used robust , contemporary statistical methods in a two part application to analyse risk factors for S . haematobium infection intensity and prevalence . This resulted in estimates of parasitic infection prevalence and intensity that could be used in control programme planning by channeling resources to areas with a known high disease burden . In this study we have looked at the intensity and prevalence of S . haematobium in relation to factors such as age , sex , education level , village type , fishing in Shire river , working in gardens , occupation and polyparasitism . Polyparasitism is the epidemiology of multiple species parasite infections . In the study , polyparasitism was based on the number of species an individual was hosting . The focus was on Hookworm , S . mansoni , S . haematobium and ascaris . The study confirms the critical importance of ascertaining the infection intensity . We found that S . haematobium infection intensity reduced with age , this confirms what previous studies found . In common intestinal helminths such as Ascaris lumbricoides ( large roundworms ) and Trichuris trichiura ( whipworm ) and also Schistosomiasis , children are more heavily affected and infected than adults [27] . Several other studies have reported that school-aged children show high infection intensity and prevalence [25] , [28] , [29] . Fishing in Shire river and working in gardens along the river were potential risk factors for exposure to schistosomes and subsequent infection because transmission requires contact with the aquatic habitat of intermediate host snails [30] . This is in line with results from a study that was conducted in western Africa [20] , that contact with water bodies that are a habitat for intermediate host snails is one of the main risk factors . Results showed low probability of infection for males compared to females . This could be explained by a number of factors including that Malawi being an agriculture based economy , and that mainly agricultural activities are carried out by females , hence they are more exposed to risk factors such as working in gardens and farming . Schistosomiasis is water dependent disease and the incidence is usually more amongst people who constantly get into contact with the schistosome infected waters through activities such as farming , fishing , swimming and washing [30] . Results from the study showed that individuals who had received chemotherapy cure for helminth showed reduced risk of infection as well as infection intensity as compared to those in the control area . Studies have shown that MDA significantly reduces Schistosomiasis infection [31] , [32] . Evidence has shown that , following chemotherapeutic cure of S . mansoni or S . haematobium infection , older individuals display a resistance to re-infection in comparison to younger children [33] . Therefore there is need to channel integrated control and interventions for helminths to areas with diseases burden in order to reduce and/or eradicate the infections - more especially towards school age children . Several studies have shown that having one infection , is a risk factor for having other infections [34] . It is conceivable that the first parasite that establishes an infection may modulate the immune response in such a way that it makes it easier for the next [22] . Worthy noting were differences that existed in associations between infection probability and infection intensity . For gender , males had a reduced risk of infection as compared to females ( negative association ) but high infection intensity ( positive association ) . This could possibly be explained by the fact that women were mostly involved in agricultural activities there by being more exposed . Also for those infected , many studies find that men visit public health care facilities much less frequently than do women [35] hence the high intensity . Poly-parasitism was positively associated with infection probability but had a negative association with infection intensity . This could be explained by the fact that having other parasites increases the chance of the body being susceptible to new parasite infections [34] . Again , secondary level of education had a positive association with infection probability but showed a negative association with infection intensity . This finding could be explained by the fact that an increase in education level corresponds to increase in age which comes with increased risk-behaviour of older school children who frequently contact schistosome-infested water for both domestic and livestock purposes relative to younger children [36] hence increased infection prevalence . At the same time , an increase in education may correspond to increased awareness and access to treatment [37] by this group hence reduced infection intensity . Those with the highest level of education , through high school , have showed the lowest mean infection intensity [37] . Being a farmer had a negative association with probability of infection and a positive association with infection intensity . The finding was in line with what was reported in [37]; farmers showed the highest levels of Schistosomiasis infection among occupational groups . Both education and occupation are proxies for the nature and intensity of water contact [37] . Individuals become infected by prolonged contact ( like irrigating farm , bathing , washing or swimming ) with fresh water containing free-swimming Cercariae [30] . We believe that the apparent dominance of agricultural , socio-economic and demographic factors in determining S . haematobium infection risk in the villages carries important implications for disease surveillance and control strategies . Prevalence of S . haematobium was highly associated with age of an individual as well as working in the garden and also number of parasites an individual hosted . Furthermore , S . haematobium infection intensity was associated with gender , education level , garden , occupation and village type ( intervention ) . Cercariae control control through environmental modifications and strategies involving socio-economic status improvement and MDA may be more promising approaches to disease control in this setting . Finally , zero adjusted methods represents a key advance in the epidemiological analysis of helminth disease data inflated with zeros . There are an increasing number of examples in the published literature where two part methods are being used for zero inflated data for helminths disease's control planning and implementation programmes [38] , [39] . Ease of implementation and straightforward interpretation of the components and its direct link with the observed data , makes the negative binomial logit hurdle model definitely a valuable alternative for researchers analysing zero-inflated count data for helminths .
|
Schistosomiasis is one of the great causes of morbidity among school aged children in the tropical region and Sub Saharan Africa in particular . It's mainly transmitted through contact with water infested with intermediate host snail Cercariae . Currently , over 200 million people are estimated to be infected in SSA alone . Here , we used robust and contemporary statistical methods in a two part application to analyse risk factors for S . haematobium infection intensity and prevalence . We found that S . haematobium was more common in younger children as compared to older children , thus making the infection and prevalence age dependent . We also found that mass chemotherapy reduced both infection prevalence and intensity . We found that dominance of agricultural , socio-economic and demographic factors in determining S . haematobium infection risk in the villages carries important implications for disease surveillance and control strategies . Therefore disease transmission and control strategies centered on improving strategies involving socio-economic status , environmental modifications to control S . haematobium intermediate host snails and mass drug administration may be more promising approaches to disease control in high intensity and prevalence settings .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"public",
"health",
"and",
"epidemiology",
"mathematics",
"epidemiology",
"statistics"
] |
2013
|
Analysis of Schistosomiasis haematobium Infection Prevalence and Intensity in Chikhwawa, Malawi: An Application of a Two Part Model
|
Theoretical models of disease dynamics on networks can aid our understanding of how infectious diseases spread through a population . Models that incorporate decision-making mechanisms can furthermore capture how behaviour-driven aspects of transmission such as vaccination choices and the use of non-pharmaceutical interventions ( NPIs ) interact with disease dynamics . However , these two interventions are usually modelled separately . Here , we construct a simulation model of influenza transmission through a contact network , where individuals can choose whether to become vaccinated and/or practice NPIs . These decisions are based on previous experience with the disease , the current state of infection amongst one's contacts , and the personal and social impacts of the choices they make . We find that the interventions interfere with one another: because of negative feedback between intervention uptake and infection prevalence , it is difficult to simultaneously increase uptake of all interventions by changing utilities or perceived risks . However , on account of vaccine efficacy being higher than NPI efficacy , measures to expand NPI practice have only a small net impact on influenza incidence due to strongly mitigating feedback from vaccinating behaviour , whereas expanding vaccine uptake causes a significant net reduction in influenza incidence , despite the reduction of NPI practice in response . As a result , measures that support expansion of only vaccination ( such as reducing vaccine cost ) , or measures that simultaneously support vaccination and NPIs ( such as emphasizing harms of influenza infection , or satisfaction from preventing infection in others through both interventions ) can significantly reduce influenza incidence , whereas measures that only support expansion of NPI practice ( such as making hand sanitizers more available ) have little net impact on influenza incidence . ( However , measures that improve NPI efficacy may fare better . ) We conclude that the impact of interference on programs relying on multiple interventions should be more carefully studied , for both influenza and other infectious diseases .
Infectious diseases continue to threaten human health throughout the world [1 , 2] . In order to help alleviate these impacts , researchers have utilized mathematical models to improve our understanding of infectious disease dynamics [3] . In many cases , these models assume that human behaviour does not change over time or respond to disease dynamics in epidemiologically relevant ways . However , individual behaviour often does both influence–and evolve in response to–disease dynamics . For example , when vaccination is not mandatory , the prevalence of an infectious disease can depend on individual decisions of whether or not to vaccinate [4 , 5] . Other behavioural practices that impact the spread of a disease include non-pharmaceutical interventions ( NPIs ) [6–10] . For susceptible individuals , NPIs can include hand washing or general avoidance of infectious individuals . For infectious individuals , these can include reducing contact with susceptible contacts , hand washing , or strict respiratory etiquette . Mathematical models of the behavioral epidemiology of infectious diseases capture interplay between disease dynamics and individual behaviour [11] ( we will refer to these as “disease-behaviour” models hereafter ) . These and similar types of models have focussed on vaccinating decisions where individual decision-making occurs according to a strategic environment or is determined by some other utility-based or rule-based mechanism [12–20] . Using such frameworks , Bauch [13] , Fu et al . [15] , Salathé and Bonhoeffer [20] , and Reluga et al . [19] use models with imitation dynamics to predict potential vaccine uptake in populations . Similarly , Xia and Liu [21] base vaccination decisions not only on minimization of the associated costs , but also the impact that social influence has on each individual . d’Onofrio et al . [22] use an information dependent model where vaccination decisions are based on private and public information gathered about a disease . Further approaches by Vardavas et al . [23] incorporate memory of past disease prevalence , and Wells and Bauch [24] include memory of previous infections to study the effect of these factors on vaccinating behaviour . Other research has explored the impact of increased individual sexual risk behaviour on disease incidence , in response to the introduction of a hypothetical HIV/AIDS vaccine [25 , 26] , or how risk perception , HIV prevalence and sexual behaviour interact with one another in a core group population [27] . Other disease-behaviour models incorporate social distancing and other NPI-related behaviours . For example , Reluga [28] analyzes a differential game in which individuals choose a daily investment in social distancing in order to reduce the risk of infection . Funk et al . [29] allow information of a disease to spread over a network , and individuals protect themselves according to the quality of information they possess . Gross et al . [30] and Shaw and Schwartz [31 , 32] study adaptive networks , where susceptible nodes rewire their connections from infectious to non-infectious nodes at a certain rate . Along the same vein , Zanette and Risau-Gusman [33] allow susceptible nodes to either permanently sever a connection with an infectious node , or rewire to another randomly chosen ( and possibly infectious ) node . Del Valle et al . [8] assume some individuals lower their contact rates once an epidemic is detected , whereas Glass et al . [34] and Kelso et al . [9] use complex contact networks which include families , schools , and workplaces to test differing social distancing methods such as school closures and the effects of staying at home while infectious . Hence , disease-behaviour models studying either vaccinating behaviour or NPI behaviour separately from one another are relatively abundant , but models incorporating both types behaviour are rare , to our knowledge . However , for many infectious diseases ( such as influenza ) both NPIs and vaccines are part of infection control strategies , and both also respond to disease dynamics [4 , 35] . Under these circumstances , it becomes important to study disease-behaviour interactions in populations where both vaccinating behaviour and NPI behaviour respond to disease dynamics . The effectiveness of one type of intervention may interfere with the effectiveness of the other intervention , through the mediator of disease dynamics . The objective of our research is to explore the interplay between individual decision-making ( which is driven by methods from decision field theory [36] ) , regarding vaccines , NPIs , and disease dynamics in the context of influenza transmission and control , and to study the implications for disease mitigation strategies .
We model the vaccination decision process using random walk subjective expected utility ( SEU ) theory , an intermediate stage in the mathematical derivation of decision field theory [36] . This approach allows us to capture the decision-making process of individuals in an uncertain environment . In the case of vaccination , the uncertainty lies in the chance of becoming infected in a given influenza season , depending on whether or not the individual is vaccinated and how effective the vaccine is . For the vaccinating decision , each susceptible individual compares the possible outcomes stemming from the “Yes” branch versus the possible outcomes stemming from the “No” branch ( Fig 1a ) . The difference in these expected utilities , or ‘valence’ ( VY ( t ) −VN ( t ) ) , updates an individual’s preference , P ( t ) , towards choosing one of these actions . The preference state of each individual is updated daily according to the rule: P ( t ) = P ( t - 1 ) + [ V Y ( t ) - V N ( t ) ] . ( 1 ) If P ( t ) reaches a specified threshold , θ , then an individual decides to become vaccinated in that influenza season . Conversely , if P ( t ) reaches −θ , the individual decides not to become vaccinated that season . Intermediate values −θ < P ( t ) < θ can be interpreted as an individual being undecided regarding whether or not to vaccinate . If a choice is made , an individual’s preference state remains constant at P ( t ) = θ or P ( t ) = −θ until the beginning of the next season , when it then resets towards ( 1−s ) Pend where Pend is the preference at the end of the last influenza season , and s is a memory decay factor . Let us now define the social utility parameters associated with the vaccinating decision . The quantity EI < 0 is the negative utility received ( cost incurred ) when an individual gets infected; EV < 0 is the negative utility received ( cost incurred ) when an individual vaccinates; ES > 0 is the positive utility received when an individual takes actions that they perceive will inhibit the spread of infection and therefore saves others from becoming infected; and EH < 0 is the negative utility received ( cost incurred ) when an individual believes they are responsible for harming a neighbour by infecting them [37] . The baseline values for these and other parameters can be found in Table 1 . If an individual chooses to not vaccinate during a season , they may become infected that season . If an individual instead chooses to vaccinate , the vaccine is effective for that season with probability ϵV ( the vaccine efficacy ) and otherwise the individual remains susceptible for the remainder of the season . The wi parameters ( 0 ≤ wi ≤ 1 ) represent the ‘subjective probability weights’ that determine the possible outcomes that are considered on a given day . In the case of vaccination , w1 is the probability an individual perceives of being infected if they do not vaccinate in a particular season . We assume w1 depends on how many of an individual’s neighbours have been infected in the current season , as well as their memory of the cumulative incidence from previous seasons: w 1 = σ M H ( X n ) + ( 1 - σ ) ξ n - 1 , ( 2 ) where σ controls the relative importance of incidence from current versus past seasons , Xn is the cumulative number of neighbours who have become infected in the current influenza season n , MH ( x ) = 1−e−κH x where κH is a proportionality constant controlling the perceived chance of becoming infected in a season , and ξn−1 is an individual’s memory of incidence from past seasons: ξ n = σ M H ( Y n ) + ( 1 - σ ) ξ n - 1 , ( 3 ) where Yn is the cumulative number of an individual’s neighbours , including themselves , that have been infected by the end of season n . In this way , the memory of past influenza incidence declines with time according to σ . Individuals may vaccinate during certain days of the year 280 < t or t < 40 ( Fig 2 ) , where t = 0 is January 1st , and the influenza season from the previous year is considered to end on day t = 285 . We use this constraint because it reflects the typical timing of public influenza vaccination programs in fall and winter in many northern hemisphere countries , such as the United States and Canada . At the end of an influenza season , we set Yn = Xn and then incorporate Yn into the individual’s memory . The second subjective probability weight is w 5 = M C ( N s u s c + ( 1 - ϵ V ) N v a c ) , ( 4 ) where MC ( x ) = 1−e−κC x , Nvac and Nsusc are the number of currently vaccinated and susceptible neighbours , respectively , and κC controls the perceived probability of infection . We interpret w5 as an individual’s perceived probability of infecting one or more neighbours , and the term w5 EH captures the future outcome of an individual potentially infecting his/her neighbours that season after becoming infectious themselves , ultimately leading to a negative utility . To complete the outcomes of this branch , we note that the perceived probability of not becoming infected when choosing to not vaccinate is simply w2 = 1−w1 . This outcome leads to a utility of 0 . On the ‘Yes’ branch , we define w3 = ϵV as the perceived probability that an individual is efficaciously vaccinated , thus w4 = 1 − ϵV . In both cases , an individual knows that they must absorb the vaccine cost , EV . In the case of efficacious vaccinating , a positive utility ES is also gained by protecting their neighbours for the remainder of the influenza season , which serves a similar function to the w5 EH term stemming from the ‘No’ branch where an individual is considering future outcomes . In the case of inefficacious vaccination , an individual assumes that they may still become infected with a probability that increases with past and current disease incidence . This is represented by the term VN ( t ) , the valence of the ‘No’ branch . We model the NPI decision process using sequential SEU theory , a method similar to random walk SEU theory , but excludes the possibility that the preference state may start from a non-neutral initial value [36] . We use sequential SEU theory because we assume individuals make social distancing decisions on a day-to-day basis , dependent only on the current state of infection amongst their respective neighbours , whereas in the case of vaccination , the tendency to vaccinate or not can be carried over from one season to the next . Each infectious individual decides whether or not to practice NPIs to protect their neighbours for the duration of their illness ( Fig 1b ) , and each susceptible individual decides whether to practice NPIs to protect themselves from their infectious neighbours that day ( Fig 1c ) . On the ‘Yes’ branch for the infectious NPI decision ( Fig 1b ) , we have the probability of efficaciously using NPIs , w7 = ϵNPI , or inefficaciously using them , w8 = 1−ϵNPI . If NPIs are efficacious , they receive a positive utility ES for saving susceptible neighbours from infection . However , if NPIs are inefficacious , they receive the valence VN ( t ) of the ‘No’ decision , on the branch associated with w8 , representing that the outcome is the same as if they had never practiced NPIs to begin with . An individual believes that their use of interventions during their illness will be either fully effective or ineffective on all of their neighbours . Regardless of whether NPIs work or not , the infectious individual who practices NPIs pays a cost ( ED ) ( NTot ) for having to utilize NPIs to protect their NTot neighbours , where ED < 0 is the negative utility received ( cost incurred ) per neighbour . For the ‘No’ branch where the individuals decides not to practice NPIs , they may infect a neighbour with probability w5 , receiving a negative utility ( cost incurred ) of EH < 0 due to feeling responsible for spreading the infection . On the other hand , they infect no neighbours with probability w6 = 1−w5 , thereby receiving a utility of zero . NPI decisions are made by susceptible individuals who seek to protect themselves from their infectious neighbours in a similar way ( Fig 1c ) . On the ‘Yes’ branch , an individual believes their NPIs will be efficacious with probability w7 , receiving utility zero . If the NPIs are not efficacious , they receive the valence VN ( t ) from the ‘No’ branch . In either case , they pay a cost ED in order to practice NPIs . On the ‘No’ branch , an individual who does not practice NPIs that day becomes infected with probability w9 = MC ( NInf ) , and receive a negative infection utility EI . With probability 1−w9 , they believe they will not become infected and receive utility zero . We do not apply social utilities EH and ES in the susceptible NPI decision process because we assume as a first-order approximation that their decision focuses on short term outcomes ( NPIs are only effective for the duration of infection , and may have to be repeated several times in the season , whereas a one-time vaccination decision will protect their neighbours from infection for the duration of the season ) . Also , if an individual is vaccinated , their perceived probability of becoming infected that day is reduced by 1−ϵV . This reflects the fact that these individuals will believe themselves to have less chance of becoming infected than those who have not vaccinated . We note that infectious individuals practice NPIs on all of their neighbours , whereas susceptible individuals only practice them on their infectious neighbours . Infectious persons may stay home from work , and their hand washing benefits all susceptible persons with whom they come in contact with . In contrast , susceptible persons can be selective about avoiding infectious persons , or hand-washing after contact with infectious persons . In our model , individuals are not aware of their own or their neighbours’ true susceptibility statuses . That is , they will make their intervention decisions based only on their own acquired knowledge which assumes everyone is susceptible at the beginning of each influenza season . This is in contrast to the true state of the system , which incorporates factors such as waning immunity . Moreover , the data we present on susceptible NPI rates only reports for those who are truly susceptible . The vaccination and NPI decision-making processes are embedded in an agent-based simulation model of influenza transmission through a static contact network . The contact network consists of 10 , 000 nodes through which influenza is transmitted , and was constructed by sampling a subnetwork from a larger contact network derived from census data from Portland , Oregon [38–40] . Previous research has confirmed that the subnetwork is a good approximation to the full network [24] . We assume a Susceptible-Infectious-Recovered-Vaccinated-Susceptible ( SIRVS ) natural history . Individuals move from the susceptible state S to the infectious state I with probability P r ( t , N i n f ) = 1 − ( 1 − β ( t ) ) N i n f per day , where Ninf is the number of infectious neighbours around the susceptible person on day t , and β is the transmission probability which varies seasonally according to β ( t ) = β 0 ( 1 + Δ β c o s ( 2 π t 365 ) ) [41] . If either the susceptible person or the infected person has opted to practice NPIs that day , then NPIs are effective with probability ϵNPI , and that infected neighbour is not included in Ninf for the purposes of computing Pr ( t , Ninf ) . Infectious individuals recover ( move from state I to state R ) in a number of days sampled from a Poisson distribution with mean λ days . Individuals who have been efficaciously vaccinated with probability ϵV are moved to the V state , 14 days after being vaccinated [53] . Both recovered and vaccinated individuals become susceptible again at the beginning of each new season ( day 285 of each year ) with probabilities ρ and ω , respectively . In order to capture seasonal case importation , 5 randomly chosen susceptible individuals are made infectious every 10 days , from day 330 to 360 . Each day , the following sequence of events occurs: ( 1 ) each susceptible individual decides whether or not to practice NPIs on that day; ( 2 ) the following occur in a random order for each randomly chosen individual in the population: ( i ) If an individual is susceptible , they update their vaccination preference , ( ii ) if an individual is susceptible , they may become infected and make an infectious NPI decision , ( iii ) if an individual is infectious , they may recover . To construct a baseline scenario , we calibrated the transmission probability β , amplitude of seasonality Δβ , and case importation rate η to the targets: ( 1 ) a cumulative seasonal incidence of approximately 15% to 20% in the absence of vaccination , and ( 2 ) infection prevalence that peaked , on average , between December and January of each year [24 , 42 , 54 , 55] . These estimates come from North American ( primarily , United States ) populations . The calibrated value of Δβ was constrained on the interval ( 0 . 15 , 0 . 3 ) [41] . We also calibrated the preference state threshold for vaccinating θV , per season memory decay rate s , and proportionality constant for the perceived chance of becoming infected in an influenza season κH to the targets: ( 1 ) vaccine coverage of 30% to 40% per season , with ( 2 ) the majority of vaccinations occurring in October and November . These targets provide disease dynamics very similar to seasonal influenza . The utilities EI and EV were set according to Ref . [24] , based in turn on Ref . [42 , 50 , 55–59] . Finally , the social utilities ED , ES , and EH were calibrated to the targets: ( 1 ) an infectious individual practices NPIs with 87% probability , and ( 2 ) a susceptible individual practices NPIs with a 66% probability [60] . After obtaining this baseline scenario , we conducted three-point estimation Monte Carlo probabilistic sampling using triangular distributions to obtain sets of parameter values that yielded outputs within acceptable ranges . The triangular distributions were defined around the most uncertain baseline parameter values . Very broad ranges were used for the most uncertain parameters , to reduce model fitting issues due to having more parameters than calibration targets ( generally , for each set of calibrated parameter values described above , there was one less target data point than the number of parameters to be calibrated ) ( Table 2 ) . We repeatedly sampled parameter values from these distributions , and ran simulations using the sampled parameter sets . We discarded any parameter sets that yielded outcomes outside of a feasible range for vaccine uptake and NPI practice rates ( Table 3 ) . We accepted a larger range of NPI practice rates than for vaccine uptake , due to the greater uncertainties about the frequency of NPI practice for seasonal influenza [60] . In total , 2250 simulations were tested , providing a target number of 100 parameter sets yielding feasible outcomes . All simulations ran for 50 seasons with an initial population of susceptible individuals having no preference towards vaccinating and no perceived probabilities of becoming infected . For our results , the first 20 seasons of each simulation were discarded to remove transient effects . The data shows the average value per season over all 100 parameter sets , for vaccine coverage , infection incidence , probability of susceptible individuals practicing NPIs given that they encounter one or more infectious individuals on a given day ( “susceptible NPI practice” ) , and probability of infectious individuals practicing NPIs while ill ( “infectious NPI practice” ) .
When vaccination is first introduced to the population , vaccine coverage climbs rapidly and peaks in the first few years after the vaccine becomes available , as members of the population adopt vaccination to avoid infection ( Fig 3 ) . As a result , seasonal influenza incidence decreases , which in turn causes a decrease in the probability that susceptible persons practice NPIs if they have an infected neighbour . This occurs because the reduced infection incidence due to the vaccine reduces the perceived infection risk among susceptible individuals , and thus makes them less willing to practice NPIs . After the initial peak in vaccine coverage and the corresponding dip in NPI practice , the vaccine uptake , infection incidence , and NPI practice rates equilibrate . In contrast to susceptible NPI practice , the infectious NPI practice rate stays almost constant during the whole period , due to the relative stability of the utilities found in the decision branches regarding this decision ( Fig 1b ) . For example , an infectious individual deciding to use NPIs will always look to protect all of their neighbours , and this does not depend strongly on population-level incidence of infection that season . This is in contrast to a susceptible individual’s decision whether to use NPIs , which depends on how many of their neighbours are perceived to be infected . Next , we conducted a univariate sensitivity analysis , evaluating the impact of changes in baseline parameter values corresponding to measures that public health might take in order to improve outcomes . Increasing the utility for saving others from infection ( ES ) causes a significant reduction in infection incidence ( Fig 4a and 4b ) . It also causes a slight decrease in NPI practice by susceptible individuals , but this is outweighed by large increases in both vaccine uptake and NPI practice by infectious persons that are sufficient to cause a net decline in infection incidence . These results illustrate a tradeoff whereby vaccine uptake , NPI practice among infectious individuals , and NPI practice among susceptible individuals cannot be simultaneously increased by changing ES . A reduced incidence due to expanding any one of these interventions will reduce the perceived infection risk and make individuals incrementally more complacent about preventing infection , which in turn reduces the uptake rates for the other interventions . Hence , each intervention tends to interfere with the other . Our focus in the remainder of this paper is to determine the conditions under which the interference between the two intervention types is strongest , which model parameters are most subject to interference , and how to bring about the greatest net reductions in infection incidence , despite interference . How interference plays out over time has already been described ( Fig 3 ) . In contrast to significant reductions in infection incidence caused by increasing ES , increasing the cost for harming others ( EH ) causes only small net reductions in incidence , because a large increase in the NPI practice among infectious persons is strongly offset by a modest decline in vaccine uptake , while the rate of NPI practice by susceptible persons remains relatively constant ( Fig 4c and 4d ) . Similarly , attempting to reduce incidence by decreasing the perceived cost of practicing NPIs ( ED ) also causes only a small net reduction in incidence , since the resulting increases in NPI practice among infectious and susceptible persons are again offset by reductions in vaccine uptake ( Fig 4e and 4f ) . Decreasing the cost of vaccination ( EV ) –for instance by making the vaccine more easily accessible–results in significant reductions in infection incidence , because the significant increase in vaccine uptake is only partly offset by the resulting decline in NPI practice ( Fig 5a and 5b ) . Likewise , increasing the perceived cost of infection ( EI ) causes an increase in both vaccine uptake and susceptible NPI practice , although infectious NPI practice remains relatively unchanged . The effect is a significant net decrease in incidence ( Fig 5c and 5d ) . In summary , increasing the utility for saving others from infection ( ES ) , decreasing the perceived cost of vaccination ( EV ) , or increasing the perceived cost of infection ( EI ) , are more effective in reducing infection incidence than changing perceived harms ( EH ) or perceived cost of NPI ( ED ) , despite interference . In order to better understand how NPIs interfere with vaccine uptake , we compute the difference ΔV in vaccine uptake between the baseline scenario where individuals are free to practice susceptible and infectious NPIs versus a hypothetical scenario where they cannot practice either form of NPI . Similarly , to determine how vaccination interferes with susceptible ( and infectious ) NPI practice , we compute the difference ΔNS in susceptible NPI practice rates ( similarly , ΔNI for infectious NPI practice rates ) between the baseline scenario where individuals are free to choose vaccination versus a hypothetical scenario where vaccination is not available . We also compute the difference in seasonal incidence ΔI between the baseline scenario and all of these hypothetical scenarios . To understand the source of this asymmetry between the two interventions , we vary the NPI efficacy ( ϵNPI ) and the vaccine efficacy ( ϵV ) ( Figs 8 and 9 ) . As the NPI efficacy increases , the proportion of individuals practicing NPIs increases significantly ( Fig 8b ) and the vaccine uptake decreases in response , while the infection incidence also declines ( Fig 8a ) . Similarly , when the NPI efficacy is very high , removing NPIs has a much larger impact on incidence than removing vaccination , the latter of which has almost no effect . But when the NPI efficacy is very low , the situation is reversed , and removing vaccination has a much bigger impact on incidence than removing NPIs ( Fig 8c and 8d ) . These results show that feedback between interventions operates such that , if a less efficacious intervention is removed , the resulting increased uptake of the more efficacious intervention is sufficient to prevent a net increase in incidence . In contrast , if a more efficacious intervention is removed , the resulting increased uptake of the less efficacious intervention is not adequate to prevent an increase in incidence . Similar patterns hold when the vaccine efficacy ( ϵV ) is varied , for similar underlying reasons ( Fig 9 ) . However , a secondary factor working in favour of vaccination is that vaccination–if efficacious–protects individuals throughout the influenza season , whereas NPI practice needs to be efficacious every time there is an infection in a network neighbour , in order for an individual to avoid infection throughout the entire season . The difference in vaccine uptake with and without NPIs is highest for intermediate values of ϵV . In general , this occurs because when intervention efficacy is very low , individuals will not adopt it , regardless of whether there is an alternative or not . Therefore , even if the alternative intervention is removed , individuals will tend not to increase adoption of the inefficacious intervention ( Fig 9c , lowest values of ϵV ) . On the other hand , if an intervention is significantly more effective than the alternative intervention , or not very costly , then it will continue to be used by a large proportion of the population , and will not experience significant interference from the less effective alternative intervention which does not significantly change incidence ( Fig 9c , highest values of ϵV ) . In summary , these results show that , the more efficacious an intervention is , the less its effectiveness will be compromised by the other intervention , but the more it will compromise the effectiveness of other intervention . The central role of intervention efficacy also explains why the highest reductions in incidence occur when utilities that support vaccination only are changed ( e . g . the vaccine cost is reduced , Fig 5a and 5b ) , or when utilities that support both vaccination and NPI practice are changed ( e . g . when the payoff for saving others from infection is increased , Fig 4a and 4b , or the perceived cost of infection is increased , Fig 5c and 5d ) . In contrast , decreasing the perceived cost of social distancing ED has little impact on incidence ( Fig 4e and 4f ) , since this NPI practice is interfered with by the mitigating response of vaccine uptake .
We have constructed a seasonal influenza transmission model that incorporates how behavioural decisions for both individual vaccinating decisions and individual NPI practice ( hand-washing , social distancing ) respond to changes in infection incidence . Our population was distributed across an empirically-based network , and parameter values were based either on literature [24 , 41–52 , 55] , or were calibrated to typical influenza seasonal patterns using a probabilistic sampling approach . These results illustrate how vaccine uptake and NPI practice interfere with one another . If vaccine coverage increases , the resulting change in transmission patterns causes a decrease in the practice of NPIs . This is especially true for susceptible individuals , since susceptible NPI practice is more sensitive to population incidence . Similarly , if NPI practice expands , vaccine coverage will decrease by a roughly similar amount . Although susceptible NPI practice and vaccine coverage have similar impacts on each other’s uptake , the impact on incidence is highly asymmetrical between the two interventions: the effectiveness of NPI practice is strongly mitigated by the response of vaccine uptake , whereas the effectiveness of vaccination is only weakly mitigated by the response of NPI practice . This asymmetry is driven by the differing efficacies of the two types of interventions: the higher the efficacy of the intervention ( ϵV , ϵNPI ) , the less its effectiveness in terms of reducing influenza incidence will be mitigated by the other intervention , but the more it will mitigate the effectiveness of other intervention . Because influenza vaccine efficacy is generally higher than NPI efficacy , these effects have potentially important implications for influenza mitigation strategies . Efforts to boost NPI practice could be strongly counteracted by the resulting declines in vaccine coverage , hence boosting NPI practice could be counter-productive . However , boosting vaccine coverage can still be productive since the resulting response of NPI practice will not as strongly mitigate the effectiveness of expanded vaccine coverage . Because of the role of efficacy , increasing NPI efficacy through fostering better hand-washing techniques or respiratory etiquette might more useful than only increasing NPI uptake rates . As a result of this asymmetry , we observed that increasing the utility for saving others from infection , ES was the most effective way of decreasing incidence because it supports both vaccination and NPI practice . From the standpoint of an advertising campaign , this would mean highlighting the fact that saving friends and family from becoming ill , both through vaccination and through NPIs , would be effective . In contrast , attempting to expand NPI practice without simultaneously encouraging vaccine uptake ( for example through making hand sanitizer stations more widely available ) could be counter-productive since NPI efficacy is not as high as vaccine efficacy . Similarly , increasing the perceived cost of infection , EI was also found to be an effective way to reduce incidence , since both vaccination and NPI practice are thereby supported . The asymmetry also explains why decreasing the cost of vaccination , EV , was observed to be effective . The resulting expansion in vaccine uptake suppresses NPI practice to some extent , but because vaccine efficacy is higher than NPI efficacy , decreasing EV still causes net reductions incidence . Hence , reducing the perceived cost of the vaccine by expanding availability ( through more seasonal influenza vaccine clinics ) or decreasing its price will reduce influenza incidence , despite interference . Our model makes several simplifying assumptions with respect to decision making . Firstly , we group all NPIs into two categories: those utilized by susceptible individuals , and those utilized by infectious individuals . In reality , however , an infectious individual may practice combinations , such as choosing to practice strict respiratory etiquette , but not staying home to isolate themselves . Both forms of NPI likely have differing efficacy , hence our grouping assumption could influence results . Similarly , we used a common cost parameter , ED , for all NPIs , but different forms of NPI would likely impose varying costs . We also allowed our population to have knowledge of both vaccine and NPI efficacies . Additional factors that could impact the decision making processes are misinterpretations of influenza-like illnesses ( ILIs ) as being cases of influenza . Often , individuals may mistake other respiratory illnesses for influenza , artificially inflating perceived infection numbers and impacting perceived vaccine efficacy . Similarly , we assume only a single strain of influenza , when in reality , there are often multiple circulating strains . The model could be improved in future research by adding further heterogeneity such as age structure , making perceived efficacy depend on individual experience with interventions , and introducing a probability of ILI being mistaken for influenza , or vice versa [24] . The contact network could also be modified to include family and work structures , which may in turn influence memories of previous infections and perceived infection risk . For example , individuals may weigh the fact that they have had a family member who was recently infected more so than if a casual contact like a co-worker recently fell ill . Moreover , individuals may be more inclined to practice NPIs around family members than their other contacts . Finally , personal infection history may be considered to be significantly more important than neighbour infection history when evaluating perceived risks , which is not accounted for in the model . Being a highly parametrized model , there are several drawbacks associated with calibrating the model to empirical targets in order to obtain a baseline parameter set . We took parameter values directly from estimates in the available literature whenever possible , but ( especially for NPIs ) there is a dearth of information regarding intervention behaviour and impact for seasonal influenza [44] , necessitating calibration . As a result , we had more calibrated parameters than calibration targets ( see Methods ) , meaning that alternative baseline parameter sets could have matched the data almost as well as the baseline set that we used . Our adoption of very broad sampling intervals for the probabilistic uncertainty sampling partially addresses this since the broad intervals will include those parameter values , however , the resulting frequency distribution of outcomes could still vary depending on what baseline parameter set is used as the baseline for defining the triangular distributions . Future work could explore alternative parameterization methods , such as Latin hypercube sampling , which might help overcome this limitation . In conclusion , interference stemming from feedbacks between interventions and disease dynamics can comprise the realized effectiveness of those interventions for reducing influenza incidence , depending on the clinical efficacy of the interventions in individuals . Health authorities and epidemiologists should further explore the potential for interference between different interventions for the same infectious disease , and formulate infection control strategies accordingly .
|
The spread of infectious diseases can be inhibited by both vaccines and non-pharmaceutical interventions ( NPIs ) such as hand-washing , respiratory etiquette , and social distancing . Theoretical models of disease spread have incorporated how individuals make decisions concerning these interventions in the face of disease risks and intervention costs . However , most previous models have considered these two interventions separately from one another . Here we combine decision-making processes for both interventions in a single model that simulates influenza spread through a network . Individuals choose interventions based on their past and present experiences with influenza , and the personal and social impacts of their choices . Our model indicates that , due to feedback loops between the interventions via their mutual impact on disease levels , efforts to reduce influenza spread by expanding NPI practice are almost completely mitigated by the resulting drop in vaccine coverage , whereas efforts to expand vaccine coverage are only weakly affected by the response of NPI practice . Furthermore , strategies such as making vaccines more available , stressing the harms of being infected , or stressing the social benefits of preventing infection through both interventions will prevent more disease than only expanding NPI practice through making hand sanitizers more widely available , for example .
|
[
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] |
[] |
2015
|
Disease Interventions Can Interfere with One Another through Disease-Behaviour Interactions
|
Candida albicans fungemia in cancer patients is thought to develop from initial gastrointestinal ( GI ) colonization with subsequent translocation into the bloodstream after administration of chemotherapy . It is unclear what components of the innate immune system are necessary for preventing C . albicans dissemination from the GI tract , but we have hypothesized that both neutropenia and GI mucosal damage are critical for allowing widespread invasive C . albicans disease . We investigated these parameters in a mouse model of C . albicans GI colonization that led to systemic spread after administration of immunosuppression and mucosal damage . After depleting resident GI intestinal flora with antibiotic treatment and achieving stable GI colonization levels of C . albicans , it was determined that systemic chemotherapy with cyclophosphamide led to 100% mortality , whereas selective neutrophil depletion , macrophage depletion , lymphopenia or GI mucosal disruption alone resulted in no mortality . Selective neutrophil depletion combined with GI mucosal disruption led to disseminated fungal infection and 100% mortality ensued . GI translocation and dissemination by C . albicans was also dependent on the organism's ability to transform from the yeast to the hyphal form . This mouse model of GI colonization and fungemia is useful for studying factors of innate host immunity needed to prevent invasive C . albicans disease as well as identifying virulence factors that are necessary for fungal GI colonization and dissemination . The model may also prove valuable for evaluating therapies to control C . albicans infections .
Candida albicans is a ubiquitous commensal organism that can cause serious disseminated infections in cancer patients [1 , 2] . Candida spp . are the fourth leading cause of nosocomial bloodstream infections in the United States , with treatment costs estimated to be more than $2–$4 billion annually [3] and with attributable mortality rates estimated to be between 38% to 49% [4] . Among the various invasive fungal infections reported in cancer patients , candidiasis is the most common infection ( 58%–69% ) [5–7] , and over the past decade , the incidence of invasive fungal infections in this population has increased significantly [8] . The presumed mechanism for all invasive C . albicans disease involves initial mucosal surface colonization followed by invasion into the adjacent tissues and organs . In cancer patients , C . albicans usually colonizes the gastrointestinal ( GI ) tract with subsequent translocation into extraintestinal organs ( i . e . , mesenteric lymph nodes , blood stream , liver , and spleen ) in the setting of chemotherapy-induced neutropenia and GI mucosal damage [9] . The three primary mechanisms that promote pathogenic microbial ( bacterial and fungal ) translocation in animal models are: 1 ) disruption of the normal GI microbiologic equilibrium allowing intestinal overgrowth of pathogens , 2 ) increased permeability of the intestinal mucosal barrier , and 3 ) deficiencies in the host immune defenses [10 , 11] . Not surprisingly , common risk factors for developing candidemia in human patients include neutropenia , mucositis , use of broad spectrum antibiotics , and invasive medical procedures [6 , 12] . The majority of murine models of disseminated candidiasis have employed the administration of a chemotherapeutic agent ( e . g . , cyclophosphamide ) followed by the subsequent intravenous injection of C . albicans [13 , 14] or simply the intravenous injection of high inocula of C . albicans [15 , 16] . Therefore , a murine model that first establishes GI colonization followed by translocation and dissemination via disruptions of select components of the innate host defense would afford valuable opportunities for studying the details of C . albicans pathogenesis , as well as delineating the relative roles of the major immune compartments or the actual immune mechanisms responsible for killing and/or controlling translocating C . albicans [10] . To evaluate normal host factors that must be disrupted to allow for GI colonization , fungal dissemination and significant morbidity or mortality due to candidemia , we have developed a reproducible mouse model wherein C . albicans GI colonization is first established and subsequent fungal dissemination is achieved following induction of immunosuppression and disruption of mucosal integrity . Both neutropenia and GI mucosal damage appear to be necessary for fungal dissemination in this murine model . Finally , we evaluated C . albicans mutants that had varying abilities to switch between the yeast and the hyphal growth form of C . albicans and found changes in virulence associated with an inability to switch to the hyphal phase , indicating the utility of this animal model for studying different aspects of the pathogenic process of C . albicans in the setting of GI colonization and dissemination .
Wild-type C . albicans strains SC5314 ( a strain that has been frequently used in various C . albicans murine models [17–20] ) and CAF2–1 consistently colonized the mouse GI tract at comparable levels: SC5314 ( median = 2 . 24 × 107 cfu/g stool , first quartile = 1 . 07 × 107 , third quartile = 3 . 75 × 107 ) and CAF2–1 ( median = 2 . 60 × 107 cfu/g stool , first quartile = 2 . 28 × 107 , third quartile = 3 . 19 × 107; Figure 1 ) . When tested up to 21 d later , there was no significant change in fungal colonization levels of the stool . Mice that were colonized with C . albicans strains SC5314 and CAF2–1 , as well as non-Candida colonized control mice that were only decontaminated with antibiotics , were separated into 17 groups ( 8 mice/group , with the exception of 4 mice/group in Rag−/− mice ) , treated with immunosuppressive regimens ( Table 1 ) , and the effect on dissemination of the different methods of immunosuppression determined ( Table 2 ) . Mice that were only decontaminated with antibiotics all survived after administration of immunosuppressive regimens—with the exception of 1 death in the group given mAb RB6-8C5 plus methotrexate ( MTX ) ( Group 7 ) . All groups that were colonized with C . albicans strains SC5314 and CAF2–1 then given immunosuppression showed levels of GI colonization comparable to those of mice not given immunosuppression ( results shown in Figure 1; data for immunosuppressed mice not shown ) . Mice that were only given antibiotic decontamination and no immunosuppression after colonization with C . albicans ( Group 1 ) also all survived . We analyzed the effects of immunosuppression and/or mucosal damage treatments on translocation of C . albicans from the GI tract . All mice in these studies were colonized with C . albicans SC5314 and had levels of GI colonization comparable to that shown in Figure 1 ( data not shown ) . Mice given no immunosuppression ( Group A ) all survived , and none appeared moribund and none grew C . albicans from the blood , spleen or MLN . Of note , fungi were found at very low levels in the livers of 4 ( out of 20 ) healthy-appearing mice ( 9 , 42 , 208 , and 369 cfu/g ) . The mice given mAb RB6-8C5 treatment only ( Group B ) all survived , and none appeared moribund . Similar to Group A , C . albicans was not found in the blood , spleen or MLN of mice only given the neutropenia-inducing mAb , but fungi were occasionally found in the liver ( 5 out of 20; median 9 . 88 × 102 cfu/g , first quartile = 4 . 09 × 102 , third quartile = 2 . 68 × 103 ) ; these are clearly higher levels than those found in the livers of mice in Group A which did not receive any immunosuppression . In contrast , 3 of the 4 mice from the group given neutropenia-inducing mAb RB6-8C5 plus MTX ( Group C ) died by day 4 post-immunosuppression , and 4 of 4 of these mice had died by day 5 . All of these mice showed significant levels of C . albicans in the liver ( median 8 . 77 × 103 cfu/g , first quartile = 3 . 45 × 103 , third quartile = 2 . 51 × 104 ) . Finally , the mice given Cy only ( Group D ) all had C . albicans in the liver by days 4 and 5: ( median 7 . 71 × 104 cfu/g , first quartile = 3 . 87 × 104 , third quartile = 1 . 13 × 105 ) . In order to assess the role of fungal morphogenesis as a virulence determinant in our murine model , we tested three C . albicans strains for their ability to colonize and disseminate following immunosuppression: strain CAF2–1 is a wild-type organism that was chosen because it , like the additional mutants we tested , only has one copy of the URA3 gene [21]; strain HLC54 ( cph1/cph1 efg1/egf1 exhibits decreased filament formation , hereafter referred as Δefg1/cph1 [22]; and strain BCa-210 ( tup1/tup1 , hereafter referred to as Δtup1 ) exhibits constitutive filamentous growth [23] . As shown in Figure 4 , both CAF2–1 ( median = 2 . 99 × 107 cfu/g , first quartile = 2 . 38 × 107 , third quartile = 3 . 30 × 107 ) and strain Δefg1/cph1 ( median = 5 . 32 × 107 cfu/g , first quartile = 4 . 69 × 107 , third quartile = 1 . 03 × 108 ) were able to colonize the GI mucosa at comparable levels , while the Δtup1 strain colonized at a level 2-logs lower ( median = 7 . 71 × 104 cfu/g , first quartile = 5 . 0 × 104 , third quartile 1 . 66 × 106 , p = 0 . 0003 by Mann Whitney test compared with other two strains ) . This discrepancy is explained by the fact that the maximum concentration achievable of Δtup1 in water was approximately 2-logs lower than the levels for the wild-type and Δefg1/cph1 strains—a difference which was consistent throughout the experiment . Interestingly , strain Δefg1/cph1 caused less mortality compared to wild-type C . albicans ( 8 of 16 mice given Δefg1/cph1 died versus 14 of 16 mice infected with wild-type , p = 0 . 02 , Fisher's exact test ) , whereas in spite of the 2-log lower levels of strain Δtup1 in feces it caused mortality comparable to that of wild-type ( 7 of 8 mice died; Figure 5 ) . In preliminary experiments with wild-type C . albicans strains SC5314 and CAF2–1 , if we administered immunosuppression ( e . g . , mAb RB6-8C5 plus MTX ) before final levels of GI colonization ( between 107 and 108 cfu/g ) had been established , dissemination was not induced , and all mice survived ( data not shown ) . All of these surviving mice continued to be colonized with their respective strains of C . albicans at levels comparable to that achieved before administration of mAb and MTX . Eight additional mice were administered higher concentrations of Δtup1 ( 5 . 5 × 105 cfu/ml ) in the drinking water , and GI colonization levels achieved with this higher dose were comparable to those of the wild-type C . albicans strain ( median = 1 . 48 × 107 cfu/g , first quartile = 5 . 20 × 106 , third quartile = 4 . 08 × 107 ) , indicating that the level of GI colonization can be increased for this strain . To determine if this constitutively filamentous strain could by itself induce mucosal damage sufficient to achieve dissemination , we administered the mice 200 μg of mAb RB6-8C5 only . No enhanced mucosal disruption sufficient to allow fungal dissemination in most mice was achieved by constitutive hyphal expression , as 7 of 8 colonized mice made only neutropenic still survived .
In this study , we attempted to devise a mouse model to study C . albicans pathogenesis and host factors leading to susceptibility to disseminated infection that emulate the pathophysiology that takes place in a human host , wherein receipt of broad-spectrum antibiotics , extensive hospitalization , or administration of immunosuppressive agents makes patients more vulnerable to invasive disease , often associated with intercurrent mucosal disruption owing to surgery , tumor invasion , or chemotherapy [24 , 25] . In neutropenic patients , the role of the gut as a source for disseminated candidiasis has been supported from autopsy studies [26] . Several murine models of GI-derived C . albicans fungemia and sepsis have been reported previously [27–32] . The levels of GI colonization we obtained were substantially higher than those achieved by prior investigators ( 1 to 3 log-fold higher ) [31 , 33] and were maintained at least up to 21 d ( our latest verification date ) . This higher colonization level is most likely due to the addition of penicillin to the drinking water . Penicillin most likely led to reductions of endogenous anaerobic bacterial flora , promoting subsequent intestinal overgrowth by C . albicans [34] . One other model did achieve comparable colonization levels by also administering adjunctive oral antibiotics [32] . The need for reduction of indigenous flora , particularly gram-negative bacteria , is critical given that bacteria are just as likely as the C . albicans to translocate and disseminate when immunosuppression is administered [35] . We were able to show this model recapitulates important aspects of human susceptibility to candidiasis , although there are some limitations to consider . The model was developed using adolescent/young adult 6- to 8-wk-old female C3H/HeN mice , and its applicability to other mouse strains or mice of other ages is not fully known . It would be difficult to use neonatal or infant mice in this model as others have done with acute C . albicans infections [28–30] due to the need for the animals to be able to drink both antibiotic water and fungi in the water and then give the immunosuppressive/mucosal disrupting agents , which would not be feasible with mice prior to weaning . C3H/HeN mice were used because of experience in a previous model of P . aeruginosa GI colonization and neutropenia-induced dissemination [35] , and the fact that this strain has no major underlying immune deficit and is highly susceptible to DSS-induced colitis [36] . While we have not extensively studied other mouse strains , initial studies using Swiss Webster and C57/BL mice showed that comparable levels of GI colonization , mortality , and liver dissemination can be achieved when using Cy as the immunosuppressive/mucosal damaging agent ( AYK and GBP , unpublished data ) . C . albicans administration via drinking water limits the ability to control the infecting dose as might be achieved by administering fungal cells by gavage . However , administering C . albicans in the drinking water for 5 d led to a reproducible and consistent GI colonization level ( Figure 1 ) associated with dissemination following the disruption of specific host defenses . Additionally , we did not confirm the entire spectrum of bacterial decontamination achieved by the antibiotics , as we did not utilize anaerobic culture conditions or media specific for all types of aerobic and anaerobic flora . Therefore , we cannot be absolutely sure that we truly eradicated all indigenous aerobic and anaerobic flora . In any case , whatever residual aerobic or anaerobic flora that remained after antibiotic decontamination does not appear to have any pathogenic significance , as evidenced by the fact that mice given only antibiotic contamination and subsequent immunosuppression all appeared healthy and exhibited 0% mortality . We also chose to measure fungal levels in livers as an indicator of dissemination . In other murine models of intravenous C . albicans infection fungal levels in the kidneys were used to confirm systemic infection was achieved . In preliminary experiments , we found that in mice that were colonized with C . albicans , given subsequent immunosuppression ( cyclophosphamide or RB6-8C5 + MTX ) , and had no other detectable infectious cause of death ( e . g . bacterial dissemination ) that the liver ( presence of C . albicans in 100% of livers from deceased mice ) was a more reliable organ for confirmation of dissemination compared to the kidneys ( presence of C . albicans in 50%–80% of kidneys in deceased mice; AYK and GBP , unpublished data ) . One other model using C . albicans GI colonization and chemotherapy-induced dissemination also noted 100% recovery from the livers but significantly less recovery from the kidneys [32] . Finally , for practical purposes we had to use organs for CFU enumeration from moribund then euthanized mice or mice that died between observations whose carcasses were frozen as close to the time of death as feasible . However , in limited studies we did compare the C . albicans yields from organs resected after freezing and storage with yields from organs resected from freshly euthanized mice and found the differences in CFU were not statistically different ( AYK and GBP , unpublished data ) . Nonetheless , because there could be effects from post-mortem fungal growth or losses upon storage at −20°C one must be cautious in using levels of C . albicans in the organs as a measure of virulence . We also attempted to determine how C . albicans might spread from the GI lumen to internal organs . Given the sporadic and low levels of hepatic dissemination in mice colonized with C . albicans but not receiving any immunosuppression , it is conceivable that C . albicans is able to translocate to the liver via the portal circulation or via the biliary tree in the absence of immunosuppression , but does not cause widespread disease because of a competent immune system that is able to prevent significant dissemination of the fungi . In our study , mice colonized with C . albicans and given only RB6-8C5 mAb also showed sporadic levels of hepatic dissemination , albeit at higher levels than in mice that did not receive immunosuppression . The higher fungal burden is most likely secondary to the lack of neutrophils and thus the subsequent diminished ability of the immunosuppressed animal to clear the C . albicans from the liver . But even in the absence of neutrophils , the fungal burden was not enough to cause frequent morbidity or death . Not surprisingly , when we administered systemic chemotherapeutic agents such as cyclophosphamide ( resulting in both neutropenia and GI mucosal damage ) we were able to reproducibly achieve widespread dissemination with C . albicans . In the final analysis we found both neutropenia and disruption of the integrity of the GI mucosa were needed for fungal dissemination . In several intravenous murine models of C . albicans systemic infection , neutrophil depletion notably increases fungal burden and mortality [37 , 38] . In this murine model , however , neutropenia alone is not sufficient for extra-intestinal dissemination most likely due to the fact that C . albicans is still unable to breach the intact GI epithelium . Although one prior study implicated the importance of CD4+ lymphocytes in protective immunity to systemic C . albicans infection [39] , several studies in SCID mice have shown that a defect in the TH1 CD4+ T-cell response to C . albicans results in mucosal or esophageal candidiasis but not in systemic dissemination . Even in C5 deficient DBA/2 mice [38] and a recent murine HIV model [40] , T cell depletion also does not result in disseminated C . albicans disease . Although the RB6-8C5 mAb at the dose used also induces some depletion of lymphocytes [35] , we found that both C3H/HeN and RAG−/− mice that lack mature lymphocytes were not susceptible to fungal disease following GI colonization and neutropenia alone . Thus , our murine model correlates with these previous studies: lymphopenia alone or in combination with macrophage depletion , neutrophil depletion , or GI mucosal disruption without neutropenia did not result in fungal dissemination or death of the colonized mouse . Similarly , depleting macrophages only did not lead to any susceptibility to systemic fungal infection . Although murine peritoneal macrophages [41 , 42] , pulmonary alveolar macrophages [43] , Kuppfer's cells [44] and human peripheral blood monocytes [45] have the ability to phagocytose and kill C . albicans in vitro , in vivo studies on disseminated candidiasis have generated evidence both supporting [42 , 46–48] and refuting the importance [49–51] of macrophages . Of the two studies utilizing selective macrophage depletion [48 , 50] , only one administered liposomal clodronate followed by intravenous fungal injection to produce a disseminated candidiasis . Spleens from the mice treated with clodronate lost their ability to trap yeast [48] . Furthermore , when macrophage-depleted mice were systemically challenged with C . albicans , not only was clearance of yeast decreased in blood , but kidneys had higher fungal burdens and overall mouse survival was decreased . Although we used the same dose of clodronate , we also utilized FACS analysis to quantify the macrophage depletion ( 37% depletion ) rather than measuring peripheral blood monocyte counts ( 30%–85% depletion ) . Ultimately , our lack of finding a significant role for macrophages in host resistance to C . albicans may be related to the use of different models . In an intravenous model with direct inoculation , this degree of macrophage depletion may be enough of a deficit to lead to more severe disseminated disease . In our murine model , it may be that macrophages do play a role , but more substantial macrophage depletion may be necessary to see this effect . Finally , while DSS has been used for experimental murine models of human inflammatory bowel disease [52 , 53] , the histopathologic damage induced by DSS has some similarities to chemotherapy-induced mucositis in that both result in denudation and ulceration . Interestingly , when we maintained the same degree of neutropenia but increased the GI mucosal damage ( by changing DSS from 2 . 5% to 5% ) , we notably increased the mortality to 100% , and this is most likely secondary to the increased mucosal damage caused by the higher concentration of DSS . These studies further support our hypothesis that it is a combination of neutropenia and GI mucosal damage that are critical for C . albicans dissemination in our murine model . In the host , C . albicans grows as all known morphologic forms ( budding yeast , pseudohyphal filaments , and true hyphae ) , and it has been postulated that the ability to induce hyphal formation is a critical virulence determinant [22] . Mutant strains of C . albicans that are incapable of hyphal formation have been found to be avirulent in murine models of disseminated candidiasis [54–56] . Whether it is simply that hyphae formation leads to an increased ability to invade host epithelial cells , or induces greater cytotoxicity , or whether the advantage of filamentous growth is in providing greater resistance to phagocytosis [22] , the definitive mechanistic explanation linking hyphal formation to virulence is lacking . When testing the morphogenesis mutants , Δefg1/cph1 , which exhibit decreased filament formation [22] , and Δtup1 , which exhibits constitutive filamentous growth [23] , we noted a nearly 2-log lower GI colonization level of the Δtup1 mutant compared to the wild-type strain , CAF2–1 . As noted earlier , this discrepancy in colonization level was most likely due to the difficulty of suspending Δtup1 in drinking water and achieving higher concentrations . All of the mice colonized with strain Δtup1 had colonization levels lower than the 107 to 108 cfu/g achieved with C . albicans strains SC5314 and CAF2–1 , yet 7 of the 8 mice colonized with the Δtup1 strain died following immunosuppression and GI mucosal damage . Since it is difficult to titrate inocula in our murine model , this finding of dissemination and mortality in the setting of significantly lower GI colonization levels may suggest that the Δtup1 mutant is more virulent in this setting . The appropriate quantification of filamentous strains , however , is confounded by the fact that cells may not be properly separated , and thus CFUs are underestimated . In addition , filamentous strains such as Δtup1 may adhere to the GI mucosa more tenaciously , and thus levels in the stool could be an underestimation of actual GI colonization levels . However , increased filamentous growth of the Δtup1 mutant alone is not sufficient to disrupt the GI mucosa and allow for fungal dissemination even in the setting of neutropenia . Our findings with CAF2–1 and Δefg1/cph1were consistent with previous investigators that utilized an intravenous model of C . albicans dissemination [22 , 57 , 58] . However , in a model of hypoxic-induced GI translocation [59] and another study using oral inoculation followed by intraperitoneal ( IP ) injections of dexamethasone for immunosuppression [57] , Δefg1/cph1 was found to be more invasive compared to the wild-type CAF2–1 . Whereas both of these studies used the presence of C . albicans in extraintestinal organs to define invasiveness or virulence , we used death attributable to C . albicans dissemination . Furthermore , we utilized a different means of immunosuppression and attempted to verify that bacterial translocation was not a confounding factor . Therefore , our findings with Δefg1/cph1 support the hypothesis that the ability to form filaments is important for translocation and dissemination with C . albicans . In conclusion , we have developed a murine model of C . albicans GI colonization following anti-microbial agent reductions in the indigenous flora and systemic spread during neutropenia that additionally requires GI mucosal damage . Neutropenia alone is not sufficient to produce disseminated C . albicans disease in this murine model . Being able to control these host factors should allow for a more detailed study of host and fungal factors needed to achieve GI colonization and systemic dissemination . These factors should thus be useful for evaluating pathogenesis as well as therapies to control C . albicans invasive infections .
The strains of C . albicans used are listed in Table 3 . C . albicans strains were grown overnight at 37°C in yeast extract-peptone-dextrose ( YPD ) broth , harvested by centrifugation , washed with PBS , and resuspended in PBS . C . albicans concentration was determined by use of a hemocytometer . Six- to 8-wk-old female C3H/HeN mice ( Harlan , http://www . harlan . com/models/c3h . asp ) were housed as groups of 4 in sterilized cages equipped with filter hoods . In some experiments , 6-wk-old female recombinase activating gene deficient mice ( Rag−/− , http://jaxmice . jax . org/strain/002216 . html ) were used . Mice were supplied with sterile bedding , sterile water and sterile mouse chow and maintained under specific pathogen-free conditions at the ARCM-MCP animal facility at Harvard Medical School in compliance with the Harvard Medical Area Institutional Animal Care and Use Committee guidelines . To deplete the indigenous GI bacterial and fungal flora , mice were fed sterile water with 2 mg streptomycin/ml ( Research Product International , Mt . Prospect , IL ) , 1500 U penicillin G/ml ( Sigma-Aldrich , St . Louis , MO ) , and 0 . 250 mg fluconazole/ml ( Roxanne Laboratories , Columbus , OH ) for 3 d , then switched to the same concentrations of streptomycin and penicillin G in their drinking water for one more day . Stool was collected from individual mice ( 0 . 030–0 . 050 g per stool pellet ) , homogenized in 1 ml 1% protease peptone , and 100 μl of the homogenate was spread and plated on yeast extract-peptone-dextrose ( YPD ) , trypticase soy ( TSA ) , or MacConkey agars to verify reduction of the targeted indigenous GI microbial flora . C . albicans strains ( grown as described above ) were added to sterile water with 2 mg streptomycin/ml , and 1 , 500 U penicillin G/ml at approximately 107 cfu/ml , and then C . albicans was administered via the drinking water to mice for 5 d . Candidal levels were constant in the drinking water and therefore water bottles were not changed during this time . After 5 d of exposure to Candida , stool was again collected , homogenized in 1 ml 1% protease peptone , serially diluted and plated on YPD agar with 0 . 010 mg vancomycin/ml and 0 . 100 mg gentamicin/ml to measure levels of GI colonization by C . albicans . We then induced immunosuppression by treating the mice with the immunosuppressive agents listed in Table 1 . After administration of immunosuppressive agents , mice were fed sterile water with 2 mg streptomycin/ml , 1 , 500 U penicillin G/ml , and 0 . 2 mg gentamicin/ml for the remainder of the experiment and monitored for morbidity for 7 d . Moribund mice were euthanized and along with mice found dead between observation periods , the carcasses were frozen at minus 20°C , later thawed , livers resected , homogenized in 1 ml 1% protease peptone , serially diluted and 100 μl of the homogenate was spread-plated on YPD with 0 . 100 mg gentamicin/ml ( Research Product International , Mt . Prospect , IL ) and 0 . 010 mg vancomycin/ml ( Sigma-Aldrich ) , and additional 100 μl amounts plated onto TSA and MacConkey agar plates . Growth media were incubated at 37°C overnight under aerobic conditions . The presence of a homogeneous population of creamy-white colonies on YPD with gentamicin and vancomycin was used for confirmation of C . albicans systemic dissemination ( Figure 6 ) . The RB6-8C5 monoclonal antibody ( mAb ) specific for the Ly6 antigen highly expressed by polymorphonuclear neutrophils ( PMN ) was produced by growth of hybridoma cells in culture ( Dulbecco's modified Eagle's medium with 10% fetal calf serum [FCS] ) followed by purification of antibody by affinity chromatography , as previously described [35] . A single dose of 200 μg of RB6-8C5 was administered to C3H/HeN mice to produce a severe neutropenia ( absolute neutrophil count < 100/mm3 ) for 5 d [35] . CL2MBP ( clodronate ) was a gift from Roche Diagnostics ( Mannheim , Germany ) . Preparation of liposomes containing CL2MBP was done as described previously [60] . For assessment of macrophage depletion , three uninfected mice per group were inoculated IP with 200 μl of PBS liposomes or with CL2MBP liposomes ( 2 mg ) [61] . At 2 d later , macrophages were quantified in the bone marrow and spleens using FACS analysis . Bone marrow and spleens were harvested from euthanized mice . To prepare single cell suspensions , spleens were sliced into small pieces with a scalpel , and the pieces placed at the ends of autoclaved frosted-glass microscope slides ( Fisher Scientific , Pittsburgh , PA ) , previously immersed in Hank's balanced salt solution ( HBSS; Gibco/Invitrogen , Carlsbad , CA ) with 10% FCS . Splenic fragments were disrupted by opposing the slide ends and applying gentle pressure in round circular movements . The resulting splenic cell suspension was collected in a Petri dish and finally run over a 70 μm nylon cell strainer ( BD Biosciences , San Diego , CA ) . For bone marrow cells , single cell suspensions were prepared by running bone marrow cells resuspended in HBSS with 10% FCS over a 70 μm nylon cell strainer . Cells from the three mice were pooled . Erythrocytes were eliminated using lysing solution ( BD Biosciences , San Diego , CA ) . The remaining hematopoetic cells were washed and resuspended in PBS supplemented with 2% FCS , 1% bovine serum albumin , and 0 . 1% NaN3 ( FACS buffer ) . Cells were blocked with Mouse Fc Block ( BD Biosciences , San Diego , CA ) at 1 μg per 106 cells for 5 minutes at 4°C . Alexa-488-conjugated antibody to the mouse macrophage cell surface glycoprotein F4/80 ( mAb BM8 , Caltag/Invitrogen , Carlsbad , CA ) and/or phycoerythrin-conjugated antibody to mouse CD11b ( BD Biosciences , San Diego , CA ) at 10 μg/ml/test were added to 1 × 106 cells in 100 μl of FACS buffer then incubated at 4°C for 30 min followed by washing with FACS buffer . To determine background levels of fluorescence , non-antigen specific isotype controls ( Alexa-488-conjugated-IgG2a or phycoerythrin-conjugated-IgG2b antibodies ) were used as primary antibodies . After staining , cells were fixed with 1% paraformaldehyde . Fluorescence was detected on a Dako MoFlo High Performance Cell Sorter ( Dako , Fort Collins , CO ) , and data analysis was performed using Summit Software v 4 . 3 ( Dako , Fort Collins , CO ) . Gates were set to include CD11b and F4/80 double positive cells . Livers and spleens from mice inoculated IP with 200 μl of PBS liposomes or CL2MBP liposomes ( 2 mg ) 48 h prior to sacrifice were resected , fixed in Bouin's solution , and paraffin-embedded sections cut and mounted on slides . 5 μm sections of hepatic and splenic tissue were deparaffinized in xylene and rehydrated in a series of reverse ethanol washes ( 100% , 95% , and 80% ethanol sequentially ) . The samples were then blocked by incubation in histology blocking buffer ( PBS containing 1% bovine serum albumin and 2% normal rat serum [Sigma-Aldrich] ) for 15 min at 37°C . Samples were then incubated with 1 μg Alexa-488-conjugated antibody to the macrophage F4/80 antigen/ml for 90 min at 37°C . To detect background fluorescence , an appropriate isotype control antibody ( 1 μg rat IgG2a/ml ) was used . The samples were then washed and sections visualized with the 10× alpha-plan lens on a Zeiss Axioplan 2 microscope , fitted with filter sets for GFP fluorescence , and a cooled CCD Hamamatsu Orca camera . Images were acquired and processed using MetaMorph ( Molecular Devices Corporation , Sunnyvale , CA ) and Adobe Photoshop software [62] . To verify we had induced GI mucosal damage/disruption with dextran sodium sulfate ( DSS ) , C3H/HeN mice ( 6- to 8-wk-old females ) colonized with C . albicans strain SC5314 and maintained on sterile water with streptomycin , penicillin , and gentamicin to prevent bacterial recontamination were divided into four groups and sacrificed for cecal histology after the following specific treatments: Group A was given RB6-8C5 mAb once; Group B was given 5% DSS for 3 d; Group C was given 5% DSS for 7 d; and Group D was given 5% DSS water for 7 d , and after the third day of DSS water , mice were given mAb RB6-8C5 once , and observed for 96 h after the RB6-8C5 dose . Ceca were resected and immediately fixed in Bouin's solution . Sections were stained with hematoxylin and eosin and reviewed by a veterinary pathologist . To determine if the C . albicans colonizing the GI tract was translocating to extra-gastrointestinal sites , mice were colonized with C . albicans strain SC5314 using the protocol described above then organized into the following four groups containing 24 mice per group: Group A , no immunosuppression; Group B , neutropenia induced by IP injection of 200 μg RB6-8C5 mAb once; Group C , neutropenia and GI mucosal disruption induced by one IP injection of 200 μg RB6-8C5 mAb and a contemporary , single IP injection of 150 mg/kg MTX; and Group D , neutropenia and mucosal damage induced by three IP injections of 150 mg/kg Cy given every other day ( Table 4 ) . Four mice from each group were sacrificed on days 0 , 1 , 2 , 3 , 4 , and 5 after immunosuppression was initiated . Blood ( 100 μl ) was drawn from an ethanol-cleaned tail vein and spread-plated onto YPD agar with vancomycin and gentamicin , TSA , and MacConkey agar plates . Mice were then sacrificed . Mesenteric lymph nodes ( MLN ) , spleens , and livers were immediately resected; organs were homogenized in 1 ml of 1% protease peptone; and 100 μl of homogenate was spread-plated on YPD agar with vancomycin and gentamicin , TSA , and MacConkey agar plates . The presence of a homogeneous population of creamy-white colonies on YPD with gentamicin and vancomycin were used for confirmation of the presence of C . albicans . Serial dilutions and plating for fungal enumeration were also done to obtain quantitative data ( CFU per gram tissue and CFU per ml blood ) . Survival data were analyzed by Fisher's exact test and the survival curve was analyzed by the Kaplan-Meier log rank test using the GraphPad Prism software ( San Diego , CA ) . Two-way comparisons of GI colonization levels were carried out using the Mann-Whitney tests and Prism software , and when multiple comparisons or more than two groups were analyzed , Bonferroni's correction to the significance level α was invoked .
|
Candida albicans is a fungus that lives harmlessly in the gastrointestinal ( GI ) tracts of humans . In cancer patients and patients undergoing bone marrow transplantation , however , the anti-cancer drugs that are administered to these patients also cause the undesired effect of suppressing the human immune system . The treatments allow C . albicans to spread into the blood and other organs and cause a severe disease . We found we could colonize the GI tracts of mice with C . albicans and then suppress the immune system with anti-cancer drugs to determine which components of the innate immune system ( neutrophils , lymphocytes , macrophages , or GI tract integrity ) are critical for preventing C . albicans from speading from the GI tract . We found that lowering the neutrophil counts and damaging the GI tract were both needed to cause systemic infection with C . albicans . We also found that the ability of C . albicans to switch from the yeast ( spherical ) form to the filamentous form is also important for establishing invasive disease . Our study provides new insights into the process of how a typically harmless microorganism inhabiting the GI tract can cause severe invasive disease once critical components of the host immune system are compromised .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"oncology",
"infectious",
"diseases",
"yeast",
"and",
"fungi",
"immunology",
"mus",
"(mouse)",
"homo",
"(human)"
] |
2008
|
Mucosal Damage and Neutropenia Are Required for Candida albicans Dissemination
|
Roberts syndrome ( RBS ) is a human disease characterized by defects in limb and craniofacial development and growth and mental retardation . RBS is caused by mutations in ESCO2 , a gene which encodes an acetyltransferase for the cohesin complex . While the essential role of the cohesin complex in chromosome segregation has been well characterized , it plays additional roles in DNA damage repair , chromosome condensation , and gene expression . The developmental phenotypes of Roberts syndrome and other cohesinopathies suggest that gene expression is impaired during embryogenesis . It was previously reported that ribosomal RNA production and protein translation were impaired in immortalized RBS cells . It was speculated that cohesin binding at the rDNA was important for nucleolar form and function . We have explored the hypothesis that reduced ribosome function contributes to RBS in zebrafish models and human cells . Two key pathways that sense cellular stress are the p53 and mTOR pathways . We report that mTOR signaling is inhibited in human RBS cells based on the reduced phosphorylation of the downstream effectors S6K1 , S6 and 4EBP1 , and this correlates with p53 activation . Nucleoli , the sites of ribosome production , are highly fragmented in RBS cells . We tested the effect of inhibiting p53 or stimulating mTOR in RBS cells . The rescue provided by mTOR activation was more significant , with activation rescuing both cell division and cell death . To study this cohesinopathy in a whole animal model we used ESCO2-mutant and morphant zebrafish embryos , which have developmental defects mimicking RBS . Consistent with RBS patient cells , the ESCO2 mutant embryos show p53 activation and inhibition of the TOR pathway . Stimulation of the TOR pathway with L-leucine rescued many developmental defects of ESCO2-mutant embryos . Our data support the idea that RBS can be attributed in part to defects in ribosome biogenesis , and stimulation of the TOR pathway has therapeutic potential .
Cohesin is a protein complex that adheres sister chromatids from the time of their replication until their division [1] , [2] . Cohesion between sister chromatids is facilitated by acetylation of the Smc3 subunit of the complex by the ECO1 acetyltransferase during S phase [3] , [4] , [5] . Human developmental syndromes such as Roberts syndrome ( RBS ) and Cornelia de Lange syndrome , termed cohesinopathies , arise from mutations in cohesin genes [6] . ESCO2 , which is a human ortholog of ECO1 in the yeast Saccharomyces cerevisiae , is inactivated in RBS [7] . RBS is an autosomal recessive , multi-system disorder characterized by prenatal growth retardation ( ranging from mild to severe ) , limb malformations ( including bilateral symmetric tetraphocomelia or hypomelia caused by mesomelic shortening ) , craniofacial abnormalities and mental retardation [8] , [9] , [10] , [11] , [12] . Previous studies have reported loss of ESCO2 acetyltransferase activity in RBS [13] . Chromosomes show a characteristic pattern of heterochromatin repulsion with the regions affected including centromeres and NORs ( nucleolar organizing centers or rDNA ) . A previous report revealed that mutations in yeast ECO1 and human ESCO2 impaired ribosomal RNA ( rRNA ) production and protein synthesis in budding yeast and human immortalized RBS cells [14] . Also , mutations in cohesin are associated with aberrant nucleolar morphology in yeast [15] . Cohesin binds to the rDNA in every organism studied , giving cohesin the potential to affect the structure and function of the nucleolus . We hypothesized that defective ribosome biogenesis contributes to the etiology of the RBS disorder . Perturbation of ribosome biogenesis is thought to lead to nucleolar stress and p53 activation . The mechanism appears to be the specific binding of ribosome proteins to Mdm2 , which inhibits its E3 ubiquitin ligase function toward p53 , leading to p53 stabilization and activation [16] , [17] , [18] . This binding happens when there is an imbalance of ribosomal proteins . Once p53 is stabilized , it will act to promote the transcription of Mdm2 in a feedback loop , as well as several other genes such as p21 and p27 , cyclin-dependent kinase inhibitors [19] . Depletion of ribosomal proteins such as Rpl5 , Rpl11 , or Rps7 induces p53 upregulation in various cell lines [17] , [20] , [21] , [22] . Loss of Rpl11 impaired zebrafish embryonic development via a p53-dependent apoptotic response [22] . Furthermore , defects in ribosomal proteins such as Rps6 ( S6 ) , Rps19 and Rpl24 have been implicated in congenital malformations and aberrant growth during fetal development [17] , [18] , [22] , [23] , [24] , [25] . Taken together , these studies indicate a strong link between p53 activation and the process of ribosome biogenesis . The TOR ( target of rapamycin ) pathway is a major node of control for protein translation and ribosome biogenesis . mTOR allows eukaryotic cells to adjust their protein biosynthetic capacity [26] , [27] , [28] through downstream effectors: ( 1 ) S6K1 , a kinase that phosphorylates Rps6 and promotes protein synthesis and cell proliferation , and ( 2 ) eukaryotic translation initiation factor 4E-binding protein 1 ( 4EBP1 ) , a protein that prevents translation when its unphosphorylated form interacts with eIF4E [29] , [30] , [31] . The TOR pathway can regulate intracellular processes such as transcription by RNA polymerase I [32] , [33] , [34] based on extracellular signals such as amino acid availability [35] , [36] , [37] or intracellular stress . mTORC1 can be stimulated by L-leucine through a mechanism that involves the leucyl tRNA synthase promoting the activity of GTP activating proteins that act on mTORC1 [38] , [39] . Given the hypothesis that RBS is associated with defects in rRNA production and ribosome biogenesis [40] , we wanted to test whether the TOR pathway was inhibited in RBS and if so , whether stimulation of TOR by L-leucine might rescue some of the defects associated with RBS . In this study , we found that translational efficiency and rRNA production are impaired in primary human RBS cells . Furthermore , nucleoli are highly fragmented . RBS cells showed an activation of p53 , and inhibition of mTOR . Inhibition of p53 activity with pifithrin-alpha ( Pifα ) , or stimulation of mTOR with L-leucine ( L-Leu ) both partially rescued proliferation in RBS cells . L-Leu also partially improved rRNA production and protein synthesis of RBS cells . Using zebrafish models for RBS , we found similar p53 activation and TOR pathway inhibition . L-Leu partially stimulated the TOR pathway in zebrafish RBS models and partially rescued several aspects of development .
In this study , we used RBS cells from 3 different sources . Immortalized skin fibroblasts came from a two-month old male Roberts syndrome patient homozygous for the mutation 877_878 delAG in exon 4 ( reported in [7] ) . De Winter and colleagues constructed a “corrected” RBS line in which the SV40 immortalized ESCO2-deficient fibroblasts were stably transfected with a cDNA construct encoding V5-tagged wild-type ESCO2 protein . Cytogenetic analysis of chromosomes demonstrated that the tagged ESCO2 protein compensated for the loss of ESCO2 activity in the mutant cells [41] . For untransformed primary fibroblasts , the donor subject was homozygous for a 5 bp deletion at nucleotide 307 in exon 3 of the ESCO2 gene ( c . 307_311delAGAAA ) resulting in a frameshift that leads to a truncated protein ( p . I102fsX1 ) . For untransformed amniocytes , the donor subject was a compound heterozygote: one allele has a 1 bp deletion at nucleotide 752 in exon 3 of the ESCO2 gene ( c . 752delA ) resulting in a frameshift leading to a premature stop codon and a predicted protein truncation ( p . K253fsX12 ) ; the second allele has an A>G substitution in intron 6 [c . IVS6-7A>G ( c . 1132-7A>G ) ] which activates a cryptic splice site ( p . I377_378insLX ) . Production of ribosomal RNA ( rRNA ) and protein synthesis are reduced and there are fewer actively translating ribosomes in eco1 mutant budding yeast and human immortalized RBS fibroblasts [14] . To extend this work , we conducted similar experiments with primary human RBS cells and carefully monitored their proliferation . Proliferation of three different sources of RBS cells ( immortalized fibroblasts , primary skin cells , and primary amniotic fluid cells , described above ) was much slower than wild type ( WT ) cells ( data shown only for immortalized cells , Figure 1A ) . Morphologically , the RBS AFCs appeared longer than normal AFCs ( data not shown ) . Re-introduction of ESCO2 to the immortalized line , indicated as ESCO2-corrected RBS cells , rescued proliferation . To further examine proliferation , we used FACScan to detect the cell cycle profile for these cell lines . There was an increase in G2/M cells in the RBS lines ( greater than 2 fold increase for immortalized fibroblasts and over 10 fold increase for untransformed AFC cells ) ( Figure 1B ) . Additionally , we performed 3H-uridine labeling to measure total rRNA production and 35S-methionine incorporation to quantify protein synthesis in the untransformed RBS cells . The results showed both rRNA and protein synthesis were significantly downregulated ( Figure 1C–D ) , similar to the previous report for the immortalized RBS cells . These data collected from three independent sources of RBS cells suggest that reduced ( 1 ) proliferation with a G2/M delay , ( 2 ) rRNA production , and ( 3 ) protein synthesis are general features of RBS cells . The cell cycle delay and the previously reported sensitivity to DNA-damaging agents in human RBS cells [41] prompted the question whether p53 activation might contribute to poor proliferation in RBS cells . p53 activation has been previously reported for zebrafish models of RBS [42] . By Western blot analysis , we found p53 was activated in immortal and untransformed RBS cells , and the levels of proteins encoded by genes positively regulated by p53 are higher in the mutant cells as well , as represented by induction of the targets Mdm2 , p21 , and p27 ( Figure 2A–D ) . To test whether p53 activation in RBS might be contributing to poor proliferation , WT and RBS cells were incubated with the p53 inhibitor pifithrin-alpha ( Pifα , 10 µM ) . This concentration was selected based on reports in the literature [43] , [44] as well as our own titrations ( data not shown ) . Cell proliferation and survival were quantified every two days using a bright-line hemocytometer with trypan blue staining to discriminate viable and dead cells . At 8 days , cell counting indicated the number of RBS cells was about 36% compared to WT cells ( normalized to 100% ) ( Figure 1A ) . This number will reflect both cell division and cell death . Inhibition of p53 activity partially restored proliferation of RBS cells as shown in the cell growth curve ( Figure 1A ) , suggesting that upregulation of p53 contributes to proliferation defects . But Pifα treatment did not suppress the elevated rate of death of the RBS cells ( Figure 2E ) . Overall , cell division and cell death both contributed to poor proliferation for RBS cells ( Figure 1A and Figure 2E ) , with cell death contributing approximately ∼15% . Pifα partially rescued the G2/M delay in the RBS cells , as shown in Figure S1A . However , Pifα did not rescue rRNA production or protein synthesis ( Figure S1B–C ) . One mechanism by which p53 can become activated is nucleolar stress [17] , [18] , [20] , [45] . Studies from mouse mutants showed that nucleolar disruption resulted in increased p53 levels and inhibition of mTOR activity , leading to mitochondrial dysfunction and increased oxidative stress , and contributing to neurodegenerative disease [46] , [47] , [48] . To examine the state of the nucleoli in RBS cells , we analyzed the distribution of the nucleolar proteins fibrillarin and nucleolin using immunofluorescence in the immortalized WT , ESCO2 mutant and corrected cells . These experiments showed that compared with WT cells and ESCO2 corrected cells , ESCO2 mutant cells have very fragmented signals for fibrillarin and very little distinct signal for nucleolin ( Figure 3 ) . Average size measurement of individual nucleolus area based on the fibrillarin staining showed that ESCO2 mutant cells have much smaller nucleolar size ( 1 . 4±0 . 1 µm2 ) compared with WT cells ( 8 . 3±1 . 0 µm2 ) , but re-introduction of ESCO2 protein partially rescued the nucleolar size in the ESCO2 corrected cells ( 3 . 7±0 . 2 µm2 ) . The highly fragmented nucleoli in RBS cells are consistent with the idea that p53 activation is caused in part by nucleolar stress . Furthermore , this result is consistent with our working model that cohesion at the rDNA may normally contribute to efficient nucleolar morphology and function . Inhibition of p53 did not rescue the aberrant nucleolar morphology in human RBS cells ( Figure S2 ) , suggesting that p53 activation may be a downstream consequence of the nucleolar fragmentation caused by ESCO2 mutation . Given the defect in protein synthesis in RBS cells , we examined the state of the TOR ( target of rapamycin ) pathway . Experiments carried out with an eco1-W216G yeast mutant revealed that the growth of this mutant was more sensitive to rapamycin treatment than a WT strain ( Figure S3A–B ) . This mutation has been associated with RBS and compromises the acetyltransferase activity of the protein [13] , [49] . Since rapamycin inhibits the TOR pathway , this result suggested that the TOR pathway might already be partly compromised in the mutant background . The mRNA for TOR1 was reduced 30% in the eco1-W216G mutant cells [14] . We asked whether the TOR pathway was affected in human RBS cells . Whole cell extracts were made from immortalized WT , ESCO2-mutant , ESCO2-corrected human fibroblasts , and the two untransformed human normal and RBS cell lines , all from Figure 1 . The concentrations of total protein were measured to normalize all samples and tubulin was used as a loading control . Extracts were used for Western blotting to measure individual members of the mTOR signaling pathway associated with protein translation . As shown in Figure 4A–D , ribosomal S6 kinase ( Thr389 of S6K1 , a site regulated by mTOR [50] , [51] ) was hypophosphorylated in the ESCO2 mutant strains , which is associated with reduced activity . Consistently , phosphorylation of ribosomal protein S6 ( S6 ) at Ser235/236 , which is a direct target of S6K1 and a component of the 40S ribosomal subunit , was reduced in ESCO2 mutant strains . This phosphorylation event normally promotes protein translation . Notably , phosphorylation of mTOR at Ser2448 , which is regulated by amino acid availability/energy status [52] , Akt [53] , [54] and cellular damage stress [55] was low , indicating that mTOR activity is impaired in RBS cells . The effect of ESCO2 mutation on the phosphorylation state of Eukaryotic translation initiation factor 4E-binding protein 1 ( 4EBP1 ) was examined . 4EBP1 is one member of a family of translational repressor proteins . The protein directly interacts with eukaryotic translation initiation factor 4E ( eIF4E ) , which is a limiting component of the multi-subunit complex that recruits 40S ribosomal subunits to the 5′ end of mRNAs . Interaction of this protein with eIF4E inhibits complex inhibits assembly and represses translation . Phosphorylation results in its dissociation from eIF4E and activation of mRNA translation . Phosphorylation of 4EBP1 decreases its electrophoretic mobility during SDS-polyacrylamide gel electrophoresis . In RBS cells , the faster mobility band α/β was dramatically increased that corresponds to the unphosphorylated form of 4EBP1 . This indicates that 4EBP1 exists in a state that will inhibit translation . Previous polysome analysis has shown that the fraction of actively translating ribosomes is reduced in both an eco1-W216G mutant yeast strain and RBS cells [14] . Therefore , the data collectively support the idea that translation and the TOR pathway are inhibited in RBS cells . We wondered whether stimulation of protein synthesis could rescue proliferation of RBS cells . L-leucine ( L-Leu ) has been reported to enhance translation initiation and protein synthesis [56] via induction of the mTOR pathway [57] , [58] , [59] . We supplemented the culture medium with 10 mM L-Leu or D-leucine ( D-Leu ) up to 8 days , and cell proliferation was quantified using a bright-line hemocytometer with trypan blue staining to discriminate viable and dead cells . The culture medium also contained other amino acids , such as L-glutamine ( L-Glu ) , which helps with the uptake of L-Leu . L-Leu significantly rescued proliferation of RBS cells ( Figure 4E ) . Both cell death ( Figure 4F ) and the G2/M delay ( Figure S4A ) were partially suppressed . Immunoblot analysis confirmed that phosphorylation of S6 in RBS cells was partially rescued by L-Leu treatment , but p53 was not altered ( Figure 4G ) . Notably , L-Leu supplementation in WT cells did not increase the levels of phosphorylated S6 ( Figure 4G ) , which could indicate that the TOR pathway is already fully active . Moreover , 3H-uridine and 35S-methionine metabolic labeling experiments indicated that the defects in rRNA production and protein synthesis in RBS cells were also partially restored by L-Leu ( Figure S4B–E ) . Rapamycin treatment curtailed the effect ( Figure S4C , E ) , suggesting L-Leu is acting through mTOR . Thus , the proliferation defects associated with RBS could be partially corrected through L-Leu stimulation of the mTOR pathway and translational induction . However , L-Leu treatment did not rescue nucleolar fragmentation ( Figure S5 ) . We speculate that the fragmentation is a molecular defect caused by ESCO2 mutation; L-Leu may stimulate mTOR and its downstream effectors without rescuing nucleolar fragmentation or p53 activation . We wanted to explore upstream effectors of mTOR in RBS cells . AMP-activated kinase ( AMPK ) activity can impede mTORC1 signaling through tuberous sclerosis complex ( TSC ) 1/2 as a repressor of mTOR function [55] , [60] . TSC2 has GTPase-activating protein ( GAP ) activity toward the Ras family small GTPase Rheb ( Ras homolog enriched in brain ) , and TSC1/2 antagonizes the mTOR signaling pathway via stimulation of GTP hydrolysis of Rheb [61] , [62] , [63] , [64] , [65] , [66] . AMPK activates TSC2 phosphorylation to catalyze the conversion of Rheb-GTP to Rheb-GDP and thus inhibits mTOR [67] . In two different sources of RBS cells , AMPK , its substrate Acetyl-CoA carboxylase ( ACC ) , and TSC2 ( Ser1387 ) were phosphorylated ( Figure S6A–B ) . Thus , TOR inhibition may be due in part to these upstream effectors . In some cancer cell lines , p53 activation triggers downregulation of the mTOR pathway [68] , [69] , but p53 and mTOR are not connected in all types of cells [55] , [70] . The phosphorylation of AMPK , ACC , and TSC2 were only slightly attenuated with siRNA directed at p53 ( Figure S6C ) , suggesting that the signals that are promoting the inhibition of mTOR are at least partly independent of p53 activation . This is consistent with the finding that L-Leu stimulation of mTOR did not reduce p53 activation , but boosted proliferation , rRNA production , and protein synthesis . mTOR can physically interact with RNA polymerase I and III promoters and stimulate transcription [71] , which could partly compensate for the poor production of rRNA and ribosomes associated with ESCO2 mutation . The phosphorylation of AMPK can be triggered by an increased AMP to ATP ratio [72] which can be caused by cellular stress . We found evidence for an increase in reactive oxygen species ( ROS ) associated with RBS cells ( Figure S6D–E ) . The ROS could be produced as part of the stress associated with mutations in ESCO2 . For example , ESCO2 mutant cells have defects in DNA replication and DNA damage repair [41] , [73] , and in this report we demonstrate defects in nucleolus formation . Any or all of these could contribute to the production of ROS [46] , [74] , [75] , [76] . Our results suggest that ESCO2 mutation is associated with oxidative stress , and this could contribute to the inhibition of mTOR . To further examine the molecular etiology underlying the developmental defects in a whole animal model for RBS , we utilized both ESCO2 morphants and ESCO2 homozygous transgenic mutant zebrafish . Morphants ( MO ) were created by microinjecting zebrafish embryos ( 1–4 cells ) with a morpholino to knockdown ESCO2 as previously described [42] . The control morpholino had 5-base mismatches ( ESCO2-5mis ) . The ESCO2-5mis injected embryos did not show any phenotypic changes compared with uninjected WT embryos . As previously reported , both the ESCO2 morphant and mutant embryos had multiple developmental defects such as underdeveloped head and eyes , cardiac edema , short body length , curved tail , and loss of skin pigmentation ( Figure 5A , B , E ) [42] , [77] . Western blot analysis showed similar findings with respect to the p53 and mTOR pathways as observed in human cells ( Figure 5C , D , F ) . Furthermore , inhibition of the TOR pathway , as monitored by phosphorylation of S6 , was found to increase in a dose dependent manner with the ESCO2 morpholino ( Figure 5C ) . We used the ESCO2-splice morpholino ( ESCO2-Splice MO ) , which shows a similar effect on embryo development as the ESCO2-ATG translation blocking morpholino ( Figure S7A ) , to test ESCO2-mRNA rescue . The results showed that the defects of ESCO2-morphants and mutant embryos were markedly rescued following injection of in vitro transcribed RNA encoding ESCO2 ( Figure S7A , B , C ) . To address whether the administration of L-Leu could rescue the phenotypic effects of ESCO2-defective zebrafish embryos , ESCO2 mutant embryos or ESCO2 morphants were treated with L-Leu , or D-Leu plus L-glutamine ( L-Glu 4 mM ) for 2 days post fertilization ( d . p . f . ) . L-Glu has been demonstrated to promote the import of amino acids [36] . ESCO2-morphants and ESCO2-mutant embryos raised in egg water with the control amino acid D-Leu showed profound developmental failure compared with ESCO2-5mis control morphants or WT embryos ( Figure 6A , C , E , F ) , including shortened body length and underdeveloped head and eyes . Interestingly , ESCO2 morphant and mutant embryos treated with L-Leu showed a remarkable improvement for all developmental deficiencies compared with those raised in egg water with D-Leu treatment ( Figure 6A , B , C , E , F ) . To further test the cooperative effects of L-Glu and L-Leu , ESCO2-morphants were treated with L-Glu alone or L-Leu alone . Only the combination showed developmental rescue ( Figure S7F ) , and this combination was used in all the experiments that follow . The effect was also specific to L-Glu and L-Leu since there was no improvement when morphants were treated with D-Leu , D-His , or L-Thr ( Figure S7G ) . We assigned morphant embryos to one of three classes: 1 ) mildly affected , 2 ) severely affected , and 3 ) dead . L-Leu not only rescued severe malformation of ESCO2 morphants , but also promoted their survival ( Figure 6A–B ) . Next we tested whether L-Leu was rescuing development of ESCO2-morphants by stimulating TOR signaling . Embryos treated with L-Leu or D-His were collected and analyzed by immunoblot analysis . Phosphorylation of S6 is strongly reduced in ESCO2 morphant and mutant embryos . However , L-Leu partially restored phosphorylation of S6 in ESCO2 morphant and mutant embryos ( Figure 6D , G ) . The ESCO2 mutant embryos were identified by fin clip and PCR genotyping analysis . Rapamycin treatment enhanced the defects observed in ESCO2 morphants at a concentration that did not affect WT embryos . Consistent with L-Leu acting through mTOR , rapamycin curtailed the rescue by L-Leu in ESCO2 morphants ( Figure 6E ) , but did not curtail the rescue by ESCO2 mRNA ( Figure S7C ) . The data indicate L-Leu stimulation of the TORC1 pathway can improve development and survival in ESCO2 defective embryos . To further study the effect of L-Leu on development of ESCO2 defective embryos , we assessed cartilage formation by staining with alcian blue at 5 days post fertilization ( d . p . f . ) . We observed marked abnormalities in the craniofacial elements in ESCO2 morphant and mutant embryos compared with control embryos ( Figure 7A–B ) . The effect was quantified using the relative ratio of the sum of the pq ( palatoquadrate ) cartilage and mc ( Meckel's cartilage ) divided by cranial length , based on a method developed in a recent report [78] ( Figure 7C ) . Impaired mTOR activity via Akt knockout has previously been shown to result in delayed skeletal development and poor cartilage matrix in mice [79] , [80] . With L-Leu treatment , cartilage length and head size were markedly recovered ( Figure 7A–C ) . ESCO2 morphants had previously been reported to show a global increase in apoptotic cells by TUNEL assay and acute elevation of caspase3/7 activity [42] . We were able to recapitulate these findings in ESCO2 morphant embryos using acridine orange staining of apoptotic cells ( Figure 7D–E ) and an assay for caspase3/7 activity ( Figure 7F ) . The cell death was reported to be p53 independent because a p53 mutation failed to rescue the observed apoptosis [42] . We also observed that knockdown of p53 levels had only a small rescue effect on ESCO2 deficient embryos ( Figure S8A–C ) . p53 knockdown did not significantly suppress caspase 3/7 activation in the ESCO2 morphants ( Figure S8D ) . While L-Leu significantly rescued cranial length , p53 knockdown did not ( Figure S8E–F ) . Since L-Leu significantly suppressed apoptosis levels in ESCO2 depleted embryos and caspase 3/7 activation , apoptosis may be coupled to regulation of the mTOR pathway . Measurements of embryo length showed that L-Leu partially restored body growth in ESCO2 morphant embryos ( Figure 7G ) . ESCO2 depleted zebrafish embryos exhibit a cell cycle block in G2/M phase [42] . Given our results in human RBS cells , we wondered if the mitotic delay could be overcome by L-Leu supplementation during embryo growth . The ESCO2 morphants were immunostained with phospho-Histone H3 ( pH3 ) in the presence of D-Leu or L-Leu , and the pH3 positive cells were quantified . Mitotic cells were robustly elevated in ESCO2 depleted embryos , but L-Leu treatment significantly reduced their number ( Figure 8A–B ) . Knockdown of p53 had a similar , but milder , effect ( Figure S9 ) .
The cohesinopathies have been proposed to arise from defects in transcription during embryogenesis . However , the mechanism by which mutations in the cohesin ring or cohesin associated factors caused these transcriptional defects has proven elusive . Our results demonstrate the activation of p53 and inhibition of mTOR in RBS . Furthermore , the defect in protein translation in RBS can be targeted via stimulation of the TOR pathway with L-Leu . Although p53 activation and nucleolar fragmentation persist with L-Leu treatment , L-Leu boosts TOR function which can then increase protein synthesis and cell proliferation . This “band-aid” is sufficient at the organismal level to partially rescue development ( see working model , Figure 8C ) . In contrast , p53 inhibition partially rescues cell division , but has less effect on cell death , protein synthesis , rRNA production , and developmental phenotypes in zebrafish . Our study represents the first RBS preclinical animal model in which L-Leu supplement produces an improvement in the developmental defects associated with RBS . ESCO2-inactivated mice showed termination of embryogenesis in the pre-implantation period due to a prometaphase delay and loss of cell viability [81]; in the future it would be interesting to test whether L-Leu could rescue embryogenesis in ESCO2-inactivated mice . Our study suggests some of the differential gene expression during embryogenesis in RBS may be due to translational defects . While RBS is rare , Cornelia de Lange syndrome ( CdLS ) is a more common cohesinopathy ( 1 in 10 , 000 births ) , and is also associated with mutations in cohesin genes . Depletion of cohesin genes ( Rad21 , Smc3 , Smc1 ) is associated with upregulation of p53 in zebrafish morphants [42] , [82] , [83] , [84] . Smc3 and Smc1 are mutated in CdLS [85] and Rad21 is mutated in a related cohesinopathy [86] . It will be important to explore whether p53 activation in these cases is due in part to nucleolar stress and defects in ribosome biogenesis . A CdLS mutation in Smc1 in yeast resulted in reduced rRNA , protein synthesis , and actively translating ribosomes [87] , demonstrating ribosome defects can be associated with Smc1 mutation . A CdLS mouse model ( NIPBL+/− ) showed decreased spontaneous adipogenesis and adipocyte differentiation as well as reduced body fat [88] . While p53 mRNA is not upregulated in zebrafish morphants for NIPBL [89] , the mTORC1-S6K1 pathway is needed for commitment to adipogenesis and the generation of de novo adipocytes [90] . Therefore , inhibition of the TOR pathway could potentially contribute to the lean body habitus of the NIPBL+/− mouse . In addition , cohesin binds to the c-Myc gene locus [82] , [91] , [92] , [93] , [94] , and c-Myc gene expression is downregulated in various CdLS cells with SMC1A and SMC3 mutations and a zebrafish CdLS model with NIPBL deficiency [83] , [95] , [96] . Since c-Myc serves as a direct positive regulator of ribosome biogenesis and protein synthesis [97] , we speculate that CdLS may also be associated with reduced ribosome and protein biosynthesis . It will be important in the future to determine which mutations in cohesin are associated with nucleolar stress , inhibition of the TOR pathway , and reduced translation . There are several disorders associated with defects in ribosome biogenesis , including Treacher Collins syndrome [98] , 5q-syndrome [99] , Diamond-Blackfan anemia ( DBA ) [100] , Shwachman-Bodian-Diamond syndrome ( SBDS ) [101] , and dyskeratosis congenita [102] . There is evidence for involvement of p53 and mTOR in these ribosomopathies . In all of these diseases p53 is upregulated , possibly due to nucleolar stress [103] , [104] , [105] , [106] . Some developmental phenotypes are shared in common between RBS and these disorders , including craniofacial , cardiac , urogenital , and limb defects [45] , [106] , [107] . The craniofacial dysmorphology of Treacher Collins syndrome , which can be caused by mutation in the nucleolar phosphoprotein Tcof , has been modeled in mice . The developmental defects in Tcof mice can be partially rescued by inhibition of p53 function [106] . Gross developmental phenotypes associated with mouse models for DBA and 5q-syndrome are rescued by loss of p53 function [45] , [108] , but anemia is not [109] . In contrast , p53 inhibition showed weak rescue in a zebrafish model for SBDS [110] , more similar to the observations for the zebrafish model for RBS , suggesting both p53 dependent and independent pathology for ribosome biogenesis defects . Interestingly , the amino acid L-leucine ameliorates developmental defects and anemia associated with mouse and zebrafish models for DBA and 5q-Sydrome , through the mTOR pathway [111] , [112] . We propose that RBS may share common underlying physiology with other ribosomopathies , with similar pathways being affected . Future work will reveal whether manipulating the p53 and TOR pathways will provide rescue for other cohesinopathies and ribosomopathies . In summary , we demonstrate for the first time in patient cells and an animal model of RBS that the TOR pathway is strongly impaired . Furthermore , L-leucine treatment of ESCO2 deficient zebrafish embryos and human cells activates the TOR pathway and relieves proliferation and developmental defects linked to RBS phenotypes . Some of the transcriptional changes that occur in RBS may occur as a result of translational defects . Our observations support the hypothesis that RBS may be partially attributed to defects in translation .
All animals were handled in strict accordance with good animal practice as defined by the relevant national and/or local animal welfare bodies , and all animal work was approved by the Stowers Institute for Medical Research , Institutional Animal Care and Use Committee . Zebrafish ( AB strain ) and esco2-transgenic mutant line were maintained by Reptile & Aquatics facility at the Stowers Institute for Medical Research as described previously [113] . Genotyping is shown in Figure S10 . ESCO2 genotyping primer-1 , 5′ GTACTATTCTACCCGGTAAGTGG 3′ ESCO2 genotyping primer-2 , 5′ GACGAGCTAATCTGCAGTTCAAG 3′ ESCO2 genotyping primer-3 , 5′ GCCAAACCTACAGGTGGGGTC 3′ Human RBS fibroblasts were a gift from Johan P . de Winter [41] . Briefly , primary skin fibroblasts from a two-month old male RBS patient homozygous for the mutation 877_878 delAG in exon 4 , were immortalized by transfection with a plasmid encoding the SV40 large-T antigen . Several weeks after transfection colonies of transformed cells appeared , which were mixed and further propagated . The transformed cells have been in continuous culture for over 60 passages and were therefore considered immortal . This immortalized RBS cell line ( VU1199-F SV40 ) , primary RBS fibroblasts ( VU1174-F ) , and wild type cell lines ( SV40 ) were cultured in Ham's F10 medium ( Gibco , Paisley , UK ) supplemented with 10% fetal bovine serum ( FBS , Hyclone , Logan , USA ) . Stable cell lines were generated by transfection of PvuI linearized expression vector pIRESneo containing cDNAs encoding V5-ESCO2 . These stable cell lines were cultured in complete medium containing G418 at 150 µg/ml ( Invitrogen , U . S . A ) . Cells were trypsinized with 0 . 05% Trypsin-EDTA ( Invitrogen , Grand Island , NY , USA ) , sub-cultured , and maintained in a humid incubator ( 37°C , 5% CO2 ) . Other RBS cells ( GM21873 and GM21872 ) were purchased from Coriell Institute for Medical Research . GM21872 is an untransformed skin fibroblast line from a 25 weeks female fetal RBS patient . These cells were cultured in Dulbecco's Modified Eagle Medium plus 15% fetal bovine serum . GM21873 is untransformed amniotic fluid-derived cell line from 20 weeks old female fetal RBS patient . These cells were cultured with AmnioMAX II Complete Medium . Both GM21872 and GM21873 are mutant for ESCO2 . The control cells were GM00957 and I91S-05 , normal human amniotic fluid-derived cells and untransformed fetal skin fibroblasts , respectively . Human RBS cells were transfected with p53-siRNA ( #6231 , Cell Signaling Technology , Inc . ) or control siRNA-A ( sc-37007 , Santa Cruz Biotechnology , Inc . ) using siPORT NeoFX Transfection Agent ( Ambion | Life Technologies , Inc . ) following the manufacturer's instructions . 24 hours after transfection , the cells were harvested and subjected to Western blot analysis . Antisense morpholino oligonucleotides ( MOs ) were obtained from GeneTools , LLC based on Mönnich M et al [42] . For microinjection , 2 nL of morpholino solution diluted in Danieau's buffer was injected into the yolk of wild type embryos at the 1 to 2-cell stage . Morpholino sequences and effective amounts were ESCO2-ATG-MO , 5′-CTCTTTCGGGATAACATCTTCAATC-3′ ( 1 ng–4 ng ) , and ESCO2-5mis-MO , 5′-GTAAACTACACAATGTTACCTCTCG-3′ ( 1 ng–10 ng ) , and ESCO2-Splice-MO , 5′-GTAAACTACACAATGTTACCTCTCG-3′ ( 6 ng–10 ng ) . ESCO2-ATG-MO targets the ATG start codon , and ESCO2-Splice-MO targets the 5′ donor of the exon/intron boundary of intron 2 of ESCO2 , to create knockdown “morphant” embryos . p53-morpholino , 5′-GCGCCATTGCTTTGCAAGAATTG-3′ ( 1 ng–4 ng ) Control morpholino is a Random Control MO produced by GeneTools , LLC . Full-length zebrafish ESCO2 mRNA was subcloned into pCS2+ . The ESCO2 mRNAs for microinjection were generated using a mMessage mMachine Kit ( Ambion ) by in vitro transcription , and 200 pg of each was injected into ESCO2 morphants and the mutant embryos at the 1-cell stage . Primers: ESCO2 full length mRNA primer-F: 5′ CGGGATCCCGATGTTATCCCGAAAGAGAAAAC 3′ ESCO2 full length mRNA primer-R: 5′ CCATCGATGGTTAGCTGATGAAGTTGTACACC 3′ Control mRNA primer-F: 5′ CGGGATCCCGATGTGTGACGACGACGAGACT 3′ Control mRNA primer-R: 5′ CCATCGATGGTTAGAAGCACTTGCGGTGGA 3′ Sodium dodecyl sulfate-polyacrylamide gel electrophoresis ( SDS-PAGE ) was performed using NuPAGE Novex 4%–12% Bis-Tris precast gels ( Invitrogen ) . Western blotting was performed according to standard protocol using a nitrocellulose ( Whatman , Protran ) or PVDF ( Millipore , Immobilon-P ) membrane . The following antibodies were used: phospho-S6K1 ( Thr389 ) , S6K1 , phospho-S6 ( Ser235/236 ) , 4EBP1 , p53 , phospho-AMPKα , AMPKα , phospho-Acetyl-CoA Carboxylase ( p-ACC ) , Acetyl-CoA Carboxylase ( ACC ) , phospho-Tuberin/TSC2 , Tuberin/TSC2 ( Cell Signaling Technology , Beverly , MA , USA ) , α-EFO2 ( ESCO2 ) , S6 ( Santa Cruz Biotechnology , U . S . A ) , α-tubulin ( Sigma , MO ) . Secondary antibodies were HRP linked , anti-rabbit IgG ( from donkey ) and anti-mouse IgG ( from sheep ) , ( GE Healthcare , NA934V , NA931V , and NA935V , respectively ) . In Figures 2 and 4 , the quantitative analysis of protein expression was performed by ImageQuant TL software . Cultured wild-type ( WT ) , ESCO2-mutant and ESCO2-corrected human RBS fibroblasts , two pairs of normal and RBS cells were grown in leucine-free Dulbecco's modied Eagle's medium ( DMEM ) plus 10% FBS . Before labeling , the cells were incubated with indicated L-Leu or D-Leu for 24 hrs . Cells were then washed in PBS twice , switched to 3 mL Met/Cys-free DMEM containing 10 µM MG-132 , a proteasome inhibitor . and pulsed with 30 µCi of 35S-methionine for the indicated time ( 4 hrs ) . Cells were lysed in RIPA buffer ( 50 mM Tris , pH 7 . 2; 150 mM NaCl; 1% sodium deoxycholate; 0 . 1% SDS; 1% Triton-X 100; 10 mM NaF; 1 mM Na3VO4 ) . Proteins were precipitated by the addition of hot 10% trichloroacetic acid . After centrifugation , the precipitate was washed twice in acetone . The precipitate was dissolved in 100 µL of 1% SDS and heated at 95°C for 10 min . An aliquot of the SDS extract was counted in Ecoscint for 35S radioactivity in a liquid scintillation spectrometer to determine the amount of 35S-methionine incorporated into proteins . Methods for rRNA labeling were derived from a previous report [114] . Cells were grown as for protein labeling . 3H-uridine ( 5 µCi ) was then incubated with 106 cells from each group for two hours . Total RNA was isolated with TriZol reagent ( Invitrogen , U . S . A ) and the concentration of each RNA sample was measured by OD260/280 . 1 µg of each sample was counted in a Beckman LS 6500 multipurpose scintillation counter to determine the amount of 3H-uridine incorporated . Three independent cultures were labeled to derive the standard deviation . Significance relative to WT was calculated using an unpaired t test . Cells seeded on coverslips were washed in PBS , fixed for 10 min at room temperature 20–22°C with 4% paraformaldehyde , permeabilized for 5 min in PBS containing 0 . 5% Triton X-100 and washed in PBS . After blocking for 30 min in PBS containing 1% BSA at room temperature , the preparations were incubated overnight at 4°C with the following antibodies diluted in PBS containing 1% BSA: mouse anti-fibrillarin with 1∶500 dilution , rabbit anti-nucleolin ( H-250; Santa Cruz ) with 1∶500 dilution . The coverslips were then washed in PBS and incubated for 30 min at room temperature with the secondary antibodies Alexa Fluor 488 goat anti-mouse and Alexa Fluor 555 donkey anti-rabbit ( Molecular Probes ) diluted in PBS containing 1% BSA and then washed in PBS . The coverslips were mounted on slides and analyzed by fluorescence imaging with a Zeiss Axioplan II confocal microscope . For p53 inhibitor ( Pifα ) treatment , cells were seeded in DMEM supplemented with 10% FBS in 6-well plates at a density of 5×104 cells/well ( in triplicate ) and grown overnight at 37°C in a humidified incubator with 5% CO2 . Treatments added the next day included Pifα ( 10 µM ) , DMSO , L-Leu , or D-Leu . Cells were trypsinized and stained with trypan blue , and the number of viable cells and dead cells were quantified using a Brightline hemocytometer . Human RBS cells were cultured and 3×106 cells for each sample was used for analysis . Cells were washed in PBS , then fixed with the addition of 1 mL of 70% cold at room temperature for 30 minutes . After washing with PBS , the cells were incubated with Rnase A and propidium iodide staining solution for 2 hrs . Samples were analyzed by flow cytometry . Data analysis was performed using FlowJo Version 7 . 6 . 4 software . Embryos were collected and most of the liquid was removed . 1 ml 2%PFA/1XPBS pH 7 . 5 was added to the embryos followed by nutation for 1 hr . Embryos were washed 1×10 mins with 100 mM Tris pH 7 . 5/10 mM MgCl2 , followed by addition of 1 ml of 0 . 04% alcian/10 mM MgCl2 stain pH 7 . 5 and overnight incubation with nutation . Embryos were washed with 80% ETOH/100 mM Tris pH 7 . 5/10 mM MgCl2 for 5 mins , then 50% ETOH/100 mM Tris pH 7 . 5 for 5 mins , then 25% ETOH/100 mM Tris pH 7 . 5 for 5 mins . 1 ml 3%H202/0 . 5%KOH was added . Open tubes were incubated for 10 mins , followed by 2×10 mins with 1 ml 25% glycerol/0 . 1% KOH . Bleach was rinsed out and 1 ml 50% glycerol/0 . 1%KOH was added . Embryos were mutated 10 mins followed by a wash with fresh 50% glycerol/0 . 1%KOH . Dechorionated embryos with or without ESCO2-knockdown were stained with phosphorylated histone H3 ( pH3 ) to measure mitotic cells in G2/M stage at 24 h . p . f . Phospho-Histone H3 was detected using an anti-Phospho-Histone H3 ( Ser10 ) antibody ( Cell Signaling ) , followed by detection with anti-rabbit-HRP ( Sigma ) . The number of pH3-staining positive cells was quantified on a consistent field ( end of yolk extension to end of tail in stage-matched embryos ) for 5 embryos per group . To compare cell counts between the samples , a two-sided t-test was used . The morphants were stained with acridine orange at 2 d . p . f . to detect apoptotic cells , which show green fluorescent granulated spots . The number of apoptotic cells was quantified on a consistent field ( end of yolk extension to end of tail in stage-matched embryos ) for 5 embryos per group . To compare cell counts between the samples , a two-sided t-test was used . Stained embryos were fixed in 4% PFA , and preserved and imaged in 90% glycerol . Live morphants were anesthetized in 0 . 4% tricaine in egg water for 3 minutes and immobilized in 1 . 5% methylcellulose . Embryos were visualized with a Leica Stereoscope ( Leica MZFLIII or Leica MZ16FA ) , a Leica DFC310FX camera and Leica application suite software . Images were processed using Photoshop CS5 ( Adobe ) . Embryo body length was measured at low magnification ( 3 . 2X–6 . 3X ) using a microscope ruler with Leica microsystems . The measurements were analyzed using a 2-way ANOVA with Bonferronis posttest analysis to assess the effects of ESCO2-knockdown , L-leucine , and time in combination . In addition , severely deformed and dead embryos were counted and statistically analyzed using a 2-way ANOVA one-way model to compare the differences . 30–50 embryos were homogenized in 1 mL of Trizol with a 20-gauge needle and syringe . The sample volume did not exceed 10% of total volume . Homogenized samples were incubated for 5 minutes at RT ( room temperature ) . 200 µL chloroform was added followed by a brief vortex and ∼3 minute incubation at room temperature . The sample was centrifuged at 12000 g for 10 minutes at 4°C . Following centrifugation , the aqueous phase was transferred to a fresh tube . An equal volume of isopropyl alcohol ( 2-propanol ) was added to the sample followed by mixing and incubation for 10 minutes at room temperature . The sample was centrifuged at 12000 g for 10 minutes at 4°C . The supernatant was removed and discarded . 1 mL 75% Ethanol ( RNase free ) was added to the sample . The sample was centrifuged at 12000 g for 10 minutes at 4°C . The supernatant was removed and discarded . The pellet was briefly air dried and then resuspended in 10 µL RNase-free water . Samples were stored at −80°C . An overnight culture of yeast was grown , followed by dilution to OD600∼0 . 1 . 1 mL of the diluted culture was spread onto a plate . Excess culture was removed and the plates were allowed to dry for ∼1 hr . 10 ul of rapamycin at the desired concentration was spotted onto sterile filter disks ( DIFCO concentration disks , 1/4″ diameter ) . With sterile tweezers , the disk was placed onto the lawn of cells . Plates were incubated for 1–2 days . The diameter of each halo ( region of no cell growth ) around the filter disk was measured . Cells were incubated with 10 µM of carboxy-H2DCFDA ( C400 , Invitrogen Corporation ) in the culture medium at 37°C for 3 hours , and the resulting fluorescence measured with a Synergy HT Multidetection Microplate Reader at an excitation wavelength of 485/10 nm and an emission wavelength of 528/20 nm . The results are reported as mean values±standard error ( mean±s . e . ) . Statistical analysis was performed by Student's t-test with SigmaPlot-Systat Software ( Sigmaplot Software Inc ) . An ANOVA two-way model was used to compare continuous variables . A P value <0 . 05 was considered statistically significant .
|
Roberts syndrome is a human developmental disorder caused by mutations in the ESCO2 gene . This gene encodes an acetyltransferase that acetylates the cohesin ring complex to promote a locked configuration . The cohesin complex binds to many locations on chromosomes and mutations that affect its function result in changes in gene expression . In fact , Roberts syndrome and other diseases caused by mutations in cohesin are associated with differential gene expression . We wanted to understand how mutations in ESCO2 affect two important molecular pathways that detect cellular stress , the p53 and mTOR ( mammalian target of rapamycin ) pathways . We report that mutations in ESCO2 are associated with p53 activation and inhibition of mTOR in human cells and zebrafish . We tested the rescue effect of p53 inhibition and mTOR activation on human Roberts syndrome cells and zebrafish models for Roberts syndrome . While both treatments displayed rescue effects , the activation of mTOR provided more significant rescue . Our work suggests that stimulation of the mTOR pathway with the amino acid L-leucine has therapeutic potential for Roberts syndrome . In addition , our work suggests that some of the differential gene expression in Roberts syndrome may be explained by translational inhibition connected with the inhibition of the mTOR pathway .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Stimulation of mTORC1 with L-leucine Rescues Defects Associated with Roberts Syndrome
|
Corticostriatal synapse plasticity of medium spiny neurons is regulated by glutamate input from the cortex and dopamine input from the substantia nigra . While cortical stimulation alone results in long-term depression ( LTD ) , the combination with dopamine switches LTD to long-term potentiation ( LTP ) , which is known as dopamine-dependent plasticity . LTP is also induced by cortical stimulation in magnesium-free solution , which leads to massive calcium influx through NMDA-type receptors and is regarded as calcium-dependent plasticity . Signaling cascades in the corticostriatal spines are currently under investigation . However , because of the existence of multiple excitatory and inhibitory pathways with loops , the mechanisms regulating the two types of plasticity remain poorly understood . A signaling pathway model of spines that express D1-type dopamine receptors was constructed to analyze the dynamic mechanisms of dopamine- and calcium-dependent plasticity . The model incorporated all major signaling molecules , including dopamine- and cyclic AMP-regulated phosphoprotein with a molecular weight of 32 kDa ( DARPP32 ) , as well as AMPA receptor trafficking in the post-synaptic membrane . Simulations with dopamine and calcium inputs reproduced dopamine- and calcium-dependent plasticity . Further in silico experiments revealed that the positive feedback loop consisted of protein kinase A ( PKA ) , protein phosphatase 2A ( PP2A ) , and the phosphorylation site at threonine 75 of DARPP-32 ( Thr75 ) served as the major switch for inducing LTD and LTP . Calcium input modulated this loop through the PP2B ( phosphatase 2B ) -CK1 ( casein kinase 1 ) -Cdk5 ( cyclin-dependent kinase 5 ) -Thr75 pathway and PP2A , whereas calcium and dopamine input activated the loop via PKA activation by cyclic AMP ( cAMP ) . The positive feedback loop displayed robust bi-stable responses following changes in the reaction parameters . Increased basal dopamine levels disrupted this dopamine-dependent plasticity . The present model elucidated the mechanisms involved in bidirectional regulation of corticostriatal synapses and will allow for further exploration into causes and therapies for dysfunctions such as drug addiction .
The present study reviews electrophysiological studies on corticostriatal synapse plasticity of medium spiny neurons and molecular biological studies focused on intracellular signaling cascades involved in this plasticity .
All signaling pathway reactions shown in Fig . 2 are represented by binding and enzymatic reactions . Binding reaction of molecule A and molecule B to form molecule AB ( 1 ) where and are rate constants for forward and backward reactions , is simulated by the ordinary differential equation: ( 2 ) The rate constants and were related to the dissociation constant and the time constant , i . e . , and . An enzymatic reaction of substrate S with enzyme E to produce product P was simulated by a collection of two elementary processes: 1 ) enzyme E bound to substrate S to form the enzyme-substrate complex ES; and 2 ) the complex ES dissociated into enzyme E and product P . The chemical equation can be written as ( 3 ) The Michaelis-Menten formulation was avoided due to problems with the steady-state assumption [21] , [43] . However , many papers and databases have provided only and the Michaelis constant rather than and . In such cases , it was assumed that was four times larger than ( i . e . and ) , based on the default setting in GENESIS/Kinetikit simulator . ( Tables S1 , S2 , S3 ) . Postsynaptic spines receive two presynaptic inputs: glutamatergic terminals from the cerebral cortex and dopaminergic terminals from the substantia nigra pars compacta . Plasticity of corticostriatal synaptic input results from phosphorylation of AMPA-type glutamatergic receptors , which promotes insertion into the postsynaptic membrane [20] , [44] , [45] . Below , the pathways linking glutamatergic and dopaminergic input to phosphorylation of AMPA receptors are delineated . The above-described signaling cascade , which links glutamatergic and dopaminergic inputs to AMPA receptor regulation , includes multiple excitatory and inhibitory pathways and feedback loops . This makes logical or intuitive inference of network behaviors virtually impossible; the outcomes depend on the strength and delay associated with each arrow in the diagram . However , logical or intuitive inference of network behaviors becomes virtually impossible , because the outcomes depend on strength and delay associated with each arrow in the diagram . This necessitates numerical simulation of a quantitative model of a signaling cascade to understand and prediction the dynamic behavior . Therefore , the present study designed a kinetic model of the cascade with the concentrations of intracellular calcium and extracellular dopamine as the inputs and AMPA receptor concentration in the postsynaptic membrane as the output . However the cascade , which links glutamate stimulation to calcium response was not included in this model but will be addressed in a future study . Similar to most large-scale cascade models , many reactions were adopted from previously published model [21] , [22] , [55] , [66] , [68] , [69] or deposited the DOQCS database [42] . When available , models of striatal spiny neurons were utilized ( e . g . , DARPP-32 , D1R , and AC5 ) . Otherwise , Otherwise , hippocampal neuron models were adopted ( e . g . , CaM , CaMKII , PP2B , I-1 , and AMPA receptor ) by assuming that molecular processes are common between different brain areas . If no previous model was available ( e . g . , PP2A , PP1 , CK1 , and Cdk5 ) , a reaction model was designed based on previous literature . Because many of the reactions remain poorly understood , a number of assumptions and simplifications were necessary to design the cascade models . For instance , although DARPP32 contains at least four phosphorylation sites that affect its enzymatic properties , phosphorylation of Ser102 by CK2 , which facilitates phosphorylation of Thr34 by PKA , was not modeled [70] . This was because the upstream regulation mechanisms for CK2 are now well known . Therefore , an 8-state model was designed for DARPP-32 , with three phosphorylation sites: Thr34 , Thr75 , and Ser137 . CK1 activation is required for Cdk5 activation [64] . Although the cascade linking these two molecules has not yet been identified , a direct pathway from CK1 to Cdk5 has been hyphothesized [23] . Although reports have described PP1 phosphorylation by Cdk5 [71] , a simple model was adopted from the DOQCS database , where only inhibition and disinhibition by I-1 and Thr34 were taken into account [72] . AMPA receptor trafficking in the postsynaptic membrane was modeled using the state transition diagram shown in Fig . 3 . AMPA receptors contain two phosphorylation sites - Ser845 phosphorylated by PKA and Ser831 phosphorylated by CaMKII . Therefore , a serial phosphorylation model was proposed for hippocampal neurons [73] where Ser831 was phosphorylated after Ser 845 phosphorylation . Initially , the model was tested to determine whether it reproduced known features of calcium- and dopamine-dependent plasticity in medium spiny neurons . Subsequently , the dynamic characteristics of the model were analyzed to predict effects of experimental manipulation . The entire model consisted of 72 reactions , with 132 reaction parameters . Among these , 83 parameters were retrieved from literature and model database . The remaining 49 parameters were hand-tuned to qualitatively reproduce the following properties: Forms and parameters of all reactions are listed in Tables S1 , S2 , S3 . Because many of the parameters affected multiple features of the model behavior , it was difficult to specify which parameter was responsible for the replication of each property . Numerical simulations were implemented by GENESIS/kinetikit ( http://www . genesis-sim . org/GENESIS/ ) . It was assumed that the postsynaptic spine was a homogeneous volume of ( cubed ) , so that each molecular species concentration represented the state variables . The two inputs to the cascade model comprised the concentrations of intracellular calcium , which were evoked by cortical glutamatergic input , and extracellular dopamine , which were evoked by nigral dopaminergic input . The time courses of the concentrations were approximated by the alpha function ( 4 ) which takes a maximum value of 1 when . The intracellular calcium concentration induced by a train of cortical spikes , which begin at time with inter-spike interval ( ISI ) , was simulated by ( 5 ) where and were the basal level and stimulus amplitude of calcium concentration , respectively ( Fig . 4A ) . The maximum function , rather than temporal summation , of calcium transients was used to replicate calcium response data from D1R-expressing striatal neurons [76] . The time constant of the alpha function was [77] , [78] . spikes at ISI ( 100 Hz ) were simulated and repeated six times with 10-sec intervals ( Fig . 4B ) . The concentrations used in the simulation were as follows: and . The extracellular dopamine concentration , which was induced by a single presynaptic spike at time , was simulated by: ( 6 ) where and were the basal level and stimulus amplitude of dopamine concentration , respectively ( Fig . 4C ) . The time constant of the alpha function was [78] , [79] . Dopamine input simulation was repeated six times with 10-sec intervals ( Fig . 4D ) . The concentrations used in the simulation were as follows: and .
The activities of intracellular molecules were simulated in response to four input conditions: i ) weak calcium input alone ( and ) ; ii ) strong calcium input alone ( and ) ; iii ) dopamine input alone ( and ) . iv ) weak calcium input coincident with dopamine input ( and ) ; The detailed input forms are explained by Eqs . ( 4 ) – ( 6 ) in Materials and Methods , and the transient time courses are shown in Fig . 4 . Results are shown in Fig . 5 . Direct downstream of calcium , CaM ( Fig . 5A ) , PP2B ( Fig . 5B ) , and PP2A ( Fig . 5C ) were moderately activated by weak calcium input ( cyan ) , but more highly activated by strong calcium input ( blue ) . In contrast , CaMKII ( Fig . 5D ) , which self-phosphorylates , did not respond to weak calcium input ( cyan ) , but responded drastically to strong calcium input ( blue ) . The differential activation profiles of PP2A , which dephosphorylates AMPA receptors , and CaMKII , which phosphorylates AMPA receptors , can be a source of bi-directional plasticity due to calcium input . CK1 ( Fig . 5E ) was activated by PP2B , but the response to strong calcium input was saturated due to a self-inhibitory mechanism . CK1 subsequently activated Cdk5 ( Fig . 5F ) and the Ser137 phosphorylation site of DARPP-32 ( Fig . 5G ) . Phosphorylation of Thr75 in DARPP-32 ( Fig . 5H ) increased with weak calcium input ( cyan ) via Cdk5 activation ( Fig . 5 ) , but decreased with strong calcium input ( blue ) via PP2A activation ( Fig . 5C ) . This bi-directional calcium effect on Thr75 was consistent with experiments showing phosphorylation of Thr75 with a glutamate receptor agonist [17] , [18] . Downstream of the D1Rs , AC5 ( Fig . 5I ) increased with dopamine input , but decreased with strong calcium input due to calcium inhibition . cAMP concentration ( Fig . 5J ) increased or decreased depending on AC5 activation level , and subsequently slowly decayed . Phosphorylated PKA ( Fig . 5K ) decreased with weak calcium input ( cyan ) and increased with strong calcium input ( blue ) , mirroring the bi-directional changes of Thr75 ( Fig . 5H ) . PKA increased at a slower rate with dopamine input ( red ) , subsequent to increased cAMP . Simultaneous stimulation of weak calcium and dopamine resulted in a bi-phasic response , including an initial dip followed by a sustained elevation . PDE activation ( Fig . 5L ) was similar to the activation profile of PKA . Dopamine input ( red ) resulted in increased Thr34 phosphorylation of DARPP-32 ( Fig . 5M ) via PKA activation . Calcium input ( cyan and blue ) reduced Thr34 phosphorylation due to stronger inhibition by PP2B . The decreased Thr34 phosphorylation due to calcium input was consistent with experimental results utilizing AMPA and NMDA [17] . Coincident calcium input ( magenta ) reduced the response of Thr34 to dopamine input ( red ) . These results were consistent with experimental responses to different levels of dopamine and NMDA inputs [74] . Dopamine input alone increased phosphorylation of Inhibitor-1 ( I-1 ) ( Fig . 5N ) via PKA activation . However , I-1 phosphorylation decreased due to either weak or strong calcium input , or simultaneous calcium and dopamine inputs , via PP2B inhibition . Phosphorylation of PP1 ( Fig . 5O ) was opposite to that of I-1 by dopamine input ( red ) , but similarly phosphorylated by both strong ( blue ) and weak ( cyan ) calcium inputs , even under simultaneous dopamine input ( magenta ) . Finally , via phosphorylation by CaMKII ( Fig . 5D ) and PKA ( Fig . 5K ) , and dephosphorylation by PP2A ( Fig . 5C ) and PP1 ( Fig . 5O ) , AMPA receptor phosphorylation at Ser845 decreased due to weak calcium input , but increased due to strong calcium input and simultaneous calcium and dopamine inputs ( Fig . 5P ) . Fig . 6A shows the time course of synaptic efficacy ( AMPA receptor concentration in the post-synaptic membrane ) induced by different levels of dopamine input coincident with a weak calcium input . While the absence of dopamine input caused depression of the synapse ( solid ) , increased dopamine levels resulted in potentiation . Fig . 6B shows the time course of synaptic efficacy in three different levels of calcium input without dopamine input . While weak calcium input causes depression , increased calcium input resulted in potentiation . Synaptic efficacy was evaluated 10 min after conditioning as an index of long-term synaptic plasticity . Synaptic efficacy increased with increasing dopamine input coincident with calcium input ( Fig . 6 ) . In conditions of dopamine depletion , where both and were set at , the calcium input did not alter synaptic efficacy . These results were in accordance with dopamine-dependent synaptic plasticity [27] , as characterized in Fig . 1A . Fig . 6D shows synaptic plasticity dependence on calcium input levels in the absence of dopamine input . Weaker calcium input resulted in LTD , but stronger calcium input caused LTP . These results were consistent with previous experimental observations [11] , [28] , [29] , as schematized in Fig . 1B . To further clarify the interactions between calcium and dopamine inputs and the roles of molecules in the signaling cascade , 2D maps of synaptic plasticity were plotted with different levels of calcium and dopamine inputs using standard and modified models . Fig . 7A shows synaptic plasticity after 10 minutes stimulation in the standard model . LTD was induced by weak calcium input in the absence of dopamine ( blue area ) , and LTP was induced by either strong calcium or strong dopamine input ( red area ) . When CaMKII activation was fixed at a steady-state level ( Fig . 7B ) , increased calcium input did not induce LTP . Rather , LTD occurred only at low levels of dopamine input . When PKA was fixed at the steady-state level ( Fig . 7C ) , dopamine-dependent plasticity disappeared . Fixing PP1 produced LTP , regardless of the strength of calcium and dopamine inputs ( Fig . 7D ) . The potentiation induced by strong dopamine alone disappeared , because the disinhibition due to decreased PP1 ( corresponding to the red line in Fig . 5O ) was removed .
Several studies have modeled signal transduction in medium spiny neurons [21]–[23] . The novelty of the present model is the incorporation of AMPA receptor phosphorylation and membrane trafficking to directly assess the effects of cascade dynamics on striatal synaptic plasticity . This allowed for the reproduction of both LTD and LTP in calcium- and dopamine-dependent plasticity and to predict interactions between calcium and dopamine inputs , as shown in Fig . 7 , and effects of various manipulations on striatal synaptic plasticity . Embedding of the present model in a complete neuronal model , or even a neural network model , enables the assessment of the role of calcium- and dopamine-dependent plasticity in cellular and network functions . The model can also serve as the basis for building simplified signaling cascade models for large-scale simulation and theoretical analysis . The present signaling cascade model involving DARPP-32 differs from previous models in several points . The factors incorporated by this model but not by existing models [21]–[23] were inhibition of PDE by PKA , Ser137 effect on Thr34 , and inhibition of PP1 by I-1 . The CK1-Cdk5 pathway , which has been previously described [23] , was critical for reproducing bidirectional phosphorylation of Thr75 , which was dependent on calcium input intensity . In addition , the present study performed a rigorous analysis of bistability of positive feedback loop formed by PKA , PP2A , and DARPP-32 on Thr75 , which was a source of a threshold-like response of PKA activity to both dopamine and calcium inputs . The model prediction of Thr34 and Thr75 responses to dopamine and calcium input were consistent with the Fernandez model [21] if the calcium input levels from the Fernandez model were regarded as the strong calcium input for the present model . However , simultaneous calcium and dopamine inputs resulted in Thr34 dephosphorylation in the present model , but phosphorylation in the Lindskog model [22] . This discrepancy could be due to inactivation by the calcium-PP2B-Thr34 pathway was stronger than activation by the PKA-Thr34 pathway in present model . DARPP-32 phosphorylation on Thr75 has been shown to because of glutamate , AMPA , or NMDA exposure , but returns to normal level within 10 min [17] , [18] . In addition , an mGluR agonist has been shown to potentiate Cdk5 activation and phosphorylation of DARPP-32 on Thr75 and Ser137 , and returns to baseline levels after peaking at 2 min [64] . Assuming that an mGluR agonist induced weak calcium levels , and glutamate , AMPA , or NMDA produced strong calcium input , those experimental results were consistent with the present results , as shown in Fig . 5H . In present model , phosphorylation of DARPP-32 on Thr75 , as a result of weak calcium input , takes place through the CK1-Cdk5 pathway . Although CK1 activation is required for Cdk5 activation through signaling from mGluR [64] , it is not known whether the pathway from CK1 to Cdk5 is direct . Similar to a previous model , the present study assumed direct activation of Cdk5 by CK1 for simplicity [23] . Alternative mechanisms for inhibition of PP2A dephosphorylation on Thr75 exist - either through the calcium-AC5-cAMP-PKA pathway or the calcium-CaM-PDE-cAMP-PKA pathways . More quantitative data on the strengths of these pathways and additional in silico experiments are necessary to definitely determine the role of the CK1-Cdk5 pathway in calcium-dependent LTD . AMPA receptor trafficking in the present model was derived from Hayer's model [82] . The primary modification comprised sequential phosphorylation of Ser845 by PKA followed by Ser831 phosphorylation by CaMKII , as proposed by Lee et . al . [83] . However , the LTP mechanism in the present striatal model differed from the hippocampal LTP by Lee et . al . [83] . Previous results demonstrated that the phosphorylation of Ser845 did not increase during LTP [83] , and the present model showed that the phosphorylation of Ser845 increased during dopamine-dependent LTP , but did not increase during calcium-dependent LTP . In addition , PKA was required for striatal LTP [75] To address this feature in the present striatal model , most of the AMPA receptors were dephosphorylated at the baseline . This prediction was consistent with the lower phosphorylation level of Ser845 by reduced PKA levels due to inhibition by DARPP-32 in the striatum [69] . It should be noted , however , that the observation of sequential AMPA receptor phosphorylation by Lee et al . [83] in the hippocampus did not exclude a parallel phosphorylation model . It could be interpreted as a result of high PKA and low PPI concentration at the baseline in the hippocampus . It is a subject of future study whether a parallel phosphorylation model can also reproduce the striatal synaptic plasticity . D1-type neurons express GluR1 and GluR2/3 in the spines [84] , [85] . A previous hippocampal study [86] showed that GluR1 subunit trafficking was a result of stimulation , but that GluR2 subunit trafficking was constitutive . In addition , chronic treatment with the antidepressant maprotiline increases GluR1 , but not GluR2 [87] . Moreover , GluR2-lacking AMPA receptors exhibit larger single-channel currents than GluR2-expressing AMPA receptors [88] . For these reasons , trafficking of GluR1 , but not GluR2 , was modeled in the present study to ascertain whether synaptic plasticity responded to stimulus . Some theoretical studies [89] , [90] have predicted that NMDA receptor-mediated calcium influx results in bidirectional synaptic change . However , these studies modeled only AMPA receptor phosphorylation , but not trafficking , and also did not consider striatal synaptic plasticity . Although the present model considered the number of AMPA receptors in the postsynaptic membrane as a measure of synaptic efficacy , previous studies have suggested that the conductance of AMPA receptor varies according to the phosphorylation state . For example , Ser831 phosphorylation increases conductance [91] and Ser845 phosphorylation increases open probability [92] , [93] . If these effects are taken into consideration , the amplitude of LTP could be larger , as observed in experiments [8]–[11] . Threshold dynamics due to the bistability of the positive feedback loop of PKA , PP2A , and Thr75 on DARPP-32 played an important role in reverting the LTD to LTP in dopamine-dependent plasticity . However , when embedded into the entire system , the loop did not exhibit complete bistability , as demonstrated by gradual conversion of synaptic conductance to baseline levels ( Fig . S2 ) . The possible mechanisms for longer-lasting synaptic plasticity are described below . First , bistability of some proteins in the cascade has been reported , such as the bistability of CaMKII phosphorylation [82] . However , CaMKII activity did not last for an extended period of time in the present model . This was consistent with a previous study [94] , which reported that CaMKII activity returns to baseline within 2–5 min . Hayer et . al . observed bistability of AMPA receptor phosphorylation and Catellani et . al . [73] mathematically determined bistability in the sequential AMPA receptor phosphorylation model . These bistable mechanisms were not incorporated in the present model , but may contribute to synaptic changes over longer periods of time . Second , the present model did not consider increased levels of AMPA receptors and other proteins as a result of gene transcription . A possible link from the current model to longer-term synaptic plasticity is cAMP-response element binding protein ( CREB ) , which controls gene transcription for longer-term synaptic plasticity in the striatum [95] . CaMKII , PKA , and PP1 directly activate CREB , but also indirectly via extracellular signal-regulated kinase ( ERK ) , which activates CREB [35] , [96] . In addition , calcium activates mitogen-activated protein kinase kinase ( MEK ) , which activates ERK [97] . PP1 activates striatal enriched phosphatase ( STEP ) [98] , which inhibits ERK , and PKA inactivates STEP . As a result , CREB is inhibited by PP1 and activated by CaMKII and PKA . Therefore , activation of CaMKII and PKA , as well as inhibition of PP1 , which results in AMPA receptor phosphorylation , can also trigger gene transcription through CREB activation . Approximately half of the model parameters were based on previous reports and databases [21] , [22] , [42] , [55] , [66] , [68] , [69] , and the remaining half were designed to reproduce experimental findings [15]–[18] , [39] , [46] , [64] , [74] . Model behavior robustness was determined by altering the kinetic parameters of the PKA-PP2A-Thr75 loop up to ten-fold ( Fig . 13 ) . Persistence of nonlinear threshold behavior , despite shifts in thresholds , was also verified . Although the present model parameters reflected some uncertainty , the model served as a useful starting point for exploring the mechanisms influencing corticostriatal synaptic plasticity by testing alternative parameter values or incorporating additional pathways . The present model did not include a number of known pathways such as the effect of DARPP-32 Ser102 on phosphorylation of Thr75 [99] . Membrane potential of striatal medium spiny neurons shifts between up- and down-states , depending on cortical inputs [100] . During the up-state , LTP is induced by cortical stimuli , even without dopamine input [31]–[34] . LTP is also induced by cortical stimulation in a magnesium-free solution [11] , [28] , [29] . Both cases reflect calcium-dependent plasticity because of the large calcium influx through NMDA receptors . Two types of medium spiny neurons exist: D1 receptor-expressing neurons that project to the direct pathway , and D2 receptor-expressing neurons that project to the indirect pathway [101] , [102] . In D1 neurons , dopamine increases cAMP via G-proteins and AC5 , similar to the present model . However , in D2 neurons dopamine inhibits AC5 and decreases cAMP so the effect of dopamine input is opposite to that in D1 neurons . Schultz et . al . recorded the activities of dopamine neurons in the substantia nigra in monkeys and found that dopamine neurons encode error signals of reward prediction [103] . The reinforcement learning model of the basal ganglia posits that striatal neurons learn to compute expected reward based on the reward prediction error signal carried by dopamine neuron firing [103] . Dopamine-dependent synaptic plasticity plays a major role in learning . The medium spiny neurons are depolarized by glutamatergic inputs from the cortex that represent a sensory or a contextual state . When the acquired reward is more than expected , phasic dopamine neuron firing would induce LTP of the activated cortico-striatal synapses . On the other hand , if the reward is less than expected , a pause in dopamine neuron firing would cause LTD of those synapses . The glutamatergic input would not only cause depolarization and firing , but also induce changes in molecular states , such as the phosphorylation level of DARPP-32 and/or shift the threshold of the positive feedback loop , which would serve as the short-term memory of preceding states . To support this scenario , the temporal order of calcium and dopamine input is a critical factor . Assuming that calcium flux by glutamatergic input is a fast process , the synaptic efficacy should be potentiated when calcium input ( associated with a sensory or contextual state ) precedes dopamine input ( associated with a reward prediction error signal ) . Our model is consistent with this point ( Fig . S1 ) . On the other hand , our model also predicts that the effect of the temporal order on synaptic plasticity is not strong enough . This suggests additional interactions between dopamine and calcium signaling . For example , dopamine facilitates L-type calcium channels , which affect the calcium influx through the interaction of glutamate receptor activation and and back-propagating action potentials . To more precisely simulate calcium dynamics , we have to construct a whole neuron model [104] and combine it with the signaling cascade model . There are several interaction pathways between calcium and dopamine signaling . In the upstream of PKA , calcium directly inhibits AC5 and indirectly cAMP through CaM and PDE . While calcium inhibition of AC5 depended on the timing between calcium and dopamine , PDE inhibition of cAMP did not depend on this timing very much . The stronger interaction of dopamine and calcium on PKA was through DARPP-32 . Weak calcium input inhibited PKA through the phosphorylation of Thr75 by Cdk5 , but strong calcium input activated PKA through the dephosphorylation of Thr75 by PP2A . While dopamine input reduced the increase of Thr75 by a weak calcium input , it did not affect the decrease of Thr75 by a strong calcium input . Furthermore , the subsystem around the PKA-PP2A-DARPP-32 positive feedback loop showed bistability while PKA activity showed a threshold like response to cAMP activation by dopamine input . However , this loop became mono-stable with both activation of Cdk5 by a weak calcium input , leading to a low level of PKA , and by activation of PP2A by a strong calcium input , leading to a high level of PKA . Addictive drugs ( e . g . cocaine and amphetamine ) increase the basal level of dopamine by inhibiting the reuptake of dopamine and facilitating the release of presynaptic dopamine [5] . They ultimately decrease DARPP-32 phosphorylation on Thr75 and increase it on Thr34 [105] . In our model , increased basal dopamine levels caused LTD with the calcium and dopamine inputs which caused LTP under control conditions ( Fig . 14 ) . This result is consistent with the theory that the value of everything except for drugs decreases because of the impairment of appropriate learning in drug addiction [106] .
|
Recent brain imaging and neurophysiological studies suggest that the striatum , the start of the basal ganglia circuit , plays a major role in value-based decision making and behavioral disorders such as drug addiction . The plasticity of synaptic input from the cerebral cortex to output neurons of the striatum , which are medium spiny neurons , depends on interactions between glutamate input from the cortex and dopaminergic input from the midbrain . It also links sensory and cognitive states in the cortex with reward-oriented action outputs . The mechanisms involved in molecular cascades that transmit glutamate and dopamine inputs to changes in postsynaptic glutamate receptors are very complex and it is difficult to intuitively understand the mechanism . Therefore , a biochemical network model was constructed , and computer simulations were performed . The model reproduced dopamine-dependent and calcium-dependent forms of long-term depression ( LTD ) and potentiation ( LTP ) of corticostriatal synapses . Further in silico experiments revealed that a positive feedback loop formed by proteins , the protein specifically expressed in the striatum , served as the major switch for inducing LTD and LTP . This model could allow us to understand dynamic constraints in reward-dependent learning , as well as causes and therapies of dopamine-related disorders such as drug addiction .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"computational",
"biology/computational",
"neuroscience",
"computational",
"biology/signaling",
"networks"
] |
2010
|
A Kinetic Model of Dopamine- and Calcium-Dependent Striatal Synaptic Plasticity
|
Leaf-cutter ants are one of the most important herbivorous insects in the Neotropics , harvesting vast quantities of fresh leaf material . The ants use leaves to cultivate a fungus that serves as the colony's primary food source . This obligate ant-fungus mutualism is one of the few occurrences of farming by non-humans and likely facilitated the formation of their massive colonies . Mature leaf-cutter ant colonies contain millions of workers ranging in size from small garden tenders to large soldiers , resulting in one of the most complex polymorphic caste systems within ants . To begin uncovering the genomic underpinnings of this system , we sequenced the genome of Atta cephalotes using 454 pyrosequencing . One prediction from this ant's lifestyle is that it has undergone genetic modifications that reflect its obligate dependence on the fungus for nutrients . Analysis of this genome sequence is consistent with this hypothesis , as we find evidence for reductions in genes related to nutrient acquisition . These include extensive reductions in serine proteases ( which are likely unnecessary because proteolysis is not a primary mechanism used to process nutrients obtained from the fungus ) , a loss of genes involved in arginine biosynthesis ( suggesting that this amino acid is obtained from the fungus ) , and the absence of a hexamerin ( which sequesters amino acids during larval development in other insects ) . Following recent reports of genome sequences from other insects that engage in symbioses with beneficial microbes , the A . cephalotes genome provides new insights into the symbiotic lifestyle of this ant and advances our understanding of host–microbe symbioses .
Ants are one of the most successful insects on earth , comprising up to 20% of all terrestrial animal biomass and at least 25% of the entire animal biomass in the New World Tropics [1] . One of the most conspicuous and prolific Neotropical ants are the leaf-cutters ( Tribe: Attini ) , so-called because of their leaf-cutting behavior [2] . Leaf-cutters are unique among ants because they obligately farm a specialized , mutualistic fungus that serves as their primary food source [3] . Using a complex system of trails , foraging ants seek out and cut leaves ( Figure 1A ) that they use to manure a fungal crop in specialized subterranean fungus gardens ( Figure 1B ) within their colonies . Fungus farming by ants is exclusive to the New World and is thought to have evolved once 50 million years ago [4] , culminating in the leaf-cutter ants . A single mature colony of the genus Atta can fill a volume of up to 600 m3 and their fungus gardens can support millions of workers capable of harvesting over 400 kg of leaf material ( dry weight ) annually [1] . These ants are thus one of the most widespread and important polyphagous insect herbivores in the Neotropics . The importance of leaf-cutter ants in Neotropical rainforest ecology lies in their ability to substantially alter arboreal foliage through their extensive leaf-cutting activities . Estimates suggest that leaf-cutter ants remove 12–17% of the total leaf production in tropical rainforests [1] . As a group , they harvest more plant biomass than any other Neotropical herbivore including mammals and other insects . As a result , leaf-cutter ants are a major human agricultural pest , responsible for billions of dollars in economic loss each year [5] . These ants do , however , have a positive impact on rainforest ecosystems , as they contribute to rapid soil turnover through their nest excavation activities [6] , stimulate plant growth by cutting vegetation [7] , and help to recycle organic carbon [1] . In addition to their importance in Neotropical ecosystems , leaf-cutter ants also serve as a model for understanding the ecology and evolution of host-microbe symbioses [8] . In return for receiving a continuous supply of leaf-material , protection from competitors , and dispersal , the fungus these ants grow provide nutrients in the form of specialized hyphal swellings called gongylidia . Gongylidia , which contain a mixture of carbohydrates , amino acids , proteins , lipids , and vitamins [9] , is the sole food source for developing larvae . The fungus garden is also known to harbor other microbial symbionts including nitrogen-fixing bacteria that provide both fungus and ants with nitrogen [10] , and a diverse community of fungus garden bacteria that appear to help the fungus degrade plant biomass [11] . The complexity of the leaf-cutter ant symbiosis is further highlighted by the presence of a specialized microfungal pathogen that exploits the ant-fungus mutualism [12] , [13] . As a result , the leaf-cutter ant symbiosis comprises at least three established mutualists and one specialized pathogen . With the reported presence of additional microbial symbionts from Acromyrmex leaf-cutter ants [14]–[19] , and the isolation of numerous microbes from other fungus-growing ants [20]–[22] , this ant-microbe symbiosis is perhaps one of the most complex examples of symbiosis currently described . Leaf-cutter ants in the genus Atta are also known for their morphologically diverse caste system ( Figure 1C ) , which reflects their complex division of labor [23] , [24] . For example , the overall body size of Atta cephalotes workers varies tremendously ( i . e . , head widths ( HW ) ranging from 0 . 6 mm to 4 . 5 mm [23] ) , and these differences correspond to the tasks performed by workers . The smallest workers ( HW 0 . 8–1 . 6 mm ) engage in gardening and brood care as their small mandibles allow them to manage the delicate fungal hyphae and manipulate developing larvae . Some of these workers are also responsible for processing plant material collected by foragers by clipping large pieces of leaf material into smaller fragments to manure the fungus . Larger workers ( HW >1 . 6 mm ) are responsible for foraging , as they have mandibles powerful enough to cut through leaves and other vegetation [24] . The largest workers form a true soldier caste , which are involved primarily in nest excavation and colony defense [23] , [24] . To gain a better understanding of the biology of leaf-cutter ants , we sequenced the genome of Atta cephalotes using 454 pyrosequencing technology [25] and generated a high-quality de novo assembly and annotation . Analysis of this genome sequence reveals a loss of genes associated with nutrient acquisition and amino acid biosynthesis . These genes appear to be no longer required because the fungus may provide these nutrients . With the recent reports of genomes from other social hymenopterans [26] , [27] and insects that engage in microbial mutualisms [28] , [29] , the A . cephalotes genome contributes to our understanding of social insect biology and provides insights into the interactions of host-microbe symbioses .
Three males from a mature Atta cephalotes colony in Gamboa , Panama were collected and sequenced using 454-based pyrosequencing [25] with both fragment and paired-end sequencing approaches . A total of 12 whole-genome shotgun fragment runs were performed using the 454 FLX Titanium platform in addition to two sequencing runs of an 8 kbp insert paired-end library , and one run of a 20 kbp insert paired-end library . Assembly of these data resulted in a genome sequence of 290 Mbp , similar to the 300 Mbp genome size previously estimated for A . cephalotes [30] . The genome is spread across 42 , 754 contigs with an average length of 6 , 788 bp and an N50 of 14 , 240 bp ( Table 1 ) . Paired-end sequencing ( 8 kbp and 20 kbp inserts ) generated 2 , 835 scaffolds covering 317 Mbp with an N50 scaffold size of 5 , 154 , 504 bp . The disparity between contig and scaffold size may be accounted for by the number of repeats present in this genome ( see below ) leading to an inflated assembly size due to chimeric contigs . Based on the total amount of base pairs generated and its predicted genome size , we estimate that the coverage of the A . cephalotes genome is 18-20X . To determine the completeness of the A . cephalotes genome sequence , we performed three analyses . First , we compared the A . cephalotes genome annotation against a set of core eukaryotic genes using CEGMA [31] , and found that 234 out of 248 core proteins ( 94% ) were present and complete , while 243 ( 98% ) were present and partially represented . Second , we analyzed the cytoplasmic ribosomal proteins ( CRPs ) in the A . cephalotes genome and identified a total of 89 genes ( Text S1 ) . These encode the full complement of 79 CRPs known to exist in animals , nine of which are represented by gene duplicates ( RpL11 , RpL14 , RpS2 , RpS3 , RpS7 , RpS13 , RpS19 , RpS28 ) or triplicates ( RpL22 ) . The presence of a complete set of these numerous genes , which are widely distributed throughout the genome , confirmed the high-quality of the A . cephalotes genome sequence ( Text S2 ) . Finally , we found that the genome of A . cephalotes contains 66 of the 67 known oxidative phosphorylation ( OXPHOS ) nuclear genes in insects ( Text S3 ) . The only OXPHOS gene missing , cox7a , we found to also be missing in the two ants Camponotus floridanus and Harpegnathos saltator and the honey bee Apis mellifera . The presence of this gene in the jewel wasp Nasonia vitripennis ( along with other holometabolous insects ) , suggests an aculeate Hymenoptera-specific loss , rather than a lack of genome coverage for A . cephalotes . We also generated an annotation for the A . cephalotes genome using a combined approach of electronically-generated annotations followed by manual review and curation of a subset of gene models . Expressed Sequence Tags ( ESTs ) generated from a pool of workers consisting of different ages and castes from a laboratory-maintained colony of A . cephalotes was used in conjunction with the MAKER [32] automated annotation pipeline to generate an initial genome annotation . This electronically-generated annotation set ( OGS1 . 1 ) contained a total of 18 , 153 gene models encoding 18 , 177 transcripts ( See Materials and Methods ) , 7 , 002 of which had EST splice site confirmation and 7 , 224 had at least partial EST overlap . The MAKER-produced gene annotations were used for further downstream review and manual curation of over 500 genes across 16 gene categories ( Table S1 ) . Significant findings from this annotation are highlighted below , with additional details of our full analysis described in Text S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 , S12 , S13 , S14 , S15 , S16 , S17 , S18 , S19 , S20 ) . In addition to the A . cephalotes genome sequence , we also recovered an 18-20X coverage complete and circular mitochondrial genome , which showed strong whole sequence identity to the mitochondrial genome sequence reported for the solitary wasp Diadegma semiclausum [33] . A synteny analysis of the predicted genes on the A . cephalotes mitochondrial genome showed near-identical gene order with that of A . mellifera [34] ( Text S4 ) . The A . cephalotes assembly contains 80 Mbp of repetitive elements , which accounts for 25% of the predicted assembly ( Table S2 ) . The large majority of these are interspersed repeats , which account for 70 Mbp ( 21% ) . Many of these repeats are transposable elements ( TEs ) , with DNA TEs the most abundant and accounting for 14 . 3 Mbp ( 4 . 5% ) . A large number of retroid element fragments were also identified , with Gypsy/DIRS1 and L2/CR1/Rex as the most abundant . However , the majority of interspersed elements ( 51 . 8 Mbp ) were similar to de novo predictions that we could not be classified to a specific family ( Table S2 ) . Improvements to the assembly , integration of repeat annotation evidence , and manual curation will be necessary to determine if these elements represent new TE families or complex nests of interspersed repeats . Given the obligate association between A . cephalotes and its fungal cultivar , we investigated the possibility that the A . cephalotes genome might contain transposable elements commonly found in fungi . This was done by re-analyzing the genome using a TE library optimized for the detection of Fungi and Viridiplantae . We did not find evidence for any high-scoring or full-length retroid or DNA TEs from either of these taxa present in the A . cephalotes genome . Our estimate that 25% of A . cephalotes assembly contains repetitive elements may be ambiguous because our assembly spans 317 Mbp and the estimated genome size for A . cephalotes is 300 Mbp [30] . These predictions are , however , more similar to other ant species [27] and N . vitripennis [35] than to A . mellifera [28] , which lacks the majority of retroid elements and other transposable elements ( TE ) found in A . cephalotes . Eukaryotic genomes can be understood from the perspective of their nucleotide topography , particularly with respect to their GC content . Previous work has shown that animal genomes are not uniform , but are composed of compositional domains including homogeneous and nonhomogeneous stretches of DNA with varying GC composition [36] . A global composition analysis was performed for A . cephalotes and the compositional distribution was compared to those of other insect genomes , as described in Text S5 . This analysis revealed that A . cephalotes has a compositional distribution similar to other animal genomes , with an abundance of short domain sequences and few long domain sequences . A . cephalotes also has the largest number of long GC-rich domain sequences when compared to other insect genomes , with over six times the number of long GC-rich domain sequences than the N . vitripennis genome . When genes are mapped to compositional domains in the A . cephalotes genome , we find that they are uniformly distributed across the entire genome , in contrast to N . vitripennis and A . mellifera , which have genes occurring in more GC-poor regions of their genomes . The methylation of genes has been reported for other hymenopterans including A . mellifera [37] and N . vitripennis [35] . In insects , it is thought that this process contributes to gene silencing [37] , but recent reports suggest a positive correlation between DNA methylation and gene expression [38] , [39] . DNA methylation is thought to involve three genes: dnmt1 , dnmt2 , and dnmt3 [40] , although the precise role of dnmt2 remains unresolved . We found all three genes as single copies in A . cephalotes , which is similar to the other ants [27] but in contrast to A . mellifera and N . vitripennis where dnmt1 has expanded to two and three copies , respectively [35] ( Text S6 ) . Dnmt3 is known to be involved in caste development in A . mellifera [41] , and the presence of this gene in A . cephalotes may therefore indicate a similar role . RNA interference is a mechanism through which the expression of RNA transcripts is modulated [42] . We annotated a total of 29 different RNAi-related genes in A . cephalotes , including most of the genes involved in the microRNA pathway , the small interfering RNA pathway , and the piwi-interacting RNA pathway ( Text S7 ) . All detected RNAi genes were found as single copies except for two copies of the gene loquacious . One of these contains three double-stranded RNA binding domains characteristic of loquacious in D . melanogaster [43] , whereas the other contains only two . It is not known what role this second loquacious-like gene plays in A . cephalotes and future work is needed to deduce its role . The insulin signaling pathway is a highly-conserved system in insects that plays a key role in many processes including metabolism , reproduction , growth , and aging [44] . An analysis of the insulin signaling system in A . cephalotes reveals that it has all of the core genes known to participate in this pathway ( Text S8 ) . One of the hallmarks of A . cephalotes biology is its complex size-based caste system and , although virtually nothing is known about the genetic basis of caste development in this ant , it is currently thought that it is intrinsically linked to brood care and the amount of nutrients fed to developing larvae [1] . Given the importance of the insulin signaling system in nutrition , it is likely that this pathway is involved in caste differentiation in A . cephalotes , as has been shown for A . mellifera [45] . The yellow/major royal jelly proteins are encoded by an important class of genes and in A . mellifera they are thought to be integral to many major aspects of eusocial behavior [46] . For example , members of these genes are implicated in both caste development and sex determination . An analysis of this gene family in A . cephalotes revealed a total of 21 genes , 13 of which belong to the yellow genes and 8 of which encode major royal jelly proteins ( MRJP ) ( Text S9 ) . In general , the yellow genes display one-to-one orthology with yellow genes in other insects like Drosophila melanogaster and N . vitripennis . With eight members in the MRJP subfamily , which is restricted to Hymenoptera , the number of MRJP genes in A . cephalotes is similar to the number reported for other Hymenoptera [35] , [46] . However , five of the eight genes in A . cephalotes are putative pseudogenes . This may indicate that a high copy number of MRJPs may be an ancestral feature and that Atta is in the process of losing these genes . The loss of MRJPs may be a common theme among ants , as the recently reported genome sequences for C . floridanus and H . saltator revealed only one and two MRJP genes , respectively [27] . Wing polyphenism is a universal feature of ants that has contributed to their evolutionary success [1] . The gene network that underlies wing polyphenism in ants responds to environmental cues such that this network is normally expressed in winged queens and males , but is interrupted at specific points in wingless workers [47] . We therefore predict that the differential expression of this network between queens and workers may be regulated by epigenetic mechanisms as has been demonstrated in honey bees [41] . In A . mellifera , developmental and caste specific genes have a distinct DNA methylation signature ( high-CpG dinucleotide content ) relative to other genes in the genome [48] . Because A . cephalotes has more worker castes than other ant species [23] ( Figure 1C ) , we predict that the DNA methylation signature of genes underlying wing polyphenism will also be distinct relative to other genes in its genome . To test this prediction , we analyzed the sequence composition of wing development genes in A . cephalotes , and found that they exhibit a higher CpG dinucleotide content than the rest of the genes in the genome ( Text S10 ) . Previous experiments have shown that genes with a high-CpG dinucleotide content can be differentially methylated in specific tissues or different developmental stages [49] . Therefore , DNA methylation may facilitate the caste-specific expression of genes that underlie wing polyphenism in A . cephalotes . This may be a general feature of genes that underlie polyphenism . An important aspect of the eusocial lifestyle is communication between colony members , specifically in differentiating between individuals that belong to the same colony and those that do not . Nestmate recognition in many ants is mediated by cuticular hydrocarbons ( CHCs ) [50] , and nearly 1 , 000 of these compounds have been described . In ants , CHC biosynthesis involves Δ9/Δ11 desaturases , which are known to produce alkene components of CHC profiles [51] . We analyzed the Δ9 desaturases in the genome of A . cephalotes and detected nine genes localized to a 200 kbp stretch on a single scaffold in addition to four other Δ9 desaturase genes on other scaffolds ( Text S11 ) . In contrast , the seven genes found in D . melanogaster are more widely distributed along one chromosome . The number of Δ9 desaturase genes in A . cephalotes is similar to the 9 and 16 found in A . mellifera and N . vitripennis , respectively . A phylogenetic analysis of these genes supports their division into five clades , with eight Δ9 desaturase genes falling in a single clade suggesting an expansion of these genes possibly related to an increased demand for chemical signal variability during ant evolution ( Text S11 ) . Interestingly , the phylogeny also supports an expansion in this type of Δ9 desaturase genes within N . vitripennis but not in A . mellifera . All insects have innate immune defenses to deal with potential pathogens [52] and A . cephalotes is no exception with a total of 84 annotated genes found to be involved in this response ( Text S12 ) . These include the intact immune signaling pathways Toll , Imd , Jak/Stat , and JNK . When compared to solitary insects like D . melanogaster and N . vitripennis , A . cephalotes has fewer immune response genes and better resembles what is known for the eusocial A . mellifera [53] . The presence of other defenses in A . cephalotes , such as antibiotics produced by metapleural glands [54]–[56] , may account for the paucity of immune genes . Furthermore , social behavioral defenses may also participate in the immune response , as has been suggested for A . mellifera [53] . A set of shared orthologs was determined among A . cephalotes , A . mellifera , N . vitripennis , and D . melanogaster ( Figure 2 ) . A total of 5 , 577 orthologs were found conserved across all four insect genomes , with an additional 1 , 363 orthologs conserved across the three hymenopteran genomes . A further , 599 orthologs were conserved between A . cephalotes and A . mellifera , perhaps indicating genes that are specific to a eusocial lifestyle . We also found 9 , 361 proteins that are unique to A . cephalotes , representing over half of its predicted proteome . These proteins likely include those specific to ants or to A . cephalotes . We then analyzed the proteins that were found to be specific to A . cephalotes and determined those Gene Ontology ( GO ) [57] terms that are enriched in these proteins , relative to the rest of the genome ( Table S3 ) . We found many GO terms that reflect the biology of A . cephalotes and ants in general . For example , we find proteins with GO terms that reflect the importance of communication . These include proteins associated with olfactory receptor activity , odorant binding function , sensory perception , neurological development , localization at the synapse , and functions involved in ligand-gated and other membrane channels . To focus on Hymenoptera evolution , we compared the A . cephalotes genome to 4 other hymenopterans including the ants C . floridanus and H . saltator , the honey bee A . mellifera , and the solitary parasitic jewel wasp N . vitripennis . We used the eukaryotic clusters of orthologous groups ( KOG ) ontology [58] to annotate the predicted proteins from all of these genomes and performed an enrichment analysis by comparing the KOGs of the social insects A . cephalotes , C . floridanus , H . saltator , and A . mellifera against the KOGs of the non-social N . vitripennis as shown in Table S4 . A detailed analysis of KOGs within each over- and under-represented category is highly suggestive of A . cephalotes biology ( Table S5 ) . One of the most over-represented KOGs in A . cephalotes includes the 69 copies of the RhoA GTPase effector diaphanous ( KOG1924 ) . In contrast , all of the other hymenopteran genomes have substantially less copies of this gene . RhoA GTPase diaphanous is known to be involved in actin cytoskeleton organization and is essential for all actin-mediated events [59] . The large number of these genes in A . cephalotes may relate to the extensive cytoskeletal changes that occur during caste differentiation . One of these genes ( ACEP_00016791 ) was found to exhibit high single nucleotide polymorphism ( SNPs ) ( Text S13 ) . Given that genes involved in caste development in other social insects like A . mellifera also have high SNPs [60] , [61] , this may indicate that this gene is important for caste determination in A . cephalotes . A . cephalotes is also significantly over-represented in the dosage compensation complex subunit ( KOG0921 ) , the homeobox transcription factor SIP1 ( KOG3623 ) , the muscarine acetylcholine receptor ( KOG4220 ) , the cadhedrin EGF LAG seven-pass GTP-type receptor ( KOG4289 ) , and the calcium-activated potassium channel slowpoke ( KOG1420 ) , relative to N . vitripennis . Many of these genes have been implicated in D . melanogaster larval development , specifically during nervous system formation [62] , [63] . As a result , an over-representation of these genes in A . cephalotes relative to N . vitripennis may indicate their association with a eusocial lifestyle , and in particular , caste and subcaste differentiation . Genes that were found to be under-represented in A . cephalotes relative to N . vitripennis include core histone genes , nucleosome-binding factor genes , serine protease trypsins , and cytochrome P450s ( Table S5 ) . These findings were confirmed by a domain-based comparison between A . cephalotes and all other sequenced insects ( Text S14 ) . One of the most under-represented KOGs is trypsin , a serine protease used in the degradation of proteins into their amino acid constituents . Trypsins in N . vitripennis are known to be part of the venom cocktail injected into its host , which helps necrotization and initiates the process of amino acid acquisition for developing larvae [35] , [64] . In contrast to the protein-rich diet of N . vitripennis , A . cephalotes feed on gongylidia produced by their fungus , which represents a switch to a carbohydrate-rich ( 60% of mixture ) diet [65] . These differences in diet may explain the under-representation of trypsin in A . cephalotes , as trypsin is likely not the primary mechanism used to digest nutrients obtained from the fungal cultivar . Our analysis also revealed a reduction of trypsin genes in the other social insects relative to N . vitripennis , and this may also reflect their diets . For example , honey dew is a major component of the diet of C . floridanus and contains primarily sugars [1] , while the honey/pollen diet of A . mellifera is composed primarily of carbohydrates , lipids , carbohydrates , vitamins , and some proteins [66] . Because this under-representation of trypsin is consistent across social insects when compared to other sequenced insects ( Table S5 , Text S14 ) , this reduction may reflect the specific dietary features of these insects , or could indicate a loss of these genes across eusocial insects . In addition to trypsin , cytochrome P450s were also found to be under-represented in both A . cephalotes and A . mellifera , relative to N . vitripennis , with reductions in both CYP3- and CYP4-type P450s ( Table S5 ) . P450s in insects are important enzymes known to be involved in a wide range of metabolic activities , including xenobiotic degradation , and pheromone metabolism [67] . We identified a total of 52 and 62 P450s in A . cephalotes and A . mellifera , respectively , which is similar to the low numbers reported for another insect , the body louse Pediculus humanus [29] . These values represent some of the smallest amounts of P450s reported for any insect genome , and may represent the minimal number of P450s required by insects to survive . Comparison of the A . cephalotes P450s against those of A . mellifera and P . humanus reveals that while there are some shared P450s , many are specific to each insect ( 15 ) . In A . mellifera , the paucity of P450s is thought to be associated with the evolutionary underpinnings of its eusocial lifestyle [68] , although an enrichment of P450s in the ants C . floridanus and H . saltator [27] would seem to contradict this prediction . It is therefore unclear why A . cephalotes has a small number of P450s relative to other ants , and future work will be necessary to provide insight into this apparent discrepancy . A SNP analysis of the P450 genes in A . cephalotes did reveal that one of these , ACEP_00016463 , has 20 SNPs/kbp ( Text S13 ) . Since P450s are known to undergo accelerated duplication and divergence [67] , the high number of SNPs in this particular P450 may reflect positive selection for new functions . Given the tight obligate association that A . cephalotes has with its fungal mutualist , one might predict that it acquires amino acids from its fungus in a manner similar to that of the pea aphid Acyrthosiphon pisum , which obtains amino acids from its bacterial symbionts [28] . To test this , we performed a metabolic reconstruction analysis using the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) [69] . A . cephalotes contains a nearly identical set of amino acid biosynthesis genes as A . mellifera , C . floridanus , H . saltator , and N . vitripennis , all of which are incapable of synthesizing histidine , isoleucine , leucine , lysine , methionine , phenylalanine , threonine , tryptophan , and valine de novo . The only exception is arginine , and only A . cephalotes was found to lack the genes necessary for its biosynthesis ( Figure 3 ) . Arginine , which is produced through the conversion of citrulline and aspartate [70] , [71] , is predicted to be synthesized at levels too low to support growth in insects [72] . In A . cephalotes the 2 genes that catalyze the synthesis of arginine , argininosuccinate synthase ( EC 6 . 3 . 4 . 5 ) and argininosuccinate lyase ( EC 4 . 3 . 2 . 1 ) , were not found ( Figure 3 ) . The loss of these two genes suggests a dependence on externally-acquired arginine , which we hypothesize , is provided by their fungus . In the carpenter ant C . floridanus , arginine is thought to be synthesized from citrulline provided by its endosymbiont Blochmannia floridanus [73] , and this dependency is predicted to play an essential role in maintaining the carpenter ant-bacteria mutualism . An extreme case has been reported for the pea aphid , which has lost its urea pathway and depends entirely on its endosymbiont , Buchnera aphidicola , for arginine [28] . The loss of arginine biosynthesis in Atta may similarly be important for maintaining the leaf-cutter ant-fungus mutualism . In line with this prediction , the fungus the ants cultivate contains all of the amino acids that A . cephalotes can not synthesize , including arginine [65] . In addition to arginine biosynthesis , A . cephalotes may have also lost the need to rely on hexamerins as a source of amino acids during development . In many insects , hexamerin proteins are synthesized by developing larvae and used as amino acid sources during development into the adult stage [74] . Four hexamerins are commonly found across insects , including hex 70a , hex 70b , hex 70c , and hex 110 . Comparison among the hymenopteran genomes reveals the presence of all hexamerins in varying copy number across all genomes except for A . cephalotes , which is missing hex 70c ( Figure 4 ) ( Text S16 ) . In A . mellifera , hexamerins are expressed at different times , with hex 70a and hex 110 expressed during the larval , pupal and adult stage of workers , and hex 70b and hex 70c only expressed during the larval stage [74] . The specific expression of hex 70b and hex 70c in larvae may reflect the increased need for these nutrients during early development . Given that A . cephalotes larvae feed primarily on gongylidia , it is possible that amino acids supplemented by the fungus over the millions of years of this mutualism has relaxed selection for maintaining larval-stage hexamerins , and thus hex 70c may have been lost . Future expression analyses of these genes at different life stages , in different castes , and under different nutritional conditions will likely confirm and elucidate their role . Here we have presented the first genome sequence for a fungus-growing ant and show that its genomic features potentially reflect its obligate symbiotic lifestyle and developmental complexity . An initial analysis of its genome reveals many characteristics that are similar to both solitary and eusocial insect genomes . One hypothesis , based on the obligate mutualism of Atta cephalotes and its fungus , is that its genome exhibits reductions related to this relationship . We have provided some evidence that A . cephalotes has gene reductions related to nutrient acquisition , and these losses may be compensated by the provision of these nutrients from the fungus . For example , the extensive reduction in serine proteases may reflect the lack of proteins in its diet since the fungus primarily provides nutrients in the form of carbohydrates and free amino acids . Furthermore , the loss of the arginine biosynthesis pathway in A . cephalotes may indicate the obligate reliance that it has on the fungus , as arginine is part of the nutrients that it provides to the ant . This type of relationship appears to be conserved in other insect-microbe mutualisms , specifically in the pea aphid [28] and the carpenter ant [73] . Finally , A . cephalotes appears to have lost a hexamerin protein that is conserved across all other insect genome sequences reported to date . Loss of this protein , which is associated with amino acid sequestration during larval development , may be tolerated because larvae have a ready source of amino acids from the fungus . These genomic features may serve as essential factors that have stabilized the mutualism over its coevolutionary history . The sequencing and analysis of this genome will be a valuable addition to the growing number of insect genomes , and in particular will provide insight into both host-microbe symbiosis and eusociality in hymenopterans .
Three males from a single mature Atta cephalotes colony were collected in June 2009 in Gamboa , Panama ( latitude 9° 7′ 0″ N , longitude 79° 42′ 0″ W ) and designated males A , B , and C . Genomic DNA from these males was extracted using a modified version of a Genomic-tip extraction protocol for mosquitoes and other insects ( QIAGEN , Valencia , CA ) . Sequencing was performed using the 454 FLX Titanium pyrosequencing platform [25] at the 454 Life Sciences Sequencing Center ( Branford , CT ) as follows . A whole-genome shotgun fragment library was constructed for male A and sequenced using a single run , generating 539 , 113 , 701 bp of sequence . For male B , a whole-genome shotgun fragment library was also constructed and sequenced using 11 runs , generating a total of 4 , 209 , 396 , 304 bp of sequence . An 8 kbp insert paired-end library was also generated for male B and sequenced using two runs , generating a total of 818 , 851 , 400 bp of sequence . A 20 kbp paired-end library was generated for male C , and sequenced using a single run , generating 349 , 435 , 001 bp . In total , 5 , 916 , 796 , 406 bp of sequence were generated for all three ants . All generated sequences were assembled using the 454 GS de novo assembler software ( March 06 2010 R&D Release ) . The Atta cephalotes whole genome shotgun project has been deposited at DDBJ/EMBL/GenBank under the project number 48117 and accession ADTU00000000 . The version described in this paper is the first version , ADTU01000000 . Workers from a healthy Atta cephalotes colony ( JS090510-01 ) collected from Gamboa , Panama and maintained in the laboratory of Cameron Currie at the University of Wisconsin-Madison were used to generate transcript sequences . A pool of 169 workers across different age and size classes was selected and total RNA was extracted using a modified version of a phenol-chloroform protocol previously described [75] . This sample was normalized and a fragment library was generated before subsequent sequencing using a single run of a 454 FLX Titanium pyrosequencer [25] at the Genome Center at Washington University ( St . Louis , MO ) , generating a total of 462 , 755 , 799 bp of sequence . Transcript sequences were assembled using the Celera assembler ( wgs-assembler 6 . 0 beta ) [76] with standard assembly parameters . Annotations for the Atta cephalotes genome was generated using the automated genome annotation pipeline MAKER [32] . The MAKER annotation pipeline consists of 4 general steps . First , RepeatMasker ( http://www . repeatmasker . org ) and RepeatRunner [77] were used to identify and mask repetitive elements in the genome . Second , gene prediction programs including Augustus [78] , Snap [79] , and GeneMark [80] were employed to generate ab-initio ( non-evidence informed ) gene predictions . Next a set of expressed sequence tags ( ESTs ) and proteins from related organisms were aligned against the genome using BLASTN and BLASTX [81] , and these alignments were further refined with respect to splice sites using the computer program Exonerate [82] . Finally , the EST and protein homology alignments and the ab-initio gene predictions were integrated and filtered by MAKER to produce a set of evidence informed gene annotations . This gene set was then further refined to remove all putative repeat elements and to include gene models initially rejected by MAKER but found to contain known protein domains using the program InterProScan [83] . The resulting gene set ( OGS 1 . 1 ) then became the substrate for further analysis and manual curation . Over 500 genes in OGS 1 . 1 were manually curated ( Table S1 ) , producing OGS 1 . 2 , which is publicly available at the Hymenopteran Genome Database ( http://HymenopteraGenome . org/atta/genome_consortium ) . The general manual curation process used for generating OGS 1 . 2 was based on a standardized protocol and conducted as follows . For each gene family , query sequences were obtained first from FlyBase [84] and supplemented with known gene models from the other sequenced hymenopteran genomes , Apis mellifera [26] and Nasonia vitripennis [35] . BLAST was used to align these gene models against putative sequences in the A . cephalotes genome predicted by MAKER . The sequence analysis program Apollo [85] was then used by all annotators to contribute their annotations to a centralized Chado [86] database . In general , putative gene models in A . cephalotes were confirmed by investigating the placement of introns and exons , the completeness of sequences , evaluating sequencing errors , and syntenic information . A final homology search was also performed with the putative A . cephalotes gene model by comparing it against the non-redundant protein database in NCBI to confirm its match against known insect models . We used PILER-DF [87] , RepeatModeler ( Smit A , Hubley R , Green P . RepeatModeler Open-1 . 0 . 2008-2010 http://www . repeatmasker . org ) , RECON [88] , and RepeatScout [89] to generate de novo transposable element ( TE ) predictions . We found 1 , 381 de novo repeat predictions including 264 from RepeatScout , 26 from PILER-DF , and 1091 from RECON . We simplified the complexity of our de novo TE predictions by removing elements that were over 80% similar over 80% of their length [90] and also screened out elements with more than 50% sequence identify to Uniprot [91] genes . This resulted in a final A . cephalotes-specific repeat library containing 1 , 252 elements ( 1048 RECON , 195 RepeatScout , 9 PILER-DF ) , which were then classified using RepeatMasker ( Smit AFA , Hubley R , Green P . RepeatMasker Open-3 . 0 . 1996-2010 <http://www . repeatmasker . org> ) and custom scripts that identify TIR and LTR sequences . This curated library was converted to EMBL format , appended to a RepBase [92] library and used to mask the A . cephalotes genome assembly . An orthology analysis was performed between the proteins from Atta cephalotes ( OGS1 . 2 ) , Apis mellifera ( preOGS2 ) [26] , Nasonia vitripennis ( OGSI 1 . 2 ) [35] , and Drosophila melanogaster ( Release 5 . 29 ) [93] . Using these protein sets , we reduced each dataset to contain only the single longest isoform using custom Perl scripts . An all-by-all BLAST was performed using the computer program OrthoMCL [94] and the best reciprocal orthologs , inparalogs , and co-orthologs were determined . We used the MCL v09-308 Markov clustering algorithm [95] to define final ortholog , inparalog , and co-ortholog groups between the datasets . For all OrthoMCL analyses , the suggested parameters were used . We then annotated those proteins in A . cephalotes that did not have any orthologs to the 3 other insects and performed a gene ontology enrichment analysis . This was done by annotating all A . cephalotes proteins using Interproscan [83] to generate Gene Ontology ( GO ) [57] terms . This resulted in 6 , 971 ( 41% ) proteins receiving at least one GO annotation . GO-TermFinder [96] was then used to determine those proteins that were enriched for specific GO terms in the A . cephalotes-specific proteins , relative to the entire A . cephalotes OGS1 . 2 dataset . We performed a eukaryotic orthologous groups ( KOG ) [58] enrichment analysis for the genomes of Atta cephalotes , Camponotus floridanus [27] , Harpegnathos saltator [27] , Apis mellifera [26] , and Nasonia vitrepennis [35] . The KOG database was obtained from NCBI and RPSBLAST [97] ( e-value: 1e-05 ) was used to compare the predicted proteins from A . cephalotes ( OGS1 . 2 ) , C . floridanus ( OGS3 . 3 ) , H . saltator ( OGS3 . 3 ) , A . mellifera ( preOGS2 ) , and Nasonia vitripennis ( OGS r . 1 ) . Each KOG hit was tabulated according to its gene category , and Fisher's exact test was then applied to determine which categories were over- or under-represented . This was done for A . cephalotes , C . floridanus , H . saltator , and A . mellifera against N . vitripennis , respectively , as shown in Table S4 . We then determine for each over- and under-represented KOG category in A . cephalotes relative to N . vitripennis , the specific KOGs within each category that were significantly enriched or under-enriched . This was done by comparing the total number of A . cephalotes KOGs within each of these categories against those in N . vitripennis using Fisher's exact test , as shown in Table S5 . The predicted peptides for Atta cephalotes were used to reconstruct putative metabolic pathways using the Kyoto Encyclopedia of Genes and Genomes [69] . This was performed using the KEGG Automated Annotation Server ( KAAS ) , which annotates proteins according to the KEGG database and reconstructs full pathways displaying them as maps . Similar maps were also constructed using KAAS for the predicted peptide sequences of Camponotus floridanus ( OGS3 . 3 ) and Harpegnathos saltator ( OGS3 . 3 ) . These maps were compared against the maps currently available in KEGG for Apis mellifera , Drosophila melanogaster , and Nasonia vitripennis . For proteins in A . cephalotes that were not found in our KEGG reconstruction analysis , relative to other insects ( e . g . argininosuccinate synthase ( EC 6 . 3 . 4 . 5 ) and argininosuccinate lyase ( EC 4 . 3 . 2 . 1 ) ) , we investigated those reads that were not incorporated into the A . cephalotes assembly to confirm that these did not contain potential gene fragments corresponding to these genes .
|
Leaf-cutter ant workers forage for and cut leaves that they use to support the growth of a specialized fungus , which serves as the colony's primary food source . The ability of these ants to grow their own food likely facilitated their emergence as one of the most dominant herbivores in New World tropical ecosystems , where leaf-cutter ants harvest more plant biomass than any other herbivore species . These ants have also evolved one of the most complex forms of division of labor , with colonies composed of different-sized workers specialized for different tasks . To gain insight into the biology of these ants , we sequenced the first genome of a leaf-cutter ant , Atta cephalotes . Our analysis of this genome reveals characteristics reflecting the obligate nutritional dependency of these ants on their fungus . These findings represent the first genetic evidence of a reduced capacity for nutrient acquisition in leaf-cutter ants , which is likely compensated for by their fungal symbiont . These findings parallel other nutritional host–microbe symbioses , suggesting convergent genomic modifications in these types of associations .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"organismal",
"evolution",
"evolutionary",
"ecology",
"genome",
"evolution",
"genome",
"sequencing",
"coevolution",
"genome",
"complexity",
"forms",
"of",
"evolution",
"comparative",
"genomics",
"biology",
"evolutionary",
"genetics",
"animal",
"evolution",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"genetics",
"and",
"genomics"
] |
2011
|
The Genome Sequence of the Leaf-Cutter Ant Atta cephalotes Reveals Insights into Its Obligate Symbiotic Lifestyle
|
Codon models of evolution have facilitated the interpretation of selective forces operating on genomes . These models , however , assume a single rate of non-synonymous substitution irrespective of the nature of amino acids being exchanged . Recent developments have shown that models which allow for amino acid pairs to have independent rates of substitution offer improved fit over single rate models . However , these approaches have been limited by the necessity for large alignments in their estimation . An alternative approach is to assume that substitution rates between amino acid pairs can be subdivided into rate classes , dependent on the information content of the alignment . However , given the combinatorially large number of such models , an efficient model search strategy is needed . Here we develop a Genetic Algorithm ( GA ) method for the estimation of such models . A GA is used to assign amino acid substitution pairs to a series of rate classes , where is estimated from the alignment . Other parameters of the phylogenetic Markov model , including substitution rates , character frequencies and branch lengths are estimated using standard maximum likelihood optimization procedures . We apply the GA to empirical alignments and show improved model fit over existing models of codon evolution . Our results suggest that current models are poor approximations of protein evolution and thus gene and organism specific multi-rate models that incorporate amino acid substitution biases are preferred . We further anticipate that the clustering of amino acid substitution rates into classes will be biologically informative , such that genes with similar functions exhibit similar clustering , and hence this clustering will be useful for the evolutionary fingerprinting of genes .
Modern molecular evolution has benefited greatly from the development of a sound probabilistic framework for modeling the evolution of homologous gene sequences [1] . In particular , codon substitution models [2] , [3] have facilitated the estimation of the ratio of non-synonymous to synonymous substitution rates ( referred to as ) , which can be interpreted as an indicator of the strength and type of natural selection ( see [4] or [5] for recent reviews ) . Codon models are fundamentally mechanistic because they use the structure of the genetic code to partition codon substitutions into classes . Initially , and in most subsequent applications of codon models , all one-nucleotide substitutions were stratified into synonymous ( rate , using the notation of [2] ) and non-synonymous ( rate ) classes . Despite several early attempts , e . g . [3] , none of the widely-adopted codon models incorporated physicochemical properties of the two residues being exchanged . In contrast , most protein substitution models are derived by estimating the relative rates of amino-acid substitutions in large protein databases [6]–[8] , and consistently report dramatic differences in the relative replacement rates of different residues . The persisting dissonance between how codon and protein models approach amino acid substitution rates has fostered multiple recent efforts to develop what we will call multi-rate codon models ( or more accurately , multi- nonsynonymous rate models ) , in contrast to the existing single-rate model . These models divide amino acid pairs ( or codon pairs ) into multiple rate categories , such that every category has its own rate which governs substitutions between the pairs in that category . In the most extreme case , every amino acid or codon pair belongs to a different category and thus has its own rate – potentially leading to a very large number of parameters that need to be estimated . Several strategies have been proposed for limiting the number of parameters in multi-rate models . Doron-Faigenboim et al . [9] proposed to overlay existing empirically derived amino acid substitution matrices ( e . g . [7] or [8] ) onto single-rate codon models by weighted partitioning of the empirical rate of substitution between two protein residues . Kosiol , Holmes & Goldman [10] directly estimated all codon-to-codon substitution rates in an empirical codon model – a codon equivalent of the nucleotide GTR model [11] , assuming the universal genetic code . However , this effort required a truly massive training dataset encompassing alignments from protein families of the Pandit database [12] . The resulting empirical codon model ( ECM ) encodes evolution patterns averaged over many proteins . However , no single empirically-derived substitution rate matrix appears to be generalizable across multiple genes and taxonomic groups , as evidenced by a plethora of specialized substitution models , e . g . for mammalian mitochondrial genomes [13] , plant chloroplast genes [14] , viral reverse transcriptases [15] or HIV-1 genes [16] . More mechanistic parameters can be introduced to improve biological realism of codon-models . The linear combination of amino acid properties ( LCAP ) model [17] expresses exchangeability of a pair of codons as an ( exponentiated ) linear combination of differences in five independently validated amino acid physicochemical properties . This parameterization incorporates weighting ( or importance ) coefficients inferred from the data to allow for differences in protein evolution between genes , shown to be significant and biologically meaningful in yeast proteins [18] , and once again underscoring the utility of gene-specific evolutionary models . All multi-rate codon models published to date have shown clear improvements in model fit over the single-rate model . However , multi-rate models in which substitutions were randomly assigned to classes easily outperform the single-rate model [19] and thus it is a poor performance benchmark . At the other extreme of model space is the full time-reversible codon model , with parameters ( or , if only single nucleotide substitutions are modeled ) , which will certainly suffer from massive over-fitting on single gene alignments . Over-parameterization can be reduced by “smoothing” , i . e . by grouping the rates into exchangeability classes based on the physicochemical properties of amino acids [20] . However , without a rigorous model selection framework , it is difficult to ascertain how well any particular smoothing approach fits the data . To appreciate how large the space of potential models is , consider that there are are approximately possible multi-rate codon models with nonsynonymous rate classes , and approximately possible models for . Given such a large search space it is impossible to evaluate even a small fraction of possible models exhaustively , and one cannot presume that any given model or a small set of models are sufficiently representative without exploring the alternatives . Huelsenbeck et al . [21] examined a Bayesian approach to estimate empirical amino acid substitution models in which amino acid exchangeability classes are assigned using a Dirichlet process . However , a prior distribution needs to be specified for the number of classes ( = 2 , 5 , or 10 ) , and mechanistic features of codon evolution are excluded . Models which combine empirical codon models and mechanistic parameters , such as and transition-transversion bias [10] , have been shown to outperform the models which include only a single effect . This evidence highlights the necessity to model both mutational effects , which result in substitution preferences for particular amino acids , and selective effects , the result of fitness differences of alternate phenotypes . In this manuscript , we present an information-theoretic model selection procedure that extends the concept of ModelTest [22] , formulated for nucleotide model selection , to codon models . Unlike ModelTest , which examines a priori defined models , we use a Genetic Algorithm ( GA ) to search the combinatorially large set of codon models ( i . e . select the number of rate classes ) , to assign amino acid substitution rates to these classes , infer rate parameters and , finally , report a set of credible models given the data . Our group has successfully applied GAs to a variety of problems in evolutionary biology , including inference of lineage-specific selective regimes [23] , detecting recombination in homologous sequence alignments [24] , and model selection for paired RNA sequences [25] , where the GA was able to recover biologically relevant properties and outperformed all known mechanistic models . Using simulated data , we demonstrate that GA model selection ( under a sufficiently stringent model selection criterion ) is not susceptible to over-fitting , and that codon alignments of typical size contains sufficient signal to reliably allocate non-synonymous substitutions into a small number of rate classes , typically . On empirical data sets , GA-selected codon substitution models consistently outperformed published empirical and mechanistic models . In addition to selecting a single best fitting model , the GA also estimates a set of credible models for an alignment . A weighted combination of models in the credible set enable model averaged phylogenetic [26] and substitution rate matrix [25] inference and further reduces the risk of over-fitting . We anticipate that improvements in model realism will translate into improved sequence alignment , phylogeny estimation , and selection detection . Moreover , we hypothesize that the clustering of non-synonymous substitution rates into groups with the same rate parameter is shared by genes with similar biological and structural properties , and hence this clustering is informative for improving evolutionary fingerprinting of genes [27] .
Models considered in this paper assume that codon substitutions along a branch in a phylogenetic tree can be described by an appropriately parameterized continuous-time homogeneous and stationary Markov process; an assumption ubiquitous in codon-evolution literature . The substitution process is uniquely defined by the rate matrix , , whose elements denote the instantaneous substitution rate from codon to codon . Using to label the amino-acid encoded by codon , and assuming a universal genetic code with three stop codons ( other codes can be handled with obvious modifications ) , matrix comprises 61×61 such elements , where ( 1 ) Here , denote equilibrium frequency parameters , denote nucleotide mutational biases , and denote the substitution rates between amino acids encoded by codons and . How to infer is the primary focus of this paper . We consider two different parameterizations of : the GY parameterization [3] , where is the equilibrium frequency of the target codon , and the MG parameterization [2] , where is a nucleotide frequency parameter for the position that is being substituted ( ; ) . For the GY parameterization , we estimate codon equilibrium frequencies by their proportions in the data ( the F61 estimator , 60 parameters for the universal genetic code ) . For the MG parameterization , we estimate the nine frequency parameters by maximum likelihood [28] . The equilibrium frequency of codon can then be computed aswhere and . Finally , we set , and estimate other rates ( ) by maximum likelihood; this parameterization follows the MG94REV model from [29] . By varying the parametric complexity of the non-synonymous substitution rate encoding in equation ( 1 ) , we can span the range of models from the single rate model ( SR , current default standard , 1 non-synonymous rate parameter ) , to the general codon time-reversible model ( REV ) with each amino-acid pair substitution exchanged at its own rate . Only 75 out of 190 total amino-acid pairs can be exchanged via a single nucleotide substitution , for example and are one such pair , but and are not . Consequently , the REV model has non-synonymous rate parameters . The purpose of our study is to explore the model space between these two extremes , taking into account the limitations of information content in single gene alignments . Note that most existing multi-rate models can be represented with an appropriate choice of in equation ( 1 ) . Empirical models ( e . g . ECM ) replace with numerical values estimated from large training data sets , whereas mechanistic models ( e . g . LCAP ) assume that rates can be modeled via a function measuring differences/similarities in physicochemical properties of residues ( Table 1 ) . We focus on structured ( or rate clustering ) models: those which assume that substitution rates can be partitioned/structured into classes , where each class has a single estimated rate parameter . These structured models may be defined using amino acid similarity classes [30] , but instead of adopting a priori classes of rates , we propose to infer their number and identity from the data . A structured model with substitutions ( e . g . for the Universal genetic code ) in classes can be represented as a vector of length , where each element is an integer between and labeling the class . For example if the vector entries corresponding to , and substitutions have values and , then and . As an analogy , the HKY85 nucleotide model [31] is a structured model with vector , , where the substitutions between 6 nucleotide pairs ( indicated by a subscript ) are placed into transition ( 1 ) and transversion ( 0 ) classes . Given the structure of a codon model , e . g . , it can be fitted to the data using standard maximum likelihood phylogenetic algorithms , e . g . as implemented in HyPhy [32] . The resulting set of rate estimates instantiate a structured model and induce a corresponding empirical model , e . g . . Because the space of structured codon models is combinatorially large , we utilize a GA previously used to solve an analogous model selection problem for paired RNA data [25] . Parameter space is defined by two components: a discrete component which assigns pairwise non-synonymous substitutions between codons to rate classes using the structured vector described above , and a continuous component comprising a vector of branch lengths , nucleotide substitution rates , frequency parameters and non-synonymous rates . The discrete component is optimized by the GA , while the continuous component is estimated using numerical non-linear optimization procedures , given the structure of the model . We initially approximate branch lengths using the SR model and update them whenever the GA iteration improves the fitness score by more than 50 points ( see below ) as compared to the most recent model for which branch lengths have been estimated . Further details of the genetic algorithm are described in detail in [25] , and for the sake of brevity we do not present it here . We are left with the problem of inferring the number of rate classes . This is done by starting with and iteratively proposing to increment . For each proposal , the model with rate classes is optimized using the optimized -class model as initialization . If the proposal results in a model with a better fitness value ( see below ) , it is accepted and a new proposal generated . The process terminates when the -class proposal does not beat the -class model . We initially assigned a fitness value to each model using where is the sample size and is the number of parameters in the model [33] . The “sample size” of a sequence alignment is difficult to quantify with a single number , since it depends on both the number of sequences in the alignment and the lengths of those sequences . We use the number of characters to approximate “sample size” to make the model selection criterion maximally conservative . While it is straightforward to count the number of estimated parameters in any given structured model , setting to that number leads to model over-fitting ( results not shown ) , because the topological component ( the assignment of rates to classes ) adds further “degrees of freedom” to the model . To determine the appropriate penalty term , we conducted simulations; there is precedent for this in statistical literature on generalized information criteria ( e . g . [34] ) . We removed the effect of phylogeny by simulating nine sets of two-sequence alignments ( divergence ) : each set of simulations consisted of replicates with between and codons ( in increments ) . The sets had 1 to 5 rate classes ( Figure 1 ) , representing rate classification problems that ranged from easy ( large numerical differences between class rates , e . g . and ) to difficult ( small numerical differences , e . g . and ) . We constructed generating multi-rate models by assigning rates to bins randomly with equal probability . For each simulation set we plotted the difference in log likelihood ( scaled by the sample size = log of characters ) between the correct model ( rates ) , and models with and rates , respectively . Simulations indicated that doubling the number of parameters in the BIC penalty term ensured sufficient power , and controlled false positives for all simulation sets ( Figure 1 ) . We used this modified BIC , to assign fitness to every model examined by a GA run and select those with the lowest . We also simulated realistic “gene-size” alignments on and taxon trees . Nucleotide frequencies were uniform ( ) for each position , and the nucleotide bias component was set to HKY85 with transition/transversion ratio , . We generated data sets for each :rate vector combination , under the single rate , and a fixed Random-K model ( Table 2 ) . These data allowed us to assess the performance of the model when the true underlying model was known . For each simulation scenario , we report the proportion of replicates for which the GA inferred the correct number of rate classes , the proportion of underfitted replicates ( too few rate classes were inferred ) and the proportion of overfitted replicates ( too many rate classes were inferred ) . For the replicates where the correct number of rate classes was inferred , we computed the Rand statistic ( , [35] ) on the generating and inferred model structures to quantify the similarity between two clusterings rates . The Rand statistic quantifies the similarity between two clusterings ( ) of the same set of objects and can be defined as , where is the number of objects ( pairs of substitution rates ) that belong to different classes in both A and B , ( ) is the number of objects that belong to different ( same ) classes in A , but the same ( different ) class in B , and is the number of objects that belong to the same class in both A and B . Clearly , for perfect agreement ( ) and for perfect disagreement ( ) . We prepared a collection of reference empirical data sets ( see Table 3 ) , to be used for benchmarking GA , published and extreme-case models . The collection included three protein family alignments from Pandit [12] selected randomly from all alignments with taxa , a randomly selected Yeast protein alignment [18] , a group M HIV-1 pol alignment [36] and an Influenza A virus ( IAV ) HA alignment comprising H3N2 , H5N1 , H2N2 and H1N1 serotypes . The latter was assembled by random selection of post-2005 sequences for each serotype from the NCBI Influenza database [37] . Finally , we examined the vertebrate rhodopsin protein , recently analyzed for molecular mechanisms of phenotypic adaptation by [38] . We inferred a structured multi-rate model for each of these data sets using the genetic algorithm and model fitness function defined above . A comparison of the GA-fitted model against existing models is unfair , since the former was selected among a set of candidate models using the test alignment . To confirm that GA models were generalizable , we evaluated the fit of the GA models and that of existing models for both the reference datasets , and independent test alignments for the same taxonomic groups ( validation data sets ) . Two HIV-1 pol gene alignments were obtained for subtypes B [39] and C [40] . Subtype assignments were confirmed using the SCUEAL sub-typing tool [36] , and inter- and intra-subtype recombinants were pruned from the analysis . For IAV HA we used independent alignments for serotypes H5N1 and H3N2 , filtered from the NCBI Influenza database [37] , and from [41] , respectively . We fitted five reference models to each dataset: ( i ) the single-rate model , ( ii ) a Random- and a Random- model , ( iii ) the empirical codon model ( ECM , [10] ) , ( iv ) the Linear Combination of Amino Acid Properties ( LCAP ) model [17] , [18] , and ( v ) the reversible ( REV ) model ( see Table 1 ) . For every dataset , the corresponding GA-run was processed to obtain three different alignment-specific multi-rate models . We used both BIC [33] and Likelihood ratio tests , where appropriate , for model comparison . These goodness-of-fit comparisons allowed us to evaluate whether a model estimated on reference alignments yielded a significant improvement over the other models when fitted to independent alignments for the same taxonomic groups . All models were implemented with the F61 frequency parameterization , in addition to their original frequency parameterizations , because the methodology used to estimate the ECM model precluded the use of other frequency parameterizations for across-the-board comparison . Alignments and phylogenetic trees were provided for the Pandit data set . In all other cases , alignments were generated using codon alignment tools implemented in HyPhy [32] . Maximum likelihood phylogenetic trees were estimated using PhyML [43] under a GTR [44] model of nucleotide substitution and among-site rate variation modeled as a discretized gamma distribution with 4 rate-classes [45] . Empirical alignments and trees are available at http://www . hyphy . org/pubs/cms/ . The entries of the substitution rate matrix can be used to estimate the expected number of substitutions per site per unit time , , and to determine the value of the time parameter ( assuming all other parameters are known ) which yields . Furthermore , the expression for the number of expected one-nucleotide substitutions between codons and , in time , at a site is given by ( the simplification is the consequence of time-reversibility ) . Given two amino-acid residues and which can be exchanged by a single nucleotide substitution , we can further define , where denotes the residue encoded by codon . Consider a element substitution spectrum vector , which describes the relative abundance or paucity of a particular type of amino-acid pair substitution under the model defined by . Given two models , and , we propose to compare their similarity by computing the distance between the corresponding substitution spectrum vectors evaluated at the corresponding “normalized” times: ( 2 ) Any norm on the standard dimension real valued vector space can be used , but for the purposes of this paper we consider the norm , and the corresponding induced Euclidean distance metric . All models and data sets utilized in this study are implemented as scripts in the HyPhy Batch Language ( HBL ) , and are be available with the current source release of HyPhy [32] . In addition , we have made the GA codon model selector available as an analysis option at http://www . datamonkey . org [46] . The GA model selection code requires an MPI cluster environment with typical runtimes of approximately 36–48 hours for an intermediate-sized alignment ( 50 taxa ) and 32 compute nodes .
Results from both two- and multi-taxon simulations ( Table 2 , Figure 1 ) indicated that controlled the rates of overfitting , defined as the proportion of replicates that overestimated the number of rate classes , . For null ( single-rate model ) simulations ( ) , false positive rates were for two-taxon simulations and for -taxon simulation . Neither two- nor multi-taxon simulations showed over-fitting across any simulation scenarios ( Table 2 ) . We deliberately designed the procedure to be conservative , since over-fitting is a major concern in statistical model selection . The power to select the correct number of rate classes ( ) behaved as expected: increasing , and eventually reaching , given sufficiently divergent sequences and well resolved rate classes ( Table 2 ) . Indeed , the limited information content of alignments where simulated rate classes are similar ( i . e 3 rates of ) , and/or where pairwise sequence divergence is low ( 0 . 2 ) , was evident as increased model under-fitting ( Table 2 ) , . Model under-fitting was substantially reduced when information content was increased , either by boosting the disparity in rate classes , or by elevating sequence divergence and/or number of taxa ( Table 2 ) . Further evidence that the GA procedure has high power is provided by the positive association of the difference between scores of the correct model with rates , and one with -1 rates , and separation between simulated rates , pairwise sequence divergence or number of taxa ( Table S1 ) . The ability to assign individual rates to the correct group ( as measured by the Rand statistic ) was similarly improved , while the variance in numerical rate parameter estimates decreased , for more divergent sequences and rate classes , suggesting that the GA search procedure recaptures most of the rate class structure , given sufficient information . We compared the fit of codon substitution models ( Table 1 ) on empirical data sets ( Table 3 ) , spanning a range of proteins , taxonomic groups and divergence levels , using the BIC to measure goodness-of-fit . Using the GA procedure , we inferred distinct multi-rate models from of these data sets ( labelled with asterisks in Table 3 ) . The remaining alignments were used for validation such that we could determine the generalizability of two of the GA-fitted models ( HIV and IAV ) to other alignments from the same taxonomic groups . In cases , the GA model outperforms every other model ( often by a large margin ) , and in cases it comes in second after the parameter rich REV model ( Table 4 ) . Note that the GA model outperforms REV in all cases under the more conservative criterion ( which was used to inform the GA ) . Data set specific GA models consistently fit the data better than state-of-the-art empirical ( ECM ) and mechanistic ( LCAP ) models . An intuitive understanding of the model selection process via the GA may be gained by thinking of it as a non-linear curve fitting problem , where the “true” curve is the unobserved distribution of biological substitution rates ( Figure 2 ) . We consider the substitution rate matrix for a codon model , extract non-synonymous rates for the above-diagonal entries which correspond to one-step non-synonymous substitutions and rank them in an increasing order to obtain monotonically increasing rate curves as shown in ( Figure 2 ) . Note that because the ratios for all substitutions between the same pair of amino-acids ( of which there are pairs ) are identical , this will create steps in such curves . In the case of one non-synonymous substitution rate ( SR ) the curve is a flat line at the estimated average non-synonymous substitution rate across all residue pairs . This is easily improved on by a random model which assigns non-synonymous substitutions randomly to one of 5 rate classes . At the other extreme lies the general time reversible models with estimated rates . Since we have no a priori reason to believe that any two non-synonymous substitution rates will be exactly the same , REV is the most biologically realistic of the models which assume time-reversibility and only single nucleotide substitutions . However , fitting the parameter rich REV model to limited data is statistically unsound . The GA-approach , instead , searches for the best ( in an information theoretic sense ) step-wise smoothing of the biological distribution given the data available ( Figure 2 ) . The “generalist” ECM model sacrifices gene-level resolution , in some cases so dramatically that it underperforms the single-rate model , even with the correction factor ( Table 4 ) . For instance , ECM appears to be ill suited for the analysis of viral genes . LCAP , on the other hand , performs poorly for highly divergent data sets; indeed the original validation of LCAP took place on relatively closely related yeast species [18] , and the mechanistic properties assumed by the model may be insufficient in alignments spanning multiple genera and taxonomic groups . To test whether GA structured models are generalizable , we estimated two viral models: one for HIV-1 polymerase and one for human IAV hemagglutinin . We then applied each of these models ( holding the inferred class structure fixed ) to two additional samples of sequences from the same gene , obtained independently from the training sample . In all cases outperformed ECM , ECM+ and LCAP by wide margins , lending credence to the claim that data-driven structured models recover substitutional biases that are shared by other samples shaped by similar evolutionary parameters . Curiously , for very low divergence ( and low information content ) intra-serotype IAV alignments , the single rate model was preferred to all other models by BIC , suggesting that there are biologically interesting alignments , which do not contain sufficient amino-acid variability to indicate the use of a multi-rate model . As a test of protein-specificity of models , we randomly selected four Pandit data sets to assess how well models inferred from unrelated proteins fitted these data ( Table S2 ) . Not surprisingly , ECM was the best model in cases , because it was derived as the best “average” protein model . LCAP topped the list in one case , but placed outside the top three in the other three cases . The GA structured models , being tailored to specific proteins , tended to differ from each other ( Table S3 ) and did not perform well on proteins from different families . However , the GA structured models for ATP cone and Transketolase C did outperform the LCAP model in cases , which suggests some similarity between the respective protein families in those cases . This indicates the GA models fitted to different proteins may be generalizable , with the degree limited by taxonomy , protein function or both . The generalizability of GA models could further be quantified by evolutionary fingerprinting of genes [27]; see also Figure 3 ( b ) . A GA search run typically examines between two- and a hundred-thousand potential models , e . g . models with to rate classes for the HIV-1 group M pol dataset . , which we compared to existing models in the previous section , is simply the single “best” model , i . e . the model that minimized the criterion among all those examined during the run . Further , we estimate the credible set of models as those models whose evidence ratio versus the best model is sufficiently large ( see methods ) . Among models fitted to HIV-1 pol by the GA , belonged to the credible set . Given sufficient data and knowing that the true model is in the set examined by the GA , e . g . in the long 2-sequence simulations discussed above , the size of the credible set frequently shrinks to ( the true model ) . These structured ( ) and model-averaged ( ) models can be analyzed further to draw inferences of the substitution process . For instance , the structured model identifies which residue pairs are exchanged rarely , relative to the baseline synonymous rate . In Figures 4 and 5 we cluster the pairs of residues which have the same rate of non-synonymous substitution; residues are labelled by Stanfel class and physicochemical properties . Note that the same residue can be present as a node in multiple clusters because the GA partitions residue pairs ( i . e . the rates between them ) , not the residues themselves . The model reveals a startling heterogeneity of substitution rates in HIV-1 pol: the single rate estimate of is resolved into rate classes ( Figure 4 ) , with relative non-synonymous substitution rates ranging from ( 20 residue pairs ) to ( 3 residue pairs ) ; a similar range is revealed for other datasets ( Table 3 ) . It is remarkable that some of the non-synonymous substitutions occur at rates matching or exceeding the gene-average rate of synonymous substitutions . This can be interpreted , for instance , as lack of selective constraint on particular residue substitutions gene-wide , or evidence of directional selection when some residues are preferentially replaced with others . Regardless of how this result is interpreted , a remarkable complexity of substitution patterns is revealed by the analysis . We hypothesize that such patterns reflect complex dynamics of substitutional preferences that may be shared by multiple samples of the same genes . This hypothesis is supported ( by the goodness-of-fit of vs other models ) on HIV-1 and IAV samples in this study ( Table 4 ) , and we are currently undertaking the GA analysis of several thousand alignments to confirm this finding . One of the benefits of using the model instead of REV or other models is that the former model automatically classifies all substitutions into similarity groups , supplying a data-driven analog of “conservative” or “radical” substitutions , previously defined a priori based on chemical properties of the residues , or a more sophisticated multi-property basis defined in the LCAP model . For example , the substitution rates are partitioned into seven classes in the model inferred from HIV-1 pol , and into rate classes for the model fitted to a smaller , but more divergent vertebrate rhodopsin alignment ( Figure 5 ) . Multi-model inference is instrumental in assessing how robust the clustering assignment made by is . In Figures 4 and 5 , we present this information by labeling individual substitution rates with their model averaged values . An examination of the numerical differences between rate estimates ( for a particular amino-acid pair ) obtained under and can reveal ambiguities in assigning a particular rate to a class . More formally , we can compute a model averaged support for the probability that rates and ( for residues , ) are in the same class , as described above , or that the corresponding edges and are in the same component of the rate graph ( Figure 4 ) . If is a cluster defined by ( with the number of nodes in , ) , we define the cluster affinity of an edge as the mean of the model averaged estimates of the probabilities that edge and other edges in belong to the same cluster:If is below a certain threshold , for instance for majority rule , then cluster membership of edge is ambiguous . For example , the substitution pair with a model-averaged non-synonymous rate of is one of two rate pairs with low ( ) cluster affinity for HIV-1 ( Figure 4 ) . Two of the inferred rate classes have non-synonymous rates of and , respectively , and the placement of model-averaged rate for between the two values is indicative of the alternate assignment of this substitution pair to these two rate classes among models of the credible set . A larger training data set may be able to infer an additional intermediate rate class between and . While yields more robust numeric estimates of substitution rates for a single data set , has better fit on validation HIV and IAV alignments ( results not shown ) . The expectation that substitutions which preserve amino acid physicochemical properties occur at a lower rate than property-altering substitutions has previously been evaluated in the maximum likelihood codon model context [20] , [47] . However , in published work , property-altering and property-conserving amino acid classes are defined a priori , whereas in the GA approach amino acid substitution pairs are first partitioned into classes based on rate similarity , and thereafter property preserving versus property-altering rates can be compared . The increased substitution rate of property preserving substitutions , holds largely – but not universally – for and rates , as evidenced in Figures 6 and 7 . For example , in the vertebrate rhodopsin sample , the median rate of charge-changing substitutions is significantly lower than the charge-preserving substitutions , but the two medians are not significantly different in the HIV-1 pol sample . The rates were negatively correlated ( , one-sided Pearson product moment test , no multiple test correction ) with out of property-based distances ( polarity , volume , isoelectric point and hydropathy ) that form the basis of the LCAP model . However , while the broad pattern follows the expectation , the consistently better fit of -based models , and the presence of strong outliers , such as and in the cluster of HIV-1 rates ( Figure 4 ) , suggests that our data driven approach detects significant deviations from purely biochemical rate expectation . These deviations could be attributed to selective pressures which promote property changes , or could arise because not all biologically relevant important properties have been included into structured models . One benefit of our approach over the “amino acid class” models [20] , [47] is that transitivity of rates ( i . e . the requirement that if , and are in the same rate class , then so is ) is not enforced by the GA models . Because we focus on modeling single-nucleotide substitution rates only , the structure of the genetic code itself contradicts transitivity . For instance both ( encoded by ) and are one-step substitutions , but is not . Further , since amino acid class models only estimate two non-synonymous rates ( within and between classes ) , it is a necessary condition that non-synonymous rates which change amino acid property be shared irrespective of how much the property is being changed . For instance , substitution rates which change charge from negative to positive will be the same as those which change charge from negative/positive to uncharged . If amino acid substitutions that result in a positive charge are favored , then these transitive conditions are not representative of the substitution process . Furthermore , the amino acid class models assume all substitutions within classes occur at the same rate . This is a very strong assumption since some amino acids with the same physicochemical property class are separated by more than one nucleotide substitution , e . g . positively charged amino acids and . Although we do not account for multiple nucleotide substitutions in the GA model directly ( but see below ) , previous work has demonstrated that these occur at lower rates than single-nucleotide substitutions [9] , [10] , [48] . Using the substitution spectrum distance defined in Equation ( 2 ) , it is easy to construct a hierarchical clustering of several models fitted to the same dataset , as well as between models fitted to different datasets . The former is useful to interpret how much difference in predicted substitution patterns over a unit of evolutionary time there is between different descriptions of the same data , whilst the latter naturally extends the concept of evolutionary fingerprinting of non-homologous genes [27] . For HIV-1 pol ( Figure 3 ) , and models both clustered closely with the rate substitution pattern predicted by the REV model , followed by LCAP , ECM+F+ , and finally – distant single rate models . The similarity between REV and GA models was especially strong for the parameterization , under which the GA models were inferred . In a between-genes model comparison ( Figure 3 ) , the two viral alignments clustered together , as did the two most divergent alignments ( ATP-cone and Transketolase C ) . Statistical inference procedures based on phylogenetic models have varying degrees of robustness with respect to the substitution rate matrix used in the analysis . For a multi-rate model , it is intuitively clear that the types of inference that rely on “mean” rates should be minimally affected , whereas those that depend on the individual residue rates can be affected significantly . We examine several such measures inferred from two of the datasets in this study . Branch length estimates are essentially unchanged when moving from the single-rate ( SR ) model to a model . On the example HIV-1 pol dataset , the total tree length changed from to ( SR ) expected substitutions/nucleotide to ( ) , and the lengths of individual branches were nearly perfectly linearly correlated with linear regression slope of , intercept of and . Ancestral character reconstruction is considerably more sensitive to the substitution model . In the vertebrate rhodopsin data set , for example , the joint maximum likelihood ancestral reconstruction [49] under SR and models differed in the number of inferred non-synonymous substitutions at sites , with more non-synonymous substitutions in cases under . At sites substitutions were mapped to a different set of branches . Site-specific diversifying selection screens are likely to be profoundly affected by a switch from single- to multi-rate models . Consider the FEL method [50] , where the SR model is fitted site-by-site and a likelihood ratio test ( LRT ) is used to test whether . First , because defines multiple substitution classes , one can now apply a variety of tests to see which non-synonymous rates at a given site exceed the baseline synonymous rate . To explore this approach for a rate multi-class model applied to the vertebrate rhodopsin alignment , we performed LRT tests , where we independently constrained each non-synonymous rate parameter ( , Table 1 ) to be equal to at a site ( neutral evolution in class ) , vs an unconstrained parameter alternative . This is analogous to performing a test for selection at a site by constraining the non-synonymous rate to be equal to the synonymous rate , and comparing the fit to the unconstrained model ( FEL ) , except that we only place the constraint on one rate class at a time . At , the standard ( SR ) FEL reported ( codon 54 ) sites as being under diversifying selection ( positively selected ) . However , for the model , there were and positively selected sites for the four substitution classes ( Figure 5 , increasing rate magnitude ) , respectively at the Bonferroni corrected of . Codon 54 was selected only with the fastest rate class ( ) , because the signal of selection is driven by a large number of substitutions . Only one codon ( ) was selected with two or more different tests ( rate classes and ) . We remark that the effects of site-to-site rate variation and multiple non-synonymous rates appear to be largely additive , and not confounded . This is a critical observation: if the effects are confounded , then we cannot justify inferring the multi-rate model independently assuming no site-to-site rate variation , as is done in this manuscript for computational expedience . To illustrate , we fitted both a constant rate model and the general bivariate distribution [27] , with and without accounting for multiple non-synonymous rate classes ( Table 5 ) . The constant rate model assumes all sites share the same rate of substitution , whereas a general bivariate distribution infers the number of site-to-site variation classes from the data [27] . These models were fitted to the vertebrate rhodopsin alignment , which exhibits extensive site to site rate heterogeneity . The inferred 4 non-synonymous rate classes for the rhodopsin alignment , whereas the single has one , resulting in three degrees of freedom for the comparison of these models . When the general bivariate model was fitted with a single or , 6 and 7 site classes were inferred , respectively , resulting in 4 degrees of freedom for the comparison of single and models ( 3 rate and 1 site class are added to the model ) . The important observation is that the addition of site-to-site rate variation component resulted in a significant improvement in log likelihood scores , regardless of the underlying substitution model ( single or ) . This suggests that by allowing multiple rate classes , we are not merely fitting variability in site-to-site selective constraints . However , as the cost of computing cores in clusters decreases , we expect that it will become practical to infer models with the site-to-site rate variation component included directly in the search procedure . Recent extensions of codon models which permit multiple instantaneous nucleotide substitutions [9] , [10] , [48] tend to fit the data better than their traditionally parameterized counterparts . We explored whether this observation held for models using a straightforward extension of the rate matrix in Equation ( 1 ) , following the ideas of [9] . We introduce four new independently estimated parameters to model the relative rates of synonymous ( ) and non-synonymous ( ) substitutions which replace two or three nucleotides , and modulate them by the product of the corresponding nucleotide rates and the target codon frequency ( assuming the GY parameterization with the F61 estimator ) . For instance the rate of synonymous substitution ( Serine ) from to is , while the rate of non-synonymous substitution ( Lysine to Proline ) is . Table 6 summarizes the effect of adding multi-step substitutions to SR and models for the vertebrate rhodopsin alignment . Much as was the case for site-to-site rate variation , the effects of multiple single-step non-synonymous rates and the non-zero rates of two or three nucleotide substitutions are additive at the level , and the estimates of single-step substitution rates were minimally influenced by the presence of the multi-step component ( results not shown ) . The model augmented to allow multi-step substitutions can be directly compared to the Mechanistic-Empirical codon ( MEC ) model [9] coupled with the LG [51]empirical amino-acid substitution model ( selected as the best fitting empirical model using the procedure implemented on http://www . datamonkey . org . Assuming no site-to-site rate variation , BIC of the MEC model is , while that of the +multi-step model using the HKY85 nucleotide component ( a direct analog to the MEC model ) is , once again highlighting how strongly the substitution process in an individual gene appears to deviate from the “average” encoded by empirical protein models . The GA could be modified to search for optimal partitions among all pairs of rates , for example using the above parameterization , but as the rhodopsin example indicates , the single-step and multi-step rate rate components appear to be effectively independent . We will explore this option in future versions of the model selection GA .
In this manuscript we have developed , validated and benchmarked a procedure to quickly and reliably infer a multi-rate model from the combinatorially large class of general time-reversible codon substitution models . Using extensive simulations , we demonstrated that our conservative model selection criterion controls over-fitting and has excellent power on data sets of biologically realistic size , inferring the exact model simulated given sufficient sequence divergence and length . We have previously argued against using the single rate model as a benchmark against which multi-rate models should be compared , since it is trivial to improve upon using a random assignment of substitutions to rate classes [19] . We reiterate this argument here , and suggest we should rather consider how well a multi-rate model approximates the REV model ( Figure 2 ) , given the limitations posed by the information content in an alignment . On a diverse collection of biological data , models consistently outperform the best-in-class empirical and mechanistic models , and match the performance of fully parameterized general time reversible models with only a few biologically relevant rate parameters ( Table 4 ) . Therefore , the provides goodness of fit matching or exceeding that of REV , with substantially fewer parameters and is thus computationally and statistically feasible for downstream analyses . ModelTest [22] has been universally adopted to mitigate the effect of model misspecification on statistical inference from nucleotide data , and we posit that a robust codon model selection procedure , for example the one offered in this paper , will play a similar role for codon data . In the same vein as ModelTest , we infer the best model ( which we term the ) for an alignment , and also utilize model averaging [26] to achieve more robust estimates of biologically relevant parameters . Certain applications of codon models , such as divergence estimation , appear unaffected by the gross biological over-simplification of single-rate models , because they are only influenced by the mean of substitution rates . Others , including ancestral sequence reconstruction ( e . g . for guided site directed mutagenesis , [38] ) , substitution mapping ( e . g . for co-evolutionary analysis , [52] ) and character sampling ( e . g . for data augmentation modeling approaches , [53] ) can see moderate effects . Applications which are tightly integrated with the substitution model and the interpretation of its parameters , such as site-by-site positive selection detection ( e . g . [50] , [54] ) , will be profoundly affected by the introduction of multiple rates . Our results strongly argue against the prospect of deriving a single “generalist” model of codon evolution , that is capable of fitting most protein alignments well . Hence we should strive to fit both gene and taxonomy specific models of codon evolution . We further hypothesize that independent alignments representing a gene or a protein family will share most of the model structure and confirm this with HIV-1 polymerase and Influenza A virus hemagglutinin examples . While significant further validation is required and is currently underway , we assert that a collection of substitution models inferred from carefully selected training datasets can provide a useful library of organism and gene-specific models to be used in inference on codon sequences . This is conceptually similar to a library of Hidden Markov profile models , inferred from seed alignments , used for detecting protein domain homology in the Pfam database [55] . In order to facilitate the process of generating gene and taxonomic specific multi-rate codon models we have implemented the GA on our free analysis webserver ( http://www . datamonkey . org , [46] ) , and have begun to assemble a library of representative multi-rate substitution models that are needed to reduce biases in those procedures that are sensitive to model misspecification . The inference of the multi-rate codon models should be considered more than just a necessary step for downstream applications . By examining the structure of inferred rate classes , we argue that the captures the a priori expectation that radical changes in one or more biochemical properties of a residue happen relatively infrequently , but also that a mere reliance on such data-abstract mechanistic properties misses out important gene and organism specific peculiarities of the evolutionary process . For instance the elevation of substitution rates between amino acids that do not preserve physicochemical properties may be indicative of selective pressures which promote property changes . These selective pressures are of crucial importance in understanding evolution in viruses , such as HIV-1 , known to evade host immune response [56] . We anticipate that considering specific substitution types when estimating selective pressures will improve power , as demonstrated with our multi-rate FEL analysis of vertebrate rhodopsin . However , this may also increase the rate of false positives , a conjecture that can be evaluated with straightforward , but laborious simulations . Finally , we demonstrate how simple metrics on models inferred from different ( e . g . non-homologous ) alignments can be used to obtain an objective measure of similarity and disparity in substitutional preferences in different proteins and thus improve the resolution in evolutionary fingerprinting of genes [27] .
|
Evolution in protein-coding DNA sequences can be modeled at three levels: nucleotides , amino acids or codons that encode the amino acids . Codon models incorporate nucleotide and amino acid information , and allow the estimation of the rate at which amino acids are replaced ( ) versus the rate at which they are preserved ( ) . The ratio has been used in thousands of studies to detect molecular footprints of natural selection . A serious limitation of most codon models is the unrealistic assumption that all non-synonymous substitutions occur at the same rate . Indeed , amino acid models have consistently demonstrated that different residues are exchanged more or less frequently , depending on incompletely understood factors . We derive and validate a computational approach for inferring codon models which combine the power to investigate natural selection with data-driven amino acid substitution biases from alignments . The addition of amino acid properties can lead to more powerful and accurate methods for studying natural selection and the evolutionary history of protein-coding sequences . The pattern of amino acid substitutions specific to a given alignment can be used to compare and contrast the evolutionary properties of different genes , providing an evolutionary analog to protein family comparisons .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"computational",
"biology/evolutionary",
"modeling",
"evolutionary",
"biology/bioinformatics"
] |
2010
|
CodonTest: Modeling Amino Acid Substitution Preferences in Coding Sequences
|
Heterochromatin silencing is pivotal for genome stability in eukaryotes . In Arabidopsis , a plant-specific mechanism called RNA–directed DNA methylation ( RdDM ) is involved in heterochromatin silencing . Histone deacetylase HDA6 has been identified as a component of such machineries; however , its endogenous targets and the silencing mechanisms have not been analyzed globally . In this study , we investigated the silencing mechanism mediated by HDA6 . Genome-wide transcript profiling revealed that the loci silenced by HDA6 carried sequences corresponding to the RDR2-dependent 24-nt siRNAs , however their transcript levels were mostly unaffected in the rdr2 mutant . Strikingly , we observed significant overlap of genes silenced by HDA6 to those by the CG DNA methyltransferase MET1 . Furthermore , regardless of dependence on RdDM pathway , HDA6 deficiency resulted in loss of heterochromatic epigenetic marks and aberrant enrichment for euchromatic marks at HDA6 direct targets , along with ectopic expression of these loci . Acetylation levels increased significantly in the hda6 mutant at all of the lysine residues in the H3 and H4 N-tails , except H4K16 . Interestingly , we observed two different CG methylation statuses in the hda6 mutant . CG methylation was sustained in the hda6 mutant at some HDA6 target loci that were surrounded by flanking DNA–methylated regions . In contrast , complete loss of CG methylation occurred in the hda6 mutant at the HDA6 target loci that were isolated from flanking DNA methylation . Regardless of CG methylation status , CHG and CHH methylation were lost and transcriptional derepression occurred in the hda6 mutant . Furthermore , we show that HDA6 binds only to its target loci , not the flanking methylated DNA , indicating the profound target specificity of HDA6 . We propose that HDA6 regulates locus-directed heterochromatin silencing in cooperation with MET1 , possibly recruiting MET1 to specific loci , thus forming the foundation of silent chromatin structure for subsequent non-CG methylation .
Chromatin modification is epigenetic information that has evolved in diverse eukaryotes adding another layer of information to the DNA code . In higher eukaryotes , histone modification and DNA methylation are involved in numerous biological processes such as development , regeneration , and oncogenesis [1] , [2] . In addition , the eukaryotic genome has evolved epigenetic mechanisms to silence potentially harmful transposable elements ( TEs ) and the repetitive elements that constitute a large proportion of the genome [3] . Heterochromatin formation , a striking function of the eukaryotic genome , is intricately controlled through repressive histone modification and DNA methylation [4] . Thus , mutations that affect the status of chromatin structure often result in strong phenotypic alterations or inviability , because of aberrant regulation of gene expression or distorted genome stability [5]–[8] . The flowering plant , Arabidopsis thaliana , is a model organism particularly suited for epigenetic research due to the availability of viable and heritable null mutants of histone modifying enzymes and DNA methyltransferases . Recent genome-wide studies on epigenetic marks of gene silencing in plants have focused on DNA methylation or repressive histone methylations [9]–[13] . Few studies , however , have focused on histone deacetylation , which is crucial for epigenetic regulation in eukaryotes [14] , [15] . Investigating histone deacetylation and DNA methylation in Arabidopsis could contribute not only to our understanding of plant biology , but also to a broad range of essential biological processes in mammals and therapeutic applications in humans [16] , [17] . Gene silencing has been investigated extensively in Arabidopsis . Plants have evolved gene silencing machinery called RNA-directed DNA methylation ( RdDM ) . Plant-specific RNA POLYMERASE IV ( Pol IV ) , RNA-DEPENDENT RNA POLYMERASE 2 ( RDR2 ) and DICER-LIKE 3 ( DCL3 ) are involved in the production of 24-nt small interfering RNAs ( siRNAs ) that guide DNA methyltransferases , DOMAINS REARRANGED METHYLTRANSFERASES 1/2 ( DRM1/2 ) , to the corresponding genomic DNA for de novo DNA methylation in all cytosine contexts ( CG , CHG , CHH; H: A , T , or C; [18] ) . METHYLTRANSFERASE 1 ( MET1 ) , a homolog of mammalian DNMT1 , is primarily responsible for the maintenance of genome-wide CG methylation [19]–[22] . KRYPTONITE ( KYP ) , a member of the Su ( var ) 3–9 class of histone methyltransferases , contributes an epigenetic mark of constitutive heterochromatin , histone H3 Lys 9 dimethylation ( H3K9me2 ) [23] , [24] . CHROMOMETHYLASE 3 ( CMT3 ) , a plant-specific DNA methyltransferase , maintains CHG methylation via H3K9me2 dependence mediated by KYP [23] , [25] , [26] . Histone Deacetylase 6 ( HDA6 ) , a homolog of yeast RPD3 and mammalian HDAC1 , is involved in gene silencing and RNA-directed DNA methylation [27]–[30] . Of the 16 Arabidopsis histone deacetylases [31] , the importance of HDA6 in gene silencing was discovered by identification of HDA6 in three independent genetic screens of gene silencing [27] , [32] , [33] . In each case , hda6 mutant plants lacking histone deacetylase activity ( sil1 , axe1 , and rts1 ) were shown to exhibit reactivation of transcription on target transgenes . Analyses of the endogenous function of HDA6 have been limited , thus far , to the regulation of chromatin at repetitive sequences such as rDNA loci [28] , [30] , [34] , [35] , transposable elements and centromeric satellite repeats [29] , [36] . However , the positions of the loci silenced by HDA6 have yet to be determined genome-wide . Various effects of the hda6 mutations on cytosine methylation have been observed previously . Several transposable elements were hypomethylated in sil1 [36] . Reduction of DNA methylation has been reported for the siRNA-directed NOS promoter in rts1 , predominantly at CG and CHG sites [27] . Similarly , a reduction in CG and CHG methylation was observed in axe1-5 , sil1 , and rts1 mutants at rDNA repeats , although the demethylation was much less than that observed in the DNA hypomethylation mutant ddm1 [28] , [30] . In contrast to these observations , a drastic reduction in CHG methylation , but not CG methylation , was observed in a Sadhu-type transposable element in axe1-5 [37] , and 5S rDNA in sil1 [35] . Furthermore , few changes in DNA methylation were observed in the centromeric repeats or transgene region in sil1 , although their silencing was lost [29] , [38] . These various effects of the hda6 mutations on DNA methylation might be due to locus dependence rather than differences in the mutations themselves , because similar effects were observed between the mutants [28] , [30] . Because previous studies have focused on only a few specific loci , precisely how the hda6 mutation influences DNA methylation in general remains obscure . Therefore , a genome-wide analysis of HDA6 target loci is vital to improve our understanding of the mechanistic basis for HDA6-mediated gene silencing via DNA methylation and histone modification . In this study , aimed at understanding the silencing mechanism mediated by HDA6 , we identified HDA6 transcriptionally repressed loci across the genome and determined the direct targets of HDA6 . We also studied the regulation mechanisms involved in histone modification and DNA methylation on HDA6 direct targets . Our data show that the hda6 mutation causes loss of heterochromatic marks and aberrant enrichment for euchromatic epigenetic marks at HDA6 direct targets . Furthermore , we present evidence that the upregulated loci in hda6 overlapped with those in met1 , and that the hda6 mutation causes the complete loss of DNA methylation on some HDA6 target loci . These results suggest that a strong functional connection between HDA6 and MET1 exists . Remarkably , hypomethylation only occurred in hda6 on the HDA6 target loci where surrounding MET1 targets were absent . We propose , therefore , that HDA6 is required for gene silencing and that it acts in cooperation with MET1 to build the infrastructure of heterochromatin .
To identify target loci for HDA6 binding in the Arabidopsis genome , we first performed a genome-wide comparison of RNA accumulation between the wild-type plant ( DR5 ) and the hda6 mutant , axe1-5 [33] , using a whole-genome tiling array . This approach identified 157 statistically significant loci that were transcriptionally upregulated in axe1-5 compared with wild-type plants ( >3 fold , p-initial<10−6 , FDR α = 0 . 05 ) ( Figure 1A and 1B ) . RT–PCR of a random selection of these loci was used to confirm their up-regulation in the axe1-5 mutant ( Figure S1 ) . Among these loci , nearly half ( 81 genes; see Table S1 ) were annotated by the Arabidopsis Genome Initiative ( http://www . arabidopsis . org/; hereafter referred to as AGI genes ) . The other half ( 76 genes; Table S2 ) were intergenic non-AGI annotated transcriptional units ( non-AGI TUs ) identified using the ARTADE program [39] ( Figure 1B ) . It is noteworthy that only a small fraction of transcripts ( 5 loci: 3% of all differentially expressed loci ) were classified as having reduced levels of expression in axe1-5 ( Figure 1A , Table S3 ) . We consistently found that the loci upregulated in axe1-5 were strongly silenced in wild-type plants ( Figure 1C ) and consisted predominantly of TE fragments and genes for unknown proteins ( 79% ) ( Figure 1D ) . A survey of the TE fragments [40] , mapping on or around the loci upregulated in axe1-5 ( from 1 kb upstream to 1 kb downstream ) , showed that a significant number of the fragments ( 342 TE fragments ) were located on or around such loci ( Table S1 and S2 ) . These results show that HDA6 regulates gene silencing on a genome-wide scale . Forward genetic screens for plants deficient in RNA-mediated transcriptional silencing identified HDA6 as an essential component of the RdDM pathway [27] , [41] . To address whether the endogenous HDA6 target loci were also directed by the RdDM pathway , siRNAs from the ASRP database [42] were mapped to the loci derepressed in axe1-5 . Consistent with knowledge that 24-nt-long siRNAs are required for the establishment of RdDM , the most abundant siRNAs mapping to upregulated loci in axe1-5 are 24-nt long ( Figure S2A ) . These 24-nt siRNAs were hardly found in the rdr2 and dcl3 mutants ( Figure S2B ) , suggesting that loci derepressed in axe1-5 contain siRNA sequences produced by RDR2 and DCL3-dependent pathways , as previously predicted [27] , [41] . Thus , our previous study for the targets of RDR2 [43] , that were identified using same growth conditions and array technology as in this study , were compared with the genes derepressed in axe1-5 . This revealed , surprisingly , despite the loss of 24-nt siRNAs from those genes were observed in the rdr2 mutant ( Figure S2B ) , the majority of the loci derepressed in axe1-5 are kept in a silenced state in the rdr2 mutant ( Figure 2A ) . In fact , elevated transcript levels were not detectable at many loci in the mutants deficient in siRNA production ( rdr2 , and nrpd1; 10 and 11 respectively , out of 13 genes tested; Figure 2B ) . There also was evidence that small subsets of the HDA6-mediated gene silencing showed dependence on RdDM pathway and the overlapped genes between axe1-5 and rdr2 was confirmed for their accumulated transcripts ( AT3TE60310 , At1g67105 , and At3g28899 ) in the RdDM mutants ( rdr2 and nrpd1; Figure 2B ) . Interestingly , larger overlap to the triple mutant drm1 drm2 cmt3 ( ddc ) involved in siRNA-directed non-CG methylation was observed ( 5 out of 13 genes; Figure 2B ) . Taken together , these results indicate the partial involvement of RdDM pathway in HDA6-mediated endogenous gene silencing . We also examined the effects of mutations in other chromatin modifying enzymes on the silencing of putative HDA6 target loci . Strikingly , 10 out of 13 of the putative HDA6 target loci were also upregulated in the met1-3 mutant ( Figure 2B ) . To address whether HDA6 and MET1 share common target loci genome-wide , we also identified differentially regulated loci in met1-3 using a tiling array ( Tables S4 , S5 , S6 and S7 ) , and compared the upregulated loci in met1-3 with those in axe1-5 . A significant overlap of upregulated loci in axe1-5 to those in met1-3 was observed ( Hypergeometric distribution , P = 1 . 08E−54; Figure 2C ) . Furthermore , the DNA methylation status of the loci derepressed in axe1-5 was also investigated using publicly available DNA methylation datasets [13] . Most of the genes upregulated in axe1-5 ( i . e . 70% of the upregulated AGI genes ) were substantially methylated in the wild-type plants with more than 50% of all cytosines at regions surrounding transcriptional start sites methylated ( Figure 2D ) . Cytosine methylation in the wild-type plants was predominantly found at CG , to a lesser extent at CHG , and least of all at the CHH sites of derepressed AGI genes in axe1-5 ( Figure 2E ) . A large proportion of the cytosine methylation on derepressed AGI genes in axe1-5 appears to be highly dependent on MET1 because the drastic reduction in cytosine methylation was observed not only at CG , but also CHG and CHH sites ( Figure 2E ) . In contrast , CG methylation in the ddc mutant remained at similar levels as the wild-type plants ( Figure 2E ) . Thus , these data demonstrate that the CG DNA methyltransferase MET1 is required for HDA6-mediated epigenetic gene silencing . Identification of the direct targets of HDA6 is a crucial step in providing mechanistic insight into HDA6 function in transcriptional control , chromatin regulation , and DNA methylation . To determine the HDA6 target loci using chromatin immunoprecipitation ( ChIP ) , we raised a specific antibody against HDA6 . The epitope for the antibody was designed against the C-terminal region of the HDA6 protein that is absent in axe1-5 . We verified that the peptide sequence was not similar to any other sequence of annotated Arabidopsis proteins . Thus , comparisons made between wild-type and axe1-5 for the level of enrichment of immunoprecipitated DNA using the antibody allowed us to exclude any non-specific signals to identify HDA6 binding sites ( Figure 3A ) . Western blot analysis confirmed the specificity of the antibody , which detected a unique band of 53 kDa in the wild-type , but not in axe1-5 ( Figure 3B ) . Using the HDA6 antibody , we performed ChIP assays and quantitative PCR ( qPCR ) . Three genes , AT3TE60310 ( At3g42658 ) , AT3TE76225 ( At3g50625 ) and At5g41660 were selected from genes upregulated in axe1-5 , representing RdDM dependent , MET1 independent , and MET1 dependent genes , respectively ( Figure S1 ) . Three primer sets were designed for each gene within the promoter , 5′ and 3′ regions of the genes ( Figure S3 ) . HDA6 binding levels in the wild-type plants were significantly higher than in axe1-5 for all of the genes tested , regardless of the dependence on RdDM pathway or MET1 . This indicates that HDA6 binds directly to all such genes . In addition , preferential binding of HDA6 was observed within the 5′ regions of the genes . Therefore , a further screening of HDA6 target loci was performed for the 5′ regions of selected loci from those upregulated in axe1-5 ( Figure S1; Figure 3C ) . As a negative control , we tested one gene that exhibited no apparent transcriptional change ( At5g55670 , Figure S1 ) , and found only a minimal difference . Our experiments show that HDA6 binding levels were enriched by 2 to 12-fold in the wild-type plants relative to axe1-5 , with statistical significance observed for 17 of the loci ( Figure 3C ) . 5 loci did not show significant differences between the wild-type and axe1-5 mutant . Heterochromatic or repressive regions are associated with H3K9me2 and/or H3K27me3 , whereas euchromatic or transcriptionally active regions are associated with H3K4me3 and H4 tetra-acetylation ( H4 tetra-acetylated on K5 , K8 , K12 , and K16 ) [44] . Furthermore , many endogenous RdDM targets are known to be associated with euchromatic modification H3K4me3 [45] . To elucidate HDA6 function and chromatin status , the effects of the hda6 mutation on histone modification was analyzed using ChIP-qPCR on its direct targets . The results show that , regardless of the dependence on siRNAs or positions on the chromosome , weak H4 tetra-acetylation and H3K4me3 and significantly high levels of heterochromatic modification , H3K9me2 , were observed in the wild-type plants at all of the loci tested ( Figure 4A , 4B and 4C; Figure S4 ) . In the axe1-5 mutant , the active marks strongly increased ( H4 tetra-acetylation at range 5 to 30 fold and H3K4me3 at 8 to 78 fold , respectively ) , and the levels of H3K9me2 were drastically reduced ( range 2 to 53 fold ) compared with the wild-type plants ( Figure 4A , 4B and 4C; Figure S4 ) . H3K27me3 , another repressive modification , are highly enriched in wild-type plants predominantly on genes within the chromosome arm regions , such as At1g67105 and At5g41660 , and drastically reduced in axe1-5 ( Figure 4D ) . Consistent with this observation , the enrichment of H3K27me3 in the euchromatic arm regions was also seen in the previous genome-wide studies of H3K27me3 [11] , [12] . These results indicate that the hda6 mutation caused an alteration of the chromatin status from a heterochromatic to euchromatic state that was concomitant with transcriptional release of HDA6 target loci . HDA6 deacetylase activity has been reported for H3K9 , H3K14 , H4K5 and H4K12 , as well as for H4 tetra-acetylation [34] . The activity at other residues , however , is currently unknown . To assess the HDA6 deacetylase activity at other lysine residues , we investigated all of the potential acetylation sites of the H3 and H4 N-tails , including pre-determined sites using ChIP-qPCR . Three HDA6 target loci ( AT3TE76225 , At5g41660 , and G683 ) were examined . The results showed that in axe1-5 , the acetylation levels significantly increased at H3K9 , H3K14 , H3K18 , H3K23 , H3K27 , H4K5 , H4K8 , and H4K12 residues relative to those in wild-type plants at all three loci tested ( Figure 5A , 5B and 5C , Figure S5 ) . Interestingly , among these residues , H3K23ac levels showed the highest enrichments in axe1-5 , for the three loci , AT3TE76225 , At5g41660 and G683 ( at 10 , 14 , 5 fold respectively ) . The acetylation levels were not significantly altered in axe1-5 for the control genes At5g55670 ( Figure 5D ) and ACT2 ( Figure S5 ) . It is noteworthy that the deacetylase activity observed for HDA6 at H3K27ac as well as H3K9ac , are both likely to be important for the subsequent histone methylation of H3K27me3 and H3K9me2 , respectively [46] . Interestingly , residues that showed increased levels of acetylation in axe1-5 , were identical to the target residues in yeast RPD3 deacetylation [47] . These results indicate that the deacetylase activity of HDA6 occurred on all of the lysine residues in H3 and H4 N-tails , except H4K16 . The different effects that hda6 mutations impose on DNA methylation have been reported as described above . To assess the function of HDA6 on DNA methylation and the relationship between HDA6 and MET1 , the DNA methylation status of HDA6 direct targets was investigated . We used the endonuclease McrBC ( Figure 6A ) , which preferentially cleaves methylated DNA , and a Chop-PCR assay using methylation sensitive restriction enzymes ( Figure 6B ) , whose cleavage is blocked by DNA methylation . In the McrBC assay shown in Figure 6A , no strong bands were detected in the wild-type plants after McrBC digestion , indicating that the direct targets of HDA6 are highly DNA methylated in the wild-type plants . This is consistent with dense DNA methylation on the loci upregulated in axe1-5 ( Figure 2D and 2E ) . However , in axe1-5 , we observed two types of McrBC sensitivity among the HDA6 direct target loci . The Group A genes substantially lost cytosine methylation , as demonstrated by the presence of strong bands of similar intensity in both Group A genes and the non-digested control ( Figure 6A , left panel ) . In contrast , the genes in Group B retained DNA methylation in axe1-5 in common with the wild-type plants as no strong bands were detected ( Figure 6A , right panel ) . We also performed Chop-PCR assays to investigate in which cytosine contexts are dependent on HDA6 ( Figure 6B ) . The methylation sensitive restriction enzymes used were HpaII , MspI and HaeIII , which reports CG , CHG , and CHH methylation , respectively . From each group categorized in Figure 6A , four representative HDA6 target loci were tested and using ACT2 as a negative control . PCR amplification of ACT2 was undetectable regardless of the genotypes or the enzymes tested ( Figure 6B ) , confirming substantial cleavage by the enzymes had occurred . In the wild-type plants , strong amplification , at a similar level as the undigested control was detected after digestion with HpaII ( CG methylation ) . Strong amplification was observed with MspI ( CHG methylation ) , but it was mostly less than the undigested control; least amplification of all was observed with HaeIII ( CHH methylation ) ( Figure 6B ) . The results of this experiment are consistent with the results shown in Figure 2E and Figure 6A , where HDA6 target loci were shown to be significantly methylated in wild-type plants , predominantly at CG sites , and to a similar or lesser extent at CHG sites , but least of all at CHH sites . In common with the results obtained for the McrBC assay ( Figure 6A ) ; the results from the Chop-PCR assays split the HDA6 target loci into two separate groups . In Group A ( AT3TE60310 , AT3TE76225 , At3g54730 and At5g41660 ) , complete demethylation in axe1-5 occurred in all sequence contexts and amplification after digestion of each enzyme was undetectable ( Figure 6B , Group A ) . In the Group B genes ( AT3TE63935 , AT5TE43385 , At1g67105 , and At3g44070 ) , for axe1-5 , however , CG methylation was mostly sustained at the similar level as the wild-type ( Figure 6B , lower panel , HpaII digest ) , although a drastic reduction in CHG and CHH methylation was detected [Figure 6B , lower panel CHG ( MspI ) and CHH ( HaeIII ) ] . Bisulfite sequencing analyses of wild-type , axe1-5 and met1-3 further confirmed that Group A genes ( At5g41660 and AT3TE76225 ) lost DNA methylation in all sequence contexts in axe1-5 , and a Group B gene ( AT3TE63935 ) lost CHG and CHH methylation but sustained CG methylation in axe1-5 ( Figure S6 ) . Collectively , the methylation analyses of the HDA6 targets demonstrated that , 1 ) CG and CHG sites were predominantly highly methylated in the wild-type plants; 2 ) CHG and CHH methylation was substantially reduced at all target loci in axe1-5; and , 3 ) CG methylation was lost at some loci but sustained at others in axe1-5 . Why were two different CG methylation states observed in axe1-5 ? Interestingly , according to the public database for DNA methylation [9] , a clear correlation was observed between a loss of CG methylation on the HDA6 targets in axe1-5 and the absence of methylated DNA regions around the HDA6 target loci ( Figure S7 ) . HDA6 target loci in Group A ( with loss of DNA methylation ) were shown to be isolated from other methylated DNA regions , whereas HDA6 target loci in Group B ( with persistent CG methylation ) were surrounded by other methylated DNA regions . We confirmed the presence or absence of DNA methylated regions around HDA6 target loci ( Figure 6C ) . The 1 . 5 kb regions upstream and downstream of the HDA6 target loci were analyzed to determine their DNA methylation status using McrBC assays . We found evidence of robust DNA methylation around the HDA6 target loci in Group B ( with persistent CG methylation in axe1-5 ) ( Figure 6C , Group B ) . These methylated regions often contained other TE fragments adjacent to HDA6 target loci . These TEs were densely DNA methylated dependently on MET1 but independently of HDA6 , since the PCR amplification after McrBC digestion was detected only in met1-3 ( Figure S8 ) . We also confirmed that these adjacent TE fragments were not targeted by HDA6 using ChIP-qPCR assays; these results showed no enrichment of HDA6 binding to adjacent TE fragments in the wild-type plants compared with axe1-5 ( Figure S9 ) . As a result , HDA6 target loci with sustained CG methylation in axe1-5 ( Group B ) must harbor the flanking TE fragments that are highly DNA methylated by MET1 independently of HDA6 . On the other hand , we found that the target loci in Group A ( with loss of DNA methylation in axe1-5 ) were isolated from other DNA methylated regions as no substantial methylation was detected around the target loci ( Figure 6C , Group A ) . We confirmed the absence of DNA methylation around the HDA6 target loci in Group A by bisulfite sequencing analysis of the upstream region of a Group A gene , At5g41660 ( data not shown ) . Thus , we deduced a requirement for HDA6 involvement in CG methylation by MET1 for HDA6 target loci , in the absence of other flanking DNA methylated regions .
We have identified 157 loci that require HDA6 for epigenetic silencing in Arabidopsis . This is the first report to identify derepressed loci in axe1-5 on a genome-wide scale . Our study revealed several interesting features of HDA6 target loci , mapped large numbers of TE fragments and DNA methylation sites in wild-type Arabidopsis plants . The derepressed loci in axe1-5 overlapped significantly with the derepressed loci in met1-3 , a CG DNA methyltransferase mutant , rather than the RdDM deficient mutants rdr2 and ddc , suggesting that HDA6 plays an important role in gene silencing , in cooperation with MET1 . We also identified 17 direct targets of HDA6 using ChIP-qPCR assays with a HDA6 specific antibody . We found that HDA6 was required for heterochromatic histone modifications and DNA methylation in the target loci . Interestingly , HDA6 deficiency resulted in aberrant enrichment for euchromatic epigenetic marks and DNA hypomethylation at HDA6 targets , along with ectopic expression of these loci . DNA hypomethylation at CG sites in axe1-5 occurred at some HDA6 target loci , but only where isolated from other MET1 target loci , possibly indicating the requirement of HDA6 for the recruitment of MET1 to specific loci . We showed that all of the HDA6 direct targets tested in this study colocalized with a constitutive heterochromatin mark , histone H3K9me2 ( 7 out of 7; Figure 4C ) , rather than another mark of repressive chromatin , H3K27me3 ( Figure 4D , 3 out of 7 ) . We saw no evidence of euchromatic modification H3K4me3 ( 0 out of 7; Figure 4B ) . Moreover , our results suggested the deacetylase activity of HDA6 against H3K9ac and H3K27ac , will be important for subsequent histone methylations of H3K9me2 and H3K27me3 , as well as H3K14ac , H3K18ac , H3K23ac , H4K5ac , H4K8ac and H4K12ac . These results suggest that H3K9 and H3K27 deacetylation by HDA6 is essential for the establishment of the heterochromatic and repressive marks mediated by H3K9me2 and H3K27me3 , and HDA6 deficiency resulted in loss of heterochromatic histone marks and aberrant enrichment for euchromatic marks at HDA6 target loci . It is noteworthy that the enrichments observed on H3K23ac were the highest among the possible acetylation sites , which may indicate the importance of H3K23 deacetylation on heterochromatic gene silencing . The abundance of silenced TE fragments and genes for unknown proteins among the loci derepressed in axe1-5 ( Figure 1C and 1D ) supports the role of HDA6 in silencing at heterochromatic regions . In addition , an important role for HDA6 in CHG methylation is indicated by the observation that all of the direct targets of HDA6 tested in this study lost CHG methylation in axe1-5 ( Figure 6B; Figure S6 ) . Several papers report that CHG methylation maintained by CMT3 is dependent on H3K9me2 and that CMT3 is recruited to methylated histones [24]–[26] . Taken together , this indicates that HDA6 deacetylase activity against its target loci is required for establishment of the heterochromatic and repressive marks H3K9me2 and H3K27me3 , and CHG methylation by CMT3 . The separation of endogenous HDA6 target loci from endogenous RdDM target loci were investigated in this study . Our results show that endogenous HDA6 target loci were associated with a constitutive heterochromatic mark , H3K9me2 ( Figure 4C ) , but not a euchromatic mark , H3K4me3 ( Figure 4B ) . However , many of the endogenous target loci of RdDM components ( such as Pol V and DRD1 ) were found to be associated with euchromatic histone modification H3K4me3 , but not H3K9me2 [45] . Surprisingly , genome-wide identification of the loci derepressed in axe1-5 revealed that these loci overlap with only a small fraction of the genes upregulated in rdr2 ( Figure 2A ) . However , considering the recent studies proposing the role of siRNAs in re-establishment of DNA methylation and gene silencing when DNA methylation was lost in the DNA methylation deficient mutants like met1 and ddm1 [48]–[50] , the siRNAs found on the HDA6 target loci might also have a role in this mechanism , and therefore the double mutants of hda6 and siRNA deficient mutants might result in larger release of gene silencing . In addition , cell-type specific regulation of siRNAs and TEs silencing especially in the gametes has been proposed recently [51] . In this case , DDM1 expression is downregulated in the pollen vegetative nucleus , which accompanies the sperm cells . Considering that HDA6 has common features with DDM1 [18] , [29] , [36]–[38] , it would also be interesting to see if HDA6 also have a role in regulating transposon silencing in gametes . An important functional connection between HDA6 and MET1 was revealed by observing the significant overlap of the loci upregulated in axe1-5 , with the loci upregulated in met1-3 ( Figure 2B and 2C ) . Indeed , we found several HDA6 target loci ( AT3TE60310 , AT3TE76225 , At3g54730 and At5g41660 ) that require HDA6 for MET1 CG methylation ( Figure 6B; Group A ) . Thus , we propose that HDA6 acts in cooperation with MET1 , possibly as a recruiter or as a component of the silencing machinery with MET1 ( Figure 7 ) . Actually we observed the loss of HDA6 binding on several HDA6 targets in the met1-3 mutant ( Figure S10 ) . It indicates the requirement for MET1 and/or CG methylation to facilitate HDA6 binding , suggesting the cooperative interplay between HDA6 and MET1 . Further support for this hypothesis , is the observation that HDA6 targets isolated from other flanking MET1 targets experienced a loss of CG methylation in the absence of HDA6 ( Figure 6B and 6C ) . It is also supported by several papers evidencing the physical interactions that occur between histone deacetylases and DNA methyltransferases in mammals [52]–[54] . The sustained CG methylation status of the other HDA6 targets in the axe1-5 mutant was found to correlate with the existence of other flanking MET1 target loci in neighboring regions of the HDA6 targets ( Figure 6B and 6C , Group B; Figure S7 , S8 , S9 , S11 ) . Thus it appears likely , therefore , that MET1 could be recruited to the neighboring regions of the HDA6 targets and pass through the HDA6 targets ( Figure S11 ) . Another possibility is that MET1-dependent CG methylation is the primary repressive modification to silence those HDA6 targets . There also were many loci that were derepressed only in the met1-3 mutant ( Figure 2C , Figure S12 ) . It is noteworthy that the intensities of the RT-PCR bands were greater in met1-3 than in axe1-5 for the genes with sustained CG methylation ( i . e . At1g67105 , AT3TE63935 , At2g15555 , At3g44070 , and G683; Figure 2B ) , indicating that sustained CG methylation on the HDA6 target loci can be repressive to some extent . However , there is clear evidence that these genes are transcriptionally derepressed strongly in the absence of HDA6 or MET1 , regardless of the CG methylation status . We deduce , therefore , that both HDA6 histone deacetylation and MET1 CG methylation are essential for the silencing of HDA6 target loci . In addition , for most of the HDA6 target loci , CHG and CHH methylation was lost in both hda6 and met1 mutants ( Figure 6B ) . A drastic reduction in CHG and CHH methylation was also observed for the loci silenced by HDA6 in the met1 mutant ( Figure 2E ) . These observations indicate that CHG and CHH methylation on HDA6 targets require the presence of epigenetically silent chromatin associated with both histone deacetylation and CG methylation . Because HDA6 was identified as RTS1 ( RNA-mediated transcriptional silencing 1 ) and MET1 as RTS2 in RdDM screens [22] , [27] , we strongly suggest that these two genes are deeply connected to each other , as proposed previously [22] , [36] . These insights have an important evolutionary implication; that the histone deacetylase superfamily , one of the most ancient enzymes in eukaryotes [16] , may build the foundations for gene silencing and concomitant CG DNA methylation . CG DNA methylation is a conserved modification in higher eukaryotes , quite distinct from the siRNA derived CHG or CHH methylation found only in plants [55] , [56] . It is noteworthy that HDA6 has high specificity for target loci within the genome . Moreover , we found that HDA6 binds only to its target loci , not the flanking TE fragments ( Figure S9 ) . Relatively low numbers of loci were derepressed in axe1-5 ( 157 loci ) , contrasting with the met1-3 mutant , where a total of 1215 loci were derepressed . We consistently detected an insignificant difference in the amount of total methylated DNA in axe1-5 , whereas a severe reduction was detected in the met1-3 mutant compared with wild-type plants ( Figure S13 ) . It was also reported that the total amount of histone H4 tetra-acetylation or H3K4me3 did not change between axe1-5 and the wild-type plants [28] . From this , we conclude that the target specificity of HDA6 is likely to be precisely controlled , with the effect of the hda6 mutation manifesting itself only in local areas . Furthermore , HDA6 , but not MET1 , has the ability to trigger de novo chromatin silencing , because only backcrossed hda6/wild-type plants were able to restore DNA methylation , low H3K4me3 levels and a silent transcriptional state , comparable with wild-type plants , unlike met1/wt [36] , [37] . These findings , taken together , indicate that HDA6 is a regulator of locus-directed heterochromatin silencing in cooperation with MET1 , where it acts possibly as a recruiter or as a component of the chromatin silencing machinery with MET1 , thus establishing the foundations for silent chromatin status for the subsequent heterochromatin mark H3K9me2 and non-CG methylation ( Figure 7 ) . Because MET1 is the primary CG DNA methyltransferase encoded in Arabidopsis , regulation of the proper and specific distribution of MET1 CG methylation by employment of HDA6 and/or other possible factors would be an efficient way for the Arabidopsis genome to adapt to several developmental and environmental effects . It will be interesting to see if HDA6 and MET1 form a complex as is the case in mammals [52]–[54] , or what information is recognized by these factors to trigger the sequential silencing mechanism . In this study , we identified dozens of loci transcriptionally silenced by HDA6 and several loci directly targeted by HDA6 . These results will undoubtedly contribute to an understanding of the complex interplay between histone deacetylation and DNA methylation , revealing mechanistic insights into heterochromatin silencing in higher eukaryotes .
Seeds were surface-sterilized and stratified for 4 days at 4°C in the dark . The seeds were then grown in tissue culture plates on MS agar ( 0 . 8% ) medium supplemented with 1% sucrose under 16 h light/8 h dark for 15 days at 22°C . All experiments used axe1-5 [33] , met1-3 [21] , ddc ( drm1-2 drm2-2 cmt3-11 , [57] ) , kyp ( SALK_069326 [58] ) , rdr2-1 ( SAIL_1277_H08 [42] ) , nrpd1a-3 ( SALK_128428 [59] ) mutants or DR5 [33] and Col-0 wild-type plants . All plants were the Columbia ecotype . The GeneChip Arabidopsis tiling array set ( 1 . 0F Array and 1 . 0R Array , Affymetrix ) was used . Total RNA was extracted using Isogen reagent ( Nippon Gene ) . Probe synthesis , hybridization , detection , data evaluation with U-test ( FDR α = 0 . 05 ) and P initial value ( P<10−6 ) was conducted essentially as described previously [60] . Three independent biological replicates were performed for each strand array . Detection of intergenic transcribed units was performed as described previously [39] , [60] based on the TAIR8 annotation . A threefold increase or decrease in RNA accumulation was taken as additional criteria for defining the loci upregulated or downregulated in the axe1-5 and met1-3 mutants . Tiling array data are available at the GEO website under the accession number GSE23950 . Total RNA was extracted using Plant RNA Purification Reagent ( Invitrogen ) and subjected to cDNA synthesis using the QuantiTect Reverse Transcription Kit ( Qiagen ) , according to the manufacturer's instructions . The PCR conditions were as follows; pre-incubation for 5 min at 94°C , 30 cycles at 94°C for 30 sec , 58°C for 20 sec , 72°C for 40 sec and a final extension at 72°C for 4 min . Primers are listed in Table S8 . The amplified DNA was visualized on a 2% agarose gel stained with ethidium bromide . Antibodies against HDA6 were generated as follows; a peptide ( DEMDDDNPEPDVNPPSS ) corresponding to the C-terminus of HDA6 was synthesized , HPLC purified , conjugated to Bovine Serum Albumin ( BSA ) and used to immunize two rabbits ( Scrum ) . The antiserum obtained was affinity-purified and used for ELISA and western blot analysis . Total protein extraction was performed on 15-day-old seedlings . Seedlings were ground in liquid nitrogen , suspended in PBS supplemented with 1 mM PMSF , centrifuged , and the supernatant used as a total protein extract . The protein concentration was analyzed using the BioRad Bradford reagent and 50 ug of protein was used for western blot analysis . Western blots prepared by the iBlot Dry Blotting system ( Invitrogen ) were blocked and incubated with the HDA6 antibody diluted at 1∶500 , washed , and incubated with anti-rabbit IgG HRP-conjugated antibodies ( GE Healthcare ) diluted 1∶5000 . The results were visualized using ECL Plus Western Blotting Detection Reagents ( GE Healthcare ) . ChIP assays were performed essentially as described previously [61] . The antibodies used in this study were: anti-H3K4me3 and H3K9me2 [62]; anti-H3K9ac ( ab4441 ) and H3K14ac ( ab1191 ) from Abcam; anti-H4 tetra-acetylation ( 06-866 ) , H3K27me3 ( 07-449 ) , H3K18ac ( 07-328 ) , H3K23ac ( 07-355 ) , H3K27ac ( 07-360 ) , H4K5ac ( 07-327 ) , H4K8ac ( 07-328 ) , and H4K12ac ( 07-595 ) from Millipore , and H4K16ac ( CB-SC-8662-R ) from Santa Cruz . The precipitates were analyzed with quantitative PCR ( Power SYBR real time reagent and ABI Prism 7000 , Applied Biosystems ) and the relative amount of each modification was estimated as described previously [63] . Statistical significance of the wild-type plants compared with axe1-5 was determined by Kruskal–Wallis test ( P<0 . 05 ) . The primers used are listed in Table S8 . Genomic DNA was extracted using a Phytopure DNA extraction kit ( GE Healthcare ) and 5 µg of genomic DNA was linearlized with 20 U BamHI for 3 hours at 37°C . McrBC assays were performed by incubating 30 U of McrBC per 1 µg of BamHI digested genomic DNA at 37°C for 16 hours before PCR amplification as described for RT-PCR with a 1 min extension time . Chop-PCR assays [30] were performed using the methylation sensitive restriction enzymes HpaII , MspI , and HaeIII ( NEB ) . Linearlized genomic DNA was incubated with the enzymes ( 30 U/µg ) at 37°C for 3 hours and subjected to PCR analysis . The amplified DNA was visualized on a 1 . 0% agarose gel stained with ethidium bromide .
|
Eukaryotes are defended from potentially harmful DNA elements , such as transposons , by forming inactive genomic structure . Chromatin , which consists of DNA and histone proteins , is densely packed in the silent structure , and chromatin chemical modifications such as DNA methylation and histone modifications are known to be essential for this packing . In plants , small RNA molecules have been thought to trigger DNA methylation and resulting silent chromatin formation . We revealed that elimination of specific histone modifications concomitant with DNA methylation is pivotal for the silent chromatin . Furthermore , the histone deacetylase was shown to have more profound target specificity than the DNA methyltransferase and is required for locus-directed DNA methylation , implying the involvement of the histone deacetylase for targeting the DNA methyltransferase to specific places on the genome . These proteins and their functions for gene silencing are evolutionarily conserved in higher eukaryotes , and several proteins involved in small RNA production are plant-specific . Thus , we present a hypothesis that the plant genome may build the protecting foundation by the conserved genome surveillance in eukaryotes , and the reinforcing machinery involving small RNAs could be evolutionarily added to the plant heterochromatin silencing system .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome",
"expression",
"analysis",
"rna",
"interference",
"plant",
"biology",
"histone",
"modification",
"plant",
"science",
"plant",
"genomics",
"epigenetics",
"chromatin",
"transposons",
"gene",
"expression",
"plant",
"genetics",
"biology",
"dna",
"modification",
"molecular",
"biology",
"genetics",
"genomics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2011
|
Arabidopsis HDA6 Regulates Locus-Directed
Heterochromatin Silencing in Cooperation with MET1
|
Rapid reinfestation of insecticide-treated dwellings hamper the sustained elimination of Triatoma infestans , the main vector of Chagas disease in the Gran Chaco region . We conducted a seven-year longitudinal study including community-wide spraying with pyrethroid insecticides combined with periodic vector surveillance to investigate the house reinfestation process in connection with baseline pyrethroid resistance , housing quality and household mobility in a rural section of Pampa del Indio mainly inhabited by deprived indigenous people ( Qom ) . Despite evidence of moderate pyrethroid resistance in local T . infestans populations , house infestation dropped from 31 . 9% at baseline to 0 . 7% at 10 months post-spraying ( MPS ) , with no triatomine found at 59 and 78 MPS . Household-based surveillance corroborated the rare occurrence of T . infestans and the house invasion of other four triatomine species . The annual rates of loss of initially occupied houses and of household mobility were high ( 4 . 6–8 . 0% ) . Housing improvements did not translate into a significant reduction of mud-walled houses and refuges for triatomines because most households kept the former dwelling or built new ones with mud walls . Our results refute the assumption that vector control actions performed in marginalized communities of the Gran Chaco are doomed to fail . The larger-than-expected impacts of the intervention program were likely associated with the combined effects of high-coverage , professional insecticide spraying followed by systematic vector surveillance-and-response , broad geographic coverage creating a buffer zone , frequent housing replacement and residential mobility . The dynamical interactions among housing quality , mobility and insecticide-based control largely affect the chances of vector elimination .
Neglected tropical diseases ( NTDs ) stem from the complex interactions among social , economic , political , cultural and environmental determinants acting at various temporal and spatial scales [1] . These complex interactions explain why some groups of people , including indigenous communities and the rural poor , are most affected by NTDs and frequently suffer co-infections as part of the cycle of poverty [2 , 3] . A case in point is the Gran Chaco ecoregion , including sections of Argentina , Bolivia and Paraguay home to numerous indigenous peoples [4–6] . This NTD hotspot is characterized by high prevalence rates of human infection with Trypanosoma cruzi ( Chagas disease ) , geo-helminthiases , and unsatisfied basic needs [7] . High levels of house infestation with Triatoma infestans , historically the main vector in the Southern Cone countries , are still present in sections of the Gran Chaco despite of the multiple insecticide-based control campaigns conducted over nearly 70 years [5 , 8–13] . In the absence of effective vaccines and given the limitations of current drugs for massive chemotherapy , Chagas disease prevention efforts have historically relied on residual insecticide spraying and screening of blood donors . Vector control campaigns contracted the geographic range of T . infestans [14–16] and suppressed the domestic transmission of T . cruzi mediated by T . infestans in various countries and provinces since the mid-1990s [6 , 15] . However , persistent house reinfestation after insecticide application fueled by peridomestic foci in different areas of the Argentine Chaco [17] , combined with growing evidence of the existence of sylvatic foci of T . infestans [18–21] and the emergence of pyrethroid insecticide resistance ( reviewed in [22] ) , cast doubts on how feasible is to achieve the elimination of T . infestans in the Gran Chaco [23] , one of the initial goals of the Southern Cone Initiative in 1991 . Lack of consistent state policies and of a sustainable vector surveillance-and-response system in resource-constrained rural areas , among other factors , contribute to a persistent or recurrent public health issue [4] . The widespread problem of native , highly competent triatomine species that reinfest insecticide-treated dwellings ( e . g . , [24] ) should be addressed by an integrated vector management strategy [25] that considers the eco-bio-social determinants of health vulnerability [26] and house infestation with triatomines [27–29] . Two major components of such strategy are community participation and housing improvement . Community participation may assist in the design and implementation of locally adapted , sustainable and more effective vector control strategies , especially in remote , deprived areas [30] . Housing improvement [25] tends to reduce triatomine infestation in domestic and peridomestic structures [29 , 31–36] . The historical trend toward rural housing improvement , albeit at a widely different pace across regions and countries , suggest any assessment of the effectiveness of vector control actions should also encompass structural determinants of infestation such as housing quality , type of occupancy , and host availability over time . As part of a longitudinal program on the eco-epidemiology and control of T . infestans and Chagas disease in the Argentine Chaco , we detected higher-than-expected post-spraying house infestation rates and apparent residual foci related to moderate ( incipient ) pyrethroid resistance in two large rural sections ( denominated Areas I and II ) of Pampa del Indio having 400–500 houses each [37 , 38] . A residual focus is defined as a post-spraying infestation derived from triatomines that survived the insecticide spraying at house level . These patterns suggested that similar rates of house reinfestation would occur in another large rural section ( Area III ) where the same community-wide insecticide spraying interventions were simultaneously implemented . All these rural areas had poor housing conditions suitable for triatomines , and high infestation with T . infestans before control actions . Seroprevalence of human T . cruzi infection averaged 30% [39] . Two of the rural sections were mainly inhabited by an indigenous people ( Qom ) that displayed intense residential mobility , especially within Area III [40] . Unexpectedly , the intervention there exerted immediate impacts on house infestation as revealed by timed-manual searches; the few foci detected after blanket insecticide spraying were mainly assigned to external sources as determined by wing geometric morphometry [41] . Identifying processes and mechanisms underlying the successful vector control status achieved in Area III is relevant to the goals of suppressing the major vectors of Chagas disease and interrupt domestic transmission , in concert with the Sustainable Development Goals and the London’s Declaration on NTDs [42] . Here we extend the follow-up of Area III to further investigate the house reinfestation process in connection with baseline pyrethroid resistance , housing dynamics ( construction , destruction , abandonment and improvement ) and household mobility over a seven–year period . Instead of assuming static environmental conditions and focusing only on vector-related issues , we adopted a more systemic approach and analyzed other household-level , time-variable processes related to the dynamics of housing quality ( habitat ) and residential mobility , which also affects host availability . Population movement is relevant to the control of NTDs and other infectious diseases in a context-specific manner [43] , but the very few studies that examined this subject in connection with Chagas disease vectors did it at a village level [44] or in urban-type settings [45–46] . Our study provides evidence of sustained vector control despite moderate pyrethroid resistance in T . infestans populations; corroborates the virtual absence of post-spraying foci and the house invasion of other four triatomine species through community-based surveillance , and links the lack of post-spraying foci to frequent housing replacement , residential mobility , and broad geographic coverage of sustained vector control .
Fieldwork was conducted in a 95 km2 rural section of Pampa del Indio municipality ( 25° 55’ S 56° 58’ W ) , Chaco province , Argentina , denominated Area III . The rural area of the municipality was divided in four sections ( I-IV ) in which a similar intervention protocol was performed ( Fig 1 ) . Area III included 404 occupied ( inhabited ) house compounds in seven communities where Qom households predominated over a Creole minority as of October 2008 ( baseline ) . The main features of the study area and its population were described elsewhere [40] . The study houses were surrounded by agricultural fields mixed with patches of native forest subject to variable degrees of degradation . No significant environmental changes were observed in the study communities during the follow-up . Official census information at the municipality scale indicated a very large annual population growth rate ( 4 . 9% ) over the 2001–2010 period . An average Qom household had 6 . 4 occupants in a 43 . 4-m2 sleeping quarter ( domicile ) , whereas local Creoles had 4 . 1 occupants in a nearly twice as large area . Houses lacked access to safe water and a sewage or garbage disposal system within their premises [40] . The last insecticide spraying campaign targeting house infestation with T . infestans in Pampa del Indio municipality had taken place in 1997–1998 . Additionally , a few clusters of Area III houses were selectively sprayed by local healthcare personnel in 2000 ( 36 houses ) , 2006 ( 49 houses ) , and July 2008 ( 45 houses from two communities ) . Although these actions were not recorded by the provincial Chagas control program , local householders’ reports to the research team agreed with the information provided by local healthcare agents [40] . A longitudinal study was conducted to monitor house infestation with triatomines before and after community-wide spraying with insecticides ( Table 1 ) and any parallel change in housing structure and household composition . Our usage of household- and housing-related terms follow definitions appearing elsewhere [47 , 48] . The condition of each house unit was registered at each vector survey and classified as follows: i ) occupied: inhabited; ii ) vacant: uninhabited , with clear signs of current vacancy , often corroborated by neighbors; iii ) closed: showing signs of being occupied , often confirmed by neighbors , with residents absent after one to three re-visits; iv ) demolished: no longer existing at its exact previous location , and v ) new house unit: no previous record at its exact current location . A household was defined as the people ( related or not ) sharing food and other housekeeping activities despite the fact that sometimes there were two or more structurally separate sleeping quarters occupied by related family ( i . e . , extended family ) . In our specific context one household equaled one house unit . A house compound consists of one or more separate human sleeping quarters ( domicile or domestic premises ) , a patio and nearby peridomestic buildings for human use and animals ( e . g . , kitchen , storeroom and chicken coop , among others ) as illustrated elsewhere [40 , 49] . The study protocol was approved by the Dr . Carlos A . Barclay Independent Ethical Committee for Clinical Research , Buenos Aires , Argentina ( IRB No . 00001678; Protocol N ° TW-01-004 , Revision N° 863-32-2011 ) . During our first visit to the area , we explained the purpose of the study to the heads of each household invited them to participate and asked them for an oral consent . A small aluminum plate was nailed to the front door to identify each house unit for detailed follow-up . The geographic location of each house compound was recorded with a GPS ( Garmin Legend ) and maps were generated with QGIS 2 . 18 . 16 . The geographical coordinates of each house compound were transformed to preserve the privacy of the households involved in this study , as described elsewhere [40] . All house compounds were inspected for triatomines by timed-manual searches using a dislodjant aerosol ( 0 . 2% tetramethrin ) ( Espacial , Argentina ) before ( October 2008 ) and periodically after community-wide insecticide spraying ( Table 1 ) : August 2009 ( 10 MPS ) , April 2010 ( 18 MPS ) , November 2012 ( 49 MPS ) and April 2015 ( 78 MPS ) . Each domicile and peridomestic structure at each compound was searched by one person for 15 min; searches and insecticide applications were performed by skilled personnel from the national or provincial vector control programs [40] . One member of the research team was present in almost every inspected or sprayed house during the entire follow-up . Closed houses were re-visited at least once , and sometimes two or three times , to achieve full-inspection coverage at each survey . Public buildings and vacant houses were inspected for triatomine infestation when they were occasionally used as a dwelling or as resting places by free-ranging domestic animals , respectively; because none of them were ever found infested , they were excluded from house infestation estimates . T . infestans was found by timed-manual searches in 28 . 0% of the occupied houses inspected at baseline , and in 31 . 9% of them when all vector detection methods were considered [40] . Triatomine abundance was calculated as the number of bugs collected by timed-manual searches per unit of catch effort . Colonization was defined as the presence of any nymphal instar among the triatomines collected . The community-wide insecticide spraying campaign covered 96 . 0% of all occupied houses in October 2008 ( 0 MPS ) ( Table 1 ) . Houses infested with T . infestans over 2009–2015 were selectively sprayed with pyrethroid insecticide immediately after each survey . In order to promote community-based vector surveillance , householders were asked for the presence of triatomines in their dwellings and shown dry specimens of the local species at every vector survey , and were instructed to collect any triatomine they may find and bring it to the local healthcare post or hospital . This information was used to build two house infestation indices: householders’ bug collection ( when they delivered triatomines ) and householders’ bug notification ( when only reports were given ) . In October 2011 ( 36 MPS ) and September 2013 ( 59 MPS ) , local healthcare agents aimed at visiting every household to ask residents for the presence of triatomines in their dwellings; no dry triatomine specimens were shown . Research team members surveyed additional households on December 2011 ( 38 MPS ) to increase the coverage of residents’ notifications; in total , 85% of all occupied house units ( 373 ) were surveyed over October-December 2011 . During these visits , timed-manual searches were performed in a selected sample of the occupied houses that either reported the presence of triatomines and/or had a previous infestation . Additionally , a random sample of dwellings with suitable conditions for infestation ( i . e . , mud or unplastered walls , tarred-cardboard roofs , high refuge availability , large household size , as defined below ) were also inspected by timed searches . In September 2013 ( 59 MPS ) , 200 houses ( 48% of all occupied house units ) were visited by local healthcare workers to canvas householders on house infestation; timed-manual searches were later performed following the same criteria described above . All triatomines collected during the follow-up were carried to the field laboratory for identification to species , stage and sex , and then were preserved at -20°C . Feces of all live third-instar nymphs and later stages were examined for infection with T . cruzi at 400× within 2–4 weeks of bug collection as described elsewhere [50] . Detailed socio-demographic and environmental surveys were performed in parallel to vector surveys in October 2008 ( 0 MPS ) , November 2012 ( 49 MPS ) and April 2015 ( 78 MPS ) . At each occupied house , an adult household member fluent in Spanish was asked for information on demographic ( i . e . , number of residents by age class and gender ) , economic ( i . e . , number and type of domestic animals , and their resting places ) and environmental variables ( e . g , construction features , insecticide use ) to monitor changes over time and investigate their association with house infestation [40] . For each domicile , the building materials used in roof and walls , presence and type of wall plaster , condition of wall surface , type of floor , and number of sleeping quarters were recorded in a form . The availability of refuges for triatomines was categorized into five levels by a skilled member of the research team [49] . Additional socio-demographic variables were registered from 49 MPS on , including land ownership , educational level of each household member , and whether a new domicile or house was provided by a government-sponsored rural housing program or not . These data were used to compute household educational level , defined as the mean number of schooling years attained by household members aged 15 years old or more , and the overcrowding index , defined as the number of human occupants per sleeping quarter [40] . Housing improvement referred to houses with mud-walled domiciles ( or more rarely , built with other materials such as wood or plastic ) that shifted to having one or more new human sleeping quarters with brick-and-cement walls and a corrugated metal-sheet roof ( the only type of improved roof recorded , frequently denominated modern houses ) , regardless of whether they had been provided by the rural housing program , house residents or any third party . These changes were assessed in relation to the status at the preceding survey in which housing-related variables were registered ( 0 , 49 and 78 MPS surveys ) . When more than one building material or more than one domicile existed at baseline , the maximum quality level qualified the status of the house unit . For example , if a house had one mud-walled domicile and a second one with brick-and-cement walls , the house unit was taken to have brick-and-cement walls . Only stable houses provided the data needed for this analysis ( see definitions below ) . The demographic composition and location of each household , and the residential destination of those who moved elsewhere between surveys and their main reasons , were recorded at 49 and 78 MPS . These data were used to classify household mobility for the entire follow-up ( 0–78 MPS ) and for each study period ( 0–49 MPS and 49–78 MPS ) , and varied slightly from the classification we used previously [40]: movers were households that changed its exact residential location within Area III ( i . e . , local movers ) ; non-movers were households that remained at the exact residential location; out-migrants were households that changed its residential location to ( peri- ) urban sections of the municipality or other cities within the country; new households , those established in Area III during the follow-up ( owing to the formation of a new family or by in-migration , regardless of their origin ) , and households that ceased to exist ( owing to separation , demise of the only resident , or permanent out-migration from Area III ) . Housing stability between any two surveys ( 0 , 49 and 78 MPS ) was classified in three levels: stable ( i . e . , a permanently occupied house unit ) , non-stable ( a demolished or vacant house unit ) , and new ( i . e . , a newly-built house unit , regardless of the source , type or state of building materials ) . A sample of the T . infestans collected at baseline ( 64 females from 22 ( 18% ) house units in 5 of the 7 infested communities ) was tested for pyrethroid resistance at the Center for Research on Plagues and Insecticides ( CIPEIN/CONICET , Buenos Aires , Argentina ) using standardized methods [51] . Screening coverage ranged between 13% and 25% of the infested houses across communities . The screening bioassays consisted in the application of a discriminating dose ( DD ) of 2 ng of technical-grade deltamethrin ( 99% ) , provided by Ehrenstorfer ( Augsburg , Germany ) per insect , which causes 99% mortality of the susceptible strain ( DL99 ) [52] . First-instar nymphs of T . infestans ( 5–7 days old , mean weight 1 . 3 ± 0 . 2 mg , unfed ) ( F1 ) received a topical application of 0 . 2 μl of deltamethrin ( 0 . 01 mg/ml ) in analytical grade acetone ( Merck , Buenos Aires , Argentina ) on the dorsal abdomen using a 10-μl Hamilton syringe ( Hamilton PB-600-1 , Nevada , USA ) with an automatic repeating dispenser [51 , 79] . The treated insects were kept inside a plastic container with folded paper at 28–30 °C and 50–70% RH . Mortality was evaluated after 24 h by placing the insects at the center of a circular filter paper of 11 cm diameter; nymphs able to walk to the border were taken as survivors [51] . The bioassays consisted of three replicates containing at least 10 insects for each bug population . A laboratory-reared , deltamethrin-susceptible colony of T . infestans ( denominated CIPEIN SRL , susceptible reference lineage ) , was used as a negative control , and a laboratory-reared pyrethroid-resistant T . infestans colony ( from Salta , Argentina ) was used as a positive control . A triatomine population was considered resistant to the insecticide tested if mortality was less than 90% in two out of three independent assays , i . e . , at least one survivor in two of 3 trials [52] . When the number of first-instar nymphs available only allowed one or two independent assays to be performed , the outcome was taken to be resistant or susceptible pending confirmation . Houses were grouped according to median mortality in the bioassays as follows: susceptible ( 91–100% ) , moderate ( 76–90% ) , reduced ( 50–75% ) . No sample showed mortality rates fewer than 50% . The percentage of new , occupied house units at the survey conducted at time t was calculated relative to the total number of occupied house units enumerated at t . The percentage of demolished houses at t was computed from the number of occupied houses enumerated at t-1 that ceased to exist at t relative to the number of occupied houses enumerated at t-1 . All statistical analyses were conducted using Stata 15 . 1 [53] . The distribution of household characteristics over the follow-up was examined with χ2 tests . The Kruskal-Wallis test was used for comparison of household size over time . The agreement between householders’ notifications of triatomine presence and direct assessments of house infestation was measured using the kappa index . An exact Fisher’s test was used to examine the association between house infestation status before and after insecticide spraying . A Cox regression survival analysis was used to test whether the loss rate of house occupancy differed between infested and non-infested houses at baseline . Global spatial analyses ( univariate and bivariate ) were performed using the weighted K-function implemented in Programita [54] . Random labeling was selected to test the null hypothesis of random occurrence of events among the fixed spatial distribution of all houses . The selected cell size was 200 m ( assuming that each house had at least three neighbors at the minimum distance of analysis ) , and the maximum distance was set at 6 km ( i . e . , half of the dimension of the area ) [55] . Monte Carlo simulations ( n = 999 ) were performed and the 95% ‘confidence envelope’ was calculated with the 2 . 5% upper and lower simulations . We created heat maps ( i . e . density maps ) to visualize the spatial aggregation of housing instability ( i . e . , non-stable houses ) over 0–78 MPS . The analysis was implemented in QGIS using a kernel density estimation algorithm within a radius of 200 m .
The total number of houses increased from 411 to 485 in a roughly linear fashion over the seven-year period , as did the frequency of occupied , new and demolished houses ( Fig 2A and 2B ) . The observed frequencies and linear regression equations appear in S1 Table . In total , 30 . 7% of the occupied houses at baseline were lost ( i . e . , vacant or demolished ) over the follow-up; the annual loss rate was 4 . 6% ( standard error , 0 . 36 ) as determined by ordinary linear regression ( F = 168 . 1 , df = 1 and 3 , p < 0 . 001 , adj R2 = 0 . 977 ) . Among the 123 baseline-infested houses ( as determined by any collection method ) , 22 . 0% and 32 . 5% subsequently became vacant or were demolished at 49 and 78 MPS , respectively , whereas 19 . 9% and 28 . 6% of the 266 non-infested houses at baseline were no longer occupied at those time points , respectively ( Fig 2C ) ( Cox regression test , χ2 = 0 . 2 , df = 1 , p = 0 . 7 ) . The prevalence of house units having at least one domicile with mud walls decreased marginally from 78 . 9% to 72 . 4% over 0–78 MPS ( χ2 = 4 . 9 , df = 2 , p = 0 . 09 ) , whereas the same metric for brick-and-cement walls increased highly significantly from 25 . 8% to 41 . 9% ( χ2 = 24 . 5 , df = 2 , p < 0 . 001 ) ( Table 2 ) . These changes were closely related to a government-sponsored rural housing program , which increasingly covered up to 18 . 6% of existing houses as of 78 MPS and benefited 77 Qom and two creole households . This increase in the proportion of brick-and-cement walled domiciles was even observed in the group of baseline-infested houses ( S1 Text ) . Nonetheless , most of the new house units ( 236 ) built during the follow-up had unplastered mud-walled domiciles ( S1 Text ) . For all houses with an improved domicile at 78 MPS , most of the new premises had plastered walls ( 73 . 3% ) and absence of cracks ( 81 . 7% ) , whereas 66–94% of them also retained the former mud-walled structure at 49 and 78 MPS . Almost 60% of all mud-walled houses had unplastered walls as of 78 MPS . The trend toward housing improvement was also expressed in the highly significant drop in houses having tarred-cardboard roofs , from 52 . 9% to 31 . 6% over 0–78 MPS ( χ2 = 40 . 5 , df = 2 , p < 0 . 001 ) , which were replaced by corrugated metal-sheet roofs ( Table 2 ) . In spite of these improvements , the mean score of domestic refuge availability for triatomines remained high and tended to increase ( Kruskal-Wallis test , χ2 = 6 . 8 , df = 2 , p = 0 . 03 ) . Householders’ application of domestic insecticides mainly included low-concentration pyrethroid sprays; their frequency of use highly significantly increased over 0–49 MPS ( χ2 = 26 . 4 , df = 1 , p < 0 . 001 ) . Household size averaged 6 residents and remained approximately constant ( Kruskal-Wallis test , χ2 = 4 . 7 , df = 2 , p = 0 . 09 ) , as did residential overcrowding and household educational level . Most households ( >78 . 5% ) had stable residence over 0–49 and 49–78 MPS ( Table 3 ) . Movements mainly occurred within the study area ( i . e . , local movers , ~11% ) , and slightly less often included out-migration to large cities or ( peri- ) urban areas of Pampa del Indio ( 7 . 5–9 . 2% ) . The annual rate of household mobility averaged 5 . 1% and 8 . 0% over 0–49 and 49–78 MPS , respectively . Most movers comprised Qom households ( 96 . 0% , 95/99 ) , who torn down ( 77 . 8% ) their former dwelling and occupied a new house ( 98 . 3% ) rather than an existing one . Significantly fewer ( 54 . 9% , 39/71 ) out-migrant households demolished their houses compared to movers ( χ2 = 10 . 0 , df = 1 , p = 0 . 002 ) . Nearly half ( 45 . 3–51 . 0% ) of the new houses were occupied by movers ( Table 3 ) . The formation of new households from current Area III residents and from in-migrants substantially increased ( by 3 . 9–11 . 6× ) over 49–78 MPS relative to 0–49 MPS . Most in-migrants ( 86% , 18/21 ) occupied a new house . Overall , 47% of the households that changed their residential location over 49–78 MPS had also moved or migrated at least once over 0–49 MPS ( repeat movers ) . House infestation decreased from 31 . 9% to 0 . 7% at 10 MPS and then remained marginal , with no infested house detected at 59 and 78 MPS ( Fig 3A ) . Vector surveys inspected 93 . 8–96 . 3% of the occupied house units over the follow-up; less than 1 . 6% of them refused searches for triatomines ( S1 Table ) . No public building or vacant house was ever found infested with T . infestans . The few occupied houses that were neither inspected for infestation nor sprayed with insecticides at baseline were subsequently inspected for triatomines at least once , and none of them was positive for T . infestans except one at 10 MPS ( which had neither been inspected nor sprayed at baseline due to its remote location ) . The apparent increase in house infestation ( 2 . 3% , only including two infested houses ) at 38 MPS was related to targeted searches of higher-risk dwellings . Pre- and post-spraying infestation mostly occurred in human sleeping quarters . Only 10 ( 1 . 6% ) of 615 occupied houses inspected at least once over the follow-up were positive for T . infestans by timed-manual searches , and 9 of them had infested domestic premises . The median relative abundance per infested collection site over the follow-up was low ( 2 . 0 triatomines; first-third quartiles , 1–3 per unit of catch effort ) , and peaked at 18 MPS owing to a high-density triatomine population in a peridomestic habitat used by chickens ( Fig 3A ) . Colonies of T . infestans were found in 8 of the 10 infested houses as determined by any detection method . The stages most frequently captured by timed-manual searches were females ( 33% ) and males ( 17% ) . Triatoma sordida ( a secondary vector candidate for domestication ) was found by timed-manual searches in 4 . 4% of the occupied houses at baseline , and then fluctuated between 1 . 7% and 5 . 3% over the follow-up ( excluding surveys in which only a sample of houses was inspected ) ( Fig 3B ) . T . sordida mainly occurred in habitats used by chickens , and was collected inside a human sleeping quarter or a kitchen ( both adults and nymphs ) only in two occasions . The median relative abundance per infested collection site was 7 triatomines ( first-third quartiles , 1–14 per unit of catch effort ) over the follow-up; colonies were found in 90 . 2% of all infested sites as determined by any method . Fifth- ( 32% ) and fourth-instar nymphs ( 23% ) were the stages most frequently captured by timed searches . Baseline and post-spraying house infestation with T . infestans were not significantly associated among 379 houses inspected for triatomines on both periods ( exact Fisher’s text , p = 0 . 683 , df = 1 ) ( S2 Table ) . Three of the post-spraying infested houses had also been infested at baseline , but in between these occasions there were 1 or 2 surveys in which no T . infestans was collected by any method , suggesting post-spraying infestations were unlikely to be residual foci . The three infested houses at 10 MPS ( two baseline-negative and one baseline-no data ) were not strictly defined residual foci because they had not been sprayed with insecticides at baseline , although two of them might have been residual foci from a prior non-professional spraying . The remainder was neither inspected nor sprayed at baseline due to its remote location ( footnotes 3 and 4 in S2 Table ) . The space-time series of observations and subsequent interviews to householders suggested putative cases of passive or active dispersal of T . infestans among infested dwellings of one community ( Pampa Grande ) . Four of the five infested houses detected at 38 and 49 MPS were located in a well-defined sector in which only one house was heavily infested ( i . e . , the putative index case ) . Householders reported that one resident of the heavily infested house moved temporally with his belongings to two nearby houses , which subsequently appeared infested , suggesting passive bug transport . The other infested house was at 700 m of the heavily infested one , and within the flight range of T . infestans . Three of the 10 infested house units detected over the follow-up harbored at least one T . infestans infected with T . cruzi whereas this fraction ( 34/75 , including only infested houses with bug infection data ) was higher at baseline though not significantly so ( Fisher’s exact test , p = 0 . 5 ) . This downward trend holds both in domiciles ( 33% , 3/9 vs . 45% , 28/62 ) and peridomiciles ( 0% , 0/3 vs . 44% , 8/18 ) , respectively . The overall infection prevalence in post-spraying T . infestans was 11% ( n = 44 insects examined ) . None of the 42 T . sordida collected from 6 houses at 10 MPS were infected with T . cruzi . Householders’ notifications of the presence of any triatomine species were significantly more frequent than the direct assessments by any detection method at every post-spraying survey ( χ2 test , df = 1 , p < 0 . 01 ) , except at 10 MPS ( χ2 test , df = 1 , p = 0 . 4 ) ( Fig 4 ) . There was a poor agreement between post-spraying householders’ notifications and other vector detection methods ( kappa coefficients < 0 . 2 ) . Nonetheless , the relative odds of post-spraying house infestation with T . infestans was approximately 10–13 times higher when householders notified the presence of any triatomine than when they did not ( 10 MPS: OR , 10 . 1; 95% CI: 0 . 9–118 . 5; 18 MPS: OR , 12 . 6; 95% CI: 0 . 7–211 . 9 , and 49 MPS: OR , 11 . 7; 95% CI: 1 . 03–134 . 3 ) . Householders’ notification of T . infestans also dropped below 1% at 10 MPS , as did timed-manual searches , and then increased at 49 MPS ( 6 . 5% ) and 78 MPS ( 3 . 6% ) ( Fig 4 ) , matching the frequent catch of T . sordida by timed searches in both surveys ( Fig 3B ) and householders’ collections of T . sordida or other triatomines at 49 MPS . Householders reported the presence of T . infestans before timed searches in five of the 10 houses ever found infested post-spraying; the remainder had either a peridomestic infestation only or very low domestic bug abundance . Householders collected 45 triatomines at 25 different dwellings and other Reduviidae bugs at 15 houses over the surveillance phase ( S1 Text ) . In total , 55% of the 22 populations of T . infestans individually screened for pyrethroid resistance had reduced mortality ( Table 4 ) . None of the post-spraying infested houses had information on pyrethroid resistance at baseline . Moreover , none of the 12 houses with reduced mortality to pyrethroids was found infested after the community-wide spraying . Housing instability was mainly concentrated in a few clusters of houses , as was housing improvements . Both events partially overlapped , particularly in a section with high-density residential mobility and housing improvement ( Fig 5A and 5B ) . The spatial distribution of non-stable houses and of improved houses did not differ significantly from a random pattern . However , a qualitative spatial overlap between a high density of improved houses , housing instability and baseline house infestation is apparent ( Fig 5A ) , and wanes for post-spraying house infestation ( Fig 5B ) . No significant association was found between post-spraying house infestation and household instability or housing improvements in univariate analyses ( χ2 = 0 . 5 , df = 1 , p = 0 . 5; χ2 = 0 . 8 , df = 1 , p = 0 . 4 , respectively ) . Both the houses with any evidence of pyrethroid resistance and those screened for the latter were dispersed throughout the study area ( Fig 5C ) , showing no apparent spatial association with post-spraying infestation . The relationship between post-spraying domestic infestation with T . infestans , housing stability and housing improvement is shown in Table 5 . Non-stable and stable houses displayed virtually the same prevalence of domestic infestation , which was twice as large as that of new houses . None of the 29 houses subjected to improvement became infested . When differential exposure time was accounted for , non-stable , non-improved houses had a much higher rate of domestic infestation than stable , non-improved houses ( ~×4 ) or new houses ( ~×2 ) . This ranking was not affected by whether the three putative residual foci were excluded or not ( footnote of Table 5 ) .
Both household mobility and housing quality are key to understand house infestation dynamics in the context of vector control interventions and demographic change , and the details of their interaction with other factors matter . For example , whether a vacant house is immediately re-occupied by newcomers ( rather than being torn down ) , or whether movers always build the new house from scratch will determine the fate of pre-existing infestations; so does whether housing improvements are combined with insecticide spraying or not . These aspects pose additional challenges to traditional housing and vector control programs since for instance , many sprayed ( presumably protected ) houses disappear while new ( unprotected ) houses built with used or new materials emerge . The relationship between housing instability , household mobility and infestation requires further investigation in different endemic and cultural settings . Underlying these patterns , our study discloses a chronic housing crisis in rural areas of the Argentine Chaco , tightly linked to one of the goals in the 2030 Agenda for Sustainable Development: ensure access for all to adequate , safe and affordable housing and basic services . Which vector surveillance strategy is appropriate for scenarios such as those in the rural Chaco ? In the context of low house infestation levels after interventions over extended , at times inaccessible areas , annual house searches for triatomines are neither cost-effective [10 , 86 , 87] nor sustainable [8] . One of the challenges vector control programs face there is how to sustain the initial progress in the face of their limited operational capabilities and the competing demands posed by recurring outbreaks of dengue and other mosquito-borne diseases [4] . Mixed approaches including community-based vector surveillance showed promising results [5 , 33 , 34 , 36 , 72 , 86–88] . Early detection of house infestation with triatomines and prompt insecticide treatment can be achieved through coordinated local efforts among householders , the primary healthcare system and other grassroots organizations , especially when rural communities are disperse and access is difficult . School-based health education interventions may assist in fostering community participation in vector surveillance and control . A crucial point in such community-based programs is related to providing an appropriate response to householders’ notifications ( service delivery ) : who will provide gear , insecticide and apply it properly [5 , 88] . The implementation , maintenance and supervision of the vector surveillance-and-response system are essential for long-term disease control in high-risk contexts [20 , 88 , 89] .
|
Efforts to prevent the transmission of the parasite that causes Chagas disease have been directed at eliminating its insect vector species from human dwellings via insecticide applications . The outcome of these interventions has usually been measured through vector-related metrics in prospective studies ranging up to a few years . Longer-term intervention studies that additionally monitor other social determinants of house infestation , such as housing quality and population movement , are lacking . Our seven-year study addresses this gap in a remote rural area of the Argentine Chaco inhabited by deprived indigenous communities . We show sustained vector control headed towards local elimination despite the highly adverse social context and the occurrence of moderate pyrethroid resistance in vector populations . Housing-quality dynamics and household mobility displayed complex patterns that may affect domestic triatomine populations . Household mobility within the area was intense: movers usually torn down their previous precarious houses and built new , equally precarious houses . Housing dynamics included structural improvements , disappearance and construction of new dwellings elsewhere within the study area . We infer that these eco-bio-social factors , including the broad geographic coverage of sustained vector control creating a peripheral buffer zone , substantially increased the long-term effectiveness of the intervention program .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"demography",
"tropical",
"diseases",
"social",
"sciences",
"anthropology",
"parasitic",
"diseases",
"rural",
"areas",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"insect",
"vectors",
"housing",
"human",
"geography",
"agrochemicals",
"geography",
"infectious",
"diseases",
"indigenous",
"populations",
"protozoan",
"infections",
"disease",
"vectors",
"agriculture",
"insecticides",
"people",
"and",
"places",
"chagas",
"disease",
"earth",
"sciences",
"geographic",
"areas",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"triatoma"
] |
2018
|
Beating the odds: Sustained Chagas disease vector control in remote indigenous communities of the Argentine Chaco over a seven-year period
|
Malaria-transmitting mosquitoes are continuously exposed to microbes , including their midgut microbiota . This naturally acquired microbial flora can modulate the mosquito's vectorial capacity by inhibiting the development of Plasmodium and other human pathogens through an unknown mechanism . We have undertaken a comprehensive functional genomic approach to elucidate the molecular interplay between the bacterial co-infection and the development of the human malaria parasite Plasmodium falciparum in its natural vector Anopheles gambiae . Global transcription profiling of septic and aseptic mosquitoes identified a significant subset of immune genes that were mostly up-regulated by the mosquito's microbial flora , including several anti-Plasmodium factors . Microbe-free aseptic mosquitoes displayed an increased susceptibility to Plasmodium infection while co-feeding mosquitoes with bacteria and P . falciparum gametocytes resulted in lower than normal infection levels . Infection analyses suggest the bacteria-mediated anti-Plasmodium effect is mediated by the mosquitoes' antimicrobial immune responses , plausibly through activation of basal immunity . We show that the microbiota can modulate the anti-Plasmodium effects of some immune genes . In sum , the microbiota plays an essential role in modulating the mosquito's capacity to sustain Plasmodium infection .
The malaria parasite has to go through series of complex developmental transitions within the mosquito vector before it can be transmitted to the human host . The major bottleneck for Plasmodium's development occurs during the ookinete invasion of the midgut epithelium , prior to the development of oocysts on the basal lamina [1] . The factors that are believed to contribute to parasite losses at this stage are digestive enzymes , the mosquito's immune defenses and the intestinal microbial flora [2]–[4] . Large communities of diverse microorganisms reside in insects with a major concentration in the intestinal sections [5] . While much research has been focused on the microbiota of the mammalian intestine and its role in defense against pathogenic microorganisms [6] , studies of insect gut microbiota have mainly concentrated on the contribution of microbial endosymbionts to the host's nutritional homeostasis [5] . However , the microbiota of the insect gut has also been shown to play a pivotal role of preventing development of pathogens . Studies have reported the wide spread of various species of Gram-negative bacteria in the midguts of both laboratory-reared and field derived mosquitoes , and some of this flora has been associated with an inhibitory activity on the sporogonic development of the Plasmodium parasites in the mosquito midgut [7]–[11] . However , these studies have not identified the causal mechanisms through which the presence of bacteria negatively impacts on malaria parasite development . Bacteria within the midgut lumen may directly interact with , and adversely affect , the different malaria parasite stages within the bloodmeal through the production of various enzymes and toxins or physical barriers that hinder the interaction between Plasmodium ookinetes and the midgut epithelium ( reviewed in [12] ) . Alternatively , the effect of bacteria on parasite development may occur indirectly through alterations in the physiology of the mosquito host itself , possibly through induction of immune responses that are cross-reactive between bacteria and malaria parasites , and/or changes of host metabolism that would affect the composition of mosquito derived molecules that are essential for Plasmodium development . Some studies have indicated that some of the mosquitoes' immune factors induced by bacterial challenge are involved in the killing of parasites at the pre-oocysts development stages [13]–[15] . Indeed , a great overlap , at the functional level , between antibacterial and anti-Plasmodium immune responses has been observed and suggests that mosquitoes lack highly specific mechanisms for defense against malaria parasites , but are using their anti-bacterial mechanisms to limit Plasmodium infection [14] , [16] , [17] . A reasonable hypothesis is that the presence of bacteria activates the mosquito's antimicrobial immune responses and the synthesized antimicrobial peptides and other immune factors will act against co-infecting Plasmodium parasites . Indeed , a complex interplay between the mammalian immune system and the intestinal microbiota is essential for protection from infectious pathogenic microorganisms [18] . Some intestinal microbial species induce innate immune effector molecules which can kill competing bacterial species , including pathogens ( reviewed in [19] ) . The composition of mosquito midgut microbiota is much less complicated than that of mammalian intestine microbiota which makes it as a good model for dissecting the dynamics between the host innate immune system , natural bacterial flora , and the pathogenic microorganisms . Besides , mosquitoes transmit a broad range of human parasitic and viral diseases , within which malaria is still one of today's most devastating infectious diseases . A better understanding of the roles of microbiota in the exploiting host immunity in defending against pathogens could potentially lead to the development of new malaria control strategies . We have examined the influence of the mosquito's midgut microbial flora and the derived antibacterial immune responses on malaria parasite infection through a series of infection assays in conjunction with functional genomics analyses .
To gain a better understanding of the potential fluctuations in microbial load and species composition between laboratory reared mosquitoes of different generations and within the same generation , we monitored the bacterial loads and species composition in individual five-day-old female A . gambiae mosquitoes of five consecutive generations . In accordance to previous studies our results showed a great variability in both parameters [8] , [9] , [20]–[22] . Interestingly , these variations were also observed between mosquitoes originating from the same generation and cage ( Figure 1 ) . This intriguing pattern may in some way relate to the equally broad distribution of Plasmodium infection intensities among mosquitoes that have fed on the same gametocyte culture . On average , individual mosquitoes carried around 40 , 000 colony forming units ( CFU ) . Similarly to previous studies , the majority of the isolated bacteria were Gram-negative suggesting that the midguts of mosquitoes have more optimal growth conditions for this type , especially those from the Enterobacteriaceae family . This strong bias is also likely to have been attributed , to some degree , to the LB agar– based aerobic culturing method that was used for these assays . Sequence analyses of the 16s ribosomal genes from morphologically distinct bacteria colonies identified the following five different species as dominant in all assayed generations: Enterobacter asburiae ( 98% ) , Microbacterium sp . ( 98% ) , Sphingomonas sp . E- ( s ) -e-D-4 ( 2 ) ( 100% ) , Serratia sp . ( 99% ) and Chryseobacterium meningosepticum ( 100% ) . The C . meningosepticum and Serratia sp . species were dominant within all five generations and the former was the most abundant , especially within the second generation . Other bacteria identified from different generations were: Asaia bogorensis ( 99% ) , Bacillus subtilis ( 99% ) , Enterobacter aerogenes ( 98% ) , Escherichia coli ( 91% ) , Herbaspirillum sp . ( 99% ) , Pantoea agglomerans ( 98% ) , Pseudomonas fluorescens ( 99% ) , Pseudomonas straminea ( 99% ) , Phytobacter diazotrophicus ( 97% ) and Serratia marcescens ( 99% ) . Interestingly , when C . meningosepticum became the dominant bacterium of the midgut flora , the growth of other bacterial species , that could be cultured on LB agar , was usually limited suggesting that this species may possess some competitive advantages in the gut environment . Our LB agar –based culture assays have some limitations in providing the complete picture of the composition of the mosquito midgut microbiota since a large fraction of bacteria are likely to be un-culturable , similarly to the human intestinal microbiota [23] . Future high throughput sequencing -based metagenomics approaches are likely to provide comprehensive information on the composition of the midgut microbiota . Nevertheless , as a proof of principle , our approach shows the great variations in both load and composition of the microbiota between different individuals and generations of insectary-reared mosquitoes . We assessed the impact of the mosquito's natural microbial flora on P . falciparum's capacity to establish infection through the removal of bacteria with antibiotic treatment , according to the established methodology [24] , [25] . Provision of antibiotic through the sugar meal effectively eliminated all detectable bacteria from mosquitoes fed on either sugar or human blood ( Figure 2A ) . The average bacterial load of sugar fed mosquito midguts was 104 CFU , and those fed on blood contained as many as 106 CFU ( Figure 2A ) . After antibiotic treatment mosquitoes became aseptic and are referred as aseptic mosquitoes , while untreated mosquitoes are referred as septic . Aseptic mosquitoes were significantly more susceptible to P . falciparum infection , as a measure of oocysts numbers on the midgut , compared to the septic mosquitoes ( p<0 . 01 ) ( Figure 2B ) . To gain a better understanding on the infection stage– specificity of this anti-Plasmodium action , we compared infection intensities between the septic and aseptic mosquitoes at two time points after ingestion of infected blood: at 28 hrs when ookinetes are still invading the midgut epithelium and at 10 days when all viable parasites have developed into oocysts on the basal side of the midgut epithelium . A significant larger number of ookinetes were found in the midgut epithelium of aseptic mosquitoes compared to the septic at 28 hrs after ingestion , suggesting that the bacteria-mediated anti-Plasmodium action has already taken place at pre-oocyst stages ( p<0 . 01 ) ( Figure 2B ) . Parasite losses during the transition from the ookinete to the oocyst stage were comparable between the septic and aseptic mosquito cohorts , suggesting that the presence of a microbial flora has little influence on parasite elimination at the early to late oocyst stages . A few aseptic mosquitoes displayed a very low infection level , while other had as many as 200 oocysts; this variation could be explained by potential differences in genetic background of individual mosquitoes . Future analyses will also address the impact of the microbiota on the later parasite stages in the mosquito . To test whether the observed differences in infection levels between septic and aseptic mosquitoes could have been attributed to a direct interaction between the antibiotic and the parasite or mosquito , we re-challenged antibiotic treated aseptic mosquitoes with bacteria that had been previously isolated from midguts of adult females , prior to infection with Plasmodium ( Figure 2B ) . The results from this assay suggested that the increased levels of oocyst infection in aseptic mosquitoes resulted from the absence , or at least a significantly decreased level , of bacteria , rather than a direct effect of the antibiotic itself on either the malaria parasites and/or the mosquito vector . The lower levels of oocysts in re-challenged mosquitoes compared to the untreated septic mosquitoes are likely to result from a higher bacterial load or the differences of the compositions of re-challenged bacteria to the natural flora . Interestingly , the presence of the microbial flora influenced the mosquito's longevity upon Plasmodium infection; approximately 60% of the infected septic mosquitoes died by day 7 post-infection ( fed with 1% P . falciparum gametocytes ) , in contrast to only 40% of the aseptic group despite an approximately 5-fold higher infection level ( Figure S1 and Figure 2B ) . The mortality of the septic and the aseptic mosquitoes after feeding on non-infected blood did not differ significantly suggesting that the increased mortality of septic Plasmodium infected mosquitoes was caused in some way by the co-occurrence of bacteria and malaria parasites ( data not shown ) . Interestingly , malaria-infected aseptic mosquitoes in which the midgut bacteria had been re-introduced exhibited reduced levels of mortality compared to untreated septic mosquitoes , possibly due to the presence of residual antibiotic in the tissues of these mosquitoes ( Figure S1 ) . This observation further supports the crucial impact of the microbiota on the mosquito's vector competence . In concordance with the previously described experiments , co-introduction of live or heat-inactivated bacteria with P . falciparum gametocytes in the midgut through feeding will result in a significantly decreased susceptibility to Plasmodium infection compared to the controls ( Figure 3A ) ; a 4-fold fewer oocysts developed in mosquitoes that had co-fed on live bacteria ( p<0 . 01 ) , and a 2 . 2-fold fewer oocysts developed in mosquitoes that had co-fed on heat-inactivated bacteria ( HIA ) ( p<0 . 05 ) compared to the control mosquitoes . These and the previously described results suggest that the bacteria in the midgut lumen exert an anti-Plasmodium effect that could either involve a mosquito response or a direct interaction with the parasite . The frequency distribution of oocysts demonstrated that co-feeding with either live or heat-inactivated bacteria and pre-injection of live bacteria ( discussed below ) resulted in an over dispersion of oocysts , with the majority of mosquitoes having very few oocysts ( Figure 3 ) . The decreased numbers of developing oocyst on the midguts of mosquitoes that had been exposed to bacteria suggested that the bacteria-mediated inhibitory activity on the parasite is acting prior to the oocyst stage . To test whether the negative effect of bacteria on malaria parasite development was to some degree attributed to a direct interaction by which the bacteria kill Plasmodium , we monitored P . falciparum development within the midgut lumen and epithelium of the four cohorts of mosquitoes ( septic , aseptic , aseptic mosquitoes re-challenged with natural flora bacteria , or septic mosquitoes co-fed with experimental bacteria ) . The prevalence of ookinetes in the blood-meal at 24 hrs after ingestion showed no significant difference between the four cohorts , suggesting that the bacteria had no effect on the pre-invasive stages . However , the number of ookinetes observed within the midgut epithelium was significantly higher in the aseptic mosquitoes , by approximately a 2 . 5-fold compared to the cohorts that contained bacteria ( Figure 4 , upper panel ) . The morphology of ookinetes was similar in the four cohorts ( Figure 4 , lower panel ) . These results suggest that the effect of bacterial exposure on mosquito susceptibility to P . falciparum occurs during ookinetes invasion , most likely through a mosquito response to the bacteria challenge which is likely to entail components of the mosquito innate immune system . Previous studies have indeed shown that the mosquito uses some of the same immune factors to combat bacteria and Plasmodium parasite infection [14] , [26] . Another possibility is that the bacteria form a physical barrier which blocks the parasite's access to the epithelium; this is a common mechanism by which the vertebrate microbiota protect against pathogenic bacterial infection ( reviewed in [19] ) . However , our current data does not directly support this hypothesis . To provide further clues on this anti-parasitic mechanism we looked at the effect of hemocoel injected live or heat inactivated bacteria on the P . falciparum development . Injection of live bacteria at 24 hrs prior to feeding on a gametocyte culture resulted in a significant reduction of oocysts ( p<0 . 05 ) while injection of heat inactivated bacteria had an insignificant effect on Plasmodium infection ( p>0 . 05 ) , compared to the PBS injected controls ( Figure 3B ) . This result further supports that the anti-Plasmodium activity of bacteria is indirect and involves a response by the mosquito vector since the injected bacteria are unlikely to directly interact with the parasites that are confined within the midgut epithelium or under the basal lamina . It is more likely that the systemic infection will induce a battery of defense molecules in the hemolymph , from where they can attack the midgut-stage parasites on the basal side of the gut , or even within the epithelium by diffusion through the basal labyrinth . Indeed our previously published studies showed that injected bacteria induced a battery of anti-Plasmodium immune factors [14] . The stronger anti-Plasmodium effect of either injected or co-fed live bacteria , compared to heat inactivated bacteria , suggest that a factor which is more specific for live bacteria may be responsible for the inhibitory effect . Alternatively , the stronger effect of live bacteria may simply reflect their proliferative capacity which resulted in multiplication of their numbers to induce a much stronger immune response from the mosquito host . Mosquitoes , as all other higher organisms , are continuously exposed to a variety of microbes . And we have shown that this exposure , whether it originates from the midgut lumen or the hemolymph , can influence the mosquito's permissiveness to P . falciparum infection . We have also shown that this effect is likely to be mediated through a mosquito response to the bacterial exposure . To better understand this response we have performed a series of genome-wide expression analyses to assess the regulation of the mosquito transcriptome upon microbial exposure . We used a microarray-based genome-wide gene expression strategy to compare transcript abundance between septic and aseptic adult female mosquitoes that had been fed on either sugar or non-infected blood ( Figure 5 and Tables S2 , S3 ) . The presence of the endogenous bacteria flora in sugar fed mosquitoes resulted in the differential regulation of some 185 genes; 121 genes were up-regulated and 64 genes were down-regulated compared to antibiotic treated aseptic mosquitoes . A similar number of 195 genes were regulated by the presence of the endogenous microbial flora after feeding on non-infected blood; 137 genes were up-regulated and 58 genes were down-regulated ( Figure 5A ) . The relatively small number of genes that were regulated as a consequence of the presence of the endogenous microbial flora most likely indicates a symbiotic relationship that has led to the adaptation of the mosquito to this flora . This hypothesis is strengthened by subsequent experiments that investigated the effect of ingested non-natural bacteria on the mosquito's transcriptome ( see below ) . The mosquitoes' responses to natural microbiota when fed with either sugar or non-infected blood were quite divergent with only limited overlap in gene expression ( Figure 5A ) , that comprised 21 induced and 1 repressed gene , corresponding to approximately 6 . 5% of the total regulated genes . However , one third of the commonly induced genes belonged to the immunity class . The regulated genes represented a variety of functional classes with a general strong bias and over-representation of innate immunity genes ( Figure 5B and 5C ) . Several of these immune genes have been previously shown to be transcriptionally-induced during malaria parasite infection , and to mediate anti-Plasmodium activity ( Tables S2 , S3 ) . The septic mosquitoes displayed elevated expression of genes code for the antimicrobial peptides Cecropins 1 ( Cec1 ) and 3 ( Cec3 ) , Defensin 1 ( Def1 ) and Gambicin; the signal transducing serine proteases SP5 , ClipA9 , ClipA7 and ClipB8 , and various pattern recognition receptors including AgMDL8 , CTLMA4 , FREP7 and FBN51 , Tep4 and Tep5 , Galectin 8 , and PGRP-LB , PGRP-LC2 and PGRP-S3 [14] , [27]–[31] . Surprisingly , the expression of the anti-Plasmodium factors FBNs 6 , 9 , and 36 were decreased in the septic mosquitoes ( Tables S2 , S3 ) . The immune responsive Lysozyme c-1 ( LYSC1 ) which previously has been linked to melanization reactions [32]–[34] , was up-regulated in septic sugar-fed mosquitoes; lysozymes are key antibacterial factors . These results suggest that the natural microbiota play an important role in stimulating a basal immune activity which in turn is likely to contribute towards the determination of the mosquito's susceptibility to various pathogens , and hence their vectorial capacity . In fact a recent study has established that Plasmodium development is significantly more influenced by the mosquito's basal level immunity rather than the induction of immune responses upon parasite infection [35] . Of particular interest was the elevated expression of the peritrophic matrix protein gene Ag-Aper1 in septic mosquitoes that had fed on either sugar or uninfected blood , and several other genes encoding proteins with peritrophin-like , laminin-EGF-like and chitin-binding like domains ( Tables S2 , S3 ) [36] . Ag-APer1 and proteins containing chitin-binding domains may function as structural components of the insect cuticle , the peritrophic matrix and/or as pattern recognition receptors . The elevated expression of Ag-Aper1 in septic mosquitoes may indicate a role of the peritrophic matrix in protecting the epithelium from the infection of midgut flora bacteria . The natural microbial flora also stimulated expression of several metabolic genes involved in glycolysis , gluconeogenesis and sugar transport and this may relate to digestion of midgut bacteria that function as a food source for the mosquitoes [37] ( Tables S2 , S3 ) . The genes exhibiting the greatest fold-differences in expression between septic and aseptic mosquitoes were of unknown function ( Figure 5C ) . To investigate the mosquito's global transcriptional response to exposure to non-natural midgut flora we compared transcript abundance between mosquitoes that had fed on blood supplemented with a mixture of both Gram-negative ( E . coli ) and Gram-positive ( Staphylococcus aureus ) bacteria and control mosquitoes that had fed on uninfected blood with PBS . These treatments resulted in a much broader response . The ingestion of these bacteria triggered the regulation of as many as 656 and 520 genes in the midgut and carcass , respectively ( Figure 5 and Tables S4 , S5 ) . In the midgut , 458 genes were up-regulated and 198 genes were down-regulated . As expected , fewer genes were regulated in the carcass compared to midgut tissue which was in direct contact with the ingested bacteria; 224 genes were up-regulated and 296 were repressed . Among the immune genes exhibiting differential expression between sterile-blood-fed and bacteria-blood-fed mosquitoes were several that have previously been shown to mediate anti-Plasmodium immune responses and to be transcriptionally up-regulated during Plasmodium parasite infection ( Tables S4 , S5 ) . The ingestion of bacteria stimulated an elevated expression of genes code for the antimicrobial peptide IRSP1 , the signal transducing serine proteases ClipB16 , and inhibitor SRPN6 and SRPN7 , and various pattern recognition receptors including AgMDLs 4 , 6 , and 7 , CTL , CTLGA1 , CTLGA3 , and CTLMA6 , FBNs 9 , 20 , 21 , and 51 , LRRD8 , PGRP-LB , PGRP-LC2 and PGRP-S3 , Tep11 and Toll6 [14] , [38]–[42] . Only four immune genes , SP5 , TPX4 , DCCE2 , and FBN51 were induced by both the natural flora and the ingested non-natural bacteria , while Tep-like , PGRP-LD , and FBN9 displayed an opposite pattern of regulation ( Figure 5A and Tables S4 , S5 ) . As mentioned above , these differences are likely to reflect an adaptation of the mosquito to its natural microbial flora . Potential differences in the dosage of bacterial exposure may however also have influenced the quite different outcome . Depletion of several immune factors through RNAi-mediated gene silencing has been shown to result in a proliferation of bacteria in the hemolymph as a result of a compromised immune system [43] , [44] . To test whether the immune genes that are induced by the natural flora are indeed implicated in defending against opportunistic bacterial infections , we assayed the proliferation of the mosquito midgut flora upon their silencing . We subjected 12 genes to this test of which Cec1 , Cec3 , Def1 , ClipA9 , Gambicin , PGRP-LB , and FBN9 were induced by the presence of the natural bacterial flora , and the remaining LRRD7 , LRRD19 , TEP1 , Rel1 and Rel2 genes represented anti-Plasmodium pattern recognition receptors or immune signaling pathway factors [35] , [45]–[47] . Depletion of Cec3 , Gambicin , PGRP-LB , LRRD7 , TEP1 , and Rel2 resulted in the significant proliferation of the natural bacterial flora in the mosquitoes' midguts . Gene silencing of Cec1 , Def1 , ClipA9 , FBN9 , and LRRD19 also resulted in some increase of bacterial loads in the midgut; however these effects were not statistically significant ( Figure 6 ) . The lack of significant bacterial proliferation in these knock-down mosquitoes could also be explained by the lower efficacy of gene silencing in the midgut tissue compared to the abdominal and thoracic compartment ( Figure S3 , Table S1 ) . These results show that the mosquito's innate immune system is actively involved in controlling the bacterial load in the midgut lumen in a constitutive fashion , and that exposure to increased bacteria will result in increased production of some of these anti-Plasmodium factors . We believe that this is the mechanistic basis of how the mosquito's endogenous flora is important in priming an anti-Plasmodium defense . The dual role of anti-Plasmodium factors in defending against both the parasite and bacteria , and the influence of bacteria on Plasmodium development , suggests the existence of complex interactions and relationships between the parasite , the microbiota and the mosquito's innate immune system . For example , the anti-Plasmodium activities of certain genes might be modulated by their parallel activities against bacteria . To assess such complexities and interactions , we studied the effect of various immune genes on P . falciparum's capacity to establish infection in the midgut tissue of both septic and aseptic mosquitoes through RNAi gene silencing approach ( Figure 7 ) . RNAi-mediated depletion of the antimicrobial peptides Cec1 , Def1 , and Gambicin had no statistically significant effect on the levels of P . falciparum oocyst infection in either mosquito groups ( data not shown ) , while gene silencing of Cec3 and PGRP-LB resulted in an increased susceptibility to P . falciparum only in the aseptic mosquitoes ( Figure 7A ) . This result may suggest that the depletion of these two immune genes in septic mosquitoes resulted in a proliferation of the microbial flora which in turn may have counteracted , or masked , the potential decrease of anti-Plasmodium immune responses . Another striking example of how important the microbial flora is in regulating anti-Plasmodium activity of immune genes is represented by the serine protease ClipA9 . When this factor was depleted in septic conditions , the mosquitoes became significantly less susceptible to P . falciparum infection ( p<0 . 05 ) . In contrast , when ClipA9 was silenced in aseptic mosquitoes it had no significant effect on susceptibility to the parasite ( Figure 7B ) . This observation suggests that the ClipA9-mediated anti-Plasmodium defense is exerted through the microbial flora and not directly against the parasite . ClipA9 is likely to be more specific for antibacterial defense and its depletion , under septic conditions , will hence result in the proliferation of bacteria which will exert strong anti-Plasmodium activity . Alternatively , it may mediate some direct Plasmodium protective activity which is abolished in the absence of bacteria . Interestingly the malaria parasite infection phenotype of ClipA9 gene silencing is opposite to that observed for the serine protease inhibitor SRPN6 , suggesting that SRPN6 may function in the same cascade as an inhibitor of ClipA9 [30] , [39] . In conclusion , similarly to humans , the mosquito intestine harbors a natural microbiota which is necessary for maintaining normal physiological functions including host metabolism and immune homeostasis . Accordingly , we have shown that the mosquito's natural bacterial flora show great variability between mosquitoes originating from the same colony and that it is an important regulator of mosquito permissiveness to Plasmodium . The mosquito's natural microbiota and artificially introduced non-natural bacteria negatively affected malaria parasite development through a mechanism that appears to implicate in the innate immune system , and not a direct killing of Plasmodia by the bacteria . The natural bacterial flora is essential in inducing a basal level immunity that in turn enhances the mosquito's ability in defending against the infection from the malaria parasites [35] . Interestingly , the effect of certain immune genes on Plasmodium infection is dependent on the presence of the microbial flora , suggesting that their mode of action is complex . This finding suggests that future studies on gene specific anti-Plasmodium action should also consider the complex interplay between the microbiota and the mosquito's immune defenses against the Plasmodium parasite . This relationship is further corroborated by observations from Dr . Barillas-Mury's group , where RNAi gene silencing of one immune gene facilitated the proliferation of microbial flora but reduced the Plasmodium infection . The natural bacterial flora has also been shown to be involved in the suppression of other pathogenic organisms in other mosquito species . Tetracycline treatment of Culex bitaeniorhynchus rendered this mosquito more susceptible to the Japanese encephalitis virus [48] and the Aedes aegypti mosquito microbial flora has been shown to stimulate a basal-level immunity which suppresses dengue virus infection [25] .
All animals were handled in strict accordance with good animal practice as defined by the relevant national and/or local animal welfare bodies , and all animal work was approved by the appropriate committee . A . gambiae Keele strain mosquitoes were maintained on a 10% sugar solution in laboratory culture at 27°C and 70% humidity with a 12 hrs light/dark cycle according to standard rearing procedures [49] . A single cohort of adult female mosquitoes were collected immediately after eclosion , and either maintained under normal , non-sterile insectary conditions or placed into a sterile environment . Following , adult female mosquitoes were daily given fresh filtered sterilized 10% sucrose solution containing 15 µg gentamicin sulphate ( Sigma ) and 10 units/10 µg of penicillin-streptomycin ( Invitrogen ) per ml , respectively . Each cohort of mosquitoes was simultaneously membrane-fed freshly washed human erythrocytes resuspended to 40% haematocrit using human serum . As far as possible , every care was taken to maintain the sterility of the blood and membrane-feeding apparatus used to feed the mosquitoes , in order to prevent the antibiotic-treated mosquitoes acquiring bacterial infection during the process of membrane-feeding . The mosquitoes were starved for 8 hrs before feeding to encourage engorgement , and sugar solution was replaced once blood feeding had finished . At 24 hrs after blood feeding , 20 mosquitoes from each replicate of each cohort was collected and dissected on ice . RNA was extracted from dissected tissues at the assayed time points using the RNeasy kit ( Qiagen ) . The quantification of RNA concentrations was performed using a Spectrophotometer ( Eppendorf ) . Probe sequence design and microarray construction were kept the same as described in [14] . Probe preparation and microarray hybridizations were performed essentially as previously described with some modifications [14] . Briefly , Cy3-labeled control cRNA probes and Cy5-labeled treatment cRNA probes were synthesized from 2–3 µg of RNA using the Agilent Technologies low-input linear amplification RNA labeling kit according to the manufacturer's instructions . Hybridizations were performed with the Agilent Technologies in situ hybridization kit according to the manufacturer's instructions with 2 µg of cRNA probes and 16 hrs after hybridization the microarray slides were washed and dried with compressed air . Microarrays were scanned with an Axon GenePix 4200AL scanner using a 10 µm pixel size ( Axon Instruments , Union City , California , United States ) . Laser power was set to 60% , and the photomultiplier tube ( PMT ) voltage was adjusted to maximize effective dynamic range and minimize the saturation of pixels . Scanned images were analyzed by using GenePix software , and Cy5 and Cy3 signal and ratio values were obtained and subjected to statistical analysis with TIGR MIDAS and MeV software [50] . The minimum signal intensity was set to 100 fluorescent units , and the signal to background ratio cutoff was set to 2 . 0 for both Cy5 and Cy3 channels . Three or four biological replicates were performed for each experimental set . The background-subtracted median fluorescent values for good spots ( no bad , missing , absent , or not-found flags ) were normalized according to a LOWESS normalization method , and Cy5/Cy3 ratios from replicate assays were subjected to t tests at a significance level of p<0 . 05 using cutoff value for the significance of gene regulation of 0 . 7 and 0 . 8 in log2 scale , for septic mosquitoes and mosquitoes co-fed with experimental bacteria respectively , according to previously established methodology [51] . Microarray-assayed gene expression of 6 genes was further validated with quantitative RT-PCR and showed a high degree of correlation with the Pearson correlation coefficient ( p = 0 . 84 ) , the best-fit linear-regression analysis ( R2 = 0 . 70 ) , and the slope of the regression line ( m = 1 . 247 ) demonstrated a high degree of correlation of the magnitude of regulation between the two assays ( Figure S2 ) . Primers' sequences for validation of microarray hybridization data were as described in [14] . And new primers for RNAi gene silencing and verification were designed with Primer 3 Program on a web-based server ( http://frodo . wi . mit . edu/ ) . All the primer sequences were listed in Table S1 . Real-time quantitative PCR ( qRT–PCR ) to check the efficiency of gene silencing were done essentially according to [14] . The quantification was performed using the QuantiTect SYBR Green PCR Kit ( Qiagen ) and ABI Detection System ABI Prism 7300 . All PCR reactions were performed in triplicate . Specificity of the PCR reactions was assessed by analysis of melting curves for each data point . The ribosomal protein S7 gene was served as internal control for normalization of cDNA templates . Sense and antisense RNAs were synthesized from PCR-amplified gene fragments using the T7 Megascript kit ( Ambion ) . The sequences of the primers are listed in Table S1 . dsRNA mediated gene silencing was done according to [14] , [28] . About 80 4-d-old female mosquitoes were injected , in parallel , with GFP dsRNA as a control group or with target gene–specific dsRNA for the experimental group . Gene silencing in the whole mosquitoes was verified 3 to 4 d after dsRNA injection by qRT-PCR , done in triplicate , with the A . gambiae ribosomal S7 gene as the internal control for normalization . Gene silencing efficiency were listed in Table S1 with standard errors shown ( KD%±SE ) . RNAi gene silencing in the midguts was verified by RT-PCR , 10 midguts were used for each replicate and at least two replicates were included with only one replicate shown ( Figure S3 ) . At least 50 control ( GFP dsRNA–injected ) and 50 experimental ( gene dsRNA–injected ) mosquitoes were fed on the same P . falciparum NF54 gametocytes culture at 3–4 d after the dsRNA injection . 24 hrs post blood feeding ( pbf ) , the unfed mosquitoes were removed and the fed-mosquitoes were dissected at 7–8 d after feeding and midguts were stained with 0 . 2% mercurochrome [43] . Oocyst numbers per midgut were determined using a light-contrast microscope ( Olympus ) . The median number of oocysts per midgut was calculated for each tested gene and for GFP dsRNA–injected control mosquitoes . The results for equal numbers of midguts from all three independent biological replicates were pooled . The dot plots of the oocysts number in each midgut within each treatment were presented by MedCalc software with the median value of the oocysts indicated . The Kruskal-Wallis ( KW ) test and Mann-Whitney test were used to determine the significance of oocysts numbers ( p<0 . 05 ) . About 80 4-day-old mosquitoes were first injected with PBS as control , or a mixture of live bacteria with approximately 30 , 000 E . coli and 60 , 000 S . aureus , or a mixture of heat-inactivated bacteria with the same number as the live ones . 24 hrs or 48 hrs after injection , mosquitoes were fed with P . falciparum NF54 gametocytes culture which were carried out according to our establish protocols [14] . For the co-feeding assay , the same sets of control PBS or bacteria were mixed in the blood meal to result in the same amount of either bacterium in the mosquito midguts . Unfed mosquitoes were removed , and the rest were kept in 26°C for 8 days before the oocysts counts . The infection phenotypes were determined as described above . Isolation and colony forming units ( CFU ) enumeration of bacteria from midguts of untreated control , antibiotic-treated mosquitoes and gene-silenced mosquitoes were done essentially according to [43] with modifications . The midguts from surface sterilized mosquitoes were dissected with sterilized PBS 4 d after dsRNA injection , and CFU were determined by plating the homogenate of the midguts with series dilutions on LB agar plates and incubating the plates at 27°C for 2 days . Each assay was performed with one midgut and at least 10 independent replicates were included for each gene . The species of the isolated bacteria were determined by amplifying a region of the 16s rDNA as described by using primers 27f and 1492r [52] . PCR products were sequenced and blasted against Nucleotide collection ( nr/nt ) database to verify the species . The early stages of P . falciparum development within untreated , antibiotic-treated and bacteria co-feeding mosquito midguts were compared by using the immuno-staining of ookinetes with anti-Pfs25 antibody ( MRA-28 , provided by MR4 ) . Preparation of samples for immuno-fluorescence microscopy of malaria parasite within the bloodmeal was carried out based on [53] with substantial modifications . Sterile 0 . 5 ml “non-stick” low retention hydrophobic tubes ( Alpha Laboratory Supplies ) and sterile “non-stick” low retention hydrophobic pipette tips ( Alpha Laboratory Supplies ) were used to minimize malaria parasite loss during sample preparation due to their adhesion to plastic surfaces . The midguts including the entire bloodmeal contents were individually homogenized and diluted in 280 µl of PBS . 10 µl was then spotted , in duplicate , onto Teflon®-printed microwell glass slides ( VWR International ) previously coated with 3-aminopropyltriethoxysilane ( APES ) according to the supplier's instructions ( Sigma ) . The sample slides were then air-dried , fixed in ice cold acetone for 2 mins and subjected to blocking in 10% goat serum for 1 hr , followed by the incubation with primary antibodies at 1∶400 dilutions for 2 hrs . After three PBS washes , sample slides were incubated with secondary antibodies ( Molecular Probes , 1∶1000 ) for 2 hrs with Alexa Fluor 488-conjugated ( green ) goat anti-mouse antibody ( 1∶500 dilution ) . After another three PBS washes , sample slides were analyzed under a Nikon E800 upright microscope with epi-fluorescence . The total number of round forms , retort-forms and mature ookinetes in each spotted sample was counted . Average values for the densities of each malaria parasite stage present within each midgut examined were calculated from the three replicates . For checking the ookinetes and early oocysts in the midgut epithelium cells , at 24–30 hrs or 8 d after blood feeding , the midguts were dissected in 1% paraformaldehyde and washed with 3 times of PBS to remove the blood content and were subjected to the fixation in 4% paraformaldehyde ( in PBS ) for 1 hr and followed with 2 PBS washes . The midguts were then subjected to blocking and immune staining with primary antibody and secondary antibody as mentioned above . Midguts stained with pre-immune of anti-Pfs25 antibody were used as control . Midgut samples were mounted using the ProLong Antifade Kit ( Molecular Probes ) with DAPI staining of the cell nuclei and analyzed with same microscopy set as described above .
|
The Anopheles gambiae mosquito that transmits the malaria-causing parasite Plasmodium has an intestinal bacterial flora , or microbiota , which comprises a variety of species . Elimination of this microbiota with antibiotic treatment will render the Anopheles mosquito more susceptible to Plasmodium infection . In this study we show that these bacteria can inhibit the infection of the mosquito with the human malaria parasite Plasmodium falciparum through a mechanism that involves the mosquito's immune system . Our study suggests that the microbial flora of mosquitoes is stimulating a basal immune activity , which comprises several factors with known anti-Plasmodium activity . The same immune factors that are needed to control the mosquito's microbiota are also defending against the malaria parasite Plasmodium . This complex interplay among the mosquito's microbiota , the innate immune system , and the Plasmodium parasite may have significant implications for the transmission of malaria in the field where the bacterial exposure of mosquitoes may differ greatly between ecological niches .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases",
"microbiology/immunity",
"to",
"infections",
"genetics",
"and",
"genomics/functional",
"genomics",
"computational",
"biology/transcriptional",
"regulation",
"immunology/immune",
"response",
"microbiology/innate",
"immunity",
"immunology/innate",
"immunity",
"microbiology/parasitology"
] |
2009
|
Implication of the Mosquito Midgut Microbiota in the Defense against Malaria Parasites
|
In the case of an influenza pandemic , the current global influenza vaccine production capacity will be unable to meet the demand for billions of vaccine doses . The ongoing threat of an H5N1 pandemic therefore urges the development of highly immunogenic , dose-sparing vaccine formulations . In unprimed individuals , inactivated whole virus ( WIV ) vaccines are more immunogenic and induce protective antibody responses at a lower antigen dose than other formulations like split virus ( SV ) or subunit ( SU ) vaccines . The reason for this discrepancy in immunogenicity is a long-standing enigma . Here , we show that stimulation of Toll-like receptors ( TLRs ) of the innate immune system , in particular stimulation of TLR7 , by H5N1 WIV vaccine is the prime determinant of the greater magnitude and Th1 polarization of the WIV-induced immune response , as compared to SV- or SU-induced responses . This TLR dependency largely explains the relative loss of immunogenicity in SV and SU vaccines . The natural pathogen-associated molecular pattern ( PAMP ) recognized by TLR7 is viral genomic ssRNA . Processing of whole virus particles into SV or SU vaccines destroys the integrity of the viral particle and leaves the viral RNA prone to degradation or involves its active removal . Our results show for a classic vaccine that the acquired immune response evoked by vaccination can be enhanced and steered by the innate immune system , which is triggered by interaction of an intrinsic vaccine component with a pattern recognition receptor ( PRR ) . The insights presented here may be used to further improve the immune-stimulatory and dose-sparing properties of classic influenza vaccine formulations such as WIV , and will facilitate the development of new , even more powerful vaccines to face the next influenza pandemic .
The first cases of human infection with highly pathogenic avian influenza ( HPAI ) H5N1 virus occurred in 1997 during an outbreak in Hong Kong [1] . Since then HPAI H5N1 has spread across Asia , Europe , Africa and the Pacific , and has caused a cumulative number of 338 laboratory confirmed human cases of infection , with a fatality rate of >60% [2] . Although no sustained human to human transmission has been observed yet , the threat of an imminent H5N1 pandemic requires maximum preparedness [3] . Vaccination is considered the cornerstone of protection against epidemic and pandemic influenza . However , an anticipated scarcity of the antigenic vaccine components and a narrowed time window between vaccine production and deployment puts special constraints on the vaccine formulation to be used in a pandemic situation [4] , [5] . Consequently , pandemic vaccine formulations should ideally be dose sparing and uncomplicated to produce [6] , [7] . Whole inactivated virus ( WIV ) vaccines consisting of formalin-inactivated whole virus particles were the first registered influenza vaccines licensed in 1945 in the United States [8] . However , the use of this vaccine formulation caused a relatively high incidence of adverse events , including local reactions at the site of injection and febrile illness , particularly among children [9] , [10] . In the 1960 and 1970s , WIV vaccines were therefore largely replaced by less reactogenic split virus ( SV ) and subunit ( SU ) formulations [8] . SV and SU vaccines contain detergent- and/or ether-disrupted ( split ) virus particles or purified viral haemagglutinin ( HA ) and neuraminidase ( NA ) proteins , respectively . Apparently , disruption of whole inactivated influenza virus particles diminishes the reactogenicity of the vaccines . In primed individuals , unadjuvanted WIV , SV , and SU vaccines in general induce similar immune responses in terms of haemagglutination inhibition ( HI ) titres ( for a meta-analysis over 24 studies see [11] ) . However , in individuals that have not been exposed to the vaccine antigens before , WIV vaccines are more immunogenic than SV and SU vaccines [9] , [11] , [12] . Similarly , in naïve animals immunization with WIV raises stronger immune responses than immunization with SV or SU [13]–[15] , especially after a single administration . In the case of an H5N1 pandemic , the majority of the population is expected to be immunologically naïve to the H5N1 subtype . In this scenario , use of WIV as basis for an optimized vaccine may be of advantage , for its immunogenic superiority seems to rely on the ability to activate unique mechanisms in the priming event of the immune response . Thus , WIV seems to harbour an intrinsic immune-potentiating component that is lost during processing of inactivated virus particles to SV and SU vaccine formulations . In earlier experiments , we and others observed that immunization of mice with WIV vaccine results in a Th1-skewed immune response and strong antibody induction with high levels of IgG2a antibodies [14]–[16] . This response type was found irrespective of the murine genetic background or subtype of virus ( either H1N1 or H3N2 ) and conferred protective immunity against challenge with homologous virus [15] , [16] . By contrast , immunization with SU vaccine yielded responses of a Th2 phenotype with lower antibody levels mainly consisting of the IgG1 subtype , which did not lead to protection . “Empty” reconstituted viral envelopes ( virosomes ) resembling intact virus particles but devoid of the viral nucleocapsid elicited responses similar to those after vaccination with SU formulations [15] . This identifies the viral nucleocapsid which contains the viral genomic ssRNA as the immune-potentiating component of WIV . In the past decade , it has become increasingly clear that the acquired immune response to microbial infection is regulated through recognition of pathogen-associated molecular patterns ( PAMPs ) by Toll-like receptors ( TLRs ) and other pattern recognition receptors of the innate immune system [17]–[20] . However , the importance of TLR signalling in immune responses to vaccines remains largely unclear . A recent study showed that TLR signalling is not important for the antibody-enhancing effect of classical vaccine adjuvants such as Complete Freund's adjuvant ( CFA ) [21] . Since CFA contains dried mycobacteria , and therefore mycobacterial PAMPs [17] , this observation casts doubt on the importance of PAMPs and TLRs in augmenting immune responses to vaccination . Influenza viral genomic ssRNA is a natural PAMP recognized by TLR7 [22] . Here , we investigate whether PAMP recognition by TLRs , in particular recognition of viral ssRNA by TLR7 , is responsible for the superior response to WIV vaccines compared to SV and SU influenza vaccine formulations .
To analyze the role of ssRNA and other PAMPs in the response to influenza vaccines in detail , we immunized wild-type C57BL/6 mice , TLR7 knock-out mice , and MyD88/TRIF double knock-out mice with different vaccine formulations . MyD88 ( myeloid differentiation factor 88 ) is an adaptor molecule which functions downstream of all known TLRs and IL1R family members with the exception of TLR3 , which instead recruits a MyD88-related adapter molecule , TRIF ( TIR domain-containing adaptor protein inducing interferon β ) [17] . Consequently , a deficiency of both MyD88 and TRIF excludes signalling by all TLRs . Mice were immunized intramuscularly with β-propiolactone-inactivated H5N1 ( NIBRG-14 ) WIV , SV , or SU vaccine . Quantitative PCR using primers specific for segment 7 of the viral genome revealed that WIV contained per vaccine dose at least 5×108 copies of viral RNA , the natural ligand of TLR7 . In SU or SV vaccine the amount of RNA was 500 and 5 , 000 times lower than in WIV , respectively . Four weeks after immunization , serum and spleen cells were collected for evaluation of humoral and cellular immune responses . Serum HI titres in WIV-immunized TLR7−/− mice and MyD88−/−/TRIF−/− mice were found to be significantly lower than in WIV-immunized wild-type mice ( Figure 1A; p = 0 . 021 and p = 0 . 001 , respectively ) . Although sera from TLR7−/− mice immunized with WIV showed a higher geometric mean titre ( GMT ) than sera from WIV-immunized MyD88−/−/TRIF−/− mice , this difference was not significant ( p = 0 . 053 ) . Most of the HI titres of SV- and SU-immunized wild-type mice were below detection level , precluding evaluation of the effect of the knock-out mutations on the HI responses to these vaccines . Similar to the HI titres , virus neutralization ( VN ) titres of pooled serum samples from mice immunized with WIV were lower in the knock-out groups than in the wild-type group ( Table 1 ) . These results clearly show that TLR signalling is critically involved in the response to WIV immunization . Yet , in the knock-out groups , VN titres obtained after immunization with WIV were still modestly higher than those obtained after vaccination of wild-type mice with the other vaccines . This points to TLR-independent pathways contributing to the superior antibody response to WIV vaccine . Serum titres of H5N1-specific IgG were determined by ELISA . In accordance with the HI and VN results , IgG titres were significantly decreased in WIV-immunized TLR7−/− and MyD88−/−/TRIF−/− mice compared to wild-type mice ( Figure 1B; p = 0 . 010 and p = 0 . 001 , respectively ) . However , like the VN titres , the IgG titres in the WIV-immunized mutant mice were still significantly higher than those induced by SV ( TLR7−/−: p = 0 . 001; MyD88−/−/TRIF−/−: p = 0 . 005 ) or SU ( TLR7−/−: p = 0 . 005; MyD88−/−/TRIF−/−: p = 0 . 021 ) immunization again indicating involvement of TLR-independent pathways . The relative contributions of TLR-dependent and -independent mechanisms to the superior IgG response to WIV can be estimated by comparing the difference in geometric mean titre ( GMT ) between WIV-immunized wild-type and MyD88/TRIF-deficient mice with the difference between WIV-immunized wild-type mice and SV- or SU-immunized wild-type mice . Using this procedure the TLR-dependent contribution was calculated to be 73% and 83% for WIV versus SV and WIV vs SU , respectively ( for calculation , see Text S1 ) . The IgG responses to SV and SU vaccine in both TLR7−/− or MyD88−/−/TRIF−/− mice did not differ from those in wild-type mice , except for the IgG response to SU in TLR7−/− mice , which was slightly but significantly decreased ( p = 0 . 038; Figure 1B ) . Together with the HI and VN results , these findings demonstrate that the superior antibody response to WIV is predominantly regulated by TLRs , TLR7 in particular , while TLRs do not seem to play a prominent role in SV and SU antibody responses . We next investigated the role of TLRs in the Th1 polarization of the response characteristically found after WIV vaccination . We first assessed numbers of IFNγ- and IL4- producing T cells ( Th1 and Th2 cells , respectively ) in a cytokine-specific Elispot assay , after re-stimulation of spleen cells from immunized mice with H5N1 SU vaccine . Numbers of Th1 cells were significantly decreased in WIV-immunized knock-out mice compared to wild-type mice ( p = 0 . 003 and p = 0 . 010 for TLR7−/− and MyD88−/−/TRIF−/− mice , respectively ) , and matched those found in SV- and SU-immunized wild-type mice ( Figure 2 ) . No difference was found between TLR7−/− and MyD88−/−/TRIF−/− mice . Numbers of influenza-specific IL4-producing cells were extremely low in all animals for all vaccine formulations without significant differences between knock-out and wild-type mice ( not shown ) . These data indicate that stimulation of TLR7 by ssRNA is the predominant determinant of the strong Th1-type cellular response induced by WIV . We further determined the subtype profiles of H5N1-specific serum IgG by ELISA ( Figure 3 ) . IFNγ is known to stimulate production of IgG2a subtype antibodies by activated B cells , while IL4 stimulates IgG1 secretion [23] . In C57BL/6 mice , however , the IgG2c subtype is produced instead of IgG2a [24] , [25] . Hence , a predominance of IgG2c or IgG1 is indicative of a Th1- or Th2-type response , respectively . WIV immunization of TLR7−/− mice as well as MyD88−/−/TRIF−/− mice resulted in significantly reduced IgG2c levels as compared to wild-type mice ( Figure 3; p = 0 . 001 for both types of knock-out mice ) , supporting a role for TLR7 in Th1 polarization . IgG1 was increased in WIV-immunized TLR7−/− mice ( P = 0 . 050 ) , adding to the preponderance towards a Th2-type response to WIV in these mice . The average of ratios of serum IgG2c and IgG1 concentrations ( determined with appropriate IgG subtype protein standards ) was 17 . 82 ( SD 8 . 44 ) for the wild-type mice immunized with WIV , compared to 0 . 53 ( SD 0 . 41 ) for TLR7−/− mice immunized with WIV . SV and SU vaccines induced predominantly IgG1 and low levels of IgG2c , consistent with a Th2-type response ( Figure 3 ) . For reasons unknown , SU vaccine induced lower IgG1 titres in both types of knock-out mice compared to the wild-type mice ( TLR7−/−: p = 0 . 050; MyD88−/−/TRIF−/−: p = 0 . 014 ) . Whether the presence of some residual RNA in SU vaccine might play a role remains to be shown . The response characteristics of the different H5N1 vaccines in wild-type mice were well in line with those previously found for other influenza subtypes [15] , [16] . This consistency is supportive of a general mechanism underlying the differences in responses to WIV , SV and SU vaccine , which operates irrespective of the virus subtype used to vaccinate . The above results demonstrate that TLR signalling plays an important role in the magnitude and Th1 skewing of the response to WIV influenza vaccines . Yet , in TLR-ko mice , WIV remained more immunogenic than SV and SU vaccines , inducing significantly higher titres of total IgG ( Figure 1B ) and Th1-type antibody subtypes ( IgG2b , IgG2c , IgG3; Figure 3; p<0 . 05 for all comparisons ) . Thus , next to TLR-dependent mechanisms , a ( minor ) TLR-independent factor seems to contribute to the superior magnitude and Th1-skewing of the immune response to WIV . Type I interferons , including IFNα , have been shown to stimulate antibody responses and isotype switching to IgG2a when added to influenza subunit vaccine or other protein antigens [26] , [27] , even without the need for additional TLR stimuli . We have previously shown for an H3N2 influenza virus strain that , unlike SU vaccine , WIV vaccine efficiently induced interferon α ( IFNα ) production in plasmacytoid dendritic cells ( pDCs ) in vitro [15] . We therefore evaluated the induction of IFNα by the H5N1 influenza vaccine formulations used in this study and its TLR7 dependency in vitro . In pDCs of wild-type mice cultured from bone marrow cells ( Figure 4A , black bars ) or enriched from splenocytes ( Figure 4B , black bars ) WIV but not SV or SU induced IFNα production . In bone marrow-derived pDCs from TLR7−/− mice , IFNα production upon incubation with WIV was strongly decreased as compared to wild-type DCs ( Figure 4A ) , confirming the results of others [22] . However , spleen-derived pDCs from TLR7−/− mice exposed to WIV produced similar amounts of IFNα as compared to pDCs from wt mice ( Figure 4B ) . Thus , while in pDCs cultured from bone marrow induction of IFNα production by WIV is strictly dependent on TLR7 , in pDCs enriched directly from spleen cells it is independent of TLR7 . This implies that bone marrow pDCs and spleen pDCs are not completely identical . In line with this notion , bone marrow pDCs and spleen pDCs were earlier found to respond differently to HSV virus infection with respect to the TLR9 dependency of the IFNα response [28] . Our results show that WIV is indeed able to induce IFNα in a TLR7-independent way . This may also be the case in the in vivo situation , where in accordance with its well-described adjuvant functions IFNα may lead to the production of Th1 type antibodies in TLR-deficient mice [26] . Possible TLR-independent pathways activated by WIV may involve the retinoic acid-inducible gene ( RIG-I ) [29]–[32] . RIG-I is a cytoplasmic RNA-helicase that recognizes influenza virus by binding viral ssRNA bearing 5′-triphosphates which leads to IFNα production [33] , [34] . The inactivated virus particles in WIV vaccine retained their membrane-fusion property ( Text S2 ) and part of the viral genomes could therefore have entered the target cell cytoplasm to be sensed by RIG-I . Taken together our observations show that the superior immune response to WIV , relative to that to SV or SU vaccines , is driven primarily by TLR-dependent mechanisms . Herein the presence of the viral RNA in the vaccine seems to play a crucial role . In contrast to SV and SU vaccines WIV contains substantial amounts of viral RNA . Removal of ssRNA from WIV by detergent solubilization and ultracentrifugation followed by reconstitution of the viral membrane envelopes to virosomes abolishes the capacity of the vaccine to induce production of IFNα by pDCs in vitro ( Text S3 and Figure S1A ) and type 1 immune responses in vivo [15] . On the other hand , ssRNA purified from WIV and condensed with polyethylenimine ( PEI ) did induce IFNα production in vitro ( Text S3 and Figure S1B ) . Obviously , exposure of the viral RNA to β-propiolactone in the course of virus inactivation leaves the RNA intact to trigger TLR7-mediated signaling pathways ( Figure 4 ) , which translates into a strong and Th1-skewed antibody response to WIV in wild-type mice . In addition , the viral RNA may contribute to the TLR-independent part of the response to WIV since TLR7-independent production of IFNα could only be induced in pDCs by WIV and not by formulations ( SV , SU , or reconstituted viral envelopes ) which lack viral RNA ( Figure 4B ) [15] . These lines of evidence point to the ssRNA in WIV as the key component that enhances and steers the adaptive immune response by involvement of innate immune mechanisms . IFNα induction in pDCs clearly discriminates WIV from SV and SU vaccines but seems to occur independent of TLR7 . The fact that the immune response to WIV is predominantly dependent on TLR7 then suggests that other TLR7-mediated mechanisms , possibly involving conventional DCs and B cells , critically contribute to the immune reaction . Recently , an in vitro study on B cells showed that TLR7 stimulation or CD40-CD40L binding by itself triggers IgG1 antibody production , but when simultaneously present induce proliferation and a switch to IgG2a production [25] . Additional stimulation of IFNα/β receptors on the same cells further drives the production of IgG2a at the expense of IgG1 antibodies [25] . Although this model might represent an over-simplification of the in vivo situation , it is in line with our data . The different scenarios encountered upon immunization of wild-type and mutant mice with WIV , SV , or SU are summarized in Table 2 . WIV provides the ssRNA for direct triggering of TLR7 in B cells as well as the CD40 ligand for CD40 stimulation on B cells through strong T helper cell induction , which was shown also to depend on TLR7 signalling . Together with IFNα produced by TLR7-mediated and/or TLR7-independent mechanisms , these signals will lead to the enhanced and strongly polarized Th1-type antibody responses characteristic for WIV . In the absence of TLR7 , WIV-induced IFNα can still stimulate moderate production of Th1 type antibodies and increase the total IgG . In contrast , SV and SU vaccines are poor inducers of T helper cells and IFNα , and cannot stimulate B cells directly via TLR7 . Consequently , SV and SU vaccines induce lower and more Th2-polarized antibody responses . Our data provide mechanisms which explain the superiority of WIV vaccine to prime HA-specific immune responses in mice . Whether similar mechanisms are operational in humans and contribute to the stronger immunogenicity of WIV compared to SV or SU in unprimed individuals remains to be elucidated . Despite the favourable immunogenic properties of WIV , recent clinical trials performed in the context of pandemic vaccine development show that even with WIV at least two immunizations with a substantial amount of antigen ( 15–30 µg ) and/or the addition of adjuvants will probably be required to achieve immune responses that comply with the CPMP criteria . If TLRs are involved in the priming of humans with WIV , their role during recall responses may be less critical , given the fact that in general WIV , SU , and SV induce similar HI titres in primed populations [11] . Use of WIV derived from wild-type virus instead of recombinant vaccine strains resulted in good antibody titres even without the addition of adjuvants and might thus be an option to obtain satisfying immune responses [35] . Evaluation of adjuvants in combination with WIV in clinical trials is so far restricted to aluminium salts . However , where adjuvanted and non-adjuvanted WIV were compared side-by-side , effects of this Th2 adjuvant on vaccine efficacy were absent , poor , or inconsistent [36] . So , better adjuvants have to be found that work synergistically with WIV in order to exploit the full potential of intact inactivated virus particles as vaccines . In conclusion , our data reveal , for the first time to our knowledge , that TLRs play an eminent role in the immune responses to a classic influenza vaccine . Of the three influenza vaccine formulations studied here , only WIV efficiently triggered TLR7-mediated mechanisms leading to superior immune responses . Processing of inactivated whole virus particles into SV or SU eliminates the immuno-potentiating effect of the viral ssRNA , the primary PAMP in WIV vaccine , and results in a loss of quantity and shift in the quality of the immune response . Thus , TLR-dependent mechanisms appear to form the basis for WIV's antigen-sparing quality and hence its recognized strong potential as a pandemic vaccine candidate [7] , [12] . Optimizing TLR7-signalling by rational vaccine design may produce even more potent vaccines , which are urgently needed in the face of the current influenza pandemic threat .
H5N1 virus ( NIBRG-14 , a 2∶6 recombinant of A/Vietnam/1194/2004 [H5N1] and A/PR/8/34 [H1N1] virus produced by reverse genetics technology ) was provided by the National Institute for Biological Standards and Controls ( NIBSC; Potters Bar , UK ) , propagated on embryonated chicken eggs , inactivated with 0 . 1% β-propiolactone to obtain WIV , and processed into split virus vaccine or subunit vaccine according to standard procedures [37] , [38] . The haemagglutinin protein concentration in the vaccines was determined by single radial immunodiffusion ( SRID ) [39] . Endotoxin levels in all vaccines met the requirements of the European Pharmacopoeia standard . ( If , nevertheless , contamination of endotoxin [signalling via TLR4] would have played an important role we should have observed substantial differences in the response between TLR7-deficient mice [capable of signalling via TLR4 ) ]and MyD88/TRIF-deficient mice [deficient in all TLR-derived signalling] . However , such differences were not found for any of the vaccines . ) CpG DNA ( ODN D19 ) was purchased from Eurogentec ( Seraing , Belgium ) . For immunization experiments , C57BL/6 , TLR7−/− and MyD88−/−/TRIF−/− mice ( generated from MyD88−/− mice [40] and TRIF−/− mice [41] ) were bred at the University of Massachusetts Medical School ( Worcester , MA ) . For in vitro studies , 10- to 12-week-old female C57BL/6 mice were purchased from Harlan Netherlands B . V . ( Zeist , The Netherlands ) , and TLR7−/− mice ( a gift from S . Akira and C . Reis e Sousa ) were bred at the University Medical Center Groningen . All experiments were conducted with approval of the local Institutional Animal Care and Use Committees . Mouse groups were matched for sex and age . Groups ( n = 6–8 ) of C57BL/6 , TLR7−/− , and MyD88−/−/TRIF−/− mice were intramuscularly injected with 50 µl of PBS in each calf muscle containing a total of 5 µg haemagglutinin protein per mouse of either WIV , SV , or SU vaccine formulation or no vaccine as a control . At 28 days after immunization , sera and spleens were collected for evaluation . Relative viral RNA content of the different vaccines was determined using a two-step real-time RT-PCR assay amplifying a 193-bp fragment within the M1 gene of influenza A viruses . For this purpose RNA was extracted from WIV , SV , or SU ( 5 µg HA ) with the QIAamp viral RNA Mini Kit ( QIAGEN , Venlo , The Netherlands ) , cDNA synthesis was performed on 5 µl of viral RNA ( one-tenth of the final elution volume ) using the Verso cDNA kit from ABgene ( Westburg , Leusden , The Netherlands ) , and 1 µM UNI12 primer ( 5′-AGCAAAAGCAGG-3′ , corresponding to viral noncoding nucleotides 1 to 12 [42] ) . Real-time PCR was performed with 200 nM M1-FOR primer ( 5′-CCTGGTATGTGCAACCTGTG-3′ ) and M1-REV primer ( 5′-AGCCTGACTAGCAACCTCCA-3′ ) ; purchased from Eurogentec , and the Absolute QPCR SYBR Green Mix ( ABgene ) . Amplification was performed on a StepOne apparatus ( Applied Biosystems ) , and consisted of 15 min initial activation at 95°C , followed by 40 thermal cycles of 15 sec at 95°C and 60 sec at 60°C . In each experiment , a standard curve ( R2>0 . 99 within the range of 1×102 to 1×109 copies per reaction ) was drawn to convert the respective cycle threshold ( Ct ) values into the number of viral genome copies . This standard consisted of a pCR2 . 1-TOPO plasmid construct in which was cloned a 473-bp sequence of influenza A/Puerto Rico/8/34 segment 7 . The HI assay was performed as described before [15] . Briefly , heat-inactivated mouse serum was absorbed to 3 volumes 25% kaolin/PBS ( Sigma-Aldrich , Inc . , St . Louis , MO ) , 20 min at room temperature ( RT ) . After centrifugation , 50 µl of supernatant was serially diluted two-fold in a round-bottom microtitre plate ( Costar , Corning Inc . , Corning , NY ) , in duplicate . Subsequently , 50 µl PBS was added containing 2 HAU of H5N1 ( NIBRG-14 ) virus and incubated for 40 min at RT . We used 2 HAU of virus instead of the standard 4 HAU to increase the sensitivity of the assay . Finally , 50 µl of 1% guinea pig erythrocytes ( Harlan ) in PBS was added to each well and HI titres were determined after 2 h incubation at room temperature . HI titres are given as the reciprocal of the highest serum dilution producing complete inhibition of haemagglutination . The levels of virus-neutralizing ( VN ) serum antibodies were determined with a VN assay [15] , [43] . The VN titre was defined as the reciprocal of the highest serum dilution capable of inhibiting 200 TCID50 of H5N1 vaccine strain virus ( NIBRG-14 ) from infecting Madin-Darby canine kidney cell monolayers in a microtiter plate . Infection was measured by an ELISA on intracellularly produced viral NP protein . Inhibition of infection by simultaneous incubation with mouse serum was established if the ELISA absorbance value ( A492 ) measured was below the cut-off value , determined by the equation: [ ( average A492 of the positive controls ( infected cells ) minus average A492 of the negative controls ( non infected cells ) ) divided by 2] plus the average A492 of the negative controls . Serum samples were tested in quadruplicate . Microtitre plates ( Greiner , Alphen a/d Rijn , The Netherlands ) were coated with 0 . 2 µg influenza H5N1 ( NIBRG-14 ) subunit vaccine per well in 100 µl coating buffer , overnight . After blocking with 2% milk in coating buffer for 45 min , 100 µl of two-fold serial dilutions of serum samples in 0 . 05%Tween 20/PBS ( PBS/T ) were applied to the wells and incubated for 1 . 5 h , in duplicate . Subsequently , 100 µl of horseradish peroxidase-conjugated goat anti-mouse IgG-isotype antibody ( Southern Biotech , Birmingham , Alabama ) was applied for 1 h . All incubations were performed at 37°C . Staining was performed using o-phenylene-diamine ( OPD ) ( Eastman Kodak Company ) and absorbance was read at 492 nm ( A492 ) with an ELISA reader ( Bio-tek Instruments , Inc . ) . After subtraction of background levels , serum dilutions yielding an OD of 0 . 2 were calculated using linear regression , of which the reciprocal of the average of the duplicates represents the titre . This assay was performed as described previously [15] . In short , erythrocyte-depleted splenocytes were seeded at a concentration of 5×105 cells in 100 µl medium per well , in triplicate in a microtitre plate ( Greiner ) , which was pre-coated with anti-IFNγ or anti-IL4 capture antibody ( Pharmingen , San Diego , CA ) and blocked with 4% BSA/PBS ( Sigma-Aldrich ) . Cells were stimulated with 1 µg H5N1 ( NIBRG-14 ) subunit vaccine per well , overnight in a humidified CO2 incubator at 37°C . Cells were lysed with 100 µl of H2O per well and plates were washed extensively , after which 100 µl of biotinylated anti-IFNγ or anti-IL4 ( Pharmingen ) in 2% BSA/PBS was added 1 h at 37°C . Subsequently , the plates were incubated with 100 µl of alkaline phosphatase conjugated streptavidin ( Pharmingen ) in 2% BSA/PBS for 1 h at 37°C , spots were visualized with 5-bromo-4-chloro-3-indolylphosphate ( Sigma-Aldrich ) substrate immobilized in solidified agarose . Plates were scanned and spots were counted manually . Plasmacytoid DCs were generated from bone marrow cells of C57BL/6 or TLR7−/− mice by seeding 1–2×106 bone marrow cells per well of a 24-well plate and culturing the cells for one week in Iscove's Modified Dulbecoo's Medium ( IMDM ) with 10% FCS and 100 ng/ml FLT3L ( R&D Systems , Abingdon , UK ) [22] . Single splenocyte suspensions were produced by collagenase D ( Roche Diagnostics GmbH , Germany ) treatment of the spleens , and spleen cell populations enriched for plasmacytoid DCs ( pDCs ) were obtained after magnetically labelling of pDCs with anti-mPDCA-1 antibody conjugated MicroBeads ( Miltenyi Biotech GmbH , Germany ) and separation over a MACS Column ( Miltenyi ) , according to the manufacturers protocol . Percentages of pDCs in the positively selected population were determined by FACS analysis using anti-mPDCA-1-PE antibody ( Miltenyi ) and anti-CD11c-FITC ( GeneTec Inc . , Canada ) . Cell suspensions containing 1–2×105 pDCs in 100 µl were seeded in a microtitre plate and stimulated in triplicate with an equal volume containing 1 . 0 µg HA of either WIV , SV , or SU vaccine , or 1 . 0 nmol CpG DNA . After 20 h of incubation in a humidified CO2 incubator at 37°C , supernatants were collected and subjected to the IFNα ELISA . IFNα detection in cell-culture supernatants was performed using a sandwich ELISA as described previously [15] . IFNα concentrations were calculated from a recombinant IFNα ( HyCult , Biotechnology , Uden , The Netherlands ) standard curve performed in quadruplicate using linear regression , and expressed in units per ml . Statistical analysis on HI titres , antibody titres , and Elispot counts was performed with SPSS ( SPSS 1202 Inc . , Chicago , IL ) using the Mann-Whitney U test with a CI of 95% . All p values are two-tailed . Statistical significance was defined as p<0 . 05 .
|
The rise and spread of the highly pathogenic avian H5N1 influenza virus has seriously increased the risk of a new influenza pandemic . However , the number of vaccine doses that can be produced with today's production capacity will fall short of the demand in times of a pandemic . Use of inactivated whole virus ( WIV ) vaccines , which are more immunogenic than split virus or subunit vaccines in an unprimed population , could contribute to a dose-sparing strategy . Yet , the mechanisms underlying the superior immunogenicity of WIV vaccine formulations are unknown . Here , we demonstrate that the viral RNA present in inactivated virus particles is crucial for the improved immunogenic properties of WIV in mice . By triggering Toll-like receptor 7 ( TLR7 ) , the viral RNA activates innate immune mechanisms that augment and determine subsequent adaptive responses . Efficient TLR7 signalling is lost in split virus and subunit vaccines with the processing steps that lead to disruption of the integrity of the virus particle and exclusion of the RNA . Our results prove for the first time to our knowledge that the immune-potentiating mechanism of a classic vaccine is based on activation of the innate immune system by one of its structural components . These findings may reflect a general principle for viral vaccines and provide a rational basis for further improvement of influenza vaccines , which are urgently needed in the face of the current H5N1 pandemic threat .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[
"microbiology/immunity",
"to",
"infections",
"microbiology/applied",
"microbiology",
"immunology/immunomodulation",
"virology/vaccines",
"immunology/immune",
"response",
"microbiology/innate",
"immunity",
"immunology/innate",
"immunity",
"virology/animal",
"models",
"of",
"infection",
"infectious",
"diseases/viral",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases/respiratory",
"infections",
"immunology/leukocyte",
"signaling",
"and",
"gene",
"expression",
"immunology/immunity",
"to",
"infections",
"microbiology/medical",
"microbiology",
"public",
"health",
"and",
"epidemiology/immunization"
] |
2008
|
Superior Immunogenicity of Inactivated Whole Virus H5N1 Influenza Vaccine is Primarily Controlled by Toll-like Receptor Signalling
|
Mathematics is often used to model biological systems . In mammary gland development , mathematical modeling has been limited to acinar and branching morphogenesis and breast cancer , without reference to normal duct formation . We present a model of ductal elongation that exploits the geometrically-constrained shape of the terminal end bud ( TEB ) , the growing tip of the duct , and incorporates morphometrics , region-specific proliferation and apoptosis rates . Iterative model refinement and behavior analysis , compared with biological data , indicated that the traditional metric of nipple to the ductal front distance , or percent fat pad filled to evaluate ductal elongation rate can be misleading , as it disregards branching events that can reduce its magnitude . Further , model driven investigations of the fates of specific TEB cell types confirmed migration of cap cells into the body cell layer , but showed their subsequent preferential elimination by apoptosis , thus minimizing their contribution to the luminal lineage and the mature duct .
Mathematical models have been used to inform basic biological research for well over 100 years [1] . Today , mathematical and computational models are particularly useful as a complementary approach to lab-based studies in developmental biology . Such models have been developed to investigate morphogenetic phenomena [2 , 3] in a broad sense , and in particular , for developmental pattern formation ( e . g . [4 , 5] ) , vascular branching ( e . g . [6] ) , cell morphology ( e . g . [7] ) , angiogenesis ( e . g . [8 , 9] ) , as well as both normal ( e . g . [10–13] ) and cancerous ( e . g . [14–16] ) development and regeneration/renewal ( e . g . [17–19] ) in many tissue types , including breast [20 , 21] . Branching morphogenesis specifically has been modeled ( see , e . g . [22] for a review ) in the lung [23] , kidney [24] , vasculature structures ( e . g . [6] ) , plants ( e . g . [25 , 26] ) , and to a certain extent , the mammary gland [11 , 27–30] ) . In most of these studies , the modeling framework includes reaction-diffusion equations to account for soluble factors driving the growth ( e . g . VGEF , FGF ) or playing a regulatory role ( e . g . Wnt , TGFβ ) . With respect to the normal mammary gland , several groups have modeled normal physiological processes . Grant et al . developed a model of mammary ductal morphogenesis adapted from a cellular-automaton model of vascular morphogenesis [27] . While this model was capable of approximating mammary branching gross morphology , it did not take into account the arrangement of the cells at small scales , or use any experimentally derived measurements . Nelson et al . also investigated branching processes in the mammary gland with engineered epithelial tubes . Although the process described does not directly mimic mammary development in vivo , they were able to mathematically model spacing of separate ductal units [31] . At smaller scales of interest , Tang et al . created an agent based model of breast acinus formation in vitro , in which they were able to determine the proliferation and apoptotic dynamics required for proper lumen formation and DCIS development [11] , whereas Rejniak et al . have proposed a sophisticated single deformable cell based model to derive conditions for acinus structure and lumen stability [30] . Although the works cited above have been successful in modeling particular aspects of normal mammary development ( i . e . ductal tree and acinus formation and some of the corresponding regulation ) , there are currently no published models of mammary ductal elongation , a critical process required for , and intimately coupled with , branching morphogenesis [32] . In this work , we sought to fill this gap by generating a mathematical model of mammary ductal elongation that not only incorporates the structure responsible for the majority of pubertal development , the terminal end bud ( TEB ) [33] , but also accounts for all of the main biological processes occurring within the TEB that are required for its growth . In the mammary gland , the rudimentary mammary duct formed during embryonic development remains relatively quiescent until puberty . Ductal elongation during puberty is driven by , and dependent upon , TEBs [32] . TEBs are bulbous , multi-layered structures that direct the growth of the duct throughout the fat pad during puberty . Mammary gland organogenesis in the virgin animal is dependent largely on this process of ductal elongation . During elongation TEBs are responsible for fat pad invasion , bifurcation of main ducts , lumen formation , basement membrane deposition , and the recruitment of periductal stroma and a blood supply . These critical functions , and the geometrically-constrained manner in which they are accomplished , make the TEB an ideal structure on which to base a mathematical model for ductal elongation . In addition , the ductal elongation rates in the murine mammary gland far exceed the allometric growth of other organs in the postnatal mouse , making this structure of unique interest . Lastly , TEBs have been postulated as a major site susceptible to carcinogenesis , leading to tumors later in life [34 , 35] . These characteristics make the TEB not only a good model for normal ductal morphogenesis , but also an ideal ground state for modeling breast tumorigenesis . In order to begin to model this structure we base the equations on the following established TEB biology . TEBs have two main cellular compartments . The outer monolayer of the TEB is known as the cap cell layer , which differentiates into the myoepithelial cell layer of the mature duct [33 , 36 , 37] , and is a putative reservoir for mammary stem cells [33 , 38 , 39] . Underlying the cap cell layer is a multi-cellular layer called the body cell layer . The cells in the body cell layer give rise to the layer of luminal epithelial cells lining the interior of the duct [36 , 37] . Apoptosis in the body cell layer has been postulated as the primary mechanism of lumen formation . Finally , high levels of proliferation in these two compartments of the TEB produce the cells that form the mature duct . We employed an investigative and integrative approach [19] using repeated iterations of biological measurements and corresponding parameter estimation to identify key processes required for ductal elongation . Through mathematical output calculation , we show that evaluation of just five experimentally measurable parameters , ( cell type-specific proliferation , cell type-specific apoptosis , cell size and number , and duct tortuosity ) , are sufficient to predict ductal displacement and elongation rates accurately . These data , and the associated modeling iterations required to interpret them , indicate that current common interpretations of developmental phenotypes can be inaccurate and require careful reconsideration . This model serves as a baseline model upon which agent-based , cellular automaton , and other mathematical models can be built to explore both normal mammary gland and breast cancer development .
In order to translate the current knowledge and concepts of mammary gland ductal development into a mathematical model for ductal elongation and displacement , we started the modeling process with a description of the time-invariant geometry of TEBs . We defined the stereotypical morphology of a single TEB in a moving frame ( i . e . invading through the fat pad with the speed associated with the ductal elongation ) and divided the TEB into eight regions ( Fig 1A and S1 Fig ) . Within the TEB structure , each region contains cells ( treated as a single population ) that are capable of proliferation and apoptosis . We then used a population dynamics model to describe the fate ( proliferation or death ) of these cells . In order to maintain a constant number of cells in each region , and a constant size of the regions , proliferation in a given region generates a flux of cells from that region into an adjacent one ( Eq 1 ) . Ductal elongation is the net result of these regional fluxes: the number of newborn cells in the TEB generates cell fluxes ( based on local conservation laws ) , with deposition of these cells in the mature duct , thus increasing the length of the subtending duct as the TEB invades ( S1 Fig ) . Thus , the net flux from the entire TEB into the mature duct dictates the rate of elongation . We model each region i of the TEB as a stationary compartment that contains a time-invariant number of cells and write the corresponding balance of gain ( cell proliferation and influxes ) and loss ( cell death and outfluxes ) terms: Model 1: Base equation . Ni : time-invariant cell number in Region i ri : average population growth ( proliferation ) rate in Region i di : average population death ( apoptotic ) rate in Region i φl→i , φi→l : cell influxes ( from adjacent regions l to i ) and outfluxes ( from i to l ) The derivation of the ductal elongation rate can be performed by considering the growth of either the outer myoepithelial ( basal ) layer of the duct ( Eq 1 for Regions i ∈ {1 , 2 , 3} ) or the inner ( luminal ) layer ( Eq 1 for Regions i ∈ {5 , 6 , 7} ) . For the purposes of the initial model , we assumed that cap cells are entirely responsible for the formation of the myoepithelial layer of the mature duct , and that the body cells are entirely responsible for the formation of the luminal layer of the mature duct . Such assumptions result in the only non-zero fluxes to be φ1→2 , φ2→3 and φ3→4 within the basal layer , and φ5→6 , φ6→7 and φ7→8 within the luminal layer ( S1 Fig ) . These fluxes correspond to the net production , by unit time , of newborn cells in each region , which result in the net cell fluxes Φbas ( Eq 2 ) corresponding to the elongation of Region 4 ( the basal layer ) and Φlum ( Eq 3 ) corresponding to the elongation of Region 8 ( the luminal layer ) . Cell flux from the TEB to the mature duct . Based on the assumption of homeostasis of the mature duct ( i . e . , apoptosis and proliferation processes balance in Regions 4 and 8 ) , the duct elongates only because cells are generated within the TEB , with time-invariant rates Φbas and Φlum for the basal and luminal layers , respectively . These two values , by definition , must match since the two layers adhere to one another , and one layer does not outpace the other during the process of elongation , as observed experimentally . The ductal elongation rate of each layer is finally found by equalizing the lateral surface area of the mature duct in a 2D cross-section to the surface monolayer covered by N adjacent cells and by taking the time derivative of these expressions ( Eq 4 ) . L ( t ) : length of the mature duct N ( t ) : cell number along the mature duct l : length of a single cell Φ=dNdt : time-invariant rate of cells provided to the mature duct by the TEB λ=dLdt : ductal elongation rate ( for a monolayer ) To determine the morphological characteristics that define the geometric framework of the model , manual measurements of each region and the individual cells within those regions were taken from histological sections of 5–6 week old paraffin embedded terminal end buds ( see Methods: Image Analysis and Measurement and Definition of Terminal End Bud Regions ) . Regional measurements are represented in Fig 1A as a scaled model of the TEB . Cellular dimensions are represented in Fig 1B and 1C as box plots . In Regions 1–4 , the cells become thinner ( from 10 . 1μm to 3 . 0μm ) and the intra-nuclear distance increases ( from 7 . 6 to 21 . 5μm ) as they mature in Region 4 ( Fig 1B ) . The cells in Regions 5–8 are relatively the same size ( Fig 1C ) . Regions 1 and 5 are represented as hollow half-discs and make up the leading tip of the TEB . The total length of these two regions is 85 . 8μm ( 75 . 7+10 . 1 ) , and the width of these two regions is 164 . 8μm ( 144 . 6+ ( 10 . 1x2 ) ) ( Fig 1A ) . Regions 2 and 6 are represented as trapezoids , with the larger end joining Regions 1 and 5 and sharing the same total width . The smaller end of Regions 2 and 6 join with Regions 3 and 7 at a total width of 91 . 3μm ( 70 . 9+ ( 10 . 2x2 ) ) ( Fig 1A ) . Regions 3 and 7 maintain approximately this width for a distance of 216 . 5μm , which we defined as the length of duct refractory to ovarian hormone treatment ( S2 Fig ) . The mature regions of the duct , Regions 4 and 8 , have a total width of 66 . 5μm ( 60 . 5+ ( 3 . 0x2 ) ) ( Fig 1A ) . All measurements are summarized in Table A in S1 Text . To determine the proliferation rate and cell cycle duration within each of the defined regions of the TEB , we conducted a dual pulse labeling experiment in which proliferating cells were labeled at time zero by incorporation of 5-ethynyl-2-deoxyuridine ( EdU ) . Proliferating cells were subsequently labeled at two hour increments over 24 hours with a 5-bromo-2-deoxyuridine ( BrdU ) pulse , and glands were harvested two hours after BrdU pulse labeling ( see Methods Section: Determination of Proliferation Rates and Cell Cycle Dynamics ) . Representative images from each time point are presented in Fig 2A . In all regions , the number of EdU/BrdU double positive cells was highest at time 0 , and reached its lowest levels eight hours after EdU labeling , indicating an S -phase duration of about six hours ( +/- ~30 minutes ) . In Regions 1 , 2 , 3 , and 5 , the number of double positive cells peaked 16 ( Regions 1 , 2and 5 ) , 18 ( Region 3 ) , or 22 ( Region 3 ) hours after EdU labeling , suggesting that cells in each region re-enter the cell cycle . In Regions 6 and 7 , the number of double positive cells remained low , suggesting that these cells do not re-enter the cell cycle at high frequency ( Fig 2B ) . Total proliferation levels did not change substantially over time in each region ( S3A Fig ) . Additional time points out to 120 hours were also investigated ( S5 Fig ) . To determine the duration of G2/M phase in the EdU labeled cells , we performed further EdU/BrdU pulse time points at 30 min , 1 hour and 1 . 5 hours and performed immunofluorescent ( IF ) staining for phospho-histoneH3 ( pHH3 ) colocalization with EdU . Levels of EdU+pHH3+ cells peaked in all regions two hours after EdU labeling . By eight hours after EdU labeling , cells were no longer co-stained with pHH3 . Together with the previous finding that the S phase is ~6 hours long , this indicates a G2/M phase duration of two hours +/- approximately 30 min ( S3B and S3C Fig ) . To determine the level of apoptosis occurring within each region of the TEB , we performed IF staining for cleaved caspase III ( CC3 ) , a marker of mid-stage apoptosis ( See Methods Section: Immunofluorescent Staining Of Tissue ) . Representative images from each time point are shown in Fig 3A . The majority of apoptosis occurs in the inner regions of the TEB ( 4 . 0% ±0 . 5 in Region 5 , 5 . 3% ±0 . 4 in Region 6 , and 3 . 5% ±0 . 5 in Region 7 ) ( Fig 3B ) . Although the location of apoptosis is consistent with published TUNEL assay data ( an indicator of late stage apoptosis ) [40] , the total levels are ~50% lower probably due to differences in apoptotic stage and the possibility of non-CC3-mediated cell death mechanisms being used . Apoptosis levels in the outer basal layer were less than 1% ( 0 . 1% ±0 . 1 in Region 1 , 0 . 3% ±0 . 1 in Region 2 , and 0 . 5% ±0 . 3 in Region 3 ) . Additionally , the time of CC3 positivity after EdU labeling was found to peak between 6 and 10 hours ( S4 Fig ) . To calculate the rate at which the mammary duct elongates , measurements corresponding to the current scientific knowledge ( proliferation and apoptosis rates ( S1 Text , section A ) , cell cycle and apoptosis times , and regional cell numbers ( Table B in S1 Text ) were used to calculate net population growth rates in the basal regions 1 , 2 , and 3 and the luminal regions 5 , 6 , and 7 . These growth rates were input into the base mathematical equation ( Model 1 ) ( Eqs 2–4 ) . The predicted linear elongation rate for the basal layer was 1 . 24 ( ±0 . 09 ) mm/day , while the predicted elongation rate for the luminal layer was 0 . 78 ( ±0 . 07 ) mm/day . This difference indicated that these initial parameters , which encompass much of the known biology of TEB , were not sufficient to model elongation because the results indicate that the two layers do not grow in a coordinated fashion , but rather that they elongate independently , which is not in accordance with observed biology . This result indicated that our model is incomplete . To address what process ( es ) were missing from our model , we sought to quantify a previously unappreciated experimental observation that cap cells migrate into the body cell layer [33] . In 1983 , Williams and Daniel used time-lapse video to study TEBs and noted that cap cells “migrate from their peripheral location into the deeper regions of the end bud” [33] . Indeed , we found many SMA and p63 positive cells within Regions 5 and 6 ( Fig 4A and 4B ) ; however , the fate of these cells was still unknown . This cap cell migration introduces an additional flux of cells from the basal to the luminal regions into the model , and would therefore decrease the predicted elongation rate for the basal layer , while potentially increasing the predicted elongation rate for the luminal layer . One hypothesis is that migrating cap cells contribute to the body cell layer and may represent mammary multipotent progenitors or stem cells [33 , 41] . To address the fate of these cells , we performed staining for SMA and either BrdU or pHH3 . Quantification of these images revealed that while SMA positive cells in Regions 1 and 2 have high levels of BrdU incorporation ( 37 . 2% ±4 . 5 and 31 . 9% ±3 . 9 respectively ) ( Fig 4C ) , SMA positive cells in Regions 5 and 6 had significantly lower levels ( 7 . 1% ± 1 . 8 and 4 . 0% ± 2 . 0 respectively , p<0 . 0001 ) ( Fig 4C ) , implying that cap cells present in the body cell layer are not as proliferative . Additionally , staining for SMA and pHH3 indicate that cap cells in the body cell layer were not undergoing mitosis ( 0 . 0% in Region 5 and 6 versus 7 . 6% ± 1 . 0 in Region 1 and 8 . 0% ± 1 . 1 in Region 2 , p<0 . 05 ) ( Fig 4C ) . To determine if migrating cap cells undergo apoptosis , we stained for SMA and CC3 . Quantification of these images found that indeed cap cells present in Regions 5 and 6 were undergoing cell death at a much higher rate than SMA positive cells in Regions 1 and 2 ( 46 . 9% ± 4 . 0 in Region 5 and 42 . 0% ± 4 . 1 in Region 6 versus 1 . 8% ± 0 . 4 in Region 1 and 2 . 3% ± 0 . 4 in Region 2 , p<0 . 0001 ) ( Fig 4D ) . Overall , these data suggest that cap cells present in the body cell layer are no longer proliferative , and are actively undergoing apoptosis . Thus , they do not appear to contribute substantially to the luminal layer of the mature duct . These results are contrary to the current consensus that cap cells contribute substantially to the luminal lineage . This result also indicated that the mathematical description of cap cell movement from the basal layer into the luminal layer , with subsequent elimination of those cells from the system , was an important missing component to our model . We accounted for cap cells migrating into the body cell layer by considering a flux of cells φbas→lum ( = φ1→5 + φ2→6 + φ3→7 , see S1 Fig ) , which leads to the modified fluxes Φbas − φbas→lum and Φlum + φbas→lum to be considered for the evaluation of the basal and luminal elongation rates ( Model 2 ) , respectively ( S1 Text , section B ) . When cap cell specific death is accounted for , our model revealed that the apoptotic index experimentally measured for the luminal layer is underestimated , resulting in inconsistencies in the model . To address this issue , an apoptotic correction factor ( δ = 97% ) was introduced ( S1 Text , section C and Table C ) into the equation for the luminal layer ( Model 3 ) , yielding an apoptotic index of 8 . 5% . Inclusion of the additional flux and apoptotic correction factor into our equations increased the predicted elongation rate for the luminal layer from 0 . 78 to 0 . 81 ( ±0 . 08 ) mm/day , and decreased the rate for the basal layer from 1 . 24 mm/day to 0 . 76 ( ±0 . 12 ) mm/day , bringing these two calculated elongation rates much closer together in accordance with the observation that the layers elongate in a coordinated fashion . To validate the accuracy of our prediction we sought to measure ductal elongation in vivo . To validate the model prediction that the gland elongates at a rate in the range of 0 . 76–0 . 81 mm per day , we measured the rate at which the TEBs are displaced during pubertal elongation in vivo . To do this we measured outgrowths in both inguinal glands of FVB mice from the nipple to the ductal front . The length of ductal outgrowth was then compared across each time point to provide the TEB displacement rate ( Fig 5 ) . The rate of displacement from weeks 5 to 6 ( 0 . 57mm/day ) , 6 to 7 ( 0 . 25mm/day ) , and 7 to 8 ( 0 . 82mm/day ) averages 0 . 54 mm/day over the three week period , a rate significantly less than the predicted elongation rate of 0 . 76–0 . 81mm/day . Our model prediction represents the linear elongation rate of a new duct formed as a straight tube ( or the speed at which the new duct forms in a straight line ) , which is different from the displacement rate that we measure here . The displacement rate is not a measure of the total length of duct , but rather it is based on the shortest distance between the nipple and the ductal front and does not take into account the actual path ( i . e . turning ) of the TEB . In order to compare our mathematical elongation rate prediction to the measured displacement rate , we must convert the linear output to its displacement equivalent . Tortuosity refers to the characteristic of a winding or twisting path as opposed to a straight path . The path of the TEB is tortuous due to both bifurcation events and natural turning . Because bifurcation is a frequent event during elongation , we sought to determine the effect that bifurcation has on the straight line route of a TEB’s growth path . We measured the angle at which a TEB is deflected off its path at the time of bifurcation by measuring the angle between the original path and the path after bifurcation ( Fig 6A ) ( see Methods: Image Analysis and Measurement ) . We found that the average angle of deflection is 35 . 5° ( ±1 . 9° , n = 62 ) ( Fig 6B ) . In addition , we evaluated the frequency of bifurcation during pubertal outgrowth . This was determined by measuring the distance ( path length ) between bifurcation events ( Fig 6C ) . The average distance between bifurcation forks was 1141μm ( ±99 . 1 , n = 23 ) ( Fig 6D ) . Additionally , we compared path tracing measurements to corresponding displacement measurements between bifurcation events and found that the displacement measurements consistently underestimate the total distance by 6 . 1% ( ±0 . 9% , n = 23 ) ( Fig 6E ) . We used the measurements of TEB path tortuosity to convert ( S1 Text , section D ) our previously predicted linear elongation rates ( 0 . 81 and 0 . 76 mm/day ) into the measured displacement rate using our ultimate model ( Model 4 ) ( Eqs 5–6 ) which accounts for the cap cells dropping into the body cell layer ( S1 Text , section D ) : Model 4 . λbas=lbas2T ( ∑i=1 , 2 , 3 ( ri−di ) Ni−φbas→lum ) ( 5 ) λlum=llum2T ( ∑i=5 , 6 , 7 ( ri−di ( 1+δ ) ) Ni+φbas→lum ) ( 6 ) where , Tortuosity due to bifurcation ( a factor of 1 . 23 ) and tortuosity due to turning ( a factor of 1 . 06 ) are calculated using measurements determined in the previous experiment ( S1 Text , section D ) . The predicted rate is converted to displacement rate by dividing our previous predictions of 0 . 81 and 0 . 76 mm/day by the total tortuosity ( T = 1 . 31 ) . This conversion gives a displacement rate of 0 . 62mm/day and 0 . 58mm/day for the luminal and basal layers , respectively ( Table D in S1 Text ) . The difference between the experimentally measured displacement rate of 0 . 54mm/day and our model’s prediction indicate that these parameters , including the addition of a novel cap cell flux with the luminal apoptotic correction factor , are sufficient to account for the kinetics of ductal elongation . These results also indicate that displacement measurements can underestimate total duct length by 24% . Our initial model , Model 1 , yielded single value predictions for the luminal and basal layers that were incompatible with known biology , and with our experimentally measured displacement rate of 0 . 54mm per day ( Model 1—Fig 7A ) . With the addition of the cap cell flux ( Model 2—Fig 7A ) the model outputs become dependent on the value of this flux , which then allows us to perform a model fit analysis: for a fixed value of δ , φbas→lum ( and therefore , the corresponding cap cell fraction X , see S1 Text , section E ) can be determined by using the matching condition λbas = λlum . Without accounting for cap cell specific death ( δ = 0 ) , the fit predicts an elongation rate that is incompatible with our experimentally measured rate ( Model 2—Fig 7A ) . However , addition of the minimal apoptotic correction factor of δ = 0 . 97 ( S1 Text , section C ) yielded a cap cell flux value ( corresponding to X = X2 = 49% ) leading to an elongation rate still outside the error of our experimentally derived levels ( Model 3—Fig 7B ) . When we account for tortuosity , the linear elongation rate is converted to a displacement rate ( Model 4—Fig 7C ) and we find agreement between the model fitting and our experimentally measured rate , however with a very large range of admissible values of the cap cell flux ( Fig 7C red error bar ) . The ultimate model , when constrained by specific values estimated experimentally ( Model 4—Fig 7D ) for X = X* = 54% and for δ = 0 . 97 ( S1 Text , section E ) , yields values ( λlum = 0 . 62mm / day and λbas = 0 . 58mm / day , Fig 7D blue and black dots ) that fall within the error of the experimentally measured displacement rate . When we compare these values to those found from model fitting with δ = 0 . 97 ( Fig 7D blue solid line and white dot ) we also find a rate within this error box . Finally , when we apply an apoptotic correction factor ( δ = 1 . 55 ) equivalent to the 11% apoptotic index reported by Humphreys et al [40] ( Fig 7D blue dotted line ) , we find a flux value by model fitting that exactly matches for both TEB layers ( white dot ) , and lies within the same error box . These results illustrate that our final model predictions , based on experimentally estimated values of X and δ , provides values of the elongation rates that 1 ) lie within the error range of our measurement and 2 ) are consistent with values forced by fitting but ignoring the biological cause , thus strengthening our final model and our modeling approach’s validity .
The TEB is the driving force for mammary duct invasion and elongation into the fat pad . Here , we exploited the geometric constraints on TEB size to develop a mathematical model for ductal elongation and displacement . This iterative modeling approach identified a set of experimentally measurable parameters required for accurate modeling . Our base model ( Model 1 ) incorporating only region-specific cell number , proliferation rate ( BrdU ) and apoptotic index ( CC3 ) , predicted two different elongation rates for the basal and luminal layers . This prediction was inconsistent with the biological observation that the two cellular compartments grow in a coordinated fashion , and required identification and evaluation of additional processes and parameters . Parameters were added sequentially to the model until the predicted elongation rates of the two cellular compartments could be matched ( Models 2 and 3 ) , thereby demonstrating that our investigative approach of ductal elongation helped identify the major biological processes involved , similar to the approach by Hoehme et al . [19] in the context of liver regeneration . Once the predicted elongation rate was established , conversion to an observed displacement rate ( Model 4 ) required measurement of average branch angle and duct tortuosity . The results from this process have a significant impact on how developmental phenotypes in elongating structures are interpreted generally , as well as for our understanding of cellular dynamics within the TEB . Cap cells have long been referred to as putative reservoir for mammary stem cells , and have been credited with contributing to both the basal and luminal lineages [33 , 38 , 42 , 43] . In 1983 , Williams and Daniel described time lapse video of cap cells migrating into the body cell layer and out of view . They hypothesized that cap cells that migrate into the body cell layer differentiate into luminal cells , and that they are examples of multipotent progenitor cells [33] . An alternative hypothesis is that migrating cap cells rejoin the basal layer as the duct elongates . Numerous microscopy analyses have shown the presence of cap cells in the body cell layer [44–46]; however , none of these studies have investigated the fate of cap cells once actually there . More recent lineage tracing experiments have led to conflicting conclusions about the contribution of cap cells to the luminal lineage . The most recent studies by Rios et al maintained the historical view that cap cells in the TEB give rise to luminal cells [36 , 37] . However their lineage tracing data did not address whether this contribution is due to migration of cap cells into the body cell layer , or due to asymmetric division directly from the cap cell layer into the body cell layer . The failure of Model 1 along with the conflicting positions on cap cell biology led us to investigate Williams and Daniel’s observation further . We demonstrate that the migrating cap cells are no longer proliferative by both decreased BrdU labelling and low pHH3 , and in fact , are actively undergoing programmed cell death at a high rate . Using the measurements of proliferation and the number of cap cells present in the body cell layer , we were able to calculate the percentage of newborn cap cells that migrate into the body cell layer to die . If we assume that cap cells are dropping in from only Regions 1 and 2 we calculate that a striking 54% of the daughter cells actually migrate into the body cell layer to undergo apoptosis ( 39% if region 3 is included 3 ) ( S1 Text , section E ) , and thus are not used productively for duct formation . Together with recent lineage tracing data [37] , these data strongly suggests that few , if any , cap cells contribute to the body cell layer , or rejoin the basal layer . The high levels of CC3+ cap cells present in the body cell layer forced us to reexamine the dynamics within the body cell layer and adjust the mathematical model . Although several groups have previously investigated cell death in the body cell layer and accepted it as a major mechanism of lumen formation [40 , 44 , 47 , 48] , none have confirmed the lineage of these dying cells . Our studies have shown the majority of CC3+ cells to be of cap cell origin , leaving only a small percentage of true body cells undergoing caspase-3-mediated cell death . When this phenomenon was introduced into the mathematical model in Model 3 , a correction factor of 97% was needed for the luminal layer in order to keep the mathematical model consistent ( S1 Text , section C ) . This correction factor represents an underestimation of total cell death by CC3 positivity . These model manipulations provide mathematical evidence that true body cell clearance levels are higher than our biological measurements have indicated . Given our results , it is possible that CC3-mediated apoptosis is a mechanism for removal of cap cells specifically , rather than a mechanism of luminal cell clearance . Previous work indicated that cap cells , when detached from the cap cell layer , undergo apoptosis in response to anoikis . In Netrin-1 KO mice , unattached cap cells die at a rate 17-fold higher than attached cap cells . This coincides with a 2-fold increase in the number of cap cells present in the body cell layer [49] . Additionally , targeted disruption of BIM , a pro-apoptotic factor , leads to a 5 . 6-fold increase in p63 positive cells in the TEB [44] . It remains unclear whether non-CC3-mediated cell death mechanisms in the body cell layer may play a role in lumen formation . In our ultimate model , Model 4 , the net increase in cell number , cell migration , and the geometric constraints within the TEB creates cellular fluxes into the mature duct , which results in forward movement of the TEB as it elongates through the fat pad . However , the ultimate displacement within the fat pad is modulated by changes in direction due to branching and turning ( tortuosity ) . The behavior of our final model has illustrated a significant problem with regard to the use of traditional measurements and terminology for characterizing developmental changes . Mammary gland researchers tend to describe mutant developmental phenotypes as ‘delayed’ or ‘promiscuous’ when they see a decrease or increase , respectively , in the distance from the nipple ( or lymph node ) to the ductal front ( a direct measurement of displacement ) , or in the percent fat pad filled ( an indirect measurement of displacement ) [50–52] . This is misleading language born from a failure to make the distinction between elongation rate and displacement rate and leads to a misunderstanding of the true nature of ductal development . As demonstrated by our measurements and mathematical model , the total distance a TEB must traverse as the duct elongates is significantly greater ( ~24% ) than a traditional displacement measurement would suggest . For example , in the Plk2 knockout ( KO ) mouse , a “delay” and “impairment” in ductal elongation was reported , with Plk2 KO glands filling the fat pads 2 weeks later than WT mice . However , increased proliferation and hyperbranching were also observed concomitant with the elongation “delay” [50] . In light of our demonstration that bifurcation decreases the measured displacement by 19% , it is not surprising that glands with increased branching take longer to fill the fat pad . In fact , this has been observed in a number of other mouse models with both gain- and loss-of-function mutations usually involving growth factor signaling [53 , 54] . While similar phenotypes have traditionally been labeled as developmental delays , they actually could represent a completely different kind of alteration in ductal development . The precise characterization of these phenomena will require a more careful analysis of the kinetics of ductal morphogenesis . In its current state , our model has some limitations . The first is that the current model only allows for changes in histology ( cellular number and arrangements ) not gross morphology ( increased branching ) . This is problematic because increased proliferation is frequently coupled with increased branching in the animal . The ultimate goal of this project is to create a multi-scale model which functions at different levels of development , from histology ( current model ) to cellular behavior and signaling , and gross morphology , which will address these limitations . A second limitation is that the parameters were evaluated as averages and assumed to be constant over time , while variations in these values could change the predicted elongation rate and may account for the natural variation in growth rates ( Fig 5 ) . We wish to note that our model is fully integrative , in the sense that it has not been ‘fit’ to the data , nor does it have degrees of freedom within the parameters , rather the data have been used exactly to inform the model . However , we were able to demonstrate that fitting our model to previously published data does validate our experimentally measured parameters , at least in the context of apoptotic index and cap cell migration ( Fig 7 ) . Because of the natural genetic variation in mammary ductal development it will be of interest to apply this mathematical model to other mouse strains that exhibit markedly different kinetics of ductal morphogenesis [55] . This model could also be adapted for modeling other branching organs , such as the salivary gland , lung , and prostate . In addition , because the TEB shares several important features with tumors ( heterogeneous cell populations , high proliferation levels , invasive behavior , and angiogenic properties ) , this model should also provide an ideal ground state for modeling and investigating ductal carcinoma in situ ( DCIS ) development as well as growth and behavior of invasive cancers in the breast .
FVB/NJ and FVB . Cg-Tg ( ACTB-EGFP ) B5Nagy/J mice were used for all experiments except to study cap cell migration into the body cell layer , for which Balb/C mice were used . All animal work was performed according to IACUC approved protocols . Female FVB virgin mice were harvested at exactly five , six , seven , and eight weeks of age . Total fat pad length was measured prior to dissection . Glands were harvested and fixed in 4% Paraformaldehyde for 3 hours and cleared in 50% Glycerol 50% PBS solution overnight . Both inguinal glands were stained with DAPI according to the protocol previously described [56] . Images were taken using a Leica MZ16F dissecting microscope equipped with a Leica DFC300 FX camera and analyzed with MetaMorph 7 . 1 ( Molecular Devices Inc . ) . The total length of the fat pad and the distance from the nipple to the ductal front were measured using MetaMorph 7 . 1 ( Molecular Devices Inc . ) . The true distance of ductal growth was calculated by multiplying the percent fat pad filled by the original length of the fat pad prior to harvest . A 2D schematic of a TEB medial longitudinal section was generated based on experimentally measured histological and morphological features of a typical TEB . Outer and inner compartments ( myoepithelial and luminal ) were designated as separate , then each was divided into regions based on morphology . To evaluate proliferation rates and cell cycle parameters , we conducted a dual pulse cell labeling experiment [57] . Five to six week old mice were pulsed with 50μg/ body weight EdU ( 25mg/ml in PBS ) by intraperitoneal injection at time 0 . An injection of 2mg/g body weight BrdU ( 25mg/ml in PBS ) was given to cohorts of mice in two hour increments after time 0 for a total of 24 hours . ( Additional long term pulse experiments were conducted with EdU given to all mice at time 0 and BrdU given at 24 hours increments for 120 hrs . ) Glands were harvested two hours after receiving the BrdU pulse ( 3 animals per time point ) . Glands were fixed in 4% PFA for 3 hours and stored in 70% EtOH before paraffin embedding . EdU was visualized using the Invitrogen Click-iT Kit . All images were counted manually . Harvested glands were fixed in 4% paraformaldehyde for 3 hours and then stored in 70% EtOH . Samples were embedded in paraffin and serially sectioned . Sections were deparaffinized in NeoClear and decreasing concentrations of EtOH ( 100% , 95% , 80% , and 70% ) . Antigen retrieval was accomplished by incubating slides in citrate buffer pH6 in a pressure cooker . Slides were cooled in water and washed in TBST before staining . Slides were blocked in 10% normal goat serum and MOM IgG blocker for 1 hour at room temperature . The different primary antibody combinations were incubated over night at 4°C . Sections were rinsed for 10 min in TBST and incubated with secondary antibodies 1:500 at room temperature for 1 hour . The following primary antibodies were used: anti-BrdU ( Invitrogen B35128 ) , anti-phospho Histone H3 ( Milipore 06–570 ) , anti-Cleaved Caspase III ( Cell Signaling 9961s ( D175 ) ) anti-p63 , anti-Cytokeratin 5 ( AbCam ab52635 ) , anti-Smooth Muscle Actin ( Sigma Aldrich A2547 ) . The CK8 monoclonal antibody developed by Phillippe Brulet and Rolf Kemler was obtained from the Developmental Studies Hybridoma Bank , created by the NICD of the NIH and maintained at The University of Iowa , Department of Biology , Iowa City , IA 52242 . The following secondary antibodies were used: anti-mouse Alexa Fluor 594 ( Invitrogen A11005 ) , and anti-rabbit Alexa Fluor 594 ( Invitrogen A11012 ) . Nuclei were stained with Dapi and slides were mounted using Vectashield ( Vector Laboratories ) . All measurements were done manually because automated methods do not perform well with histological sections of mammary tissue due to complex image regions/scenarios ( i . e . cells close together , apoptotic clumps of debris , multiple foci within one nucleus ) . Further , all measurements within a given experiment were done by a single individual in order to eliminate variation in methodology from individual to individual . All methods of measurements were reviewed for consistency and spot-verified by a second individual , as well as by the senior co-authors , over the course of the study . All raw images are available from the corresponding author upon request . The medial longitudinal section of each TEB was selected from serial sections and imaged . Imaged TEBs were broken down into regions 1–8 manually , and cells counted manually based on staining combinations . Sections were imaged using a Zeiss Axioskop 2 Plus fluorescent microscope equipped with an AxioCam MRm camera . Regional and cellular dimensions were measured using Zeiss Axiovision 4 . 8 software . Whole mount images were taken with a Leica MZ16F dissecting microscope equipped with a Leica DFC300 FX camera and analyzed with MetaMorph 7 . 1 ( Molecular Devices Inc . ) . Bifurcations were chosen when the TEBs could be traced back to the bifurcation event . The growth path just prior to bifurcation was traced , and the subsequent angles off of that path by each new TEB were measured using the Metamorph 7 . 1 ( Fig 6A ) . TEB regions were defined by morphological cues . A line drawn across a longitudinal histological section of a TEB at the widest point will divide Regions 1 and 5 from Regions 2 and 6 . The Regions forward of the line are Region 1 ( a single cell outer most layer ) and Region 5 ( which includes all cells between the lumen and the outer single layer ) . A line across the duct at the neck of the TEB will divide Regions 2 and 6 from Regions 3 and 7 . This line should be drawn at the location where the narrowing stops ( after which the duct becomes a uniform thickness ) . This length can vary between TEBs and an average value is presented in Fig 1 . Region 2 is the single cell outer layer within this length and Region 6 contains all the cells between the lumen and the single cell outer layer within this length . Regions 3 and 7 are defined by the boundary between TEB-proximal ductal cells lacking differentiation capacity when exposed to pregnancy hormones relative to more distal ductal cells that are capable of morphological differentiation in response to hormones , as demonstrated in S2 Fig . In our experiments with FVB mice , the length of these two regions is 216 . 48μm . Region 3 is the single cell outer layer within this length and Region7 contains all cells between the lumen and the single cell outer layer . Regions 4 and 8 are mature duct distant to the TEB and were measured from sections not containing TEB features .
|
Our paper describes a mathematical model of mammary ductal elongation during pubertal development . We make several conclusions that will be of interest to scientists studying mammary gland biology , epithelial tube formation , and branching morphogenesis . First , our model indicates that a common measurement of developmental outgrowth ( ‘percent fat pad filled’ ) underestimates the total growth and leads to mischaracterization of mutant phenotypes . Second , we show that cap cells , a population enriched with putative mammary stem cells , do not contribute to the luminal lineage as previously hypothesized . Further , we find that a high percentage of proliferation in these cells is not used productively to actually form the mammary duct . We believe our model has future application to other branching organs and also for the modeling of disease states in the breast .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"cell",
"death",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"mathematical",
"models",
"developmental",
"biology",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"lipids",
"fats",
"mathematical",
"and",
"statistical",
"techniques",
"exocrine",
"glands",
"breast",
"tissue",
"biochemistry",
"cell",
"staining",
"cell",
"biology",
"anatomy",
"apoptosis",
"cell",
"migration",
"biology",
"and",
"life",
"sciences",
"mammary",
"glands"
] |
2016
|
A Geometrically-Constrained Mathematical Model of Mammary Gland Ductal Elongation Reveals Novel Cellular Dynamics within the Terminal End Bud
|
Demographic , economic and behavioural factors are central features underpinning the successful management and biological control of dengue . This study aimed to examine these factors and their association with the seroprevalence of this disease . We conducted a cross-sectional telephone survey of households in a 3 km radius of the schools where we had conducted serological tests on the student population in a previous study . Households were surveyed about their socio-demographics , knowledge , practices , and Health Belief Model ( HBM ) constructs . The results were then associated with the prevalence rate of dengue in the community , as marked by IgG seropositivity of the students who attended school there . A total of 1 , 400 complete responses were obtained . The community's IgG seropositivity was significantly positively associated with high household monthly income , high-rise residential building type , high surrounding vegetation density , rural locality , high perceived severity and susceptibility , perceived barriers to prevention , knowing that a neighbour has dengue , frequent fogging and a higher level of knowledge about dengue . In the multivariate analyses , three major correlates of the presence of IgG seropositivity in the community: ( 1 ) high-rise residential apartment house type or condominium buildings; ( 2 ) the main construct of the HBM , perceived severity and susceptibility; and ( 3 ) the additional constructs of the HBM , lack of preventive measures from the community level and having a neighbour with dengue as a cue to action . Weak correlations were found between self-practices to prevent dengue and the level of dengue seropositivity in the community , and between HBM constructs and knowledge ( r = 0 . 09 ) . The residential environment factor and the constructs of the HBM are useful and important elements in developing interventions to prevent and control dengue . The study also sheds light on the importance of the need for approaches that ensure the translation of knowledge into practice .
The incidence of dengue has grown dramatically in recent decades [1] . Over 2 . 5 billion people , 40% of the world's population , are at risk from dengue , and each year an estimated 50–100 million dengue infections are reported worldwide [1] . The disease is currently endemic in more than 125 countries in Africa , the Americas , the Eastern Mediterranean , Southeast Asia and the Western Pacific [1] . Since the first reported case of dengue fever in Malaysia in 1902 , dengue has remained a serious public health problem in this country . Malaysia has experienced several major outbreaks , which were reported in 1974 , 1978 , 1982 and 1990 [2] . Likewise in Malaysia , the incidence of dengue fever and its more severe forms have increased dramatically in recent decades . The incidence rate of dengue shows an increasing trend from 44 . 3 cases/100 , 000 population in 1999 to 181 cases/100 , 000 population in 2007 [3] . High dengue IgG seropositivity ( >91% ) was found in a sample of Malaysian adults aged 35 to 74 years old in a recent study [4] . Serological tests conducted in Lundu District in Sarawak , Malaysia , found that in >23% of 215 samples , individuals had a history of dengue [5] . In another study , a nationwide sample of 1 , 410 children aged 7 to 18 years was surveyed and 11 . 0% were found to be positive for dengue IgG ( Tiong et al . , unpublished data ) . Among the most frequently applied methods of controlling or preventing dengue is the control of dengue virus transmission through mosquito reduction activities . Human behaviour is an important contributor to creating breeding grounds for mosquitoes and sustaining mosquito populations [6] . It has long been recognized that socio-demographic characteristics , beliefs and practices about dengue have an impact on dengue prevention and control . The basic concepts of the Health Belief Model ( HBM ) feature individual consideration of the likelihood ( susceptibility ) and seriousness ( severity ) of illness and the capacity of the individual to adopt the desired behaviour to prevent it [7] . Since its original conception , two additional concepts have been added to the HBM: self-efficacy , or one's confidence in the ability to successfully perform an action; and cues to action , external events that prompt readiness to make a health change [8] . A handful of studies have specifically applied the HBM in attempts to understand perceptions of risk and sustained dengue prevention [9]–[11] . In the context of health communication strategies for dengue prevention and control , the HBM provides a framework for understanding how to effectively structure messages and influence behavioural change [12] . Previous studies have primarily focused on investigating the constructs of the HBM in conceptualizing health beliefs and knowledge about the threat of dengue , the barriers to engaging in the desired preventive behaviour , and the subsequent development of self-efficacy to initiate the desired behaviours . This study differs from previous field investigations in that it focuses on investigating the HBM constructs and their association with the seroprevalence of dengue virus-specific IgG in the community . To the best of our knowledge , such a study has never been conducted elsewhere .
The seroprevalence data were obtained from our previous study ( Tiong et al . , unpublished data ) . Between 2008 and 2009 , anonymized serum samples of school children , aged 7 to 18 , from 26 schools throughout Malaysia were examined for dengue virus-specific IgG . Dengue IgG capture ELISA ( Standard Diagnostics , Korea , Cat . no . 11EK10 ) was used to test for the presence of anti-dengue-specific IgG antibodies . Of the 1 , 411 samples from 26 schools , 156 ( 11% ) were positive for dengue-specific IgG antibodies . The prevalence of seropositivity for dengue-specific IgG in the student population , calculated for each school , ranged from 0% to 25±1% . The percentage of seropositivity for dengue-specific IgG in the surrounding community ( within 3 km ) was assumed to be the same as that of the student population in the nearby school . Households within a 3 km radius of a school were therefore assigned a corresponding percentage of seropositivity for dengue-specific IgG . In the current study , the level of IgG seropositivity in the study participants was categorized as being ‘absent’ or ‘present’ . Further , a K-means cluster analysis procedure was also used to identify groups by the percentage of positive IgG results in the study participants . The households in a 3 km radius of the schools were surveyed about their knowledge and health belief constructs , the results of which were associated with the absence or presence of seropositivity and with the seropositivity groups identified in cluster analysis . The seroprevalence of dengue amongst the students was used as a surrogate for the prevalence of dengue in the community , as students were not likely to have extensively travelled outside their respective communities . The community around the school was chosen because Malaysian schools usually admit only those students living within a 10 km radius of the respective school . Because of the unmanageably large number of households within a 10 km radius of the schools , this study surveyed only those households within a 3 km radius . The study samples for the present study were households within a 3 km radius of the schools . Residential communities within a 3 km radius of the schools were first identified . Subsequently , we performed a cross-sectional study by contacting all households in the identified residential community with a registered landline telephone ( which served as the study sample of the population ) . To be eligible for a telephone interview , participants had to be Malaysian , aged between 18 and 60 years old , and resident in the contacted household . Only one person per household was surveyed . If more than one eligible person was found in a household , one person was selected randomly by using a random number table . Interviews were conducted between 5 . 30 p . m . and 10 . 00 p . m . on weekdays in order to avoid over-representation of unemployed participants , and from 10 . 00 a . m . to 7 . 00 p . m . on weekends or on public holidays . Interviewers made three attempts to call unanswered telephones on different days before regarding them as non-responses . After the survey questions were constructed , a panel of experts , consisting of four members , was assembled to investigate the content validity of the survey . Two parameters were measured in the content validation study: ( 1 ) the necessity of each survey item , and ( 2 ) the clarity of each survey statement . The panel was asked to comment independently on the necessity and clarity of the items in order to calculate the content validity ratio ( CVR ) and the content validity index ( CVI ) , respectively . The necessity of the items was assessed by using a three-point rating scale: ( 1 ) essential; ( 2 ) useful , but not essential; and ( 3 ) not necessary . The clarity of the items was also assessed by using a 3-point rating scale: ( 1 ) clear , ( 2 ) item needs revision , and ( 3 ) not clear . Following the experts' assessments , a CVR for the total scale was computed . The CVR in this study for the total scale was 0 . 61 , indicating a satisfactory result . For the necessity parameter , all CVRs were 1 . 0 , thus leading to a CVI of 1 . 0 . The overall CVI for the clarity of survey statements was 0 . 87 . A satisfactory level of agreement was found ( CVI>0 . 80 ) among panellists , suggesting that the scale had good content validity [13] . Several questions were rephrased , as suggested by the panellists , to improve clarity . The modified questionnaire was executed in the form of a preliminary study , which included a random sample of 50 people to investigate the possible problems of the questionnaire and its reliability . The questionnaire inquired about socio-demographic characteristics , house and surrounding environment , beliefs regarding dengue fever , self-reported preventive practices against dengue fever and knowledge of dengue fever . Belief questions were based on the HBM constructs [7]: Self-reported practices for the prevention and control of dengue fever , namely ( 1 ) prevention of mosquito breeding and ( 2 ) prevention of mosquito bites , were assessed by using 6-item and 7-item questions , respectively . The options for practices ( not at all , rarely , sometimes , often , and not applicable ) were assigned penalty points of 4 , 3 , 2 , 1 and 0 , respectively . The possible scores ranged from 0 to 36 for mosquito breeding preventive practices , and 0 to 28 for mosquito bite preventive practices . A higher number of penalty points indicates fewer preventive practices . As the dengue prevention scales were newly developed , an initial pilot test was performed to ensure test-retest reliability , the result of which was found to be acceptable ( correlation coefficient >0 . 70 ) . Cronbach's alpha measurements were also performed . Cronbach's alpha coefficients for prevention of mosquito breeding and mosquito bites were 0 . 791 and 0 . 898 , respectively , demonstrating good internal consistency . The scale for the measurement of knowledge of dengue fever consisted of 43 items . For each statement , the respondents could choose between three response categories: yes , no and don't know . For the analyses , the responses were scored as 1 for a correct response and 0 for an incorrect response or a ‘don't know’ response . Possible scores ranged from 0 to 43 . The higher scores indicate greater knowledge about dengue fever . Cronbach's alpha was 0 . 916 , showing high internal consistency . The study was approved by the Medical Ethics Committee of the University of Malaya Medical Centre , Kuala Lumpur , Malaysia ( MEC Ref No . 896 . 15 ) . Due care was taken to ensure that all those who agreed to participate in the study did so voluntarily . Respondents were assured that their responses would remain confidential and anonymous , and that they were free to withdraw from the interview at any time . As written informed consent is not practical in a telephone survey , verbal informed consent was obtained from the respondents before the beginning of an interview . The verbal consent procedure was approved by the Medical Ethics Committee . The 26 schools throughout Malaysia were sampled randomly from six zones ( North , East , West , Centre and South of Peninsular Malaysia , as well as Sabah in East Malaysia ) . Therefore , conventional statistical analyses with underlying distributional assumptions were inappropriate for variance estimation and statistical testing because of the multistage probability sampling design . Sampling weights were incorporated into the analyses to produce representative estimates . Each observation corresponding to the IgG of the respective school was proportionally weighted according to the overall school samples in the respective zone before analysis to account for the complex sampling design . The dependent variable ( percentage of positive IgG ) corresponding to each respondent was compared with the independent variable ( socio-demographic characteristics , HBM constructs , dengue prevention practices and knowledge ) by using t-test analysis , analysis of variance or chi-square analysis to see how the variables were associated independently of level of seroprevalence . In multivariate analysis , multiple linear regression standard errors are computed by using a sandwich estimator . A generalized linear model was used in which the outcome was the presence of a level of dengue seroprevalence vs . its absence . The covariance matrix was estimated by using the robust estimator method . This sandwich standard error estimator assumes independence of the clusters . Correlations of individual level data were conducted with Spearman's rank correlation coefficient to examine the association between self-reported practices to prevent mosquito breeding and mosquito bites and ( 1 ) HBM constructs and ( 2 ) knowledge score . Effect sizes of 0 . 20 or lower were considered to demonstrate ‘small’ relationships , while those ranging from 0 . 30 to 0 . 50 were considered to demonstrate ‘medium’ relationships and those of 0 . 80 or greater ‘large’ relationships [14] . All statistical analyses were performed with SPSS 16 . 0 ( SPSS Inc . , Chicago , IL ) . In all analyses , a P-value of less than 0 . 05 was considered statistically significant .
There were no significant differences in the percentage of positive IgG results in the community by ethnicity , gender , educational level or household size ( Table 1 ) . Dengue-specific IgG was present in 83 . 0% of the group earning a household income of more than MYR4000 monthly ( USD1313 ) and in 70 . 4% of the group earning a household income of less than MYR2000 monthly ( USD657; P<0 . 001 ) . A significantly higher proportion of respondents from the community presenting dengue IgG seropositivity were found to be residing in high-rise residential apartment or condominium buildings than in single ( 67 . 9% ) or terraced ( 61 . 1% ) houses . There were no clear ascending or descending trends between the proportion of respondents with dengue IgG present and the surrounding vegetation density levels reported , but participants reporting no vegetation in the surrounding environment were found to have the highest proportion of dengue IgG present ( 82 . 1% ) compared with those reporting a high vegetation density in the surrounding environment ( 77 . 1% ) . The proportion of respondents with dengue IgG present was significantly higher among those from high-rise apartment or condominium buildings who reported moderate surrounding vegetation density ( 66 . 0% ) than it was for those who reported no vegetation density ( 15 . 0% ) . The proportion of respondents with dengue IgG present was also significantly higher in rural than in urban areas ( 76 . 5% versus 55 . 4% ) . A higher proportion of rural respondents ( 44 . 7% ) reported a high level of vegetation density in their surrounding environment than did their urban counterparts ( 15 . 8% ) . The mean ( SD ) rating of perceived severity of dengue fever on a scale from 1 to 10 was 7 . 45 ( SD = 2 . 90 ) for respondents from communities where positive dengue IgG results were present and 6 . 88 ( 3 . 58 ) for respondents from communities where positive dengue IgG results were absent . Among respondents from communities where dengue IgG was present , the majority ( 81 . 7% ) rated the severity of dengue fever as 7 and 8 . The mean ( and SD ) rating of the perceived susceptibility score was 5 . 40 ( SD = 2 . 32 ) for respondents from communities where dengue IgG was present and 5 . 18 ( SD = 2 . 52 ) for respondents from communities where dengue IgG was absent . The mean ( SD ) rating for the perceived barriers to sustain dengue prevention was also higher for respondents from communities where dengue IgG was present ( 4 . 48 , SD = 2 . 39 ) than for respondents from communities where positive dengue IgG was absent ( 3 . 79 , SD = 2 . 70 ) . A higher proportion of respondents from communities where positive dengue IgG results were present reported that they either strongly agreed or agreed ( 78 . 3% ) that they lacked efficacy in taking preventive measures , as compared with respondents from communities where positive dengue IgG results were absent ( 21 . 7% ) . Mosquito problems in the neighbourhood were reported as severe by 84 . 9% respondents from communities where dengue IgG was present compared with only 15 . 1% among respondents from communities where dengue IgG was absent . A higher proportion of respondents from communities where dengue IgG seropositivity was present ( 79 . 4% ) reported that they were aware that dengue was in their community , as compared with respondents from communities where dengue IgG was absent ( 20 . 6% ) . Respondents from communities where dengue IgG was present reported a higher frequency of mosquito fogging in their community than did those from communities where dengue IgG was absent . A higher proportion of respondents from communities where positive dengue IgG results were present reported that they either strongly agreed or agreed ( 75 . 6% ) that there was a lack of preventive measures at the community level , as compared with respondents from communities where positive IgG results were absent ( 24 . 4% ) . Similarly , a higher proportion of respondents from communities where IgG was present reported that they either strongly agreed or agreed ( 74 . 1% ) that there was a lack of preventive measures from the authorities , as compared with respondents from communities where IgG was absent ( 25 . 9% ) . Thirty-seven per cent of respondents ( 350 of 944 ) did not periodically examine mosquito breeding places in their surroundings , 20% ( 167 of 849 ) did not cover their water storage containers , and 9% ( 86 out of 959 ) did not periodically change water stored at home . The total responses do not add up to 1 , 400 because some respondents did not practise storing water at home . A higher proportion of urban respondents ( 53 . 9% ) did not practise storing water compared with their rural counterparts ( 34 . 4% ) . The mean ( SD ) penalty points for self-reported practices to prevent mosquito breeding was slightly higher among respondents from communities where positive dengue IgG results were present ( 13 . 8 , SD = 6 . 7 ) than among respondents from communities where positive IgG results were absent ( 13 . 2 , SD = 5 . 9 ) . Using cluster analysis , we identified three distinct percentages of positive IgG seropositivity clusters . Table 2 shows the characteristic differences of respondents in three subgroups by percentage level of IgG ( % ) identified by cluster analysis . Cluster I , consisting of nine schools , has the lowest mean percentage of IgG ( 2 . 37 , SD±2 . 39 ) , whereas cluster III ( total of six schools ) has the highest mean percentage of IgG ( 25 . 2 , SD±3 . 02 ) . Cluster III had a higher proportion of high-rise house types ( 15 . 6% ) and village houses ( 61 . 7% ) . Across clusters ( Cluster I to Cluster III ) , an increasing proportion of respondents reported “a lot” of vegetation density . Likewise , an increasing proportion of respondents across clusters agreed that there was a lack of preventive measures at the community level and a lack of self-practice to prevent mosquito bites , as well as fogging . In contrast , we found a decreasing proportion of respondents with a high knowledge score from Cluster I to Cluster III . The Spearman correlation between scores for self-prevention practices and the constructs of the HBM , and between scores for self-preservation practices and knowledge scores , are shown in Table 3 . The effect size , Spearman's rho , indicates that the correlations between self-prevention practices ( both prevention against mosquito breeding and against mosquito bites ) and the constructs of the HBM can be considered a small effect , according to Cohen's criteria on effect size . Significant small effect sizes were also found in the correlations between knowledge scores and preventive practices . Findings from multivariate logistic regression analysis ( Table 1 ) indicate that the income group of MYR2001–4000 monthly ( USD757-1313 ) was less likely than the income group above MYR4000 monthly ( odds ratio [OR] = 0 . 34 , 95% CI , 0 . 22–0 . 86 , P<0 . 01 ) to be seropositive for dengue IgG . Likewise , the income group below MYR2000 monthly was also less likely than the group above MYR4000 monthly ( USD1313 ) to be seropositive for dengue IgG ( OR = 0 . 49 , 95% CI , 0 . 34–0 . 99 , P<0 . 05 ) . Compared with those who reported a high level of vegetation density , those who reported moderate vegetation density ( OR = 0 . 62 , 95% CI , 0 . 40–0 . 98 , P<0 . 05 ) had a lower likelihood of being seropositive for dengue IgG . Those living in an urban area ( OR = 0 . 40 , 95% CI , 0 . 30–0 . 74 , P<0 . 001 ) had a lower likelihood of being seropositive for dengue IgG compared with those living in the reference rural area . The results of multivariate logistic regression analysis also indicated that the two main constructs of the HBM ( perceived severity and susceptibility ) were significant correlates of seropositivity for dengue IgG . Those with a lower perceived severity ( level of severity 7–8 ) had a higher likelihood ( OR = 1 . 84 , 95% CI , 1 . 25–2 . 87 , P<0 . 001 ) of being seropositive for dengue IgG compared with those with a reference level of severity of 9–10 . Likewise , those with a lower perceived susceptibility ( level 3–4 ) had a higher likelihood ( OR = 4 . 50 , 95% CI , 1 . 95–10 . 99 , P<0 . 01 ) of being seropositive for dengue IgG compared with those with a reference level of 9–10 . Having perceived barriers to sustain prevention ( on a scale of 1–10 ) was not found to be significantly associated with being seropositive for dengue IgG in the logistic regression model . Respondents who rated the mosquito problem as 'moderate' had a lower likelihood of being dengue seropositive ( OR = 0 . 45 , 95% CI , 0 . 20–0 . 97 , P<0 . 05 ) than did those who rated it as 'severe' . Respondents who knew dengue was in their community had an increased likelihood of being dengue seropositive ( OR = 1 . 45 , 95% CI , 1 . 28–2 . 53 , P<0 . 01 ) . There was a significant relationship between the gradual increase in the likelihood of being seropositive for dengue IgG and the frequency of fogging . Self-reported practices to prevent mosquito breeding was not a significant correlate of being seropositive for dengue IgG; however , having lower mean penalty points in the prevention of mosquito bites was associated with a lower likelihood of being seropositive for dengue IgG ( OR = 0 . 54 , 95% CI , 0 . 49–0 . 98 , P<0 . 001 ) . Lack of preventive measures from authorities was not a significant correlate of being seropositive for dengue IgG; however , lack of preventive measures at the community level was significantly associated with IgG seropositivity ( OR = 3 . 1 95% CI , 1 . 58–3 . 85 , P<0 . 001 ) . A low mean knowledge score ( 0–14 ) was associated with a lower likelihood of seropositivity for dengue IgG ( OR = 0 . 30 , 95% CI , 0 . 25–0 . 65 , P<0 . 001 ) .
Our multivariate analyses revealed three major correlates of IgG seropositivity that should be the prime focus in dengue prevention and control: ( 1 ) high-rise residential apartment or condominium building house type; ( 2 ) the main construct of the HBM , perceived severity and susceptibility; and ( 3 ) the additional constructs of the HBM , lack of preventive measures at the community level and knowing a neighbour has dengue as a cue to action . These findings also suggest that constructs of the HBM can be integrated as ways of motivating the adoption of preventive practices against dengue , and they may work best by complementing other advocacy and mobilization approaches . Another significant outcome of this study is that it sheds light on the importance of the need for approaches that ensure the translation of knowledge into practice . Most important , the findings have profound implications for future studies that should seek more accurate and confirmatory evidence by investigating the socio-demographic and HBM constructs and their association with individual levels of IgG seropositivity .
|
Demographic , economic and behavioural factors are important to successful control of dengue . This study aimed to examine these factors and their association with the prevalence rate of dengue in a community marked by IgG seropositivity of students attending schools there , as assessed in our previous study . Telephone interviewing was used to try to reach the households in a 3 km radius of the schools where serological tests were conducted . Results showed three major correlates of the presence of IgG seropositivity in the community: ( 1 ) high-rise residential apartment house type or condominium buildings; ( 2 ) the main construct of the HBM , perceived severity and susceptibility; and ( 3 ) the additional constructs of the HBM , lack of preventive measures at the community level and having a neighbour with dengue as a cue to action . The residential environment factor and the constructs of the HBM are useful and important elements in developing interventions to prevent and control dengue .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"biology",
"and",
"life",
"sciences",
"behavior",
"psychology",
"social",
"sciences"
] |
2014
|
Community Knowledge, Health Beliefs, Practices and Experiences Related to Dengue Fever and Its Association with IgG Seropositivity
|
Constitutive overexpression of the MDR1 ( multidrug resistance ) gene , which encodes a multidrug efflux pump of the major facilitator superfamily , is a frequent cause of resistance to fluconazole and other toxic compounds in clinical Candida albicans strains , but the mechanism of MDR1 upregulation has not been resolved . By genome-wide gene expression analysis we have identified a zinc cluster transcription factor , designated as MRR1 ( multidrug resistance regulator ) , that was coordinately upregulated with MDR1 in drug-resistant , clinical C . albicans isolates . Inactivation of MRR1 in two such drug-resistant isolates abolished both MDR1 expression and multidrug resistance . Sequence analysis of the MRR1 alleles of two matched drug-sensitive and drug-resistant C . albicans isolate pairs showed that the resistant isolates had become homozygous for MRR1 alleles that contained single nucleotide substitutions , resulting in a P683S exchange in one isolate and a G997V substitution in the other isolate . Introduction of these mutated alleles into a drug-susceptible C . albicans strain resulted in constitutive MDR1 overexpression and multidrug resistance . By comparing the transcriptional profiles of drug-resistant C . albicans isolates and mrr1Δ mutants derived from them and of C . albicans strains carrying wild-type and mutated MRR1 alleles , we defined the target genes that are controlled by Mrr1p . Many of the Mrr1p target genes encode oxidoreductases , whose upregulation in fluconazole-resistant isolates may help to prevent cell damage resulting from the generation of toxic molecules in the presence of fluconazole and thereby contribute to drug resistance . The identification of MRR1 as the central regulator of the MDR1 efflux pump and the elucidation of the mutations that have occurred in fluconazole-resistant , clinical C . albicans isolates and result in constitutive activity of this trancription factor provide detailed insights into the molecular basis of multidrug resistance in this important human fungal pathogen .
The yeast Candida albicans is usually a harmless commensal in many healthy people where it resides on mucosal surfaces of the gastrointestinal and urogenital tract , but it can also cause superficial as well as life-threatening systemic infections , especially in immunocompromised patients [1] . Infections by C . albicans are commonly treated with the antimycotic agent fluconazole that inhibits the biosynthesis of ergosterol , the major sterol in the fungal cell membrane . However , C . albicans can develop resistance to fluconazole , especially during long-term treatment of oropharyngeal candidiasis , which frequently affects HIV-infected persons and AIDS patients [2] . Molecular fingerprinting of serial C . albicans isolates from recurrent episodes of oropharyngeal candidiasis has shown that fluconazole resistance usually develops in previously susceptible strains , and such serial isolates from the same patient , so-called matched isolates , have proved an excellent tool to study the molecular basis of drug resistance [3–10] . Fluconazole resistance can be caused by different mechanisms , including alterations in the sterol biosynthetic pathway , increased expression of the ERG11 gene that encodes the target enzyme of fluconazole , sterol 14α-demethylase ( Erg11p ) , mutations in the ERG11 gene that result in reduced affinity of Erg11p to fluconazole , and overexpression of genes encoding membrane transport proteins , which transport fluconazole out of the cell . In clinical C . albicans strains , several of these mechanisms are often combined to result in a stepwise development of clinically relevant fluconazole resistance ( for a review , see [11] ) . A major mechanism of drug resistance in C . albicans is the constitutive upregulation of genes encoding efflux pumps that actively transport fluconazole and many other , structurally unrelated toxic compounds out of the cell . Two types of efflux pumps have been identified in C . albicans [12–15] . The CDR1 and CDR2 genes encode ATP-binding cassette ( ABC ) transporters , whereas MDR1 encodes a multidrug efflux pump of the major facilitator superfamily . In drug-susceptible C . albicans strains , MDR1 is expressed at low or non-detectable levels in standard laboratory media , but its expression can be induced when the cells are grown in the presence of certain toxic compounds , like benomyl , hydrogen peroxide , or diamide [16–20] . In contrast , many fluconazole-resistant , clinical C . albicans isolates constitutively overexpress MDR1 [3 , 4 , 7–10] . Inactivation of MDR1 in such MDR1 overexpressing C . albicans isolates increased their susceptibility to fluconazole , confirming that MDR1 overexpression contributed to fluconazole resistance [21] . The increased resistance of such isolates to other metabolic inhibitors , like cerulenin , brefeldin A , and diamide was completely abolished after MDR1 deletion , indicating that the resistance of the strains to these toxic compounds was mainly or exclusively mediated by Mdr1p [22 , 23] . Comparison of the MDR1 promoter sequences in matched pairs of fluconazole-susceptible and MDR1 overexpressing , fluconazole-resistant C . albicans isolates from the same patient demonstrated that the constitutive MDR1 upregulation in the resistant isolates was not caused by promoter mutations but by alterations in trans-regulatory factor ( s ) [24] . Several groups have identified sequences in the MDR1 promoter region that mediate its upregulation in drug-sensitive strains in response to inducing chemicals and its constitutive activation in drug-resistant strains [19 , 20 , 25 , 26] . However , in contrast to the ABC transporters CDR1 and CDR2 , whose expression has recently been shown to be controlled by the transcription factor Tac1p , which is mutated in CDR1/CDR2 overexpressing C . albicans strains [27 , 28] , the regulatory factors controlling MDR1 expression and the mutations that are responsible for its constitutive overexpression in drug-resistant clinical isolates have not yet been identified . In this study , we compared the alterations in gene expression occurring in three different MDR1 overexpressing , clinical C . albicans isolates on a genome-wide scale to identify genes that are commonly upregulated with MDR1 . This approach led to the identification of a central regulator of MDR1 expression and to the elucidation of the molecular basis of MDR1 overexpression and multidrug resistance in clinical C . albicans isolates .
To identify genes that are coordinately upregulated with MDR1 , we compared the transcriptional profiles of three matched pairs of fluconazole-susceptible and fluconazole-resistant clinical C . albicans isolates . F1 and G1 are the first , drug-susceptible isolates of two well-characterized series of clinical C . albicans isolates and do not detectably express MDR1 , while F5 and G5 are the last isolates in each series that overexpress MDR1 and have become multidrug-resistant [8 , 21 , 23 , 24] . An additional isolate pair from another patient , isolates 5833 ( no MDR1 expression ) and 6692 ( MDR1 overexpression ) , was obtained from Martine Raymond and has been described recently [29] . As can be seen in Figure 1 , a common set of 21 genes was consistently upregulated in all three MDR1 overexpressing isolates , including all the genes encoding proteins that had previously been identified as upregulated in isolates F5 and G5 by proteome analysis [23] . In addition , seven genes were downregulated in all resistant isolates . The complete data set for all differentially expressed genes in the pairwise comparisons can be found in Table S1 . Interestingly , one gene , orf19 . 7372 ( IPF1266 ) , which was moderately upregulated in the MDR1 overexpressing strains , encodes a predicted zinc cluster transcription factor that has been given the preliminary name ZCF36 in the Candida Genome Database ( http://www . candidagenome . org/ ) . As transcription factors often regulate their own expression in addition to that of their target genes , we hypothesized that this transcription factor might be involved in MDR1 expression and thereby control multidrug resistance . The results shown below demonstrate that this was indeed the case and we have therefore named orf19 . 7372 as MRR1 , for multidrug resistance regulator . To investigate if MRR1 affects drug resistance in C . albicans , we deleted the gene in the C . albicans model strain SC5314 as well as in the drug-resistant clinical isolates F5 and G5 using the SAT1-flipping strategy ( [30] , Figure 2A and 2C , lanes 1–3 , and unpublished data ) . From each parental strain , two independent homozygous mrr1Δ mutants were constructed ( see Table S2 ) and tested for their susceptibilities to various metabolic inhibitors to which MDR1 overexpression confers resistance [22 , 23 , 31] . Inactivation of MRR1 in the drug-susceptible strain SC5314 , which does not detectably express MDR1 under standard growth conditions , did not affect its susceptibility to the tested compounds , except for diamide , which is known to induce MDR1 expression ( see also below ) . In contrast , the mrr1Δ mutants of the MDR1 overexpressing isolates F5 and G5 completely lost their resistance to cerulenin , brefeldin A , and diamide and became as susceptible to these inhibitors as mdr1Δ mutants derived from these strains or the matched , drug-susceptible isolates F2 and G2 , which were the last isolates in each series that did not detectably express MDR1 [8] ( Figure 3 ) . The heterozygous mrr1 mutants exhibited intermediate resistance . Resistance of isolates F5 and G5 to these compounds was mediated mostly or exclusively by MDR1 overexpression , since resistance was lost after deletion of MDR1 [22 , 23] . In contrast , MDR1 overexpression contributes only partially to fluconazole resistance of these isolates , which is caused by a combination of different mechanisms [8 , 21] . Interestingly , deletion of MRR1 in isolates F5 and G5 had an even stronger effect than inactivation of MDR1 , suggesting that MRR1 is required for various mechanisms of fluconazole resistance . The increased fluconazole resistance of isolate F5 as compared with isolate F2 was completely lost in the mrr1Δ mutants derived from this strain , while the mrr1Δ mutants of isolate G5 , which also contains a mutation in the ERG11 gene that results in reduced affinity of fluconazole for its target enzyme , were still more resistant than the matched isolate G2 , which does not contain this mutation [8] . The results presented above suggested that MRR1 affects drug resistance by controlling expression of the MDR1 efflux pump . Therefore , we compared MDR1 promoter activity in the mrr1Δ mutants and their wild-type parental strains . For this purpose , a PMDR1-GFP ( green fluorescent protein ) reporter fusion from plasmid pMPG2S ( see Materials and Methods ) was integrated into the genome of these strains , and GFP expression was quantified by flow cytometry . As can be seen in Figure 4A , the strong MDR1 promoter activity in isolates F5 and G5 was completely abolished after deletion of MRR1 , demonstrating that MRR1 is required for the constitutive activation of the MDR1 promoter in these clinical isolates . The heterozygous MRR1/mrr1Δ mutants exhibited reduced MDR1 promoter activity , showing that both MRR1 alleles contributed to MDR1 overexpression in the drug-resistant isolates . The loss of MDR1 expression in the mrr1Δ mutants was also independently confirmed by quantitative real-time reverse transcription ( RT ) -PCR ( see Figure S1A ) . As mentioned above , strain SC5314 does not detectably express MDR1 , and no significant MDR1 promoter activity was seen in the corresponding reporter strains ( Figure 4B ) . However , MDR1 expression can be induced in drug-susceptible strains when the cells are exposed to certain chemicals [16–19] . We used two such compounds , benomyl and H2O2 , to investigate whether MRR1 is also required for inducible MDR1 expression . These chemicals were chosen because it has recently been reported that different regions in the MDR1 promoter mediate its upregulation by benomyl and H2O2 [20] . While the deletion of one MRR1 allele had no effect on MDR1 promoter activity under these conditions , activation of the MDR1 promoter by either of the two compounds was almost completely abolished in the homozygous mrr1Δ mutants ( Figure 4B ) . Therefore , MRR1 is required for both the constitutive MDR1 overexpression in drug-resistant , clinical C . albicans isolates and the inducible MDR1 expression in a drug-susceptible strain . Since MRR1 was upregulated in the MDR1 overexpressing clinical C . albicans isolates , we investigated whether artificial overexpression of MRR1 would result in activation of the MDR1 promoter . For this purpose , we placed the MRR1 coding region from strain SC5314 under the control of the strong ADH1 promoter in plasmid pZCF36E1 ( see Materials and Methods ) . The PADH1-MRR1 fusion was integrated into strain CAG48B , a derivative of the fluconazole-sensitive laboratory strain CAI4 , which expresses the GFP reporter gene from the endogenous MDR1 promoter ( see Table S2 ) . The resulting strain did not detectably express the GFP gene , similar to a control strain that carried an otherwise identical construct without the MRR1 gene ( unpublished data ) , suggesting that overexpression of MRR1 was not sufficient to activate the MDR1 promoter and that the drug-resistant isolates F5 and G5 might carry gain-of-function mutations in MRR1 . Therefore , we cloned and sequenced the MRR1 alleles of isolates F2 , F5 , G2 , and G5 ( see Table S3 ) . Isolate F2 contained two polymorphic MRR1 alleles . The coding region of allele 1 ( MRR1F2–1 ) was identical to the MRR1 sequence of strain SC5314 , while allele 2 ( MRR1F2–2 ) differed from it at 26 positions , with three of the polymorphisms resulting in amino acid exchanges . Only one allele ( MRR1F5 ) was obtained from the matched resistant isolate F5 and this allele corresponded to allele 1 of isolate F2 except for a single C-T mutation at position 2047 , which resulted in a proline-serine substitution at position 683 of Mrr1p . Direct sequencing of the PCR products confirmed the loss of heterozygosity in isolate F5 and the absence of the mutation in the MRR1 alleles of isolate F2 . A similar situation was found for the isolate pair G2/G5 . Isolate G2 contained two polymorphic MRR1 alleles ( MRR1G2–1 and MRR1G2–2 ) , which differed from one another at 28 positions , with two of the polymorphisms resulting in amino acid differences . G5 contained only one MRR1 allele ( MRR1G5 ) that was identical with allele 2 of isolate G2 except for a single G-T mutation at position 2990 , which resulted in a glycine–valine substitution at position 997 of Mrr1p . Therefore , both resistant isolates F5 and G5 had become homozygous for a mutated MRR1 allele , suggesting that these mutations might have caused the constitutive MDR1 overexpression and the resulting multidrug resistance . To directly test whether the P683S and G997V mutations in Mrr1p are responsible for MDR1 overexpression and drug resistance , we introduced the mutated MRR1 alleles from isolates F5 and G5 as well as the corresponding wild-type alleles from isolates F2 and G2 into the mrr1Δ mutants of strain SC5314 . All four MRR1 alleles were integrated into one of the inactivated mrr1Δ alleles to ensure expression from the endogenous MRR1 promoter ( see Figure 2B and 2C , lanes 4–7 ) . In each case , two independent correct transformants were used for further analysis . Figure 5A shows that , as noted above , deletion of MRR1 in strain SC5314 did not affect its susceptibility to cerulenin , brefeldin A , and fluconazole , but the mutants displayed increased sensitivity to diamide , a compound that induces MDR1 expression and is an Mdr1p substrate . Insertion of the MRR1–1 allele from isolate F2 ( which is identical to MRR1 of strain SC5314 , so this also represented a reinsertion of the original allele ) or the MRR1–2 allele from isolate G2 complemented the hypersusceptibility of the mutants to diamide , but did not increase resistance to cerulenin , brefeldin A , and fluconazole . In contrast , insertion of the mutated alleles from isolates F5 and G5 resulted in enhanced resistance to all compounds . To obtain direct evidence that the mutated MRR1 alleles confer drug resistance by activating the MDR1 promoter and , thus , mediate overexpression of the Mdr1p efflux pump , we integrated the various MRR1 alleles into a derivative of the laboratory strain CAI4 that expresses the GFP reporter gene from the endogenous MDR1 promoter . For this purpose , we first inactivated the MRR1 wild-type alleles in the reporter strain CAG48B , as described above for the other strains , and then reinserted one of the four different MRR1 alleles . Two independent transformants were kept for each MRR1 allele . Figure 5B shows that , while the MRR1F2–1 and MRR1G2–2 alleles from the susceptible isolates had no effect , expression of the corresponding mutated MRR1F5 and MRR1G5 alleles resulted in strong MDR1 promoter activity . MDR1 upregulation by the mutated MRR1 alleles was also independently confirmed by comparing MDR1 mRNA levels in strains carrying wild-type or mutated MRR1 alleles by quantitative real-time RT-PCR ( see Figure S1B ) . These results demonstrated that the P683S and G997V mutations in Mrr1p caused constitutive MDR1 overexpression and multidrug resistance . A mutation in the transcription factor TAC1 has recently been shown to cause constitutive upregulation of the ABC-transporters CDR1 and CDR2 as well as drug resistance in certain C . albicans strains , but only after the strains had become homozygous for the mutated allele [28] . Since the resistant isolates F5 and G5 also had become homozygous for mutated MRR1 alleles , we tested whether these alleles could mediate MDR1 overexpression and drug resistance in the presence of a non-mutated , wild-type allele . Therefore , to produce strains that contained both a wild-type and a mutated allele , the MRR1 alleles from isolates F5 and G5 were inserted into the inactivated mrr1 allele in the heterozygous MRR1/mrr1Δ mutants derived from strain SC5314 ( see Figure 2B and 2C , lanes 8 and 9 ) . The mutated alleles conferred drug resistance also in the presence of a wild-type MRR1 allele , although we observed a slightly reduced resistance as compared with the strains containing only a mutated MRR1 allele , especially for the MRR1F5 allele ( see Figure 5A ) . To directly compare MDR1 promoter activity in strains carrying only a mutated MRR1 allele or both a mutated and a wild-type MRR1 allele , the mutated MRR1 alleles were also integrated into the inactivated mrr1 allele of the heterozygous MRR1/mrr1Δ mutant with the PMDR1-GFP reporter fusion . Figure 5B shows that both mutated MRR1 alleles were able to activate the MDR1 promoter in the presence of a wild-type MRR1 allele , but the degree of activation was lower than in strains containing only a mutated allele . Again , this effect was more pronounced for the MRR1F5 allele . Taken together , these results demonstrate that the MRR1F5 and MRR1G5 alleles can act in a semi-dominant fashion and mediate MDR1 overexpression and multidrug resistance in the presence of a nonmutated MRR1 allele , but the presence of a wild-type MRR1 allele reduces the activity of the MRR1 alleles containing gain-of-function mutations . That deletion of MRR1 from the drug-resistant C . albicans isolates F5 and G5 affected fluconazole resistance more strongly than deletion of MDR1 suggests that Mrr1p controls the expression of additional genes that contribute to the increased fluconazole resistance of these isolates . Therefore , to identify the set of genes controlled by the transcription factor Mrr1p , we compared the gene expression profiles of isogenic strains expressing either a wild-type MRR1 allele ( MRR1F2–1 or MRR1G2–2 ) or its constitutively active , mutated counterpart ( MRR1F5 and MRR1G5 , respectively ) . In addition , we compared the gene expression profiles of the clinical isolates F5 and G5 , which carry the gain-of-function MRR1 alleles , with those of their mrr1Δ derivatives . As shown in Figure 6A and 6B , 20 and 27 genes were consistently upregulated in the transformants expressing the MRR1F5 and MRR1G5 alleles , respectively , and 19 of the 28 total genes were commonly upregulated by both gain-of-function alleles . As expected , CDR1 and CDR2 were not among the Mrr1p target genes . Strikingly , the 11 most strongly upregulated genes were the same in strains expressing the MRR1F5 or the MRR1G5 allele . All of these ( plus three additional commonly upregulated genes ) were also significantly overexpressed in the drug-resistant clinical isolates F5 and G5 as compared with the matched susceptible isolates F1 and G1 , respectively , and downregulated again after deletion of MRR1 from the resistant isolates F5 and G5 . These 14 genes , therefore , represent a core set of genes whose expression is controlled by Mrr1p , and some of them might contribute to fluconazole resistance in the clinical isolates ( see Discussion ) . Twenty-eight genes ( including MRR1 ) were consistently downregulated after deletion of MRR1 in the resistant isolates F5 and G5 ( see Table 1 ) A complete list of all genes that were found to be differentially regulated in the mrr1Δ mutants as compared with their wild-type parental strains is provided in Table S4 . Fourteen of the 28 genes were also upregulated by the MRR1F5 and MRR1G5 alleles in the SC5314 background and in the resistant isolates F5 and G5 as compared with the matched susceptible isolates F1 and G1 , respectively . The 14 genes were mainly those most strongly affected by the MRR1 deletion . The other genes were either not consistently upregulated above the threshold level in the SC5314 transformants expressing only one mutated MRR1 allele ( e . g . , MRR1 itself ) or possibly had a strain-specific dependence on MRR1 . Altogether , the results of the transcriptional profiling experiments provided a comprehensive list of genes that are regulated , directly or indirectly , by Mrr1p in various C . albicans strain backgrounds . The identification of these genes , in turn , provided clues about how gain-of-function mutations in MRR1 contribute to fluconazole resistance of clinical C . albicans strains , besides causing overexpression of the MDR1 efflux pump ( see Discussion ) .
Since the initial report by Sanglard et al . more than ten years ago [3] , many studies have shown that the major mechanism of resistance to the widely used antifungal agent fluconazole in clinical C . albicans isolates is the constitutive overexpression of efflux pumps , which actively transport this drug and other toxic substances out of the cell , thereby conferring multidrug resistance . It is well established that mutations in trans-regulatory factors are responsible for the upregulation of genes encoding efflux pumps in drug-resistant C . albicans isolates [24 , 32]; however , until recently the identity of these regulators has remained elusive . Two main factors , namely 1 ) the discovery of a cis-regulatory element in the promoters of CDR1 and CDR2 with features that are typical of binding sites of zinc cluster transcription factors , and 2 ) the observation that homozygosity at the mating type locus was linked with the development of azole resistance in certain clinical strains , led Coste et al . to exploit the available C . albicans genome sequence information to systematically search for candidate transcription factors that might regulate the expression of these efflux pumps . This strategy resulted in the identification of the zinc cluster transcription factor TAC1 , which is located near the mating type locus , as the major regulator of CDR1 and CDR2 [27] . In contrast , no obvious criteria for a similar in silico search for a regulator of MDR1 , the other efflux pump that mediates drug resistance in many clinical C . albicans isolates , were evident . Promoter deletion analyses performed by several research groups have identified three different activating regions in the MDR1 promoter , two of which contain binding sites for the transcription factors Cap1p and Mcm1p [19 , 20 , 25 , 26] . However , each of these regions could be individually deleted from the full-length MDR1 promoter without abrogating its constitutive activation in MDR1 overexpressing C . albicans isolates , which suggested that a transcription factor other than Cap1p or Mcm1p causes the upregulation of MDR1 in drug-resistant strains . Assuming that a common mechanism might be responsible for the constitutive MDR1 upregulation in such strains , we compared the alterations in gene expression that occurred in three different MDR1 overexpressing , drug-resistant C . albicans isolates . This approach led to the identification of MRR1 , a zinc cluster transcription factor that was moderately upregulated in all three resistant isolates as compared with matched , drug-susceptible isolates , suggesting that this transcription factor might contribute to MDR1 overexpression and/or drug resistance . A role of MRR1 in fluconazole resistance would not have been easily inferred from genetic analysis of commonly used C . albicans laboratory strains , because deletion of MRR1 from the model strain SC5314 did not result in hypersusceptibility of the mutants . However , the development of methods to inactivate genes in clinical C . albicans strains [21 , 30] allowed us to demonstrate the essential role of MRR1 in drug resistance of two different MDR1 overexpressing C . albicans isolates . Similar findings were previously obtained with the efflux pump MDR1 itself , whose disruption in a C . albicans laboratory strain did not result in fluconazole hypersusceptibility , whereas MDR1 inactivation in MDR1 overexpressing strains reduced or abolished their resistance to fluconazole and other metabolic inhibitors [14 , 21 , 33] . We found that MRR1 is not only responsible for the constitutive overexpression of MDR1 in drug-resistant isolates , but also mediates the inducible MDR1 expression in a drug-susceptible strain . The transcription factors Cap1p and Mcm1p have been implicated in the induction of MDR1 expression by H2O2 and benomyl , respectively [20] . Interestingly , deletion of MRR1 almost completely abolished the induction of the MDR1 promoter in response to both of these stimuli , indicating that Mrr1p has a more central and essential role in the control of MDR1 expression and , depending on the environmental conditions , may cooperate in different ways with these other transcription factors to regulate expression of the efflux pump . The identification of MRR1 as the central regulator of MDR1 expression also enabled us to elucidate the genetic alterations that had occurred in drug-resistant C . albicans isolates and were responsible for MDR1 overexpression and multidrug resistance . In two clinical isolates we found different single nucleotide substitutions in MRR1 that resulted in amino acid exchanges in Mrr1p . The ability of the mutated MRR1 alleles to activate the MDR1 promoter and confer drug resistance when expressed in a C . albicans laboratory strain confirmed that these were indeed gain-of-function mutations that resulted in constitutive activation of the transcription factor . The two mutations , which were found in different regions of Mrr1p , most likely relieve the transcription factor from repression by an autoinhibitory domain or by another negatively acting factor that keeps Mrr1p in its inactive state in the absence of inducing signals . In both cases the MDR1 overexpressing , drug-resistant isolate had become homozygous for the mutated MRR1 allele , which suggested that Mrr1p containing a gain-of-function mutation would not be able to activate the MDR1 promoter in the presence of wild-type Mrr1p . However , we found that both mutated MRR1 alleles were able to induce MDR1 expression and cause drug resistance in the presence of a wild-type allele , albeit at slightly reduced levels as compared with strains containing only the mutated allele . This indicates that the gain-of-function alleles acted in a semi-dominant fashion . The comparison of heterozygous and homozygous mrr1Δ mutants with their drug-resistant parental strains showed that the presence of two rather than only one mutated MRR1 allele resulted in higher MDR1 promoter activity and drug resistance . Therefore , the increased activity of the transcription factor in strains carrying two copies of a mutated MRR1 allele instead of only one , coupled with a slight negative effect of wild-type Mrr1p on the activity of the activated form , appears to provide sufficient advantage during antimycotic therapy to select for the loss of heterozygosity once a gain-of-function mutation has occurred in one of the two MRR1 alleles . Such a loss of heterozygosity readily occurs in C . albicans either by loss of one chromosome and duplication of the homologous chromosome or by mitotic recombination between the two homologous chromosomes , as has been well documented in several recent studies [28 , 34 , 35] . Interestingly , deletion of the transcription factor MRR1 from MDR1 overexpressing C . albicans isolates reduced fluconazole resistance of these strains even more than deletion of MDR1 itself , suggesting that the gain-of-function mutations in MRR1 contribute to fluconazole resistance of these strains by other mechanisms , in addition to causing overexpression of the efflux pump . We , therefore , aimed to find out which of the many alterations in gene expression seen in the drug-resistant C . albicans clinical isolates ( see Figure 1 and Table S1 ) were caused by the MRR1 mutations . When introduced into strain SC5314 , both mutated MRR1 alleles caused the upregulation of 19 genes including MDR1 ( note that some genes of the IFD family may in fact be alleles of the same gene , e . g . , those that were originally designated as IFD1 and IFD5 in CandidaDB but have now been assigned the same orf19 name ) . Some additional genes were reproducibly upregulated by only one of the two mutated alleles ( one for MRR1F5 and eight for MRR1G5 ) , which could be explained by differential effects of the P683S and G997V mutations on Mrr1p activity at the respective target promoters . The expression of 14 of the 19 genes that were upregulated by both mutated MRR1 alleles in the SC5314 background was found to be downregulated after deletion of MRR1 in both drug-resistant clinical isolates F5 and G5 . A considerable number of additional genes were affected by inactivation of MRR1 in the clinical isolates ( see Table S4 ) , but only 28 genes were downregulated in the mrr1Δ mutants of both parental strains , indicating that the other effects of MRR1 deletion depended on the strain background . It is striking that of the core set of Mrr1p target genes ( i . e . , the 14 genes that were upregulated by the mutated MRR1 alleles in the SC5314 background as well as upregulated in the drug-resistant isolates F5 and G5 and downregulated in the mrr1Δ mutants of these strains ) many encode putative oxidoreductases ( see Table 1 ) . IPF7817 is a member of the NAD ( P ) H oxidoreductase family that is strongly induced during oxidative stress [36 , 37] . It is supposed to be involved in the regulation of intracellular redox homeostasis , as mutants in which the gene was inactivated had increased intracellular levels of reactive oxygen species ( ROS ) and , presumably as a compensatory mechanism , upregulated other redox-related genes [38] . IPF17186 is a member of the ThiJ/PfpI protein family [39] . Its ortholog in S . cerevisiae , also named HSP31 , is induced by the transcription factor Yap1p in response to oxidative stress . Also , an hsp31Δ mutant is hypersensitive to a subset of ROS generators , suggesting that Hsp31p may protect the cell against oxidative stress [40] . GRP2 is a homolog of S . cerevisiae GRE2 , which encodes a stress-induced NADH-dependent methylglyoxal reductase that is regulated by the pleiotropic drug resistance regulator Pdr1p . A gre2Δ mutant exhibited a growth defect under conditions of membrane stress and activated ERG genes as a compensatory mechanism , and it displayed an increased sensitivity to ergosterol biosynthesis inhibitors [41] . IFD1 , IFD4 , IFD5 , IFD6 , and IFD7 encode proteins of the aldo-keto reductase family and are homologs of the putative aryl-alcohol dehydrogenase YPL088w of S . cerevisiae . Interestingly , YPL088w is regulated by the transcriptional regulators Yrr1p and Yrm1p , which are involved in the control of multidrug resistance [42] . Similarly , IPF5987 encodes a member of the aldo-keto reductase family , and its ortholog in S . cerevisiae is also transcriptionally regulated by Yrr1p and Yrm1p [42] . MDR1 and other Mrr1p target genes , almost all of which have also been found to be upregulated in another MDR1 overexpressing clinical C . albicans isolate [18] , are induced in the presence of chemicals that exert oxidative stress upon the cells , like hydrogen peroxide or diamide . They are also induced by the microtubule destabilizing agent benomyl [16 , 18–20] . Interestingly , benomyl treatment causes lipid peroxidation and glutathione depletion in rats and these effects were blocked by treatment with antioxidants , suggesting that the in vivo toxicity of benomyl may be associated with oxidative stress to cellular membranes [43] . Therefore , it seems possible that the activation of Mrr1p by all these compounds may be a response to oxidative damage of the cells , and the function of many of the target genes that are induced by activated Mrr1p could be to restore the intracellular redox balance . This may also explain how the upregulation of these genes contributes to fluconazole resistance of clinical C . albicans strains containing MRR1 gain-of-function mutations . In addition to inhibiting ergosterol biosynthesis , azoles have been shown to increase the level of endogenous reactive oxygen species in C . albicans cells , and the decrease in cell viability associated with miconazole treatment was significantly prevented by addition of an antioxidant [44] . These observations suggest that ROS plays a role in the mechanism of action of azole antifungal agents and that ROS detoxification mechanisms may contribute to azole resistance . In contrast to other compounds that cause oxidative stress , like hydrogen peroxide or diamide , fluconazole does not induce MDR1 expression [8 , 16] . Therefore , the constitutive upregulation of MDR1 and other Mrr1p target genes in strains containing MRR1 gain-of-function mutations provides protection of the cells against this antifungal agent . A major regulator of the oxidative stress response in C . albicans is the bZIP transcription factor Cap1p [36 , 45 , 46] . Like MDR1 , several other Mrr1p target genes ( IPF7817 , IPF17186 , and GRP2 ) contain a YRE element , the putative binding site for Cap1p , in their promoters . Their expression is induced by hydrogen peroxide in a Cap1p-dependent manner [18 , 20 , 36 , 45] . Therefore , Cap1p and Mrr1p may act together to regulate expression of these genes in response to oxidative stress . Cap1p also controls the expression of other genes that are involved in the oxidative stress response , like the thioredoxin reductase TRR1 , the glutathione reductase GLR1 , the glutathione S-transferase GTT1 , and the superoxide dismutase SOD2 [36 , 45] , and which were not found among the Mrr1p target genes . Unlike CAP1 inactivation [45] , MRR1 deletion or MRR1 gain-of-function mutations had no effect on the susceptibility of C . albicans to H2O2 ( unpublished data ) , which in contrast to diamide is not a substrate of the Mdr1p efflux pump [23] . Therefore , the Cap1p target genes that are not controlled by Mrr1p and that are typical oxidative stress-response genes seem to be more important for the resistance of C . albicans to hydrogen peroxide , while the Mrr1p target genes contribute to fluconazole resistance , presumably because the two compounds cause different types of damage within the cells . The precise function of most of the Mrr1p target genes is currently unknown and their potential involvement in an oxidative stress response remains speculative . Alternatively , it is possible that fluconazole treatment causes the accumulation of other toxic molecules that are eliminated by the combined action of the oxidoreductases and other gene products whose expression is regulated by Mrr1p . The identification of MRR1 as the major regulator of MDR1 expression and the elucidation of the mutations in clinical isolates that cause constitutive activity of this transcription factor represent a major step forward in our understanding of multidrug resistance development in C . albicans . Important questions that can now be addressed include how Mrr1p is normally activated in response to inducing signals , how gain-of-function mutations cause constitutive activation of the transcription factor , and how Mrr1p interacts with other putative MDR1 regulators like Cap1p and Mcm1p to control expression of its target genes .
C . albicans strains used in this study are listed in the supporting Table S2 . All strains were stored as frozen stocks with 15% glycerol at −80 °C . The strains were routinely grown in YPD medium ( 10 g yeast extract , 20 g peptone , 20 g glucose per liter ) at 30 °C . To prepare solid media , 1 . 5% agar was added before autoclaving . For induction of the MDR1 promoter with benomyl or H2O2 , overnight cultures of reporter strains were diluted 10−2 in three flasks with fresh YPD medium and grown for 3 h . Fifty μg/ml of benomyl or 0 . 005% H2O2 was then added to one of the cultures and the cells were grown for an additional hour . The fluorescence of the cells was quantified by FACS analysis . The coding region of the MRR1 gene of C . albicans strain SC5314 was amplified by PCR with the primers ZCF36–1 and ZCF36–2 ( primer sequences are given in Table S5 ) . The PCR product was digested at the introduced SalI and BglII restriction sites and substituted for the OPT4 open reading frame ( ORF ) in the XhoI/BglII-digested pOPT4E1 [47] to generate pZCF36E1 . The sequence of the cloned MRR1 gene was identical to that found in the genome sequence of C . albicans strain SC5314 ( orf19 . 7372 ) . An MRR1 deletion construct was generated in the following way: A SacI-SacII fragment containing MRR1 upstream sequences from positions −314 to +15 with respect to the start codon was amplified with the primers ZCF36–3 and ZCF36–4 , and an XhoI-ApaI fragment containing MRR1 downstream sequences from positions +3260 to +3557 was amplified with the primers ZCF36–5 and ZCF36–6 . The MRR1 upstream and downstream fragments were cloned on both sides of the SAT1 flipper cassette in plasmid pSFS1 [30] to result in pZCF36M2 in which the MRR1 coding region from positions +16 to +3259 ( 66 bp before the stop codon ) is replaced by the SAT1 flipper ( see Figure 2A ) . DNA fragments containing the N-terminal part and upstream sequences of the MRR1 alleles of the clinical C . albicans isolates F2 , F5 , G2 , and G5 were amplified with the primers ZCF36–3 and ZCF36seq6 , digested at the introduced SacI site and at an internal EcoRI site , and cloned in pBluescript to generate pZCF36NF2A and D , pZCF36NF5A , pZCF36NG2B and D , and pZCF36G5A , respectively . DNA fragments containing the MRR1 C-terminal part and downstream sequences were amplified with the primers ZCF36–1 and ZCF36–6 , digested at the internal EcoRI site and at the introduced SacI site , and cloned in pBluescript to obtain pZCF36CF2B and C , pZCF36CF5C , pZCF36CG2A and D , and pZCF36CG5A , respectively . The complete ORFs and upstream sequences of the MRR1–1 alleles of isolates F2 and G2 were also amplified with the primers ZCF36–3 and ZCF36–8 and cloned in pBluescript to yield pZCF36TF2–1 and pZCF36TG2–1 , respectively . To express wild-type and mutated MRR1 alleles in mrr1Δ mutants , an EcoRI-SalI fragment from pZCF36E1 containing the C-terminal part of MRR1 , the ACT1 transcription termination sequence , and part of the caSAT1 selection marker was cloned into the EcoRI/SalI-digested pZCF36NF2A to obtain pZCF36K1 . An MRR1 downstream fragment was then amplified from SC5314 genomic DNA with the primers ZCF36–7 and ZCF36–6 , digested at the introduced NsiI and ApaI sites , and cloned together with a BglII-PstI fragment from pZCF36E1 containing the ACT1 transcription termination sequence and the caSAT1 selection marker into the BglII/ApaI-digested pZCF36K1 to generate pZCF36K2 , which contains the MRR1–1 allele of isolate F2 ( which is identical to the MRR1 gene of strain SC5314 ) . Plasmid pZCF36K3 , which contains the mutated MRR1 allele from isolate F5 , was obtained by substituting an EcoRI-PstI fragment from pZCF36CF5C for the corresponding fragment in pZCF36K2 . For expression of the MRR1–2 allele of isolate G2 , an NdeI-EcoRI fragment from pZCF36NG2B and an EcoRI-PstI fragment from pZCF36CG2D were substituted for the corresponding region in pZCF36K2 , resulting in pZCF36K4 . Replacement of the EcoRI-PstI fragment in this plasmid by the corresponding region from pZCF36CG5A generated pZCF36K5 , which contains the mutated MRR1 allele from isolate G5 . Plasmid pMPG2S , which contains a PMDR1-GFP reporter fusion , was constructed by substituting the caSAT1 selection marker from pSAT1 [30] for the URA3 marker in the previously described plasmid pMPG2 [25] . C . albicans strains were transformed by electroporation [48] with gel-purified inserts from the plasmids described above . Nourseothricin-resistant transformants were selected on YPD agar plates containing 200 μg/ml nourseothricin ( Werner Bioagents ) as described previously [30] . The correct genomic integration of all constructs was confirmed by Southern hybridization . Genomic DNA from C . albicans was isolated as described previously [49] . 10 μg of DNA was digested with appropriate restriction enzymes , separated on a 1% agarose gel and , after ethidium bromide staining , transferred by vacuum blotting onto a nylon membrane and fixed by UV cross-linking . Southern hybridization with enhanced chemiluminescence-labeled probes was performed with the Amersham ECLTM Direct Nucleic Acid Labeling and Detection System ( GE Healthcare ) according to the instructions of the manufacturer . Stock solutions of the drugs were prepared as follows . Fluconazole ( 1 mg/ml ) and diamide ( 20 mg/ml ) were dissolved in water , while cerulenin ( 5 mg/ml ) and brefeldin A ( 5 mg/ml ) were dissolved in DMSO . In the assays , serial 2-fold dilutions in the assay medium were prepared from the following initial concentrations: cerulenin , 200 μg/ml; brefeldin A , 200 μg/ml; diamide , 800 μg/ml; fluconazole , 200 μg/ml . Susceptibility tests were carried out in high resolution medium ( 14 . 67 g HR-Medium [Oxoid GmbH] , 1 g NaHCO3 , 0 . 2 M phosphate buffer [pH 7 . 2] ) , using a previously described microdilution method [50] . Readings were done after 48 h . Fluorescence-activated cell sorter ( FACS ) analysis was performed with a FACSCalibur cytometry system equipped with an argon laser emitting at 488 nm ( Becton Dickinson ) . Fluorescence was measured on the FL1 fluorescence channel equipped with a 530-nm band-pass filter . Twenty thousand cells were analyzed per sample and were counted at low flow rate . Fluorescence and forward scatter data were collected by using logarithmic amplifiers . The mean fluorescence values were determined with CellQuest Pro ( Becton Dickinson ) software . The nucleotide sequences corresponding to 6 , 165 ORFs for C . albicans were downloaded from the Galar Fungail European Consortium ( Assembly 6 , http://www . pasteur . fr/Galar_Fungail/CandidaDB ) . Following the Affymetrix Design Guide , we designed two separate probe sets for each ORF , each consisting of 13 perfect match and 13 mismatch overlapping 25 bp oligonucleotides , to the 3′ 600 bp region . For ORFs less than 600 bp in length , the sequence was divided in two equal segments for subsequent design procedures . For quality control and normalization purposes , we made 2–3 additional probe sets spanning the entire sequence of the C . albicans 18S rRNA ( GenBank Accession M60302 ) , genes encoding GAPDH , actin and Mdr1p ( Bmr1p ) in addition to the standard Affymetrix controls ( BioB , C , D , cre , DAP , PHE , LYS , THR ) . The probe selection was performed by the Chip Design group at Affymetrix , Inc . using their proprietary algorithm to calculate probe set scores , which includes a probe quality metric , cross-hybridization penalty , and gap penalty . The probe sets were then examined for cross-hybridization against all other sequences in the C . albicans genome as well as a number of constitutively expressed genes and rRNA from other common organisms . Consequently , for some target regions we were not able to design high quality probe sets . In the end , the GeneChip contained 10 , 736 probe sets including 9 controls , 6 , 123 unique ORFs , and duplicate probe sets for 4 , 604 ORFs . The duplicate probe sets are made to distinct regions of the ORF , thereby allowing 2 independent measurements of the mRNA level for that particular gene . The C . albicans custom Affymetrix NimbleExpress Arrays ( CAN04a530004N ) were manufactured by NimbleGen Systems [51] per our specification . Total RNA was isolated using the hot SDS-phenol method [52] . Frozen cells were suspended in 12 ml of 50 mM sodium acetate ( pH 5 . 2 ) , 10 mM EDTA at room temperature , after which 1 ml of 20% sodium dodecyl sulphate and 12 ml of acid phenol ( Fisher Scientific ) were added . This mixture was incubated 10 min at 65 °C with mixing each minute , cooled on ice for 5 min , and centrifuged for 15 min at 12 , 000g . Supernatants were transferred to new tubes containing 15 ml of chloroform , mixed , and centrifuged at 200g for 10 min . The aqueous layer was removed to new tubes , RNA was precipitated with 1 vol isopropanol and 0 . 1 vol 2 M sodium acetate ( pH 5 . 0 ) , and then collected by centrifugation at 17 , 000g for 35 min at 4 °C . The RNA pellet was suspended in 10 ml of 70% ethanol , collected again by centrifugation , and suspended in nuclease-free water . Immediately prior to cDNA synthesis , the purity and concentration of RNA samples were determined from A260/A280 readings , and RNA integrity was determined by capillary electrophoresis using the RNA 6000 Nano Laboratory-on-a-Chip kit and Bioanalyzer 2100 ( Agilent Technologies ) as per the manufacturer's instructions . First and second strand cDNA was synthesized from 15 μg total RNA using the SuperScript Double-Stranded cDNA Synthesis Kit ( Invitrogen ) and oligo-dT24-T7 primer ( PrOligo ) according to the manufacturer's instructions . cRNA was synthesized and labeled with biotinylated UTP and CTP by in vitro transcription using the T7 promoter-coupled double-stranded cDNA as template and the Bioarray HighYield RNA Transcript Labeling Kit ( ENZO Diagnostics ) . Double-stranded cDNA synthesized from the previous steps was washed twice with 70% ethanol and suspended in 22 μl of RNase-free water . The cDNA was incubated as recommended with reaction buffer , biotin-labeled ribonucleotides , dithtiothreitol , RNase inhibitor mix , and T7 RNA polymerase for 5 h at 37 °C . The labeled cRNA was separated from unincorporated ribonucleotides by passing through a CHROMA SPIN-100 column ( Clontech ) and ethanol precipitated at −20 °C overnight . The cRNA pellet was suspended in 10 μl of RNase-free water and 10 μg was fragmented by ion-mediated hydrolysis at 95 °C for 35 min in 200 mM Tris-acetate ( pH 8 . 1 ) , 500 mM potassium acetate , 150 mM magnesium acetate . The fragmented cRNA was hybridized for 16 h at 45 °C to the C . albicans NimbleExpress GeneChip arrays . Arrays were washed at 25 °C with 6 × SSPE , 0 . 01% Tween 20 followed by a stringent wash at 50 °C with 100 mM MES , 0 . 1 M NaCl , 0 . 01% Tween 20 . Hybridizations and washes employed the Affymetrix Fluidics Station 450 using their standard EukGE-WS2v5 protocol . The arrays were then stained with phycoerythrein-conjugated streptavidin ( Molecular Probes ) and the fluorescence intensities were determined using the GCS 3000 high-resolution confocal laser scanner ( Affymetrix ) . The scanned images were analysed using software resident in GeneChip Operating System v2 . 0 ( Affymetrix ) . Sample loading and variations in staining were standardized by scaling the average of the fluorescent intensities of all genes on an array to a constant target intensity ( 250 ) . The signal intensity for each gene was calculated as the average intensity difference , represented by [Σ ( PM − MM ) / ( number of probe pairs ) ] , where PM and MM denote perfect-match and mismatch probes . The scaled gene expression values from GeneChip Operating System v2 . 0 software were imported into GeneSpring 7 . 2 software ( Agilent Technologies ) for preprocessing and data analysis . Probe sets were deleted from subsequent analysis if they were called absent by the Affymetrix criterion and displayed an absolute value below 20 in all experiments . The expression value of each gene was normalized to the median expression of all genes in each chip as well as the median expression for that gene across all chips in the study . Pairwise comparison of gene expression was performed for each matched experiment . Among direct comparisons between matched clinical isolates , genes were considered to be differentially expressed if their change in expression was ≥ 1 . 5-fold in three independent experiments . An aliquot of the RNA preparations from the samples used in the microarray experiments was saved for quantitative real-time RT-PCR follow-up studies . First-strand cDNAs were synthesized from 2 μg of total RNA in a 21-μl reaction volume using the SuperScript First-Strand Synthesis System for RT-PCR ( Invitrogen ) in accordance with the manufacturer's instructions . Quantitative real-time PCRs were performed in triplicate using the 7000 Sequence Detection System ( Applied Biosystems ) . Independent PCRs were performed using the same cDNA for both the gene of interest and the 18S rRNA , using the SYBR Green PCR Master Mix ( Applied Biosystems ) . Gene-specific primers were designed for the gene of interest and the 18S rRNA using Primer Express software ( Applied Biosystems ) and the Oligo Analysis & Plotting Tool ( QIAGEN ) and are shown in Table S5 . The PCR conditions consisted of AmpliTaq Gold activation at 95 °C for 10 min , followed by 40 cycles of denaturation at 95 °C for 15 s and annealing/extension at 60 °C for 1 min . A dissociation curve was generated at the end of each PCR cycle to verify that a single product was amplified using software provided with the 7000 Sequence Detection System . The change in fluorescence of SYBR Green I dye in every cycle was monitored by the system software , and the threshold cycle ( CT ) above the background for each reaction was calculated . The CT value of 18S rRNA was subtracted from that of the gene of interest to obtain a ΔCT value . The ΔCT value of an arbitrary calibrator ( e . g . , untreated sample ) was subtracted from the ΔCT value of each sample to obtain a ΔΔCT value . The gene expression level relative to the calibrator was expressed as 2−ΔΔCT .
The coding sequences of the MRR1 alleles described in this study have been deposited in GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/index . html ) with the following accession numbers: EU139261 ( MRR1F2–1 ) , EU139262 ( MRR1F2–2 ) , EU139263 ( MRR1F5 ) , EU139264 ( MRR1G2–1 ) , EU139265 ( MRR1G2–2 ) , EU139266 ( MRR1G5 ) .
|
The Candida albicans MDR1 ( multidrug resistance ) gene encodes a multidrug efflux pump of the major facilitator superfamily that is constitutively overexpressed in many fluconazole-resistant strains . Although MDR1 overexpression is a major cause of resistance to this widely used antifungal agent and other metabolic inhibitors , so far the molecular basis of MDR1 upregulation in resistant strains has remained elusive . By comparing the transcription profiles of MDR1 overexpressing , clinical C . albicans isolates and matched , drug-susceptible isolates from the same patients , we identified a transcription factor , termed multidrug resistance regulator 1 ( MRR1 ) , which was upregulated in all resistant isolates and turned out to be a central regulator of MDR1 expression . Resistant isolates contained point mutations in MRR1 , which rendered the transcription factor constitutively active . Introduction of these mutated alleles into a susceptible strain caused MDR1 overexpression und multidrug resistance . Inactivation of MRR1 in clinical isolates abolished MDR1 expression and affected fluconazole resistance even more strongly than deletion of the MDR1 efflux pump itself , indicating that additional Mrr1p target genes , which were identified by genome-wide gene expression analysis , contribute to fluconazole resistance . These findings provide detailed insights into the molecular basis of multidrug resistance in one of the most important human fungal pathogens .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"infectious",
"diseases",
"yeast",
"and",
"fungi",
"microbiology",
"eukaryotes",
"genetics",
"and",
"genomics",
"saccharomyces"
] |
2007
|
The Transcription Factor Mrr1p Controls Expression of the MDR1 Efflux Pump and Mediates Multidrug Resistance in Candida albicans
|
The dynamics and regulation of HIV-1 nuclear import and its intranuclear movements after import have not been studied . To elucidate these essential HIV-1 post-entry events , we labeled viral complexes with two fluorescently tagged virion-incorporated proteins ( APOBEC3F or integrase ) , and analyzed the HIV-1 dynamics of nuclear envelope ( NE ) docking , nuclear import , and intranuclear movements in living cells . We observed that HIV-1 complexes exhibit unusually long NE residence times ( 1 . 5±1 . 6 hrs ) compared to most cellular cargos , which are imported into the nuclei within milliseconds . Furthermore , nuclear import requires HIV-1 capsid ( CA ) and nuclear pore protein Nup358 , and results in significant loss of CA , indicating that one of the viral core uncoating steps occurs during nuclear import . Our results showed that the CA-Cyclophilin A interaction regulates the dynamics of nuclear import by delaying the time of NE docking as well as transport through the nuclear pore , but blocking reverse transcription has no effect on the kinetics of nuclear import . We also visualized the translocation of viral complexes docked at the NE into the nucleus and analyzed their nuclear movements and determined that viral complexes exhibited a brief fast phase ( <9 min ) , followed by a long slow phase lasting several hours . A comparison of the movement of viral complexes to those of proviral transcription sites supports the hypothesis that HIV-1 complexes quickly tether to chromatin at or near their sites of integration in both wild-type cells and cells in which LEDGF/p75 was deleted using CRISPR/cas9 , indicating that the tethering interactions do not require LEDGF/p75 . These studies provide novel insights into the dynamics of viral complex-NE association , regulation of nuclear import , viral core uncoating , and intranuclear movements that precede integration site selection .
HIV-1 enters and travels through the cytoplasm of an infected cell , reverse transcribes its genomic RNA into double-stranded DNA and forms a preintegration complex ( PIC ) , crosses through a nuclear pore , and integrates its DNA into the host genome ( reviewed in [1] ) . Although movement of fluorescently labeled HIV-1 complexes in infected living cells has been described [2–4] , the dynamics with which individual HIV-1 complexes encounter nuclear pores and are imported into the nucleus , as well as the molecular events that regulate these dynamics , are not well understood , largely because these events have not been extensively studied in living cells . HIV-1 viral cores are large ( 61-nm width , 120-nm length ) compared to the 40-nm size limit for translocation through a nuclear pore complex ( NPC ) [5 , 6]; thus , it is generally assumed that viral complexes must be disassembled before nuclear import can take place ( reviewed in [7] ) . If viral complexes are disassembled in the cytoplasm and are converted to a form/size that is competent for nuclear import , then one might expect that the viral complexes exhibit a very short residence time at the nuclear envelope ( NE ) . This would be a scenario that is similar to adeno-associated virus 2 complexes , which are 25-nm diameter in size , and other large cellular cargos , which dock at the NE and are transported through the nuclear pore within milliseconds [8–10] . On the other hand , if the viral complexes undergo capsid disassembly at the NE then they would be expected to reside at the NE for a long time prior to import , during which uncoating occurs to generate a viral complex that can be translocated through a nuclear pore . A third possibility is that uncoating can occur either in the cytoplasm or at the NE; in this scenario , one would expect that the length of time a viral complex is in the cytoplasm would be inversely correlated with their residence time at the NE . Thus , live-cell imaging analysis of the length of time viral complexes reside in the cytoplasm and at the NE can provide valuable insight into not only the process of nuclear import , but also the process of viral core uncoating , and facilitate the identification of viral and host factors that regulate these events . Currently , the translocation of viral complexes into the nucleus and their nuclear movements after import have not been observed . Understanding these processes can provide insights into the dynamics and molecular interactions of viral complexes with chromatin or other macromolecules that precede integration site selection and provirus formation . Although currently there is a debate as to whether HIV-1 integrates into genes near the periphery of the nucleus [11 , 12] or in genes throughout the nucleus [13] , the kinetics of association of viral complexes with chromatin prior to integration , and viral and host factors that may be essential for tethering viral complexes to chromatin , have not been investigated . Analysis of the intranuclear movement of viral complexes can provide insights into the extent and the kinetics with which viral complexes can freely diffuse in the nucleus , whether they form multiple transient contacts with the chromatin before selecting the site of integration , and whether they stably associate with specific sites on chromatin before integration . The intranuclear location of viral complexes can also provide information regarding the nuclear location of integration sites relative to the nuclear point of entry . Lastly , we previously observed that only 1 in 50 viral complexes in the cytoplasm are imported into the nucleus , which is similar to the number of proviruses that integrated and expressed a reporter gene [14]; thus , visualizing the nuclear import of viral complexes can help to identify and study the complexes that are likely to complete viral replication away from the vast majority of the cytoplasmic viral complexes that do not complete viral replication . APOBEC3F ( A3F ) and APOBEC3G ( A3G ) are host restriction factors that are incorporated into virions in virus producing cells and potently inhibit virus replication [15–19] . We [20 , 21] and others [22] recently demonstrated that A3F and A3G inhibited viral integration and that A3F has a higher affinity for double-stranded DNA than A3G , suggesting that A3F and A3G might stay associated with the viral complex in the nucleus . Subsequently , we produced virions labeled with yellow fluorescent protein ( YFP ) -tagged A3F ( A3F-YFP ) or A3G ( A3G-YFP ) , infected cells , and then analyzed their association with viral complexes . We found that A3F-YFP , and to a lesser extent A3G-YFP , remained stably associated with viral complexes and could be used to visualize viral complexes in the nuclei of infected cells [14] . We also found that A3F-YFP colocalized with viral nucleic acid in the nuclei of infected cells , indicating that A3F-YFP indeed remains stably associated with nuclear viral complexes . Some recent studies have shown that a Vpr-integrase-green fluorescent protein fusion protein ( Vpr-IN-GFP ) could be used to label viral complexes [23 , 24]; Vpr-IN-GFP is incorporated into virions , and proteolytic processing between Vpr and IN-GFP by viral protease during virion maturation results in the release of IN-GFP , which can be used to detect nuclear HIV-1 complexes . Cyclophilin A ( CypA ) is a host peptidylprolyl isomerase that binds to an exposed loop on the surface of CA and is incorporated into virions [25–27] . The CA-CypA interaction can be disrupted by treatment with cyclosporine A ( CsA ) , which binds to CypA , or by introducing mutations into the CA loop that is involved in binding to CypA ( reviewed in [7] ) . Disruption of the CA-CypA interaction results in a reduction in viral titers in primary CD4+ T cells , monocyte derived macrophages , and in some T cell lines , but has minimal effects on the viral infectivity in other cell lines such as HeLa cells [28–31] . It is also known that disruption of the CA-CypA interaction influences nuclear import and the requirement for some nuclear pore proteins [32–35] . However , the influence of CypA on the timing and regulation of HIV-1 nuclear import has not been studied . The detailed dynamics of HIV-1 association with the NE and their import into the nucleus in living cells have not been previously analyzed . Arhel and colleagues [2] , and more recently Dharan and colleagues [36] , observed the docking of a few HIV-1 complexes with the NE in living cells , but they did not determine the functional relevance or the kinetics of these associations . We and others have observed association of viral complexes with the NE in fixed cells; however , these studies cannot provide insights into the kinetics and stability of NE association [3 , 14 , 37–42] . Here , we analyzed living cells infected with HIV-1 particles labeled with either A3F-YFP or Vpr-integrase-YFP fusion protein ( hereafter referred to as IN-YFP ) , and characterized the dynamics with which HIV-1 complexes stably associate with the NE , translocate from the cytoplasm into the nucleus , move after nuclear entry , and the viral and host factors that regulate these events . Our studies indicate that long residence times at the NE involving CA and host nuclear pore protein Nup358 are required for nuclear import and that the timing of nuclear import is regulated by CA-CypA interactions , but not reverse transcription . Additionally , the nuclear import of viral complexes is correlated with substantial loss of CA , indicating that at least a portion of the viral core uncoating occurs during nuclear import . After nuclear entry , the viral complexes exhibited a brief fast phase during which the viral complexes may move away from the NE , followed by a long slow phase during which the viral complexes move at a rate similar to integrated proviruses , supporting the hypothesis that the viral complexes are tethered to chromatin and/or other large macromolecules and remain at or near their future site of integration .
To visualize viral complexes in infected cells , we labeled HIV-1 virions with A3F-YFP that were pseudotyped with vesicular stomatitis virus G protein ( VSV-G ) . VSV-G pseudotyped virions enter cells through endocytosis , and fusion of the viral and endosomal membranes leads to the release of the viral core into the cytoplasm ( reviewed in [43] ) . We determined that the infectivity of the virions containing A3F-YFP was modestly reduced by threefold compared to control virions without any label ( Fig 1A ) . To ensure that we visualized the movement of viral complexes after fusion and not those trapped in endosomes , we determined the efficiency and kinetics of fusion . For these experiments , the virus particles were labeled with A3F-YFP , which is incorporated into the virion core , as well as S15-mCherry; S15-mCherry is a protein that is non-specifically incorporated into membranes , and serves as a marker that can be used to distinguish post-fusion viral complexes ( S15-mCherry– ) from those that remain in endosomes ( S15-mCherry+ ) ( Fig 1B ) [44] . We determined the proportion of all A3F-YFP labeled virions that remained 20 , 40 , and 60 minutes after infection relative to the number of virions at the time of infection ( 0-min time point , which was set to 100%; Fig 1C ) . There was progressive loss of A3F-YFP labeled virions , and ∼44% of the A3F-YFP labeled virions remained 60-min after infection . A3F-YFP labeled virions may be degraded after endocytosis or the YFP signal may be lost due to diffusion of A3F-YFP away from some viral complexes , such as immature virions with incompletely closed viral cores . Approximately 50% of the A3F-YFP labeled virions were also labeled with S15-mCherry at the time of infection and , at the 60-min time point , the proportion of dual-labeled virions was reduced to ∼9% of the total A3F-YFP labeled particles at the time of infection . The efficiency of fusion was determined by dividing the number of dual-labeled particles by the total A3F-YFP labeled particles at each time point , and setting the ratio at the 0-min time point to 100% ( Fig 1D ) . The percentage of dual-labeled particles remaining relative to the 0-min time point was reduced to ∼20% at the 40- and 60-min time points , providing a fusion efficiency of ∼80% , which is consistent with previous reports [44 , 45] . Based on the observed timing of fusion events , we initiated most live-cell microscopy experiments ∼45 min after infection so that we were primarily visualizing post-fusion viral complexes; because these experiments were initiated after most viruses had fused their membranes with endosomal membranes , virions for these experiments were made without S15-mCherry . To verify that A3F-YFP signals in infected cells were viral complexes , we fixed infected cells at 0 , 1 , and 3 hrs post-infection ( hpi ) and detected HIV-1 capsid ( CA ) by immunofluorescence staining and confocal microscopy ( Fig 1E ) . Approximately 80% of the A3F-YFP signals in the cytoplasm at the 0 hpi were colocalized with detectable levels of CA , verifying that most of the YFP signals in the cytoplasm were viral complexes ( Fig 1F ) . At 1 and 3 hpi , ∼60% of the viral complexes in the cytoplasm and at the NE were colocalized with CA , which was significantly different from the 0-hr time point , but not significantly different from each other . In contrast , only ∼35% of the YFP signals in the nuclei at the 3-hr time point were associated with CA ( Fig 1F ) . The proportions of CA+ viral complexes at the 1-hr time point were not determined because very few nuclear viral complexes were detected at this early time point . We also found that the average relative intensity of the CA signals at the 1-hr time point in the cytoplasm and at the NE were ∼50% of the average CA signal intensities at the 0-hr time point ( Fig 1G ) . Since only ∼1500–2500 of the ∼5000 CA molecules that are packaged into virions form the viral core [5 , 46] , we expect that ∼50 to 70% of the CA protein in virions are monomers that will diffuse away from the viral complexes upon fusion of the viral and host membranes . In addition , the loss of CA signal at the 1-hr time point could be due to loss of CA from the viral cores due to uncoating . We did not observe any significant loss of average CA signal intensity from viral complexes in the cytoplasm or at the NE between 1-hr and 3-hr time points . However , we observed 2 . 5-fold lower CA signal intensity associated with nuclear complexes compared to NE-associated complexes , indicating that a significant amount of CA is lost upon nuclear entry ( Fig 1G ) . The detailed kinetics of association of individual HIV-1 complexes with the NE and their residence time at the NE have not been examined , although association of a few viral complexes with the NE has been observed [2 , 36] . To address these questions , we generated HeLa cells that stably expressed POM121-mCherry , a nuclear pore marker ( Fig 2A and 2B ) . We infected these cells with VSV-G pseudotyped virions containing A3F-YFP at a multiplicity of infection ( MOI ) of ∼1 . 1 , defined as the number of proviruses per cell that expressed a GFP reporter gene ( S1 Table ) . We previously estimated that in our imaging experiments 1 in 50 A3F-YFP labeled particles that entered each cell was imported into the nucleus [14] . To determine the fate of WT viral complexes that formed semi-stable ( >1 min to < 20 min ) and stable ( >20 min ) associations with the NE , we acquired time-lapse images ( z-stack every 1 min for 2 hrs ) and analyzed the residence time for each HIV-1 complex that associated with the NE for at least 1 min ( 2 consecutive frames; Fig 2C ) . Most WT HIV-1 complexes ( 171/194 = 88% , obtained from analysis of 23 cells ) resided at the NE ≤ 20 min ( avg . residence time = 5 . 0 min ) , whereas few complexes ( 23/194 = 12% ) resided at the NE for >20 min ( avg . residence time ≥ 51 . 4 min ) ( Fig 2C ) . Of these 23 particles , 10 particles also dissociated from the NE and no nuclear import was observed before the end of the observation time ( 120 min ) , suggesting that association with the NE for >20 min does not necessarily result in the nuclear import of the viral complex . These data strongly indicate that forming semi-stable associations with the NE ( >1 min to < 20 min ) is necessary but not sufficient for making stable associations lasting > 20 min . To determine the role of HIV-1 CA in forming stable associations with the NE , we analyzed the association of HIV-1 complexes of a CA mutant ( K203A ) which forms unstable viral cores , and a CA mutant ( E128A/R132A ) that forms hyperstable viral cores [14 , 47] . We previously observed that these mutants were defective in nuclear import and in associating with the NE in a fixed-cell assay [14] . We previously proposed that an interaction between CA and the nuclear pore complex is essential for nuclear import , and that the hypostable mutant is defective in nuclear import because there is little or no CA associated with the viral complexes; in addition , the hyperstable mutant was defective in inducing structural/conformational changes in the viral core that are required to increase access to determinants that associate with the NE . We confirmed that these mutants had much lower levels of infectivity compared to wild-type HIV-1 ( Fig 2D ) . We compared the ability of WT virus and CA mutants K203A and E128A/R132A to form stable associations with the NE ( >20 min ) . We analyzed 19–27 cells for WT and CA mutants and observed that for both CA mutants the proportion of viral complexes that had NE residence times of >20 min was reduced to 0 . 2–0 . 4 particles/cell compared to 1 . 0 particle/cell for wild-type virus ( Fig 2E ) . These observations indicated that the CA mutants that are defective in nuclear import are also defective in forming stable associations with the NE . To identify host factors important for stable association of HIV-1 complexes with the NE , we carried out RNAi-mediated knockdown of Nup358 ( S1A Fig ) , a nuclear pore protein that has been implicated to be directly [34 , 36 , 37 , 48] or indirectly [49] important for nuclear import and HIV-1 infectivity . Depletion of Nup358 significantly reduced infectivity compared to the control siRNA ( Fig 2D; S1A Fig ) , steady-state NE association in a fixed-cell assay ( S1B Fig ) , and nuclear import of HIV-1 complexes ( S1C Fig ) . We determined the effects of Nup358 knockdown on cell viability using the ATPlite luminescence assay ∼48 hrs after transfection of control and Nup358 siRNAs; we did not find detectable levels of cytotoxicity at the time of our live-cell experiments , which were performed ∼48 hrs after siRNA transfection ( S1D Fig ) . Similar to the CA mutations , Nup358 knockdown also reduced the proportion of HIV-1 complexes that stably associated with the NE to ∼0 . 4 particles/cell compared to 1 . 0 particle/cell for the siRNA control ( Fig 2E ) . These results showed that both HIV-1 CA and Nup358 play important roles in the formation of stable associations between HIV-1 complexes and the NE . This result is consistent with the reduction in the steady-state level of HIV-1 complexes that were at the NE in Nup358-depleted cells ( S1B Fig ) , but is in contrast to a recent report that showed a perinuclear accumulation of HIV-1 complexes at the NE in Nup358-depleted cells [36] . We also examined whether endosomal viral complexes can be transported to the NE and the extent to which they colocalized with the NE by analyzing endosomal viral complexes produced in the absence of VSV-G ( Fig 2E ) . We found that the endosomal viral complexes were inefficient at forming stable associations with the NE compared to post-fusion complexes ( ∼0 . 2 vs . 1 . 0/cell ) , indicating that stable association with the NE is largely a property of post-fusion viral complexes . The results also indicated that endosomal viral complexes may be associated with the NE at a low efficiency; we expect that our fusion efficiency is 80% , and 20% of the A3F-YFP labeled complexes may be trapped in endosomes . Since the endosomal complexes associated with the NE for >20 min with a ∼5-fold lower efficiency ( ∼0 . 2 vs . 1 . 0/cell ) , we estimate that only ∼4% of the viral complexes observed at the NE may be unfused viral complexes in endosomes . To gain insights into transient associations of viral complexes with the NE that result in semi-stable associations ( > 1 min ) , and to determine whether CA mutations and Nup358 knockdown influence the transient associations , we captured 1-min movies ( 10 frames/sec ) of a single focal plane near the equatorial region of the nucleus from 45–120 minutes post-infection ( mpi ) , after most virus particles ( ∼80% ) fused and entered the cytoplasm ( S2A Fig ) . A 1-μm wide mask of the NE was created using the POM121-mCherry signal to facilitate automated image analysis ( S2B Fig ) . Within the focal volume ( ∼0 . 8 μm in height ) , particles that moved in the x , y , and z dimension could be followed , provided they generated tracks of sufficient length for analysis ( ≥ 5 steps ) . Overlapping of the first and last frames from the 1-minute movie indicated that the POM121-mCherry signal from the NE did not significantly move during the 1-min movie ( S2C Fig; S1 Movie ) . We employed single-particle tracking combined with a population-based approach to analyze the movement of A3F-YFP labeled viral particles near the NE . A computer simulation to determine the length of time a complex would associate with the NE by random chance showed that very few ( 0 . 9% ) of the randomly moving particles remained in contact with the NE for >5 sec ( S2D Fig ) . We therefore defined contacts with the NE for >5 sec and < 1 min as transient associations . Qualitatively , the movies indicated that few HIV-1 complexes formed transient associations with the NE ( S2E Fig [left and middle panels] , S2 and S3 Movies ) , and most contacts of viral complexes with the NE lasted for <5 seconds ( S2E Fig [right panel] , S4 Movie ) . To quantify transient NE association , we analyzed the association of HIV-1 complexes of wild-type HIV-1 and the CA mutants K203A and E128A/R132A in ∼20 cells per condition , and performed 3–5 independent experiments , constituting analysis of viral complexes from 60–100 cells for each experimental condition . We determined the longest consecutive residence time at the NE for each HIV-1 complex that colocalized with the POM121 mask for at least one frame ( 0 . 1 sec ) during the observation time . We found that most of the 2 , 017 HIV-1 complexes that contacted the NE ( 92% ) made contacts with the NE that remained for <5 sec , and only 8% formed transient associations ( S2F and S2G Fig ) . Both CA mutants were defective for transiently associating with the NE; only 2 . 4 and 3 . 4% of the K203A and E128A/R132A complexes that contacted the NE , respectively , formed transient associations ( S2F and S2G Fig ) . We never observed nuclear import of viral complexes in the movies used to analyze transient associations of viral complexes with the NE . Similar to the CA mutations , Nup358 depletion also reduced the proportion of HIV-1 complexes that transiently associated with the NE compared to the control siRNA ( ∼3% vs . ∼8%; S2F and S2G Fig ) . These results , in addition to the results described in Fig 2 , indicate that both HIV-1 CA and Nup358 play important roles in the formation of both transient and stable associations between HIV-1 complexes and the NE . Although nuclear HIV-1 complexes have been observed by fluorescence microscopy in living cells [2] , the translocation of HIV-1 complexes from the cytoplasm to the nucleus has not been observed . To visualize the entry of HIV-1 complexes into the nucleus and gain insights into the kinetics of nuclear import and nuclear movements , we acquired z-stacks of cells every 3 min starting as early as 10 min after infection for up to 10 hrs . Each z-stack , which covered an area ∼4 μm in height , was centered near the equatorial plane of the nucleus; the top and bottom of the nuclei were not imaged to minimize photobleaching during the long movies . The movement of the nucleus during these long movies was corrected ( S5 Movie ) prior to 3D single-particle tracking of viral complexes . We observed 7 A3F-YFP labeled HIV-1 complexes enter the nucleus in ∼210 hrs of movies of 28 cells ( Fig 3A and 3B and S3A and S3B Fig; Table 1 and S2 Table ) . Viral complexes were stably associated with the NE for ∼20 min ( S6 Movie ) to >3 hours ( S7 Movie ) immediately prior to import , indicating that the viral complexes that were observed to enter the nucleus had long residence times at the NE with wide variation in the length of time at the NE prior to import . In addition to the 7 particles that were observed to translocate from the cytoplasm to the nucleus , we observed an additional 37 A3F-YFP labeled complexes in the nuclei of 28 cells that were analyzed ( ∼1 . 6 HIV-1 complex/nucleus ) . Thus , translocation into the nucleus was visualized for ∼15% of the particles that were imported into the nuclei during the observation time ( 10-min to 10-hrs post-infection ) . After import , the viral complexes remained near the periphery of the nucleus throughout the observation time ( Table 1 and S2 Table; S3C Fig ) , consistent with previous observations by us [14] and others [11 , 12 , 23] . The viral complexes exhibited long and variable residence times at the NE ( Table 1 and S2 Table ) and were imported 1–6 hrs post-infection , a time frame which is consistent with the kinetics of nuclear import [14 , 23] . Because A3F-YFP , a protein which can inhibit viral integration [20 , 21] , could potentially alter the behavior of nuclear HIV-1 complexes , we also sought to observe the nuclear import of HIV-1 complexes using IN-YFP . We observed that viruses labeled with IN-YFP did not exhibit a defect in infectivity ( S4A Fig ) ; however , in our experiments , 8% of the viruses were labeled with IN-YFP , as determined by single virion microscopy analysis , and we cannot exclude the possibility that IN-YFP labeled virions are replication defective [23] . In this regard , previous studies have reported a modest decrease ( ∼2–3 fold ) in infectivity of viruses labeled with Vpr-IN-GFP [24] . Although a lower proportion of virions were labeled with IN-YFP than with A3F-YFP ( 8% vs . 54% , respectively; S4B and S4C Fig ) , the average YFP fluorescence intensities of the labeled particles were similar ( S4D Fig ) , suggesting the A3F-YFP and IN-YFP complexes would be detected with similar efficiencies . Upon infection of cells with equivalent amounts of A3F-YFP- or IN-YFP-labeled virions , we observed a similar number of YFP particles/cell , a similar percentage of cytoplasmic YFP particles that were at the NE , and a similar percentage of YFP particles that were in the nucleus throughout the time-course ( S4E–S4G Fig; P > 0 . 05 , t-test ) . Thus , ∼7-fold higher amount of virus was required to achieve a similar number of YFP-labeled particles in infected cells using IN-YFP compared to A3F-YFP ( [100/8] ÷ [100/54] = 6 . 75 ) , but IN-YFP and A3F-YFP remained associated with HIV-1 complexes in infected cells with a similar efficiency . Next , we visualized the translocation of 14 IN-YFP labeled HIV-1 complexes from the NE to the nucleus in 910 hrs of movies of 91 cells ( Table 1 and S2 Table; Fig 3C and 3D; S8 Movie ) . After import , both A3F-YFP and IN-YFP labeled viral complexes remained near the periphery of the nucleus throughout the observation time ( average and maximum nuclear penetration distances are 1 . 4 ± 0 . 4 μm and 2 . 1 ± 0 . 6 μm from NE , respectively [Table 1 and S2 Table] ) , but moved away from the point of nuclear entry ( average and maximum distances of 2 . 3 ± 0 . 8 and 3 . 2 ± 1 . 1 μm , respectively ) . The A3F-YFP and IN-YFP complexes exhibited long and variable NE residence times prior to nuclear import ( average 1 . 5 ± 1 . 6 hours ) , and were imported an average of 4 . 3 ± 2 . 6 hours after infection . Importantly , no significant differences were observed between A3F-YFP and IN-YFP labeled viral complexes regarding nuclear penetration distance , distance from point of entry , time in cytoplasm prior to NE association , NE residence time , and time of nuclear import ( P > 0 . 05 , t-test or Mann Whitney test; S2 Table ) , indicating that the method of labeling viral complexes did not have any discernible effect on these aspects of viral replication . Immediately following nuclear import , there was a brief fast phase ( <9 min [3 frames] ) in which the viral complexes exhibited higher mobility , followed by a second phase of slower mobility for the remainder of the observation time ( Fig 3E and 3F ) . On average , the YFP intensity was similar just before and after nuclear import , indicating that both A3F-YFP and IN-YFP remained stably associated with the viral complexes through nuclear import ( Fig 3G ) . We analyzed nuclear viral complexes for which the translocation from the NE to the nucleus was not observed . These particles may have entered the nucleus from above or below the z-stack of observed nuclear volume; alternatively , these particles may have entered the nucleus with a very short NE residence time ( < 3 min ) such that their NE docking was not captured in the 3-min/frame movies . We identified 80 such particles in movies discussed in Fig 3 as well as in movies that were manually analyzed to determine NE residence time ( discussed below ) . Particles that were in the nucleus at the start of the movies were excluded from this analysis . We found that 22 of these nuclear viral complexes first appeared in the bottom z-slice , 57 first appeared in the top z-slice , and 1 appeared in the center of the z-stack but could not be confidently tracked because of low signal intensity . Viral complexes that entered the nucleus with very short NE residence times ( < 3 min ) , would be expected to appear in the equatorial plane of the nucleus and it would not be possible to track them to the NE or the top or bottom z-slices . With the exception of one dim particle , 79 of these 80 particles in the nucleus were tracked to the Z-slice above or below the equatorial Z-stack . Thus , we conclude that most ( if not all ) nuclear viral complexes for which the nuclear entry was not observed ( S3A Fig ) entered the nucleus from above or below the z-stack . In addition , we observed 30 viral complexes that exhibited long NE residence times and entered the nucleus ( 21 tracked particles described above and 9 additional particles from experiments discussed below; S2 Table ) . Therefore , <1/110 nuclear viral complexes may have entered the nucleus with short NE residence times ( <3 min ) , strongly supporting the view that a long NE residence time is a requirement for successful nuclear import . To gain further insight into the movement of viral complexes in the long slow phase , we performed an ensemble mean square displacement ( MSD ) analysis of nuclear viral complexes ( Fig 4A ) . The diffusion rates of nuclear A3F-YFP labeled and IN-YFP labeled viral complexes were low ( 1 and 0 . 6 × 10−4 μm2/sec , respectively ) , but were 2- to 3-fold higher than when they were docked at the NE prior to import ( 0 . 3 × 10−4 μm2/sec ) . A linear relationship between MSD values and time indicate free diffusion . However , the MSD graph lines indicate a sub-linear relationship between the MSD values and time , indicating that the viral complexes inside the nucleus exhibited restricted diffusion . The diffusion coefficients of A3F-YFP and IN-YFP labeled HIV-1 complexes in the long slow phase were within twofold of the diffusion coefficients that have been reported for genes ( reviewed in [50] ) , supporting the hypothesis that the viral complexes are tethered to chromatin . Interestingly , the diffusion rate of A3F-YFP labeled viral complexes was twofold higher than that of IN-YFP labeled viral complexes . The observation suggests that virion incorporation of A3F-YFP alters the structure of the viral complex in a way that reduces its ability to tether to the host chromatin . This effect of A3F may be related to the ability to A3F to inhibit HIV-1 integration by interfering with the IN-mediated 3’ processing reaction [20 , 21] . To further examine the movement of nuclear viral complexes in the long slow phase , we sought to determine whether their movement would be similar to nascent HIV-1 RNA that is tethered to the proviral DNA transcription sites in chromatin until transcription is completed and the RNA is released . If the viral complexes exhibited a diffusion rate that is faster than that of proviral transcription sites , the result would suggest that the viral complexes dissociate/reassociate with multiple chromatin sites before integration and provirus formation . We detected HIV-1 RNA transcribed from the proviral DNAs using a previously described strategy [51] . Briefly , the viral genome was engineered to encode 18 RNA stem-loops that are specifically recognized by the Escherichia coli BglG protein that was tagged with YFP ( Fig 4B ) . It has been previously shown that the strongest RNA signals in the nuclei represent nascent RNA transcripts that are retained at the transcription site until they are released [52–54] . Single cell clones containing one or two proviruses encoding stem-loops that bind to BglG were selected and expanded , the integrated proviral transcription sites were identified by detection of the brightest RNA signals in the nuclei after expression of the BglG-YFP fusion protein ( Fig 4C ) . The movements of 11 transcription sites in living cells ( totaling 47 hours of movement ) were analyzed . The diffusion coefficient of the HIV-1 transcription sites ( 0 . 6 × 10−4 μm2/sec; Fig 4A ) was nearly identical to that of IN-YFP labeled viral complexes and within 2-fold of the A3F-YFP labeled viral complexes , and in agreement with previously reported diffusion coefficients of genes ( reviewed in [50] ) . The results support the hypothesis that the viral complexes are tethered to chromatin and that the movement in the long slow phase was largely due to the movement of the chromatin . We also observed several faint RNA spots in the cells that contained HIV-1 proviruses and expressed the BglG protein , which we hypothesize are HIV-1 ribonucleoprotein complexes ( Fig 4C; [51 , 55] ) . These RNA spots exhibited much faster movement than the RNA transcription sites , and their movements could not be analyzed from the 1 frame/3 min movies . We captured additional movies at 10 frames/sec , performed single particle tracking followed by MSD analysis of their movements ( Fig 4D ) . The results indicated a diffusion rate of 2 × 10−2 μm/sec , which is significantly faster than the diffusion rate of HIV-1 transcription sites ( 0 . 6 × 10−4 μm/sec; Fig 4A ) ; this diffusion rate is in general agreement with previously reported diffusion coefficients for nuclear ribonucleoprotein complexes [54 , 56] . Because of the significantly slower movements of HIV-1 transcription sites , their MSD plot was not significantly different from immobile virus particles on a glass slide at these time lags . Importantly , the MSD analysis can clearly distinguish between HIV RNA transcription sites and HIV ribonucleoprotein complexes . Next , we compared the intranuclear movements of viral complexes in the long slow phase in cells that were treated with RT inhibitor NVP ( Fig 4E ) , IN inhibitor RAL ( Fig 4F ) , or CsA , which disrupts CA-CypA interaction ( Fig 4G ) . The results showed that these treatments had no effect on the intranuclear movements of viral complexes , suggesting that the absence of reverse transcription does not affect the apparent tethering of viral complexes with chromatin , and that unintegrated viral complexes exhibit movements that are similar to those of chromosomal HIV-1 RNA transcription sites . Finally , the intranuclear viral complexes in the presence of CsA presumably did not have any CypA associated with them; the presence or absence of CypA in association with the intranuclear viral complexes did not have any impact on their nuclear movements . Similarly , the CA mutant P90A , which is defective for binding to CypA , exhibited a diffusion rate that is similar to the WT viral complexes and viral complexes in CsA-treated cells ( Fig 4G ) . Lens epithelium-derived growth factor/p75 ( LEDGF/p75 ) is a nuclear protein that binds to nucleosomes of transcriptionally active genes and is thought to tether proteins or protein complexes to chromatin ( reviewed in [57 , 58] ) . LEDGF/p75 binds tightly to HIV-1 IN and facilitates HIV-1 replication by directing integration into transcriptionally active genes [59] . We sought to determine whether interaction between LEDGF/p75 and HIV-1 IN is essential for tethering of viral complexes to chromatin during the long slow phase of nuclear movement . We generated HeLa cells in which a portion of LEDGF/p75 encoding gene ( PSIP1 ) that codes for the integrase binding domain of LEDGF/p75 ( IBD ) was deleted by using CRISPR/cas9 ( Fig 5A ) . Transduction of HeLa cells with CRISPR/cas9 and two different gRNAs resulted in a HeLa cell clone ( HLKO ) with deletion of a 678-bp fragment in two alleles , and inversion of a 648-bp fragment in another allele , resulting in a complete knockout of the IBD ( Fig 5B and S5 Fig ) . Western blotting analysis with an antibody targeting an epitope near the C-terminus of LEDGF/p75 confirmed that LEDGF/p75 expression was undetectable ( Fig 5C ) . Infection of the HLKO cells with a virus that expresses a luciferase reporter gene showed that luciferase expression was reduced by ∼20-fold ( Fig 5D ) . We infected the HLKO cells with A3F-YFP and IN-YFP labeled virions , and compared the movements of nuclear viral complexes to those in WT HeLa cells ( Fig 5E ) . The results showed that deletion of LEDGF/p75 from the infected cells did not have any effect on the movement of A3F-YFP or IN-YFP labeled nuclear viral complexes in the long slow phase . These results suggest that LEGDF/p75 is not required for the proposed tethering of viral complexes to chromatin and that other host factors may be involved in this interaction with the viral complexes . Finally , we compared the nuclear penetration distance of viral complexes in WT and HLKO cells ( Fig 4F ) . We found that the viral complexes were located near the periphery of the nuclei , regardless of the presence or absence of LEDGF/p75 ( median penetration distances of 1 . 7 and 1 . 8 μm , respectively ) . Overall the results indicated that the interaction of viral complexes with LEDGF/p75 is not a determinant of the proposed tethering of viral complexes to chromatin or their nuclear penetration distance . We sought to determine whether reverse transcription and/or CypA associated with viral complexes regulate the dynamics of nuclear import . We previously showed that inhibiting reverse transcription with reverse transcriptase ( RT ) inhibitor nevirapine ( NVP ) or using a catalytic site mutant of RT did not influence the efficiency of nuclear import , leading us to conclude that reverse transcription was not required for nuclear import [14] . However , our previous studies did not rule out the possibility that reverse transcription can regulate the timing of NE docking and/or nuclear import . In addition , the CA-CypA interaction has been shown to modulate capsid uncoating and alter the dependence on some nuclear pore proteins for nuclear import [25 , 29 , 32–34] . Therefore , we sought to determine whether CypA binding and/or reverse transcription can regulate the timing of NE docking and nuclear import . For these studies , we identified 9 additional nuclear import events ( 6 labeled with IN YFP and 3 labeled with A3F-YFP ) by manual analysis of new movies of infected cells ( 1 frame/3 min from 10 min– 10 hrs after infection ) in addition to 17 of the 21 tracked viral complexes described above , providing a total of 26 viral particles that were observed to enter the nucleus in untreated cells ( for this analysis 4 A3F-YFP labeled particles that were at the NE at the beginning of the movies were excluded [S2 Table] ) . Determining the time of nuclear import and docking of viral complexes at the NE relative to the time of infection allowed us to determine the length of time the viral complexes were in the cytoplasm ( time in cytoplasm ) and the length of time the viral complexes were associated with the NE prior to import ( NE residence time ) ( Fig 6A ) . In addition to the 26 particles analyzed from untreated cells , we analyzed 31 , 42 , and 21 nuclear import events of viral complexes in cells treated with NVP to block reverse transcription , in cells treated with CsA to disrupt the CA-CypA interaction , and viral complexes derived from CypA-binding mutant P90A , respectively ( Fig 6B ) . We observed that compared to untreated cells , NVP treatment did not significantly alter the time in cytoplasm , NE residence time , and time of import , indicating that inhibiting reverse transcription did not regulate nuclear import ( Fig 6B–6D ) . However , when compared to untreated cells , disruption of CA-CypA interaction by treatment of target cells with CsA or the CypA-binding CA mutant P90A significantly decreased the time of import; the mean time of import for wild-type viral complexes was 4 . 3 ± 2 . 6 hours post-infection ( hpi ) , which was reduced to 1 . 9 ± 1 . 3 hpi and 2 . 3 ± 2 . 0 hpi for viral complexes in CsA-treated cells or cells infected with the P90A mutant of CA , respectively ( Fig 6B ) . These results indicate that CypA binding to viral capsid results in slowing down nuclear import , whereas disruption of the CA-CypA interaction results in faster nuclear import . The observation that CsA treatment and the P90A mutation had very similar effects on the time of import strongly indicates that the faster nuclear import is not the result of an indirect effect of CsA treatment or an indirect effect of the P90A mutation on the viral capsid structure/function . Next , we determined if the faster nuclear import in CsA-treated cells was due to a decrease in the cytoplasmic residence time prior to NE association and/or a decrease in the NE residence time . We found that treatment with CsA led to a decrease in the time in cytoplasm ( 1 . 2 ± 0 . 9 hpi vs . 2 . 8 ± 1 . 9 hpi , P = 0 . 0001; Fig 6C ) as well as a decrease in the NE residence time ( 0 . 7 ± 1 . 0 hpi vs . 1 . 5 ± 1 . 6 hpi , P = 0 . 0001; Fig 6D ) . Similarly , the P90A mutation also led to a decrease in the time in cytoplasm ( 1 . 7 ± 2 . 1; P = 0 . 0091 ) as well as a decrease in the NE residence time ( 0 . 6 ± 0 . 7 hpi; P = 0 . 0010 ) . A comparison of the shorter times of import , time in cytoplasm , and NE residence time in CsA-treated cells and the P90A mutant indicated that there were no significant differences in these two conditions , further supporting a direct role for CA-CypA interaction in the regulation of nuclear import . We also determined whether CsA treatment or the P90A mutation affected the efficiency of nuclear import in HeLa cells by determining the level of nuclear import in fixed cells at early ( 2 and 6 hpi ) and late ( 24 hpi ) time points . The level of nuclear import was higher at 2 and 6 hpi with CsA treatment or the P90A mutation , which is consistent with our results obtained from the visualization of nuclear import in living cells , but was similar at 24 hpi ( S6A Fig ) . In addition , the infectivity of WT virus in untreated cells , WT virus in the presence of CsA treatment , or the P90A mutant virus was similar in HeLa cells ( S6B Fig ) . These results indicate that although the speed of nuclear import increases with disruption of the CA-CypA interaction , the efficiency of nuclear import for viral complexes and infectivity is similar with or without disruption of the CA-CypA interaction . Our results indicated that lower levels of CA are associated with nuclear viral complexes compared to NE-associated viral complexes ( Fig 1F and 1G ) , suggesting that a portion of viral core uncoating occurs during nuclear import . Previous studies have suggested the CypA stabilizes viral complexes by slowing down uncoating , and that CsA treatment may accelerate viral core uncoating [60] . To explore the mechanism by which the CypA-CA interaction influences the kinetics of nuclear import , we determined the level of CA associated with viral complexes at 2 and 6 hours post-infection in the presence or absence of CsA by immunofluorescence staining ( Fig 6E and 6F ) . At the 2-hour time point , the percentage of cytoplasmic and nuclear A3F-YFP signals that colocalized with CA was similar with and without CsA treatment ( Fig 6E ) , and a modest reduction in the percentage of NE-associated A3F-YFP signals that colocalized with CA was observed with CsA treatment ( 65% vs . 50%; P < 0 . 05 ) . At the 6-hour time point , the proportions of cytoplasmic and NE-associated A3F-YFP signals that colocalized with CA were reduced with CsA treatment . There were no differences in the proportion of nuclear A3F-YFP signals that colocalized with CA at 2 or 6 hours after infection . Next , we compared the CA signal intensities that colocalized with A3F-YFP signals in the cytoplasm , NE , and the nucleus with and without CsA treatment at the 2-hour and 6-hour time points ( Fig 6F ) . We observed that the CA signal intensities that colocalized with the A3F-YFP signals were reduced in the cytoplasm and at the NE at both 2-hour and 6-hour time points , but there were no differences in CA signal intensities associated with the A3F-YFP signals in the nuclei . The reduced CA signal intensities in the cytoplasm and at the NE in CsA-treated cells indicate that CsA treatment facilitated faster viral core uncoating , which may have resulted in faster docking at the NE ( reduced time in cytoplasm ) , as well as reduced NE residence time prior to nuclear import . We sought to determine whether a longer time in cytoplasm is correlated with changes in the viral core that increase speed of nuclear import ( reduce NE residence time ) . As we noted earlier , most viral complexes that formed semi-stable and stable associations with the NE dissociated from it and were not imported into the nucleus . It is not known whether the viral complexes that dissociated from the NE are defective and cannot reassociate with the NE , or whether successive cycles of NE association/dissociation contribute to the dynamics of nuclear import of viral complexes , and result in shorter NE residence times and faster nuclear import . If viral complexes that are imported undergo successive cycles of association/dissociation at the NE , a longer time in cytoplasm might be correlated with more cycles of NE association/dissociation; therefore , we determined whether the time in cytoplasm correlated with a shorter NE residence time prior to import . We found no correlation between the length of time in cytoplasm and NE residence time for each viral complex that we observed enter the nucleus within the four different populations of viral complexes ( WT complexes , WT+CsA , P90A , and WT+NVP population; Fig 6G; Pearson Correlation , P > 0 . 05 ) . The lack of correlation between time in cytoplasm and NE residence time suggests that viral complexes in the cytoplasm , or those that make multiple transient contacts with the NE without being imported , did not accumulate any structural/conformational changes that shortened the NE residence time . Overall , these results suggest that NE docking and translocation from the NE to the nucleus are independent events , both of which are regulated by the CA-CypA interaction . It has been reported that infectivity of CA mutant N74D is independent of TNPO3 when using VSV-G-mediated entry , but dependent on TNPO3 when using HIV-1 envelope ( Env ) mediated entry [61] . To determine whether virions that use the HIV-1 Env and fuse at the plasma membrane have different kinetics of NE docking and nuclear import , we compared the kinetics of HIV-1 NE docking and nuclear import in HeLa cells for VSV-G mediated entry to HIV-1 Env mediated entry ( Fig 7A , upper and middle panels ) . For these experiments , we constructed a TZM-bl HeLa cell line that expresses POM121-mCherry ( Fig 7A middle panel ) ; TZM-bl cells express HIV-1 Env receptor CD4 and co-receptor CCR5 on their cell surface and virions with HIV-1 Env can infect these cells through fusion at the plasma membrane . We did not find any significant difference between VSV-G pseudotyped virions and HIV-1 Env virions with regard to the time of nuclear import ( Fig 7B ) , time in cytoplasm ( Fig 7C ) , or NE residence time ( Fig 7D ) , indicating that the envelope used for entry into cells does not have any significant effect on the kinetics of NE docking or nuclear import . Finally , we sought to rule out the possibility that expression of POM121-mCherry , a nuclear pore protein , influenced the kinetics of NE docking and/or nuclear import . We constructed a HeLa cell line that expresses nucleosomal protein H2B fused to mCherry ( H2B-mCherry ) ; H2B-EYFP expressing cells were previously used to study HIV-1 nuclear import [23] . We infected the H2B-mCherry expressing cells ( Fig 7A , bottom panel ) with HIV-1 virions pseudotyped with VSV-G and compared the kinetics of NE docking and nuclear import to HeLa cells that expressed POM121-mCherry; we did not find any significant differences regarding the time of nuclear import ( Fig 7B ) , time in cytoplasm ( Fig 7C ) , or NE residence time ( Fig 7D ) , indicating that expression of POM121-mCherry did not significantly influence the kinetics of NE docking or nuclear import .
Because nuclear import of HIV-1 has never been observed in living cells , the behavior of the viral complexes at the NE prior to nuclear import is not known . We observed the translocation of 30 A3F-YFP or IN-YFP labeled WT HIV-1 complexes from the cytoplasm to the nucleus , determined the NE residence times of 26 viral complexes , and the intranuclear movements of 21 viral complexes . The behavior of IN-YFP labeled HIV-1 complexes was indistinguishable from the behavior of A3F-YFP labeled viral complexes with respect to their NE residence time , time of import , nuclear penetration distance , and distance traveled from the point of nuclear entry . Our results provide strong evidence that a long NE residence time prior to import ( ∼1 . 5 ± 1 . 6 hrs ) is likely to be a characteristic of viral complexes that are imported and go on to integrate and form a provirus that can express its genome . We also determined that viral complexes cannot enter the nucleus with short NE residence times ( <3 min ) , since nearly all nuclear viral complexes ( 109/110 ) , with the exception of one dim particle , could be tracked to the NE or to the top or bottom z-slices of the observed nuclear volume . This is in stark contrast to the nuclear import of AAV2 and other large cellular cargos , which dock at the NE and are transported through the nuclear pore within milliseconds [8–10] . Our results show that long and stable associations are necessary but not sufficient for nuclear import and are dependent upon CA and Nup358 . The HIV-1 complexes that were imported into the nucleus did not show any apparent lateral movements on the nuclear membrane , suggesting that they were docked at the nuclear pore through which they eventually entered the nucleus . HIV-1 viral cores are presumably disassembled before nuclear import due to their large size ( 61-nm width , 120-nm length ) compared to the 40-nm size limit for translocation through a nuclear pore [5 , 6] . The intracellular location at which viral core uncoating occurs has not been fully established . Several studies have suggested that uncoating occurs in the cytoplasm [40 , 45] and is temporally associated with reverse transcription [62–65] , while another study has suggested a role for the cyclophilin homology domain of Nup358 in capsid disassembly at the nuclear pore [48]; however , a direct role for Nup358 in engaging the capsid prior to nuclear import remains controversial [49] . We propose that viral cores may require a long NE residence time , during which they undergo extensive CA dissociation and/or conformational rearrangements that are a prerequisite for nuclear import . Consistent with this hypothesis and in agreement with others [39 , 66] , we observed reduced intensity of CA signal associated with nuclear viral complexes compared to viral complexes docked at the NE , suggesting that one of the viral core uncoating steps occurs during nuclear import . It should also be noted that in addition to viral core uncoating , structural/conformational changes to the NPC may be required to allow nuclear import of HIV-1 complexes . Our results show that the CA-CypA interaction , but not reverse transcription , regulates the dynamics of nuclear import . Disruption of the CA-CypA interaction with CsA treatment or with the P90A mutation resulted in faster nuclear import of viral complexes by decreasing both the time in cytoplasm ( faster NE docking ) and NE residence time ( faster translocation into the nucleus ) . This result is consistent with a previous report , which concluded that CypA affects nuclear entry of viral cDNA [35] . Although disruption of the CA-CypA interaction reduced the NE residence time from 1 . 5 hrs to 0 . 6–0 . 7 hrs , their NE residence time ( ∼40 min ) is still substantially longer than the milliseconds required for transport of adeno-associated virus or other cellular cargos . Faster nuclear import of viral complexes in CsA-treated cells may be related to the fact their import is insensitive to depletion of nuclear pore proteins Nup153 , Nup358 and TNPO3 [32–34] . CsA treatment inhibits HIV-1 replication in some T cell lines and primary CD4+ T cells [29–31 , 35] , suggesting that CypA binding evolved to facilitate HIV-1 replication in its natural target cells of infection . Our observation that disrupting the CA-CypA interaction in HeLa cells leads to faster NE docking and nuclear import indicates that one function of CypA binding is to slow down nuclear import . We hypothesize that one function of CypA binding is to protect incompletely synthesized viral DNA from host nucleases and/or DNA repair enzymes in the nucleus , leading to more efficient viral replication . Our hypothesis is consistent with a previous proposal that CypA binding keeps viral nucleic acid sequestered in the viral core and evades innate immune nucleic acid sensors [28] . Development of assays to visualize the nuclear import of HIV-1 in T-cells and monocyte-derived macrophages will be required to test these hypotheses . Several studies have shown a link between reverse transcription and viral core uncoating and/or conformational changes in the viral core [39 , 62 , 63 , 67] . While these studies clearly showed that reverse transcription triggers loss of CA and/or conformational changes in the viral core , the significance of these structural changes in the core to NE docking or nuclear import was not examined . Interestingly , our results showed that inhibiting reverse transcription did not alter the time in cytoplasm or NE residence time , indicating that the uncoating and/or conformational changes in the viral core that are triggered by reverse transcription are not required for NE docking or nuclear import . We speculate that the changes in the viral core eventually occur in the absence of reverse transcription in a manner that does not affect the kinetics of nuclear import . We observed that right after import the nuclear viral complexes exhibit relatively fast mobility ( brief fast phase ) followed by a second phase of slower mobility ( long slow phase ) , suggesting that during the brief fast phase viral complexes have not associated with chromatin or large macromolecules and may be diffusing freely . The viral complexes exhibited slower movements in the long slow phase , and their diffusion rate was similar to the rate we observed for HIV-1 transcription sites ( ∼0 . 6 × 10−5 μm2/sec ) and the rates previously reported for genes ( reviewed in [50] ) . These observations support the hypothesis that the viral complexes are tethered to chromatin and their movement in the long slow phase is due in large part to the movement of the chromatin . Intriguingly , treatment of target cells with NVP , RAL , or CsA did not affect the mobility of the nuclear viral complexes , indicating that the completion of reverse transcription or successful integration are not required for the proposed tethering to chromatin . Although LEDGF/p75 is known to interact with HIV-1 IN and target viral complexes to their integration sites , our result suggested that LEDGF/p75 is not the primary determinant of the proposed tethering of viral complexes to chromatin after nuclear import . It is noteworthy that LEDGF/p75 is dispensable for targeting HIV-1 to less condensed euchromatic regions [68] . In future studies , it will be of interest to determine whether CPSF6 , a host nuclear protein that is known to bind CA and affect HIV-1 integration [69] , HRP-2 [70] , or other host proteins that have been reported to interact with integrase [71 , 72] affect the mobility of HIV-1 nuclear complexes and integration site selection . Our data support the hypothesis that viral complexes become tethered to chromatin at or near their sites of integration . The peripheral location of viral complexes that we observed is consistent with previous reports indicating that HIV-1 integrates at the nuclear periphery [11 , 12 , 73] , but in contrast to another report which suggested that integration in wild-type HIV-1 occurs randomly throughout the nucleus [13] . Our observation that the viral complexes may be tethered to chromatin shortly after entering the nucleus and remain near the nuclear periphery is in line with the models proposed by Marini et al . [12] , Lelek et al . [73] , and Di Primio [11] . However , our data also indicates that the virus-chromatin complex can move away from the nuclear point of entry , albeit slowly , so it may not be correct to assume that a viral complex entered through the nearest nuclear pore . HIV-1 and other retroviruses have low ratios of infectious to non-infectious particles; consequently , most of the observed viral complexes in microscopy studies may be non-infectious . We previously determined that in our microscopy experiments , ∼50 particles enter each cell , of which ∼2 enter the nucleus , and ∼1 forms a provirus that can express a GFP reporter gene [14] . We determined that the multiplicity of infection ( MOI ) of A3F-YFP labeled virions in the imaging experiments was ∼1 . 1 , defined as proviruses/cell that express a GFP reporter gene ( S1 Table ) . In the current live-cell microscopy studies , we observed ∼1 . 6 A3F-YFP labeled viral particles per nucleus ( 44 A3F-YFP labeled complexes/28 cells ) , which is close to the MOI of ∼1 . 1 . However , since our A3F-YFP labeling efficiency was ∼50% , we estimate that ∼3 . 2 viral complexes entered per nucleus , of which ∼1 . 1 generated a provirus that expressed GFP , indicating that ∼1 of 3 nuclear viral complexes were infectious . Thus , a high percentage of nuclear viral complexes , which may be unlabeled or labeled with A3F-YFP , are infectious . Similar overall behavior of A3F-YFP labeled and IN-YFP labeled viral complexes that entered the nucleus ( time in cytoplasm , NE residence time , time of import , nuclear penetration distance , distance from point of entry , and the short fast phase and long slow phase of nuclear movements ) suggests that these observed characteristics reflect the average behavior of infectious viral complexes . In our microscopy studies , only a small proportion of cytoplasmic or NE-associated viral complexes enter the nucleus , indicating that only a minority of cytoplasmic and NE-associated viral complexes are infectious . The amount of CA associated with cytoplasmic and NE-associated viral complexes reflects the average of all viral complexes , and the possibility that this average differs from the amount of CA associated with the infectious viral complexes cannot be excluded . Methods to efficiently and quantitatively detect CA in living cells , such as tetracysteine-tagged CA [74 , 75] , and the recently developed dsRed-tagged CypA [63] , are needed to determine the state of the viral cores of infectious viral particles at the NE . In summary , we determined the kinetics of NE docking of HIV-1 complexes , and showed that the stable associations which involve CA and Nup358 are functionally important for nuclear import . We also observed the translocation of HIV-1 complexes from the cytoplasm to the nucleus and found that the CypA-CA interaction , but not reverse transcription , regulates the timing of nuclear import , by increasing the time the viral complex resides in the cytoplasm as well as at the NE immediately prior to nuclear import . We observed two phases of nuclear movement , suggesting the viral complexes associate with chromatin at or near the site of integration shortly after entering the nucleus . Future studies of the dynamics with which HIV-1 complexes associate with the NE , enter the nucleus , move in the nucleus and interact with chromatin will increase our understanding of these essential steps in HIV-1 replication . Finally , these studies may provide insights into the mechanisms involved in the nuclear import of large cytoplasmic macromolecular complexes .
HeLa , TZM-bl , and 293T cells were maintained as previously described [14 , 76] . HeLa cells or TZM-bl cells stably expressing POM121-mCherry were created by transduction with a lentiviral vector containing the human POM121-mCherry expressed from the ubiquitin C promoter . HeLa cells stably expressing H2B-mCherry were created by transduction with a lentiviral vector containing the human H2B-mCherry expressed from the CMV promoter ( Addgene plasmid #20972; [77] ) . Immunofluorescence staining was performed as previously described [14]; AG3 . 0 antibody was used to detect HIV-1 CA ( NIH AIDS Reagent Program ) . NVP and RAL were obtained through the NIH AIDS Reagent Program and was used at a final concentration of 5 μM and 10 μM , respectively . CsA ( Millipore ) was used at a final concentration of 5 μM . Cell viability was determined using the ATPlite Luminescence Assay System ( PerkinElmer ) according to manufacturer’s protocol . HeLa cells were reverse-transfected with siRNA targeting Nup358 or control siRNA ( Ambion ) using RNAiMax ( Invitrogen ) as previously described [14] . Cell lysates were harvested and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by Western Blot analysis using the Odyssey System ( Li-Cor ) as previously described [78] . Antibodies were obtained for specific detection of Nup358 ( Abcam ) , LEDGF/p75 ( Cell Signaling ) , or α-tubulin ( Sigma ) . Primary antibodies were detected using either infrared dye-labeled ( 800C ) goat anti-rabbit or infrared dye-labeled ( 680 ) goat anti-mouse secondary antibodies ( Li-Cor ) . HeLa cells were transfected ( TransIT-LT1 Transfection Reagent; Mirus Bio LLC ) with 0 . 5 μg each of Cas9-GFP ( Addgene #44719 ) plus plasmids expressing gRNA LFor ( 5’- GGCTAAGTATAATGAATTAG -3’ ) and gRNA LRev ( 5’- TTTAGAACATGTTCTTGGT-3’ ) . The gRNAs were positioned to bind in intron 11 and exon 14 of LEDGF to delete the IBD domain ( Fig 4D ) . Single cell clones were isolated by limiting dilution . Genomic DNA isolated from single cell clones was amplified by PCR using primers outside the gRNA binding sites ( Lfor 5’-GGGACTGGAGAGCAGAGGAGTTATC-3’ and Lrev 5’- GCTCTAGTCCTTCAATAGGCCCATCAGA-3’ ) , producing either undeleted ( 1158 bp ) or deleted IBD bands ( ∼500 bp ) . Clones positive for deleted IBD domains were further sequenced ( Macrogen ) to verify Cas9/gRNA cleavages . Cell lysates from positive clones were analyzed by Western Blot analysis ( described above ) . Fluorescent protein-labeled HIV-1 particles were prepared by co-transfection of 293T cells with the HIV-1-derived vector pHDV-EGFP ( 10 μg; [79] ) or pH-EGFP ( 10μg; pH-EGFP is derived from pHL [80] except that vif and vpr were deleted and EGFP replaced the luciferase gene ) , HCMV-G ( 2 . 0 μg , VSV-G; [81] ) , and A3F-YFP ( 1 . 25 μg ) or Vpr-IN-YFP ( 5 . 0 μg ) as previously described [14] . In some experiments , an HIV-1 envelope expression plasmid ( AD8; 2 . 0 μg ) was used instead of HCMV-G [82] . Vpr-IN-YFP is similar to Vpr-IN-GFP [23] , except that GFP was replaced with YFP and the HIV-1 protease cleavage site between Vpr and IN ( IRKVL/FLDGI ) is preceded by a flexible glycine-rich linker ( “/” indicates protease cleavage ) . In some experiments , the viral membrane marker S15-mCherry ( 5 μg ) was also used [44] . In some experiments , the CA region in HDV-EGFP was replaced with CA mutants K203A and E128A/R132A [14] . The CA region in pH-EGFP was replaced with CA mutant P90A , which was created using site-directed mutagenesis ( Agilent ) . For additional virus characterization , a 1:1 mixture of unlabeled Gag ( pHDV-EGFP ) and Gag-iCFP ( pGag-iCFP; contains a CFP between MA and CA and is flanked by HIV-1 protease cleavage sites [kindly provided by Dr . Marc Johnson , University of Missouri] ) was used during virus production so that HIV-1 particles could be identified by CFP fluorescence . pC-Help ( an HIV-1 helper construct that lacks several cis-acting elements needed for viral replication , including the packaging signal and primer-binding site ) , GagCeFP-BglSL ( HIV-1 construct containing 18 copies of Bgl stem loops ) , and BglG-YFP expression vector were previously described [83] . HIV-1 particles containing the genomes with Bgl stem loops were prepared by co-transfection of 293T cells with GagCeFP-BglSL ( 8 μg ) , pC-help ( 3 μg ) , and HCMV-G ( 2 . 0 μg , VSV-G ) . HeLa cells were seeded in ibiTreated μ-slides ( 3 × 104 cells/well ) one day prior to infection . Cells were infected with a normalized number of fluorescently-labeled particles via spinoculation at 16°C , which permitted virion binding to cell membranes but prevented virion endocytosis as previously described [14 , 84] . After centrifugation , the media was replaced with prewarmed media to allow internalization of the virus ( defined as the 0-h time point ) and thereafter incubated at 37°C . Infections with A3F-YFP labeled virions were performed at a multiplicity of infection ( MOI ) of ∼1 ( S1 Table ) . Time-lapse images of the infected cells were acquired by epifluorescence microscopy ( described below ) or the cells were fixed at various time points post-infection with 4 . 0% ( wt/vol ) paraformaldehyde ( PFA ) and imaged by confocal microscopy ( described below ) . To determine infectivity for the viruses containing a GFP reporter gene ( viruses made using pHDV-EGFP ) , the percentage of GFP+ cells was determined by flow cytometry analyses performed on a FACSCalibur system ( BD Biosciences ) 48 hrs post-infection . Confocal images were acquired of fixed cells using an LSM710 laser scanning confocal microscope ( Zeiss ) with a Plan-Apochromat 63x N . A . -1 . 40 oil objective , using 405-nm , 515-nm , 561-nm , and 633-nm lasers for illumination or a Nikon Eclipse Ti-E microscope equipped with a Yokogawa CSU-X1 spinning disk unit with a Plan-Apochromat 60x N . A . 1 . 40 oil objective , using 405-nm , 514-nm , and 633-nm lasers for illumination . The diffraction-limited spots were detected and their positions were determined in each image using Localize [85] . The positions of the spots were used to determine colocalization; spots were considered colocalized if the centers of the spots were within 3 pixels . Colocalization of the YFP particles with the NE mask ( based on Lamin A/C immunofluorescence staining ) was determined using a custom-written MATLAB program ( Mathworks ) . A mask of the nucleus interior was also created using the NE mask . The percentage of cytoplasmic YFP particles that colocalized with the NE mask and the percentage of YFP particles that colocalized with the nucleus mask were determined . A custom-written MATLAB program was used to determine the colocalization of A3F-YFP with HIV-1 p24 capsid ( CA ) signal in the cytoplasm , at the NE , and inside the nucleus of infected cells . Because of variable background intensity across different regions of the cell , the intensity values were determined in the Cy5 channel ( channel used to detect CA ) at random positions in the cytoplasm , at the NE , and inside the nucleus . The threshold intensity values were determined as the mean + 1 SD of the random intensities for each region . The intensity values of the Cy5 channel at the position of each A3F-YFP particle was determined; A3F-YFP particles co-localizing with Cy5 signals that were above the threshold intensity value were considered positive for CA . Epifluorescence microscopy was performed as previously described , with some modifications [51] . Briefly , we employed an inverted Nikon Eclipse Ti microscope and a 100 × 1 . 45-N . A . oil objective , using 514-nm and 594-nm lasers for illumination . A 1x tube lens ( pixel size = 0 . 160 μm ) or a 1 . 5x tube lens ( pixel size = 0 . 107 μm ) was used . Digital images were acquired using an Andor iXon3 897 Camera and NIS-Elements software ( Nikon ) with emission filters of 542/27 nm and 650/75 nm , respectively . Simultaneous dual-color imaging was performed by using a Four-Channel Simultaneous-Imaging System ( QV2; Photometrics ) ; the separate channels were manually aligned using fluorescent virus particles on a slide prior to each experiment . In addition , the images of separate channels were further aligned with single-pixel accuracy using a custom-written MATLAB program . The localization precision of the microscope was determined by acquiring time-lapse images of immobilized A3F-YFP labeled virus particles on a slide under the same conditions as live-cell imaging experiments . Single-particle tracking of A3F-YFP particles ( described below ) was performed and the single-step jump distances ( i . e . the distance a particle moves between two consecutive frames ) were determined . The average 1-step jump distance for all YFP ( ∼43 nm ) signals indicates a localization precision for each channel that is suitable for single-particle tracking . Virus particle movement was followed by acquiring time-lapse images at 9 . 8 Hz with a 100-ms integration time and ∼2-ms overhead between frames , resulting in an overall 102-ms frame time . Since high frame rates were used to capture viral movements , we were limited to short , 1-min movies due to photobleaching . Therefore , several 1-min movies were taken of the infected cells ( 1 movie per cell ) from 45 mpi to 2 hpi . The focal plane was selected at approximately equatorial plane of the cells . It was not necessary to deconvolve images because contribution of out-of-focus light was minimal ( signal-to-noise ratio >4 ) . Single-particle tracking was performed with MATLAB code based on the available tracking algorithms [86] , with maximum single-step displacement of five pixels ( 0 . 54 μm ) when using the 1 . 5x tube lens and a minimum track length of five consecutive frames . The positions of the diffraction-limited spots in the tracks were refined with 2D Gaussian fit [87] . Colocalization of the A3F-YFP tracks with the POM121-mCherry signal was determined using a custom-written MATLAB program . First , it was determined that there is no detectable cell movement within the 1-min movies ( S1A Fig; S1 Movie ) . Therefore , a mask of the NE using the POM121-mCherry signal was created using an average projection of the mCherry channel of the entire time series . This mask was approximately 1 μm in width and was centered on the POM121-mCherry signal ( Fig 2B ) . Occasionally , there were regions of the NE in which the POM121-mCherry signal was not well-defined and therefore a mask could not be accurately created . These ill-defined regions were manually selected and any tracks that entered these regions were excluded from the analyses . The longest consecutive residence time for each A3F-YFP particle that colocalized with the POM121 mask for at least 1 frame was determined . To determine the kinetics of association of randomly moving particles with the NE , a simulation was performed using a custom-written MATLAB program . Briefly , a mask of the nucleus was created using the NE mask ( described above ) . The masks were made using the POM121-mCherry signal obtained during a typical experiment . Then , a simulation of randomly moving particles was performed with 2 parameters; particles could not leave the outer limits of the image and could not colocalize with the nucleus mask ( since in a typical 1-min movie we did not observe translocation of HIV-1 complexes from the NE to the nucleus ) . For each experiment , the random movement of particles around 3 different nuclei was simulated , resulting in ∼550 particles that contacted each NE mask during the 1-min simulation ( >1600 total particles ) . Next , the longest consecutive residence time for each particle that contacted the NE mask for at least 1 frame was determined . The same random movement simulation was repeated two more times and analyzed . To visualize the association of HIV-1 complexes with the NE over a longer time period , a 9-slice z-stack ( taken at 0 . 4-μm step intervals ) was acquired every 1 min for 2 hrs starting at 10 mpi to 6 hpi . Each z-stack , which covered an axial depth of ∼4 μm , was centered near the equatorial plane of the nucleus; the top and bottom of the nuclei were not imaged to minimize photobleaching . The residence time for each particle that was associated with the NE for 2 consecutive frames ( 1 min ) was determined manually . To observe nuclear import of A3F-YFP-labeled or IN-YFP-labeled particles , a 9-slice z-stack ( taken at 0 . 4-μm step intervals ) was acquired every 3 min shortly after infection for up to 10 hrs . Each z-stack , which covered an axial depth of ∼4 μm , was centered near the equatorial plane of the nucleus; the top and bottom of the nuclei were not imaged to minimize photobleaching during the long movies ( S3A Fig ) . As a result , particles that entered the nucleus from the top or bottom were not detected . Deconvolution was applied to the images to remove the out of focus noise using Huygens software ( SVI , Netherlands ) . The lateral movement of the nucleus during each movie was corrected using the POM121-mCherry signal and a custom-written MATLAB program ( S5 Movie ) . After adjusting for the movement of the nucleus during the time-course , 3D localization ( x , y , z ) of the YFP-labeled particles was conducted using FISH-QUANT software [88] . U-track algorithm was used for 3D single-particle tracking [89] . Ensemble MSDs were calculated from positional coordinates as previously described [90] . In free diffusion , the MSDs [r2 ( t ) ] for 3D tracks are linearly related to time ( t ) and diffusion coefficient ( D ) by the formula r2 ( t ) = 6Dt . The MSDs for 2D tracks are linearly related to time and diffusion coefficient by the formula r2 ( t ) = 4Dt . The diffusion coefficients were calculated using the first four time lags . The nuclear penetration distance of the HIV-1 complexes imported into the nuclei was determined using a custom-written MATLAB program . A mask of the movement-corrected nucleus ( POM121-mCherry images; described above ) was created for each time-point . Then , a series a 2-pixel wide ( 0 . 32 μm ) concentric rings were created; the particle positions ( as determined by 3D single-particle tracking ) were used to determine the colocalization with the nuclear rings ( S3C Fig ) . For each HIV-1 complex that we observed enter the nucleus , the average nuclear penetration distance ( calculated by averaging the nuclear penetration distance for all frames in which the HIV-1 complex was located in the nucleus ) , maximum nuclear penetration distance , average distance from the nuclear point of entry ( calculated by averaging the distances between each of the positions of the HIV-1 complex inside the nucleus and the position of the HIV-1 complex on the NE one frame before nuclear import ) , maximum distance from the nuclear point of entry , observation time at the nuclear envelope and inside the nucleus , time of NE association ( relative to time of infection ) , and time of nuclear import ( relative to time of infection ) were determined .
|
Although nuclear import of HIV-1 is essential for viral replication , many aspects of this process are currently unknown . Here , we defined the dynamics of HIV-1 nuclear envelope ( NE ) docking , nuclear import and its relationship to viral core uncoating , and intranuclear movements . We observed that HIV-1 complexes exhibit an unusually long residence time at the NE ( ∼1 . 5 hrs ) compared to other cellular and viral cargos , and that HIV-1 capsid ( CA ) and host nuclear pore protein Nup358 are required for NE docking and nuclear import . Soon after import , the viral complexes exhibit a brief fast phase of movement , followed by a long slow phase , during which their movement is similar to that of integrated proviruses , suggesting that they rapidly become tethered to chromatin through interactions that do not require LEDGF/p75 . Importantly , we found that NE association and nuclear import is regulated by the CA-cyclophilin A interaction , but not reverse transcription , and that one of the viral core uncoating steps , characterized by substantial loss of CA , occurs concurrently with nuclear import .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"nuclear",
"import",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"hela",
"cells",
"pathogens",
"cell",
"processes",
"biological",
"cultures",
"microbiology",
"viral",
"structure",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"luminescent",
"proteins",
"rna",
"viruses",
"yellow",
"fluorescent",
"protein",
"cell",
"cultures",
"epigenetics",
"cellular",
"structures",
"and",
"organelles",
"chromatin",
"research",
"and",
"analysis",
"methods",
"viral",
"core",
"chromosome",
"biology",
"proteins",
"medical",
"microbiology",
"hiv",
"gene",
"expression",
"microbial",
"pathogens",
"hiv-1",
"cell",
"lines",
"virions",
"cytoplasm",
"biochemistry",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"cultured",
"tumor",
"cells",
"lentivirus",
"organisms"
] |
2017
|
Dynamics and regulation of nuclear import and nuclear movements of HIV-1 complexes
|
The extraction of hidden information from complex trajectories is a continuing problem in single-particle and single-molecule experiments . Particle trajectories are the result of multiple phenomena , and new methods for revealing changes in molecular processes are needed . We have developed a practical technique that is capable of identifying multiple states of diffusion within experimental trajectories . We model single particle tracks for a membrane-associated protein interacting with a homogeneously distributed binding partner and show that , with certain simplifying assumptions , particle trajectories can be regarded as the outcome of a two-state hidden Markov model . Using simulated trajectories , we demonstrate that this model can be used to identify the key biophysical parameters for such a system , namely the diffusion coefficients of the underlying states , and the rates of transition between them . We use a stochastic optimization scheme to compute maximum likelihood estimates of these parameters . We have applied this analysis to single-particle trajectories of the integrin receptor lymphocyte function-associated antigen-1 ( LFA-1 ) on live T cells . Our analysis reveals that the diffusion of LFA-1 is indeed approximately two-state , and is characterized by large changes in cytoskeletal interactions upon cellular activation .
The lateral mobility of cell-surface proteins plays a critical role in mediating the biological functions of membrane proteins [1] . The diffusion of membrane components is affected by factors including the viscosity of the membrane , clustering of the receptor , and binding to cellular components . The spatio-temporal dynamics of membrane-associated receptors are therefore of considerable interest as they can provide crucial insight into cellular signal transduction . A variety of biophysical techniques , particularly fluorescence microscopy experiments , have been extensively utilized to quantify the lateral mobility of membrane proteins . The complementary techniques of single particle tracking ( SPT , reviewed in Ref . [2] ) and fluorescence recovery after photobleaching ( FRAP , reviewed in Ref . [3] , [4] ) probe these dynamics at different length scales . FRAP captures the behavior of a population of labeled particles on a spatial scale of a few microns , while SPT records the dynamics of individual molecules or small macromolecular clusters over lengths of tens to hundreds of nanometers . In a typical SPT experiment , a membrane-associated protein is labeled , either fluorescently or with an antibody conjugated bead , and imaged using high speed video microscopy with a temporal resolution of tens of milliseconds or less . The spatial coordinates of the particle can be determined to a sub-optical resolution of tens of nanometers , permitting a detailed examination of the particle's motion [5] , [6] . The enhanced spatial resolution of SPT , as well as its non-ensemble nature , make the technique attractive for detailed single molecule studies of cell surface receptor dynamics . The analysis of particle trajectories is commonly based on a classification into different modes of motion , such as Brownian , hop diffusion , confined motion or directed diffusion based on fits to their mean squared displacement ( MSD ) over time [7] , [8] . Brownian diffusion is characterized by a linear increase in MSD with time with a slope proportional to the diffusion coefficient . The timescale of diffusion is often treated by analyzing diffusion over short time periods ( typically 1–4 timesteps or tens of milliseconds ) , referred to as microdiffusion , or longer time periods ( typically on the order of seconds ) , referred to as macroscopic diffusion . Deviations from linearity are ubiquitous in time versus MSD data for membrane-associated proteins . Such deviations are variously attributed to flow , the presence of obstacles , membrane compartmentalization or changes in membrane lipid organization [9] , [10] . Numerous modelling studies have examined the effect of membrane structure on particle trajectories and have proposed methods to identify structural features of the plasma membrane responsible for the observed diffusion [11]–[16] . Further difficulties in the analysis of SPT data arise as individual trajectories often show evidence of heterogeneity that is not easily resolved [17]–[21] . Thus new methods of analyzing particle trajectories are needed to extract and interpret subtle changes in diffusive behavior . Both FRAP and SPT experiments on adhesion receptors commonly show a large reduction in receptor mobility upon binding with cytoskeletal components . Therefore , receptor motion may involve multiple states ( i . e . bound or unbound ) that contribute to the diffusion of the receptor in different ways . In a previous study of the T cell integrin receptor , LFA-1 , particle trajectories were acquired with a temporal resolution of 1000 frames/s using antibody-conjugated beads [22] . Macroscopic diffusion coefficients calculated using an MSD analysis were shown to be distributed in two distinct subpopulations . Relative contributions of the two subpopulations varied when the cells were treated with different pharmacological agents , and when different conformations of the protein were preferentially labeled . These results suggested a dynamic equilibrium of LFA-1 between two states with distinct mobilities . Using cytoskeletal inhibitors , it was shown that the cytoskeleton was largely responsible for the state with low mobility . The existence of multiple states with distinct diffusive properties has also been observed for the CD2 receptor on the surface of T cells [23] . In these studies , evidence of heterogenous diffusion was obtained using an MSD analysis that required a large number of replicates for a reliable identification of the underlying states . Additionally , the analysis used relied on changes in the average diffusion , making it difficult to detect subtle or transient changes in diffusivity within single trajectories . Here , we present a novel analytical framework to identify multiple diffusion states and estimate probabilities of switching between them , from particle trajectories of cell-surface proteins . Transitions between these states represent the binding and unbinding of receptors to cytoskeletal contacts or other intracellular signalling components . We introduce a new model that treats particle trajectories as the outcome of a two-state hidden Markov process , parametrized by diffusion coefficients of the two states and rates of transition between them . We adopt a likelihood maximization strategy to identify model parameters that best describe a set of tracks , thus characterizing the underlying diffusive states and the kinetics of the transitions between them . This analysis was first tested with a series of simulated trajectories and compared with previous approaches for isolating subpopulations . We show that our analysis achieves a more accurate and informative resolution of the underlying biophysical parameters for a complex trajectory consisting of multiple states of diffusion . We tested the applicability of this analysis to experimental data of LFA-1 particle trajectories , and found that the diffusion of this adhesion molecule can indeed be treated as a two-state process due to its interactions with cytoskeletal binding partners . Our analysis identifies the characteristic diffusion coefficient of LFA-1 in the two states , and reveals the kinetics of switching between them . The use of a likelihood-based approach further allowed us to compare multiple models for given experimental data , and identify the statistically most optimal model that captures the receptor dynamics .
We modeled single particle tracks for a labeled , membrane-associated protein that binds to a uniformly distributed intracellular substrate , such as cytoskeletal binding proteins . This binding is schematically represented by the bimolecular reaction ( 1 ) where and are the free and bound forms of the protein , and is the substrate . The kinetics of this interaction are characterized by the bimolecular forward rate constant , , and a first order unbinding rate constant , . We assume a homogeneous spatial distribution of the substrate so that at equilibrium the binding reaction is effectively first order with a rate constant , where is the equilibrium concentration of the free substrate . With this assumption , we can represent the bimolecular reaction , at equilibrium , by the unimolecular reaction ( 2 ) where and are the diffusion coefficients of the protein in its free and bound forms , respectively . We further make the simplifying assumption that the particle is imaged instantaneously , and that changes in the particle state occur only at the acquisition time , implying that the particle is entirely in one or the other state between successive image frames ( see Discussion for more details ) . For a constant frame rate , , where is the sampling interval , this assumption leads to the following fixed transition probabilities for the particle to switch its state between successive frames ( see Text S1 for a derivation ) : ( 3 ) ( 4 ) ( 5 ) In this model , the state sequence of the particle during an SPT experiment is regarded as a 2-state Markov chain . The displacement of the particle at each step is the outcome of Brownian diffusion with a diffusion coefficient corresponding to the particle state at that interval . As described in Materials and Methods , to simulate a single particle track arising from the 2-state dynamics described above , we first generated a discrete Markov chain that specifies the particle state at each time point . The initial state of the particle was chosen randomly according to the stationary probabilities of the two states , and the remaining states were determined using a discrete-time stochastic algorithm ( Algorithm 1; Fig . 1 ) . The particle displacements at each frame were sampled from a zero mean Gaussian distribution with variance proportional to the diffusion coefficient . In an experimental trajectory , only the particle position is recorded and information about the particle state must be inferred from the displacement of the particle between successive frames . Therefore , in our model , a particle trajectory is regarded as the outcome of a 2-state hidden Markov model ( HMM ) [24] consisting of a sequence of discrete states – free or bound – that are hidden from the observer , and an observable displacement at each time point from a well-defined probability distribution ( Fig . 2A ) . As demonstrated below , a traditional analysis using the mean squared displacements does not reveal the diffusion coefficients of the constituent states , the rates of transition between them , or the state-sequence underlying an observed track . Therefore , we developed a likelihood-based analysis of single particle tracks to infer these model parameters and thus quantify the underlying biophysical process . It should be noted that , though we have chosen to test the two-state model described above , the hidden Markov formulation and the associated likelihood maximization scheme is a more general and powerful technique for analyzing a wide range of models . In particular , for sufficiently well resolved data , an arbitrarily complex model with multiple states , with diffusive , confined or directed motion could be analyzed using this method . We intend to explore such general models in future studies . We first consider a trajectory arising from 2D Brownian diffusion and sampled at fixed time intervals , . For an observed sequence of independent displacements , the likelihood of a diffusion coefficient is ( 6 ) where . To calculate the maximum likelihood estimate of we define the log likelihood function ( 7 ) ( up to an additive constant ) and maximize it with respect to to obtain ( 8 ) where is the mean squared step size . This maximum likelihood estimate of the diffusion coefficient is most closely related to the microscopic diffusion coefficient obtained from an MSD analysis . The previous equation can be rewritten in the following familiar form ( 9 ) with and is the sum of squared displacements . For a particle undergoing Brownian diffusion , the single parameter sufficiently describes the particle motion . In Fig . 2B we plot for three sets of simulated trajectories , two for Brownian diffusion with a single diffusion coefficient , and one for 2-state diffusion . We note an excellent linear fit to in each case , and an excellent match between the estimated and for the Browmian diffusion trajectories . For the 2-state system described above , as the track length increases , approaches an effective diffusion coefficient , , defined as the weighted average of the diffusion coefficient in each state . For a sufficiently long track , or when averaging over multiple tracks , the particle is in state 1 for a fraction of steps , and in state 2 for a fraction of steps . Thus , the expected value of is ( 10 ) The slope of a linear fit to for the 2-state tracks in Figure 2B is indeed this weighted average for the chosen set of parameter values . This is a good descriptor for the overall mobility of a 2-state particle , but it does not reveal the underlying diffusion coefficients and their relative contributions . We now describe a likelihood maximization scheme to identify these parameters by fitting particle tracks to a 2-state hidden Markov model . The 2-state HMM is characterized by two diffusion coefficients and two transition probabilities . We parametrized the model by the parameter set , and sought to calculate the likelihood of , for an observed particle track ( 11 ) where represents a particular state sequence of the Markov chain . The probability of observing the state sequence depends only on the two transition probabilities , whereas , for that state sequence , the probability of observing the track depends only on the two diffusion coefficients . Because the possible number of state sequences grows exponentially with the number of steps in a track , a direct calculation using the above equation is computationally prohibitive . However , the forward-backward algorithm [25] , [26] efficiently calculates this probability by recursively evaluating the forward variable , defined as the probability of observing the partial sequence of steps up to step , and being in state at step , given the model parameters : ( 12 ) The probability of observing a track for a given choice of the parameters is ( 13 ) As described in Materials and Methods , we used a modified version of the forward algorithm to calculate the log likelihood of the parameter set for an observed set of particle tracks ( Algorithm 2; Fig . 3 ) . We then maximized this log likelihood with respect to the four model parameters to calculate their most likely values for a given set of tracks . We used a Markov Chain Monte Carlo ( MCMC ) algorithm ( Algorithm 3; Fig . 4 ) to maximize the log likelihood function [26] . While it is computationally less efficient than traditional gradient-based maximization schemes , this algorithm is less liable to be stuck in a local maxima because of stochastic downhill steps . Moreover , by sampling the log likelihood landscape around the maxima , this algorithm establishes the measure of uncertainty in each parameter estimate . Fig . 5 ( A and B ) show a typical MCMC trajectory for fitting a set of simulated 2-state particle tracks to a 2-state HMM . There is an initial “burn-in” phase , indicated by the shaded region containing the first 20000 MCMC steps , during which the log likelihood increases nearly monotonically as the trajectory converges toward a maximum in log likelihood . After this burn-in phase , the log likelihood value and the parameter estimates maintain relatively steady values with small stochastic fluctuations . The distributions of parameter estimates from the MCMC optimization are shown in the histograms in Fig . 5C and D . We report the mean of each parameter distribution as the maximum likelihood parameter estimate and use the coefficient of variation ( CV ) to quantify the uncertainty in this estimate . We assessed the MCMC parameter optimization scheme for a range of parameter values , using an ensemble of simulated tracks for each parameter set . The results , summarized in Table S1 ( Text S1 ) , include the maximum likelihood parameter estimates and their relative deviations from the true parameter values . For all but one parameter combination we tested , the maximum likelihood parameter estimates are remarkably close to their true values , with relative errors that are typically less than 10% . The error and dispersion in the parameter estimates are most appreciably affected by the relative magnitude of the two diffusion coefficients . In particular , as the two diffusion coefficients approach each other , the estimates of transition probabilities are progressively more error-prone and errors of as much as 70% arise . Notably , the magnitude of transition probabilities , either relative to each other - simulating a preferred state - or when they are uniformly high - simulating a frequent turnover of the particle between the two states - had only a minimal effect on the overall reliability of parameter estimates . We also tested the effects of varying the track length on the accuracy and variability of estimated parameters ( Fig . S1 , Text S1 ) . As expected , both relative errors and dispersions in the parameter estimates decreased with an increasing number of frames . In Fig . 5 ( E and F ) , we plot another measure of dispersion in parameter estimates , namely , the span of a 95% coverage of the parameter distributions , which reveals any assymmetry in the parameter distributions . For fixed values of , and , but varying ( corresponding to the first four parameter combinations in Table S1 ) , we observe increasing error and dispersion as approaches . These trends arise because the log likelihood algorithm attempts to classify each displacement as arising either from or , using equation 19 . This classification is increasingly error-prone as the two states become indistinguishable , resulting in the errors seen in Table S1 and Fig . 5 ( E and F ) . These results suggest that , if the maximum likelihood estimates of the two diffusion coefficients differ by less than two-fold , then the 2-state HMM is a poor descriptor of the system and parameter estimates ( especially the transition probabilities ) should be interpreted cautiously . The most commonly used analysis of single particle trajectories is to extract a diffusion coefficient from a linear fit to their mean squared displacement ( MSD ) over time [2] . Typically , a macroscopic diffusion coefficient , , that captures the particle behaviour on a time scale of seconds is calculated . Heterogeneities in the distribution of reveal multiple subpopulations of diffusing particles , and their relative contribution [22] . We used simulated particle trajectories to directly compare an MSD-based analysis with a 2-state HMM analysis over a range of frame rates , acquisition times and simulation parameters . Typical results are summarized in Fig . 6 ( A–D ) , with the output of analysis shown on the left ( Fig . 6A , C ) and the output of a 2-state HMM analysis shown on the right ( Fig . 6B , D ) . For simulated 2-state trajectories , we note that the distribution of values is more dispersed than the individual distributions of and ( Fig . 6A and B ) . Further , peaks of the two subpopulations constituting the distribution do not accurately report the diffusion coefficients of the two underlying states . In contrast , the HMM analysis is less error-prone and yields sharper parameter distributions . Moreover , the distribution of does not reveal the kinetics of the transition between the two states . Finally , we note that when trajectories are simulated with only a single underlying state , the analysis shows spurious subpopulations with peaks flanking the true value of the single diffusion coefficient ( Fig . 6 C ) , whereas the HMM analysis correctly reports a near complete overlap in the distributions of and , consistent with only a single identifiable diffusion coefficient ( Fig . 6 D ) . These results offer additonal validation of the proposed HMM analysis for accurate resolution of 2-state dynamics that are not well-discerned with an MSD-based analysis of particle tracks . To test the applicability of the 2-state HMM described above , we analyzed a set of experimental SPT data for the T cell integrin , LFA-1 . LFA-1 is critical for lymphocte adhesion and signaling , and has been previously studied using both SPT [22] , [27]–[29] and FRAP techniques [30] , [31] . In studies of LFA-1 lateral mobility on T cells , it has generally been observed that receptor diffusion is highly dependent upon cytoskeletal contacts . These interactions have manifested themselves in large immobile fractions and reduced diffusion coefficients . In previous work by Cairo et al . , SPT experiments showed heterogeneous LFA-1 dynamics , with two apparent populations of diffusion coeffients [22] . The relative contributions to LFA-1 mobility from these two subpopulations were found to vary according to changes in the conformation of LFA-1 and the activation state of T cells . We sought to better understand the heterogeneity present in these experiments by analyzing them with the 2-state HMM model . A typical distribution of the most likely values of and for one set of experiments is shown in Fig . 6 F , alongside the previously identified distribution of values segmented into the two subpopulations ( Fig . 6 E ) . As was the case for simulated particle trajectories , the distribution of for LFA-1 is more dispersed with a significant overlap between the two subpopulations , compared to the distributions of and from the HMM analysis . However , it must be noted that unlike simulated trajectories , experimental particle tracks are subject to greater intrinsic variability arising from differences between individual cells . It is likely that this cell-to-cell variability is partly responsible for the observed dispersion in values , whereas the maximum likelihood parameter estimates from the HMM analysis essentially ignore this variability . Thus , for experimental particle tracks , the well-resolved peaks in the estimates of the diffusion coefficients ( Fig . 6 F ) should be interpreted as their most likely values over the population of cells analyzed , while an MSD-based analysis should be used to gauge the variability within the population . We applied the 2-state HMM analysis to the data set of LFA-1 particle trajectories observed on T cells by Cairo et al . [22] . In these experiments , LFA-1 was labeled with either its cognate ligand ICAM-1 , or an antibody , TS-1/18 , known to block adhesion , and LFA-1 tracks were observed on resting cells , or those perturbed by various pharmacological agents ( Fig . 7 ) . Maximum likelihood parameter estimates for the 2-state model are reported in Table 1 . In addition to these model parameters , we also list the stationary probabilities for the two states , a pseudo equilibrium constant for the first order reaction ( equation 2 ) , and an effective diffusion coefficient , ( equation 10 ) , that captures the overall LFA-1 mobility for each set of particle tracks . The reported in Table 1 are nearly identical to values calculated using equation 8 , indicating that these two measures of the overall mobility of a particle are consistent with each other , and may be used interchangeably . We note that for all the experiments analyzed here , the maximum likelihood estimate of is at least double that of , and typically greater by five-fold or more . This separation suggests relatively small errors in the parameter estimates ( ) , based on our tests of this analysis with simulated tracks of comparable length and sampling interval . Dispersions in the parameter distributions compare favourably with those for simulated tracks , with CV<2% for the two diffusion coefficients and CV<15% for the two transition probabilities . With the exception of ICAM-1-ligated LFA-1 in phorbol-12-myristate-13-acetate ( PMA ) -treated cells , the estimated value of was , most likely capturing the diffusion of LFA-1 on the plasma membrane with relatively little interaction with the cytoskeleton . We observe a much greater variability in the estimates of , with values spanning nearly an order of magnitude , consistent with an active engagement between LFA-1 and the actin cytoskeleton in this state , thus rendering it susceptible to factors that affect this interaction , such as cytochalasin D treatment , or PMA-induced activation . We observed that in untreated cells , ICAM-1 ligation reduces the overall mobility of LFA-1 , compared to TS-1/18-labeled LFA-1 , as assessed by the value for the two experiments ( Table 1; cf . rows 1 and 2 ) . This is consistent with the previously reported results using an MSD analysis [22] , but the HMM analysis additionally reveals that the reduced mobility is primarily due to a two-fold decrease in , and not due to an increased fraction of time spent in the bound state . The decrease in suggests that upon interaction with ICAM-1 , the integrin may bind to an additional cytoskeletal-binding protein or could increase the number of cytoskeletal contacts as part of a cluster resulting in reduced mobility [32] . Treating cells with cytochalasin D reduces the lifetime of the bound state , with approximately 40% smaller values compared to untreated cells ( Table 1; cf . rows 1 and 3 , and rows 2 and 4 ) . Interestingly , this altered distribution between the two states is not reflected in a consistent trend in the overall mobility: is virtually unchanged upon cytochalasin D treatment for the TS-1/18 label , but increases by nearly 20% for ICAM-1-treated cells . The difference arises because is affected by changes in both the two diffusion coefficients , as well as the relative lifetimes of the two states ( equation 10 ) . In this specific case , a marginal decrease in offsets the shift in the equilibrium to that state for TS-1/18-labeled LFA-1 such that the overall mobility is essentially unaltered upon cytochalasin D treatment . In contrast , for ICAM-1-ligated LFA-1 , both diffusion coefficients increase upon cytochalasin D treatment ( by nearly 10% , and by over 25% ) , resulting in an increase in overall mobility . These results illustrate a significant advantage of the 2-state HMM analysis in its ability to capture subtle changes in multiple biophysical parameters , compared to an MSD-based analysis that only captures the overall mobility . PMA-induced activation of T cells lowered relative to its value in untreated cells , by over 8-fold for the TS-1/18 label ( Table 1; cf . rows 1 and 5 ) , and by nearly 2-fold for the ICAM-1 label ( Table 1; cf . rows 2 and 6 ) , albeit with important differences between the two cases . For cells labeled with TS-1/18 , the reduced mobility of the bound state is offset by a shift in the equilibrium toward the free state , resulting in no net change in the overall mobility . In contrast , when LFA-1 is ligated with ICAM-1 , and the cells are stimulated with PMA , the mobility of both free and bound LFA-1 are reduced and concurrently , there is a shift in the equlibrium toward the bound state , as seen by a two-fold increase in . In combinations , these two factors dramatically lower the overall LFA-1 mobility resulting in the lowest value across all the experiments analyzed here . Notably , the combination of ICAM-1 ligation and PMA-induced activation also increases both the transition probabilities , by nearly five-fold and by over two-fold , relative to ICAM-1 ligation alone ( Table 1; cf . rows 2 and 6 ) . PMA-activation alone however reduced these transition probabilities relative to their values in resting cells labeled with TS-1/18 , as well as decreasing by nearly 10-fold . The transition probabilities are related to the on and off rates of the LFA-1 interaction with its cytoskeletal binding partners ( equations 3 and 4 ) . With the improved resolution of the HMM analysis , we can thus discern subtle regulatory mechanisms for the integrin receptors . It is clear that LFA-1 is tightly regulated by a dynamic interaction with its cytoskeletal binding partners . The effective diffusion of the receptor is likely controlled by altering the specific binding partner , or the on- or off-rates of the interaction . We see evidence for both these putative mechanisms: the decrease in upon PMA-induced activation suggests that a different binding partner may be involved , whereas the increased transition probabilities upon the combination of ICAM-1 ligation and PMA treatment suggest that the turnover rate between the two states is altered . Thus , activation of the cell can alter either of the resolved diffusion coefficients or modify the equilibrium between the bound and free state . Together , these findings support the view that LFA-1 diffusion is a complex and dynamic process that integrates multiple biochemical cues , such as cellular activation , binding partner and conformational state , to influence T cell adhesion . The hidden Markov formulation that we used to analyze single particle tracks also allows us to identify the most likely state of the Markov chain at each step along a track . To achieve this , the forward-backward algorithm defines a backward variable ( 14 ) that is the probability of observing the partial track conditional on the particle being in state at the step and on model parameters . Therefore , the ( unnormalized ) probability of the particle being in state at step , for an observed track conditional on is ( 15 ) where is defined in equation 12 . The state that maximizes this probability is the most likely state . We modified the recursive definition of above for our likelihood-based calculation , as described in Algorithm 4 ( Fig . 8 ) , and estimated the most likely particle states for a given track , using the maximum likelihood parameter estimates , , for the calculation . We tested the performance of the segmentation algorithm for simulated trajectories that were previously used to assess the performance of the likelihood maximization algorithm ( Table S1 ) . For each set of trajectories , we used the maximum likelihood parameter estimates to identify the sequence of most likely particle state at each point along each track , and compared the prediction with the true identity of that state . Not surprisingly , the accuracy of track segmentation was strongly dependent on the accuracy of the maximum likelihood parameter estimates , and in turn on the separation between the two diffusion coefficients . When the diffusion coefficients differed by two-fold or greater , we could typically identify the true particle state more than 80% of the time . A representative simulated track , color-coded to identify the particle state at each point , is shown in Fig . 9 A , alongside the true and predicted state sequences for the trajectory depicted with state-sequence “barcodes” for an easy visual assessment of the segmentation . We applied the trajectory segmentation algorithm to LFA-1 particle tracks analyzed with a 2-state HMM . A selection of segmented LFA-1 particle tracks is shown in Fig . 9 B . We noted that for a majority of the observation period ( 4 s ) the particles were found in a single state , suggesting relatively slow switching kinetics on the time scale of these experiments . To further classify the behavior of individual trajectories , we calculated the total number of state transitions during the 4 s data acquisition period , and the fraction of that time during which a particle was in the bound state ( Fig . 10 ) . The overall mobility of an individual particle decreased with increasing fractions of time in the second state , consistent with the smaller diffusion coefficient of the second state . Interestingly , these plots reveal that on the time scale of these experiments a majority of the particles were predominantly in a single state , and only a small number of trajectories had frequent state switches ( Fig . 10B ) . This result is consistent with the generally small transition probabilities , typically , for this system ( Table 1 ) . It could also explain the relatively greater dispersion in the transition probability estimates reported here , as a substantial number of state switches would be required to estimate the transition probabilities accurately . We now address the question of how to determine whether a 2-state model is indeed the best descriptor for the observed data , given one or more alternate models . We compared different models by means of Akaike's information criterion ( , equation 21 ) and the associated Akaike weights ( equation 23; see Text S1 for details ) . We fitted simulated trajectories for pure Brownian diffusion with a 2-state model , and noted that for the maximum likelihood parameter estimates obtained in that case , the 2-state model effectively collapses to a single-state diffusion model ( Fig . S2 ) , that is preferred by the Akaike criterion . In contrast , when the trajectories are simulated from a 2-state process , the 2-state HMM outperforms a simpler 1-state model . Notably , for all LFA-1 trajectories analyzed here , the 2-state model is overwhelmingly preferred based on the Akaike criterion ( data not shown ) , thus indicating the suitability of this model over a single state model to capture LFA-1 dynamics . To determine whether a 2-state model is sufficient to describe the data , we attempted to further resolve the two states into component “sub-states” . After the intial segmentation of an ensemble of trajectories , we assembled all the displacements ascribed to into a single trajectory , and likewise , all the diplacements ascribed to into another trajectory . The two resulting trajectories were then further analyzed with both a 1-state and a 2-state model . We found that it is indeed possible to further resolve the each of these trajectories with a 2-state model ( Fig . S3 ) , suggesting some heterogeneity in the two states originally identified . Importantly however , the separation between the diffusion coefficients of the sub-states is much smaller ( approximately a factor of two ) relative to the separation between the diffusion coefficients of the two original states ( greater than an order of magnitude ) . As noted above , a small separation between the two diffusion coefficients implies that a 2-state model is an unreliable descriptor of the data . Thus , we conclude that our initial resolution of the data into two component states is sufficent to characterize the experimental trajectories . When this procedure was applied to an ensemble of simulated trajectories generated using the maximum likelihood parameter estimates for the data , we found that for the two virtual trajectories , the 2-state model effectively collapsed to a 1-state model .
In this study , we examined single particle trajectories for a membrane-associated protein that interacts with cytoskeletal binding proteins . Adhesion proteins at the cell membrane regulate a variety of biological phenomena including inflammation and antigen-presentation . Using a hidden Markov formulation to model 2D trajectories of a membrane protein , we outlined a systematic and easily-implemented procedure to parameterize a two-state model of diffusion and binding . Parameter estimates for this model can be used to identify the most probable state at each frame of the trajectory and thus divide it into mobile and immobile fragments . To establish the applicability of this analysis , we rigorously tested it with simulated trajectories for a range of parameter values . The HMM analysis revealed the diffusion coefficients of the individual states and identified transient state changes within single trajectories . Hidden Markov models have been previously used to analyze actomyosin and kinesin-microtubule movement data [36] , [37] , and DNA looping kinetics [38] in single-molecule microscopy experiments , but not to our knowledge , to analyze the lateral diffusion of membrane proteins . Thus , we have developed a novel methodology to analyze and interpret single particle trajectories of cell-surface molecules . Our method expands upon the standard MSD analysis for SPT experiments , and provides previously inaccessible information about hetereogeneous diffusion . We are able to confidently detect the presence of two diffusion coefficients ( and ) , the transition probabilities for switching between these states ( and ) , and an apparent equilibrium constant based on these probabilities ( ) . In previous studies of LFA-1 diffusion , a population-based MSD analysis was used to infer the presence of multiple states of diffusion . Our new analysis reveals that there are indeed two states responsible for the lateral-mobility of LFA-1 , and that individual trajectories show a mixture of both states ( Fig . 9 ) . We are able to resolve the detailed state-switching behaviour of individual trajectories ( Fig . 10 ) . These values are accessible only in the aggregate using an MSD analysis , therefore , the method described here provides a new window into single-molecule experimental data . As noted above , the parameters provided by the HMM are inaccessible to a standard MSD analysis , and may be used to resolve changes in the identity or rate of specific interactions through changes in diffusion coefficients and transition probabilities , respectively . We made two key simplifying assumptions: first , that the particle transitions between the two states with first order kinetics , and second , that all transitions occur at the sampling time . First order kinetics are justifiable when there is an excess of binding sites , but without direct experimental data , it is difficult to judge the merit of this assumption . Thus , the transition probabilities reported here must be interpreted with care , as they depend on , and therefore on the equilibrium substrate concentration , . This caveat is especially important if transition probabilities reported here are used to derive first order on and off rates by solving equations 3 and 5 for and . Nonetheless , given a measurement of the substrate concentration , and assuming that it doesn't change dramatically over the course of the 4 second particle track , our method could be used to estimate the true bimolecular on-rate for the interaction . The assumption that transitions in the particle state occur on order of the sampling time is more easily justified in light of the relatively low transition probabilities that we observe ( less than once every 100 frames ) . For infrequent transitions relative to the frame rate , the exact transition moment should not significantly alter our analysis . The validity of this assumption must be checked a-posteriori for a given experimental setup , by confirming that the transition probabilities are indeed small ( ) for the chosen frame rate . We plan to expand our analysis to the more general case when the transition rates are comparable to the acquisition frame rates and the transitions occur at intermediate times . Our analysis offers some distinct advantages over an MSD-based approach . Firstly , by examining the diffusive behaviour of a particle at each step along a trajectory , heterogeneous diffusion is efficiently resolved . Secondly , unlike the distribution of from an MSD analysis , the distributions of HMM parameter estimates quantify not only the diffusion coefficients of the underlying states , but also the kinetics of transitions between them . With some notable exceptions [39]–[41] , these kinetic parameters are typically inaccessible in traditional analyses of SPT ( or FRAP ) experiments . There is mounting evidence that interprotein interactions affect the mobility of membrane proteins [42] , [43] , and some progress has been made toward modelling these effects [44] . In our analysis , we explicitly considered the effect of a binding interaction on the local diffusive behaviour of a molecule at short time scales and inferred the most likely parameter estimates for this interaction . Of the two states identified in our analysis , the one with greater mobility ( ) is most likely the freely diffusing form of LFA-1 , with minimal interactions with intracellular proteins . This interpretation is well-supported by the relatively consistent value of observed across a variety of experimental conditions ( Table 1 ) . The state with low mobility ( ) , reported here and in a previous study [22] , is likely to be either an actin cytoskeleton-associated form of LFA-1 , or part of an integrin-associated signaling cluster that is slowly diffusing . Association with the actin cytoskeleton is strongly supported by the nearly twofold reduction in the pseudo-equilibrium constant , , upon cytochalasin D treatment ( Table 1 ) . As well , a majority of the trajectories in cytochalasin D treated cells , are found predominantly in the high mobility state and exhibit very few state transitions ( Fig . 10 ) , suggesting that continued actin polymerization is required for maintaining the cytoskeletal attachment . Though the specific molecular mechanisms responsible are not fully understood , there is considerable evidence for a tightly regulated interaction between integrin receptors and the actin cytoskeleton , mediated by cytoskeletal proteins such as talin [45] , [46] . Our technique thus offers the potential to resolve and quantify these interactions using SPT data for LFA-1 . We note that , the values of diffusion coefficients reported here are influenced by the use of a micron-sized bead to label the protein . The potential effects of a bead on the mobility of a membrane protein are discussed in reference [2] , and include , enhanced drag due to the interaction between the bead and the extracellular matrix , and possible artifacts from crosslinking of the protein by the antibodies used . Nonetheless , the use of a bead allows for imaging at the high frame rates used in these experiments ( 1000 frames/s ) , thus exposing the transient state switching behavior that occurs on these short time scales . Our analysis also assumes that the binding partner is homogeneously distributed , such that the transition probabilities have no spatial dependence . In this respect , it differs notably from another class of SPT analysis that has been used to resolve transient spatial confinement of particles [19]–[21] . Spatial confinement typically arises from the preferential partitioning of cell-surface receptors into or out of membrane microdomains . Such trapping or exclusion has been directly visualized for T cell signaling molecules with respect to CD2-enriched domains [43] and CD9 with respect to tetraspanin-enriched areas ( TEA's ) [41] . In another study , analysis of SPT data for a G-protein-coupled receptor showed evidence for confinement within domains that were themselves slowly diffusing ( termed as “walking confined diffusion” ) [42] . Our analysis does not directly resolve spatial confinement , but instead resolves heterogeneity in the temporal behavior of a diffusing particle . For sufficiently small confinement regions that are relatively uniformly distributed , the slow diffusing state in our model may indeed reflect the passage of a particle through such a confinement zone . But it is difficult to make such a conclusion in the absence of a secondary label used to visualize the membrane heterogeneity . We have tested simulated 2-state trajectories and experimental LFA-1 trajectories using the spatial confinement algorithms described previously [19] , [20] , but do not find any consistent patterns between the temporal state-switching in our analysis and spatial confinement as identified by these algorithms ( data not shown ) . This is not surprising , because these algorithms requires a clear separation between the macroscopic and microscopic diffusion coefficients for effective detection of confinement , and such separation is rarely observed in the LFA-1 data [22] . The LFA-1 trajectories were acquired with a very high frame rate ( 1000 frames/s ) , but for a relatively short interval ( 4 s ) . Consequently , these data are best suited for analyzing the behaviour of LFA-1 on a short time scale . This is in contrast with the typical acquisition rates of 30 frames/s or slower and acquisition times of tens of seconds that were used for the other studies cited above . These longer acquisition times allow the molecules to sample putative confinement regions and are therefore better suited to effectively distinguish short term diffusive behavior from long term confinement . In general , analyzing spatial heterogeneity in mobility with the HMM formulation would require substantially more complex models than the one presented here , as the transition probabilities themselves would vary with the location of the particle . Additional complexity would be introduced by variations in the size of confinement regions . In future studies , we intend to examine modifications to our model that rigorously address these issues . A notable advantage of the present analysis is the lack of any user-tuned parameters , such as a characteristic confinement length ( ) or a minimum segment length ( ) , used in previous studies [19] . These parameters may vary for different experimental systems and their judicious choice is essential for succesfully detecting spatial confinement . In contrast , our analysis is directly applicable to a variety of experiments without requiring significant modification from its current form . However , we note that it may be possible to extract equivalents of the confinement length or other parameters from the results of the HMM analysis . Finally , the likelihood-based approach that we adopted here is flexible and can be extended to account for other modes of motion . We tested a two-state Brownian model in this work , but the HMM approach could be used to introduce additional states or alternative models of mobility , such as directed motion . This approach has the potential to resolve extremely complex and heterogeneous trajectories . The use of likelihood as a metric for the quality of a model allows for statistically well-defined comparisons between various models , using , as described here , and other tests described elsewhere [47] . In summary , we believe that fitting particle tracking data to a well-defined model and using likelihood maximization to estimate model parameters is a natural and powerful tool for inferring and quantifying the spatiotemporal dynamics of cell surface proteins .
Experimental LFA-1 trajectories used were acquired as described in Cairo et al [22] . Briefly , 1 micron beads were labeled with either an adhesion protein ( ICAM-1 ) or a Fab fragment of an LFA-1 binding antibody ( TS1/18 ) . The beads were then blocked to prevent non-specific binding , and Jurkat T cells ( clone E6 . 1 , ATCC , Manassas , VA , USA ) were labeled with beads and observed using video microscopy [48] . Cells were treated with HBSS buffer containing either a vehicle control ( DMSO ) , phorbol-12-myristate-13-acetate ( PMA ) , cytochalasin D ( cytoD ) , or calpain inhibitor-I ( cal-I ) . Trajectories were collected on live cells at 1000 FPS ( 1 ms ) and converted to trajectories using Metamorph ( Universal Imaging , Downington , PA , USA ) . Data were analyzed by either an MSD algorithm combined with a population analysis [22] or by the HMM method described here . For a particle undergoing Brownian diffusion in a space with a diffusion coefficient , the probability density of observing a displacement after a time interval is given by: ( 16 ) In this study we are concerned with single particle tracks of a membrane-associated protein that is imaged at fixed time intervals . Thus , and is the frame rate at which the particle is imaged . A simulated track therefore consists of successive displacements , with the displacement along each dimension distributed normally , with mean 0 and variance . To simulate Brownian diffusion , we used the Matlab function normrnd to generate such a sequence of displacements and then cumulatively summed them to calculate the particle coordinates . To simulate trajectories for a particle with 2-state diffusion we first generated a Markov chain where denotes the state of the particle at the time point . The Markov transition matrix ( 17 ) is composed of the probabilities and for transitions between the two states . The Markov chain was simulated using Algorithm 1 ( Fig . 1 ) . The particle displacements were then drawn randomly from a normal distribution with 0 mean and a variance . A particle trajectory consists of a sequence of individual displacements , denoted as , where , . We calculated the log likelihood , , of parameter values for a particle track as described next . First , we defined the likelihood of a diffusion coefficient for an individual displacement as ( 18 ) and the corresponding log likelihood as ( 19 ) where . The proportionality in the first equation arises because the likelihood function is only defined up to an arbitrary multiplicative constant . Likewise , the log likelihood function is only defined to an arbitrary additive constant , and in our definition ( equation 19 ) we only retained terms that contain an explicit dependence on model parameters , ignoring coefficient such as . The log likelihood of the parameters for a sequence of displacements was calculated using Algorithm 2 ( Fig . 3 ) , which is a modified version of the forward-backward algorithm [24] , [49] . Finally , the log likelihood function for an ensemble of independent trajectories , , is simply the sum of the log likelihood function evaluated for each trajectory . ( 20 ) To estimate the maximum likelihood parameters of a 2-state HMM for a set of tracks , we used a stochastic Markov Chain Monte Carlo ( MCMC ) optimization scheme ( Algorithm 3; Fig . 4 ) . This algorithm assigns random initial values to all the parameters and iteratively traverses the parameter space through a succession of small displacements along each parameter axis . For each proposed displacement , the log likelihood function is evaluated at parameter values after the displacement and compared to the log likelihood for the current parameter values . A proposed displacement is accepted or rejected using a Metropolis rejection scheme: any proposed displacement that increases the log likelihood from its current value is accepted , but a proposed displacement that decreases the log likelihood from its current value is only accepted with a probability equal to the fractional change in the likelihood function after the proposed move . Typically the MCMC runs were steps long with an initial burn-in phase during which the MCMC trajectories approach an equilibrium . The scales of displacement , , were adjusted to achieve an acceptance rate of 20–40% along each parameter axis after the burn-in phase . The acceptance rate is defined to be the ratio of number of accepted moves to the total number of proposed moves along a parameter axis during the MCMC run . The sample means of the MCMC trajectories , after excluding the burn-in phase , were reported as the maximum likelihood parameter estimates . We also calculated the coefficient of variation ( CV ) , the ratio of the sample standard deviation to the sample mean , to measure the variability of the parameter estimates . We define as the set of maximum likelihood parameters for a given track and use a modified version of the forward-backward algorithm to estimate the most likely state , of the Markov chain at each step along the track ( Algorithm 4; Fig . 8 ) . To compare the effectiveness of different models in describing a set of tracks , we used the Akaike information criterion ( ) , defined as ( 21 ) where is the log likelihood function of the maximum likelihood parameter set for a model with parameters , given independent observations . Here , is the number of individual displacements in the trajectory . To interpret the values for different models , we use the rescaled values , defined as ( 22 ) where is the minimum value among all models under consideration , that is , . Each model is then assigned an Akaike weight ( 23 ) that measures the relative evidence in its favour . The sum in the denominator is over all the models under consideration [47] .
|
Many important biological processes begin when a target molecule binds to a cell surface receptor protein . This event leads to a series of biochemical reactions involving the receptor and signalling molecules , and ultimately a cellular response . Surface receptors are mobile on the cell surface and their mobility is influenced by their interaction with intracellular proteins . We wish to understand the details of these interactions and how they are affected by cellular activation . An experimental technique called single particle tracking ( SPT ) uses optical microscopy to study the motion of cell-surface receptors , revealing important details about the organization of the cell membrane . In this paper , we propose a new method of analyzing SPT data to identify reduced receptor mobility as a result of transient binding to intracellular proteins . Using our analysis we are able to reliably differentiate receptor motion when a receptor is freely diffusing on the membrane versus when it is interacting with an intracellular protein . By observing the frequency of transitions between free and bound states , we are able to estimate reaction rates for the interaction . We apply our method to the receptor LFA-1 in T cells and draw conclusions about its interactions with the T cell cytoskeleton .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"mathematics/statistics",
"biophysics/experimental",
"biophysical",
"methods",
"biophysics/theory",
"and",
"simulation",
"immunology/leukocyte",
"activation"
] |
2009
|
A Hidden Markov Model for Single Particle Tracks Quantifies Dynamic Interactions between LFA-1 and the Actin Cytoskeleton
|
Accumulating evidence indicates that paternal age correlates with disease risk in children . De novo gain-of-function mutations in the FGF-RAS-MAPK signaling pathway are known to cause a subset of genetic diseases associated with advanced paternal age , such as Apert syndrome , achondroplasia , Noonan syndrome , and Costello syndrome . It has been hypothesized that adult spermatogonial stem cells with pathogenic mutations are clonally expanded over time and propagate the mutations to offspring . However , no model system exists to interrogate mammalian germline stem cell competition in vivo . In this study , we created a lineage tracing system , which enabled undifferentiated spermatogonia with endogenous expression of HrasG12V , a known pathogenic gain-of-function mutation in RAS-MAPK signaling , to compete with their wild-type counterparts in the mouse testis . Over a year of fate analysis , neither HrasG12V-positive germ cells nor sperm exhibited a significant expansion compared to wild-type neighbors . Short-term stem cell capacity as measured by transplantation analysis was also comparable between wild-type and mutant groups . Furthermore , although constitutively active HRAS was detectable in the mutant cell lines , they did not exhibit a proliferative advantage or an enhanced response to agonist-evoked pERK signaling . These in vivo and in vitro results suggest that mouse spermatogonial stem cells are functionally resistant to a heterozygous HrasG12V mutation in the endogenous locus and that mechanisms could exist to prevent such harmful mutations from being expanded and transmitted to the next generation .
In order to propagate genetic information to the next generation with high fidelity , germline cells must maintain a low mutation rate . Nevertheless , maternal germline cells ( human oocytes ) are well known to transmit abnormal chromosomes to offspring , especially in advanced maternal age ( reviewed in [1] ) . Surprisingly , recent high-throughput genome analyses have revealed that men contribute a much higher number of mutations , specifically de novo single nucleotide mutations , to their children than do women [2–4] . Most strikingly , the risk of certain genetic disorders increases with advancing age of the father at the time conception of the child , referred to as the paternal age effect ( PAE ) . This phenomenon could be explained by the unique biology of paternal germline stem cells . The latter are termed spermatogonial stem cells ( SSCs ) , and , once established in the post-natal period , continue to self-renew and differentiate to supply sperm in mammals throughout adult life . This continuous self-renewal and long-term survival of SSCs may underlie the increase in mutation burden with paternal age , due to a cumulative increase in copy errors or other DNA lesions , despite the fact that the baseline germline mutation rate is thought to be lower than that of somatic cells [5] . Although the natural history of mutations in the aging testis is poorly understood , pathogenic variants are occasionally transmitted to offspring , resulting in a wide range of disorders . Among these , de novo gain-of-function mutations in the growth factor receptor-RAS signaling pathway are classically known to cause so-called PAE disorders , such as Apert syndrome , achondroplasia , Noonan syndrome , and Costello syndrome ( reviewed in [6] ) . Direct quantification of such mutations in the sperm and testes of healthy men of different ages has revealed an age-dependent increase in the mutation burden , in a manner that exceeds what would be expected from cumulative copy errors [7–9] . Moreover , in human testes , Ras pathway-associated mutations have been reported to occur in a clustered manner , suggesting that SSCs with PAE mutations are positively selected and clonally expand in normal , otherwise healthy testes over time [10–12] . We previously showed that a gain-of-function mutation in FGFR2 that causes Apert syndrome is sufficient to confer a selective advantage to murine SSCs in vitro [13] . However , no model system has been developed to interrogate mammalian SSC competition in vivo . Furthermore , no cell biological or molecular mechanisms have been described to explain this phenomenon . Although clonal expansion of stem cells with oncogenic mutations has been observed in the mouse intestinal crypt model [14 , 15] , it is not clear whether the same holds true for SSCs in the adult mouse testis . To test this long-standing hypothesis for SSC competition , we sought to establish an inducible mosaic model in which a hyperactive form of Hras could be induced within the endogenous locus in a subset of SSCs so that their long-term fate could be followed . The undifferentiated spermatogonia ( Aundiff ) represent a population of cells in the mammalian testes that is defined by morphology and function . Along with somewhat more committed cells , the Aundiff pool contains long-term self-renewing SSCs . Morphologically , the Aundiff in rodents comprises As ( single ) , Apr ( pair ) , and Aal ( aligned ) cells , which are remarkably interconvertible , with significant migratory capacity and cell fate plasticity when subject to stress [16 , 17] . Those cells reside along the basement membrane in the seminiferous tubules and are heterogeneous with respect to expression of genetic markers . Hara et al . ( 2014 ) first employed a cre driver controlled by the endogenous promoter of Gfra1 , one of the robustly expressed markers for Aundiff and demonstrated that the labeled population marked by Gfra1-creERT2 comprised the long-term stem cell fraction [16] . Therefore , in our current study , we chose the same Gfra1 cre driver to create a novel germline mosaic model . HRAS , a member of the RAS oncogene superfamily , is a monomeric GTPase and relays signals from receptor tyrosine kinases to the cell interior . It serves as a molecular switch for a MAP kinase signaling module in which HRAS is “on” when GTP is bound and “off” when GDP is bound ( as reviewed in [18] ) . The HrasG12V mutation encodes a hyperactive form of HRAS protein that is locked in a GTP-bound state and cannot hydrolyze its bound GTP to GDP [19 , 20] . HrasG12V is a rare mutation found in patients with Costello syndrome , whereas the HrasG12S mutation comprises the majority of probands [21] , and it has been demonstrated that the Hras mutation burden in sperm from healthy donors increases according to donors’ age [9] . In mice , heterozygous HrasG12V mice phenocopy human Costello syndrome [22] . It was observed that overexpression of HrasG12V using transgenes in cultured mouse SSCs caused tumor development [23] . However , the effect of one copy of hyperactive Hras expression on paternal SSCs in vivo , simulating the putative earliest events in the human gain-of-function mutation disorder , has never been addressed . In order to understand how a hyperactive HRAS affects long-term paternal stem cell fate , we induced HrasG12V at the endogenous locus in Aundiff in a mosaic manner . We found that Gfra1-creERT2 successfully drives mosaic HrasG12V activation in the adult germline in vivo . This model system allowed us to track mutated cell fate for prolonged chase periods in a quantitative manner . Surprisingly , the mutated SSC fraction persisted stably without significant expansion , suggesting that robust mechanisms exist to protect the SSC pool from harmful expansion of mutated cells and prevent transmission of deleterious alleles .
Whereas cultured SSCs are reported to express Hras [23] , the abundance of Hras transcripts in the Aundiff spermatogonia in vivo has not been reported . Therefore , we evaluated Hras expression in vivo in Aundiff , as a surrogate for SSCs . To isolate the Aundiff population by flow sorting , we stained dissociated testicular cells with an anti-MCAM antibody ( Fig 1A and 1B ) . Although MCAM-based sorting has been previously shown to enrich the Aundiff population , MCAM is also expressed in somatic cells [24 , 25] . To avoid somatic cell contamination in FACS experiments , we first obtained tdTomato-labeled germ line cells using Gfra1-creERT2; tdTomfl/- mice , in which tamoxifen induces tdTomato efficiently in the Aundiff and their progeny but not in testicular somatic cells ( Fig 1G , iWT ) . Three months after tamoxifen induction , tdTomato+ cells that had high , medium , low , or absent ( negative ) expression of MCAM were isolated by FACS and analyzed for Hras by RT-PCR ( Fig 1A and 1B ) . Hras transcripts were found throughout the different germ cell populations tested ( Fig 1C ) . To induce the HrasG12V mutation exclusively in Aundiff , we employed the Gfra1-creERT2 mouse line . A “flox-and-replace” ( FR ) model generated by Chen et al . ( 2006 ) was utilized to produce monoallelic HrasG12V ( Fig 1D ) [22] . In this mouse line , the endogenous Hras locus consists of 2 tandemly-arrayed Hras genes , the upstream of which is a WT allele flanked by loxP sites , such that the downstream HrasG12V allele is silent until removal of the upstream 2 . 5 kb WT gene by recombination . By breeding , we generated Gfra1-creERT2; tdTomfl/-; FR-HrasG12Vfl/- mice . Tamoxifen-inducible Cre-mediated recombination of the Hras locus resulted in conversion from wild-type Hras to HrasG12V and activation of tdTomato expression at the Rosa26 locus . ( Fig 1D & 1G ) . To verify the presence of the mutation in the testes , cDNA was obtained from FACS-sorted tdTomato-positive testicular cells , which consisted of pure germline cells ( since the Gfra1 promoter is not active in testicular somatic cells ) and was amplified for sequencing . Sanger sequencing confirmed that the expected nucleotide substitution ( C>A ) was present ( Fig 1E ) . From the chromatogram , not all the tdTomato-positive cells exhibited HrasG12V three months after tamoxifen induction , indicating that wild-type and HrasG12V-positive germline cells coexisted in a mosaic manner in the labeled germline cell population . Importantly , Hras mRNA levels of Aundiff in vivo ( i . e . , tdTomato+/MCAM-high population in Fig 1A ) measured by qPCR were similar between induced WT and FR mice , validating that a physiological level of Hras expression ( WT + G12V ) was achieved in the induced FR ( iFR ) model by utilizing the endogenous locus ( Fig 1F ) . To measure the fractional abundance of HrasG12V-positive cells in germline cells , two independent methods were developed ( Fig 2A–2F ) . First , Sanger sequencing-derived plots of amplified cDNA from RNA were employed to quantify the mutated nucleotide ( cytosine→adenine ) from sequencing traces and create a model equation ( S1 Fig ) . The validity of the model was confirmed by amplifying cDNA samples of RNA of known standard concentrations from WT ( HrasWT/WT or HrasFR/WT ) and Costello syndrome ( HrasG12V/WT ) mice at varying ratios ( 0–100% Costello mRNA ) ( Fig 2A–2C ) . The relationship between the HrasG12V-positive cell fraction ( x% ) and the ratio of adenine/cytosine ( A/C ) peak heights ( y ) fitted the model curve ( y = 1 . 371*x / [200-x] ) . In the second approach , sperm genomic DNA ( gDNA ) analysis was used to quantify the HrasG12V-positive cell fraction . GDNA qPCR primers were designed to detect an SV40 poly-A ( PA ) region exclusive to the FR-HrasG12Vfl locus but which is lost after recombination ( Fig 2D–2F ) . By performing qPCR with mixed gDNA comprising heterozygotic FR-HrasG12V ( i . e . , without recombination ) and wild-type sperm at different known ratios , we confirmed a linear relationship between the FR-HrasG12Vfl recombined fraction ( x% ) and the SV40 PA allelic fraction obtained by qPCR ( y%; Fig 2E ) . Since the loss of SV40 PA or FR means a gain of HrasG12V , y = 100-z was obtained , where z is the HrasG12V-positive cell fraction ( Fig 2F ) . Using these two methods , HrasG12V-positive cell fractions of tamoxifen-induced animals were measured . The calculated fractions from sperm gDNA qPCR correlated well with that from the Sanger sequencing ( Fig 2G ) . Furthermore , we observed a tamoxifen dose-dependent increase in the HrasG12V-positive fraction in sperm gDNA ( Fig 2H ) . These results indicate not only that HrasG12V was successfully induced in Aundiff by tamoxifen but also that HrasG12V Aundiff can undergo differentiation and produce sperm . Previously , Chen et al . ( 2009 ) used their murine FR-HrasG12V allele to model human Costello syndrome offspring [22] ( Fig 3A left ) . In that study , FR-HrasG12V mice were crossed with Caggs-Cre ( CC ) mice to obtain HrasG12V heterozygotes , which phenotypically recapitulated Costello syndrome ( driven by HrasG12S ) . In the prior model , HrasG12V induction takes place after fertilization , rather than in the parental germ cells . On the other hand , our inducible system enabled us to test whether mosaic HrasG12V-positive Aundiff ( in a context of neighboring wild-type germ cells and somatic cells ) can differentiate into normal sperm and give rise to F1 offspring with Costello syndrome ( Fig 3A right ) . Tamoxifen-induced Gfra1-creERT2; tdTomfl/-; FRfl/- males ( n = 3 ) were crossed with C57Bl/6J females and their pups were sacrificed at post-natal day 0 and genotyped using RNA . Genotyping revealed that Costello offspring were indeed born ( left two pups in Fig 3B ) . This result demonstrates that HrasG12V-positive Aundiff are able to self-renew , undergo differentiation , and give rise to offspring . HrasG12V has been detected in human sperm of healthy donors and is allelic to HrasG12S , the most common mutation in Costello syndrome , which increases with sperm donor age [9] . Given that gain-of-function mutations in the RAS pathway mediate positive selection of human SSCs [26 , 27] and that overexpression of an HrasG12V cDNA is tumorigenic [23] , we hypothesized that an increasing burden of HrasG12V would be detectable over time following induction in the adult testis . To test this hypothesis , our inducible mutation model enabled us to follow the fate of HrasG12V-positive Aundiff over time . The proportion of HrasG12V-positive cells in the labeled germline population was assessed at different time points after tamoxifen administration ( Fig 4A ) . Sanger sequencing showed that the HrasG12V allele fraction did not increase from 3 to 14 months ( Fig 4B , left ) . Notably , despite the long chase period , no germ cell tumors were detected in these animals , and the testes were normal at the gross histological level ( see Fig 1G ) . Sperm mutation analysis serves as a more physiologically relevant readout , since human studies on mutation burden utilized sperm samples from different age individuals [7 , 9] . Thus , to mirror the human studies , we collected sperm from different time points following tamoxifen induction . Using qPCR , we quantified the HrasG12V-positive cell fractions in gDNA of sperm obtained from the cauda epididymis at these time points . Similar to the HrasG12V-positive cell frequency among tdTomato-labeled germ cells , the HrasG12V-positive sperm proportion also did not show an increase over the 11-month chase period ( Fig 4B , right ) . Sperm studies have demonstrated large inter-individual variance in the mutation burden in the same age cohort [7 , 9] . Therefore , such static observations require a large sample size , which is not always readily achieved . Furthermore , the time at which mutations first appear in the human testis is unknown , nor is the mutational load at relevant loci at the time of early adulthood . An ideal study would examine sperm serially from same individuals , in order to capture a temporal change in the proportion of the variant cell population using a relatively small cohort . If selection for the HrasG12V genotype were to occur , one would predict an increase in the same individual over time . Thus , we established a method for in vivo serial sperm sampling . We performed microsurgical epididymal sperm aspiration [28] , in which sperm was withdrawn twice from a live single male from the ipsilateral testis ( Fig 4C & S2 Fig ) . For this study , we employed the mTmG reporter mouse line , in which membrane tdTomato expression converts to membrane GFP ( mGFP ) upon cre-induced recombination [29] . We generated Gfra1-creERT2; mTmGfl/-; FR-HrasG12Vfl/- males and induced HrasG12V and mGFP in Aundiff ( Fig 4C ) . Sperm from wild-type siblings ( i . e . , Gfra1-creERT2; mTmGfl/- ) served as controls . In the control model , the recombined mTmG fraction obtained by measuring the loss of the tdTomato allele ( as a reference ) by qPCR should not change over time unless there is competition between tdTomato+ and GFP+ Aundiff . Both the HrasG12V and reference allelic fractions were measured by qPCR as above ( see Fig 2D ) . From 4 months to 12 months , a minimal increase ( ~9% on average ) in the HrasG12V-positive cell fraction was observed ( Fig 4E ) . Of note , the reference alleles in the wild type did not show any increase over the same time period , suggesting that tdTomato+ and GFP+ germ cells are mutually neutral with respect to fitness ( Fig 4F ) . To confirm whether the HrasG12V-positive cell fraction increases , we used a different method to assess the ratios of sperm bearing mGFP vs . tdTomato , resulting from Gfra1-expressing progenitors , in which the reporter transgene either remained intact or , alternatively , underwent recombination . The mGFP+ and tdTomato+ sperm were counted at two time points , at 4 months and 12 months , and the ratio of mGFP-positive to total sperm ( mGFP-positive plus tdTomato-positive ) was calculated . There was a strong positive correlation between values obtained by sperm counting vs . qPCR with the r2 = 0 . 90 ( p<0 . 0001 ) , validating that the sperm counting method itself was as accurate as the qPCR ( Fig 4D ) . Thus , if the HrasG12V-positive sperm increase over time , we should observe an increase in the GFP+ sperm population . However , no change in the fraction of mGFP+ sperm in the mutant was detected ( Fig 4G ) . Finally , we analyzed successive litters from tamoxifen-induced Gfra1-creERT2; tdTomfl/-; FR-HrasG12Vfl/- sires; no increase was observed in the proportion of HrasG12V/WT offspring obtained at the early phase of the study interval compared to the late phase ( S1 Table ) . Taken together , these data suggest that the population structure is relatively stable and do not support a substantial change in the ratios of controls and mutant cells within the mosaic pool . To test stem cell capacity of Aundiff containing the HrasG12V mutation , competitive SSC transplant assays developed by Kanatsu-Shinohara et al . ( 2010 ) were carried out [30] . In general , transplantation assays serve to measure cell autonomous capacity for self-renewal , survival , and differentiation . By mixing neutral , unlabeled competitor cells with either labeled experimental or labeled control donor cells , respectively , this method is designed to force the donor cells in question to compete against unlabeled competitors and allow direct functional comparisons between the two different donor types . Following two courses of tamoxifen treatment in vivo , the FR locus recombined at high efficiency , concurrent with the Rosa26-LoxStopLox tdTomato locus , indicating that most of the Aundiff were positive for HrasG12V and tdTomato ( see next paragraph below ) . From the previous result above ( see Fig 4B ) , this recombination ratio did not change over an extended period , suggesting that the number of initially induced HrasG12V-positive Aundiff remained stable by balancing self-renewal and differentiation for the rest of the time course . Therefore , the timing of testicular cell collection after tamoxifen ( i . e . , length of chase ) should not affect the size of Aundiff population . Based on these findings , we chose to perform the transplantation assay 2 weeks after the start of tamoxifen administration . Dissociated testicular cells from these tamoxifen-administered WT ( iWT ) and FR ( iFR ) were mixed with non-labeled wild-type competitor testicular cells from Gfra1-creERT2 mice at a ratio of 1:1 and microinjected into busulfan-treated recipient testes ( Fig 5A ) . After 10 weeks , tdTomato-positive colonies were counted ( Fig 5B ) . Colony numbers were comparable between iWT and iFR , suggesting that the short-term cell-intrinsic capacity of stem cells of HrasG12V-positive Aundiff is not greater than that of wild-type cells ( Fig 5C ) . Activating mutations in oncogenes such as RAS family members are thought to produce relatively robust changes in cell phenotype and gene expression , particularly in tumor cells [31 , 32] . To understand the molecular features of HrasG12V+ Aundiff in vivo , we isolated tdTomato+ and MCAM bright Aundiff ( Fig 1A ) from WT and FR mice treated with high-dose tamoxifen ( i . e . , two courses ) for RNA-Seq three months after induction . In order to confirm that the HrasG12V-positive Aundiff fraction was sufficiently high in the mutant ( MUT ) group , a variant calling analysis was performed at position chr7:141192906 , where the HrasG12V ( c . 35G>T ) mutation substitutes an adenine for a cytosine . This revealed that the mutation fraction was 46–49% in the MUT group , indicating that >92% of the isolated Aundiff had undergone recombination and were positive for HrasG12V ( Fig 6A ) . Differential gene expression analysis detected only minimal differences between the two groups , suggesting that Aundiff from WT and MUT are highly similar at the transcriptional level ( S2 Table ) . Among the few differentially expressed genes , Pax7 was found to be upregulated in MUT . This result was confirmed by RT-qPCR ( Fig 6C ) . However , the remaining seven differentially expressed genes were either functionally elusive or apparently irrelevant in Aundiff . To further capture any subtle differences in phenotype between WT and HrasG12V+ Aundiff , we manually curated a gene list comprising 28 stem cell marker genes and 25 differentiation marker genes based on recent studies that employed single cell RNA sequencing of mouse Aundiff [33 , 34] and created a heat map for the differential expression ( log2 CPM ) for each sample ( Fig 6B ) . Apart from Pax7 , these 52 genes , were not differentially expressed; yet , the stem cell marker genes trended toward slight down-regulation and the differentiation markers exhibited slight up-regulation in the MUT group ( Fig 6B ) . Despite these trends , we concluded that HrasG12V does not confer large transcription alterations in mutants . Cultured SSCs serve not only as a complementary model for a complex in vivo system but also enable facile manipulation of extrinsic stimuli to enhance cell phenotypes . Thus , to elaborate on the functional effects of HrasG12V in vitro , we derived SSC cell lines from different adult mouse lines ( Fig 7A ) . Upon 4-OHT addition in vitro , WT cells activated tdTomato ( iWT1 ) or mGFP ( iWT2 ) and FR cells initiated expression of HrasG12V , in addition to tdTomato ( iFR1 ) or mGFP ( iFR2 ) ( Fig 7A ) . First , the presence of Hras transcripts in WT , iWT , FR and iFR lines was measured by designing RT-qPCR primers that recognized a common sequence between the wild-type and HrasG12V alleles . Levels of Hras transcripts were similar in these cell lines , validating that the SV40 polyadenylation signal in FR does not affect the Hras transcript ( Fig 7B ) . Since HRASG12V protein is locked in a constitutively active conformation , we examined the amount of the functionally active form of HRAS in iFR SSC lines using a pull-down assay , in which only the active form of HRAS is recognized by the RAS-binding domain of RAF1 . Active HRAS was pulled down only in the iFR but not in the iWT , demonstrating the presence of functional HRASG12V protein in this model ( Fig 7C ) . We next studied changes in the growth trajectory of HrasG12V SSCs . In normal culture conditions , iFR did not grow faster than FR ( Fig 7D ) . Several lines of evidence show that a fitness advantage may be more pronounced upon environmental challenge , such as growth factor deprivation [13 , 35] . We therefore examined whether HrasG12V SSCs outcompete wild-type SSCs when cultured at an FGF2 concentration 100-fold lower than is used in normal SSC growth media . For the following experiments , FR2 and iFR2 cell lines were employed ( Fig 7A & 7E ) . We mixed the non-induced parental wild-type line FR2 ( mtdTomato+ ) with iFR ( mGFP+ ) at a 1:1 ratio and cultured them on feeder cells in FGF2 ( 0 . 01 or 10 ng/ml ) for up to five passages . iFR did not replace its wild-type parental line over time ( Fig 7E ) . This result suggests that a low FGF2 environment does not confer a competitive advantage to HrasG12V SSCs . Hras is an important mediator molecule of FGF-RAS-MAPK signaling [23] , and FGF2 has been shown to control stem cell self-renewal and possibly differentiation in SSCs [36 , 37] . In various RASopathy models , RAS-MAPK signaling pathways are highly dysregulated [22 , 38 , 39] . However , it is unclear whether this holds true for SSCs harboring endogenous gain-of-function mutations seen in RASopathies . To understand how MAPK signaling is altered in HrasG12V SSCs in response to FGF2 , iWT and iFR SSC clones were stimulated with FGF2 and probed for pERK and total ERK . pERK was similarly phosphorylated in iWT and iFR in response to FGF2 ( Fig 7F ) . However , basal level ERK phosphorylation prior to stimulation appeared to be slightly higher in iFR as compared to iWT . Lack of enhanced ERK activity in iFR following stimulation was unexpected , considering the presence of active HRAS protein in iFR cell lines . This result suggests that active HRAS at a physiological gene dosage does not augment the signal response to FGF2 in SSCs .
The competitive interactions of wild-type and neighboring mutant germline stem cells in the adult human testis are thought to produce uneven transmission of pathogenic alleles to children , but a dearth of model systems has hindered progress toward understanding the details of selection . In this study , we created a stem cell competition system in the mouse germline , in which undifferentiated spermatogonia with endogenous expression of an oncogenic mutation , HrasG12V , compete with their wild-type counterparts in the testis . We found that the HrasG12V mutation was transmitted from the paternal germline stem cells to offspring , whereas the mutated germ cells did not exhibit a significant expansion over a year of fate analysis . For in vivo genetic manipulation of the Aundiff , which contain the stem cell pool , only a handful cre driver lines have been described , including Vasa-cre and Stra8-cre . These cre alleles are not only non-specific with respect to germ cell subpopulations but also are active during early gonadal development , making it difficult to study adult mouse SSCs . For these reasons , cultured SSCs have been employed for loss- or gain-of-function studies using knock-down/out or ectopic expression of genes [13 , 23] . Hara et al . ( 2014 ) first leveraged Gfra1-creERT2 for lineage tracing of adult Aundiff; by measuring long-term self-renewal of the population , they demonstrated that genetically marked Aundiff are functional stem cells that persist at homeostatic levels [16] . In our mouse model , therefore , we adapted Gfra1-creERT2 to induce both a mutation and a label in the Aundiff and achieved robust recombination at the target loci in Aundiff . Conveniently , the size of the recombined fraction was adjustable from ~8 to almost 100% by dose of tamoxifen . This tunability makes Gfra1-creERT2 an appropriate cre-driver to study germline mosaicism , in which only a small subset of stem cells is genetically manipulated or marked and their fate is traceable thereafter . On the other hand , with a high dose of tamoxifen , Gfra1-creERT2 is suitable to obtain a highly enriched population of genetically manipulated germline cells in vivo . Indeed , by combining Gfra1-creERT2 with a reporter and anti-MCAM staining , we were able to enrich an Aundiff sample with 96% HrasG12V positivity , as measured at the RNA level . Although Gfra1-creERT2 seems to be best available inducible driver to genetically manipulate adult Aundiff , it is notable that Gfra1 is broadly expressed in the Aundiff and that the Gfra1-expressing population comprises a heterogeneous subset of Aundiff . It remains controversial whether Gfra1-negative stem cells exist in the Aundiff . Complex engineered alleles are required to model precisely the genetics of human germline disorders . In the FR-HrasG12Vfl/- system , we sought to induce HrasG12V without affecting total gene dosage throughout the recombination process; accordingly , similar levels of Hras transcript were observed before and after cre induction . Thus , such an approach most closely recapitulates the occurrence of human de novo mutations in male germline stem cells and the resultant germline mosaicism . Also , since this genetic model entails two tandem Hras genes , both comprising endogenous exons and introns , the final transcription product is also an endogenously spliced variant , in case any alternatively spliced SSC-specific variant exists . In our model , we found that recombination at the Rosa26 reporter locus happened more efficiently than that of the FR-HrasG12V locus . This could be explained by accessibility of the locus to cre recombinase; the Rosa26 locus is known for its constitutively active promoter region [40] , whereas the activity of the Hras locus may be tightly regulated by transcription factors and/or chromatin remodeling . Classical PAE disorders , such as Noonan syndrome , Costello syndrome , and other RASopathies , are caused by de novo mutations that are found only in affected children but not in the somatic DNA of either parent . Although most such mutations are derived from the paternal germline ( as reviewed in [6] ) , the process of transmission from paternal germline stem cells to the offspring has not been widely explored . A major strength of our strategy is that it allows one to interrogate the transmission of the mutated gene induced in male germline stem cells to subsequent generations . Here , we observed that the fate of HrasG12V stem cells resulted in HrasG12V offspring with neonatal mortality , consistent with Costello syndrome . This indicates that Aundiff carrying HrasG12V can differentiate into functional sperm , and the resultant Costello embryos are capable of developing to the neonatal stage . Similarly , our germline mosaic model could be applied to a variety of other de novo disorders with PAEs . Owing to technological advancements in genome sequencing , many de novo mutations that originate in the paternal germline have been implicated as drivers of neurodevelopmental disorders ( e . g . , autism ) . Yet , adult germline mosaicism as a major source of human disease is still controversial , and its natural course has not been addressed experimentally . By inducing specific autism-driving mutations in Aundiff and tracking their fate , our mosaic model could be useful going forward to uncover the causal relationship between paternal age and neurodevelopmental disorders . A human sperm analysis revealed that healthy males carried different Hras mutations at codon 12 [9] . Although G12S was the most prevalent mutation , G12V was also substantially elevated in the sperm and the level was weakly correlated with donor age . Thus , we sought to investigate how G12V-laden germline cells expand in our experimental model . To perform stringent lineage tracing of HrasG12V-positive cells , we employed multiple independent quantification methods and their outcomes were well correlated: cDNA sequencing from labeled germ cells , sperm gDNA analysis , and sperm counting . Over a year of lineage tracing , we did not observe a significant increase in the HrasG12V cell and sperm populations . In the serial sperm analysis , a minimal increase ( ~9% ) was observed over the 8–9 month period , yet we did not observe an increase in the proportion of HrasG12V offspring from mutant sires in successive litters . Despite the fact that the HrasG12V is the most potent gain-of-function mutation among various HRAS proteins at codon 12 , these findings indicate that the heterozygous HrasG12V alone is not sufficient to drive SSC competition in the mouse testis . There are several possible explanations for the absence of competition in our model . First , there could be fail-safe mechanisms to suppress aberrant cell signaling driven by hyperactive RAS , providing a robust system that protects stem cells from harmful consequences . Second , HRAS may not have a major role in FGF-RAS-MAPK signaling in the Aundiff , which would be unexpected given previously published data [23] . Other members of the RAS superfamily may be critical for relaying FGF signals in the SSCs . Third , the HrasG12V mutation may be detrimental , such that that cell turnover could be faster than usual . Fourth , induction experiments performed at extremely low starting ( i . e . , baseline ) mutation levels ( e . g . , <1% recombination ) might reveal competitive interactions that could have been obscured in our experiments , due to as yet uncharacterized paracrine effects . Fifth , key differences could exist in the microenvironment or cellular population structure between mice and humans . Sixth , the observation period in this study ( ~one year ) could be too short to capture long-term cell competition because of the much shorter lifespan of mice than of humans , which entails decades of continuous stem cell survival and self-renewal in the testis . Regarding these last two points , age-related changes in the stem cell niche could be necessary for HrasG12V-mediated stem cell competition to become apparent . Finally , additional , yet unidentified genetic lesions could be required to confer enhanced competitiveness . An absence of cell competition was also observed in vitro . In cultured SSCs with HrasG12V , although a functionally active HRAS was detected , ERK activation following FGF2 stimulation was not enhanced . Chen et al . ( 2009 ) made a similar observation , using mouse embryonic fibroblasts ( MEFs ) with heterozygous HrasG12V [22] . Only in later passaged HrasG12V MEFs was more pERK detected than in controls . This suggests that active HRAS from one copy of HrasG12V is not adequate to activate ERK in either cultured SSCs or MEFs , and more time may be required to gain additional gene mutational hits . Furthermore , a reduced FGF2 environment did not favor selection of the HrasG12V cell population , indicating that HRAS may not be a major GTPase mediating FGF2 signaling in SSCs . On the other hand , when HrasG12V is overexpressed in cultured SSCs , increased proliferation and oncogenic transformation were observed [23] , again suggesting that the HrasG12V effect is dependent on gene dosage . The transcriptional profile between wild-type and HrasG12V Aundiff did not reveal a significant difference in gene expression . Interestingly , among a few differentially expressed genes , Pax7 was up-regulated in iFR cells . Pax7 is a transcription factor identified as a conserved marker for a particularly rare subset of Aundiff in mammalian testes [33 , 41] . Although its function in self-renewal of Aundiff is not well characterized , Pax7 may be one of the downstream genes upregulated by MAPK signaling via HRAS in a subset of the Aundiff population . Overall , the absence of large transcriptional changes in HrasG12V Aundiff could account for the fact that there was no obvious competitiveness or higher stem cell capacity in HrasG12V-positive Aundiff in the transplant assays . In conclusion , these results revealed a stable and tolerant system to prevent normal germline stem cells from being replaced by mutated cells . This unanticipated resistance to hyper-active Hras suggests inherent mechanisms within germline stem cells to suppress harmful mutations that would otherwise be propagated to offspring . In contrast , we anticipate that future studies will likely uncover factors that overcome such protective mechanisms , leading to aberrant clonal expansion . Given the increasing number of disorders ( e . g . , autism ) linked to germline mosaicism and PAEs , the inducible adult mosaic model will be invaluable to understand the earliest origins of such pathogenic gene variants .
This study was approved by the Weill Cornell Medical College IACUC ( #2010–0028 ) . Either isoflurane or Ketamine/Xylazine was used for anesthesia in combination with buprenorphine and meloxicam for analgesia . For euthanasia , mice were exposed to CO2 followed by cervical dislocation . Gfra1-creERT2 mice were a gift from Dr . Sanjay Jain [16 , 42] . Reporter mice , tdTomato ( #007914 ) [43] and mT/mG ( #007676 ) [29] , were obtained from the Jackson Laboratory . The FR-HrasG12Vfl/fl mice were previously generated by Dr . James Fagin [22] . The controls were wild-type littermates that do not have the FR-HrasG12Vfl allele but contain Gfra1-creERT2 to induce tdTomato expression . These mice were maintained on a mixed genetic background of C57BL/6J ( >50% ) , 129/Sv , and Swiss Black mice . All the experimental protocols were approved by the Weill Cornell Medicine Institutional Animal Care and Use Committee . At 6–8 weeks of age , 100 mg/kg of tamoxifen ( Sigma ) dissolved in corn oil ( Sigma ) was administered intra-peritoneally for 4 days ( one standard course ) , unless otherwise specified . At each time point , animals were euthanized and the testes and caudal epididymal sperm were harvested for downstream experiments . Detunicated testes were fixed with 4% paraformaldehyde in phosphate buffered saline overnight at 4°C , immersed in 30% sucrose , and embedded in OCT compound . After cryosectioning the samples at 10 μm , DAPI was applied . For whole-mount staining , after overnight fixation , the seminiferous tubules were untangled , washed in PBS , and blocked with 3% BSA/PBS with 0 . 1% Tween for an hour . After incubation with anti-Gfra1 antibody ( 1:200 , BD ) overnight , an anti-rabbit biotinylated secondary antibody , followed by Alexa647-conjugated streptavidin was used for detection . DAPI was used for nuclear staining . Images were captured with a Zeiss LSM 800 confocal microscope . Microsurgical epididymal sperm aspiration was performed as previously described [28] . Under deep anesthesia , through a small skin incision on the scrotum , the cauda epididymis was punctured by a syringe , and its contents were aspirated ( S2 Fig ) . The procedure was performed on ipsilateral testis at 4 months and again at 12 months after tamoxifen administration . gDNA extraction was performed using AllPrep DNA/RNA Mini Kit ( Qiagen ) with a modified protocol [44] . A whole testis dissociate was prepared using a two-step enzymatic digestion [45] . For MCAM staining , testes from Gfra1-creERT2; tdTomato mice ( n = 3 ) were dissociated >3 months after tamoxifen administration . The single-cell suspensions from two testes were incubated with Alexa Fluor 647 anti-MCAM antibody ( ME-9F1 , BioLegend ) at a concentration of 6 g/ml for 45 minutes at 4°C . After exclusion of doublets and DAPI-positive cells , Alexa Fluor 647 and tdTomato double-positive cells were gated and collected using a BD Aria flow cytometer . Total RNA was extracted from sorted testicular cells or feeder-free cultured SSCs using Arcturus PicoPure Kit ( Applied Biosystems ) or RNeasy Plus micro kit ( Qiagen ) , respectively , with an on-column DNA digestion protocol ( Qiagen ) . Reverse transcription was performed using qScript ( Quanta Biosciences ) followed by a real-time PCR using Sybr Select Master Mix ( Applied Biosystems ) with a LightCycler 480II ( Roche ) . Each technical triplicate was normalized to Actb and relative expression levels to control conditions were calculated using 2-ΔΔCt method . FACS-collected tdTomato+ cells were lysed in RLTplus buffer ( Qiagen RNA mini kit ) with 2% β-mercaptoethanol , and RNA was purified according to the manufacturer’s instructions . The RNA was reverse-transcribed to cDNA using qScript cDNA SuperMix ( Quanta Biosciences ) . A targeted region flanking HrasG12V was amplified by PCR and Sanger-sequenced . To quantify the mutated nucleotide ( cytosine→adenine ) from Sanger sequencing traces , a model curve was proposed and confirmed by amplifying cDNA samples of known concentrations from WT ( HrasWT/WT or FR/WT ) and Costello offspring ( HrasG12V/WT ) RNA ( x-axis: 0 , 25 , 50 , 75 , and 100% of Costello RNA ) ( Fig 2A ) . By using 4 pairs of different offspring samples , it was confirmed that the relationship between the HrasG12V-positive cell fraction ( x% ) and ratio of adenine/cytosine ( A/C ) peak heights ( y ) fitted the model curve y = 1 . 371*x / ( 200-x ) . All the calculations for HrasG12V-positive cell fraction were done using this equation . ImageJ was used to measure A/C ratio via peak height . All the primers used in the study are listed in S3 Table . To obtain a fractional abundance of the HrasG12V allele , qPCR primers were designed , detecting the SV40 PA region that is exclusively present in the FR-HrasG12Vfl locus and lost after recombination ( Fig 2B ) . By running qPCR with mixed gDNA from a FR-HrasG12V heterozygote ( without recombination ) and wild-type sperm collected from cauda epididymides at different known ratios ( x: 0 , 25 , 50 , 75 , 100% of FRfl/- allele ) , we confirmed a linear relationship between the FR-HrasG12Vfl recombined fraction ( x% ) and the SV40 PA recombined fraction obtained by qPCR ( y% ) ( Fig 2B left ) . Since the loss of SV40 PA or FR means a gain of HrasG12V , y = -z+100 was obtained , where z is the HrasG12V+ cell fraction ( % ) ( Fig 2B , right ) . Per reaction , 60ng of sperm gDNA in 4 ul was used . For normalization of the quantity of gDNA , Ngn3 gene primer sets were used . Light Cycle480 Software was used for analysis . To quantify a recombined fraction of a reference gene , mTmG , we employed the same strategy as the quantification for HrasG12V+ cell fraction . Primer sets were designed to detect a region of the mTomato locus that is deleted after recombination ( Fig 2B ) . Sperm obtained by microsurgical epididymal sperm aspiration was mounted on slides and imaged using a BX50 fluorescence microscope ( Olympus ) with a Spot Pursuit CCD camera ( Diagnostic Instruments Inc ) . Total of 100 to 300 sperm ( > 3 fields ) were counted . To obtain the highest labeling after the maximum dose of tamoxifen , testis cell suspensions from Gfra1-creERT2; tdTom mice and Gfra1-creERT2; tdTom; FR-HrasG12V mice ( littermate of the former ) were mixed with that of non-labeled wild-type ( Gfra1-creERT2 ) at a 1:1 ratio , respectively , and transplanted into adult busulfan-conditioned C57Bl6 recipient testes ( n = 10 or 11 mice per experiment ) . A total of 1 . 2 x 106 ( first two experiments ) or 0 . 6 x 106 ( the 3rd experiment ) mixed cells were injected per testis . Site of injection ( left vs . right side ) was alternated per genotype . Three independent transplantations were performed ( 3 donor mice per group ) . After 10 weeks , colony numbers were quantified using stereomicroscopy . WT and FR mice ( n = 3/ group ) were treated with tamoxifen for four days for twice over two weeks . The testes were dissociated ( see Fig 1A ) and RNA from tdTomato+ MCAM-bright Aundiff was isolated via Arcturus PicoPure kit with DNase I treatment . Mean RNA Integrity Number ( RIN ) was 9 . 47 ( SD 0 . 31 ) . Libraries were constructed using TruSeq Stranded mRNA . Sequencing was performed on an Illumina HiSeq 2500 ( v4 chemistry ) with a 50 bp paired-end protocol . Variant calling and filtering were carried out using Bam-Readcount and SAMtools mpileup at chr7:141192906 , where samples with heterozygous HrasG12V ( c . 35G>T ) mutation should have evidence of a C/A genotype while wild-type samples have only a C genotype . Differential expression was assessed using DESeq2 with a false discovery rate ( FDR ) of 0 . 1 . SSC lines were derived from two pairs of littermate adult mice ( see Fig 7A for specific cell lines ) and maintained on mitotically-inactivated JK1 feeders [46] . SSC growth media was StemPro-34 with additional supplements as described previously [45] . Treatment with 6 μM of 4-OHT for 4 days was repeated for a total of 3–4 times over 3–4 weeks until the HrasG12V population was confirmed to be more than 70–80% by the sequencing method ( see Fig 2A–2C ) . Between tamoxifen treatments , tdTomato-positive cells were enriched once by FACS to enhance the recombination efficiency for HrasG12V . All the experiments were performed using cells with passage number 9 to 22 . The stem cell activity was confirmed by transplantation assays . An HRAS activation assay was performed using a GST-fusion protein of the RAS-binding domain ( RBD ) of RAF1 , as instructed by the manufacturer ( Pierce Biotechnology ) . Briefly , SSCs maintained in growth media were washed with cold TBS and lysed . Active RAS was pulled down with GST-RAF1-RBD along with glutathione agarose resin , followed by Western blot detection with an anti-Hras antibody ( sc-520 , Santa Cruz Biotechnology ) . After feeder subtraction , SSCs were washed once with ice-cold TBS and lysed in lysis buffer containing PMSF ( Sigma ) , protease inhibitor ( Sigma ) , and phosphatase inhibitor cocktails ( Sigma ) . Protein concentration was quantified by the BCA method . Immunoblotting was performed according to standard procedures . Denatured samples were subjected to 12% SDS/PAGE gel and transferred to PVDF membrane . The following antibodies were used: pERK ( 9101 , Cell signaling ) , ERK ( 9102 , Cell Signaling ) , CyclophilinB ( Invitrogen ) , and Hras ( sc-520 ) . To obtain mouse HRAS protein control for western blotting , a fragment ( 606 bp ) of Hras ( NM_008284 . 2 ) was synthesized by Integrated DNA Technologies , inserted into BamHI/EcoRI sites of pCIG [47] , and overexpressed in HEK293T cells . Gfra1-creERT2; mTmGfl/-; FRfl/- ( FR2: tdTomato+ wild-type cells ) and its induced derivative iFR2 ( GFP+ HrasG12V ) cells were mixed at 1:1 ratio , cultured in low ( 0 . 1 ng/ml ) and high ( 10 ng/ml ) FGF2 media , and passaged 5 times . At each passage , GFP+ cell ratios were measured by FACS ( BD Accuri ) . Feeder-free SSCs were starved for 18 hours and then either left unstimulated ( 0’ ) or stimulated with FGF2 ( 10 ng/ml ) , and harvested 5 , 15 , and 30 minutes after the stimulation . Cell lysates were analyzed by immunoblotting . Results are presented as mean ± SD . At least three biological replicates and three technical replicates were performed for each experiment unless otherwise indicated in the text . GraphPad Prism was used for statistical analyses and generating graphs .
|
Recent research has found that advanced paternal age is associated with increased risk in children to develop a subset of congenital anomalies , such as Apert syndrome , achondroplasia , Noonan syndrome , and Costello syndrome . The causative genetic errors ( mutations ) in these disorders have been identified to originate from the fathers’ testicles and their numbers increase with fathers’ age . It has been hypothesized that the germline stem cells that continuously self-renew and differentiate to supply sperm ( referred as spermatogonial stem cells [SSCs] ) carry these mutations and have the ability to expand preferentially as compared to normal SSCs with advancing age of the father , thereby increasing the likelihood of transmission of mutant sperm to the next generation . To test this hypothesis , we created a mouse model , in which a mutation known to enhance cell proliferation is induced in a subset of SSCs , and these cells compete with the neighboring normal ( i . e . , wild-type ) stem cells . However , surprisingly , the germline cell population carrying the mutation in the testis was stable over a year of observation , suggesting that mechanisms could exist to prevent such harmful mutations from being expanded and transmitted to the next generation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"sequencing",
"techniques",
"sv40",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"microbiology",
"spermatogonia",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"germ",
"cells",
"viruses",
"stem",
"cells",
"dna",
"viruses",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"sperm",
"research",
"and",
"analysis",
"methods",
"transplantation",
"blood",
"and",
"lymphatic",
"system",
"procedures",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"polyomaviruses",
"testes",
"cell",
"biology",
"anatomy",
"stem",
"cell",
"transplantation",
"viral",
"pathogens",
"cell",
"transplantation",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"genital",
"anatomy"
] |
2019
|
Functional robustness of adult spermatogonial stem cells after induction of hyperactive Hras
|
A synergistic combination of two next-generation sequencing platforms with a detailed comparative BAC physical contig map provided a cost-effective assembly of the genome sequence of the domestic turkey ( Meleagris gallopavo ) . Heterozygosity of the sequenced source genome allowed discovery of more than 600 , 000 high quality single nucleotide variants . Despite this heterozygosity , the current genome assembly ( ∼1 . 1 Gb ) includes 917 Mb of sequence assigned to specific turkey chromosomes . Annotation identified nearly 16 , 000 genes , with 15 , 093 recognized as protein coding and 611 as non-coding RNA genes . Comparative analysis of the turkey , chicken , and zebra finch genomes , and comparing avian to mammalian species , supports the characteristic stability of avian genomes and identifies genes unique to the avian lineage . Clear differences are seen in number and variety of genes of the avian immune system where expansions and novel genes are less frequent than examples of gene loss . The turkey genome sequence provides resources to further understand the evolution of vertebrate genomes and genetic variation underlying economically important quantitative traits in poultry . This integrated approach may be a model for providing both gene and chromosome level assemblies of other species with agricultural , ecological , and evolutionary interest .
The rapid and continuing development of next-generation sequencing ( NGS ) technologies has made it feasible to contemplate sequencing the genomes of hundreds—if not thousands—of species of agronomic , evolutionary , and ecological importance , as well as biomedical interest [1] . Recently , a draft genome of the giant panda was described , based solely on Illumina short read sequences [2] . Below , we describe the genome sequence of the turkey ( Meleagris gallopavo ) determined using primarily NGS platforms . In this case , however , a combination of Roche 454 and Illumina GAII sequencing was employed . While this approach presented unique assembly challenges , the turkey sequence benefits from the particular advantages of both platforms . In addition , unlike the case for the panda , this novel approach allowed us to use a BAC contig-based physical and comparative map , along with the turkey genetic map [3] and the chicken genome sequence [4] , to align the turkey sequence contigs and scaffolds to most of the turkey chromosomes . Such an alignment is essential for making long range evolutionary comparisons and employing the sequence to improve breeding practices using , for example , genome-based selection approaches , where chromosome locations are critical . The high throughput and low cost of NGS technologies allowed sequencing the turkey genome at a fraction of the cost of other recently reported genomes of agricultural interest ( bovine and equine ) [5] , [6] . The draft turkey genome sequence represents the second domestic avian genome to be sequenced , and this permits a genome-level comparison of the two most economically important poultry species . When added to the recently published zebra finch genome [7] , analysis of the three avian genomes reveals new insights into the evolutionary relationships among avian species and their relationships to mammals . Turkeys , like chickens , are members of the Phasianidae within the order Galliformes . One estimate [8] is that the last common ancestor of turkeys and chickens lived about 40 million ( M ) years ago; however , other estimates are more recent [9] , [10] . Comparison of the turkey genome to that of the chicken provides the opportunity for high resolution analysis of genome evolution within the Galliformes . The turkey has 2n = 80 chromosomes ( chicken has 2n = 78 ) and , as for most avian species , the majority of these are small “microchromosomes” that cannot be distinguished by size alone . Although most turkey chromosomes are syntenic to their chicken orthologues , the chicken chromosome GGA2 is orthologous to two turkey chromosomes , MGA3 ( GGA2q ) and MGA6 ( GGA2p ) , due to fission at or near the centromere , while GGA4 is orthologous to MGA4 ( GGA4q ) and MGA9 ( GGA4p ) [10] , [11] .
Generally , DNA from a single inbred animal is preferential for sequencing to minimize polymorphism . For the turkey , however , such an option is not available , and thus we sequenced DNA from “Nici” ( Nicholas Inbred ) , a female turkey , which is also the source DNA for the two BAC libraries that have been characterized [12] . Nici is from a subline ( sib-mating for nine generations ) originally derived from a commercially significant breeding line , but her genome is still extensively heterozygous . A side benefit of this approach was the concomitant identification of extensive and commercially relevant single nucleotide polymorphism ( SNP ) data , as discussed below . With the exception of the BAC end sequences ( BES ) used only for chromosome alignment , the sequence data used for this assembly came solely from two sequencing platforms: the Roche/454 GS-FLX Titanium platform ( 454 Life Sciences/Roche Diagnostics , Branford , CT ) and the Illumina Genome Analyzer II ( GAII; Illumina , Inc . , San Diego , CA ) . The 454 data were generated using the latest “Titanium” protocol at Roche and the Virginia Bioinformatics Institute ( Virginia Tech ) and included both unpaired shotgun reads and paired-end reads produced from two libraries with estimated 3 kilobase pair ( Kbp ) and 20 Kbp fragment sizes . The 454 runs yielded approximately 3 M read pairs from the 3 Kbp library ( average usable read length 180 bases ) , 1 M read pairs from the 20 Kbp library ( average length 195 bases ) , and 13 M shotgun reads ( average length 366 bases ) . The Illumina sequencing data were generated at the USDA Beltsville Agricultural Research Center and the NIH National Institute on Aging from both single and paired-end read libraries with a 180 bp fragment size for the paired reads . Details on the sequence data are presented in Table 1 . These data represent approximate 5× genome coverage in 454 reads and 25× coverage in GAII reads , assuming a genome size similar to that of the chicken at 1 . 1 billion bases [4] . In addition , BACs used to generate the 40 , 000 BES alignment markers by traditional Sanger sequencing spanned ∼6× clone coverage of the genome . Since female DNA was used , coverage of the Z and W sex chromosomes was half that of autosomes; therefore the assembly of both these chromosomes was poor . A modified version of the Celera Assembler 5 . 3 [13] , [14] was used to produce the contigs and scaffolds in the assembly ( see Methods for details ) . The initial assembly contained 931 Mbp of sequence in 27 , 007 scaffolds with N50 size of 1 . 5 Mbp . The span of the scaffolds was 1 . 038 Gbp . The scaffolds contained 145 , 663 contigs with N50 size of 12 . 6 Kbp . The assembled scaffolds were then ordered and oriented on turkey chromosomes using a combination of two linkage maps and a comparative BAC contig physical map . The first turkey linkage map [3] had 405 chicken and turkey microsatellite sequences that mapped to the assembled scaffolds . The second linkage map , based on segregation of SNPs in a different population [15] , had 442 SNP markers mapped to the scaffolds . The comparative chicken-turkey physical map [16] provided turkey chromosome positions for 30 , 922 BES found in scaffolds . Comparison of scaffolds to the marker map resulted in splitting only 39 scaffolds due to inconsistencies between the assembled scaffolds and marker positions on the chromosomes . A total of 28 , 261 scaffolds containing 917 Mb of sequence were assigned to chromosomes ( Table 2 ) . Included in this number were 7 , 080 single-contig scaffolds that represented repetitive sequences but that could be linked to non-repetitive scaffolds . The remaining 5 , 858 scaffolds were pooled to form ChrUn ( unassigned ) which contains 19 Mb of sequence in comparison to about 64 Mb on the current chicken chr_Un . Analysis of the assembled contigs showed that 4 . 6% of the sequence was covered only by reads from a single sequencing platform , with 2 . 3% covered exclusively by each . If the reads covered the genome uniformly , one would expect to have missed only 0 . 67% of the genome with Roche/454 and 0 . 0006% with Illumina . The distribution of regions of exclusive coverage for both platforms ( Figure S1 ) shows there was a large number of short ( <20 bp ) gaps in coverage by Illumina sequencing , whereas the Roche/454 coverage gaps tended to be larger . Mean sequencing gaps were 46 bases for Illumina reads and 72 for the Roche/454 coverage . Coverage biases previously have been shown for both platforms [17] , but fortunately , the biases are relatively orthogonal . Therefore , it is definitely beneficial to use data from both platforms in de novo assemblies . The draft turkey assembly was compared to the chicken genome assembly ( 2 . 1 ) , which was sequenced and assembled using traditional Sanger sequencing [4] . Table 3 illustrates that assembly of NGS sequence data , although feasible , does not produce contigs and scaffolds as large as those expected from an assembly based on Sanger sequencing . However , the relatively low cost of NGS sequencing ( <$250 , 000 for the turkey ) makes such projects feasible for species with more focused interest groups and facilitates for resources to be directed toward genome analysis and interpretation as opposed to generating raw sequence data . However , chromosome assemblies currently still require the integration of multiple data types including shotgun reads and contigs , genetic linkage maps , BAC maps and BES , and cytogenetic assignments . The challenge was to develop databases and software to achieve this goal . Integrity of the assembly was validated by mapping the assembled turkey scaffolds to 197 Kbp of finished BAC sequence containing part of the MHC B-locus , GenBank accession DQ993255 . 2 . The average sequence similarity was over 99 . 5% and no inconsistencies in the 21 scaffolds that mapped to that region were observed . The extent of the genome coverage could be estimated both from the total span of the assembled scaffolds and from portions of the chicken genome with syntenic matches to the turkey scaffolds . Both methods produced consistent estimates of the size of the euchromatic portion of the turkey genome at about 1 . 05 Gbp . With 936 Mbp of sequence in the final chromosomes , including ChrUn , the assembly encompasses an estimated 89% of the total sequence of the genome . One of the striking observations in the chicken genome sequencing project was the difficulty obtaining sequences for specific regions , including the 10 smallest microchromosomes [4] . For example , the chicken genome lacks sequence orthologous to human chromosome 19q . Remarkably , these sequences appeared to be absent not only from the shotgun clone libraries used to generate the whole genome shotgun ( WGS ) reads but also from all available BAC libraries [18] . Although these regions have high GC content , it is unclear why this region of the genome is resistant to cloning in E . coli . In general , BAC coverage of microchromosomes is less than macrochromosomes in both chicken and turkey BAC libraries , although the HSA19q orthologues are an extreme example of a missing syntenic region . Since the turkey genome was sequenced without any cloning step , the assembly was tested for representation of HSA19q orthologous sequence . Presence of sequences was verified by performing a BLAT analysis of the complete HSA19q sequence against the turkey and chicken genomes ( Table S1 ) . Surprisingly , regions orthologous to HSA19q were not represented at a higher frequency in the turkey assembly versus the chicken assembly . As was observed in the chicken , regions orthologous to HSA19p and a small syntenic region from HSA19q are covered well in the turkey assembly ( MGA30 and 13 , respectively ) . These results suggest that absence of HSA19q orthologous sequences is not due to the high GC content , in that Illumina sequences show a bias towards higher coverage of GC rich regions [19] , [20] . The identification of a single BAC clone that hybridizes across the entire length of a single microchromosome in chicken [21] suggests that the occurrence of microchromosome-specific repeats might be a more likely explanation for the absence of these sequences using both traditional Sanger sequencing as well as NGS technologies . Heterozygous alleles , including both SNPs and single nucleotide insertions and deletions ( indels ) , were detected by scanning the assembled contigs for positions where the underlying reads significantly disagreed with the consensus base [22] . A previous study cataloging heterozygous alleles from assembled shotgun reads within an individual human genome used a similar approach , augmented with a set of quality criteria used to distinguish genuine biological variations from sequencing error [23] . Following this approach , a set of quality criteria was developed and implemented within the assembly forensics toolkit [24] . Two classes of SNVs were catalogued: ( 1 ) those with abundant evidence , called strong SNVs ( 601 , 490 SNVs ) , and ( 2 ) a more inclusive set called weak SNVs ( 920 , 126 SNVs total ) . In the turkey genome , transitions were roughly 2 . 4× more common than transversions: 295 , 055:122 , 731 for strong SNVs and 466 , 629:200 , 743 for all SNVs . Many single base indel positions were detected: 183 , 215 of 601 , 490 strong SNVs , and 249 , 512 out of all 920 , 126 SNVs . A very small number of SNVs ( 489 strong , and 3 , 242 all ) were detected with more than two well-supported variants , suggestive of unfiltered sequencing errors or collapsed repeats . The depth of coverage for strong SNVs ranged between 6 and 30 with mean and standard deviation of 15 . 3±5 . 3 , while the depth of coverage for all SNVs ranged between 4 and 5 , 319 with mean and standard deviation of 41 . 4±134 . 6 . The very high coverage regions are highly likely to be due to collapsed near-identical repeats . Annotation of the turkey genome sequence identified a total of 15 , 704 genes ( Table S2 ) of which 15 , 093 were distinct protein coding loci and 611 non-coding RNA genes . In addition , multiple distinct proteins produced by alternative splicing were identified for some loci , giving a total of 16 , 217 distinct protein sequences . Orthologs between turkey , chicken , and human proteins were defined using sequence homology , phylogenetic trees , and conservation of synteny . All gene annotations are available from the Ensembl genome browser version 57 ( http://e57 . ensembl . org ) . The draft turkey genome assembly was used to test the distribution of nucleotide diversity across the turkey genome by aligning SNPs covering ∼3 . 97% of the genome identified through resequencing a reduced representation library from commercial turkeys [15] . Substantial deviations were observed between regions in the genome . Chromosome Z showed the lowest nucleotide diversity , about half ( θ = 0 . 000273 ) that of the autosomes , which is likely the result of a lower effective population size of this chromosome and lower recombination rate ( Figure S2 ) [25] . The five largest chromosomes had similar nucleotide diversities as the microchromosomes . Given the higher recombination rate on the microchromosomes , the ensuing higher mutation rate [26] , and lower susceptibility to hitchhiking effects , equal rates of nucleotide diversity between micro- and macrochromosomes may seem unexpected . However , these findings are in line with observations in the chicken [27] and may be explained by higher gene density and higher purifying selection on the microchromosomes . Within chromosomes , extended regions of low nucleotide diversity were detected , many of which coincided with centromeres . Comparisons of gene family assignment statistics for the turkey and chicken genome assemblies are shown in Table S4 . Although the draft turkey sequence has fewer genes than the current chicken genome build ( 2 . 1 ) , part of the difference may be due to cutoff values used by annotation groups resulting in variation in gene number . Even with this caveat , more than half of the gene families show no change in copy number between them ( Table S5a–d ) . Overall , most families exhibiting variation have general regulatory functions related to transcription , metabolism , cation transport , cell-cell signaling , and cell development or differentiation ( Figure 3 ) . Distinct keratin families , encoding major structural proteins of chicken feathers , claws , and scales , have undergone uneven expansion or contraction with considerable variation in number among species . More than half of the innovation families ( found in turkey but not chicken ) have unknown functions , are singletons , and were annotated by mapping to the zebra finch protein prediction . Species-specific gene families in birds and mammals are summarized in Tables S6 , S7 . Of these , 881 are specific to turkeys and chickens and 271 specific to birds . The inference for bird-specific functions is of relatively high quality since the likelihood that a bird gene is not simultaneously found in all 13 non-bird species is low . Most of the rapidly evolving gene families in birds have unknown functions . Approximately 83% of the turkey/chicken-specific families and 71% of the bird-specific families have unknown functions . For the remaining families , most have well-defined roles ( Table 5 ) . Families related to egg formation ( such as avidin , ovocalyxin , and vitellogenin ) and scavenger receptors were identified as avian specific in the present and previous analyses [4] . Examination of gene family sizes between the avian species and the platypus , an egg laying mammal , found two egg-related gene families [egg envelop protein ( ENSFM00500000271806 ) and vitellogenin , an egg yolk precursor protein ( ENSFM00250000000813 ) ] to be conserved among the four egg-laying species . Both of these gene families are absent from eutherians . Other gene families specific to egg-laying species ( birds and platypus ) are mainly related to protein metabolism , cell-cell communication , and regulatory functions . Several other proteins related to egg formation , such as avidin and ovocalyxin , are found in birds but not in platypus . In contrast to unique gene families , only 70 families were completely absent in both the turkey and chicken ( 33 in all birds ) compared to the non-avian species . These include the gene family associated with enamel formation ( ENSFM00250000008876 , an enamelin precursor related to teeth ) , which is completely lost in the three avian species . Genes encoding the vomeronasal receptors and several casein related families are also completely absent in the avian species . Several olfactory receptor families specific to mammals are either absent or dramatically reduced in birds . Interestingly , the olfactory receptor 5U1 and 5BF1 gene families , reported to be dramatically expanded in chicken as compared to humans and flies [4] , is contracted in turkey . Lineage events in the turkey , chicken , and zebra finch genomes reveal significantly higher synonymous substitution rates on microchromosomes than macrochromosomes ( Figure 4a ) , with a clear inverse relationship with chromosome size . This suggests that genes on the microchromosomes are exposed to more germ-line mutations than those on other chromosomes [38] . However non-synonymous mutation rates do not seem to vary so widely and when combined show the dN/dS ratio ( a measure of selection ) to increase with chromosome size . These results are consistent with the prediction that the higher synonymous substitution rates of microchromosomes combined with the “Hill-Robertson” effect [39] of higher recombination rates on these smaller chromosomes increases purifying selection [40] on the microchromosomes ( Figure 4b and 4c ) . Theory predicts natural selection to be more efficient in the fixation of beneficial mutations in mammalian X-linked genes than in autosomal genes , where hemizygous exposure of beneficial non-dominant mutations increases the rate of fixation . This “fast-X effect” should be evident by an increased ratio of non-synonymous to synonymous substitutions ( dN/dS ) for sex-linked genes . As shown in Figure 5 , there is solid confirmation of the predicted rapid evolution in the sex-linked genes based on turkey , chicken , and zebra finch genome-wide data . These results confirm that evolution proceeds more quickly on the Z chromosome [41] , where hemizygous exposure of beneficial non-dominant mutations increases the rate of fixation . Based on the analysis of differentially evolved genes , 428 and 257 genes were identified as being under accelerated evolution in the turkey and chicken lineages , respectively . Most of the accelerated genes in the turkey lineage have gene ontology ( GO ) terms related to DNA packaging and regulation of transcription ( Figure 6a ) . In contrast , a large proportion of the accelerated genes in the chicken lineage have GO terms related to negative regulation of cellular component organization and biogenesis , proteolysis , interphase , and cell cycle arrest ( Figure 6b ) . The enrichment of KEGG pathways using DAVID supports the GO term analysis ( Table S8 ) . These results suggest that genes with a role in transcriptional regulation are key in the evolution of the turkey , whereas genes involved in protein turnover and cell proliferation have been more important in the evolution of the chicken . For genes classified as innate immune loci by InnateDB ( www . innatedb . ca ) , dN/dS ratios were calculated for each pair of species ( turkey-chicken , turkey-zebra finch , etc . ) and then compared with ratios for non-immune genes . Innate immune genes showed lower dN/dS ratios than other genes in all species-pairs of mammals and birds , except between turkey and chicken where the values are essentially equal ( Figure 7 ) . Using Wilcoxon rank sum test , it is obvious from the comparisons that the innate immune-related genes have been under more purifying selection than non-immune-related genes except between turkey and chicken ( Table S9 ) . Evolution of genes of the innate immunity system is thought to be continuous and under balancing selection [42] . However , purifying selection under the same conditions may be the dominant force acting on the vast majority of genes that function within the innate immune system [43] . Although only innate immune genes are under purifying selection by functional constraints , they are also more constrained than other genes . This relationship supports the view that the ancient innate immune system has a highly specialized function , critical for the recognition of pathogens and thus should be under purifying selection . However , unlike other species , the dN/dS ratios for innate immune genes between turkey and chicken are similar to other genes . Perhaps the adaptation of turkey and chicken to different ecological niches has exposed them to new pathogenic environments with potentially lethal pathogens having exerted selective pressures on their genomes . This thesis would suggest that there was a period of accelerated evolution of the innate immunity system after the divergence of these species 30–40 M years ago . The availability of the turkey genome for comparison to the chicken [4] and zebra finch [7] allows for interrogation of the immune gene repertoire . In general , homologs for all the innate immune gene families were found ( Table 6 ) , with smaller gene families present in birds . This finding is consistent with earlier comparisons of mammalian with the chicken genome [44] and provides greater evidence of an avian-wide phenomenon . Examples include the chemokines , TNF superfamily , and pattern recognition receptors . Inflammatory CCL chemokines , which occur in all avian and mammalian species , fall into two multigene families ( MIP and MCP; Figure S4 ) . There are four MIP family members in the chicken and the zebra finch ( CCLi1–4 ) , yet only three family members in the turkey genome build ( CCLi2–4 ) . For the MCP family , there are six ( CCLi5–10 ) , three ( CCLi5–7 ) , and five ( CCLi5–7 and 9–10 ) members in the chicken , zebra finch , and turkey genomes , respectively . The chicken genome sequence lacks TNFSF-family members TNFSF1 and TNFSF3 [44] . Presence of these lymphotoxins controls lymph node formation in mammals [45]; however , lymph nodes are absent in birds [46] . Therefore , it was not surprising that these genes were not found in any of the three avian genomes . In contrast , lack of TNFSF2 ( TNFA ) was unexpected , since it is found in many fish species [47] , and there are several reports of TNF-alpha-like activity in chickens [48] . A sequence homology search in the three avian species only detected TNFSF15 , a close relative of TNFSF2 . Loss of TNFSF1 , 2 , and 3 ( as well as TNFSF14 ) in the avian lineage could explain these observations ( Figure S5 ) . Absence of specific genes from the three avian genomes further implies that particular genomic regions are intrinsically difficult to clone and/or sequence with the traditional Sanger and NGS methods . Finally , clear differences between birds and mammals exist in the size of the pattern recognition receptor families . For example , there are only six NODLR family members in each of the three avian species , in contrast to 22 and 32 in human and mouse , respectively ( Table 6 and Figure S6 ) . These are cytoplasmic receptors that recognize a range of ligands that activate caspases , and elicit an inflammatory response . A recent analysis revealed hundreds of NODLR genes in fish [49] with homologs of all mammalian genes . It is therefore clear that NODLR genes were lost during the evolution of the avian genomes . In contrast , while similar numbers of TLRs are found in birds and mammals , evolutionary histories of gene gain , loss , and conversion are complex ( Figure S7 ) [50]–[52] . The avian TLR1A/B and TLR2A/B genes are orthologs of mammalian TLR1/6/10 and TLR2 , respectively . All three birds have lost TLR8 and 9 but retained TLR7 . The avian TLR21 is the ortholog of mouse TLR13 , which was lost in the human lineage , and TLR15 appears to be unique to the avian lineage . Approximately 6 . 94% of the turkey genome consists of interspersed repeats , most of which belong to three groups of TEs , the CR1-type non-LTR retrotransposons , the LTR retrotransposons , and the mariner-type DNA transposons ( Table 7 and Dataset S1 ) . The CR1 group of TEs is the most abundant , occupying 4 . 81% of the genome , which is likely an underestimate because a number of highly degenerate and low copy number CR1-type elements remain to be characterized . Overall , the turkey and chicken genomes are very similar with respect to repeat content and the types of predominant TEs [4] , [53] with high sequence similarities between major TEs . For example , CR1_B in turkey and chicken share ∼91% nucleotide identity over a 2 Kbp region , the Birddawg_I LTR retrotransposons share ∼89% identity over a 3 . 6 Kbp region , and the mariner transposon Galluhop shares ∼91% identity over the entire 1 . 2 Kbp of the full-length element . Similar to the chicken , the Galluhop repeat in turkey is associated with a deletion derivative of ∼550 bp . Repetitive sequences are among the fastest evolving sequences in the genome . Therefore , the conservation of the repeat elements and sequences between the turkey and chicken is indicative of very stable genomes given a divergence time of 30–40 M years . Genome projects enable the collection of large supermatrices of alignable nucleotide sequences for phylogenetic analysis . Galliform phylogeny was re-examined by collecting sequences from the turkey and chicken genomes for 42 loci . These sequences were assembled into the largest supermatrix available for the order , containing 83 galliform species representing 73 genera , with three anseriform outgroup species . With several whole mitochondrial sequences , two genomes , and repeated use in multiple studies , 37 taxa were represented by 11 or more loci , and 12 taxa by more than 20 loci , providing data-rich anchor points that bridged locus sets throughout the tree . For the turkey , the main finding was its close relationship with the Central American ocellated turkey Agriocharis ( Meleagris ) ocellata ( 94% bootstrap support ) and the relation to the grouses within the phasianids ( Figure S10 ) . The turkey-grouse clade has been recovered in several [59]–[61] but not all previous multi-locus studies . The average bootstrap support for the nodes was high and the topology reproduced many features of previous studies , with monophyly of the megapodes , cracids , numidids and odontophorids , and polyphyly of the Perdicinae and Phasianinae within the phasianids . Grouping of an African bird ( Ptilopachus petrosus ) traditionally classified as a phasianid with the New World quails as recently observed [59] is supported , with the three loci independently reproducing this clustering . The same was true when P . nahanii was used instead of P . petrosus . Polyphyly of Francolinus was expected [62]; however , the implied polyphyly of Lophura was not . Increased throughput and decreased costs of NGS technologies facilitate cost- and time-effective sequencing of genomes . The turkey genome sequence described herein represents the first eukaryotic genome completely sequenced and assembled de novo from data produced by a combination of two NGS platforms , Roche-454 and Illumina-GAII . This genome project is a first where the majority of the production cost was invested in analysis and interpretation rather than generating sequence , and that the assembly is comparable in genome coverage to the predominantly Sanger-based sequences of the chicken and zebra finch . The sequence assigned to the chromosomes covers approximately 93% of the turkey genome . The quality of this sequence makes it a valuable resource for comparative genomics including identification of thousands of SNVs amenable to whole genome analyses . The turkey sequence confirms and extends the previously known high synteny between the turkey and chicken genomes [3] . These two avian species are remarkably similar with only 30 predicted rearrangements ( mainly small inversions ) distinguishing their genomes , despite last sharing a common ancestor about twice as long ago as the common ancestor of mice and rats or humans and gibbons . Chromosome rearrangements that occurred show a trend towards more acrocentric chromosomes in the turkey than in the chicken . The stability of galliform genomes is further confirmed by the overall conservation of gene sequences and repeat families . At less than a third the size of mammalian genomes , a greater proportion of the turkey genome ( ∼10% ) is under selective constraint versus mammals where the fraction of conserved nucleotides is approximately 5% . This also reflects the reduced percentage of the turkey genome comprised of interspersed repeats ( 7% ) . Whereas genomes of close relatives allow for analysis of rapidly changing sequence , those of distant species help elucidate regions conserved during vertebrate evolution . Gene families present only in birds provide a broad perspective on lineage-specific evolution . For example , variation in gene content between birds and an egg-laying mammal ( platypus ) shows functions shared by egg-laying animals in general as well as those unique to egg-laying birds . Likewise , genes specific to mammalian characteristics such as tooth formation have been lost in avian species . Some gene families such as TLRs of the innate immune system show complex evolutionary histories of gene gain , loss , and gene conversion between mammalian and avian species . The adaptive immune system is a relatively recent innovation peculiar to the vertebrates and provides a valuable framework for genome comparisons [63] . Genes involved in the control and regulation of the immune response towards invading pathogens are subject to strong selective pressures: the so-called “arms race” between pathogen and host . The result has been exceptional sequence divergence between the immune genes of vertebrate species , in particular those between birds and mammals [64] . Additionally , many immune genes belong to gene families that have been subject to lineage specific expansions and contractions , facilitating the evolution of new functions to combat pathogenic challenges . There are many fundamental differences between the immune systems of birds and mammals , including the major histocompatibility complex ( MHC ) structure [65] , absence of lymph nodes in birds [46] , and different mechanisms of somatic recombination in the generation of antibody diversity [66] . From an evolutionary perspective , the turkey and chicken provide an interesting case for comparative study . These two genomes have undergone intense artificial selection in recent decades for similar production traits , yet their differentially evolved genes included more functioning in transcriptional regulation in turkey , and more functioning in protein turnover and cell proliferation in chicken . Comparative genomics can provide additional insights into the response of the galliform genomes to this recent period , as well as to their longer histories of domestication . The turkey genome sequence can enhance the discovery of genetic variations underlying economically important quantitative traits , further maximizing the genetic potential of the species as a major protein source .
Vertebrate whole genome sequence assembly is aided by decreased variability in the target genome . To this end , a female turkey “Nici” ( donated by Nicholas Turkey Breeding Farms ) identified as NT-WF06-2002-E0010 was chosen for sequencing; Nici is also the source DNA for the two BAC libraries that have been characterized [12] . Nici is from an inbred sub-line ( i . e . , sib-mating for nine generations ) originally derived from a commercially significant breeding line . From her pedigree , Nici has an increased inbreeding coefficient of 0 . 624 relative to the founder breeding line . As a prelude to initial genome sequencing , heterozygosity of Nici was compared with that of individuals from several breeder lines by genotyping 147 randomly distributed microsatellites [12] . Mean heterozygosity for Nici was determined to be 0 . 31 compared to 0 . 33 for other commercial birds . Further SNP genotyping results found Nici was homozygous at 293 of 333 SNPs ( 87 . 99% ) compared to an average of 275 ( 81 . 73% ) for birds from a Beltsville Small White flock closed for 30 years . Of note , all sequence data accumulated to date suggest that Nici is monomorphic at the MHC , typically the most polymorphic region of the genome [67] . It is noteworthy that NGS depth of coverage allowed for the use of a genome that was only partially inbred . Celera Assembler release 5 . 3 was used to produce the assembly . The assembly process can be summarized to the following major stages: Stage 1 ( gatekeeper ) : input of reads and quality control Stage 2 ( overlapper ) : computation of read overlaps and trimming of poor quality sequence based on the overlaps Stage 3 ( unitigger ) : initial assembly of uniquely-assemblable contiguous chunks of sequence based on the overlaps Stage 4 ( cgw ) : scaffolding of unitigs based on mate pair data , followed by merging overlapping unitigs into contigs Stage 5 ( consensus ) : computation of consensus sequences for the contigs There are multiple choices of the assembler modules available for overlapping and unitigging . The traditional OVL overlapper was originally designed for Sanger reads . The more advanced MER overlapper was designed to account for the homopolymer errors that are common in 454 read data . The MER overlapper is more accurate , although several times slower , on pure 454 assemblies . Surprisingly , with the combined Illumina and 454 Titanium data , the MER overlapper had no advantage over OVL , which suggests that the homopolymer errors are less pronounced in the latest Titanium data . Because BOG ( best overlap graph ) is more tolerant of highly variable read sizes ( 74 bp to 366 bp ) , the newer BOG unitigger was used instead of the original unitig module . Three maps were used to produce a Combined Map ( CMap ) for alignment of assembled sequence to chromosomes . The CMap had 31 , 769 markers that mapped both to the assembly and to the turkey chromosomes MGA1 through MGA30 . Maps for the sex chromosomes W and Z were not used due to fragmentary marker coverage . Instead , scaffolds that aligned only to chicken W and Z chromosomes were identified and then ordered and oriented according to the chicken coordinates . A detailed comparative BAC contig physical map for turkey [16] was generated based on over 43 , 000 BES , over 80 , 000 BAC fingerprints , and over 34 , 600 BAC locations assigned by hybridization to overgo probes corresponding to 2 , 832 loci [70] . Two different BAC libraries were used: CHORI-260 generated by the Children's Hospital of Oakland Research Institute and 78TKNMI generated at Texas A&M University . Comparative BAC contigs were assembled based on: ( 1 ) consistent ( correct strandedness and separation distance ) alignments of mate-paired BES to the chicken genome sequence ( Build 2 , May 2006 , http://genome . ucsc . edu ) , ( 2 ) hybridization to unique overgo sequence probes aligned with the chicken genome , and ( 3 ) BAC fingerprint-based contigs [16] . The BAC contig physical map , along with the BES , provides a tool for aligning scaffolds from the turkey sequence to turkey chromosome regions as well as for identification of rearrangements between the chicken and turkey genomes . ( Regularly updated versions of this map are available at http://poultry . mph . msu . edu/resources/resources . htm#TurkeyBACChicken , and it can also be accessed in graphical form at http://birdbase . net/cgi-bin/gbrowse/turkey09/ , see Text S1 . ) The current number of contigs , end sequence matches to the chicken genome and lengths are provided in Table S12 . Most gaps between contigs are due to regions of low BAC density ( particularly on microchromosomes MGA18 and 24–30 and on the sex chromosomes that are underrepresented in the BAC libraries and , in some cases , poorly assembled in the chicken sequence ) . However , some gaps are due to repetitive regions and likely sites of CNV [10] . The average size of comparative map BAC contigs on the autosomes is over 10 . 5 Mb with the N50 average autosomal contig size being about 31 Mb . Twelve chromosomes ( MGA6 , 9 , 11 , 12 , 13 , 16–21 , and 26 ) are spanned by only a single BAC contig and another five are spanned by two contigs ( MGA2 , 5 , 15 , 23 , and 30 ) . In addition to the previously known centric split of GGA2 to MGA3 and 6 and fusion of acrocentric MGA4 and 9 to create the metacentric GGA4 , the comparative map suggests movement of more interstially positioned chicken centromeres to positions at or near the telomeres on MGA2 , 7 , 10 , 11 , 12 , and 13 . Although MGA3 and 7 both contain short p arms visible in the turkey karyotype [10] , no evidence of centromeric breaks internal to sequences orthologous to that of the chicken were found on these chromosomes , although there are a couple of short terminal contigs on MGA7 that could comprise a very small p arm . Of course , there may also be repetitive sequences or other sequences that were refractory to assembly in the chicken sequence that may be located on p arms of MGA3 and 7 ( MGA8 and 14 are difficult to resolve near the likely p end telomere due to multiple rearrangements , and microchromosomes MGA18 and 25–30 tend to be fragmented in the BAC map and less well assembled in the chicken sequence due to poor BAC coverage ) . A total of 768 SNPs were genotyped on a F2 population of two genetically different commercial turkey lines that consisted of 18 full sib families with a total of 948 offspring . SNPs were genotyped using the Illumina Golden Gate assay . Of the 768 SNPs , 458 were eventually used to build linkage maps for 27 chromosomes ( MGA1–17 , 19–26 , 28 , 30 ) ( Table S13 ) . The linkage map was constructed with a modified version of CRI-MAP software kindly provided by Drs . Liu and Grosz of Monsanto . All markers were checked for non-Mendelian inheritance errors using the option “prepare . ” Linkage maps for the individual chromosomes were constructed in a number of iterative rounds using the “build” option within CRI-MAP starting with a threshold of LOD = 5 with subsequent stepwise lowering the LOD threshold until LOD = 0 . 1 . Closely linked markers not separated by recombination events were ordered according to their location on the chicken sequence map ( build WASHUC2 ) . The order of markers in the final map was verified using the “flips” option . Strong SNVs are positions at which: ( 1 ) at least three reads support each nucleotide variant , ( 2 ) the sum of the top three quality values for each variant is at least 60 , and ( 3 ) the overall depth of coverage is at most 30 . For this analysis , gaps in the multiple alignments were assigned a quality value equal to the minimum quality of the flanking bases . In addition , if the SNV was an indel within a homopolymeric run , at least one Illumina read was required supporting each variant . The support and quality value thresholds should reduce chance sequencing errors to 1/1 , 000 , 000 , and the 30-fold depth of coverage threshold should filter out apparent variations caused by near-identical repeats . The requirement for Illumina reads verifying homopolymer indels was used to filter out well-known 454 sequencing biases [71] . Weak SNVs are similar to strong SNVs but with relaxed thresholds . For weak SNVs , at least two reads had to support both variants , and the sum of the top two quality values for each variant had to be at least 45 . The restriction on the depth of coverage was removed . As with strong SNVs , if the variant was an indel within a homopolymer , at least one Illumina read supporting each variant was required . Draft annotation was generated using two independent methods . First , a draft annotation of 12 , 206 putative protein coding loci was generated by combining evidence from multiple sources using JIGSAW [72] . Evidence for genes included spliced alignments of known proteins and mRNAs from multiple vertebrates , and expressed sequence tags ( ESTs ) from chicken and turkey . Ab initio gene predictions by Twinscan [73] and GlimmerHMM [74] ( trained on chicken genes ) were also considered by JIGSAW but with a very low relative weight , such that no gene models were based solely on ab initio predictions . The protein mappings were given the highest weight , followed by full-length mRNA alignments and then EST alignments . Proteins and mRNAs were taken from the most recent Ensembl gene builds for chicken , zebra finch , and green Anole lizard , and from the GenBank RefSeq database ( the “other vertebrates” section plus mouse and zebrafish genes ) . Many thousands of genes and gene fragments were eliminated from the initial predictions if the computational evidence was insufficient . Second , the gene-finding pipeline at Ensembl [75] , which also uses a combination of known proteins , ESTs , and cDNAs to annotate genes , was used to generate an independent set of protein-coding genes and noncoding RNAs . After combining the two gene lists , the total number of distinct protein coding loci was 15 , 093 plus 611 noncoding RNA genes , for a total of 15 , 704 genes . Some loci were identified as producing multiple distinct proteins due to alternative splicing , giving 16 , 217 distinct protein sequences . Orthologs between turkey , chicken , and human proteins were defined using sequence homology , phylogenetic trees , and conservation of synteny . Homologous pairs and orthology type are available from the version 57 Ensembl Compara database ( http://e57 . ensembl . org ) . It was assumed that all 1∶1 orthologs were correct and were used to define conserved syntenic regions . Further orthologs were then defined from the one-to-many and many-to-many relationships , if the homologs mapped to a conserved syntenic region . This allowed for a 7%–8% increase in the number of defined orthologs for all species ( Table S14 ) . Multiple ( three-way ) alignments were built on the turkey , chicken [4] , and zebra finch [7] genomes using Pecan [33] . Pecan is a global multiple sequence aligner that assumes no major rearrangements in the input sequences . Thus , sets of collinear segments were defined before aligning the sequences . Searches were based on the turkey-chicken and chicken-zebra finch pairwise BLASTZ-net alignments [76] . The genomes were repeat-masked using RepeatMasker ( www . repeatmasker . org ) , followed by BLASTZ analysis , to find all highly similar regions , which were grouped in chains using the axtChain software and refined using the netChain software [77] . GERP [34] was used to get both per-base conservation scores and conserved elements . Searches were combined based on similarities and de novo repeat analysis approaches . For de novo repeat analysis , RepeatScout [80] and LTR_Struc [81] were used under default conditions . Overall , 944 repeat elements were identified , many of which were redundant . LTR_Struc uncovered three LTR retrotransposons that had both LTRs and the reverse transcriptase domain . During similarity searches , known chicken repeats ( Repbase Update , http://www . girinst . org ) as well as representatives of different types of TEs were used as query . Repeats identified by RepeatScout were classified by comparison with known chicken repeats as well as with representative LTR retrotransposon protein sequences , non-LTR retrotransposon ( or LINE ) protein sequences , and DNA transposases . Whenever possible , the chicken homolog was used to assign names for the turkey TEs . Sequences other than those derived from the turkey genome project were collected from NCBI ( www . ncbi . nlm . nih . gov ) using BLAST [97] , retrieved through searches in Ensembl ( www . ensembl . org ) , and identified by BLAT searches on the UCSC Genome Bioinformatics database ( genome . ucsc . edu ) . ESTScan v . 2 [98] , [99] was used to correct predicted coding regions for frameshift and other sequencing errors . The alignments of amino acid and coding sequences used for the analysis of gene conversion and the construction of phylogenetic trees were generated with MUSCLE v . 3 . 7 [100] . Gene trees reconciled with species trees were calculated using TreeBeST v . 1 . 9 . 2 [101] and trees were visualized using Archaeopteryx [102] . Perl scripts and modules from Bioperl [103] were used to manipulate sequence and phylogenetic data . DNA sequences for 42 high-coverage loci ( 11 mitochondrial ) were collected from GenBank , for all Galliformes and for three outgroup Anseriformes species . One representative species of each genus was selected based on frequency of coverage , and additional representatives were chosen for the known polyphylous genus Francolinus and in cases of complementary locus coverage . The 86 final species were represented by 10 . 6 kb on average . For Coturnix coturnix , loci from the complete mitochondrial sequence of C . japonica ( often considered a subspecies of the former ) were used . Northura maculosa was classified as a cracid at NCBI but did not join the other cracids in preliminary trees; this GenBank entry was apparently a misspelling and misclassification of the tinamou N . maculosa and was removed from the study; GenBank was notified of the problem . For each locus , sequences were aligned and the unmasked alignments were concatenated , partitioned by gene ( except that the two mitochondrial rRNA genes were fused as were the eight mitochondrial coding sequences ) . A maximum likelihood tree was constructed using RA×ML with the GTR-GAMMA model and 100 full bootstraps were taken .
|
In contrast to the compact sequence of viruses and bacteria , determining the complete genome sequence of complex vertebrate genomes can be a daunting task . With the advent of “next-generation” sequencing platforms , it is now possible to rapidly sequence and assemble a vertebrate genome , especially for species for which genomic resources—genetic maps and markers—are currently available . We used a combination of two next-generation sequencing platforms , Roche 454 and Illumina GAII , and unique assembly tools to sequence the genome of the agriculturally important turkey , Meleagris gallopavo . Our draft assembly comprises approximately 1 . 1 gigabases of which 917 megabytes are assigned to specific chromosomes . Comparisons of the turkey genome sequence with those of the chicken , Gallus gallus , and the zebra finch , Taeniopygia guttata , provide insights into the evolution of the avian lineage . This genome sequence will facilitate discovery of agriculturally important genetic variants .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Methods"
] |
[
"genetics",
"and",
"genomics/comparative",
"genomics",
"genetics",
"and",
"genomics/genetics",
"of",
"the",
"immune",
"system",
"computational",
"biology/comparative",
"sequence",
"analysis",
"evolutionary",
"biology/genomics",
"genetics",
"and",
"genomics/genome",
"projects",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics"
] |
2010
|
Multi-Platform Next-Generation Sequencing of the Domestic Turkey (Meleagris gallopavo): Genome Assembly and Analysis
|
Mutations in human Exostosin genes ( EXTs ) confer a disease called Hereditary Multiple Exostoses ( HME ) that affects 1 in 50 , 000 among the general population . Patients with HME have a short stature and develop osteochondromas during childhood . Here we show that two zebrafish mutants , dackel ( dak ) and pinscher ( pic ) , have cartilage defects that strongly resemble those seen in HME patients . We have previously determined that dak encodes zebrafish Ext2 . Positional cloning of pic reveals that it encodes a sulphate transporter required for sulphation of glycans ( Papst1 ) . We show that although both dak and pic are required during cartilage morphogenesis , they are dispensable for chondrocyte and perichondral cell differentiation . They are also required for hypertrophic chondrocyte differentiation and osteoblast differentiation . Transplantation analysis indicates that dak−/− cells are usually rescued by neighbouring wild-type chondrocytes . In contrast , pic−/− chondrocytes always act autonomously and can disrupt the morphology of neighbouring wild-type cells . These findings lead to the development of a new model to explain the aetiology of HME .
Mutations in human EXT1 and EXT2 confer an autosomal dominant disorder called HME [1] , [2] , [3] . Both EXT1 and EXT2 encode glycosyltransferases that together form a hetero-oligomeric complex in the Golgi and catalyse the polymerisation of sugars to form heparan sulphate ( HS ) ( for review see [4] ) . Patients with HME have a short stature and during childhood develop osteochondromas ( also called cartilaginous exostoses ) that first appear near the growth plate regions of their skeleton . Osteochondromas are made up of a cartilage cap that resembles a growth plate and a bony collar that forms a marrow cavity that is contiguous with the underlying bone . While osteochondromas are normally benign , they can lead to complications and patients have a 1–2% risk of developing chondrosarcoma or osteosarcoma . Most of the tested patients with HME are heterozygous for mutations in either EXT1 ( 41% ) or EXT2 ( 30% ) [5]–[7] . Determining the genetic basis for the cases that cannot be attributed to EXT genes ( 29% ) is essential for counselling HME patients . The sporadic and dominant nature of osteochondromas formation in HME patients has led to the proposal of two genetic models ( For discussion see [8] ) . Osteochondromas may arise from a loss-of-heterozygosity ( LOH ) at one of the EXT loci in skeletal cell resulting in unregulated growth and clonal expansion . In support of this model , LOH due to somatic mutations or aneuploidy has been identified in a small number of the osteochondromas analysed [9] , [10] . In addition , HS is absent in chondrocytes within osteochondromas which is consistent with a complete loss of EXT function due to LOH [11] . Contrary to this model , HS is secreted and it is likely that a homozygous mutant chondrocyte would be rescued by contact with neighbouring cells . The alternative model is that reduced EXT gene dosage causes reduced HS synthesis that results in a structural change in the growth plate . This change allows chondrocytes to occasionally escape normal developmental constraints to give rise to an osteochondroma . The finding that the majority of analysed exostoses do not show a second mutation in the EXT gene family lends support to the gene dosage theory [10] . Resolving between these two models could play an important role in designing future treatment for HME patients . Skeletal histology in fish is comparable to that of tetrapods [12] and the development of the cranial skeleton of zebrafish has been well described [13] , [14] . The precartilage condensations that will give rise to the cartilaginous skeleton begin to appear during the second day of development . Condensations give rise to two cell types: the cells of the perichondrium ( a sheath that encapsulates the cartilage ) and the chondrocytes that begin to secrete the cartilage matrix . As the skeleton forms , some chondrocytes flatten and intercalate to form a column that gives rise to rod shaped cartilage elements . Alternatively , chondrocytes flatten to form a single layer of tessellated cells that give rise to plate-like elements [15] . Much of the cartilaginous skeleton is then replaced by bone in a process that resembles endochondral ossification in tetrapods . These bones are referred to as cartilage bones . Also , like tetrapods , some of the bony skeleton does not form from a cartilage template . These bones are called intramembranous ( or dermal ) bones . Large-scale genetic screens have identified many genes that disrupt skeletal development in zebrafish [16]–[18] . Here we have focused on two genes that are required for skeletal development , dak and pic . Both dak and pic , along with a third gene boxer , are also required for fin development [19] , [20] and axon sorting [21] suggesting that the mutated genes act in a common pathway . We have previously shown that dak and boxer encode glycosyltransferases required for HS synthesis ( ext2/dak , extl3/boxer ) [22] . In this study we present evidence that pic encodes a putative PAPS transporter ( 3′-phosphoadenosine 5′-phosphosulfate transporter , PAPST1 ) that is required for sulphation of glycans . We show that dak and pic are required for cartilage morphogenesis , but surprisingly not for early cartilage differentiation . We show that hypertrophic differentiation of chondrocytes and subsequent cartilage bone formation is lost in mutant larvae . We also show that intramembranous bone formation is reduced due to a reduction of osteoblast differentiation . We show that dak and pic can act cell autonomously during chondrogenesis , and based upon these findings propose a model for how LOH could account for osteochondroma formation in HME patients .
Chondrocytes in osteochondromas often differ from chondrocytes found in normal growthplates . Instead of being flattened and forming long columns of cells , they are usually rounded and form clusters of cells [11] , [23] , [24] . We wondered whether chondrocytes in dak−/− and pic−/− mutants behave in a similar way . Although most of the cartilage elements are present in both dak−/− and pic−/− larvae , the elements are shorter and thicker than wild-type ( Figure 1A , D , G ) [18] . In dak−/− larvae , anterior cartilages tend to have more cells than wild-type and posterior cartilages have less . For example , in 144hpf dak−/− larvae the Meckel's cartilage much larger than in wild-type , while ceratobranchial 4 is small or even absent ( H . R . and M . W . unpublished ) . Wild-type chondrocytes flatten along the longitudinal axis ( stack ) in most elements ( Figure 1A–C ) and in rod shaped elements the cells intercalate to form a single column ( Figure 1B ) . Some elements have regions where stacking is not obvious ( arrowhead in Figure 1C ) especially in regions adjacent to joints ( arrowheads in Figure 1B ) . In all dak−/− larvae , all chondrocytes are round and do not form into columns ( Figure 1D–F ) . While pic−/− larvae show lower expressivity , most larvae have a loss of chondrocyte organisation that resembles that seen in dak−/− larvae ( Figure 1G–I ) . One striking difference between dak−/− and pic−/− larvae is that pic−/− larvae do not stain with Alcian Blue at pH1 . 0 ( Figure 1D , G ) but do stain at pH2 . 5 ( M . W . and A . C . unpublished ) . As Alcian Blue preferentially stains sulphated groups at low pH [25] , one possible explanation for the lack of staining in pic−/− larvae is a loss of sulphation of glycans and other sulphated moieties . To investigate this further , we used antibodies for HS [26] , CS [27] and KS [28] and found that whereas HS is reduced in both dak−/− and pic−/− larvae ( Figure 2A–C ) , CS and KS are reduced only in pic−/− larvae ( Figure 2D–I ) . The antibody used to detect heparin , 10E4 , recognizes an epitope that is localised to basal laminae , but not found in the developing zebrafish cartilage ( Figure S2 ) . We further analysed HS composition using heparan lyase digestion followed by HPLC [22] . Peaks generated by zebrafish larval extracts were compared to 6 known standards . pic−/− embryos show a severe reduction of sulphated disaccharides , but surprisingly also show a reduction in unsulphated disaccharides ( Figure 2J ) . This perhaps indicates that the loss of sulphation affects processing or stability of heparan . However , it is important to note that in the absence of sulphation , heparan synthesis may generate atypical disaccharides which would not be identified by this analysis [29] . Together these data confirm that pic is required for sulphation of proteoglycans . 29% of patients with HME do not carry mutations in EXT1 or EXT2 genes . In order to help identify new candidate genes , we positionally cloned pic . Using SSLP microsatellite markers , we mapped the pic locus to a 3 . 3cM interval on chromosome 20 ( Figure 3A ) . Using the zebrafish RH map , we placed a zebrafish gene with homology to human and Drosophila PAPST1 [30] , [31] in the same interval ( Figure 3C and Table S1 ) . As PAPST1 transports PAPS into the Golgi ( PAPS being the universal donor for sulphation ) , it is a good candidate gene to explain the loss of proteoglycan sulphation . We then sequenced papst1 cDNA from picto216z/to216z and picto14mx/to14mx mutant embryos ( Figure 3B ) . The picto216z allele has a nucleotide transition ( G to A ) at position 390 in the third exon , creating a stop codon . The picto14mx allele is a genomic deletion that results in an in-frame deletion of all of exon 3 in the cDNA . To confirm that mutations in papst1 result in the pic−/− phenotype , we expressed the wild-type zebrafish papst1 cDNA under the control of a heterologous promoter in pic−/− embryos . Expression of wild-type papst1 in a single cell was sufficient to rescue staining of KS in the notochord ( Figure 3D–F ) . To determine the expression pattern of papst1 , we performed wholemount in situ hybridization with the full-length cDNA . Consistent with its role as a general component of the cellular sulphation machinery , papst1 is expressed ubiquitously ( Figure S1 ) . As both alleles are predicted to result in severe truncation of the PAPST1 protein and have identical phenotypes , they are likely to be null alleles . The LOH model for HME raises the question of whether EXT−/− cells could differentiate into all the cell types that make up an osteochondroma . To address this question , we first tested whether perichondral cells and chondrocytes differentiate normally in dak−/− and pic−/− mutant larvae . Surprisingly , expression of three markers of early chondrogenesis occurs in both mutants as in wild-type larvae ( Figure 4A–I ) . To test whether the perichondrium is present , we used a marker , gdf5 , which is expressed in the perichondrium of the ceratohyal [32] . Expression is present in the ceratohyal of both mutants ( arrows in Figure 4J–L ) . The flattened cells of the perichondrium can also be seen in toluidine blue stained sections of mutant larvae , indicating that HS is dispensable for the differentiation and morphogenesis of these cells ( arrows in Figure 4M–O ) . Together , these data suggest that the cartilage and perichondral components of osteochondromas could be formed by EXT−/− cells . As osteochondromas contain a bony collar , we next tested whether bone forms normally in homozygous dak−/− and pic−/− larvae . Intramembranous ( dermal ) and cartilage bones appear during early larval development [13] . Alizarin Red staining for bone at 144hpf shows a strong reduction of calcification in both bone types ( Figure 5J–L ) . Consistent with this , there is a strong reduction of several markers for osteoblast differentiation in both mutants at 96hpf ( Figure 5A–I and see Table S2 ) . As cartilage hypertrophy precedes endochondral ossification , we also tested whether expression of the hypertrophic marker , collagen10a1 is affected at 144hpf . We found that chondrocyte expression of collagen10a1 is absent in both mutants ( arrows in Figure 5 M–O ) . Together these data suggest that EXT−/− cells in an osteochondroma would not contribute significantly to the formation or remodelling of bone . Thus it is likely that EXT−/+ cells would be recruited to an osteochondroma and take part in the formation of the bony collar . To determine when chondrocyte behaviour is first affected in dak−/− embryos , we looked at condensation formation in the jaw . Early condensations within the first arch were visualized at 45 and 50hpf using sox9a as a chondrogenic marker ( Figure 6A–D ) . Even at this early stage , dak−/− chondrocytes appeared more round than those in wild-type embryos ( as judged by nuclear morphology , arrows in Figure 6C , D ) . In anterior condensations , the level of sox9a expression is variable and usually stronger in dak−/− condensations , consistent with the increase in chondrocyte cell number seen in anterior arches ( Figure 6B , D ) . We also examined early stacking within condensations of the second arch at 54 and 58hpf . During this time wild-type chondrocytes intercalated to form a single cell layer , flattened perpendicular to the growth axis and began to secrete cartilage matrix ( Figure 6E , E' , G , G' ) . Although dak−/− chondrocytes also began secreting matrix , the cells showed no signs of undergoing morphogenesis ( Figure 6F , F' , H , H' ) . Taken together , these data suggest that the primary cartilage defect in dak−/− larvae is the loss of chondrocyte organisation . Similar results were obtained with pic−/− larvae but with lower and more variable expressivity ( data not shown ) . One caveat in the LOH model is that HS is secreted and thus an EXT−/− cell that arises would be rescued by neighbouring cells . To ascertain whether clones of dak−/− cells behave autonomously when juxtaposed to HS secreting cells , we transplanted dak−/− cells into wild-type embryos . The transplantations were done at sphere stage , then the embryos were allowed to develop for several days before fixation and analysis . We found that in most cases ( 19/24 transplants ) , the transplanted mutant cells stacked normally when juxtaposed to wild-type cells ( arrow , Figure 7B ) . This alone would argue that single EXT−/−cells in HME patients would be unable to form exostoses and thus would refute the LOH model . However in some cases ( 5/24 ) mutant cells behaved autonomously and failed to stack or intercalate ( arrowheads , Figure 7C–E ) . These mutant clones grew out from the edge of the cartilage perpendicular to the wild-type stacks . Given that HS may not diffuse far , it is plausible that dak−/− cells on the edge of the cartilage lack sufficient contact to be rescued by neighbouring wild-type chondrocytes . This explanation is consistent with studies of HME patients that have found that osteochondromas are first seen on the edge of the growth plate just beneath the perichondrium [33] , [34] . Importantly , this results shows that cells that ext2−/− chondrocytes can behave autonomously in zebrafish and thus EXT−/− chondrocytes in humans could be responsible for the formation of osteochondromas . In all transplants examined ( 39/39 ) , pic−/− cells behaved autonomously and failed to stack or intercalate ( arrowheads , Figure 7F , G ) . In addition , when juxtaposed to pic−/− cells , wild-type cells often adopted the mutant rounded morphology ( arrowheads , Figure 7H , I ) . In many cartilage elements , both stacked and non-stacked clusters of wild-type chondrocytes were seen ( 20/39 ) . Significantly , whenever wild-type cells stacked , they flattened and formed columns that were oriented to the longitudinal axis of the cartilage element , even when few wild-type cells were present in a pic−/− cartilage element ( arrow , Figure 7I ) . This suggests that there is a signal that polarizes chondrocytes so that stacking is oriented to the correct axis and that this signal is still present in pic−/− larvae . These findings are also consistent with the LOH model and suggest that patients with mutations in PAPST1 may have more severe clinical symptoms .
While much is known about the genetic basis of HME , the mechanism of osteochondroma formation is poorly understood . In this study , we show that zebrafish is an excellent model for HME and our findings support the LOH model in several ways . First , proliferating chondrocytes in osteochondromas resemble those seen in homozygous dak−/− larvae: they are rounded and do not form into columns of cells [11] , [23] , [24] . Second , we show that homozygous mutant cells differentiate into chondrocytes , despite the absence of morphogenesis . Third , transplants with dak−/− cells into wildtype animals show that although most homozygous mutant clones were rescued , some dak−/− chondrocytes behaved autonomously . The rescue of mutant cells is presumably due to HS secretion from neighbouring wild-type cells , but may also be due to other secreted factors . The results presented here as well as from other studies suggest a model for how LOH could result in osteochondroma formation ( Figure 8 ) . Although our results lend credence to the LOH model , they do not refute the gene dosage model and it is possible that both mechanisms play a role . Several studies of the role of EXT genes during mouse skeletogenesis have been published and these favour the gene dosage model for HME . In mice homozygous for a hypomorphic allele of Ext1 ( Ext1gt/gt ) or heterozygous for a targeted deletion ( Ext1+/− ) , the chondrocytes of limb growth plates show delayed hypertrophic differentiation and endochondral ossification [35] , [36] . Given that HS is known to regulate the activity of many signalling pathways , the researchers tested whether a signalling defect could explain the Ext1 mutant phenotype . Indian Hedgehog ( IHH ) , a signalling protein that normally acts in the growth plate to block hypertrophy and terminal differentiation of chondrocytes was found to have increased activity in mutant mice . The model for these results is that in wild-type animals , HS normally acts to limit diffusion of IHH and thereby allow chondrocytes to become hypertrophic [35] , [36] . The authors favour the gene dosage model for HME and propose that hereditary osteochondromas are caused in part by a delay in chondrocyte hypertrophy caused by excessive IHH signalling [35] , [36] . In contrast , mice heterozygous for a targeted deletion of Ext2 ( Ext2+/− ) have normal limb growth plates and there is no discernible effect on IHH diffusion [37] . However , Ext2+/− mice do have osteochondroma-like outgrowths on their ribs . These authors also favour the gene dosage model but do not find evidence to support a role for IHH . In comparison , dak−/− and pic−/− larvae show a more severe skeletal phenotype , perhaps due to a stronger reduction of HS . Whereas mice homozygous for null mutations in Ext1 or Ext2 in mice arrest during gastrulation [37] , [38] , dak−/− and pic−/− embryos can gastrulate probably due to maternally deposited RNAs ( [22] and Figure S1 ) . By 36hpf , HS is only weakly detectable in dak−/− and pic−/− embryos ( by immunohistochemistry , AC and MW unp ) . The early reduction of HS has enabled us to identify cartilage morphogenesis as the primary defect during skeletogenesis . Indeed , chondrocytes in both Ext1gt/gt and Ext2+/− mice show a mild disruption of the columnar organization within the growth plate [36] , [37] . Thus it is likely that complete loss of stacking , as early as the cartilage condensation phase , would be evident with a more severe reduction of mouse Ext gene function . IHH signalling is not likely to be responsible for the cartilage morphogenesis phenotype because neither of the two zebrafish IHH genes is expressed until 2 days after chondrocyte stacking begins . Furthermore , pharmacological inhibition of Hedgehog signalling during skeletogenesis does not affect chondrocyte stacking ( MW and AC unpublished ) . One plausible candidate for this signal in zebrafish is wnt5b which encodes a ligand for the non-canonical WNT signalling pathway [39] . The evidence for this is that wnt5b is expressed in cells surrounding cartilage condensations and mutations in wnt5b result in reduced chondrocyte stacking ( MW and AC unpublished ) . As the wnt5b−/− cartilage phenotype is mild compared to that of dak−/− larvae , other members of the non-canonical WNT family of genes may be redundant with wnt5b . An intriguing possibility is that WNT signalling and other components of the planar cell polarity system mediate chondrocyte stacking , just as they regulate convergence/extension movements during gastrulation . Although early chondrocyte differentiation is unaffected in zebrafish mutant larvae , we did find that expression of the hypertrophic marker , collagen10a1 , is lost . This is in agreement with results from Ext1 mutant mice which show a delay in chondrocyte hypertrophy due to increased IHH signalling [35] , [36] . The opposite result has been found for Ext2 mutant mice where a reduction in HS was shown to cause premature collagen10 expression [37] . Intriguingly , studies of HME patients have found evidence of premature hypertrophy in osteochondromas [23] , [24] . Determining why apparently conflicting results have been obtained in these systems will require more detailed analysis . HS has been implicated in osteoblastogenesis , however there has been no clear evidence for a developmental role to date [40] . Here we show that osteoblastogenesis is impaired by the reduction of HS mutant zebrafish larvae . Although previous Ext mutant mouse studies have focused on cartilage differentiation , a reduction in the bone mineral density of Ext1+/− mice has been observed [35] . Furthermore , osteopenia has been shown to be associated with HME in a family that carries a mutation in EXT1 [41] . These findings suggest a new role for HS during osteoblast differentiation . While most of the tested patients with HME are heterozygous for mutations in either EXT1 ( 41% ) or EXT2 ( 30% ) , the genetic basis of the remaining cases is unknown ( 29% ) [5]–[7] . Several EXT-like genes have been shown encode enzymes required during HS synthesis and would thus make good candidate genes ( EXTL1 , EXTL2 , and EXTL3 ) . Unfortunately , none of these have been shown to carry disease-related mutations in HME patients [7] . Although mutations in sulphate metabolism can cause hereditary skeletal disorders , none of these results in the formation of osteochondromas [42]–[44] . Here we show for the first time that PAPST1 is essential for sulphation of glycans in vertebrates and that mutations in pic confer a specific phenotype that is very similar to the dak−/− phenotype . This is a surprising result because unlike dak mutations which only reduce HS , pic mutations should reduce all sulphation in the cell . Inactivation of PAPST genes in Drosophila also results in phenotypes that resemble EXT mutant phenotypes [31] , [45] . Together , these findings suggest that although other glycans may be required during vertebrate and invertebrate development , HS is the principal glycan . Furthermore , our results suggest that PAPST1 as well as other genes involved with sulphate transport and metabolism are candidate genes for HME .
The pic locus was mapped to linkage group 20 ( lg20 ) after analysing SSLP ( simple sequence length polymorphism ) markers on 700 meioses . picto14mx maps 2 . 7cM south of z1534 ( 20/740 meiosis ) and 0 . 6cM north of z8554 ( 4/698 meiosis ) [47] . This interval on the T51 radiation hybrid ( RH ) map [48] was found to contain many ESTs . A contig in the neighborhood of one of these , fc88f04 . x1 , was assembled using traces from the Sanger Centre , and found to contain a gene having homology to a human and Drosophila PAPS Transporter 1 ( PAPST1 , also known as: Solute Carrier Family 35 , Member B2 ) [30] , [31] . To confirm the location of zebrafish papst1 , primers were designed to exon 4 and analysed using the T51 panel ( papstf3 , 5′ CGTCACCACATTCTCCGGCGT 3′ and papstr3 , 5′GTGTCTGATTTCCTGAAGTGT 3′ ) . The papst1 pattern matches the pattern of other markers in the pic interval ( see Table S1 ) . To sequence alleles , cDNA was obtained from picto14mx/to14mx , picto216z/to216z larvae and wild-type larvae and then sequenced with primers papstf1 and papstf1 . 1e ( respectively , 5′ TGGCAGTTTTGTAGAGGCGGAG 3′ and 5′ GAATGCAGACGCTGTAGAC 3′ ) . Primary antibodies were anti-HS 1∶500 ( 10E4 , Europa ) , anti-keratan sulphate 1∶100 ( KS ) ( 3H1 , Developmental Studies Hybridoma Bank ) , anti-chondroitin sulphate 1∶100 ( CS ) ( CS-56 , Sigma ) , anti-collagen type II 1∶200 ( II-II6B3 , Developmental Studies Hybridoma Bank ) and anti-GFP 1∶100 ( Torrey Pines Biolabs ) . Secondary antibodies were horse anti-mouse-HRP and goat anti-rabbit-HRP ( Vector Laboratories ) . Detection was done using DAB substrate ( Vector Laboratories ) or TSA-CY3 substrate ( Perkin Elmer ) . Larvae were mounted in 70% glycerol or Vectashield with DAPI ( Vector ) . Antisense probes were made using the following cDNAs: chondromodulin1 [49] , collagen2a1a [50] , collagen10a1 [51] , growth and differentiation factor-5 ( gdf5/contact ) [32] , crossveinless2 [52] and sox9a [53] . Cloning and characterisation of zebrafish osterix will be described elsewhere . Goat anti-DIG fab fragments and NBT/BCIP substrate ( Roche ) were used to develop the in situs . To rescue pic−/− embryos , we made an expression construct by cloning the wild-type zebrafish papst1 cDNA into an hspIG vector that contains a heat shock promoter and an IRES2 ( internal ribosome entry site ) /eGFP cassette ( gift from Dr Florian Maderspacher ) . The construct was then injected into pic−/− larvae and the wild-type papst1 cDNA was expressed by heat shocking larvae at 24 hours post-fertilisation ( hpf ) , for 1 hour at 38°C . The larvae were fixed 6 hours later and a double antibody staining was performed to check for rescue . As DNA injected at the one cell stage is inherited mosaically , the eGFP reporter was used to confirm that cells that synthesise keratan sulphate ( KS ) indeed carried the rescuing construct . First the larvae were stained using anti-KS with DAB as the substrate . After photographing , the embryos were then stained using anti-GFP with NBT/BCIP as the substrate to determine which cells carried the construct . GFP donor embryos were first injected with tetramethyl-rhodamine dextran 3% ( Invitrogen ) at 1-cell stage . Then , both donors and recipients were dechorionated in pronase . Transplantation was done in E3 from sphere stage and based upon a zebrafish fate map ( Woo and Fraser , 1995 ) . At 24hpf , each recipient was screened for the presence of fluorescent rhodamine in the neural crest cells and kept until 120hpf with their donor . Larvae were then fixed in 4% PFA and Alcian Blue staining followed by antibody staining to track the GFP transplanted cells was performed . The experiment was done with picto216z/to216z or dakto273b/to273b transplanted into wild-type and vice versa .
|
Hereditary Multiple Exostoses is a disease that causes the formation of benign bone tumours in children . Besides causing severe skeletal deformity , the bone tumours can compress nerves or other tissue resulting in chronic pain . Although the tumours can usually be surgically removed , they sometimes recur or are in positions that prevent surgery . We have identified two strains of zebrafish whose offspring have skeletal defects that resemble those of patients with Hereditary Multiple Exostoses . We have found that each strain carries a mutated form of an essential gene . Importantly , these two genes are also found in humans , and thus by analysing their function in zebrafish , we may shed light on their role in humans . Our study has elucidated the roles of these genes during normal skeletal development and has allowed us to generate a model for how genetic changes give rise to bone tumours in humans .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology/cell",
"differentiation",
"genetics",
"and",
"genomics/disease",
"models",
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology"
] |
2008
|
Regulation of Zebrafish Skeletogenesis by ext2/dackel and papst1/pinscher
|
Variation in cortical connectivity profiles is typically modeled as having a coarse spatial scale parcellated into interconnected brain areas . We created a high-dimensional common model of the human connectome to search for fine-scale structure that is shared across brains . Projecting individual connectivity data into this new common model connectome accounts for substantially more variance in the human connectome than do previous models . This newly discovered shared structure is closely related to fine-scale distinctions in representations of information . These results reveal a shared fine-scale structure that is a major component of the human connectome that coexists with coarse-scale , areal structure . This shared fine-scale structure was not captured in previous models and was , therefore , inaccessible to analysis and study .
Resting state functional magnetic resonance imaging ( rsfMRI ) reveals patterns of functional connectivity that are used to investigate the human connectome [1–3] and parcellate the brain into interconnected areas that form brain systems and can be modeled as networks [4–11] . The connectivity of a single area is considered to be relatively homogeneous and typically is modeled as a mean connectivity profile . Cortical topography , however , has both a coarse scale of cortical areas and a finer scale of multiplexed topographies within areas [12–16] . Fine-scale within-area topographies are reflected in patterns of activity that can be measured with fMRI and decoded using multivariate pattern analysis ( MVPA ) [12 , 13 , 17] . Fine-scale variation in connectivity , however , has been overlooked due to poor anatomical alignment of this variation across individual brains . We ask here whether local variation in functional connectivity also has a fine-scale structure , similar to fine-scale response tuning topographies , and whether such variation can be captured in a common model with basis functions that are shared across brains . We developed a new algorithm , connectivity hyperalignment ( CHA ) , to model local variation in connectivity profiles with shared basis functions for connectivity profiles across individuals and individual-specific local topographies of those connectivity basis functions ( Fig 1 ) . The resultant common model connectome consists of transformation matrices for each individual brain , which contain individual-specific topographic basis functions , and a common model connectome space , which contains shared connectivity profiles ( Fig 2 ) . Individual transformation matrices transform an individual brain’s connectome , in its native anatomical coordinate space , into the common model space [13 , 16] . The individual transformation matrices and common model connectivity matrix are derived iteratively from training data . Validity testing is done on connectivity profiles and other functional parameters from independent test data that are hyperaligned into the common model connectome space . The results show that CHA can derive these shared basis functions from functional connectivity derived from neural activity while watching an audiovisual movie and from neural activity in the resting state . The resultant common model connectome accounts for substantially more shared variance in functional connectivity derived from both movie fMRI data and resting state fMRI data than was accounted for by previous models . This shared variance resides in fine-scale local variations in connectivity . We show further that this local variability in functional connectivity profiles is meaningful in that it is closely related to local patterns of response that encode fine distinctions among representations . Our results indicate that shared fine-scale local variation , which was not evident in previous models , is a major component of the human connectome that coexists with shared coarse-scale areal structure . Our common model connectome makes this fine-scale local variation accessible for group-level study of its network properties .
CHA afforded large increases in ISCs of connectivity profile vectors in both the movie fMRI data and the rsfMRI data ( Fig 3 and Fig 4 ) . Increases in ISCs of functional connectivity derived from movie data were distributed across all of cortex ( Fig 3 ) . ISC at a cortical node is the correlation of the one subject’s connectivity profile with the mean of other subjects’ profiles , indexing how well other subjects’ connectivity profiles can predict an individual’s connectivity profiles . Fig 3A shows a cortical map of mean ISCs of connectivity profiles in the common model connectome space as compared to ISCs in anatomically-aligned data . Fig 3B is a scatterplot of mean ISCs for individuals after anatomical alignment and CHA , which shows that CHA increased ISC for each individual and preserved individual similarity or deviance from the group . We quantify the increases in 24 functional ROIs , identified using a meta-analytic database , NeuroSynth [21] ( Fig 3C; S1 Table ) . Mean ISC of connectivity profiles across these ROIs was markedly higher in the common model connectome than in the anatomically-aligned data ( 0 . 67 versus 0 . 15; difference = 0 . 52 , 95% confidence interval , CI = [0 . 46 , 0 . 56] ) . Increases in ISCs of resting state connectivity profiles were similarly distributed across all of cortex and replicated the findings based on ISCs of movie viewing connectivity profiles ( Fig 4 ) . Fig 4A shows a cortical map of mean ISCs of resting state connectivity profiles in the common model connectome space and in data aligned with the HCP’s MSM-All method ( multimodal surface matching [22] ) . Fig 4B is a scatterplot of mean ISCs for individuals after MSM-All alignment and CHA , which shows that CHA of resting state data increased ISC for each individual and preserved individual similarity or deviance from the group . Fig 4C shows a cortical map of within-subject correlations between connectivity profiles from different resting state sessions . We quantify the increases in 26 functional ROIs , identified using a meta-analytic database , NeuroSynth [21] ( Fig 4D; S6 Table ) . Mean ISC of connectivity profiles across these ROIs was markedly higher in the common model connectome than in the MSM-All-aligned data ( 0 . 66 versus 0 . 35; difference = 0 . 31 [0 . 30 , 0 . 33] ) . ISCs of resting state connectivity profiles in the common model connectome space are slightly higher than within-subject correlations of resting state connectivity profiles ( mean correlation = 0 . 64; CI for difference = [0 . 00 , 0 . 05] ) ( Fig 4D ) . This latter result indicates that an individual’s connectome based on resting state functional connectivity is better predicted by the common model connectome , based on other subjects’ data , than by estimates based on a typical sample of that subject’s own rsfMRI data , due to the benefit of estimating connectivity profiles based on a large number of brains and the precision of CHA . The substantial increase in ISCs with hyperalignment is due in part to discovery of shared variance that was obscured by misalignment but also to suppression of unshared variance and amplification of shared variance mediated by filtering the data in the transformation step with smaller weights for nodes with unshared or noisy variance and larger weights for nodes with shared variance . To gauge the size of the effect of filtering independent of better information alignment , we calculated ISCs in data that are filtered by our algorithm but aligned based on anatomy or MSM-All ( see methods ) . ROI mean ISCs of connectivity profiles in movie data filtered with CHA but aligned based on anatomy was 0 . 22 ( CHA versus filter-control difference = 0 . 45 [0 . 39 0 . 49] ) and for HCP resting state data filtered with CHA but aligned based on MSM-All was 0 . 41 ( CHA versus filter-control difference = 0 . 25 [0 . 23 , 0 . 27] ) . These ISCs are larger than ISCs of unfiltered , anatomically and MSM-All-aligned data but , nonetheless , still markedly lower than ISCs of connectivity profiles in the common model connectome space , which is both filtered and re-aligned by CHA . We investigated the spatial specificity of the common model connectome by computing the intersubject spatial point spread functions ( PSF ) of ISCs of connectivity profiles [16] . The PSF of connectivity profiles was computed as the correlation of the connectivity profile in a cortical surface node for a given subject with the average connectivity profiles of other subjects in the same node and nodes at cortical distances ranging from 3 to 12 mm . We similarly calculated within-subject PSFs based on within-subject correlations ( WSC ) of connectivity profiles between two resting state sessions . Fig 5A shows the slopes of connectivity profile PSFs for movie data in 24 functionally-defined ROIs , and Fig 5B shows the mean PSF across these ROIs as a function of cortical distance in the common model connectome space and in anatomically-aligned data . CHA increased the average slope of PSF across these ROIs , relative to anatomical alignment , from 0 . 013 to 0 . 105 ( difference = 0 . 092 [0 . 080 , 0 . 099] ) . Fig 5C shows the slopes of connectivity profile PSFs for resting state connectivity profiles in the 26 functionally-defined ROIs , and Fig 5D shows the mean PSF across these ROIs ( ISC or WSC as a function of cortical distance ) in the common model connectome space , in MSM-All-aligned data , and within-subject . CHA increased the average slope of PSF across these ROIs , relative to MSM-All alignment , from 0 . 012 to 0 . 065 ( difference = 0 . 053 [0 . 047 , 0 . 055] ) . The intersubject PSF slopes in the common model connectome space and the PSF within-subject ( slope = 0 . 067 ) were not significantly different ( difference = 0 . 002 [-0 . 002 , 0 . 007] ) . This fine spatial granularity was ubiquitous in cortex , with steep PSFs in sensory-perceptual areas in occipital and temporal cortices as well as in higher-order cognitive areas in lateral and medial parietal and prefrontal cortices . The mean PSFs across ROIs ( Fig 5B and 5D ) clearly show that CHA captures fine-scale variations in connectivity profiles for neighboring cortical nodes across subjects that are not captured by anatomical alignment or MSM-All alignment . The ISCs of connectivity profiles for neighboring nodes in the common model connectome are substantially lower than ISCs for the same node ( differences = 0 . 21 [0 . 18 , 0 . 23] for movie data and 0 . 09 [0 . 08 , 0 . 09] for resting state data ) . Similar fine spatial granularity is seen in the within-subject between-session PSFs for resting state connectivity profiles ( 0 . 10 [0 . 09 , 0 . 10] ) . By contrast , ISCs for connectivity profiles in the anatomically-aligned and MSM-All aligned data barely differ for nodes spaced 0 versus 1 voxel/3 mm ( differences = 0 . 005 , [0 . 005 , 0 . 005] and 0 . 004 , [0 . 004 , 0 . 015] , respectively ) and 2 voxels/6 mm ( 0 . 01 [0 . 01 , 0 . 02] and 0 . 02 [0 . 01 , 0 . 02] , respectively ) apart . Decrements for larger distances ( ISCs of nodes spaced 3 voxels/9 mm: 0 . 03 [0 . 03 , 0 . 03] and 0 . 03 [0 . 03 , 0 . 03] , respectively; and 4 voxels/12 mm: ( 0 . 05 [0 . 05 , 0 . 06] and 0 . 05 [0 . 04 , 0 . 05] , respectively ) were similarly small . Next we asked if this shared variance in fine-scale local variation in connectivity profiles carries meaning by testing whether it reflects fine-scale variations in response tuning topographies that carry fine-grained distinctions in representation . We tested whether projecting movie response data into the CHA-derived common connectome space afforded better alignment of representational geometry for movie time-points and better bsMVPC of movie time segments . Results show that shared fine-scale structure in the common model connectome is closely related to fine distinctions in representations . Fig 6A shows a cortical map of mean ISCs of local representational geometry after anatomical alignment and in the common model connectome . Representational geometry is the matrix of all pairwise similarities between patterns of response to different time-points in the movie , resulting in a matrix of more than 800 , 000 pairwise similarities ( see methods ) . Fig 6B shows a cortical map of mean bsMVPC accuracies for 15 s movie time-segments in searchlights after anatomical alignment and CHA . CHA greatly increased both ISCs of representational geometry and bsMVPC accuracies . CHA significantly increased ISCs of representational geometry in all ROIs ( ROI mean ISCs = 0 . 308 and 0 . 210 after CHA and anatomical alignment , respectively , difference = 0 . 097 [0 . 080 , 0 . 110] ) . CHA also dramatically increased bsMVPC accuracies in all ROIs ( ROI mean bsMVPC accuracies = 10 . 37% and 1 . 04% after CHA and anatomical alignment , respectively , difference = 9 . 33% [7 . 71% , 10 . 54%] ) . We tested the generalization of the common model connectome derived from resting state fMRI by applying connectivity hyperalignment parameters derived from one session of resting state data to task maps provided by the HCP database comprised of 32 task activation maps and 14 task contrast maps ( S2 Table ) . These task maps reflect simple operations and , thus , do not have the same fine-grained structure that is associated with activation by dynamic , naturalistic stimuli such as a movie . We calculated the ISC of these task maps between each subject and the average of others before and after hyperalignment . Hyperalignment improved correlations on average across all tasks and in all but two ( Face-Shapes and Body-Average , labeled ns ) task contrast maps ( Fig 7 ) . The average correlation across task maps increased from 0 . 58 to 0 . 65 ( mean difference = 0 . 07 [0 . 06 , 0 . 08] ) . Since CHA aligned fine-scale patterns of response tuning functions across subjects better than anatomy-based alignment , we asked how well it compares to our previously published response-based hyperalignment ( RHA ) [16] . Because RHA requires responses that are synchronized across subjects in time , it cannot be applied to resting state data . We compare CHA and RHA of movie viewing data on 1 ) ISC of connectivity profiles , 2 ) ISC of representational geometry , and 3 ) bsMVPC of 15 s movie segments . Results showed that both CHA and RHA increased ISCs and bsMVPC classification accuracies significantly over anatomy-based alignment , but each algorithm achieves better alignment for the information that it uses to derive a common model , namely connectivity profiles and patterns of response , respectively . ISCs of connectivity profiles are significantly higher in a common model based on CHA than in a common model based on RHA ( ROI mean ISCs = 0 . 67 and 0 . 575 , respectively; CHA-RHA difference = 0 . 095 [0 . 081 , 0 . 112] ) ( S1 Fig ) . By contrast , RHA marginally but significantly outperforms CHA on some validations based on response tuning functions , namely ISCs of representational geometry ( ROI means = 0 . 322 and 0 . 308 , respectively; RHA-CHA difference = 0 . 014 [0 . 007 , 0 . 019] ) ( S2 Fig ) , and bsMVPC of movie segments ( ROI mean accuracies = 13 . 65% and 10 . 37% , respectively; RHA-CHA difference = 3 . 28% [2 . 76% , 3 . 78%] ) ( S3 Fig ) .
These results show that fine-scale local variation in connectivity profile is a major component of the human connectome that can be modeled with shared connectivity basis functions . Each connectivity basis function has a connectivity profile that is shared across subjects and a different local connectivity topography in each individual brain . These basis functions are derived from multiple subject data in local cortical fields . An individual’s connectivity pattern in a cortical field is modeled as multiplexed or overlaid connectivity topographic basis functions , and the connectivity profile of each cortical node or voxel is modeled as a weighted mixture of local connectivity profile basis functions . Thus , the connectivity profile for each voxel or node is modeled as a high-dimensional vector of connectivity profile bases , capturing how it varies locally from its neighbors , rather than modeling the connectivity of a brain area as a single connectivity profile that is shared by all voxels or nodes . We show that these shared basis functions can be discovered with connectivity hyperalignment of data collected during viewing and listening to a rich naturalistic movie and during the resting state . These basis functions constitute a common model connectome . Shared fine-scale variation is a ubiquitous characteristic of all of human cortex and is a major component of the human connectome that coexists with shared coarse-scale areal variation . We show that patterns of connectivity exhibit fine-scale variation that is captured in the CHA-derived common model connectome . We define fine-scale structure as voxel-by-voxel or node-by-node variation in response and connectivity profiles , as compared to the coarse structure of parcels that consist of sets of voxels or surface nodes and are treated as a functional unit with a homogeneous functional profile . In Fig 8 we illustrate the fine scale structure that is captured in the common model connectome for connectivity patterns in a left lateral-occipital/inferior-temporal cortex cortical field . Quantitatively , we show that shared fine-scale structure is captured in the common model connectome with a direct measure of the spatial granularity of local variation in connectivity profiles—the intersubject point-spread function . The intersubject spatial point-spread function for variation in connectivity profiles is dramatically , six to eight-fold , steeper after data are transformed into the common model connectome than for data that are anatomically aligned . Next we show that capture of this fine-scale structure in functional connectivity generalizes to capture of fine-scale structure in neural representation . Transformation of movie data into the common model space , using matrices derived from functional connectivity in independent movie data , afford bsMVPC of time segments that are tenfold higher than for anatomically-aligned data . bsMVPC of movie time segments relies on fine-scale structure that is not well-aligned based on anatomy , nor on functional alignment using a “rubber-sheet” warping of cortical topographies , nor on hyperalignment based on responses to a limited variety of still images of visual categories [13 , 16 , 23–25] . ISCs of local representational geometries also are dramatically higher after CHA than after anatomical alignment . These local representational geometries reflect fine-scale structure that reveals how information spaces in different cortical fields vary , offering a window on how these spaces are transformed along processing pathways and reshaped by task demands [26–28] . Finally , we also show that transformations derived from rsfMRI improve alignment of topographies in task activation and task contrast maps in the HCP database . The existence and importance of fine-scale connectivity is well-recognized [29–31] but previously was not modeled in a common computational framework and , consequently , was largely overlooked . Attempts to model within-area topographies of connectivity either were limited mostly to within-subject analyses or coarser within-area topographies that could be captured with anatomy-based alignment of group data [31] . Consequently , when not simply overlooked , within-area variations in connectivity profiles were usually analyzed as gradients that have a single cycle in a cortical area , such as retinotopy or somatotopy [29–31] . Other models of shared structure in the human connectome have focused on the identification of shared functional networks that can be identified with cluster analysis ( e . g . [5 , 32 , 33] ) or independent components analysis ( ICA; e . g . [19] ) . These methods do not attempt to align the fine-scale structure within areas in these networks . In some approaches , each voxel is assigned to one cluster or system and is , thereby , associated with the time-series tuning function that characterizes that cluster [5 , 8 , 32 , 33] . Approaches that use ICA , or related componential analyses such as PCA or SVD , have the potential to capture node-by-node variation in connectivity profiles , but implementations of these approaches have not adapted them to analyze this fine-scale topographic structure . For example , dual regression could allow using group ICA as a common space for modeling each voxel in an individual as a weighted sum of independent components [34 , 35] . In practice , however , each voxel is characterized in ICA analyses by the network to which it belongs , not as a mixture of multiplexed functional topographies . Node-by-node local variation in connectivity topographies is blurred in group analyses because individual variation on independent components is projected into anatomically-aligned brains rather than into a single reference voxel space to reveal shared fine-scale structure , as we do here . A novel approach by Langs et al . [32 , 33] allows nodes to be assigned to different clusters in a common functional connectivity embedding space independently of anatomical location . The implementations of this method , however , do not attempt to discover shared fine-scale structure , and the low dimensionality of the embedding space and small number of clusters are probably insufficient to capture this level of detail . Cortical functional architecture has multiplexed topographies at multiple spatial scales . In primary visual cortex , retinotopy is multiplexed with ocular dominance columns , edge orientation , spatial frequency , motion direction , and motion velocity , among other low-level visual attributes [36 , 37] . Primary visual cortex sends coherent projections to other visual areas where these topographies are recapitulated and transformed , affording the emergence of more complex features , such as curvature , texture , shape , color constancy , and biological motion; and , subsequently , even higher-order attributes such as object categories , view-invariant face identity , and species-invariant attributes of animals such as action categories and dangerousness [12 , 15 , 26–28 , 38–40] . Similar transformations of multiplexed topographies characterize other sensory modalities and , undoubtedly , supramodal cognitive operations . Modeling inter-areal communication as a single value of connectivity strength sheds no light on how information is transformed along cortical processing pathways to allow high-order information to be disentangled from confounding attributes [41] . Multiplexed cortical topographies at multiple spatial scales can be modeled with individual-specific topographic basis functions that have shared tuning profiles [13 , 16] and shared connectivity profiles ( as shown here ) . No previous model captured multiple spatial scales of connectivity topographies with connectivity profiles that are shared across brains . By capturing coarse- and fine-scale connectivity topographies with shared basis functions , the common model connectome casts a bright light on the dominant role of fine-scale connectivity patterns in the human connectome and opens new territory for investigation of the network properties of cortical connectivity at finer levels of detail . With this new perspective , inter-areal connectivity can be modeled as more than a simple replication of global activity , as is the assumption underlying existing approaches to modeling the connectome , but , instead , as information processing operations in which functional topographies are transformed by projections between areas .
All fMRI data collection at Dartmouth College was approved by the Dartmouth Committee for the Protection of Human Subjects . Resting state data from the Human Connectome Project was approved by the Institutional Review Boards associated with that project . We scanned 11 healthy young right-handed participants ( 4 females; Mean age: 24 . 6+/-3 . 7 years ) during movie viewing . Participants had no history of neurological or psychiatric illness . All had normal or corrected-to-normal vision . Informed consent was collected in accordance with the procedures set by the local Committee for the Protection of Human Subjects . Participants were paid for their participation . These data also were used in a prior publication on whole cortex RHA [16] . In the HCP database [20] , we found unrelated subjects of age < = 35 with at least four resting state scans , yielding a list of 64 subjects . We chose the first 20 of these subjects in the sorted order of subject IDs for our analysis . For each subject , we used their cortical surfaces and fMRI data aligned to the group using MSM-All [22] with 32K nodes in each hemisphere as provided by the HCP . We used data from one resting state session [19] ( “rfMRI_REST1_LR” ) to derive CHA parameters and validated it on a different resting state session ( “rfMRI_REST2_LR” ) , and task fMRI sessions [18] ( EMOTION , GAMBLING , LANGUAGE , MOTOR , RELATIONAL , SOCIAL , and WM ) . Resting state data were acquired for 1200 TRs with a TR of 0 . 720 s in each session ( total time = 14 min 33 s ) . The data used to derive the CHA parameters and common model and the resting state data used for validation tests used the same phase-encoding direction ( LR ) . We used a single session of rsfMRI for alignment to mimic a typical resting state data acquisition which usually varies from 10–20 mins of scanning . See [19] for more details about the acquisition and preprocessing pipelines . We use CHA to derive a common model of the human connectome and the transformation matrices that project individual brains’ connectomes into the common model connectome space . The common model connectome is a high-dimensional information space . In the current implementation , the model space based on movie fMRI data has 54 , 034 dimensions , corresponding to the number of voxels in the gray matter mask , and the model space based on HCP resting state fMRI data has 59 , 412 dimensions , corresponding to the number of cortical nodes in those data . The derivation of this space starts with hyperalignment in local cortical fields , searchlights , which yields orthogonal transformation matrices for each subject in each field . These searchlights are aligned across subjects based on anatomy ( movie data ) or MSM-All ( HCP resting state data ) ; consequently , each locus within a searchlight is similarly aligned across subjects before CHA . Local transformation matrices for each searchlight map anatomically or MSM-All aligned cortical loci in a cortical field to CHA-aligned dimensions in the common model connectome . These local transformation matrices are then aggregated into a whole brain transformation matrix , which is not globally orthogonal . The whole brain transformation matrices are derived based on local hyperalignment in searchlights to constrain resampling of information to cortical neighborhoods defined by those searchlights .
|
Resting state fMRI has become a ubiquitous tool for measuring connectivity in normal and diseased brains . Current dominant models of connectivity are based on coarse-scale connectivity among brain regions , ignoring fine-scale structure within those regions . We developed a high-dimensional common model of the human connectome that captures both coarse and fine-scale structure of connectivity shared across brains . We showed that this shared fine-scale structure is related to fine-scale distinctions in representation of information , and our model accounts for substantially more shared variance of connectivity compared to previous models . Our model opens new territory—shared fine-scale structure , a dominant but mostly unexplored component of the human connectome—for analysis and study .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"nervous",
"system",
"brain",
"neuroscience",
"magnetic",
"resonance",
"imaging",
"mathematics",
"algebra",
"brain",
"mapping",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"imaging",
"techniques",
"resting",
"state",
"functional",
"magnetic",
"resonance",
"imaging",
"vector",
"spaces",
"connectomics",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"neuroanatomy",
"anatomy",
"linear",
"algebra",
"central",
"nervous",
"system",
"data",
"visualization",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences"
] |
2018
|
A computational model of shared fine-scale structure in the human connectome
|
Nonreplicating type I uracil auxotrophic mutants of Toxoplasma gondii possess a potent ability to activate therapeutic immunity to established solid tumors by reversing immune suppression in the tumor microenvironment . Here we engineered targeted deletions of parasite secreted effector proteins using a genetically tractable Δku80 vaccine strain to show that the secretion of specific rhoptry ( ROP ) and dense granule ( GRA ) proteins by uracil auxotrophic mutants of T . gondii in conjunction with host cell invasion activates antitumor immunity through host responses involving CD8α+ dendritic cells , the IL-12/interferon-gamma ( IFN-γ ) TH1 axis , as well as CD4+ and CD8+ T cells . Deletion of parasitophorous vacuole membrane ( PVM ) associated proteins ROP5 , ROP17 , ROP18 , ROP35 or ROP38 , intravacuolar network associated dense granule proteins GRA2 or GRA12 , and GRA24 which traffics past the PVM to the host cell nucleus severely abrogated the antitumor response . In contrast , deletion of other secreted effector molecules such as GRA15 , GRA16 , or ROP16 that manipulate host cell signaling and transcriptional pathways , or deletion of PVM associated ROP21 or GRA3 molecules did not affect the antitumor activity . Association of ROP18 with the PVM was found to be essential for the development of the antitumor responses . Surprisingly , the ROP18 kinase activity required for resistance to IFN-γ activated host innate immunity related GTPases and virulence was not essential for the antitumor response . These data show that PVM functions of parasite secreted effector molecules , including ROP18 , manipulate host cell responses through ROP18 kinase virulence independent mechanisms to activate potent antitumor responses . Our results demonstrate that PVM associated rhoptry effector proteins secreted prior to host cell invasion and dense granule effector proteins localized to the intravacuolar network and host nucleus that are secreted after host cell invasion coordinately control the development of host immune responses that provide effective antitumor immunity against established ovarian cancer .
Toxoplasma gondii is a ubiquitous parasite that chronically infects a wide array of warm-blooded vertebrates following the oral ingestion of infectious oocysts or tissue cysts in contaminated water or food [1] . The primary infection is typically subclinical with minor or no apparent disease due to strong immune control , yet T . gondii invariably establishes long-term infection of the host by developing latent tissue cysts [1] . Infection during pregnancy can harm the fetus , and reactivation of latent stages because of immune deficiency ( AIDS , cancer chemotherapy , transplantation ) causes severe and potentially lethal toxoplasmosis infections [2] . There are no currently approved vaccines to prevent toxoplasmosis in humans , or vaccines to prevent infection of cats which host the sexual parasite stages and disseminate infectious oocysts into the environment [3] . Remarkably , uracil auxotrophic vaccine strains of T . gondii that do not replicate or cause infection in mammals retain a dynamic ability to activate protective immunity to T . gondii [4–14] as well as protective immunity to established highly aggressive pancreatic , melanoma , and ovarian tumors [15–20] . The remarkable biological ability of T . gondii to manipulate the immune system most likely originates from its life style as an obligate intracellular parasite . The parasite as well as the host must both survive the acute infection to permit the development of latent infection that is essential for the transmission of T . gondii to new hosts [21] . To accomplish this , T . gondii extensively manipulates its host cells through the secretion of specialized effector proteins [21 , 22] . Secreted rhoptry ( ROP ) effector proteins originating from the apical rhoptry organelle are injected directly into the host cell cytosol prior to active invasion of the host cell and formation of the parasitophorous vacuole ( PV ) [23 , 24] . After host cell invasion , many of these ROP effectors traffic specifically to the nascent PV membrane ( PVM ) to establish PVM functions required for parasite replication and survival [23 , 25] . Rhoptry secreted effectors are also injected into parasite contacted host cells that are not subsequently invaded [26–28] , suggesting that parasite manipulation of host cells occurs in both the parasite invaded as well as in the parasite injected noninvaded cell populations . After PVM formation , effectors are secreted from parasite dense granules ( GRA proteins ) into the PV lumen and these GRA proteins traffic to the intravacuolar network ( IVN ) of nanotubular membranes , to the PVM and its extensions , to the host cell cytosol , or to the host cell nucleus [29] . The PV avoids acidification and fusion with host endolysosomes [30 , 31] , and host cell endoplasmic reticulum ( ER ) and mitochondria closely associate with the PVM [30 , 32 , 33] . A parasite secreted protein ( MAF1 ) mediates the association of mitochondria to the PVM and modulates host inflammatory cytokines [34] . The host rapidly recognizes T . gondii and drives a TH1-biased immune response marked by the cytokine IL-12 [35] . CD8α+ dendritic cells drive the initial IL-12 response downstream of TLR11/12 mediated activation of MyD88 [36–38] . Production of IL-12 is required for the development of protective T cell populations [11 , 12] and immune control is ultimately achieved by T cell populations that produce the cytokine interferon-gamma ( IFN-γ ) [7 , 8 , 11 , 39 , 40] , and by cytolytic CD8+ T cell populations [8 , 41] . IFN-γ activates cell autonomous host immunity related GTPases ( IRGs ) [42–47] . In addition , host guanylate binding proteins associate with the PVM [48–53] and components of autophagy also associate with the PVM after exposure to IFN-γ but fail to activate degradative macroautophagy [54–58] . T . gondii strain types exhibit variation in their level of virulence . Virulent type I strains invariably kill laboratory and outbred strains of mice , while type II and III strains are markedly less virulent [3] . Strain type dependent virulence is linked to polymorphisms in secreted proteins such as ROP5 and ROP18 that form PVM associated complexes to resist IFN-γ activated IRG dependent innate killing mechanisms [59–72] , and virulence is also correlated with the exact ROP5 and ROP18 allele combinations [73] . ROP5 and ROP18 were also recently identified as virulence factors in divergent South American strains of T . gondii [74] and in less virulent type II strains [75] . The ROP18 kinase activity directly inactivates innate immunity by phosphorylation of host IRGs [63] . ROP18 also inhibits host transcription factor NF-κβ signaling through association with its p65 subunit [76] and ROP18's association with the host ER stress sensor ATF6β interferes with antigen presentation [77] . Thus ROP18 manipulates innate and adaptive immunity as well as host cell signaling pathways . The ROP38 secreted effector molecule extensively modulates host cell transcriptional programming and gene expression profiles [78] , and while this molecule does not influence parasite virulence , it is required for establishing latent infection [75] . Dense granule ( GRA ) secreted effectors also play important roles in host cell manipulation [34 , 79–82] . In particular , ROP16 [83–85] , GRA15 [86] , and GRA24 [87] manipulate host cell signaling pathways and their associated biological outputs can either promote or inhibit production of IL-12 by the invaded host cell in a parasite strain type dependent manner . Parasite modulation of the IL-12 response is significant because T . gondii preferentially invades dendritic cells [88] that are responsible for initiating the host IL-12 response necessary for the development of protective T cell populations [89] . Consequently , active invasion of the host cell by T . gondii uniquely creates a nonfusogenic PV hospitable for intracellular parasite replication and parasite infection while extensively modifying host cell behavior through functions of ROP and GRA secreted effector proteins [21 , 22] . Uracil auxotrophic vaccine strains [5 , 13 , 14] have emerged as important models to dissect the parasite-host interaction and to investigate host and parasite mechanisms that determine the development of life-long CD8+ T cell dependent immunity [4–14] . Nonreplicating uracil auxotrophs permit the evaluation of infected host cells manipulated by secreted effector proteins in the absence of parasite replication and destruction of the host cell . Uracil auxotrophs preferentially invade myeloid cell types [4 , 16–19] and more quickly stimulate protective CD8+ T cell responses in comparison to replicating strains of T . gondii [4 , 7 , 8 , 14 , 18] . Nonreplicating uracil auxotrophs also preferentially target and invade CD11c+ and myeloid cell populations in aggressive models of murine pancreatic cancer , melanoma , and ovarian cancer to convert tolerogenic tumor microenvironments into immune stimulatory microenvironments [15 , 16 , 19 , 20] . Uracil auxotrophs stimulate high-level expression of co-stimulatory molecules CD80 and CD86 as well as IL-12 , and tumor antigen specific CD8+ T cell populations that mediate the long-term survival of tumor-bearing mice [15 , 16 , 19 , 20] . Here , we developed targeted deletion mutants using genetically tractable Δku80 uracil auxotrophic vaccine strains to explore the extent to which parasite secreted ROP and GRA effector proteins are associated with mechanisms that promote CD8+ T cell dependent therapeutic immunity to established highly aggressive ovarian tumors [16 , 90] . Our results reveal that rhoptry secretion , host cell invasion , formation of the PVM , formation of the IVN , and the secretion of dense granule proteins to the host cell nucleus are essential for the development of the antitumor response . The ROP18 kinase activity that mediates parasite virulence was not essential for this antitumor response . However , ROP18 PVM localization , other PVM localized rhoptry molecules ( ROP5 , ROP17 , ROP35 and ROP38 ) , and dense granule molecules that localized to the IVN ( GRA2 and GRA12 ) or to the host cell nucleus ( GRA24 ) were essential to trigger the development of this potent antitumor immunity .
Therapeutic vaccination of established highly aggressive and immunosuppressive ID8-Defβ29/Vegf-A ( ID8DV ) ovarian tumors [90 , 91] with the nonreplicating type I uracil auxotrophic vaccine strain CPS [5] was shown to preferentially target CD11c+ antigen presenting cells to reverse tumor associated immune suppression and promote tumor antigen specific CD8+ T cell dependent antitumor immunity [16–18] . To identify parasite and host mechanisms that mediate this potent antitumor response we utilized a genetically tractable nonreverting CPS-like type I uracil auxotrophic vaccine strain ( the OMP mutant ) [13] developed in the RHΔku80 background [92] that enables highly efficient targeted genetic manipulations . We first verified that a three-dose OMP or CPS intraperitoneal vaccination protocol delivered sequentially at 8 , 20 , and 32 days after ID8DV ovarian tumor challenge markedly prolonged the survival of tumor-bearing mice to a similar degree ( Fig 1A ) . OMP vaccination delivered systemically by the intravenous route provided a mean survival of 54 days while intraperitoneal vaccination provided a mean survival of 59 days ( S1A Fig ) . Additional OMP vaccinations delivered at 12-day intervals markedly enhanced the survival of ovarian tumor-bearing mice ( Fig 1B ) , suggesting that the underlying therapeutic immune responses were triggered even after immunity to T . gondii was established . Moreover , aged mice with established long-term vaccine induced immunity to T . gondii infection ( S1B Fig ) exhibited equivalent therapeutic benefits to those of age matched nonimmune naive mice ( Fig 1C ) . These results suggest that previous protective immune responses induced by uracil auxotrophs in tumor-bearing mice did not abrogate the efficacy of additional sequentially delivered vaccinations with uracil auxotrophs . IL-12 was previously shown to be important to the antitumor response to ID8DV tumors induced by therapeutic vaccination with uracil auxotrophs [16] . To verify the involvement of the host IL-12/IFN-γ TH1 axis in the antitumor response we vaccinated ID8DV tumor-bearing IL-12p35-/- mice with uracil auxotrophs . IL-12p35-/- mice lacking the ability to produce bioactive IL-12p70 ( IL12-p35 + IL12-p40 subunits ) mounted no detectable antitumor response ( Fig 2A ) . IL-12p40-/- mice also exhibited no antitumor response after vaccination ( S2 Fig ) . Despite the recognized importance of MyD88 signaling in driving high level IL-12 production in response to T . gondii infection [37] , vaccination of MyD88-/- mice bearing ID8DV tumors provided a therapeutic benefit equivalent to vaccination of wild-type tumor-bearing mice ( Fig 2B ) . Together these results suggested that IL-12 production via a MyD88 independent mechanism was sufficient to drive the antitumor response induced by uracil auxotrophs . The CD8α+ dendritic cell subset is a critical source of IL-12 required for immune control of T . gondii infection [36] . To address the role of CD8α+ dendritic cells in the antitumor response we established ID8DV tumors in basic leucine zipper transcription factor ATF-like 3 deficient ( Batf3-/- ) mice which exhibit a loss of CD8α+ dendritic cells [93] , and vaccinated these tumors . The antitumor response triggered by vaccination with uracil auxotrophs was severely abrogated in Batf3-/- mice ( Fig 2C ) . These data argue that the CD8α+ dendritic cell subset confers critical mechanisms required for the stimulation of antitumor immunity . Production of protective IFN-γ in response to T . gondii infection is highly dependent on production of IL-12 [35] . To investigate the role for IFN-γ in the antitumor response we established ID8DV tumors in IFN-γ deficient mice ( IFN-γ-/- ) and vaccinated these mice . Production of IFN-γ was required for the therapeutic benefit ( Fig 2D ) . In addition , IFN-γ deficient mice developed ID8DV ovarian tumors more rapidly and succumbed faster to ovarian cancer than wild-type mice ( Fig 2D ) . Collectively , these results suggested that CD8α+ dendritic cells play a key role in stimulating the IL-12/IFN-γ TH1 axis to promote antitumor immunity . We previously reported that therapeutic vaccination of established ID8DV ovarian tumors with uracil auxotrophs stimulated a significant increase in ovarian tumor antigen specific CD8+ T cells as well as total T cell populations that limited ovarian tumor development [16] . To verify the protective T cell populations required for this antitumor immunity we examined tumor development in CD8 deficient mice ( CD8-/- ) . Vaccinated tumor-bearing CD8-/- mice exhibited no therapeutic benefit , and treatment accelerated disease in CD8-/- mice ( Fig 3A ) . Similar results were observed after depletion of CD8+ T cells using anti-CD8α antibody ( Fig 3B ) . NK cells provide another early pathway of IFN-γ production in response to T . gondii infection [94] . However , depletion of this cell population using anti-NK1 . 1 antibody did not affect the antitumor response ( S3A Fig ) . Depletion of CD4+ T cells using anti-CD4 antibody also abrogated the antitumor response ( Fig 3C ) . Moreover , consistent with a requirement for CD4+ T cells , major histocompatibility complex ( MHC ) II was also required for stimulation of the antitumor response ( S3B Fig ) . Surprisingly , MHCII played a dichotomous role . The absence of MHCII in normal ( nontumor ) cells delayed development of the ID8DV tumor , suggesting that MHCII may also act to promote ovarian tumor development ( S3B Fig ) . Together these results suggested that CD8+ T cells , CD4+ T cells , and MHCII were essential to drive IFN-γ production and antitumor immunity . To examine whether active invasion of host cells was essential to the antitumor response , we heat killed uracil auxotrophs at 56°C to abolish their ability to invade host cells ( S4A Fig ) . Heat killed uracil auxotrophs exhibited no detectable antitumor response ( Fig 4A ) . However , we could not exclude the possibility that heat treatment reduced the activity of parasite molecules that would otherwise stimulate IL-12 production , since heat treatment of T . gondii has been previously shown to elicit a reduced IL-12 response in comparison to live parasites [95] . To address this , we prepared , tachyzoite excreted/secreted antigen ( ESA ) extracts , tachyzoite lysate antigen ( TLA ) extracts , and soluble tachyzoite antigen ( STAg ) extracts from uracil auxotrophs to examine whether parasite extracts not subjected to heat treatment could promote the antitumor response . ESA , TLA , and STAg extracts exhibited no detectable antitumor activity ( Fig 4B ) . We hypothesized that while heat killed uracil auxotrophs could stimulate IL-12 , the activation of the IL-12/IFN-γ TH1 axis may be incomplete . To examine this , highly immunosuppressive ID8DV tumors were established in mice for 25 days and vaccinated once with live uracil auxotrophs or heat killed uracil auxotrophs . Supernatants from the peritoneal ovarian tumor microenvironment were analyzed for IL-12 , type I and type II interferons , and inflammasome activation 18 and 66 h post-vaccination . Increased production of IL-12p40 and IL-12p70 was observed 18 h following treatment with heat killed parasites but this early IL-12 subsided by 66 h ( Fig 4C and 4D ) . As expected , higher levels of IL-12 were observed 18 h after vaccination with live uracil auxotrophs , and this response persisted for at least 66 h ( Fig 4C and 4D ) . Live invasive uracil auxotrophs elicited a slight increase in IFN-γ as early as 18 h , however , IFN-γ levels were markedly increased by 66 h ( Fig 4E ) . In contrast , noninvasive heat killed parasites failed to stimulate any detectable increase in the production of IFN-γ in the immunosuppressive ID8DV tumor microenvironment ( Fig 4E ) . In addition , type I interferon alpha ( IFN-α ) was not detected and low levels of interferon beta ( IFN-β ) present in the untreated tumor microenvironment were not affected by vaccination with either live or with heat killed parasites ( S4B Fig ) . While IL-1β was not detected in the tumor microenvironment at 18 or 66 h post-vaccination , a low level increase in IL-1α was observed at 66 h after therapeutic vaccination with live invasive uracil auxotrophs ( S4C Fig ) . CD4+ and CD8+ T cells are the major producers of IFN-γ in T . gondii infection [39] , and IFN-γ is the key inflammatory cytokine that activates the major mechanism of IRG-dependent innate immunity that effectively clears intracellular parasites and PVs [63] . Preferential invasion of CD11c+ antigen presenting cells by uracil auxotrophs was previously associated with the initiation of the T cell dependent antitumor response that targets ID8DV ovarian tumors [16] . The requirement of live invasive parasites to induce IFN-γ ( Fig 4E ) suggested that persistence of parasites in invaded host cells could be essential to drive the antitumor response . On the other hand , IFN-γ could also rapidly clear parasites from invaded cells through host IRGs and innate immunity . Type I uracil auxotrophs were recently shown to fully resist PV clearance in IFN-γ activated macrophages in vitro for at least 5 days [4] . We examined clearance of YFP+ uracil auxotrophs from invaded CD11c+ cells in the ID8DV tumor microenvironment and observed a 16 . 8-fold and 12 . 9-fold reduction in the absolute number or percentage , respectively , of YFP+CD11c+ antigen presenting cells between 18 h and 66 h after therapeutic vaccination ( Fig 4F ) . In addition , previous studies have shown that uracil auxotrophs remain in the peritoneum after vaccination and parasite invaded myeloid cells were found to rarely migrate to draining lymph nodes in either naive mice [4] or in tumor-bearing mice [19] . These results suggest that uracil auxotroph PVs in CD11c+ cells were not cleared prior to 18 h but were rapidly cleared between 18 and 66 h , potentially through an IFN-γ dependent mechanism . In view that live invasive uracil auxotrophs were associated with the potent antitumor response , we explored whether invasion and/or rhoptry secretion was required for this response . To investigate rhoptry secretion , uracil auxotrophs were treated with the chemical 4-bromophenacyl bromide ( 4BPB ) . 4BPB selectively inhibits phospholipase A2 to irreversibly block rhoptry secretion and parasite invasion of host cells ( S5 Fig ) without affecting parasite attachment to host cells [97–99] . 4BPB treatment of uracil auxotrophs completely abrogated the antitumor response ( Fig 5 ) , suggesting that rhoptry secretion was essential for this response . Toxoplasma directly injects rhoptry effector molecules into host cells prior to invasion as well as into contacted but noninvaded bystander cells [26–28] . To differentiate the role of rhoptry secretion into host cells with or without parasite invasion and PVM formation in the host cell , we treated uracil auxotrophs with the chemical mycalolide B , a compound that irreversibly blocks parasite motility thereby inhibiting invasion of host cells without blocking rhoptry secretion [100 , 101] . As expected , mycalolide B treated uracil auxotrophs were blocked in their invasion of host cells ( S5 Fig ) . Mycalolide B treated uracil auxotrophs were markedly weakened in their ability to trigger the antitumor response ( Fig 5 ) . Together these chemical inhibition experiments suggested that rhoptry secretion as well as active invasion of host cells was necessary to arm the full antitumor response . Based on the requirements for active invasion and parasite secretion , we hypothesized that specific ROP and GRA secreted effector proteins may control the antitumor response . To identify whether specific secreted effector molecules were involved in the antitumor response we used a reverse genetic approach to target complete gene deletions of selected candidate parasite ROP or GRA genes using the Δku80ΔompdcΔupΔhxgprt OMP uracil auxotroph vaccine background as the parental strain ( Table 1 ) . Targeted gene deletions were engineered using positive genetic selection to replace genes of interest ( GOI ) with the HXGPRT selectable marker [102] at the targeted gene loci ( S6 Fig ) [13 , 14 , 92 , 103 , 104] . We targeted the deletion of predicted active ROP kinases ROP21 , ROP35 and ROP38 . The deletion of ROP35 or ROP38 in the less virulent type II background was recently shown to abrogate latent infection without affecting parasite virulence during acute infection [75] . ROP35 and ROP38 are differentially expressed ( >16 fold ) between parasite strains and overexpression of ROP38 down regulates transcription of many host genes involved in proliferation , MAP kinase ( MAPK ) signaling , and apoptosis [78] . Deletion of ROP21 ( OMPΔrop21 ) did not affect the antitumor response ( Fig 6A ) . In contrast , deletion of the ROP38 gene locus ( OMPΔrop38 ) as well as ROP35 ( OMPΔrop35 ) markedly impaired the antitumor response ( Fig 6A ) . The ROP35 deleted uracil auxotroph ( OMPΔrop35 ) was complemented with the wild-type ROP35 allele to generate strain OMPΔrop35::ROP35 by targeted insertion of ROP35 ( C-terminal HA tagged ) along with the cytosine deaminase gene into the OMPDC locus , and the complemented strain was selected by growth in cytosine which is converted to uracil by cytosine deaminase ( CD ) ( S7 Fig ) [105] . Importantly , the CD gene provides a new genetic marker for positive selection , which can be used specifically in uracil auxotrophic backgrounds without affecting parasite virulence . In addition , the CD marker can also be deployed as a genetic marker for negative selection by selection with the nontoxic prodrug 5-fluorocytosine [105] . Complementation of the ROP35 deletion by expression of a HA-tagged allele of ROP35 ( OMPΔrop35::ROP35 ) ( Fig 6A ) rescued the antitumor response ( Fig 6A ) . Complementation of OMPΔrop38 was not currently feasible due to the existence of multiple related ROP38 alleles in the ROP38 gene locus and the absence of any complete cosmid , fosmid , or BAC that spans this gene locus ( ToxoDB . org ) [75] . ROP21 and ROP38 were previously shown to localize to the PVM following their injection into host cells [78] , however , the localization of ROP35 has not been reported . To determine whether ROP35 is also localized to the PVM after its injection into the host cell we used digitonin selective permeable conditions to expose the cytosolic face of the PVM [59 , 62] and found that ROP35 was localized to the PVM ( Fig 6C ) . These data suggest that specific parasite secreted ROP effectors associated with the PVM and host transcriptional manipulation , but not specifically associated with parasite virulence , were involved in the mechanism that triggers the antitumor response . Shortly after host cell invasion parasite dense granules massively secrete a large repertoire of proteins into the PV lumen and a subset of dense granule proteins including GRA2 [106] and GRA12 [107] establish association with an intravacuolar network ( IVN ) of nanotubular membranes . Deletion of IVN associated GRA proteins GRA2 ( OMPΔgra2 ) or GRA12 ( OMPΔgra12 ) markedly impaired the antitumor response ( Fig 6D ) . In contrast , deletion of PVM associated GRA3 ( OMPΔgra3 ) did not affect the antitumor response ( S8A Fig ) . Other GRA proteins such as GRA16 and GRA24 are secreted past the PVM to the host cell nucleus [80] . GRA24 deletion ( OMPΔgra24 ) markedly impaired the antitumor response ( Fig 6D ) . In contrast , deletion of GRA16 ( OMPΔgra16 ) , a secreted effector molecule that targets to the host nucleus and modulates host p53 and cell cycle [79] , did not affect the antitumor response ( S8A Fig ) . In addition , deletion of GRA15 ( OMPΔgra15 ) or ROP16 ( OMPΔrop16 ) did not affect the antitumor response ( S8B Fig ) . Since GRA15 and ROP16 were previously shown to modulate various host transcriptional pathways [86 , 108 , 109] , we also developed a double-targeted deletion of ROP16 and GRA15 ( OMPΔgra15Δrop16 ) . The OMPΔgra15Δrop16 double mutant also exhibited a normal antitumor response ( S8B Fig ) . Thus certain parasite secreted effector molecules that manipulate IL-12 ( ROP16 , GRA15 ) or that associate with the PVM ( GRA3 , ROP21 ) did not influence the antitumor response . In contrast , PVM associated ROP proteins ROP35 and ROP38 , IVN-associated GRA proteins GRA2 and GRA12 , as well as host nucleus-associated GRA24 were essential to the mechanism that triggers potent antitumor immunity . Since GRA2 was previously suggested to play a role in reducing association of IRGs to the PVM [66] , we assessed whether other parasite molecules involved in resistance to IRGs were necessary to trigger the antitumor response . PVM associated protein complexes containing ROP5 , ROP17 , or ROP18 molecules resist host IRGs [62] . Deletion of ROP5 ( OMPΔrop5 ) or ROP18 ( OMPΔrop18 ) markedly impaired the antitumor response in comparison to parental OMP , or to ROP17 deletion ( OMPΔrop17 ) , although ROP17 deletion itself also slightly impaired antitumor response ( Fig 7A ) . This profile was consistent with a potential role for IRG resistance in mediating the antitumor response . We assessed the resistance of the type I uracil auxotroph mutants to IRGs in IFN-γ stimulated mouse embryonic fibroblasts ( MEFs ) . The OMPΔrop5 and OMPΔrop18 mutants , as expected , were markedly impaired in their survival in IFN-γ-activated MEFs ( Fig 7B ) . Also as expected , the OMPΔrop17 mutant exhibited only a minor defect in IRG resistance and survival . However , other uracil auxotrophic mutants ( OMPΔrop35 , OMPΔrop38 , OMPΔgra2 , OMPΔgra12 , OMPΔgra24 ) that exhibited major defects in triggering the antitumor response fully resisted IRG mediated killing in IFN-γ activated MEFs ( Fig 7B ) . Moreover , these uracil auxotrophic mutants did not exhibit any defects in their invasion of host cells ( Fig 7C ) . In comparison , uracil auxotrophic mutants that exhibited no defects in their antitumor response retained their normal ability to resist IRG killing in IFN-γ stimulated MEFs ( S9 Fig ) . Together , these results suggested that while ROP5 , ROP17 , and ROP18 functions were important for the antitumor response , resistance to IRGs did not appear to be specifically associated with the antitumor response . Through genetic crosses , type I ROP5 has been previously associated with markedly increased virulence , whereas type II ROP5 was proposed to be avirulent [61 , 69] . Consistent with these findings , previous experiments have shown that the expression of virulent type I ROP5 in the less virulent type II strain increased parasite virulence by ~3 to 4-logs [66] , and that type II T . gondii strains poorly resisted IRG killing mechanisms in IFN-γ activated cells [44] . We recently reported that the genetic deletion of ROP5 or ROP18 in the less virulent type II strain reduced parasite virulence by ~ 2-logs and correspondingly reduced resistance to killing by host IRGs in IFN-γ stimulated MEFs by ~10% [75] . To further assess the relationship of the antitumor response and IRG resistance we examined a type II uracil auxotroph ( type II OMP ) [14] as well as attenuated type II mutants lacking genes for ROP5 or ROP18 [75] . Type II OMP [14] ( Table 1 ) elicited indistinguishable antitumor responses in comparison to type I OMP ( Fig 7D ) . In addition , the deletion of ROP5 or ROP18 in the type II background markedly impaired the antitumor response ( Fig 7D ) . To further examine the role of resistance to IRG killing mechanisms from other potential mechanisms we focused on the active ROP18 kinase that directly phosphorylates IRG effectors to inactivate host innate immunity [63] . We complemented the ROP18 deleted OMPΔrop18 strain by expression of wild-type ( OMPΔrop18::ROP18 ) or various mutant C-terminally HA-tagged ROP18 alleles , ( OMPΔrop18::ROP18KD , OMPΔrop18::ROP18RAH2 ( ATF ) , or the double mutant OMPΔrop18::ROP18RAH2 ( ATF ) , KD ) ( S10A Fig ) via targeted insertion of the ROP18 gene along with the cytosine deaminase gene into the OMPDC locus through positive selection in cytosine ( S7 Fig ) . Complementation of OMPΔrop18 with C-terminal HA-tagged wild-type ROP18 ( OMPΔrop18::ROP18 ) rescued the antitumor response ( Fig 8A ) . Remarkably , complementation of OMPΔrop18 with C-terminal HA-tagged kinase-dead ( KD ) ROP18 ( OMPΔrop18::ROP18KD ) [63] also rescued the antitumor response ( Fig 8A ) , suggesting that the ROP18 virulence function mediated by the ROP18 kinase activity was not essential for the antitumor response . The ROP18 ATF6β association domain is identical to the second arginine rich amphipathic helix ( RAH2 ) domain of ROP18 that targets ROP18 to the PVM after its injection into host cells [64 , 110] . While active wild-type ROP18 as well as kinase-dead ROP18 fully rescued the antitumor response , complementation of OMPΔrop18 with C-terminal HA-tagged RAH2 ( ATF ) deficient alleles ( designated as RAH2 ( ATF ) [75] ) of ROP18 ( OMPΔrop18::ROP18RAH2 ( ATF ) or the double mutant OMPΔrop18::ROP18RAH2 ( ATF ) , KD failed to rescue the antitumor response ( Fig 8A ) . Previous studies have shown that ectopic expression of ROP18 kinase activity in the absence of infection and PVM formation mediated the degradation of host NF-κβ [76] as well as host ATF6β [77] . In addition to ROP18 kinase activity , ROP18 PVM association is also required for resistance to host IRGs and virulence [63 , 64] . Therefore , we measured the resistance of mutant ROP18 uracil auxotrophs to IRGs and PV killing . The wild-type allele of ROP18 fully rescued resistance to IRG killing in the ROP18 deficient strain , however , the kinase-dead as well as RAH2 ( ATF ) domain deleted ROP18 alleles , as expected , failed to rescue any detectable resistance to IRG killing ( Fig 8B ) . These results were corroborated in ex vivo experiments using IFN-γ stimulated bone marrow derived macrophages . The RAH2 ( ATF ) domain deleted and the kinase-dead ROP18 alleles failed to confer resistance to host IRGs in IFN-γ stimulated macrophages , whereas the wild-type ROP18 allele was resistant to PV killing ( S10B Fig ) . In addition , the type II OMP uracil auxotroph that elicits a potent antitumor response indistinguishable from type I OMP ( Fig 7D ) , as expected , was exquisitely susceptible to IRG mediated killing in comparison to type I OMP ( Fig 8B ) . To further assess the requirement for the association of ROP18 with the PVM to trigger the antitumor response , we localized wild-type ROP18 as well as the kinase-dead ROP18 to the cytosolic surface of newly formed PVs using digitonin selective permeable conditions ( Fig 8C ) . In contrast , deletion of the RAH2 ( ATF ) domain , as expected , completely abolished the association of ROP18 with the PVM after its injection into host cells ( Fig 8C ) . These data suggested that the antitumor response was not dependent on the ROP18 kinase activity but did require ROP18 PVM association . We next asked whether injection of ROP18 into bystander host cells in the absence of invasion could trigger the antitumor response . To address this question we compared the antitumor efficacy of vaccinations using the ROP18 deleted OMP strain ( OMPΔrop18 ) compared to mycalolide B treatment of this strain . A significant increase in resistance to tumor development was observed in ROP18 deficient OMP parasites compared to mycalolide B treated ROP18 deficient OMP parasites ( Fig 8D ) . However , no significant difference in the antitumor response was observed between mycalolide B treated OMP compared to mycalolide B treated ROP18 deficient OMP parasites ( Fig 8D ) . These results suggest that ROP18 kinase activity in the absence of PVM association did not trigger the antitumor response .
Our studies extend the application of reverse genetic engineering in Toxoplasma gondii by using a genetically tractable Δku80 uracil auxotrophic vaccine background to reliably generate parasite strains with precisely targeted gene deletions of secreted parasite effectors . Targeted deletions of specific rhoptry and dense granule proteins in isogenic uracil auxotrophic backgrounds enabled a comparable measurement of the functional contribution of each of these secreted effectors in triggering the antitumor response to ID8DV ovarian tumors . These genetic studies revealed key roles for ROP5 , ROP17 , ROP18 , ROP35 , ROP38 , GRA2 , GRA12 , and GRA24 secreted parasite effectors in triggering the antitumor response . Host cell invasion by T . gondii requires motility and secretion from apical organelles [111] . Chemical treatment of T . gondii with 4-bromophenacyl bromide ( 4BPB ) irreversibly inhibits rhoptry secretion and invasion [97–99] . The antitumor response was absent in 4BPB treated uracil auxotrophs . Mycalolide B treated parasites fail to invade host cells due to a motility defect but can still secrete from their rhoptry organelles [100 , 101] . Our mycalolide B inhibition experiments suggested that rhoptry secretion in the absence of host cell invasion provided a detectable , though suboptimal , antitumor response . However , this antitumor response was markedly increased through a sequential combination of rhoptry secretion , host cell invasion , and the formation of the PV and PVM . Collectively , these findings demonstrate that secreted GRA proteins and PVM associated ROP proteins trigger a potent antitumor response from within invaded host cells in the tumor microenvironment . Therapeutic vaccination of solid tumors with uracil auxotrophs converts tolerogenic tumor microenvironments into immune stimulatory microenvironments in aggressive models of murine pancreatic cancer , melanoma , and ovarian cancer through the stimulation of tumor-specific CD8+ T cell populations [15 , 16 , 19] . Induction of the antitumor response in primary tumors did not require CD4+ T cell populations in B16 melanoma [15] or disseminated pancreatic tumors [19] , although in treated mice that survived the primary cancer , CD4+ T cells were shown to play an important role in enforcing immunity to pancreatic cancer [20] . Our data suggests that in addition to CD8+ T cells [16 , 17] , MHCII and CD4+ T cells were also necessary to mediate the antitumor response induced by uracil auxotrophs to ID8DV ovarian tumors . Uracil auxotrophic mutants were previously shown to preferentially target and invade CD11c+ antigen presenting cells in the ID8DV tumor microenvironment [16] . Our findings show that heat killed uracil auxotrophs failed to elicit an antitumor response . In addition , noninvasive heat killed uracil auxotrophs induced early IL-12 but , surprisingly , did not stimulate IFN-γ in the tumor microenvironment , suggesting that T cell populations were not highly activated in the absence of parasite invasion . Parasite extracts also failed to elicit any detectable antitumor response , reinforcing the importance of parasite invasion as a key step in the antitumor mechanism . Live invasive parasites rapidly induced high levels of IL-12 followed shortly thereafter by high-level production of IFN-γ . Moreover , high IFN-γ levels present after therapeutic vaccination with live parasites were associated with the effective clearance of intracellular uracil auxotrophs in the parasite invaded CD11c+ cell population within 66 h . Collectively , these findings indicate that parasite invasion of CD11c+ antigen presenting cells rapidly triggered the antitumor response through IFN-γ producing CD4+ and CD8+ T cell populations . High-level expression of IL-12 is triggered in CD8α+ dendritic cells by MyD88 dependent signaling downstream of TLR11/12 dependent recognition of parasite profilin [37 , 38 , 89] . However , uracil auxotroph vaccination of MyD88 knockout ( MyD88-/- ) mice was previously shown to elicit a reduced IL-12 response that was sufficient to drive IFN-γ production by protective T cell populations [10] . Our findings revealed that the antitumor response was intact in MyD88-/- mice , suggesting that the primary MyD88 dependent pathway for high level IL-12 production by CD8α+ dendritic cells was not strictly required . However , mice completely deficient in CD8α+ dendritic cells ( Batf3-/- or IRF8-/- knockout mice ) do not produce protective IL-12 or IFN-γ and rapidly succumb to T . gondii infection [36 , 112] , and our experiments ( present work ) revealed that CD8α+ dendritic cell deficiency in Batf3-/- knockout mice abrogated the antitumor response . These findings establish the essentiality of the CD11c+ CD8α+ dendritic cell in the antitumor response elicited by live uracil auxotrophs . CD11c+ CD8α+ dendritic cells have been recognized play an important role in cross presentation of antigens to prime CD8+ T cell mediated immunity to tumors [113] . We have previously shown that the invasion of ID8DV tumor associated immune suppressed CD11c+ antigen presenting cells by uracil auxotrophs triggered the expression of the T-cell receptor costimulatory molecules CD80 and CD86 [16 , 17] . Invasion of CD11c+ antigen presenting cells by uracil auxotrophs also increased CD80 and CD86 expression in melanoma and pancreatic cancer tumor microenvironments [15 , 19] , and uracil auxotroph invasion of dendritic cells in naive mice increased the expression of CD80 and CD86 as well as MHCI that is required to initiate antigen presentation that drives the development of protective T cell populations [4] . The high level induction of CD86 by T . gondii was previously shown to require active parasite invasion and host cell signaling through the JNK pathway [114] . Consistent with increased antigen presentation by uracil auxotroph invaded CD11c+ cells , antigen presenting cells harvested 18 h after uracil auxotroph vaccination of the ID8DV ovarian tumor microenvironment were previously shown to have regained a substantial ability to present the ovalbumin ( OVA OT-I ) peptide on MHCI for recognition by OVA-specific CD8+ T cells [16] . These findings showed that uracil auxotrophs reversed immune suppression in CD11c+ antigen presenting cells in the immune suppressed ID8DV tumor microenvironment [16–18] . The constitutive activation of the host unfolded protein response ER stress pathway mediated by XBP-1 in ID8DV ovarian tumor associated dendritic cells was recently shown to be a central immunosuppressive mechanism that promotes ovarian cancer progression by dampening antigen presentation [115] . Moreover , targeted deletion or inactivation of XBP-1 to quench the unfolded protein response to stress was demonstrated to restore the immunostimulatory activity of ID8DV tumor associated CD11c+ dendritic cells and promoted the development of antitumor CD8+ T cell populations [115] . T . gondii has been shown to manipulate the unfolded protein response through ATF6β [77] . These findings suggest that additional studies are necessary to further address whether the antitumor mechanism elicited by uracil auxotrophs targets host cell stress pathways . We used recently engineered Δku80 genetic models [92 , 103] to target parasite gene deletions to identify secreted parasite molecules that were required for the antitumor response . PVM associated ROP5/ROP17/ROP18 parasite effector molecules play a central role in mediating the antitumor response . In virulent type I parasite strains , ROP5 associates in high molecular weight complexes with either ROP17 or ROP18 to mediate IRG resistance and virulence mechanisms [61–63 , 69] . The kinase activity of ROP17 [62] as well as ROP18 [63] is required for IRG resistance and parasite virulence . In addition to kinase activity , PVM association of the ROP5/ROP17/ROP18 complexes has been shown to be essential for IRG resistance and parasite virulence [62–64] . Our findings show that while PVM association of ROP18 was essential to trigger the antitumor response , the ROP18 kinase activity was not required for this response . Since deletion of ROP5 also abrogated the antitumor response , it seems likely that the antitumor mechanism is mediated through the ROP5/ROP18 PVM complex . In addition to IRG resistance , the kinase activity of ectopically expressed ROP18 in the absence of invasion and the PVM has been previously shown to mediate the degradation of the host NF-κβ p65 subunit [76] as well as the host ER stress sensor ATF6β [77] . It remains to be determined whether a PVM associated kinase inactive ROP18 could associate with host ATF6β . In addition to previously recognized associations of ROP18 with host IRGs , host NF-κβ [76] , and ATF6β [77] , ROP18 was previously shown in yeast two-hybrid assays to associate with additional host cell proteins [116 , 117] . Further studies would be necessary to decipher whether the interaction of ROP18 with any of these host proteins triggers the antitumor mechanism . Our studies show that PVM associated ROP35 and ROP38 molecules were also essential to the antitumor response . The function of ROP35 is poorly understood although this molecule , like ROP38 , exhibits strain type specific variation in mRNA expression levels [78] . ROP38 has been previously shown to modulate the expression level of a large number of host genes involved in MAPK signaling cascades , apoptosis , and host cell proliferation [78] . In addition , the ROP38 gene locus [78] as well as ROP35 , were recently shown to mediate the parasites ability to establish latent infection through a mechanism that did not affect parasite virulence [75] . In addition , ROP35 and ROP38 molecules were not identified in ROP5/ROP17/ROP18 associated protein complexes [62] . These data suggest that the ROP35 and ROP38 antitumor mechanisms most likely function independently of the ROP5/ROP17/ROP18 complexes . Additional studies are thus essential to decipher the ROP35 and ROP38 antitumor mechanisms . We identified GRA24 ( also known as TgBRADIN [118] ) as a secreted dense granule molecule that was crucial to trigger the antitumor response . GRA24 was previously shown to traffic past the PVM to the host cell nucleus and this effector molecule functions to sustain the activation of host cell p38α MAPK by both type I and type II parasite strains [87] . In addition , host cell modulation of IL-12 by GRA24 was previously shown to be parasite strain type dependent and while GRA24 in type II strains stimulated IL-12 production , GRA24 in type I strains did not [87] . Our results show that type I and type II uracil auxotrophs elicited equally potent antitumor responses . These findings suggest that the GRA24 antitumor mechanism is potentially associated with p38α MAPK signaling . The secreted GRA15 molecule was previously shown to elicit IL-12 in type II strain infection but not during infection by type I strains [86 , 100 , 109] . Type II GRA15 increased host cell IL-12 production through strain type dependent manipulation of host NF-κβ signaling [86 , 100] . In contrast , type I ROP16 , but not type II ROP16 , has been shown to suppress IL-12 production [83 , 109] . Our experiments revealed that neither GRA15 or ROP16 were essential for the antitumor response . GRA15 and ROP16 effectors have been previously identified as key molecules that underpin the induction of the M2 phenotype in macrophages polarized by infection with type I parasites , whereas these secreted molecules were associated with the induction of the classical activation M1-like phenotype in macrophages polarized by infection with type II parasites [109 , 119] . In contrast , we previously reported that type I uracil auxotrophs induced a unique profile of macrophage polarization exhibiting features of both the M1 as well as the M2 polarized phenotypes [120] . In addition , vaccination with nonreplicating type I uracil auxotrophs quickly elicited IL-12 in naive mice [7] as well as after therapeutic vaccination of established tumor microenvironments ( present work ) [15 , 16 , 19] . In contrast , infection with virulent type I strains was previously shown to suppress the early production of IL-12 [121] . Together these findings suggest that parasite strain type dependent host cell modulation by certain parasite secreted effectors after vaccination by nonreplicating uracil auxotrophs is mechanistically distinct from host modulation that occurs following infection by replicating parasite strains . Further studies are therefore necessary to understand how , in the absence of parasite replication , secreted effectors that are known to be associated with virulence function to trigger antitumor immunity . In addition to secreted rhoptry effectors and the host nucleus targeted GRA24 , we identified an important role for IVN associated dense granule proteins GRA2 and GRA12 in the antitumor response . Deletion of GRA2 was previously shown to increase IRG coating of the PVM [66] , to slightly reduce parasite virulence [122] , to modulate the presentation of PV membrane bound or soluble PV lumen antigens by MHCI in parasite invaded macrophages and dendritic cells [123] , and to also induce a morphological loss of the appearance of the IVN within the PV luminal spaces without affecting the association of host ER and mitochondria with the PVM [106] . The morphology and dynamic membranes of the IVN are poorly characterized , however , recent helium ion microscopy studies have suggested that IVN structures can extend from the surface of the tachyzoite stage parasite in the PV directly to IVN junctions on the PV lumen side of the PVM [124] . Our findings suggest that the role of GRA2 in formation of the IVN [106 , 123] was essential for the antitumor response . In addition , GRA2 deletion was recently shown to reduce parasite heterophagy , a parasite process that ingests host cell cytoplasm into the lumen of the PV [125] . While the heterophagy mechanism has not yet been precisely determined , it is highly likely that this process involves the PVM , which serves as a membrane barrier between the PV lumen and the host cell cytoplasm . One previously proposed model of the PVM has suggested that invaginations of the PVM could develop into PVM tubules that extend into the PV luminal space [110] . This model hypothesized that these tubular network structures may preferentially attract molecules from the cytosolic face of the PVM that have an affinity for high negative curvature , and then retain these PVM molecules through potential binding partners associated with IVN tubular membranes [110] . While the function of GRA12 is currently unknown , like GRA2 , this molecule also localizes to the IVN structures in the PV space [107] . Remarkably , GRA12 was previously shown to associate specifically with the ROP5/ROP18 complexes that assemble on the PVM [62] . Collectively , these findings suggest that the antitumor mechanism depends on maintenance of dynamic interactions between the IVN and the PVM . Further studies of the parasite IVN and PVM would be necessary to identify the specific mechanisms that underpin the antitumor response mediated by the IVN associated GRA2 and GRA12 dense granule secreted molecules . Our results argue that secretion of rhoptry effector proteins , host cell invasion , formation of the PV , secretion of specific GRA effector proteins to the IVN structures present in the PV lumen , and the export of GRA protein ( s ) past the PVM to the host cell nucleus coordinately triggers the potent ID8DV antitumor immune response elicited by nonreplicating uracil auxotroph mutants of T . gondii .
All procedures involving mice were in accordance with the guidelines published by the Guide for the Care and Use of Laboratory Animals of the National Institute of Health . All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee of Dartmouth College under protocols bzik . dj . 1 and bzik . dj . 2 . All procedures involving mice were reviewed and approved by the Institutional Animal Care and Use Committee ( IACUC ) of Dartmouth College ( Animal Welfare Assurance Number #3259–01 ) and were in accordance with the guidelines published in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . Female 7–9 week old C57BL/6 , IL-12p40-/- , IL-12p35-/- , MyD88-/- , Batf3-/- , IFN-γ-/- , CD8-/- , and MHCII-/- mice were purchased from Jackson laboratories ( Bar Harbor , ME ) and were maintained under specific pathogen-free conditions at the Center for Comparative Medicine and Research at the Geisel School of Medicine at Dartmouth . ID8 cells were provided by Katherine Robey ( University of Kansas Medical Center ) and were transduced with Defβ29 and Vegf-A to establish the ID8-Defβ29/Vegf-A ( ID8DV ) xenograft ovarian tumor model in the C57BL/6 background [90] . ID8DV cells were cultured in Dulbecco's Modified Eagle's Medium ( DMEM ) /high glucose ( HyClone ) in 10% fetal bovine serum ( FBS ) ( Invitrogen ) and Penicillin and Streptomycin ( Corning ) ( final concentration 100 units/ml Penicillin and 100 μg/ml Streptomycin . Human foreskin fibroblasts ( HFFs ) were cultured in Eagle's Minimal Essential Medium ( EMEM ) ( Lonza ) in 10% FBS and Antimycotic-Antibiotic ( Gibco ) ( final concentration 100 units/ml Penicillin , 100 μg/ml Streptomycin , and 25 μg/ml Fungizone ) . Mouse embryonic fibroblasts ( MEFs ) in the C57BL/6 background were obtained from the ATCC and were cultured in DMEM in 15% FBS and Penicillin and Streptomycin . Toxoplasma gondii was maintained by serial passage of tachyzoites in HFF monolayers cultured in EMEM containing 1% FBS , ( 2 mM glutamine , 100 units/ml penicillin , and 100 μg/ml streptomycin ) as previously described [7] . Uracil auxotrophs were supplemented with 0 . 2 mM uracil ( Sigma-Aldrich ) . OMP [13] , CPS [5] , CPS-YFP [96] , and other strains used for treatment of ID8DV tumors were cultured in HFF monolayers and freshly lysed extracellular tachyzoites were purified by filtration through 3 . 0 μm filters ( Nuclepore ) , and washed extensively in Dulbecco's Phosphate Buffered Saline ( PBS ) prior to intraperitoneal injection in 0 . 2 ml PBS in ID8DV ovarian tumor-bearing mice . Aggressive disseminated ovarian tumors were established in mice by intraperitoneal injection of 2 X 106 ID8DV tumor cells [90] . Independent samples of identical ID8DV stocks were used for all experiments . Intraperitoneal treatment of established ID8DV tumors with uracil auxotrophs was performed using 2 x 106 tachyzoites in a one-dose ( 25 d ) , two-dose ( 8 , 20 d ) , three-dose ( 8 , 20 , 32 d ) , or five-dose ( 8 , 20 , 32 , 44 , 56 d ) schedule after ID8DV tumor challenge . Each preparation of tachyzoites was subjected to a plaque forming unit ( PFU ) assay to measure parasite viability and verify treatment dose . Tumor-bearing mice were monitored for wellness and tumor development . Parasite survival was measured in MEFs that were pre-stimulated for 24 h with murine IFN-γ ( Peprotech ) ( final concentration of 100 units/ml ) . Triplicate MEF monolayers that were stimulated or not stimulated with IFN-γ in 24 well culture plates were infected with 100–200 PFU of each parasite strain and incubated for 6 days at 37°C . All PFU in each culture was scored using light microscopy and survival percentage was calculated as the number of PFU in IFN-γ stimulated MEFs divided by the number of PFU scored in nonstimulated MEFs ( no IFN-γ treatment ) . Survival assays of parasite strains were performed in at least three independent experiments for each strain examined . C57BL/6 bone marrow derived macrophages were differentiated as previously described , and 4 x 106 macrophages were seeded in 6-well trays overnight . Macrophages were pre-stimulated , or not stimulated , for 6 h with IFN-γ ( 100 U/ml , Preprotech ) and TNF alpha ( TNF-α ) ( 10 U/ml , Preprotech ) then cultures were infected with ~ 100 tachyzoites per well and cultures were incubated for 6 days to develop PFU . Cultures were stained with coomassie brilliant blue to identify PFU . Survival percentage was calculated as the number of PFU in IFN-γ stimulated macrophages divided by the number of PFU scored in nonstimulated macrophages . Survival assays of parasite strains were performed in at least two independent experiments for each strain examined . The intracellular growth rate of parasite strains was measured in HFF cells in a 30 h growth assay using previously described methods [5] . Briefly , HFF monolayers were infection at a multiplicity of infection ( MOI ) of ~0 . 1 . After 1 h of invasion the noninvaded parasites were removed by extensive ( 4X ) washing in PBS and the number of parasites per vacuole was scored 30 h later from at least 50 vacuoles per sample . The relative rate of invasion of mutant strains was measured in MEFs in comparison to parental OMP . Two hundred tachyzoites of tested strains were added in triplicate to each monolayer of MEF cells in 24 well trays and allowed to invaded for 3 h . Noninvaded parasites were removed by extensive washing in PBS and the number of PFU per well was scored 6–7 days later by microscopy . Experiments were repeated in at least four independent assays . Uracil auxotroph tachyzoites were filtered through 3 μm nuclepore membranes and washed extensively in PBS . Tachyzoites were heat killed at 56°C for 15 minutes as previously described [126] . Excreted-secreted antigen ( ESA ) extracts were prepared by incubation of 2 x 108 tachyzoites for 3 h at 37°C in EMEM supplemented with 10% FBS followed by removal of tachyzoites by centrifugation at 1200g for 7 minutes as previously described [127] . Total Toxoplasma lysate antigen ( TLA ) extracts were prepared by sonication of tachyzoites on ice . Complete loss of viability was verified in PFU assays . Soluble tachyzoite antigen ( STAg ) extracts were prepared by high-speed centrifugation of TLA as previously described [39] . Parasite extracts were stored at -80°C . For each treatment , ID8DV tumor-bearing mice received the equivalent of 5 x 106 tachyzoites of ESA , TLA , or STAg extracts . Tachyzoites were purified using 3 . 0 μm Nuclepore filters , concentrated by centrifugation at 1000g and washed extensively in PBS . Tachyzoites ( 2 x 107/ml ) were treated with 100 μM 4-bromophenylacyl bromide ( 4BPB ) ( Sigma-Aldrich ) for 10 min at 37°C and quenched with 20 volumes of complete media and washed in PBS [126] . Tachyzoites ( 2 x 107/ml ) in PBS were treated with 3 μM mycalolide B ( Wako ) for 15 min at 37°C and quenched with 20 volumes of complete media and washed in PBS [100 , 101] . Parasite inactivation of invasion was verified in PFU assays . Purified monoclonal antibodies αCD8 ( 2 . 43 ) , αCD4 ( GK1 . 5 ) , and isotype control Rat IgG2a were purchased from BioXcell . For cell depletions , 500 μg of antibody was administered by intraperitoneal injection 1 day prior to and 250 μg of antibody was administered by intraperitoneal injection 0 and 3 d after each treatment with uracil auxotrophs . NK cells were depleted using αNK1 . 1 ( PK136 ) antibody ( a generous gift from Dr . Charles Sentman at Dartmouth , EBioscience ) , and intraperitoneal injections of 50 μg of αNK1 . 1 antibody were given on day -1 , 0 , and +3 relative to each treatment with uracil auxotrophs [19] . In all experiments target cell populations were depleted by greater than 98% . Peritoneal fluid obtained by peritoneal lavage using 5 ml PBS [16] was used for the detection of cytokines in the ovarian tumor microenvironment . IL-12p40 and IL-12p70 were determined using duplex luminex assays ( Millipore ) . IFN-γ , IL-1α and IL-1β were measured using a mouse 32-plex Luminex assay ( Millipore ) . Type I interferons were measured using a mouse IFN alpha and IFN beta 2Plex assay ( EBioscience ) . T . gondii genomic DNA was purified from tachyzoites using the DNA Blood Mini Kit ( Qiagen ) on a robotic Qiacube ( Qiagen ) . PCR products for targeting vector construction were amplified from primers ( Integrated DNA Technologies ) using High Fidelity polymerases ( Roche ) . The Toxoplasma genome resource ( www . toxodb . org ) [128] was used to identify gene loci of interest and sequences to design oligonucleotide primers . Targeted ROP and GRA knockouts were generated in the OMP parental uracil auxotroph strain RHΔku80ΔompdcΔupΔhxgprt which was generated by the deletion of HXGPRT selectable marker from the uridine phosphorylase of OMP ( strain RHΔku80ΔompdcΔup::HXGPRT ) locus using 6-thioxanthine ( 250 μg/ml ) ( Acros Biochemicals ) and uracil selection as previously described [13] . ROP and GRA gene targeting plasmids were developed in the pRS416 yeast shuttle vector using yeast recombination to fuse , in order , 3 distinct PCR products with 31 to 34 bp crossovers on DNA fragments that included a 5’ target gene flank , the HXGPRT selectable marker , and a 3’ target flank ( S6 Fig ) . Gene targeting plasmids were engineered to delete the entire ROP or GRA gene locus including 400 bp of the immediate 5’ UTR . The DNA primers used to generate the 5' and 3' gene targeting flanks and the nucleotides deleted in each ROP and GRA knockout strain are shown in supplemental material ( S1 Table ) . Targeting plasmids were linearized via a unique restriction site at the junction of the 5' targeting flank and pRS416 vector prior to transfection into RHΔku80ΔompdcΔupΔhxgprt and continuous selection in mycophenolic acid ( 25 μg/ml ) and xanthine ( 250 μM ) and uracil as previously described [13 , 14 , 92 , 103 , 104] . Isolates were subcloned 20 days after transfection by limiting dilution . Targeted knockouts were validated by genotype analysis using PCR assays ( S6 Fig ) to measure: ( i ) PCR 1 , targeted deletion of the coding region of the targeted gene ( DF and DR primers ) ; ( ii ) PCR 2 , correct targeted 5' integration ( CXF & 5'DHFRCXR primers ) ; and ( iii ) PCR 3 , correct targeted 3' integration ( 3'DHFRCXF and CXR primers ) using DNA validation primers shown in supplemental material ( S2 Table ) . Selected knockout strains were complemented with wild-type or mutant alleles of the deleted ROP gene . A new positive genetic selection was developed for uracil auxotroph backgrounds based on cytosine auxotrophy using the cytosine deaminase ( CD ) selectable marker that converts cytosine to uracil [105] . Genes for complementation analysis were C-terminally tagged with the HA peptide to allow fluorescent visualization of protein expression and the complementing genes were targeted for integration at the OMPDC gene locus that was already deleted for OMPDC and HXGPRT [13] . Complementation targeting plasmids were developed in the pRS416 yeast shuttle vector using yeast recombination to fuse , in order , a 5' OMPDC target flank , the complementing gene of interest plus the genes 5' UTR ( ~ 1 Kbp ) synthesized on one or two PCR products , the CD selectable marker , and the 3' OMPDC target flank ( S7 Fig ) . The DNA primers used to generate the complementing gene ( synthesized on one two or three PCR products ) , the CD marker , and target flanks are shown in supplemental material ( S1 Table and S3 Table ) . Mutant ROP18 alleles expressing a kinase-dead ( KD ) ROP18 , or mutant ROP18 lacking the PVM and ATF6β association domain ( RAH2ATF ) were designed using previously described mutations [63 , 77] . Targeting plasmids were linearized via a unique restriction site at the junction of the 5' targeting flank and pRS416 vector prior to transfection . Following transfection parasites were grown 1 d in uracil medium , then were switched to selection medium containing 1 mM cytosine , and isolates were subcloned 30 days later by limiting dilution . Targeted knockouts were validated by genotype analysis using PCR assays ( S7 Fig ) to measure: ( i ) PCR 4 , correct targeted 5' integration and ( ii ) PCR 5 , correct targeted 3' integration of the complementing transgene and CD at the OMPDC locus using DNA validation primers ( S4 Table ) . HFF cells were cultured on circular micro cover glass ( Electron Microscopy Sciences ) and were infected with parasites for 18–30 h . For localization of rhoptry proteins to the rhoptry organelles , infected cells were fixed with Histochoice ( Amresco ) , and permeabilized in 0 . 1% saponin ( Sigma ) for 10 min . For localization of rhoptry proteins to the PVM , infected cells were fixed in 4% paraformaldehyde and exposed to 0 . 002% digitonin for 5 minutes as previously described [62] to expose proteins associated with the cytosolic face of the PVM . All samples were blocked with 10% FBS and incubated with a 1:500 dilution of primary rabbit monoclonal α-HA-tag antibodies ( Cell Signaling ) 1 h at room temperature . Preparations were washed 3 times with PBS and incubated 1 h at RT with a 1:1000 dilution of secondary goat anti-rabbit IgG antibodies conjugated to Alexa Fluor 488 . Samples were mounted in Slowfade Gold antifade with DAPI ( Life Technologies ) and then imaged with a Nikon A1R SI confocal microscope ( Nikon , Inc . ) . Confocal images were processed with FIJI [129] . Vacuole locations were determined by differential interference contrast ( DIC ) microscopy . Antibodies used were CD16/CD32 ( 93 , ebioscience ) , APC-Cy7-conjugated anti-mouse CD45 ( 30-F11 , Biolegend ) , and AF647-conjugated anti-mouse CD11c ( N418 , Biolegend ) . FACS analysis was performed using a Miltenyi 8-color MACSQuant and data was analyzed using FlowJo v9 . 77 and v10 ( TreeStar ) . Statistical analysis was performed using PRISM software ( Graphpad Software ) . Statistical significance in acute virulence assays in mice was determined using the Log-rank Mantel-Cox test performed using the Gehan-Breslow-Wilcoxon test . P values less than or equal to 0 . 05 ( P<0 . 05 ) were considered to be significant . All other statistic analysis was performed using an unpaired two-tailed Students t test with the assumption of equal variance , and a P value less than or equal to 0 . 05 ( P<0 . 05 ) was considered to be significant .
|
Toxoplasma gondii extensively manipulates cellular signaling pathways and host immune responses through secreted effector proteins , yet the host rapidly establishes T cell immunity to control acute infection thereby permitting survival of the host as well as survival of the parasite in latent infection . Recently , vaccination of mice bearing highly aggressive ovarian cancer with a safe nonreplicating , noncyst forming , vaccine strain of T . gondii was shown to effectively reverse tumor associated immune suppression and activate potent antitumor immunity . Using a new genetically tractable Δku80 vaccine strain of T . gondii we deleted multiple parasite secreted effector molecules to explore parasite specific mechanisms associated with the development of potent antitumor immunity . Our results demonstrate that specialized effector proteins secreted by T . gondii both before and after host cell invasion trigger and coordinately control the development of a potent antitumor response . Consequently , tracking and understanding the host cell pathways manipulated by these T . gondii secreted effector proteins can reveal fundamental mechanisms controlling immunity to infection and can also identify relevant mammalian cell mechanisms as new targets for devising more effective therapies against highly aggressive solid tumors .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"nucleobases",
"viral",
"transmission",
"and",
"infection",
"toxoplasma",
"gondii",
"cancer",
"treatment",
"immunology",
"microbiology",
"parasitic",
"diseases",
"nucleotides",
"parasitic",
"protozoans",
"parasitology",
"oncology",
"apicomplexa",
"tachyzoites",
"protozoans",
"toxoplasma",
"uracils",
"cytotoxic",
"t",
"cells",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"pathogenesis",
"biochemistry",
"host",
"cells",
"cell",
"biology",
"host-pathogen",
"interactions",
"virology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2016
|
Secretion of Rhoptry and Dense Granule Effector Proteins by Nonreplicating Toxoplasma gondii Uracil Auxotrophs Controls the Development of Antitumor Immunity
|
MicroRNAs regulate networks of genes to orchestrate cellular functions . MiR-125b , the vertebrate homologue of the Caenorhabditis elegans microRNA lin-4 , has been implicated in the regulation of neural and hematopoietic stem cell homeostasis , analogous to how lin-4 regulates stem cells in C . elegans . Depending on the cell context , miR-125b has been proposed to regulate both apoptosis and proliferation . Because the p53 network is a central regulator of both apoptosis and proliferation , the dual roles of miR-125b raise the question of what genes in the p53 network might be regulated by miR-125b . By using a gain- and loss-of-function screen for miR-125b targets in humans , mice , and zebrafish and by validating these targets with the luciferase assay and a novel miRNA pull-down assay , we demonstrate that miR-125b directly represses 20 novel targets in the p53 network . These targets include both apoptosis regulators like Bak1 , Igfbp3 , Itch , Puma , Prkra , Tp53inp1 , Tp53 , Zac1 , and also cell-cycle regulators like cyclin C , Cdc25c , Cdkn2c , Edn1 , Ppp1ca , Sel1l , in the p53 network . We found that , although each miRNA–target pair was seldom conserved , miR-125b regulation of the p53 pathway is conserved at the network level . Our results lead us to propose that miR-125b buffers and fine-tunes p53 network activity by regulating the dose of both proliferative and apoptotic regulators , with implications for tissue stem cell homeostasis and oncogenesis .
MicroRNAs ( miRNAs ) are short non-coding RNA molecules that were first discovered as regulators of developmental timing , and later found to regulate complex networks of genes to orchestrate cellular functions . Lin-4 was the first miRNA gene to be discovered , and shown to regulate developmental timing by repressing its target genes at the post-transcriptional level [1] . Subsequently , miRNAs were found to regulate processes ranging from proliferation and apoptosis , to cell differentiation and signal transduction [2]–[4] . Several miRNAs are conserved in metazoan evolution , one prominent example being lin-4 whose vertebrate homologues comprise the miR-125a/b family [5] . Much like lin-4′s role of regulating the homeostasis of reiterative or self-renewing stem cells in C . elegans [6] , recent studies have shown that miR-125a/b regulates mammalian neural stem cell commitment , as well as the mammalian hematopoietic stem cell ( HSC ) pool size [7]–[10] . Although Lin28 and Bak1 have been proposed as the critical targets of miR-125a/b for regulating these stem cell compartments [8] , [9] , the hundreds of predicted targets for miR-125a/b suggest a more complex interplay between miR-125a/b and its targets in regulating proliferation and differentiation . Depending on the cell context , miR-125b has been proposed to regulate both apoptosis and proliferation . miR-125b has been shown to downregulate apoptosis in many contexts , in some cases by repressing Tp53 and Bak1 . Examples include mammalian hematopoietic stem cells , human leukemia cells , neuroblastoma cells , breast cancer and prostate cancer cells [9]–[18] . During zebrafish embryogenesis , loss of miR-125b leads to widespread apoptosis in a p53-dependent manner , causing severe defects in neurogenesis and somitogenesis [16] . On the other hand , miR-125b can also downregulate proliferation in a variety of human cancer cell-lines [19]–[23] and one of its bona fide targets Lin28 , also promotes cancer cell proliferation [24] . Therefore in different contexts , miR-125b appears to be able to regulate both apoptosis and proliferation . Another molecular pathway that regulates both apoptosis and proliferation is the highly conserved p53 network [25]–[28] . Due to the central role of the p53 network in these two processes , and because we found that miR-125b regulates both human and zebrafish Tp53 but not mouse Tp53 [16] , we sought to examine if miR-125b regulates the p53 network in a conserved manner in vertebrates . To address this question , we used a gain- and loss-of-function screen for miR-125b targets in different vertebrates , and validated these targets with the luciferase assay and a novel miRNA-target pull-down assay . We demonstrate that miR-125b directly represses 20 novel targets in the p53 network , including both apoptosis regulators like Bak1 , Igfbp3 , Itch , Puma , Prkra , Tp53inp1 , Tp53 , Zac1 , and also cell-cycle regulators like cyclin C , Cdc25c , Cdkn2c , Edn1 , Ppp1ca , Sel1l . We found that although individual miRNA-target pairs were seldom conserved , regulation of the p53 network by miR-125b appears to be conserved at the network-level . This led us to propose that miR-125b buffers and fine-tunes p53 network dosage , with implications for the role of miR-125b in tissue stem cell homeostasis and oncogenesis .
To systematically identify direct targets of miR-125b in the p53 network of vertebrates , we first employed a bioinformatics approach by identifying all predicted miR-125b targets in the p53 network , followed by three complementary methods to screen and validate these targets for both direct binding and repression by miR-125b ( Figure 1 ) . Existing databases and prediction algorithms were used to shortlist a set of p53 network genes predicted to possess miR-125b-binding sites in their 3′ UTRs . We analyzed the Ingenuity Pathways Analysis™ ( IPA ) database and the p53 Knowledgebase [29] , [30] for a list of genes and proteins that participate in the p53 network , either by regulating p53 upstream , by direct interaction with p53 protein , or by serving as effectors of p53 function downstream . We then analyzed the TargetScan and MicroCosm Target databases [31] , [32] for genes that are predicted to possess miR-125b-binding sites in their 3′ UTRs , in three vertebrate genomes: human , mouse and zebrafish . The genes at the intersection of the predicted miR-125b target list and the list of p53 network genes constituted our list of predicted miR-125b targets in the p53 network ( Table S1 ) . Next we sought to screen our list of predicted targets for significant repression by miR-125b in cells , by performing a miR-125b gain- and loss-of-function screen . Gain-of-function ( GOF ) in miR-125b was achieved by transfection of miR-125b duplex into human SH-SY5Y or mouse N2A neuroblastoma cells , whereas loss-of-function ( LOF ) in miR-125b was achieved in human primary lung fibroblasts or mouse 3T3 fibroblasts by knocking down miR-125b with an antisense ( AS ) RNA ( Figure 2A ) . We chose to perform a gain-of-function screen in human ( SH-SY5Y ) or mouse ( N2A ) neuroblastoma cells , because these cells possess low levels of endogenous miR-125b ( Figure S1A , S1B ) . For the loss-of-function screen , we chose human fetal lung ( hLF ) or mouse ( 3T3 ) fibroblasts because they possess high levels of miR-125b ( Figure S1C , S1D ) . miR-125a-AS was co-transfected with miR-125b-AS to achieve a complete silencing of the miR-125a/b family , because miR-125a , which shares the same seed sequence and the same predicted targets as miR-125b , is also highly expressed in human and mouse fibroblasts ( Figure S1C , S1D ) . Genes that were either significantly repressed by miR-125b or significantly derepressed by miR-125a/b-AS with fold-changes within the range of microRNA regulation ( P<0 . 05 , fold change > 1 . 3 ) , were selected as candidate miR-125b targets ( Figure 2B–2D ) . For zebrafish embryos , which possess high levels of miR-125b , the loss-of-function ( LOF ) screen was performed using an antisense morpholino cocktail that blocks the loop regions of all 3 pre-miR-125b hairpin precursors [16] . The gain-of-function ( GOF ) screen was performed by co-injecting miR-125b duplex with the morpholino ( Figure 2A ) . All gene expression changes were measured with at least three biological replicates using qRT-PCR . Our GOF/LOF screen revealed that in humans , out of 29 predicted targets in the p53 network , 13 genes were derepressed by miR-125a/b-AS in hLF cells and 20 genes were repressed by miR-125b in SH-SY5Y cells ( Figure 2B ) . In mice , out of 22 predicted targets in the p53 network , 11 genes were derepressed by miR-125a/b-AS in 3T3 cells and 12 genes were repressed by miR-125b in N2A cells ( Figure 2C ) . In zebrafish embryos , out of 20 predicted targets in the p53 network , 13 genes were derepressed by pre-miR-125b morpholino and 12 genes were repressed by the injection of miR-125b duplex ( Figure 2D ) . In total , 22 human genes , 13 mouse genes and 14 zebrafish genes passed the gain- and loss-of-function qRT-PCR screen . To assess which candidate miR-125b targets identified in the gain- and loss-of-function qRT-PCR screen are directly bound by miR-125b in cells , we employed a novel miRNA pull-down method developed by Lal et al . ( manuscript in preparation ) . RNA transcripts bound to biotinylated-miR-125b were pulled down with streptavidin beads and quantified by qRT-PCR relative to mRNAs bound to biotinylated-control miRNA ( log2 fold change > 0 . 5 , P<0 . 05 ) . In this assay , biotinylated miRNAs were shown to be loaded into the RNA-induced silencing complex ( RISC ) and fully functional in repressing their target mRNAs ( Lal et al . , manuscript in preparation ) . This method provides a robust and complementary method for detecting miRNAs bound to endogenous target mRNAs , and serves as a useful approach for distinguishing direct and indirect targets in the same pathway ( Lal et al . , manuscript in preparation ) . Quantification of the pulled down mRNA targets in hLF cells revealed that 13 out of 22 gene transcripts , Bak1 , Cdc25c , Edn1 , Igfbp3 , Mre11a , Ppp1ca , Ppp2ca , Prkra , Puma , Tdg , Tp53 , Tp53inp1 and Zac1 , were direct binding targets of miR-125b in human cells ( Figure 3A ) . In mouse 3T3 cells , 11 out of 13 gene transcripts , Bak1 , Hspa5 , Itch , Ppp1ca , Ppp2ca , Prkra , Puma , Sel1l , Sp1 , Tdg and Tp53inp1 , were found to be direct binding targets of miR-125b ( Figure 3B ) . In zebrafish embryos , 8 out of 14 gene transcripts , Cdc25c , Cdkn2c , Gtf2h1 , Hspa5 , Itch , Ppp1ca , Sel1l , and Tp53 , were pulled down by miR-125b ( Figure 3C ) . Tp53 mRNA was pulled down by miR-125b only in human lung fibroblasts and zebrafish embryos but not in mouse fibroblasts , consistent with previously published results [16] and the Targetscan algorithmic prediction that miR-125b targets Tp53 in humans and zebrafish but not in mice . As a final validation of the candidate miR-125b targets we have identified in the p53 network , we tested our candidate target genes with the luciferase reporter assay . Where cloning was successful , we cloned the entire 3′ UTR of selected candidate target genes into a Renilla luciferase reporter , and assayed luciferase expression following co-transfection of miR-125b duplex into HEK-293T cells . Transfection of miR-125b significantly suppressed 40-60% ( P<0 . 01 ) of the luciferase activity of many 3′ UTR reporters of the miR-125b targets we analyzed , relative to transfection of the negative control miRNA ( Figure 4A ) . For humans , the 3′ UTR reporters of Bak1 , Cdc25c , Ppp1ca , Ppp2ca , Prkra , Puma , Tdg , Tp53 , Tp53inp1 , and Zac1 were significantly suppressed by miR-125b . In mice , the 3′ UTR reporters of Bak1 , Itch , Ppp1ca , Ppp2ca , Prkra , Puma , Sel1l , Tdg , and Tp53inp1 were significantly suppressed by miR-125b ( Figure 4B ) . In zebrafish , the 3′ UTR reporters of Ccnc , Cdc25c , Cdkn2c , Gtf2h1 , Hspa5 , Ppp1ca , and Tp53 were significantly suppressed by miR-125b ( Figure 4C ) . With the exception of zebrafish Ccnc , all genes tested were positive in the miR-125b-pull-down as well as the miR-125b gain- and loss-of-function screen . Amongst these targets , we found Ppp1ca , Prkra and Tp53 to be especially interesting from the evolutionary viewpoint , since all 3 vertebrate species possess these 3 genes , but each gene shows a different pattern of evolutionary conservation with respect to miR-125b-repression . Ppp1ca is repressed by miR-125b in all 3 species , Prkra is repressed by miR-125b in humans and mice , while Tp53 is repressed in humans and zebrafish . To examine the sequence evolution of these miRNA-mRNA pairs in greater detail , we compared the Targetscan-predicted miR-125b binding sites of these genes in humans , mice and zebrafish . In Ppp1ca , the predicted binding site is 95% identical between humans and mice and 55% identical between humans and zebrafish , while the seed binding sequence is 100% conserved in all 3 species ( Figure 4D ) . In Prkra , the predicted binding site is 94% identical between humans and mice , but only 26% identical between humans and zebrafish , while the seed binding sequence is completely absent in zebrafish ( Figure 4D ) . In contrast , the predicted binding site in Tp53 is 64% identical between humans and zebrafish , and the seed binding sequence is 100% conserved between humans and zebrafish , but only 36% identical between humans and mice , while the mouse seed binding sequence has acquired 2 point mutations ( Figure 4D ) . The miR-125b-repression patterns we observed for each of these genes in the qPCR , pull-down and luciferase assays are consistent with these DNA sequence analyses , suggesting that evolution in the miRNA-mRNA binding is driving the evolution in miR-125b-repression patterns . Introduction of point mutations into the predicted seed binding sequences abrogated miR-125b-repression of each target 3′UTR luciferase reporter ( P<0 . 05 ) , validating the predicted miR-125b binding sites and confirming the miRNA-mRNA sequence evolution patterns we observed ( Figure 4E ) . Finally , we checked miR-125b regulation of protein expression in a subset of p53 network targets for which reliable Western blotting was possible . miR-125b significantly downregulated the protein levels of human BAK1 , PPP1CA , TP53INP1 , PPP2CA , CDC25C , and TP53 in SH-SY5Y neuroblastoma cells ( Figure 4F ) . In mouse N2A neuroblastoma cells , miR-125b significantly downregulated mouse BAK1 , PPP1CA , PUMA , and ITCH protein ( Figure 4G ) . Our results reveal that miR-125b regulation of the p53 network is conserved at the network-level over the course of vertebrate evolution , but individual miRNA-target pairs are evolving rapidly . To summarize our results , our list of predicted miR-125b targets in the p53 network ( Table S1 ) was filtered and reclassified according to the results of the screen and validation assays ( Figure 5 ) . From the GOF/LOF screen we were able to identify mRNAs perturbed by miR-125b . However these results did not discriminate between direct or indirect targets . To supplement these experiments the pull-down assay was used to uncover mRNAs physically associated with miR-125b . Of note , the pull-down might not identify mRNA targets that are rapidly degraded , and as such the luciferase reporter assay can complement its shortcomings . Taken together the three assays provide a powerful means to identify direct miR-125b targets . In order to minimize false positives , we counted the number of assays for which each gene target was positive , and gene targets that failed to pass at least 2 assays in at least one vertebrate species were filtered out . Predicted targets that passed 3 assays ( red ) , 2 assays ( orange ) , 1 assay ( yellow ) , or predicted targets that failed all assays but whose orthologues in other species passed 3 assays of direct regulation by miR-125b ( pink ) , were colored as indicated ( Figure 5 ) . Using our conservative estimate of miR-125b targets in the p53 network , we found that in all three vertebrates we examined – humans , mice and zebrafish – miR-125b regulates multiple p53 network genes . This shows that miR-125b regulation of the p53 network is conserved at least at the network level . However very few individual gene targets of miR-125b in the p53 network were conserved across all three vertebrates ( Figure 5; Figure 6A-6C ) . Instead , conserved miR-125b regulation of the p53 network appears to occur through evolving miRNA-target pairs in the three vertebrates – zebrafish ( Figure 6A ) , mouse ( Figure 6B ) , and humans ( Figure 6C ) . In general , we observe miR-125b regulating 2 general classes of genes in the p53 network: ( i ) apoptosis regulators like Bak1 , Igfbp3 , Itch , Puma , Prkra , Tp53inp1 , Tp53 , and Zac1 , and ( ii ) cell-cycle regulators like cyclin C , Cdc25c , Cdkn2c , Edn1 , Ppp1ca , and Sel1l . Because miR-125b represses both pro-apoptosis and anti-apoptosis genes , as well as both proliferation and cell-cycle arrest genes in all three vertebrates ( Figure 5 ) , miR-125b appears to modulate the p53 network on the whole through an incoherent feedforward loop ( FFL ) [33] , [34] acting on the cellular processes of apoptosis and cell proliferation ( Figure 6D ) . An incoherent type-2 FFL is a regulatory pattern in which X represses a target Z and also represses Y , another repressor of Z ( Figure 6D ) . Incoherent FFLs have been found in the transcription factor networks of human embryonic stem cells and hematopoietic stem cells , and have been shown to modulate E2F1 dosage in the Myc-E2F1 pathway [35]-[37] . Besides accelerating responses and acting as amplitude filters [38]-[40] , the incoherent FFL motif is also a noise buffering motif that reduces the variance of network dosage [41]–[43] . Thus our finding that incoherent FFLs fit the overall structure of network relationships between miR-125b and the p53-mediated processes , suggests that miR-125b is fine-tuning and buffering p53 network dosage .
In this study , we sought to identify direct targets of miR-125b in the p53 network of humans , mice and zebrafish , to better understand how miR-125b regulates the p53 network throughout evolution and how that might relate to its conserved role in regulating tissue stem cells . We identified 20 direct targets of miR-125b in the p53 network , including 15 novel targets like Zac1 , Puma , Itch and Cdc25c , and also targets like Bak1 and Tp53 that were identified in previous studies [9] , [16]–[18] . In general , we found that miR-125b directly represses 2 classes of genes: apoptosis regulators and cell-cycle regulators . With the exception of Ppp1ca , Itch and Edn1 , very few individual targets were strictly conserved throughout vertebrate evolution . Instead , we found that only the network-level of regulation was conserved , and miR-125b-regulation of individual apoptosis and proliferation regulators appears to be evolving rapidly from species to species . This observation suggests that , at least within the vertebrates , the 3′ UTR sequences of each gene target is evolving rapidly via neutral genetic drift . In other words , the loss or gain of a single miR-125b-binding site in the 3′ UTR of most genes appears to have a relatively insignificant effect on the fitness of an organism . On the other hand , the strict conservation of miR-125b-regulation at the network-level in humans , mice and zebrafish , suggests that natural selection acts on the network-level rather than the gene-level with regard to miRNA-target evolution . It will be interesting to see if this novel paradigm applies to other microRNAs or gene networks as well . Previous studies on miRNA evolution have suggested that a relatively poor conservation of individual miRNA-target pairs but strong conservation of a miRNA-gene network relationship is consistent with miRNAs' role as buffers of gene expression [42] , [44] , [45] . Our observation that an incoherent FFL-like network motif fits the overall structure of the miR-125b - p53 network models with respect to apoptosis and cell proliferation , lends further support to this idea since incoherent FFL network motifs are well-adapted for noise filtering [41] , [43] , [46] . It is thought that miRNAs are at least partially responsible for the phenomenon of developmental or phenotypic stability within each species [41] , [42] , termed “canalization” by C . H . Waddington [47] . These studies suggest that miRNAs have a conserved role in regulating the overall stability of pathways/networks , a role which is relatively unaffected by the loss or gain of individual miRNA-targets over the course of evolution . A network buffering function has also been suggested for the regulation of muscle development by miR-1 throughout evolution , regulation of the Wnt pathway by miR-8 , and fine-regulation of Pten dosage by a variety of miRNAs [48]–[50] . Our findings suggest that the fine-tuning of p53 network dosage by miR-125b is another example of this paradigm . Fine-regulation of p53 network dosage by miR-125b may also explain miR-125b's conserved role in regulating tissue stem cell homeostasis . In C . elegans , loss-of-function mutations in lin-4 lead to a delay in differentiation and thus expansion of vulval precursor cells , seam stem cells in the lateral hypodermis and mesoblasts , causing multiple defects in larval development [6] . In zebrafish , loss of miR-125b leads to widespread p53-dependent apoptosis with consequent defects in early embryogenesis , especially in neurogenesis and somitogenesis [16] . Overexpression of miR-125a/b causes an expansion of mammalian hematopoietic stem cells ( HSCs ) and aberrant differentiation , leading to myeloid leukemia [9] , [10] and also lymphoid leukemia if miR-125b is overexpressed in fetal liver HSC-enriched cells [12] . However , the molecular underpinnings of miR-125a/b's regulation of tissue stem cell homeostasis had remained unclear largely due to the complex nature of microRNA regulation of gene networks . The 2 classes of miR-125b targets in the p53 network , and the incoherent FFL network motifs that we found , may at least partially explain how miR-125b regulates tissue stem cells in vertebrates . By fine-tuning both apoptosis regulators and cell-cycle regulators , miR-125b may fine-tune the p53 network dosage to drive the self-renewal of tissue stem cells . It could explain how overexpression of miR-125b leads to an expansion of self-renewing hematopoietic stem cells while loss of miR-125b leads to aberrant apoptosis and proliferation , with consequent defects in tissue differentiation . Several studies have implicated miR-125b as an oncogene in a variety of mammalian tissue compartments , e . g . leukemia , neuroblastoma , prostate cancer and breast cancer [9]–[18] . These studies have ascribed miR-125b's anti-apoptotic effect as an oncogene to its direct suppression of Bak1 or Tp53 [9] , [16]-[18] . On the other hand , several research groups have also reported miR-125b's role as a potential tumor suppressor by suppressing proliferation in cell-culture models [19]–[23] . Our identification of 20 direct targets of miR-125b in the p53 network reconciles these findings because miR-125b modulates the expression of both apoptosis regulators and cell-cycle regulators . Although miR-125b's suppression of p53 itself is not conserved in mice , miR-125b's anti-apoptotic role – through suppression of multiple pro-apoptosis regulators in the p53 network – appears to be conserved in vertebrates . miR-125b's ability to fine-tune the subtle balance of apoptosis vs . cell-cycle regulators and thus buffer the p53 network dosage in different contexts , could explain why miR-125b dysregulation can lead to either tumor suppression or oncogenesis depending on the context . It is possible that this buffering feature of miR-125b represents a general principle of miRNA regulation of gene networks .
A list of p53-associated genes was compiled from the p53 Knowledgebase website [30] and from the Ingenuity Pathway Analysis™ database [29] . The targets of miR-125b in human and mouse were predicted by TargetScan [31] . The targets of miR-125b in zebrafish were predicted by MicroCosm [32] . The human homologues of mouse and zebrafish targets were identified by the DAVID gene ID conversion tool . Human lung fibroblast cells , human neuroblastoma SH-SY5Y cells , mouse neuroblastoma Neuro-2A cells , mouse fibroblast Swiss-3T3 cells and human HEK-293T cells were maintained in DMEM media , supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin ( Invitrogen ) . Neuro-2A cells , 3T3 cells , SH-SY5Y cells and human lung fibroblast cells were transfected in suspension with 5×105 cells per well in 6-well plates using lipofectamin-2000 ( Invitrogen ) . miRNA duplexes and antisense oligonucleotides ( Ambion ) were transfected at a final concentration of 80 nM . Wild-type zebrafish were maintained by standard protocols [51] . All injections were carried out at 1–4 cell stage with 2 nl of solution into each embryo . In the knockdown experiments , miR-125b morpholinos were injected at 0 . 75 pmole/embryo ( lp125bMO1/2/3 indicates the co-injection of three lp125bMOs , 0 . 25 pmole each ) ; miR-125b duplex was injected as 37 . 5 fmole/embryo . RNA was extracted from cells or zebrafish embryos using Trizol reagent ( Invitrogen ) and subsequently column-purified with RNeasy kits ( Qiagen ) . For qRT-PCR of miR-125a , miR-125b and RNU6B , 100 ng of total RNA was reverse-transcribed and subjected to Taqman microRNA assay ( Applied Biosystems ) . For qRT-PCR of mRNAs , cDNA synthesis was performed with 1 µg of total RNA using the High Capacity cDNA Archive Kit ( Applied Biosystems ) . The expression of all genes was analyzed by SYBR assay using the Applied Biosystems real-time PCR system or the Fluidigm 48x48 dynamic array system ( Fluidigm ) following the manufacturer's protocol . 50 ul of streptavidin coated magnetic beads ( Invitrogen ) were blocked with 1 mg/ml yeast tRNA and 1 mg/ml BSA in 1 ml lysis buffer ( 20 mM Tris pH 7 . 5 , 100 mM KCl , 5 mM MgCl2 , 0 . 3% NP-40 ) for 2 hours at 4°C and wash twice with lysis buffer . hsa-miR-125b or cel-miR-67 ( negative control ) duplex was synthesized with a biotin conjugated at the 3′ end of the active strand by Dharmacon Research Inc . The miRNAs were transfected into human lung fibroblasts or mouse 3T3 fibroblasts at a final concentration of 80 nM as described above . The miRNAs were also injected into zebrafish embryos at 1 to 4-cell stage at a final concentration of 37 . 5 fmole/embryo . After 24 hours , cells from 3 wells of fibroblasts or 50 zebrafish embryos were incubated with 500 ul cold lysis buffer containing freshly added 100 units/ml RNase inhibitor ( Invitrogen ) and protease inhibitor cocktail ( Roche ) for 20 minutes on ice . After the cell debris is removed by centrifugation , the lysate was incubated with pre-blocked streptavidin coated beads for 2 hours at 4°C . Subsequently , the beads were washed 5 times with cold lysis buffer and incubated with Trizol for RNA extraction . The whole 3′ UTR of target genes were cloned into the psiCHECK-2 vector ( Promega ) , between the XhoI and NotI site , immediately 3’ downstream of the Renilla luciferase gene . For selected targets , we introduced 3 point mutations into the 7-nt seed-binding sequence using inverse PCR with non-overlapping primers carrying the mutated sequences . 10 ng of each psiCHECK-2 construct was co-transfected with 10 nM miRNA duplexes or into HEK-293T cells in a 96-well plate using lipofectamin-2000 ( Invitrogen ) . After 48 hours , the cell extract was obtained; Firefly and Renilla luciferase activities were measured with the Dual-Luciferase reporter system ( Promega ) according to the manufacturer's instructions . Cells were lysed in RIPA buffer ( Pierce ) . Proteins were separated by a 10% polyacrylamide gel and transferred to a methanol-activated PVDF membrane ( GE Healthcare ) . The membrane was blocked for one hour in PBST containing 7 . 5% milk and subsequently probed with primary antibodies ( Santa Cruz ) overnight at 4°C . After 1-hour incubation with goat-anti-mouse HRP-conjugated secondary antibody ( Santa Cruz ) , the protein level was detected with luminol reagent ( Santa Cruz ) . Two-tail T-tests were used to determine the significance of differences between the treated samples and the controls where values were obtained from luciferase reporter assay or qRT-PCR . The tests were performed using Microsoft Excel where the test type is always set to two-sample equal variance .
|
MicroRNAs ( miRNAs ) are tiny endogenous RNAs that can regulate the expression of hundreds of genes simultaneously , thus orchestrating changes in gene networks and mediating cellular functions in both plants and animals . Although the identification of individual targets of miRNAs is of major importance , to date few studies have sought to uncover miRNA targets at the gene network level and general principles of miRNA regulation at the network level . Here we describe how miR-125b targets 20 apoptosis and proliferation genes in the p53 network . We found that , although each miRNA-target pair evolves rapidly across vertebrates , regulation of the p53 pathway by miR-125b is conserved at the network level . The structure of the miR-125b regulatory network suggests that miR-125b buffers and fine-tunes p53 network activity . This buffering feature of miR-125b has implications for our understanding of how miR-125b regulates oncogenesis and tissue stem cell homeostasis . We believe these findings on miR-125b support a new fundamental principle for how miRNAs regulate gene networks in general .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"rna",
"interference",
"gene",
"expression",
"genetics",
"biology",
"genetics",
"and",
"genomics"
] |
2011
|
Conserved Regulation of p53 Network Dosage by MicroRNA–125b Occurs through Evolving miRNA–Target Gene Pairs
|
Although cutaneous ulcers ( CU ) in the tropics is frequently attributed to Treponema pallidum subspecies pertenue , the causative agent of yaws , Haemophilus ducreyi has emerged as a major cause of CU in yaws-endemic regions of the South Pacific islands and Africa . H . ducreyi is generally susceptible to macrolides , but CU strains persist after mass drug administration of azithromycin for yaws or trachoma . H . ducreyi also causes genital ulcers ( GU ) and was thought to be exclusively transmitted by microabrasions that occur during sex . In human volunteers , the GU strain 35000HP does not infect intact skin; wounds are required to initiate infection . These data led to several questions: Are CU strains a new variant of H . ducreyi or did they evolve from GU strains ? Do CU strains contain additional genes that could allow them to infect intact skin ? Are CU strains susceptible to azithromycin ? To address these questions , we performed whole-genome sequencing and antibiotic susceptibility testing of 5 CU strains obtained from Samoa and Vanuatu and 9 archived class I and class II GU strains . Except for single nucleotide polymorphisms , the CU strains were genetically almost identical to the class I strain 35000HP and had no additional genetic content . Phylogenetic analysis showed that class I and class II strains formed two separate clusters and CU strains evolved from class I strains . Class I strains diverged from class II strains ~1 . 95 million years ago ( mya ) and CU strains diverged from the class I strain 35000HP ~0 . 18 mya . CU and GU strains evolved under similar selection pressures . Like 35000HP , the CU strains were highly susceptible to antibiotics , including azithromycin . These data suggest that CU strains are derivatives of class I strains that were not recognized until recently . These findings require confirmation by analysis of CU strains from other regions .
Haemophilus ducreyi classically causes chancroid , a sexually transmitted disease that presents as painful genital ulcers ( GU ) , which are often accompanied by infected regional lymph nodes . Although the current global prevalence of chancroid is undefined due to syndromic management of genital ulcer disease and lack of surveillance programs , the worldwide prevalence of chancroid has declined over the last decade [1] . In addition to causing its own morbidity , chancroid facilitates the acquisition and transmission of the human immunodeficiency virus type 1 [1] . In addition to causing chancroid , H . ducreyi has been isolated from or its DNA has been detected in chronic cutaneous ulcers ( CU ) in yaws-endemic regions in the South Pacific islands and equatorial Africa [2–7] . Yaws is a chronic infection of skin , bone , and cartilage that occurs mainly in poor communities in tropical areas of Africa , Asia , and Latin America; yaws is caused by Treponema pallidum subspecies pertenue , which is closely related to T . pallidum subsp . pallidum , the cause of venereal syphilis . A prospective cohort study by Mitjà and colleagues in yaws-endemic villages of Papua New Guinea showed that H . ducreyi is a major cause of chronic CU in children younger than 15 years old [6] . In that study , nearly 60% of patients with ulcers had detectable lesional H . ducreyi DNA , while only 34% were positive for lesional T . pallidum subsp . pertenue DNA . Approximately 2% of the total population and more than 7% of the children aged 5–15 years had ulcers positive for H . ducreyi as detected by PCR . Similar findings were reported from yaws-endemic communities in the Solomon Islands [8] . Mass drug administration ( MDA ) of oral azithromycin ( AZT ) for yaws in Papua New Guinea with a population coverage rate of 84% reduced the prevalence of CU by 90% [9] . Although MDA significantly reduced the proportion of ulcers with T . pallidum subsp . pertenue DNA , the proportion of ulcers containing H . ducreyi DNA was not affected [9] . The presence of H . ducreyi-positive CU was also reported from districts of Ghana that had received several rounds of MDA of AZT for trachoma [7] . These data raise the possibility that CU strains may be resistant to AZT , exist in an environmental reservoir , or are so infectious that MDA at the above coverage rate fails to eradicate H . ducreyi . Multilocus sequence analysis is frequently used to determine the genetic relatedness of bacterial strains . Based on analysis of 11 H . ducreyi genes , GU strains form two genetically distinct classes , designated class I and class II , which diverged from each other approximately five million years ago ( mya ) and may represent distinct species [10] . A similar analysis including four CU strains suggests that they are a subset of class I GU strains [11] . However , this analysis was limited by the fact that it was based on only three informative loci . To obtain additional insights into the evolutionary relationship of CU and GU strains , here we performed whole-genome sequencing of CU strains isolated from patients infected in Samoa and Vanuatu and archived class I and class II GU strains . Due to the persistence of CU strains after MDA of AZT , we also determined the in vitro susceptibilities of CU and GU strains to antimicrobials used for the treatment of chancroid .
The 5 CU strains used in this study were the only strains available at the time the study was initiated ( Table 1 ) ; their associated clinical features are listed in S1 Table . The class I and class II strains used in this study were chosen because these strains had been previously analyzed by multilocus sequencing ( Table 1 ) [10] . 35000HP , whose genome has been sequenced ( GenBank accession no . NC_002940 . 2 ) , was used as the reference strain in this study; 35000HP was isolated from a volunteer who was experimentally infected on the arm with strain 35000 and has been extensively characterized in human inoculation experiments [12 , 13] . The H . ducreyi strains were grown on Columbia agar plates or in Columbia broth supplemented with 1% bovine hemoglobin ( Sigma-Aldrich ) , 1% IsoVitaleX , and 5% fetal bovine serum ( Hyclone ) at 33°C with 5% CO2 . Genomic DNA was extracted from H . ducreyi strains using the DNeasy Blood & Tissue kit ( Qiagen ) and quantified using the Quant-It High Sensitivity dsDNA Assay kit ( Life Technologies ) . The sequencing libraries were prepared using the NexteraXT DNA Library Preparation kit ( Illumina , Inc . ) following the manufacturer’s instructions . Samples were multiplexed using the NexteraXT Dual Index Primer kit . Equimolar concentrations of indexed libraries were combined into a single pool and were sequenced at the Tufts University Genomics Core Facility . Paired-end 250-bp sequencing was performed on the Illumina MiSeq platform using the MiSeq V2 500 cycles chemistry . The de novo assembly was performed using Edena , with a customized bash script that optimizes the assembly process by optimizing three key Edena parameters [14] . The assembled contigs were annotated using the RAST online annotation tool [15] . A flow chart of comparative genome analysis of CU and GU strains is depicted in S1 Fig . For all comparative genome analyses in this study , the genome sequence of 35000HP was used as the reference . The de novo assembled contigs were ordered into Locally Collinear Blocks ( LCBs ) by Mauve Contig Mover ( MCM ) [16] . The breakpoints between LCBs were resolved by using BLAST analysis of the unaligned contigs produced by MCM , the breakpoint regions in the de novo assembled contigs , and by alignment of raw reads against 35000HP . After resolving the breakpoints , the ordered contigs were concatenated into draft genomes using Emboss 6 . 3 . 1 . Pairwise genome conservation distances , which represent both gene content and sequence similarity , were estimated from draft genomes using ProgressiveMauve and plotted as heat map using CIMminer [17 , 18] . The draft genome sequences for the 14 H . ducreyi strains NZS1 , NZS2 , NZS3 , NZS4 , 82–029362 , 6644 , HD183 , HMC46 , HMC56 , NZV1 , 33921 , CIP542 , DMC64 , and DMC111 were deposited in GenBank under the accession numbers CP011218 , CP011219 , CP011220 , CP011221 , CP011222 , CP011223 , CP011224 , CP011225 , CP011226 , CP011227 , CP011228 , CP011229 , CP011230 , and CP011231 , respectively . Genome rearrangements were identified from multiple alignments of the draft genomes generated by ProgressiveMauve , BLAST Ring Image Generator , and nucleotide BLAST and from the assembly of raw reads against 35000HP by SeqMan NGen [17 , 19 , 20] . Because reference-based alignment can miss additional genes that might be absent in the 35000HP genome , the de novo assembled contigs that did not align to the 35000HP genome by ProgressiveMauve were aligned against other microbial genomes using translated nucleotide BLAST . SNPs and small insertions and deletions ( indels , <10 bp ) were detected using DNASTAR Lasergene ( DNASTAR , Inc . , Madison , WI ) . Briefly , the sequenced reads were assembled by SeqMan NGen against 35000HP . SNPs and indels were discovered by Seqman Pro using default parameters except that a minimum frequency of 90% reads and a minimum coverage of 50 reads were used for the analysis . SNPs were grouped as non-coding , synonymous , or nonsynonymous . Nonsynonymous SNPs were further categorized as substitutions , no-start , no-stop , nonsense , or frameshifts . All SNPs in the genomes of the CU strains were manually verified for accuracy . Diversity analyses of whole-genome nucleotide sequences and translated concatenated coding sequences were performed using Mega 6 . 0 [21] . The reliability of the diversity analyses was tested using 1000 bootstrap replicates . Recombination analysis was performed using the Phi test implemented in PhiPack and the likelihood ratio test implemented in TOPALi v2 [22 , 23] . For both tests , a threshold P < 0 . 05 was used to define a recombination event . Phylogenetic analyses were performed using Mega 6 . 0 and Realphy [21 , 24] . Briefly , whole-genome alignments were imported into Mega 6 . 0 and subjected to model testing to identify the best-fit models of nucleotide substitution . Model testing identified Hasegawa-Kishino-Yano plus invariant sites plus gamma-distributed model as the best-fit nucleotide substitution model for our data . Using the best-fit model , phylogenetic analyses were performed with both whole-genome alignments and alignments of translated amino acid sequences from concatenated protein-coding regions using different methods of phylogeny reconstruction , including Maximum Likelihood , Maximum Parsimony , Minimum Evolution , and Neighbor Joining with different gap treatment approaches . We also inferred phylogenies using Realphy , which generates phylogenetic trees by merging alignments obtained by mapping to multiple reference genomes . A rooted Maximum Likelihood tree was reconstructed by including other Pasteurellaceae members ( Actinobacillus pleuropneumoniae , Mannheimia haemolytica , Pasteurella multocida , Aggregatibacter actinomycetemcomitans , and Haemophilus influenzae ) as outgroups . The reliability of all the trees generated was verified by 1000 bootstrap replicates . The times to the most recent common ancestor ( MRCA ) were estimated by Bayesian molecular clock method using Beast v1 . 8 . 1 [25] . Hasegawa-Kishino-Yano plus invariant sites plus gamma-distributed model and a relaxed clock model were used to account for variation in substitution rates . The results from the Beast analysis were visualized using Tracer v1 . 6 . A best-fit tree was identified from the tree data generated by Beast using TreeAnnotator and visualized using FigTree v1 . 4 . 2 . As described previously , we used a substitution rate of 4 . 5 × 10−9 per site per year to calibrate the tree [26] . Selection analyses were performed using Mega 6 . 0 and Hyphy 2 . 1 [21 , 27] . Briefly , protein-coding regions were extracted from the annotated genomes , ordered against 35000HP using MCM , concatenated using Emboss 6 . 3 . 1 , and aligned using ProgressiveMauve [16 , 17 , 28] . The alignments were manually edited for accuracy to obtain a codon-delimited alignment , which was used for all the selection analyses . Rates of nonsynonymous ( dN ) and synonymous ( dS ) substitutions are widely used as a sensitive measure of selection occurring in a protein with dN = dS , dN > dS , and dN < dS indicating neutral , positive , and negative selection , respectively . Alignment-wide evidence for selection was tested using the codon-based Z test . For the codon-based Z test , we first calculated dN and dS and their variances using 1000 bootstrap replicates . We then used this information to test the null hypothesis of neutrality ( dN = dS ) versus alternative hypothesis of positive ( dN > dS ) or negative ( dN < dS ) selection using a Z-test . A branch-site random effects likelihood test was used to test whether any of the branches in the tree are evolving under positive selection . A branchTestDNDS test was performed to test whether a prespecified branch of the tree is evolving under different selection strength than the rest of the tree [27] . Individual sites under positive or negative selection were identified using the single likelihood ancestral counting and fixed effects likelihood methods [27] . The draft genomes were interrogated for the presence of genes that are required for the virulence of strain 35000HP in the human inoculation experiments , using nucleotide BLAST [13] . For identifying sequence variation , the nucleotide sequences of virulence genes were translated into amino acids and the translated sequences were aligned using Clustal Omega [29] . The draft genomes were searched for the presence of known antimicrobial resistance genes using ResFinder with default parameters [30] . AST was performed using the agar dilution method as described previously with some modifications [31–33] . Briefly , H . ducreyi strains were grown on Columbia agar ( Difco ) containing 1% hemoglobin ( BBL ) , 0 . 2% activated charcoal ( Sigma-Aldrich ) , 5% fetal bovine serum ( Atlanta Biologicals ) , and 1% IsoVitaleX ( BBL ) for 48 h at 33°C under microaerophilic conditions . The colonies were suspended into Mueller-Hinton ( BBL ) broth containing 1% IsoVitaleX and 0 . 002% Tween-80 ( Sigma-Aldrich ) , passed through a 22-gauge needle and left at room temperature for 15 min . The optical density of the culture was adjusted to that of a 0 . 5 McFarland standard using a Spectronic 20 Plus spectrophotometer ( Milton Roy ) . AST was performed on Mueller-Hinton II medium ( BBL ) containing 33% lysed horse blood ( Remel ) , 5% fetal bovine serum , and 1% IsoVitaleX . The following antibiotics were tested: amoxicillin ( AMX ) , amoxicillin/clavulanic acid ( AMC; 2:1 ) , azithromycin ( AZT ) , ciprofloxacin ( CIP ) , ceftriaxone ( CRO ) , doxycycline ( DOX ) , erythromycin ( ERY ) , and penicillin ( PEN; all from Sigma-Aldrich ) . The H . ducreyi strains CIP542 , 35000HP , and the H . influenzae strain 49247 were used as controls . A 104/ml suspension of each strain was delivered onto each plate with a Steer’s Replicator ( CMI-Promex , Inc . ) , and the plates were dried for 15 min at room temperature . The minimal inhibitory concentrations were recorded after incubating the plates for 48 h at 33°C under microaerophilic conditions . The presence of three or fewer colonies was recorded as no growth .
Whole-genome sequencing generated between 0 . 5 and 4 million reads for each of the 14 strains ( Table 1 ) . The estimated genome sizes ranged from 1 . 52 Mb to 1 . 74 Mb , with an average GC content of 37 . 8% to 38 . 6% ( Table 1 ) . The total number of contigs for each strain ranged from 40 to 129 ( Table 1 ) . The estimated average genome coverage ranged from 92 to 596 fold ( Table 1 ) . Contig ordering generated 2 to 6 LCBs for the CU strains , 5–15 LCBs for the class I strains , and 12–16 LCBs for the class II strains . Inspection of the LCBs revealed that the majority of the putative breakpoints between LCBs occurred in genes that share high homology with other genes in the H . ducreyi genome such as lspA1 and lspA2 , genes encoding rRNAs , and bacteriophage-related genes and that the majority of the breakpoints did not contain any rearrangements . Analysis of pairwise genome conservation distance of the draft genomes showed that CU strains form a subcluster within class I strains and that class II strains form a separate cluster from CU and class I strains ( Fig 1 ) . Compared to 35000HP , all the CU strains consistently contained ~20-kb deletion ( HD1528 to HD1565 ) in a bacteriophage locus that is homologous to Pseudomonas aeruginosa bacteriophage B3 and five small deletions that ranged in size from 30–767 bp ( Fig 2 and S2 Table ) . The class I strain HMC56 contained a 50-kb deletion ( HD0897 to tRNA-Lys-1 ) in a region homologous to the H . influenzae ICEHin1056 integrative conjugative element ( Fig 2 and S3 Table ) . All the class II strains contained 3 major deletions of 37 kb ( HD0087 to HD0161 ) , 35 kb ( HD0478 to HD0495 ) and 50 kb ( HD0897 to tRNA-Lys-1 ) , which are homologous to Escherichia coli bacteriophage D108 , Haemophilus bacteriophage SuMu , and H . influenzae ICEHin1056 , respectively ( Fig 2 and S4 Table ) . The class II strains also contained several deletions ( between HD1528 and HD1618 ) in a region that is homologous to P . aeruginosa bacteriophage B3 ( Fig 2 and S4 Table ) . All the GU strains also contained several other small deletions as listed in S3 and S4 Tables . Compared to 35000HP , we did not find any inversions in CU strains with the exception of NZV1 , which contained an inversion of ~428 kb that spanned from HD0054 ( tuf ) to HD0659 ( S2 Fig ) . Among the class I strains , HMC56 contained an inversion of ~300 kb that spanned from glpA ( HD1157 ) to lspA1 ( HD1505 ) ( S2 Fig ) . HD183 contained a ~161 kb inversion that spanned from hhdA ( HD1327 ) to lspA1 ( HD1505 ) ( S2 Fig ) . All the class II strains contained an inversion of ~17 kb that spanned from HD1532 to HD1565 ( S2 Fig ) . However , BLAST analysis of the inversion breakpoints showed no major changes in their genetic content . Compared to 35000HP , the CU strains , the class I strain 82–029362 , and the class II strain CIP542 contained no additional genes in their genomes . All the remaining class I and class II strains contained several additional genes as listed in S5 Table . To get a deeper understanding of the relationship of CU strains to GU strains , we next performed whole-genome SNP analysis using 35000HP as a reference . CU strains differed from 35000HP by ~400 SNPs ( Table 2 ) . The class I strain HD183 differed from 35000HP by ~160 SNPs , while all other class I strains differed by ~2 , 000 SNPs ( Table 2 ) . The class II strains differed from 35000HP by ~30 , 000 SNPs ( Table 2 ) . Analysis of within lineage genetic diversity showed that CU strains had the least nucleotide and amino acid divergence followed by class I and class II strains ( Table 3 ) . Analysis of interlineage diversity showed that there was little divergence between CU and class I strains; however , a greater amount of divergence was observed between CU and class II strains ( Table 3 ) . Interlineage diversity analysis showed that there was high divergence between class I and class II strains ( Table 3 ) . While the Phi test showed no evidence of recombination ( P = 0 . 44 ) , the likelihood ratio test identified five putative recombination events in CU and GU strains ( S6 Table ) . Removal of recombination regions had no major effect on the overall topology of the phylogenetic tree described in the following section , except for minor differences in bootstrap values and positioning of individual species within class clades ( S3 Fig ) . In general , all methods showed that class I and class II strains formed two separate phylogenetic clusters and that CU strains formed a subcluster within the class I clade with minor differences in bootstrap values and positioning of individual species within class clades . A rooted tree generated by the Maximum Likelihood method and Pasteurellaceae members as outgroups was used as the final tree ( Fig 3 ) . To determine the approximate time to the MRCA of the CU strains , we performed a molecular clock analysis using the Bayesian method and the mutation rates proposed by Ochman et al . for calibration [25 , 26] . The divergence time of the CU strains from the MRCA of the class I strains 35000HP and HD183 was estimated as 180 , 000 years ago ( Fig 4 ) . The divergence time of the CU strains , 35000HP , and HD183 from the MRCA of other class I strains was estimated as 450 , 000 years ago ( Fig 4 ) . The divergence time of class I strains from the MRCA of class II strains was estimated as 1 . 95 mya ( Fig 4 ) . Molecular clock analysis also showed that the CU strains began to diversify from each other around 27 , 000 years ago ( Fig 4 ) . Thus , CU strains appear to have recently diverged from class I GU strains . Pairwise analysis of rates of nonsynonymous ( dN ) and synonymous ( dS ) substitutions and their variances showed that the Z-test rejected the null hypothesis of neutrality ( dN = dS ) in favor of the alternative hypothesis of negative selection ( dN < dS ) ( Table 4 ) . The dN-dS value averaging over all sequence pairs was -75 . 55 ( P = 0 . 0000000001 ) . Utilizing the rates of nonsynonymous ( dN ) and synonymous ( dS ) substitutions , we also calculated the overall mean and pairwise mean dN/dS ratios; the overall mean dN/dS ratio for all genomes was 0 . 31 and the pairwise mean dN/dS ratios for most comparisons were less than 1 ( Table 4 ) . The pairwise mean dN/dS ratio between the CU and GU lineages was 0 . 35 , between CU and class I lineages was 0 . 38 , and between CU and class II lineages was 0 . 33 . Consistent with these analyses , the single likelihood ancestral counting and the fixed effects likelihood analyses identified 141 and 132 negatively selected sites , respectively . To determine whether CU strains evolved under different selection strength than GU strains , we performed a TestBranchDNDS analysis . This analysis showed that the strength of selection in CU strains was not significantly different than in GU strains ( likelihood ratio difference = 3 . 8; P = 0 . 58 ) . We determined whether the genomes of CU strains contained the genes that are required for the virulence of strain 35000HP in the human challenge model of infection and whether there were variations in these virulence determinants compared to GU strains [13] . BLAST analysis showed that all the CU and GU strains contained all of the genes known to be required for virulence in the human challenge model ( S7 Table ) . Alignment of amino acid sequences of the virulence determinants showed that the DsrA , LspA1 , and LspA2 proteins of the CU strains differed by at least 1 amino acid from class I strains ( S7 Table ) . To determine whether CU strains were resistant to clinically relevant antimicrobials , we performed AST using the agar dilution method . The CU strains from Samoa and Vanuatu were AZT susceptible , and had similar susceptibility patterns as the type strains 35000HP and CIP542 ( Table 5 ) . With the exception of 82–029362 and 35000HP , all the class I strains were resistant to penicillin ( MIC , >256 μg/ml ) , amoxicillin ( MIC , 64–256 μg/ml ) , and doxycycline ( MIC , 8–16 μg/ml ) ( Table 5 ) . With the exception of CIP542 , all class II strains were resistant to amoxicillin and penicillin ( MIC , 128–256 μg/ml ) ( Table 5 ) . The class II strains 33921 and DMC64 were also resistant to doxycycline ( MIC , 8–16 μg/ml ) ( Table 5 ) . All the strains were susceptible to ciprofloxacin , azithromycin , erythromycin and ceftriaxone ( Table 5 ) . Consistent with their susceptibility to clinically relevant antimicrobials , the CU strains contained no horizontally acquired genes encoding antimicrobial resistance determinants in their genomes . Consistent with their resistance to penicillin/amoxicillin and doxycycline , the genomes of GU strains contained genes that confer resistance to penicillin/amoxicillin ( blaTEM-1B ) and doxycycline [tet ( B ) , tet ( 32 ) or tet ( M ) ] ( Table 5 ) .
H . ducreyi was previously thought to exclusively cause the sexually transmitted disease chancroid but has emerged as a major cause of the nonsexually transmitted CU in children in yaws-endemic regions of South Pacific islands and equatorial Africa . Here , we performed whole-genome sequencing of a limited number of CU strains and compared them to class I and class II GU strains . Comparative genome analyses showed that the CU strains are remarkably similar to class I strains . Phylogenetic analyses showed that the CU strains evolved from class I GU strains . Analysis of genome conservation of CU and GU strains showed that CU strains had 98–99% similarity to each other , 94–98% similarity to class I strains , and 81–92% similarity to class II strains . Kunin et al . , estimated genome conservation within different bacterial taxonomic ranks and found that strains within most bacterial species have a genome conservation of approximately 87% ( range , 73–101% ) [34] . Thus , the H . ducreyi genome conservation values are well within the range of those of other bacterial species . Genome conservation analysis also showed that CU strains form a subcluster within class I GU strains and that class II strains form a distinct cluster from class I and CU strains . These findings are in good agreement with the results of the whole-genome phylogenetic analysis as well as with previous multilocus sequence-based phylogenetic analysis [11] . Consistent with the genome conservation data , analysis of whole-genome genetic diversity also showed that there was smallest amount of genetic diversity within the CU strains ( dnucleotide = 0 . 000013 ) , little genetic diversity between CU and class I strains ( dnucleotide = 0 . 00012 ) , and a greater amount of genetic diversity between CU and class II strains ( dnucleotide = 0 . 0098 ) and class I and class II strains ( dnucleotide = 0 . 01 ) . Cejcova et al . , reported a whole-genome nucleotide diversity of 0 . 00033 for strains within T . pallidum subsp . pallidum and of 0 . 00032 for strains within T . pallidum subsp . pertenue [35] . Thus , the H . ducreyi genetic diversity values for CU and Class I strains are similar to those of the two Treponema species that inhabit similar ecological niches as H . ducreyi , while the diversity values between CU and class II strains and class I and class II strains are higher than those of the two Treponema species . Our study estimated that class I strains diverged from class II strains 1 . 95 mya and that CU strains diverged from class I strains 0 . 18 mya . Previous studies estimated that class I strains diverged from class II strains 5 mya and that CU strains diverged from class I strains 0 . 355 mya [10 , 11] . In our study , divergence times were estimated using entire genomes . The previous studies used only 11 H . ducreyi loci , all of which were selected to contain variant alleles to allow for epidemiological typing . Since a large proportion of genes in the genome do not contain variant alleles , averaging the variance over the entire genome would result in relatively lower divergence times than those estimated in previous studies . Although CU strains lacked additional genetic material compared to 35000HP , CU strains differed from 35000HP by ~400 SNPs . Nearly 40% of these SNPs were nonsynonymous and 25% were located in noncoding regions of the genome . Previous studies have shown that SNPs can have a profound impact on global gene expression in bacteria [36] . Whether SNPs in the CU strains would result in a different global gene expression pattern than 35000HP requires additional investigation . A large number of SNPs in the CU strains were located in 21 genes that individually or in combination are required for H . ducreyi infection in human volunteers . CU strains differed from class I strains by at least one amino acid in 3 virulence determinants , specifically DsrA , LspA1 , and LspA2 . DsrA is a surface protein and LspA1 and LspA2 are secreted proteins; all three are required for evasion of immune defenses [37 , 38] . Thus , the variations of these proteins in CU strains are likely an effect of host immune pressure . Consistent with the fact that class I strains differ from class II strains in several of the known virulence determinants and our finding that CU strains formed a subcluster under class I strains , CU strains differed from class II strains by at least one amino acid in 19 of the 21 virulence determinants [10] . In agreement with a previous study , compared to class I strains , the nucleotide sequences of DsrA and NcaA of class II strains also contained several short rearrangements including deletions and insertions [10] . Analysis of rates of nonsynonymous substitutions ( dN ) and synonymous substitutions ( dS ) in the CU and GU genomes showed that synonymous substitutions were found at a higher rate than nonsynonymous substitutions with an overall mean dN/dS ratio for all strains of 0 . 31 . Similarly , the pairwise mean dN/dS ratio between CU and GU lineages was 0 . 35 . These data suggest that CU and GU strains evolve under negative selection . Other sexually transmitted bacterial pathogens such as Neisseria gonorrhoeae and Chlamydia trachomatis also evolve under negative selection , with overall mean dN/dS ratios of 0 . 3184 and 0 . 4021 , respectively [39 , 40] . These findings are in agreement with the neutral theory of molecular evolution , which postulates that selective fixation of neutral mutations by genetic drift is the major determinant behind species divergence [41] . Our data also showed that both CU and GU strains evolve under similar selection strength , which may be due to the similar immunological pressures that these strains encounter in their respective ecological niches of human skin versus mucosal surfaces and human skin . Using SNPs , molecular dating analysis indicates that the CU strains began to diversify from each other ~27 , 000 years ago . The CU clade is characterized by several shared , derived deletions of defined lengths ( synapomorphies ) , which were most likely inherited from the common ancestor of modern CU strains . Given that these deletions were absent in all the class I GU strains including 35000HP and HD183 , we speculate that the Samoan/Vanuatu CU lineage may have existed for at least 27 , 000 years . Mitjà and colleagues hypothesized that syndromic management of genital ulcers in the South Pacific may have forced H . ducreyi into a new niche of cutaneous ulcers in children [6] . Syndromic management of GU in the South Pacific was introduced in 2002 , while CU due to H . ducreyi was first reported in 1989 [6] . The fact that the CU strains diverged from GU strains ~180 , 000 years ago and from each other ~27 , 000 years ago supports the idea that cutaneous infection with H . ducreyi preceded syndromic management of GU . A possible explanation why H . ducreyi was not recognized as a cause of CU previously is that CU in the South Pacific has traditionally been empirically treated with penicillin [42] . As CU strains are susceptible to penicillin , CU due to H . ducreyi would have responded to empirical treatment . The current World Health Organization case definition of yaws includes a patient with a chronic atraumatic skin ulcer and seropositivity for T . pallidum subsp . pertenue . In the cross sectional survey in Papua New Guinea , a reasonable proportion of children with detectable H . ducreyi DNA in ulcers were also seropositive for T . pallidum subsp . pertenue [6] and therefore would be classified as having yaws . This could account for the lack of earlier recognition of H . ducreyi as a source of CU . Although penicillin had been the cornerstone of yaws eradication efforts for the last several decades , MDA of AZT is the mainstay of the World Health Organization’s new program for the eradication of yaws [42] . MDA was given to 84% of the villagers who were studied in Papua New Guinea [9] . At 12-months follow-up , MDA reduced the prevalence of CU by 90% [9] . In those who had ulcers at follow-up , there was a significant reduction in the proportion of ulcers with T . pallidum subsp . pertenue DNA [9] . However , the proportion of ulcers containing H . ducreyi DNA was unchanged relative to the baseline level of 60% [9] . The CU strains from Samoa and Vanuatu were as susceptible to AZT as 35000HP . Whether CU strains from Papua New Guinea are susceptible to AZT is not known . If they are susceptible , their persistence after MDA suggests that CU strains may have a higher level of infectivity than T . pallidum subsp . pertenue or may be present in an environmental reservoir . Inoculation of the upper arm of human volunteers with the GU strain 35000HP produces an infection that is clinically and histopathologically nearly identical to natural chancroid [13 , 43 , 44] . Evolutionary analyses showed that CU strains are closely related to 35000HP . Similar to 35000HP , CU strains are capable of infecting nongenital skin . Our data showed that CU strains evolve under selection strength similar to that of GU strains . Due to lack of biopsy specimens , we do not know whether the histopathology of a CU lesion is similar to that of an experimental lesion caused by 35000HP or natural chancroid . Nevertheless , these data suggest that H . ducreyi likely encounters similar host pressures in the genital and nongenital skin . Placement of 106 CFU of 35000HP on intact skin does not cause disease in human volunteers; but as few as one bacterium delivered by a puncture wound causes infection [13] . These data raises the possibility that either wounds are required for CU strains to initiate infection or that CU strains possess additional genes that allow them to penetrate intact skin . Our data showed that the CU strains did not contain additional genetic elements , suggesting that CU strains likely use wounds to initiate infection . In Papua New Guinea , up to 7% of children have CU with detectable H . ducreyi DNA [6]; it is difficult to imagine that wound to wound transmission is responsible for this astoundingly high prevalence . In the Papua New Guinea study , many children infected with H . ducreyi were seropositive for T . pallidum subsp . pertenue and some ulcers contained both H . ducreyi and T . pallidum subsp . pertenue DNA [6] . Thus , T . pallidum subsp . pertenue may serve as an instigating pathogen while H . ducreyi superinfects yaws lesions . Photographs of typical CU lesions show that flies frequently land on ulcers [6] . Thus , it is possible that CU strains are transmitted from person to person by direct contact of wounds with infected lesions , or by vectors such as flies . In a randomized controlled clinical trial , treatment with 1 gram of AZT prevented experimental infection of adult volunteers with 35000HP for nearly two months [45] . Given that a 2-gram dose of AZT is being used to eradicate yaws , MDA may provide treatment and prophylaxis against CU strains for a similar period of time . These data also suggest that repetition of MDA on a bimonthly basis and/or higher coverage rates may contribute to successful eradication of CU strains from yaws-endemic areas . By PCR-based testing , 2% of commercial sex workers in a chancroid endemic region are asymptomatically colonized in the cervico-vaginal tract with H . ducreyi [46] . Whether CU strains asymptomatically colonize the skin of humans living in the tropics is unknown , but colonization would provide a source of bacteria that could enter wounds . As AZT is concentrated intracellularly especially in fibroblasts [45] , colonization of the skin surface could allow CU strains to escape AZT treatment . Our study has several limitations . We only reported draft genomes , and the genetic variation among the strains was not confirmed by PCR and sequencing . Our study involved a small number of GU strains , with limited clinical and epidemiological data . Our analysis only included CU strains that were acquired in Samoa and Vanuatu; our findings should not be extrapolated to CU strains from other regions . All strains used in this study were obtained following culture and storage; their sequences could have been affected by these factors over time . Finally , the CU strains were not compared to contemporaneous GU strains from the same or other regions; to our knowledge and due to syndromic management , few such GU strains exist . This was the first study using comparative genomics to examine a small number of cultured H . ducreyi strains isolated from CU and GU . Our findings show that CU strains are derivatives of class I GU strains whose lineage may be 27 , 000 years old . Further studies are needed to determine the phylogeny of CU strains from other endemic areas , such as Papua New Guinea , Ghana , and the Solomon Islands , and to examine strains that persist after MDA of azithromycin . Flies and nonhuman primates are thought to serve as reservoirs for T . pallidum subsp . pertenue [6 , 47]; it would be interesting to determine whether they serve as reservoirs for CU strains or whether humans who reside in endemic areas are colonized with H . ducreyi .
|
Cutaneous ulcers ( CU ) in children living in equatorial Africa and the South Pacific islands have long been attributed to yaws , which is caused by Treponema pallidum subsp . pertenue . However , PCR-based cross sectional surveys done in yaws-endemic regions show that Haemophilus ducreyi is the leading cause of CU in these regions . H . ducreyi classically causes the genital ulcer ( GU ) disease chancroid and was once thought to be exclusively sexually transmitted . We show that CU strains obtained from Samoa and Vanuatu are genetically nearly identical to class 1 GU strains and contain no additional genetic content . The CU strains are highly susceptible to antibiotics , including azithromycin . The data suggest an urgent need to obtain and analyze CU isolates from Africa and other countries in the South Pacific and to search for environmental sources of the organism .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Haemophilus ducreyi Cutaneous Ulcer Strains Are Nearly Identical to Class I Genital Ulcer Strains
|
Systematic , genome-wide loss-of-function experiments can be used to identify host factors that directly or indirectly facilitate or inhibit the replication of a virus in a host cell . We present an approach that combines an integer linear program and a diffusion kernel method to infer the pathways through which those host factors modulate viral replication . The inputs to the method are a set of viral phenotypes observed in single-host-gene mutants and a background network consisting of a variety of host intracellular interactions . The output is an ensemble of subnetworks that provides a consistent explanation for the measured phenotypes , predicts which unassayed host factors modulate the virus , and predicts which host factors are the most direct interfaces with the virus . We infer host-virus interaction subnetworks using data from experiments screening the yeast genome for genes modulating the replication of two RNA viruses . Because a gold-standard network is unavailable , we assess the predicted subnetworks using both computational and qualitative analyses . We conduct a cross-validation experiment in which we predict whether held-aside test genes have an effect on viral replication . Our approach is able to make high-confidence predictions more accurately than several baselines , and about as well as the best baseline , which does not infer mechanistic pathways . We also examine two kinds of predictions made by our method: which host factors are nearest to a direct interaction with a viral component , and which unassayed host genes are likely to be involved in viral replication . Multiple predictions are supported by recent independent experimental data , or are components or functional partners of confirmed relevant complexes or pathways . Integer program code , background network data , and inferred host-virus subnetworks are available at http://www . biostat . wisc . edu/~craven/chasman_host_virus/ .
Our work is related to methods that address several different categories of problems: finding mechanistic explanations for source-target pairs , subnetwork extraction , candidate gene prioritization , and gene set enrichment . One closely related task is to infer the physical interactions that mediate the observed direct or indirect relationships between a source gene and a target gene . The input to these methods is a set of source-target pairs and a background network consisting of unsigned protein-protein and/or protein-DNA interactions . The output is a subnetwork that provides a connection between each source and target . Most closely related to our work are the approaches that globally infer a subnetwork to account for all given pairs by providing paths between them . The Markov network-based Physical Network Model [11] and the integer programming-based SPINE [12] both infer subnetworks in which each source must be connected to its targets by one or more acyclic pathways , and in which the sign of each edge is also inferred . The Physical Network Model also infers directions for edges . Related methods for signaling network orientation [13]–[15] infer edge directions , but not edge signs or node phenotypes . Yosef et al . [16] infer rooted trees that connect a set of sources with a set of targets . Additionally , some methods account for source-target pairs separately , rather than in a global inferred subnetwork [17] , [18] . Others employ genetic interactions or correlation of mRNA expression in addition to protein-protein interactions to infer indirect or direct relationships between genes [19] , [20] . Our work has similarities to these approaches , particularly those based on integer linear programming , but differs in some key respects . In our setting , the common target of all hits – the virus – is external to the background network , and the identity of the host factors that interact with it directly must be predicted . Additionally , our background network encompasses a greater variety of biological interactions than the background networks used by these other approaches . Unlike the methods that use mRNA expression profiles as the basis for determining direct or indirect relationships between genes , ours uses only phenotypes derived from a genome-wide mutant assay . Recently , Gitter et al . [21] presented an application of their source-target pair-based method to inferring human signaling pathways involved in influenza A viral infection . In their approach , sources are human proteins that are known to directly interact with a viral component , analogous to the interfaces in our conceptual model . Targets are human genes whose expression is measured over several time points during viral infection . The method orients paths through a protein-protein interaction network from the sources to the targets , preferring paths that contain influenza-relevant genes identified by RNAi experiments . Conceptually , this method infers the signaling pathways that control the host's transcriptional response to viral infection . In this paper , we look at host-virus interactions from the opposite direction and infer the mechanistic pathways by which suppressed host genes inhibit or enable the normal viral replication cycle . Other related methods address the network extraction task: selecting specific types of connecting structures from a background network when a biologically-motivated node- and/or edge-weighting function is available . The structures include rooted trees [22] , variants on Steiner trees [23]–[26] , random walks and short paths [27] , parallel pathways [28] , dense highly-connected subnetworks [29] , and undirected subnetworks that provide connections between pairs of genes [30] . Unlike our method , these approaches do not distinguish ( or infer ) phenotype signs and edge signs , nor do they apply global constraints to the extracted subnetwork other than a global edge minimization . In contrast , we employ global constraints such as an upper bound on the number of interfaces . We do not believe that for our task it is appropriate to assume the entire network will be minimal , which is an assumption made by the Steiner tree and shortest-paths methods . Other methods apply graph kernels or flow algorithms to an interaction network to predict and prioritize additional hit genes [31]–[34] . Notably , Murali et al . [33] predict which genes modulate HIV replication in human cell lines . Like these methods , our approach uses a gene ranking method to prioritize genes for inclusion in the inferred subnetwork . However , these methods themselves do not infer consistent , directed pathways , nor do they predict which host factors directly interact with the virus . Our approach combines a gene prioritization method with a directed network inference method . More distantly related to our work , gene set enrichment techniques are widely used to interpret hit sets identified by high-throughput experiments . These methods identify which pre-defined biological components and processes , such as Gene Ontology annotations or KEGG pathways [35] , [36] , are represented in a set of genes [37] . In contrast , our method does not restrict our pool of candidate genes and interactions to predefined gene sets . Additionally , gene set enrichment-based methods are typically better suited when the task is to identify common annotations within a gene set , rather than to predict a set of high-precision additional hits or relevant mechanistic interactions among known hits .
The input to our approach consists of a set of viral phenotypes observed in a loss-of-function experiment and a background network of intracellular interactions . When available , we can also take advantage of confirmed relevant interactions curated from the literature . We have developed an integer-linear-programming-based approach to infer a directed subnetwork of interactions that are relevant to virus replication in a host cell . The approach infers subnetworks that have the following properties:
We first describe an experiment in which we assess the accuracy of our approach in predicting whether test genes with held-aside phenotypes are hits or not . We refer to this as the hit-prediction task . Previously , diffusion kernel methods like the one we use in our objective function have been successfully applied to this task , which is also called gene prioritization [31] , [33] . Using a leave-one-out methodology , we hold aside the measured phenotype for one gene at a time . The set of genes that are held-aside as test cases for the BMV data set includes 104 hits ( 49 up and 55 down ) and 1074 no-effect genes . The test set for the FHV data set comprises 55 hits ( 48 up and 7 down ) and 991 no-effect genes . We do not test weak-phenotype genes in this evaluation . When a given gene is held aside , it is treated as if its phenotype has the unobserved label , meaning that the inference process is used to predict whether or not the gene is relevant , and , if it is predicted to be relevant , its phenotype label . If the test case is included in the set of literature-curated interactions , then all interactions that involve the test case are held aside as well . We also recalculate the diffusion kernel scores for the entire network for each held-aside test case . To predict the label of a held-aside node , we use our integer programming approach to infer an ensemble of subnetworks . An individual subnetwork may include the held-aside gene and provide a predicted up or down phenotype for it , or it may exclude the gene . We assess our confidence in whether the gene is a hit or not by determining the fraction of subnetworks in which it is predicted to have an up or down phenotype . When this fraction is the same for a set of cases , the node scores computed by the kernel are used as a secondary measure of confidence . By varying a threshold on these confidence values , we can plot a precision-recall curve characterizing the predictive accuracy of our method . Recall is defined as the fraction of true hits in the test set that are predicted to be hits , and precision is defined as the fraction of predicted hits that are truly hits . In this context , we consider precision to be the more important of the two measures , as it is better to to avoid devoting follow-up experiments to false positives . As discussed in the Related Work section , several integer programming methods have been developed to infer signalling and regulatory networks from experimental data that comes in the form of source-target pairs . A key aspect of our approach is that it does not assume that targets are given . Instead , it infers the downstream interfaces . Existing IP approaches are therefore not directly applicable to our own task . However , we consider some components of existing methods that can be substituted into our integer program: namely , two alternative objective functions , and one alternative heuristic for inferring edge signs . Additionally , in this section , we explore the effect of varying some of the previously discussed parameters and constraints of our IP . One motivation for our inference approach is to make predictions about which unassayed host factors may be involved in viral replication . A number of host factors were unable to be assayed using the deletion or doxycycline-repressible mutant libraries , either because the mutant was not part of the library or did not grow under experimental conditions . As these factors cannot be assayed using high-throughput screens , there is a need to identify a high-confidence subset of them for further , lower-throughput experimentation . Toward this end , we use our approach to make predictions about host factors that were not assayed ( or not successfully assayed ) in the genome-wide BMV screens [1] , [4] . We collect an ensemble of 100 inferred subnetworks using all available phenotype data and allowing the use of 97 interfaces . We choose this number of interfaces because it is in the middle of the range tested in our cross-validation experiments , the results of which suggest that prediction accuracy is not significantly affected by a larger number of interfaces . We also seed the network with the literature-curated edges described in the Data section . Out of 1 , 821 unassayed host factors in the background network , 221 are predicted to be relevant by any of the 100 inferred subnetworks in the ensemble , and 189 receive confidence . Of these , 124 represent ORFs ( about 9% of unassayed ORFs/putative ORFs ) , and 65 represent protein complexes ( about 14% of represented protein complexes ) . Here we discuss independent evidence supporting a selection of these predictions . In numerous cases , the predicted hits include members of pathways of protein complexes known to be involved in BMV replication . In these cases , the inferred subnetworks correctly expanded the relevant complexes with other known components or functional partners that were absent from the given hit sets for technical reasons , such as non-viability of the relevant mutant strain . One example is the inferred inclusion of previously un-implicated components of the cellular ubiquitin-proteasome system , such as the 20S proteasome and components of the 19S regulator complex . While some experimental and literature-curated hits are associated with the proteasome , the predicted hits contribute several more proteasomal proteins . Recent additional experiments , including inhibitor studies and other approaches , have confirmed the involvement of the 20S proteasome , the 19S regulator and other factors in this system in multiple aspects of BMV RNA replication ( B . G . and P . A . , unpublished results ) . Even more important biological validation of our results emerged from additional experimental studies . For example , our ensemble predicts the involvement of Snf7p and Vps4p , both at 0 . 99 confidence . These are proteins in the ESCRT pathway , which is involved in membrane bending and scission events in cell division , cell surface receptor down-regulation and other processes [53] . Recent studies initiated independently of the work reported here have confirmed the predicted role of ESCRT pathway , and of Snf7p and Vps4p in particular , in facilitating BMV RNA replication ( A . Diaz , X . Wang and P . Ahlquist , manuscript in preparation ) . The predicted relevant interactions involving Snf7p and Vps4p are shown in Figure 8 . A further example is provided by the inferred involvement of Xrn1p , a protein involved in RNA degradation . An independent study confirmed the strong impact of the gene XRN1 on BMV replication by showing that a BMV mutant defective in modifying BMV RNA's by the addition of a 5' cap could not accumulate RNA in wild type yeast but did so in an xrn1 deletion strain [54] . The subnetworks inferred using our method can be used to predict which host factors are closest to a direct interaction with the virus . For this evaluation , we predict a set of high-confidence host interfaces for BMV . The ability of our methods to predict physical interfaces between host protein networks and viral components is constrained by the limits of current background knowledge , as specifically represented by the input background network of interacting host proteins . Because of such external limitations , some predicted interfaces may not represent actual host-virus interfaces , but instead approximate the host component that would most likely connect with an actual interface if the relevant subnetwork were extended to include currently unrecognized interaction partners . We consider support from domain knowledge that the predicted interfaces are plausibly close to a direct interaction with a viral component . To predict high-confidence interfaces , we infer an ensemble of 100 subnetworks for BMV-yeast interactions , applying the global constraint that only the minimum possible number of interfaces can be used ( that is , the smallest number of interfaces such that the IP remains feasible; in this case , 47 ) . We also seed the network with the literature-curated edges described previously , which include four interfaces . Over the entire ensemble , the total number of interfaces used by at least one subnetwork is 51 . We designate as “high-confidence” those interfaces that ( i ) account for more than one hit ( other than themselves ) , ( ii ) have greater than 0 . 75 confidence , and , ( iii ) are predicted to be an interface with an average of at least 0 . 75 confidence across all of the leave-one-out ensembles inferred using a minimum number of interfaces . Our method predicts 14 novel high-confidence yeast interfaces for BMV , as shown in Table 6 . We assessed these high-confidence interfaces for plausibility based on their annotated function in the Saccharomyces Genome Database [55] . The value of our subnetwork inference method is supported by the observation that several of the predicted 14 high-confidence interfaces are known interactors with BMV components and many more are closely associated with known interactors . Below we discuss available information on several classes of these predicted interfaces . One significant advantage of our approach is that it enables domain knowledge to be readily incorporated into the inferred subnetworks . Specifically , the IP can incorporate constraints that represent knowledge about host factors and interactions that are known to be involved in viral replication , thereby influencing decisions about the rest of the subnetwork . These constraints were shown in the Computational Methods section . Here , we consider the effect of seeding the subnetworks with interactions from specific host pathways that are known to be involved in BMV replication . This set of domain knowledge , which we have elicited from the relevant literature , comprises 28 interactions among 24 host factors . It also specifies several host factors that should be treated as interfaces . For comparison , we also infer BMV subnetwork ensembles that do not use the literature-curated interactions . Seeding the subnetwork with these interactions does not have any apparent effect on hit-prediction accuracy , as we discussed earlier in the Results section . However , the interactions do appear to have an influence on their local neighborhoods . In examining the 97-interface BMV subnetwork ensemble , we observe a small number of cases in which the supplied interactions and interfaces serve to provide “anchors” that allow us to explain other , related hits . One set of edges extracted from the literature connects the ubiquitin-proteasome pathway to membrane synthesis , and specifies that Ole1p is an interface to BMV . The inferred subnetwork identifies a connection between Ole1p , a fatty acid desaturase , and Acb1p , which is involved in transporting newly synthesized fatty acids; the relevant portion of the subnetwork is shown in Figure 9 . The connection between the ubiquitin-proteasome pathway and Acb1p was not identified in any subnetwork inferred without the provided literature-based interactions . Furthermore , Ole1p is not inferred to be relevant at all without the provided interactions . Another component from the literature specifies the chaperone protein Ydj1p is an interface . The inferred subnetwork , shown in Figure 10 , identifies upstream connections from the hits Hsf1p and Ure2p to Ydj1p , which were not mentioned in the paper discussing Ydj1p's relationship to BMV [45] . These inferred connections demonstrate that the inferred subnetwork can be used to predict relevant connections between well-understood components of the network and host factors that have not yet been studied in detail . To supplement our manual analysis of predicted hits and interfaces , we employ the Model Based Gene-Set Analysis ( MGSA ) tool [69] to evaluate the ability of the inferred subnetwork to better identify relevant functional categories than an analysis of the experimental hits alone . The MGSA method uses a Bayesian network to analyze the representation of all GO terms in a gene set at once . As output , it provides the marginal probability that each GO term accounts for the input gene set . We use MGSA to analyze first the experimental hits and literature-derived relevant genes for BMV that are present in the background network , and second , the experimental hits combined with the predicted hits from the 97-interface inferred subnetwork . We use a probability threshold of 0 . 25 because we are willing to tolerate a degree of redundancy in the results , in exchange for the identification of a thorough list of representative GO terms . We further assess the significance of each returned GO term by comparison to the subnetworks inferred from random data . For each GO term , we generate a p-value as the proportion of random subnetworks for which MGSA gives a greater or equal probability . Table 7 presents the GO terms returned by MGSA with probability for the combined set of experimental and predicted hits . The “Experimental Hits” columns show the number of experimental hits associated with each GO term , and MGSA's probability that the GO term explains the experimental hits alone . Similarly , the “Predicted Hits” columns show the number of additional predicted hits associated with the GO term , and MGSA's probability that the GO term explains the combined experimental and predicted hit set . The “p-value” column shows the proportion of random subnetworks with equal or greater probability for the GO term as compared to the inferred subnetwork , with asterisks indicating . As shown in Table 8 , an additional 15 GO terms are identified by MGSA for the combined hit set , but are not identified for the experimental hits alone . A number of these GO terms represent only predicted hits . Eight of the GO terms receive a from the random subnetwork analysis . This result indicates that our subnetwork inference method predicts hits that ( i ) are useful for amplifying weak functional signals among the experimental hits , and ( ii ) are among themselves functionally coherent . Several of the amplified GO terms represent protein complexes or pathways that are recognized for their role in BMV replication . Deadenylation-dependent mRNA decapping factors are also known to be relevant [44] , and the perinuclear region of the cytoplasm is the cellular location in which BMV replicates [56] . Among the novel GO terms that contain no experimental hits are represented specific parts of the ubiquitin-proteasome system and ribosome synthesis , both of which we have noted are relevant to BMV replication . One advantage of our method is that we explicitly include protein complexes as nodes in our background network . We propose that doing so allows the inferred subnetworks to provide useful information about cooperative interactions between proteins . We use two Monte Carlo tests to assess the degree to which the representation of complexes among inferred subnetworks , and specifically among inferred interfaces , is due to ( i ) topological properties of the background network and inference procedure , independent of the experimental data , and ( ii ) properties of the experimental data , independent of the inference procedure . The details of these experiments are available in Text S3 , with results in Tables S3-S4 . Considering predicted relevant complexes in the 97-interface BMV subnetwork , 17 out of 65 predicted relevant complexes receive a p-value below 0 . 05 from either Monte Carlo test . Of the high-confidence , predicted BMV-yeast interfaces that are protein complexes or are members of complexes , five out of the eight receive a p-value below 0 . 05 from either test . These results indicate that the representation of many complexes by our inferred subnetworks are well-supported by predicted hits and are not likely to be artifacts of the background network or chance .
We have presented an approach that aims to elucidate how viruses exploit their host cells . Our approach uses known host intracellular interactions to infer ensembles of directed subnetworks which provide consistent explanations for phenotypes measured in genome-wide loss-of-function assays . This approach is able to represent a rich set of interaction types , in addition to domain knowledge about specific interactions that are known to be relevant . By inferring an ensemble of subnetworks , the approach is able to quantify its certainty about the relevance of various genes and interactions . The value of the subnetworks inferred by our method is that they can be used to ( i ) predict which unassayed genes may be involved in viral replication , ( ii ) interpret the role of each hit in modulating the virus , and ( iii ) guide further experimentation . Our empirical evaluation demonstrates that , using a gene-prioritization method as a sub-component , our method is able to predict phenotypes for unassayed genes with accuracy that is comparable to the gene-prioritization method alone . We also used our method to predict host-virus interfaces and additional relevant host genes for Brome Mosaic Virus , and performed a literature-based analysis of the predicted relevant host factors . While additional experimentation is necessary to confirm our predictions , a number of them are supported by domain knowledge . Among the predicted interfaces , many are known to bind or modify RNA , localize to the site of viral replication , or act in processes that have been previously identified as involved in viral replication . Similarly , many predicted hits are members of known relevant complexes , and a few are supported by independent experiments . These results are also supported by a Gene Ontology analysis which showed that our inferred subnetworks identify more relevant functional categories than the experimental data alone . Our experiments also demonstrated that the predictions made by our inferred networks have high levels of stability given small changes to the input data . There are a number of promising directions in which we plan to extend this work . Among them are applying the method to RNAi studies in more complex host networks and incorporating literature-extracted interactions into the background network . Our supplementary website is located at http://www . biostat . wisc . edu/~craven/chasman_host_virus/ . There we provide integer program code and data in the GAMS language , and visualizations of the background network and inferred BMV subnetworks as Cytoscape [70] files .
|
Nearly every step of the viral life cycle requires the action or use of host machinery . Genome-wide suppression experiments have been used to identify individual host genes whose products are involved in viral replication . The hit sets identified by such experiments are typically fairly large and difficult to comprehend . We propose a method to infer subnetworks of intracellular interactions that explain the experimental data . These inferred subnetworks make the data more interpretable in terms of the mechanisms of viral replication and can be used to guide further experiments .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"systems",
"biology",
"computer",
"and",
"information",
"sciences",
"mathematical",
"computing",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"host-pathogen",
"interactions",
"virology",
"biology",
"and",
"life",
"sciences",
"computing",
"methods",
"microbiology",
"computational",
"biology",
"pathogenesis",
"computer",
"inferencing"
] |
2014
|
Inferring Host Gene Subnetworks Involved in Viral Replication
|
Traveling waves in the developing brain are a prominent source of highly correlated spiking activity that may instruct the refinement of neural circuits . A candidate mechanism for mediating such refinement is spike-timing dependent plasticity ( STDP ) , which translates correlated activity patterns into changes in synaptic strength . To assess the potential of these phenomena to build useful structure in developing neural circuits , we examined the interaction of wave activity with STDP rules in simple , biologically plausible models of spiking neurons . We derive an expression for the synaptic strength dynamics showing that , by mapping the time dependence of STDP into spatial interactions , traveling waves can build periodic synaptic connectivity patterns into feedforward circuits with a broad class of experimentally observed STDP rules . The spatial scale of the connectivity patterns increases with wave speed and STDP time constants . We verify these results with simulations and demonstrate their robustness to likely sources of noise . We show how this pattern formation ability , which is analogous to solutions of reaction-diffusion systems that have been widely applied to biological pattern formation , can be harnessed to instruct the refinement of postsynaptic receptive fields . Our results hold for rich , complex wave patterns in two dimensions and over several orders of magnitude in wave speeds and STDP time constants , and they provide predictions that can be tested under existing experimental paradigms . Our model generalizes across brain areas and STDP rules , allowing broad application to the ubiquitous occurrence of traveling waves and to wave-like activity patterns induced by moving stimuli .
After an initial stage of activity-independent construction [1] , the developing nervous system undergoes a period of refinement that is strongly influenced by spontaneous and evoked patterns of neural activity [2 , 3] . Traveling wavefronts are a striking feature of these activity patterns [4–9] . Within short temporal windows , wavefronts induce strong interneuronal correlations that can act through Hebbian mechanisms of synaptic plasticity to build structure into the connectivity of neural circuits [10 , 11] . This has prompted the hypothesis that correlated activity plays an instructive role for circuit refinement in the developing brain ( reviewed in [2] and [3] ) . One Hebbian mechanism that is well suited to this role and is widely reported in the brain is spike-timing dependent plasticity ( STDP ) , for which synaptic connections are strengthened or weakened depending on the relative timing of pre- and postsynaptic spikes that arrive at the synapse , typically within tens of milliseconds of each other [12–14] . In this article , we undertake a mathematical analysis of the interaction between traveling wave activity patterns and STDP , and explore the types of connectivity patterns that emerge as a result of this interaction . Past studies have demonstrated that STDP could translate correlated input patterns into structured neural circuits [15–21] , and could mediate the construction of realistic receptive fields ( RFs ) with properties that resemble those found in the visual cortex [22 , 23] . These models primarily focussed on spatial and temporal correlations as separable features when considering their interaction with STDP . However , the spatiotemporal correlations induced by traveling waves are space-time inseparable , providing additional information that may be utilized during circuit building . Space-time inseparable activity patterns map the temporal profile of the STDP rule into a spatial profile of synaptic strength changes [24] , which can be used to build circuits that mimic neuronal sensitivity to visual motion during repeated exposure to moving visual stimuli [25–28] . But despite the demonstrated applicability of STDP to specific cases of neural circuit development , a more general , formal analysis of the interaction of STDP and wave-like activity patterns is still lacking . Here , we derive a mathematical expression that accounts for the interaction of a variety of traveling wave activity patterns and STDP rules , and we examine the analytical predictions in a simple yet biologically plausible model of spiking neurons . We show that , for a broad class of experimentally observed STDP rules , such interactions build highly structured and periodic connectivity patterns into feedforward circuits , analogous to a Turing instability in reaction diffusion systems [29] , which has been applied to diverse cases of biological pattern formation [30 , 31] . We then demonstrate the robustness of this pattern formation process and how it can be utilized to construct and refine the size and shape of RFs . Our results offer theoretical insights that may advance the understanding of the role played by traveling wave phenomena in different areas of the brain . We highlight particular insights into visual system development and outline a number of predictions that may be tested experimentally .
To understand how correlated activity caused by traveling waves could influence synaptic strengths via a STDP mechanism , we consider a reduced model ( Fig 1A ) consisting of a one-dimensional ( 1D ) layer of presynaptic input cells , all of which connect via excitatory synapses , wi , j , onto a single , postsynaptic output cell . In later sections , we extend the input layer to two dimensions . When a wave traverses the input population ( Fig 1A ) , each input neuron is recruited by the wavefront ( red colored unit with rightward arrow ) and discharges a burst of spikes , which drives spiking in the output neuron . Temporal differences between the spike times of a given input neuron and the output neuron , Δt = tin−tout , determine how the respective synapse is modified in strength according to a STDP rule , K ( Δt ) ( see Methods , Eqs 21 and 22 ) . The set of spikes for all input neurons leads to a simple diagonal band structure in space and time ( Fig 1B ) . To illustrate how this input spike pattern leads to spatially structured changes in synaptic strength , we consider a STDP rule that is asymmetric in Δt ( Fig 1C , top ) . In Fig 1D , we show a snapshot of the traveling wave and its influence on the surrounding synaptic strengths . Here , an input neuron ( colored dark red ) is recruited by the wavefront and fires a burst of spikes , which in turn elicits excitatory postsynaptic potentials ( EPSPs ) that drive the output neuron to spike . For waves traveling left to right , spikes generated by input neurons to the left of the wavefront always precede output spikes . Consequently , their respective synapses are strengthened , because the STDP rule specifies strengthening for negative Δt . Likewise , synapses connecting input neurons to the right of the wavefront will be weakened . In fact , due to the wave motion , the dependence of the STDP rule on relative spike times , Δt , is mapped onto the spatial axis of the input layer , Δx , and thus the relative spike locations , as shown in Fig 1D and 1G . Consequently , as depicted in Fig 1E , all input neurons ( colored light red ) that surround a synapse will influence the net change in strength at that synapse . One might predict this change to be proportional to the integral of the STDP rule , K 0 = ∫ − ∞ ∞ d Δ t K ( Δ t ) . However , the extent to which an input influences surrounding synapses depends on its influence over the output firing rate , and thus its own synaptic strength . For example , with an asymmetric STDP rule and waves traveling left to right , a synapse will strengthen relative to K0 if the surrounding synapses are stronger to the right than to the left , as in Fig 1F . On the other hand , a synapse will be relatively weakened if the surrounding synapses are stronger to the left than to the right . By the same argument , a synapse will be relatively weakened by a temporally symmetric STDP rule if the surrounding synapses are stronger on both sides ( Fig 1G ) , and relatively strengthened if the surrounding synapses are weaker . In this way , strong synapses increasingly dictate changes to local synaptic strengths as more waves pass . Eventually , the connectivity pattern begins to form islands of strong synapses that are flanked by regions of weak synapses . In the following analytical derivation and simulations , we show how this process of wave-induced STDP results in a distinct type of pattern formation in the network for a broad class of STDP rules . Here , we derive a description for the dynamics of synaptic strengths that are driven by pairs of input and output spikes acting through a STDP rule and resulting from traveling wave activity patterns traversing the input layer . Within our framework , input and output spike trains are generated by a stochastic process with a time-dependent firing rate . The stochastic arrival of spikes results in stochastic changes to the synaptic strengths , thus posing a challenge when seeking a description for the spatial structure of synaptic strengths that evolves slowly over long periods of time , during which many traveling waves occur . It is therefore useful to separate the slow dynamics from the fast , stochastic dynamics under the assumption that , during a limited period of time , ΔT , individual changes in synaptic strengths are negligible , but accumulate slowly over multiple periods of ΔT as a result of the time-averaged input and output activity . By approximating synaptic strengths as being constant during the period ΔT , Kempter et al . [16] showed that changes in synaptic strength over this period could be described by the inner product of the STDP rule , K ( Δt ) , and the cross-correlation function , Cij ( Δt , t ) , between the spike trains of input neuron i and output neuron j: w i j ( t + Δ T ) - w i j ( t ) Δ T = Δ w i j ( t ) Δ T ≈ η ∫ t t + Δ T d Δ t K ( Δ t ) C i j ( Δ t , t ) , ( 1 ) where η is a small , positive constant that sets the required slow rate of change in synaptic strengths . The cross-correlation is given by: C i j ( Δ t , t ) = ∫ t t + Δ T d t ′ S i ( t ′ + Δ t ) S j ( t ′ ) , ( 2 ) where Si ( j ) ( t ) are ensemble averages , for example over multiple waves , of the input ( output ) spike trains and can thus be identified with the input ( output ) firing rates . It is important that the firing rates be sufficiently high for Cij ( Δt , t ) to accurately portray wave-induced correlations . In addition , η must be sufficiently small for wave-induced correlations to be recovered over several waves . Moreover , without small η , calculating Cij ( Δt , t ) becomes difficult , as Sj ( t ) would depend on stochastically changing synaptic strengths , wij ( t ) . As such , Eq 1 implements the approximation by averaging over the small stochastic fluctuations , thus providing only the mean drift in wij ( t ) . Given that we will deal with discrete waves that pass one-by-one across the input layer , it is convenient to relate the time scale , ΔT , to the passage time of just a single wave . Further assumptions are now required to uphold the validity of Eq 1 . First , in order to ensure that multiple waves do not mutually influence changes in synaptic strength , ΔT must include an amount of time , 𝓚 , both before and after the wave , where 𝓚 is the temporal width of the STDP rule which contains most of its power . More formally , we require that ∫ − 𝓚 𝓚 d Δ t ∣K ( Δt ) ∣ ≫ ∫ − ∞ − 𝓚 d Δ t ∣K ( Δt ) ∣ + ∫ 𝓚 ∞ d Δ t ∣K ( Δt ) ∣ [16] . Because we are effectively considering Δwij ( t ) for a wave in isolation , we can extend the integral limits in Eqs 1 and 2 to ±∞ . Second , with wij ( t ) effectively constant during a wave , and because Δwij ( t ) is small , we will analyze changes in wij on a slower time scale , T , which is discretized in ΔT increments . Thus , we approximate wij ( t ) with wij ( T ) and Cij ( Δt , t ) with Cij ( Δt , T ) , and therefore approximate the left-hand side of Eq 1 with ∂wij ( T ) /∂T = ∂T wij ( T ) . By considering a simple 1D chain of input neurons with a single output neuron , we replace all subscripts i with the argument , x . That is , we replace wij ( T ) with w ( x , T ) , Si ( t ) with Sin ( x , t ) , and Cij ( Δt , T ) with C ( x , Δt , T ) . Eq 1 then becomes: ∂ T w ( x , T ) ≈ η ∫ - ∞ ∞ d Δ t K ( Δ t ) C ( x , Δ t , T ) , ( 3 ) where C ( x , Δ t , T ) = ∫ − ∞ ∞ d t S in ( x , t + Δ t ) S out ( t ) , with Sin and Sout the input and output firing rates , respectively . As a final matter of notation , we will hereafter refrain from explicitly writing the dependence of w and C on T , for brevity . We make two further assumptions to simplify the analytical derivation , then relax these for a more general case: i ) the input firing rate at the wavefront can be described as a short , traveling pulse using a Dirac delta function: Sin ( x , t ) = δ ( x−vt ) , where δ ( y ) = ∞ if y = 0 and is zero otherwise , and v is the wave speed; ii ) the output neuron’s response to its input is instantaneous: S out ( t ) = ∫ − ∞ ∞ d x S in ( x , t ) w ( x ) = w ( v t ) . Using these forms for S ( in ) out , C ( x , Δt ) = w ( x−vΔt ) and we can write Eq 3 as ∂ T w ( x ) = η ∫ - ∞ ∞ d Δ t K ( Δ t ) w ( x - v Δ t ) = η v ∫ - ∞ ∞ d Δ x K ( Δ x / v ) w ( x - Δ x ) = η K v ( x ) * w ( x ) , ( 4 ) where Kv ( x ) has been introduced as a rescaled copy of K , Kv ( x ) = v−1 K ( x/v ) , and * denotes convolution . There are two key features to Eq 4 . Firstly , as illustrated above , the STDP rule can be reinterpreted as a spatial kernel as a result of the wavefront’s constant velocity , which maintains a strict relationship between space and time . Secondly , the wave’s effect on the synaptic dynamics is described by a convolution of the STDP rule with the synaptic strengths . By deriving a solution for w ( x ) in Eq 4 , we will demonstrate how convolution plays an important role in the type of connectivity patterns that w ( x ) acquires , but first we relax the two simplifying assumptions used to reach Eq 4 . Incorporating finite input bursts and the dependence of output firing rates on EPSPs adds a simple modification to Eq 4 , which becomes ∂ T w ( x ) = η K v ( x ) * α ( - x / v ) * α ( x / v ) * ϵ ( x / v ) * w ( x ) , ( 5 ) where α ( t ) describes the time dependent firing rate during an input burst , and thus captures the shape of the wavefront , and ϵ ( t ) is the EPSP . In our model , both α ( t ) and ϵ ( t ) are positive valued for t > 0 ( Methods ) , and act as low pass filters on the STDP rule . The full derivation for Eq 5 is provided in S1 Text . Note that the firing rate of the output neuron ( Rout in Methods ) acts simply as a coefficient of ϵ in Eq 5 and hence plays a similar role to η by varying the rate at which synaptic strengths are modified . Amalgamating all terms in Eq 5 , except for w ( x ) , we have ∂ T w ( x ) = η κ ( x ) * w ( x ) , ( 6 ) where κ ( x ) is the effective spike-location dependent plasticity rule , which incorporates the dynamics of input bursts and EPSPs: κ ( x ) = K v ( x ) * α ( - x / v ) * α ( x / v ) * ϵ ( x / v ) . ( 7 ) When a typical STDP rule ( Fig 2A , top; see legend for parameters ) is transformed into a spatial kernel ( Fig 2A , bottom ) by a wave traveling at 3 mm/s ( the speed of spontaneous waves in the mouse cerebellum [8] , or a 25 °/s stimulus on the kitten retina using the visual angle to space conversion in Methods ) , the kernel extends over approximately 1 mm of input space . Note that κ ( x ) preserves the asymmetric shape of the STDP rule , but is low pass filtered by α ( t ) and ϵ ( t ) . A solution for w ( x ) as T → ∞ is more easily found in the frequency domain by taking the Fourier transform , such that convolution becomes multiplication: ∂ T w ˜ ( k ) = η κ ˜ ( k ) w ˜ ( k ) , ( 8 ) where ‘∼’ denotes the Fourier transform and k is the spatial frequency . Using w0 to describe the initial synaptic strengths at time T = 0 , the solution to Eq 8 for w ~ ( k ) is w ˜ ( k ) = w ˜ 0 ( k ) e η κ ˜ ( k ) T . ( 9 ) We can reconcile Eq 9 with several previous studies of STDP by writing κ ~ ( k ) = λ ( k ) + i ϕ ( k ) to explicitly express real and imaginary components: w ˜ ( k ) = w ˜ 0 ( k ) e η λ ( k ) T e i η ϕ ( k ) T . ( 10 ) Here , the imaginary component gives rise to a spatial phase shift in w ( x ) , a feature that has been exploited to drive a spatial redistribution of synaptic strengths by asymmetric STDP rules in several models of sequence learning [26 , 32] , asymmetric shifts in hippocampal place fields [33 , 34] , and the formation of direction selective cells [24 , 27 , 28] . The focus of our results , however , is the stability of w ~ ( k ) , which is determined by the real component , λ ( k ) . We therefore consider waves that travel in both directions so that , for sufficiently small η , κ ( x ) is effectively symmetric and contains no imaginary component . This can be shown by integrating Eq 8 over two wave events , with the first wave traveling left to right , and the second wave traveling right to left ( replacing v with −v ) . After the first wave , at T = T1 , w ˜ ( k , T 1 ) = w ˜ 0 ( k ) e η κ ˜ ( k ) T 1 . ( 11 ) Using w ~ ( k , T 1 ) as the initial condition for the second wave , we have at T = T2 w ˜ ( k , T 2 ) = w ˜ ( k , T 1 ) e η κ ˜ ( k ) ¯ ( T 2 - T 1 ) , ( 12 ) where κ ~ ( k ) ¯ is the complex conjugate of κ ~ ( k ) , with conjugation resulting from the sign reversal in the wave speed . Expanding Eq 12 , we have w ˜ ( k , T 2 ) = w ˜ 0 ( k ) e η λ ( k ) T 1 e i η ϕ ( k ) T 1 e η λ ( k ) ( T 2 - T 1 ) e - i η ϕ ( k ) ( T 2 - T 1 ) = w ˜ 0 ( k ) e η λ ( k ) 2 Δ T , ( 13 ) where ΔT = T1 = T2−T1 is the time taken for one wave to cross the input layer . We can therefore write an approximation to Eq 9 for the special case in which waves travel in both directions in equal numbers: w ˜ ( k ) ≈ w ˜ 0 ( k ) e η Re [ κ ˜ ( k ) ] T . ( 14 ) If the wave direction were random , instead of alternating , Eq 14 would be valid only if a large number of waves traversed the input layer during the interval ΔT/η . Thus , for any k such that Re [ κ ~ ( k ) ] < 0 , w ~ ( k ) will be stable and decay to zero , whereas for any k such that Re [ κ ~ ( k ) ] > 0 , w ~ ( k ) will become unstable and grow exponentially . For the STDP rules used in this study ( Fig 1C ) , κ ~ ( k ) has a positive valued maximum at k* ( Fig 2B ) . Therefore , if the synaptic strengths are initially random , such that the expected form of w ~ 0 ( k ) is flat , then w ~ ( k * ) will be the fastest growing eigenmode of the synaptic strengths , and w ~ ( k ) will asymptotically approximate a delta function , δ ( k* ) . If only a single spatial frequency dominates w ~ ( k ) , w ( x ) will be sinusoidal . Thus , our derivation predicts that , for STDP rules like those reported experimentally , the synaptic strengths will develop a connectivity pattern that is periodic in space with a period of 1/k* , under the influence of traveling waves . A similar model of pattern formation describes the development of ocular dominance columns in primary visual cortex [35] , in which it is proposed that short range excitatory and long range inhibitory lateral interactions give rise to the spatially periodic dominance of eye specific afferents across primary visual cortex . In our model , it is the mapping of the STDP rule onto space that provides the lateral interactions necessary for pattern formation . More generally , the solution derived above is analogous to pattern forming solutions that result from Turing instabilities in reaction-diffusion systems [29] , whereby the initially homogeneous distribution of synaptic strengths becomes unstable , allowing the fastest growing eigenmode to dominate the resulting pattern . In systems composed of a diffusible activator and an inhibitor , Turing instabilities frequently occur when the inhibitor diffuses over greater distances than does the activator [30] . Here , the decay constants of the positive and negative STDP lobes bear similarities to the length scales of diffusion in reaction-diffusion systems . Thus , STDP rules with narrow windows for strengthening and wide windows for weakening are good candidates for pattern formation in neural circuits that support traveling wave activity patterns . If the synapses are unbounded , the STDP rules used here will always yield a sinusoidal connectivity pattern when driven by traveling waves , given sufficient time , and the mean synaptic strength will always be zero ( the mean of a sine wave ) . This is far from a physiologically plausible scenario: without bounds , the synaptic strengths would tend towards ±∞ . When bounds are imposed , however , care is needed to formulate the STDP rule , so that the synaptic strengths do not reach the upper or lower bound before the dominant eigenmode at k = k* takes hold . This can be achieved by ensuring that there is not too strong a bias for either synaptic weakening or strengthening in the STDP rule: − B L < ∫ − ∞ ∞ d Δ t K ( Δ t ) = Re [ κ ~ ( k = 0 ) ] < B U , where BL and BU are , respectively , the magnitudes of the lower and upper bounds to the STDP bias . Because synaptic strengths are constrained to be positive , it must be that BL > BU . In other words , the range of biases for synaptic weakening that will yield stable periodic patterns is greater than that for strengthening . Nevertheless , it is important to note that we need not have BU < 0 . This contrasts with previous studies that utilized different input activity patterns [16 , 17 , 19] , and that required a bias for synaptic weakening by setting ∫ − ∞ ∞ d Δ t K ( Δ t ) < 0 to stabilize the connectivity pattern . That is , in our model , it remains possible for periodic patterns to emerge even if there is a bias for synaptic strengthening in the STDP rule , so long as the synaptic strengths are not pushed to the upper bound before the dominant spatial frequency takes hold . The stability of the connectivity will also be sensitive to the learning rate , which is scaled by η in Eq 6 , and the initial conditions of the synaptic strengths . For example , we later explore formulations of κ ~ ( k ) for which there are multiple peaks that are similar in amplitude , or a single , broad peak . The stochastic dynamics introduced by spiking neurons may therefore move the slower dynamics of plasticity along a spectrum of eigenmodes , resulting in more aperiodic connectivity patterns . We later introduce a robustness measure to quantify the periodicity of a connectivity pattern and that takes these potential features of κ ~ ( k ) into account . To test the predictions of the analytical derivation above , we examined whether a periodic connectivity pattern would develop as a result of traveling waves and STDP in a simulated network of linear Poisson , spiking neurons . The simulation captures the architecture and function of the network illustrated in Fig 1A , consisting of a 1D layer of input neurons that all synapse onto a single output neuron . During simulations , synaptic strengths were modified by one of two distinct forms of STDP rule ( see Methods ) that have been reported in the literature , one that is asymmetric ( Eq 21 ) [13 , 17 , 36] and the other symmetric ( Eq 22 ) [37 , 38] in Δt . Spiking activity is generated by a traveling wavefront that moves back and forth along the input layer , eliciting EPSPs and spikes in the output neuron . We varied two parameters in the simulation: i ) the temporal window of K ( Δt ) , which we control with the decay time constant for synaptic strengthening , τ+ , and ii ) the wave speed , v . These parameters modulate the shape of κ ( x ) and , therefore , the spatial frequency of the predicted periodic connectivity pattern . As predicted by our derivation ( Eq 9 ) , the synaptic strengths in the simulation indeed developed a periodic connectivity pattern ( Fig 2C ) in the presence of traveling waves for both the asymmetric and symmetric rules . The periodic pattern was robust over a range of STDP time constants ( τ+ = 20 , 30 , 40 , 50 , 60 and 70 ms ) , keeping the wave speed constant ( v = 3 mm/s ) , and for a range of wave speeds ( v = 1 , 2 , 3 , 4 , 5 , and 6 mm/s ) , keeping τ+ constant ( τ+ = 20 ms ) . As predicted , increasing τ+ or v caused the period of the connectivity pattern to increase ( Fig 2C ) . To test the accuracy with which Eq 9 predicts the resulting spatial frequency over this range of parameters , we computed the power spectrum of the steady state synaptic strengths ( mean subtracted ) to which a Gaussian curve was fit . The spatial frequency of the connectivity pattern was taken to be the centre of the fitted Gaussian . We repeated simulations sixteen times for each set of parameters , using a different seed for the random number generators . In Fig 2D and 2E , we compare the measured spatial frequencies ( blue circles ) with the predicted spatial frequencies , k* ( grey curves ) , as a function of τ+ ( Fig 2D ) and v ( Fig 2E ) , where the predicted values were found by numerically computing Re [ κ ~ ( k ) ] and determining the spatial frequency at its peak . We measured the accuracy of our predictions by computing the coefficient of determination , R2 , between the logarithms of the predicted and measured spatial frequencies . For each panel in Fig 2D and 2E , R2 > 0 . 85 . Thus , the assumptions made to derive Eq 9 appear to maintain a veritable description of the noisy dynamics in the simulation over the range of parameters tested . In addition to the simulations , we verified that solutions obtained by numerically integrating Eq 6 ( incorporating nonlinearities such as hard bounds to w ( x ) , see Methods ) also produced periodic connectivity patterns with spatial frequencies that matched predictions ( magenta circles , Fig 2D and 2E , R2 > 0 . 92 ) . To investigate whether nonlinearities in the postsynaptic response influence the outcome of the connectivity patterns , we replaced the linear output neuron with a leaky integrate and fire ( LIF ) neuron ( see Methods ) that modeled absolute and relative refractory periods of 2 ms and 5 ms , respectively . These simulations yielded a similar relationship between the spatial frequency of the connectivity pattern , wave speed and STDP time scale ( Fig 2F and 2G ) . This similarity might be expected , as the wave correlations extend over relatively long time scales and thus smooth out any ripples in the correlation function , C ( x , Δt ) , that would be introduced on the short time scales of the refractory period or EPSP rise time . More noticeable differences may , however , be observed if wavefronts elicited very short bursts or only single spikes . The development of periodic patterns in synaptic connectivity could have wide applications throughout the nervous system of many species , particularly because of the ubiquity of both traveling waves and STDP . However , it is first necessary to determine the extent to which pattern formation is influenced 1 ) over the range of space and time scales observed in biology , 2 ) in the presence of noise , and 3 ) by 2D waves . We explore these issues in the following sections . In the previous section , we found that the wave speed and STDP time scale are key parameters that determine the spatial frequency of periodic connectivity patterns . Traveling waves in different areas of the brain are characterized by wave speeds that span at least two of orders of magnitude , from slow retinal waves with speeds on the order of 0 . 1 mm/s [39] to fast cortical waves with speeds reaching 17 mm/s [9] . On the other hand , time scales for STDP are typically 10–100 ms [40] , but time scales on the order of seconds are predicted to be relevant to retinal waves [41] . We consider in the Discussion how the theoretical results above might apply to these different biological circuits . To do this , we first obtain a more complete picture of the spatial scales of pattern formation predicted by the theory over a wide range of wave speeds and STDP time scales . We calculated the landscape of spatial frequencies as a function of τ+ and v by numerically computing κ ~ ( k ) , and finding the spatial frequency , k* , that maximizes its real component . The k* landscapes for both the asymmetric and symmetric STDP rules ( Fig 3A and 3B , respectively ) reveal a remarkably similar dependence of the predicted spatial frequency on τ+ and v , which is perhaps not surprising given the simple exponential functions underlying the STDP rules ( Eqs 21 and 22 ) . In particular , for a large region of parameter space , scaling either v or τ+ by a factor , f , simply scales k* by 1/f . We illustrate this scaling feature using triangles with two equal sides aligned to the axes , as multiplication in linear space is equivalent to addition in logarithmic space . Starting at one iso-frequency contour ( 1 cycle/mm , bottom left vertex of white triangle in Fig 3A ) , a constant step along the v-axis moves k* to the same iso-frequency contour ( 0 . 04 cycles/mm ) as does an equal step along the τ+-axis . Because the input burst acts as a low pass filter on the STDP rule , this relationship does not continue ( green triangle , Fig 3A ) when τ+ becomes shorter than the burst duration ( here 0 . 1 s ) of the input neurons . Thus , the burst duration has a strong influence on k* at time scales longer than τ+ , which is particularly the case during retinal waves when burst durations are often as long as 3 s [42 , 43] . Addressing the impact of the burst duration is the focus of the next section . The warping of the spatial frequency landscapes above shows that the burst duration also plays a role in pattern formation . Furthermore , when correlated activity patterns comprise long burst durations , STDP rules with short time scales struggle to extract information from the correlations that might be relevant for circuit development [20 , 41 , 44] . Here , we further examine the influence of burst duration on pattern formation in our model by carrying out simulations over the range of burst durations that are observed for different types of traveling wave . Waves in this set of simulations traveled with a fixed speed of 3 mm/s , and we used the asymmetric STDP rule with a fixed decay time of t+ = 20 ms . In Fig 4A , we show examples of the evolving connectivity pattern for different burst durations . For bursts lasting 0 . 03 s , the connectivity pattern had only a very weak periodic structure ( Fig 4A , left panel ) . The power spectrum of connectivity patterns when the burst duration was 0 . 03 s , averaged over repeated trials , is shown in Fig 4B ( blue ) . By normalizing the spectrum to the power at its peak , it is clear to see that power is spread over a broad range of spatial frequencies . Burst durations on the order of a few hundred milliseconds produced more distinct periodic connectivity patterns ( Fig 4A , middle panels ) , with power concentrated around the peak in the power spectrum ( Fig 4B , orange ) . However , in simulations with burst durations of 1 s or longer , the connectivity pattern became disordered ( Fig 4A , right panel ) , with some power concentrated at the lowest spatial frequencies and otherwise spread evenly across higher frequencies ( Fig 4B , black ) . To summarize the robustness with which a periodic connectivity pattern is produced under different conditions , we define robustness to be the ratio of the power spectrum amplitude at its peak to the total power in the discretized power spectrum ( see Methods ) . All robustness measures are plotted in comparison to a reference value , which we take to be the mean robustness when parameters yielded a clear periodic connectivity pattern . Here , simulations with a 0 . 1 s burst duration were used as the reference . The robustness is plotted in Fig 4C for simulations using burst durations from 0 . 03 s to 5 s ( black circles ) . In agreement with the shapes of the power spectra in Fig 4B , the robustness is relatively high for burst durations in the range 0 . 1–0 . 5 s , and relatively low outside of this range . Because non-linearities in the simulation , such as bounded synaptic strengths , allow for the building of multiple spatial frequencies into the connectivity pattern , we examined whether the concentration of power in Re [ κ ~ ( k ) ] at the dominant spatial frequency , k* , could explain the relationship between the burst duration and robustness . That is , we numerically computed Re [ κ ~ ( k ) ] over the full range of simulated burst durations , from which we calculated the ratio of power at k* to the total power ( see Methods ) . We plot this theoretical measure of robustness in Fig 4C ( pink ) , having normalized it to the theoretical robustness when the burst duration is 0 . 1 s . The theoretical robustness provides a good estimate for the robustness with which the periodic connectivity patterns are produced in the simulations . To demonstrate how different burst durations distribute power across different spatial frequency ranges , we plot examples of Re [ κ ~ ( k ) ] in Fig 4D for three burst durations: 0 . 03 s ( blue ) , 0 . 3 s ( orange ) and 3 . 0 s ( black ) . For a 0 . 03 s burst , power is distributed across a wide range of high frequencies , whereas , for a 3 . 0 s burst , most of the power is concentrated in the negative dip at the lowest frequencies . However , for 0 . 3 s bursts , power is concentrated between these two extremes , around the dominant spatial frequency of k* . This trend is not specific to the choice of α ( t ) . In S1A Fig , we show Re [ κ ~ ( k ) ] for which α ( t ) is modeled using an alpha function . Despite the poor robustness at short burst durations , the spatial frequencies that were measured from the connectivity patterns were well matched to the predicted k* for burst durations between 0 . 03–0 . 5 s ( Fig 4E ) . However , for burst durations exceeding 0 . 5 s , the connectivity patterns no longer yielded spatial frequencies that matched the theory . This is likely to be due to the similar amplitude of multiple peaks in κ ~ ( k ) for longer burst durations ( for example , the black curve in Fig 4D ) , compounded by the overall loss in robustness ( Fig 4C ) . Butts & Rokhsar [41] have shown that waves with long burst durations provide more information that is relevant for refinement when the plasticity rule has a time scale much longer than is typically seen for STDP . It might therefore be possible to rescue pattern formation for waves with long burst durations by using a longer time scale for the STDP rule . To examine this possibility , we computed the fraction of power at κ ~ ( k * ) over the range of simulated burst durations for a STDP rule with τ+ = 50 ms ( Fig 4F , green ) . The wider STDP rule exhibits a greater concentration of power at k* over a much wider range of burst durations , when compared with the STDP rule with τ+ = 20 ms , and the same trend is seen for other α ( t ) kernels ( S1B Fig ) . This includes bursts exceeding 1 s in duration . For burst durations as long as 3 s , which are common for retinal waves , STDP rules with even longer time scales would be expected , according to our model . Thus , our model supports the hypothesis of Butts & Rokhsar [41] , and provides a new approach for estimating the required time scale of STDP with which retinal waves may refine developing neural circuits . Additional contributions to wave-related correlations can arise from the noisy mechanisms of wave and spike generation during early development , which may deliver waves in quick succession as well as generate non-wave related input spikes . The extent to which our analytical results can be applied to biological systems therefore depends on their robustness to these additional contributions to C ( x , Δt ) . In the following sections , we seek to understand how the presence of noise and multiple waves may impact upon the development of periodic connectivity patterns . Thus far , we have tested theoretical predictions using idealized wavefronts . In reality , spontaneous and sensory driven waves exist among continuous background activity and travel with variable wave speeds . Here , we use simulations to establish the sensitivity of pattern formation to these sources of noise . To examine the effect of variable wave speed , we conducted a set of simulations in which a new speed was assigned to each wave from a lognormal distribution , with a mean of 4 mm/s and standard deviation ( SD ) σv . We imposed a minimum speed of 0 . 05 mm/s so that waves would not take too long to traverse the input layer . Examples of the evolving connectivity pattern in these simulations are provided in Fig 5A and , in Fig 5C , we plot the mean power spectra of the final connectivity patterns for σv = 0 . 0 , 2 . 0 and 4 . 0 mm/s . An unexpected feature of the resulting connectivity patterns was a decrease in the spatial frequency with increasing σv ( orange: σv = 0 . 0 Hz , cyan: σv = 2 . 0 Hz , black: σv = 4 . 0 Hz ) . In these simulations , the developing connectivity pattern may experience a greater influence from the faster waves that recruit more input neurons per unit time and thus drive higher output firing rates . This would correspond to our earlier results in Fig 2 , for which we showed that lower spatial frequencies are associated with higher wave speeds . In Fig 5E , we plot the robustness of the connectivity pattern to variation in wave speed , where we have referenced robustness to the case when σv = 0 . Despite the variation in wave speed , which corresponds to there being a spectrum of spatial frequencies impressed on the network , the connectivity retains a reasonably robust periodic structure . We next tested the robustness of periodic connectivity patterning to the presence of background spiking noise , which diminishes the relative contribution of wave-activity to C ( x , Δt ) in Eq 3 . To implement background noise in the simulation , input neurons that were not participating in a wave fired spontaneous spikes with a firing rate of R0 , which was varied between simulations ( see Methods ) . Examples of the evolving connectivity patterns are shown in Fig 5B ( top row ) for different background firing rates . Mean power spectra of the final connectivity patterns when R0 = 0 . 0 ( cyan ) , 1 . 0 ( orange ) and 2 . 0 Hz ( black ) are plotted in Fig 5D , and show little variation in the dominant spatial frequency when R0 < 2 . 0 Hz . However , the robustness degraded substantially when R0 ≥ 2 . 0 Hz ( black circles in Fig 5F , reference case R0 = 0 Hz ) . In these simulations , a new wave traversed the input layer every 10 s , but the burst duration was only 0 . 1 s . Therefore , the ratio of wave-related spikes to background spikes , rather than the background rate , might better parameterize when the connectivity pattern will be robust . In this case , periodic patterns lacked robustness when the ratio went below approximately 1:4 . Accordingly , the robustness could be significantly enhanced for almost all background rates by increasing the firing rate during a wave burst to 100 Hz . In this case , the robustness degraded substantially when R0 ≥ 4 Hz ( Fig 5B , bottom row; Fig 5F , grey circles ) . During early , spontaneous waves , low background firing rates are typical , and there is good reason to believe that background firing rates are low in early stages of sensory processing as well [45] . We review more evidence for this in the Discussion . Spike-spike correlations induced by the wavefront , rather than time averaged firing rates , are the driving force behind periodic patterning in our model . Additional sources of spike-spike correlations may therefore disrupt periodic patterning . We conducted a set of simulations in which inputs experienced instantaneous correlations with their neighbors ( see Methods ) while also varying the background firing rate . With non-zero background rates , increasing the correlation strength does indeed degrade the robustness of the periodic connectivity patterns ( Fig 5F , orange circles ) , referenced to simulations with no local correlations and R0 = 0 . 0 Hz . In the absence of background spiking , however , the additional local correlations act to enhance the robustness . Typically , correlations are not instantaneous but decay over time [46–49] and thus would have a reduced impact on plasticity due to the decaying amplitude of the STDP rules around Δt = 0 [18] . The intervals between consecutive spontaneous waves can be very variable , and inter-wave intervals ( IWIs ) cover a broad range , from approximately 100 ms in the cerebellum [8] to tens of seconds in the retina [39 , 50] . IWIs during sensory driven waves may also be highly variable , and are likely to match the temporal pattern of naturally occurring stimulus features to which a population of input neurons are tuned . During vision , for example , multiple luminance contours can be cast across the retina in quick succession when tracking a moving object . In the following , we investigate how the presence of multiple waves , which simultaneously drive spiking in the output neuron , impact the development of the connectivity pattern . In so doing , we test our assumption in the above derivation that waves be sufficiently isolated in time . We ran a set of simulations in which waves were generated with an approximately lognormal distribution of IWIs and varied the mean IWI , μIWI , between simulations ( see Methods ) . At a speed of 4 mm/s , waves took 2 . 5 s to traverse the input layer . Thus , to ensure that multiple waves were present for the majority of the simulation , we set μIWI ≤ 2 . 0 s across the set of simulations . Despite multiple waves driving the output cell , periodic connectivity patterns emerged in the simulations , examples of which are shown in Fig 6A ( left and center panels ) . Power spectra of the connectivity patterns had distinct peaks near the predicted spatial frequency ( 0 . 91 cycles/mm ) for μIWI as short as 0 . 5 s ( Fig 6C ) . The connectivity pattern began to degrade for μIWI < 0 . 5 s , and had little structure when μIWI ≤ 0 . 2 s ( Fig 6A , right panel ) , with power spread broadly across the spectrum ( Fig 6C ) . The loss of robustness with decreasing μIWI is summarized in Fig 6E ( black circles ) , where robustness is referenced to simulations in which a constant IWI of 5 s was used , thus ensuring that only one wave was present at a time . The loss of robustness with decreasing μIWI may be related to our requirement in the above derivation that waves be sufficiently isolated in time , with ΔT > 2𝓚 , where ΔT can be interpreted as the IWI and 𝓚 as half the temporal width of the STPD rule . Although 𝓚 is not precisely defined , values in the range 0 . 1–0 . 3 s would be reasonable for the STDP rule used here , for which the decay time of the longer , negative lobe was τ− = 0 . 04 s . An alternative explanation might be that successive waves with very short IWIs are not very different from single waves with long burst durations , as far as the timing precision of the STDP rule is concerned . Because the burst duration also influences robustness , these two possible factors would be difficult to disentangle , and efforts to do so go beyond the scope of this study . The important qualitative result is that , above a lower limit to the mean IWI of only a few tenths of a second , the succession of randomly occurring waves does not greatly degrade the robustness of periodic connectivity patterns . We next asked whether a constant IWI between waves , which would contribute a strong frequency component to κ ~ ( k ) , would lead to a similar degradation in robustness with decreasing IWI . Because wavefronts of activity in the retina readily track moving luminance edges [7] , regular waves could correspond to stimulation of the retina by luminance gratings , as used in experiments studying the development of orientation and direction selectivity [51] . We ran simulations in which waves traversed the inputs in a periodic fashion , with a constant IWI , but otherwise used the same parameters as for the simulations with irregular waves . Examples of the periodic patterns that developed for different IWIs are provided in Fig 6B , and their power spectra are shown in Fig 6D . It is clear that regular waves built robust periodic patterns , even with the shortest IWIs . Our robustness measure confirmed that this was the case across all of the IWIs tested ( Fig 6E , grey circles ) . In addition to the enhanced robustness , a distinct feature of the connectivity patterns with constant IWIs is the shift towards higher spatial frequencies for short IWIs ( Fig 6D , power spectra for IWIs of 0 . 2 s and 0 . 15 s ) . The emergence of robust periodic connectivity patterns , with increased spatial frequencies for fixed IWIs ≤ 0 . 2 s , appears at odds with our theoretical predictions . However , these features can be easily accounted for by considering how κ ( x ) is constructed in Eq 7 . To represent the periodic input bursts elicited by multiple waves , we replace α ( t ) in Eq 7 with α p ( t ) = ∫ − ∞ ∞ d t ′ ∑ n δ ( t ′ − ( n × IWI ) ) α ( t − t ′ ) , the convolution of α ( t ) with a Dirac comb . Eq 7 expresses κ ( x ) as the convolution of Kv ( x ) , αp ( x/v ) , αp ( −x/v ) and ϵ ( x/v ) , and can be solved by taking its Fourier transform , whereby convolution in the real domain is equivalent to multiplication in the frequency domain . In Fig 6F , we plot the two functions that have the greatest influence in our simulations . In black is the Fourier transform of the STDP rule , K ~ v ( k ) . The remaining curves are the Fourier transforms of the input firing patterns , α ~ p ( v k ) ( with the zeroth frequency component removed ) , for three cases: a single wave in isolation ( effectively , IWI = ∞ grey curve ) and periodic waves with IWI = 1 . 0 s ( blue ) and IWI = 0 . 15 s ( pink ) . Note that the Fourier transform of the EPSP has little effect in this calculation because its time constant is much shorter than that of the other functions . The product K ~ v ( k ) ∣ α ~ p ( v k ) ∣ 2 therefore yields a good approximation of κ ~ ( k ) , and is drawn for the three cases of α ~ p ( v k ) in Fig 6G . Because the Fourier transform of a Dirac comb is another Dirac comb , α ~ p ( v k ) produces sharp peaks in κ ~ ( k ) ( pink and blue ) at spatial frequencies that may be higher ( pink ) than the peak in the case of isolated waves ( grey ) . The large amplitude and narrow peaks in κ ~ ( k ) , when IWI = 0 . 15 s and 0 . 2 s , explain the dominance and robustness of a single spatial frequency in the respective connectivity patterns , and the positions of these peaks explain the shift of the power spectra toward higher spatial frequencies in Fig 6D . The strong dependence of the spatial frequency on the IWI for short IWIs only , as depicted in Fig 6F and 6G , suggests a critical value for the IWI . We can relate the critical value to the time taken for a wave to travel a distance equal to one cycle of the periodic pattern: IWIcrit = 1/vk* ≈ 0 . 27 s . That is to say , when IWI < IWIcrit , the spatial frequency with which waves are lined up along the input layer exceeds the dominant spatial frequency of κ ( x ) for a wave in isolation ( S2 Fig ) . This interpretation fits well with our assumption in the derivation that waves be sufficiently isolated in time ( and therefore isolated in space ) . Moreover , the existence of a critical IWI provides a means for quantifying 𝓚 , and thus provides a possible explanation for the degradation in robustness when μIWI ≤ 0 . 3 s in the simulations with irregular waves . The filtering properties of a RF result from feedforward and recurrent synaptic inputs to the neuron , as well as its intrinsic cell properties . In this section , we used simulations of spiking neurons to examine whether the interaction of traveling waves and STDP rules , which can impose spatially periodic connectivity patterns on a uniform field of synaptic strengths ( shown above ) , might be a plausible mechanism for shaping the feedforward component of RFs . As such , a RF in our model refers to the spatial pattern of synaptic strengths impinging on the output neuron from a small region of the input layer . During development , coarse RFs are thought to be set up by molecular cues before activity dependent refinement takes place [52] . Therefore , at the beginning of each simulation , we set synapses from the center of the input layer to maximum strength and the remaining synapses to zero , thus endowing the output neuron with an initial RF having a diameter , RF0 . Despite this restricted initial arrangement of synaptic strengths , we found that a periodic pattern would nevertheless emerge across the entire input array ( S3 Fig ) . This outcome is not typical of the brain , as topographic maps and the limited size of axonal arbors and dendritic trees restrict the spatial distribution of synaptic inputs to a neuron . We therefore applied an arbor function ( see Methods ) that was centered in the middle of the input layer and spanned a region greater than RF0 . Synapses that were outside the arbor were disconnected from the output neuron and prevented from strengthening . A schematic of this new network architecture is provided in Fig 7A . In the majority of the following simulations , we set RF0 = 0 . 8 mm , which approximates an area of retina that corresponds to RF sizes in developing areas of the visual system , including the cat primary visual cortex [53 , 54] , the lateral geniculate nucleus ( LGN ) in ferrets [55 , 56] , and the superior colliculus ( SC ) in adult mice that have had disrupted retinal waves . The conversion between RF size and retinal distance is described in the Methods . Furthermore , in the remaining sections , we increase the bias for synaptic weakening in the asymmetric STDP rule from A− = 0 . 51 to A− = 0 . 55 , which is useful for RF refinement . Later , we examine the influence of this bias in greater detail . By interpreting the output neuron RF as a single cycle in the type of periodic connectivity patterns obtained above , we are able to examine how properties such as the wave speed and the STDP time scale may shape RF development by controlling the characteristic wavelength , 1/k* , of the connectivity pattern . Depending on the size of 1/k* with respect to RF0 , one of three modes of RF modification were observed . In one set of simulations , the RF became larger ( Fig 7B , left ) , smaller ( Fig 7B , middle ) or split into subfields ( Fig 7B , right ) with progressively slower wave speeds and thus shorter characteristic wavelengths . The same modes of RF modification were achieved by decreasing the STDP time scales ( Fig 7C ) and holding the wave speed constant . If instead of changing k* we increased RF0 , keeping the wave speed and STDP rule fixed , the RFs were similarly modified ( Fig 7D ) . Thus , characteristic wavelengths that are long relative to RF0 caused the RF to grow , shorter wavelengths caused the RF to shrink , and even shorter wavelengths enabled multiple cycles of the periodic pattern to form within the arbor . Our results suggest that 1/k* in fact determines the size of the final RF , or each subfield . To verify this , we varied RF0 from 0 . 2 mm to 3 . 2 mm between simulations , with the wave speed and STDP rule fixed , and looked for the convergence of RF sizes , which we measured as the number of synapses with strengths > 0 . 5 . To accommodate RF0 , the width of the arbor function was set to the larger of 0 . 8 mm or RF0+0 . 4 mm . RF sizes converged to one of three sizes ( Fig 7E ) , depending on whether the RF maintained a single field ( purple ) , or split into two ( yellow ) or three ( blue ) subfields . The larger RF0 , the more subfields that emerged . Regardless of RF0 , the mean final size of each subfield was approximately the same . Integer multiples of the mean subfield size ( measured from RFs with three subfields ) are drawn on the right of Fig 7E to illustrate this fact . Small biases were evident , however , whereby larger values of RF0 tended to give rise to larger RFs or subfields . This considered , the final number of strong synapses were easily distinguishable between cases in which one , two or three subfields developed . Thus , the size and shape of the final RF tightly corresponds with the wave , STDP and initial RF properties . We showed above , in Fig 7B and 7C , that different wave and STDP parameter combinations could yield the same mode of RF modification , including RF expansion , contraction , or splitting . Here , we focus on RFs that are made to contract , which is particularly relevant to the refinement of topographic maps such as the retinotopic map in superior colliculus , and examine the range over which the wave speed and STDP time scale achieves this mode of modification . To do this , we numerically integrated Eq 6 ( see Methods ) for the wide range of wave speeds and STDP time constants used to generate Fig 3 , holding RF0 = 0 . 8 mm constant . An example of w ( x ) during the numerical integration of Eq 6 , starting with an initial RF , is shown in Fig 8A , for which parameters matched those in the central panels of Fig 7B and 7D . Using the solutions for w ( x ) , we constructed a phase diagram for the size and shape of RFs that developed , which we call the refinement phase space . In the refinement phase space , shown in Fig 8B , there is a region ( shaded blue ) corresponding to v and τ+ parameters that caused the RF to contract , with darker shades of blue indicating greater contraction , measured as a percentage of RF0 . In the region labelled α ( unshaded ) , at which κ ( x ) has higher characteristic spatial frequencies , RFs split into subfields . In the region labelled β ( unshaded ) , at which spatial frequencies are lower , RFs expanded . Overlaid are the iso-frequency contours of the predicted spatial frequencies , k* , as in Fig 3 . The boundaries between each mode of RF modification closely follow the iso-frequency contours , with greater RF contraction ( darker shades ) occurring at higher spatial frequencies . Because the outcome of the final RF depends on the initial RF size , we recomputed the refinement phase space , setting RF0 = 0 . 44 mm , which just exceeded the smallest size of RFs after contracting in solutions when RF0 = 0 . 8 mm . For RF0 = 0 . 44 mm , the region of phase space in which RFs contracted ( colored purple ) corresponded to slower wave speeds and shorter STDP time constants , i . e . higher spatial frequencies in κ ( x ) . Taken together , these results confirm that smaller initial RF sizes require higher spatial frequencies in κ ( x ) to achieve a particular mode of RF modification . The area of the blue and purple regions of phase space also show that , in order for RFs to be refined to a smaller size by the asymmetric STDP rule , the maximum possible wave speed can be just over a factor of two times greater than the minimum possible wave speed . At least the same range in τ+ can also enable RF contraction . Thus , the mechanism of RF contraction described here allows for some variation in STDP properties between synapses , variation in wave speed , or even fluctuations in both of these phenomena over time . An interesting corollary of these results is that changes in wave speed and STDP properties can follow changes in RF size , thus allowing continued contraction beyond that of a fixed v and τ+ combination . To demonstrate this point , we ran a simulation using the same parameters as those used for the central panels in Fig 7B and 7D . However , once the RF had refined to a smaller size , we decreased the wave speed from 4 mm/s to 2 . 5 mm/s . Note that , according to Fig 8B , v = 2 . 5 mm/s and τ+ = 20 ms would cause a RF with RF0 = 0 . 8 mm to split into subfields , as this point in the phase space lies to the lower left of the blue shaded region . However , as shown in Fig 8C , an initial period of contraction with v = 4 mm/s set up further contraction , after a decrease in wave speed to v = 2 . 5 mm/s , without splitting the RF into subfields . Increasing the wave speed back to 4 mm/s at a later time caused the RF to expand and return to its previous size . Analogously , changes in the time constants of STDP can achieve the same effect . The phase diagram corresponds very well with the simulation results in Fig 7 , confirming our hypothesis that κ ( x ) and RF0 jointly determine the final shape and size of the RF . This was not unique to the asymmetric STDP rule . In Fig 8D , we show a similar correspondence between refinement and the iso-frequency contours for a symmetric STDP rule . Tuning the characteristic wavelength by itself was not sufficient to achieve RF contraction . In addition , both asymmetric and symmetric STDP rules required a bias towards synaptic weakening . In Fig 8E , we plot the refinement phase space of a 0 . 8 mm RF for three asymmetric STDP rules that differed only in A− , which scaled the amplitude of the negative lobe in the rule . When A− = 0 . 5 , we have ∫ − ∞ ∞ d Δ t K ( Δ t ) = 0 , such that synapses no longer compete for connections with the output . Consequently , increasing the dominant spatial frequency of κ ( x ) would not decrease the size of the RF before causing it to split into subfields ( left panel ) . However , by increasing A− ( A− = 0 . 55 , middle panel; A− = 0 . 6 , right panel ) , such that ∫ − ∞ ∞ d Δ t K ( Δ t ) < 0 , RF contraction was possible . Furthermore , increasing A− moved the iso-frequency contours , making RF contraction possible at lower characteristic spatial frequencies and , therefore , larger v and τ+ values . As such , for a given v and τ+ pair ( red circles in Fig 8E ) , greater contraction was achieved with increased bias for synaptic weakening . Equipped with a basic understanding of the relationship between periodic patterning and RF modification , we examine how more realistic traveling waves and STDP impact the development of RFs in two dimensions in the following sections . In two dimensions , the variables x and v in Eq 6 can now be considered as 2D vectors , x = ( x , y ) and v = ( vx , vy ) , in the x-y plane . As such , we expected the connectivity pattern to adopt the characteristic spatial frequency , k* = ( kx , ky ) , and thus for the type of RF modification along a particular axis to depend on the direction of wave propagation . To test this , we generated wave stimuli in a 2D input layer consisting of 64×64 units arranged on a square lattice . Plane waves were generated in a similar fashion to the 1D scenario , moving with a constant speed in alternating directions , from one side of the input layer to the opposite side with a wave speed of 4 mm/s . The single output neuron was provided with an initial , circular RF 0 . 8 mm in diameter ( Fig 9A , left panel ) , and synapses were modified by an asymmetric STDP rule ( τ+ = 20 ms ) . When plane waves travelled along just one axis , the RF contracted along the same axis ( Fig 9A; top panels: waves travel along the horizontal axis; bottom panels: waves travel along the vertical axis ) , corresponding with the 1D results above ( Fig 8B ) . However , the RF grew along the orthogonal axis until it spanned the full extent of the arbor . This corresponds with an effectively infinite wave speed along the orthogonal axis , resulting in a periodic pattern with a spatial frequency of zero . Exposing the network to a reduced set of wave directions in this way can be used to build RFs that exhibit periodic structure along the same directions . In S4 Fig , we provide examples of RFs that developed when waves traveled in both directions along one , two and three axes . Of particular interest are RFs that exhibit multiple parallel subfields ( S4A Fig ) , which bare several similarities with oriented simple cell RFs in primary visual cortex . In the Discussion , we describe in more detail how periodic patterning might be applied to realistic simple cell RFs . When waves travelled in 16 possible directions , equally spaced around the compass and in a random sequence , the final RF maintained an approximately circular shape ( Fig 9B , left ) . However , using the same wave and STDP properties as for Fig 9A , the RF area increased ( Fig 9C , dark blue ) because contraction along the axis parallel to the wave direction was outweighed by expansion along the orthogonal axis . RF expansion could be counteracted by changing parameters that enhance contraction , as explained in the results of Fig 8 . Thus , increasing k* by decreasing the wave speed to 3 mm/s ( Fig 9B , middle ) , or increasing the amplitude for synaptic weakening in the STDP rule from A− = 0 . 55 to A− = 0 . 6 ( Fig 9B , right ) , caused RFs to contract ( Fig 9C , light blue and purple , respectively ) . This shows that the principle of using STDP and traveling waves to refine a RF extends from 1D to 2D networks for simple , idealized waves . In the following section , we will examine how this process holds up when waves follow more irregular and complex trajectories . To test the robustness of refinement under conditions in which waves are far from idealized plane waves moving with a constant velocity , we used a wave model developed by Feller et al . [39] to generate complex wave patterns in the input layer . In the Feller et al . model , waves are generated spontaneously in random locations , and propagate along winding trajectories on a 2D input layer ( see Methods ) . Due to the complexity of the model , it was not possible to set a precise wave speed . We therefore controlled the mean wave speed by temporally rescaling precomputed wave patterns , and measuring the speeds of waves that were isolated by a center of mass ( COM ) tracking algorithm ( Methods and S7 Fig ) . In this way , we generated slow , medium and fast waves with speeds of 1 . 29±0 . 01 mm/s , 2 . 58±0 . 02 mm/s and 3 . 87±0 . 03 mm/s ( mean ± SEM ) , respectively . Examples of two isolated waves are shown in Fig 10A , and a 90 s movie of isolated waves is provided in S1 Movie . Using these rescaled wave patterns as input to simulations of RF development , we found that , quite remarkably , complex waves could shape and refine RFs in much the same manner as simple plane waves . With an asymmetric STDP rule and holding t+ = 20 ms fixed , RFs could be made to expand ( Fig 10B , left ) , contract ( Fig 10B , center ) or split into subfields ( Fig 10B , right ) by the fast , medium and slow complex waves , respectively . Although slow waves did not always split the RF , the final RF area was always smaller with slow waves than with medium wave speeds . The evolution of the RF area during the simulation , averaged over repeated trials , is plotted in Fig 10C for the different wave speeds ( solid: fast waves; dashed: medium waves; dotted: slow waves ) . Similarly , STDP rules with shorter STDP time scales produced RFs with smaller areas ( Fig 10D and 10E ) , corresponding to a shorter characteristic wavelength in a periodic connectivity pattern . These results recapitulate the dependence of RF development on wave speed and STDP time scales , even when the input wave patterns followed complex and noisy trajectories .
Our theoretical results , expressed in Eqs 6 and 9 , are analogous to those derived by Swindale [35] , who modeled the development of ocular dominance columns in primary visual cortex . Swindale showed that short range excitatory and long range inhibitory lateral interactions could give rise to the spatially periodic dominance of eye specific afferents across primary visual cortex , a principle also used to model the development of orientation columns in visual cortex [57] . In our model , the lateral interactions between synapses that are necessary for pattern formation result from the timing dependence of STDP , which is mapped onto space by the spatiotemporal correlations of traveling waves . Pattern formation of this kind is analogous to a Turing instability in reaction-diffusion systems [29] , which has been applied to diverse cases of biological pattern formation [31] . A useful recipe for Turing-like pattern formation includes the presence of a diffusible activator and inhibitor , whereby the inhibitor diffuses over greater distances than the activator [30] . STDP rules emulate this feature if τ− > τ+ , i . e . the temporal window for synaptic weakening is longer than that for synaptic strengthening . Such STDP rules are widely reported throughout the brain [14 , 58] . In the alternative scenario , when τ+ > τ− , we found that STDP tends to weaken local regions of synapses that had strengthened by chance , preventing islands of strong synapses from forming . It is not impossible , however , for periodic patterns to form when τ+ > τ− , so long as we carefully choose A+ and A− , which control the overall bias for strengthening and weakening . However , we found that the stability of the pattern is very sensitive to small changes in A+ and A− in this case . Models of circuit development by Hebbian plasticity require constraints on the synapses so as to keep their strengths within biologically realistic bounds . A suitable choice of constraint , such as subtractive normalization , provides competition between synapses so that when a subset of synapses strengthen , all other synapses are suppressed [11 , 59 , 60] . Under this constraint , correlations between nearby inputs over short temporal windows encourage a localized group of synapses to strengthen , yielding a RF-like connectivity pattern [11] . This is because synapses separated by larger distances are uncorrelated within short temporal windows , so are rarely strengthened but are often weakened together . STDP rules that are biased for synaptic weakening can achieve this type of competition between uncorrelated synapses , and are therefore capable of building RF-like connectivity patterns [18] . However , this competitive process is fundamentally different to the mechanism by which STDP builds structured connectivity patterns in this paper . We recognize that a single wave can induce strong correlations between any two input neurons that it passes , irrespective of their separation . They will be correlated with a specific time lag , Δt , which scales with the separation , Δx , between the two inputs as Δx = vΔt . If Δt corresponds to synaptic weakening in the STDP rule , the synapses of the two input neurons will compete as a result of their strong correlations . Note that waves cannot produce periodic connectivity patterns if synaptic weakening does not have a particular dependence on timing , for example when competition is provided by subtractive normalization , as this will not yield the bandpass filtering that we have demonstrated for experimentally observed STDP rules ( Fig 2B ) . This underlies why the wave speed , and the range of Δt values for which the STDP rule specifies synaptic weakening or strengthening , determines the resulting spatial frequency of the connectivity pattern , and therefore influences the type of modification experienced by a RF . In Fig 5 , we showed how contributions to the correlation function that are not wave-related act to degrade the periodic connectivity pattern . The strength of wave-related correlations may be further suppressed by recurrent inputs arriving from other postsynaptic cells , which we have excluded from our model . Output spikes that are driven by recurrent activity are likely to be much less correlated with spikes in the input layer and will therefore contribute an approximately constant term to the correlation function , C ( x , Δt ) , the effect of which will be to emphasize any bias in the STDP rule towards synaptic weakening or strengthening . Moreover , excitatory recurrent connections are well known for encouraging similarities between postsynaptic RF properties , whereas inhibitory connections are known to decorrelate RFs , without drastically changing the basic RF properties of individual postsynaptic neurons [18 , 57 , 61–63] . Increasing the bias for synaptic weakening in our simulations enhanced the degree to which RFs could be refined to a smaller size ( Fig 8E ) . A bias for weakening in STDP is frequently observed in experiments [14 , 64] , but is not the only possible source of synaptic competition , which can also be achieved by homeostatic regulation and intrinsic plasticity [65 , 66] . As mentioned above , these additional sources of competition would not aid pattern formation if they do not have the appropriate spike timing dependence that yields bandpass filtering as do the STDP rules used in this study . Several studies have previously explored the interaction between space-time inseparable input patterns and STDP . They demonstrated how motion stimuli can act through asymmetric STDP rules to impart an asymmetry in the spatial profile of excitatory connection strengths . This can be utilized to endow a network with direction selectivity similar to that found in the visual cortex [25–28] . Through our analytical results , we have shown that this phenomenon is one component of a richer set of dynamics with which the connectivity pattern can evolve under space-time inseparable inputs . Specifically , synaptic strengths are modified according to the convolution of the spatially mapped STDP rule , κ ( x ) , with the synaptic strengths , w ( x ) . Because of this convolution , the imaginary component of κ ~ ( k ) results in spatial shifts , whereas the real component results in the emergence of a periodic connectivity pattern . It is possible that periodic patterning was occluded in previous modeling studies due to the setup of those models for the specific application to the development of direction selectivity . Retinal waves have been reported in the developing retina of several species [4 , 67–71] , suggesting that they play an important developmental role that has been conserved over the course of evolution . In mice , for example , disrupting normal retinal wave propagation [42 , 72] has a profound effect on the refinement of retinal ganglion cell ( RGC ) afferents to , and RFs in , superior colliculus ( SC ) [73 , 74] . One way in which retinal waves may drive this refinement is by imposing a wavelength onto the spatial structure of retinocollicular connections . To illustrate the relationship between RF sizes in SC , retinal wave speeds and STDP time scales , we computed the landscape of dominant spatial frequencies ( S5 Fig ) as a function of v and τ+ , as in Fig 3 , but using a typical retinal wave burst duration of 2 s for α ( t ) [42 , 43] . Retinal waves travel at relatively slow speeds of 0 . 1–0 . 2 mm/s , such that if the STDP rule at retinocollicular synapses is asymmetric with τ+ = 20 ms , the characteristic wavelength would be 1/k* ≈ 0 . 09 mm ( solid red rectangle in S5 Fig ) . By interpreting a RF as half a cycle in the periodic pattern , a wavelength of 0 . 09 mm corresponds to a RF size in mice of 1 . 8° , using the distance to visual angle conversion in Methods . This is much smaller than the reported RF sizes of ∼10° [74] . In order to obtain RFs ∼10° in diameter ( equivalent to a wavelength of 0 . 51 mm ) with slow retinal waves , the STDP time scale would have to be τ+ = 0 . 15 s ( dashed red rectangle in S5 Fig ) , almost an order of magnitude longer than time scales commonly reported in STDP studies . The long duration of retinal wave bursts imposes an additional constraint on the minimum STDP time scale . We showed in Fig 4 that , when τ+ = 20 ms , periodic patterning degraded when the burst duration exceeded 0 . 5 s . However , as the STDP time scale increases , so too should the limiting burst duration ( Fig 4F ) . Assuming that the limit imposed by the burst duration scales linearly with τ+ suggests that a STDP time constant exceeding 0 . 1 s should operate at retinocollicular synapses during retinal waves . Our prediction of long STDP time scales at developing retinofugal synapses supports previous work by Butts and Rokhsar [41] , who showed that retinal waves convey the most information about the relative retinotopic positions of RGCs over time scales ranging 0 . 1–2 s . This was further supported by the observation of such a rule at developing retinogeniculate synapses in the rat [44] . Evidence for long STDP time scales was recently provided in the visual cortex of mice that had a greater than normal expression of NMDA receptors ( NMDARs ) with NR2B subunits [75] . The STDP rule in these animals was temporally asymmetric , but was sensitive to remarkably long spike time differences greater than 0 . 1 s . NR2B expression is high during the stage of development when retinal waves are important for refinement , which in mice is the first postnatal week [52] . Thereafter , the number of NR2B containing NMDARs reduces , being replaced by faster acting NR2A containing NMDARs [76 , 77] . The change in NMDAR composition may explain why many STDP studies in mice , which are typically conducted in the second or third postnatal week of development , reveal STDP rules with time scales not much longer than 20 ms . We speculate that such developmental changes in NMDAR composition may provide a gradual reduction in STDP time scales that enables greater levels of refinement , as we demonstrated in Fig 8B-8C . The effect of this developmental progression may be complemented by the observed reduction in wave speeds in mice during the same period [43] . Our results describe a potentially important role for these phenomena in refinement , which can be examined experimentally . The noisy mechanism of wave generation in the developing retina adds considerable variability to the size of retinal waves [39] . In our model , it is important that the total distance travelled by a wave exceeds the characteristic wavelength , 1/k* . This is the spatial analog of requiring the wave duration to exceed the time scales of STDP in our analytical derivation . Our calculation above therefore suggests that retinal waves should travel distances greater than 0 . 51 mm in order to produce RFs 10° in diameter in the mouse SC . By tracking the center of mass of experimentally recorded retinal waves , Maccione et al . showed that a substantial majority of retinal waves indeed exceeded this distance throughout the period of major retinocollicular refinement [43] . Taken together , experimentally recorded retinal waves exhibit many of the properties required for retinocollicular refinement by a periodic patterning process . However , whether the mode of synaptic plasticity at retinocollicular synapses is suitable for this process has yet to be determined . Traveling waves occur in many different parts of the brain , and each location will have particular constraints that limit the types of patterns that could form based on the interaction with STPD rules . As discussed above , one constraint is the characteristic wavelength , 1/k* , which sets a lower bound to the distance traveled by waves if periodic patterning is to be achieved . We computed the characteristic wavelengths associated with waves in brain areas other than the retina—including the cortex , cerebellum and hippocampus—and list them in Table 1 , under the assumption of an asymmetric STDP rule like that used in many of the results above ( τ+ = 20 ms ) , and which is reported widely throughout the brain . The speeds of waves in these areas are 1–2 orders of magnitude faster than retinal waves [5 , 8 , 9] , but yield characteristic wavelengths that are similar in scale to that required for RFs in SC because of the short STDP time scale ( first four entries in Table 1 ) . Fast wave speeds and short STDP time scales place these waves in a region near the bottom right hand corner of Fig 3A ( solid red rectangle ) , assuming a burst duration of 0 . 1 s . Also listed in Table 1 are the maximum distances over which the waves in different areas may propagate . Comparing these distances with the characteristic wavelengths reveals that periodic patterning in the efferents of the cerebellum , dorsal cortex and ventral cortex would not be expected , as the maximum possible propagation distance is close to or less than the lower bound set by 1/k* . However , wave properties in the hippocampus , as in the retina , do satisfy the minimum distance constraint for the type of STDP rule assumed . More detailed analyses of waves in the hippocampus and STDP at its efferent synapses is necessary to determine whether these phenomena could drive periodic patterning in a manner useful for development . RFs of simple cells in the primary visual cortex ( V1 ) exhibit weak selectivity for orientated features in the visual environment at the onset of vision [78–80] and become more selective with visual experience [79 , 81] . Yet when an animal is raised in an environment comprising a restricted range of oriented contours , RF selectivity matures for those same orientations but not others [79–84] . A key feature of orientation tuned simple cell RFs is their composition of ON and OFF subfields , which are respectively sensitive to increments and decrements in luminosity , and are thought to reflect the visuotopic organization of ON and OFF inputs from the lateral geniculate nucleus ( LGN ) [85 , 86] . The orientation of these subfields is thought to confer the orientation preference of the simple cell [85 , 86] . However , it is not clear whether oriented features in the environment instruct the development of orientation selectivity , or whether they permit the maturation of RFs that are already selective for the same orientations . The spatial frequency tuning of simple cells may help to resolve this question , as it is well predicted by the periodic structure of ON and OFF subfields [87] . One model of simple cell RF development [57] posits that spatial frequency tuning is determined by the structure of spatial correlations between LGN cells . However , the necessary correlations were not observed in the developing LGN in later experiments [88] . Other models inspired by sparse coding schemes [89] posit that simple cell RFs result from learning the independent components of natural visual scenes [90] , a consequence of which is that RFs would exhibit tuning for velocity and spatial frequency with an inversely proportional relationship [91] . However , despite recent insights [23 , 63] , it is not yet well established how sparse coding is implemented in biological circuits . Our results provide an alternative mechanism for simple cell RF development . We have shown how oriented wavefronts can build orientated RFs ( Fig 9A ) , and how the same process can develop multiple , periodic subfields ( Fig 7 and S4A Fig ) , akin to the ON and OFF subfields of simple cells , thus providing a means for spatial frequency tuning ( Fig 8 ) . Moreover , our prediction of an inverse relationship between the wave speed and spatial frequency of the connectivity pattern provides a novel test for the role of visual experience in shaping simple cell development , and bares striking similarity to the sparse coding schemes mentioned above . Using stimuli that consist of high contrast luminance contours to elicit wave-like activity in the retina and LGN may help to test our predictions , and experimental protocols using such stimuli with young animals are already well established ( for example , see [92] ) . We can use our model to predict a range of plausible STDP parameters that would build RFs with the spatial frequency tuning of simple cells . In adult cats , spatial frequency tuning ranges from ∼0 . 2–2 cycles/° within eccentricities of ±15° [54 , 93] . We further constrain the model by considering wave speeds that match the velocity tuning of cat simple cells ( ∼0 . 5–20°/s , [94] ) . In S6 Fig , we illustrate the spatial frequencies that are obtained in selected parts of this large STDP parameter space . Assuming an asymmetric STDP rule in the cortex [36 , 95] , realistic spatial frequencies can be obtained with STDP time scales ranging ∼1–100ms , and biases for synaptic weakening in the approximate range 0 . 3 ≲ A-/A+ ≲ 0 . 7 . Realistically , STDP need not operate over such a wide range of parameters , as variability in the speeds of natural stimuli should explain most of the variability in spatial frequency tuning . The relationship between temporal correlations and the development of spatially structured connectivity patterns may be extended to other visual RF properties . For example , temporal delays between inputs with spatially offset RFs are the major components needed to build direction selective cells [96 , 97] . This kind of organization , and hence direction selectivity , can be learned with rate-dependent Hebbian plasticity [62] , utilizing diverse response latencies in the LGN [98–100] . It is feasible that STDP should also yield spatial offsets between inputs with different response latencies , given its Hebbian nature . Combined with the capacity for periodic patterning , we speculate that the interaction of traveling waves with STDP could yield connectivity patterns that are direction selective , as well as orientation and spatial frequency tuned . We found that pattern formation in our simulations was sensitive to noise: patterns began to degrade when the ratio of the background spike rate to the wave-induced spike rate became too high . It is therefore essential to consider what is known about background firing rates reported in the literature . We concentrate on background noise in the developing visual system , for which good data are available . During retinal wave activity in mice , retinal ganglion cells rarely spike outside of a burst . However , bursts that occur outside of a wave event have approximately the same firing rate as wave related bursts ( Table 1 in [42] ) . Nevertheless , wave related bursts comprise approximately 90% of all bursts in the developing retina [42] , suggesting that background spiking noise would have little impact on wave-induced correlations at this stage of visual development . Retinal waves also drive bursts in the LGN at firing rates of 10 Hz with long intervening periods of quiescence [101] . In later stages of development , visually evoked responses in the LGN reach firing rates of 10 Hz , whereas 1 Hz firing rates are typical of spontaneous activity [102] . This highly skewed distribution of firing rates is recapitulated in the developing visual cortex [103] . Thus , it is likely that , during early developmental periods when RF properties undergo refinement , spontaneous spikes contribute little to the overall activity in the visual system , and that patterned retinal waves and visual stimulation are propagated throughout the early visual system . The precise dependence of STDP on spike timing can be sensitive to many factors that can modify the shape of the STDP rule ( reviewed extensively in [14] and [58] ) . Whatever the mechanisms that underly these additional interactions , they must preserve a bandpass dependency on spike times in order for periodic patterns to form . Biophysical models of cellular activity that may underlie STDP have been proposed to account for some of these additional interactions [104 , 105] . These models rely on the summed pre- and postsynaptic contributions to the intracellular calcium concentration , the amplitude of which at any given time determines whether synapses are strengthened or weakened . As such , the shape of the STDP rule changes for different spike trains and thus would not adhere to the spike-timing dependence that is necessary for bandpass filtering and pattern formation . If pattern formation as outlined in this paper does influence circuit development , then we predict that STDP must maintain a bandpass profile at synapses that mediate wave-like activity patterns . To test this , a typical STDP experimental protocol could be performed , in which paired spike bursts are stimulated in connected pre- and postsynaptic cells over a range of temporal offsets , and the resulting change in synaptic strength measured . A key requirement of such an experiment would be to extend the temporal offsets well beyond the burst durations , in contrast to previous studies [44 , 106 , 107] , so as to detect the full temporal profile of the plasticity rule and ensure that synaptic changes decay to zero with larger offsets . If the bandpass property of synaptic plasticity is present at these synapses , then it should be reflected in the resulting STDP curve . Retinocollicular or geniculocortical synapses would be ideal substrates for testing this . The robustness of our results to various types of noise suggests that pattern formation should also be achieved in more complex models that incorporate details specific to the circuit . The developing visual system provides a promising arena in which to test our theory of pattern formation for the reasons discussed above , and we outline here future work that will facilitate this investigation . The dominant spatial frequency that characterizes a periodic pattern is strongly influenced by the stimulus response properties of the input neurons , as demonstrated by the effect of the burst duration in Fig 4 . Incorporating accurate space- and time-dependent response properties of neurons in the visual pathway , as determined by their spatiotemporal RFs [55 , 108–111] , is therefore essential to making accurate predictions about pattern formation in downstream visual targets such as the SC and V1 . For theoretical analysis , these features can be easily incorporated by reformulating α ( t ) as α ( x , t ) to take account presynaptic RF structures . Application of our model to simple cell RF development will require modeling both ON and OFF response types in the input layer , which must become spatially segregated as a result of learning . Previous proposals for the mechanism of this segregation typically rely on correlations within each ON or OFF population exceeding those between the two populations [20 , 57 , 88] . The extent to which STDP will be sensitive to ON and OFF correlations , in addition to the strong spatiotemporal correlations induced by traveling waves , requires further investigation . The linearity of our current model , though useful for theoretical analysis , prevents RFs from developing an orientation preference when waves travel in multiple directions ( Fig 9B ) , as is the case for retinal waves and natural visual environments . Any bias towards one orientation is eventually averaged away by the influence of waves with other orientations . The addition of a nonlinearity , either in the sensitivity of STDP to the output firing rate , or in the transfer function for the output firing rate itself , should help to break the symmetry in wave directions and bias the RF towards any asymmetric shape that is built into it by chance . In this way , a population of output neurons might be able to acquire different orientation preferences when exposed to the same input pattern . In a similar fashion , a spectrum of spatial frequency preferences could be acquired throughout the population , rather than the uniform preference for a single spatial frequency , as we observed in simulations in which the wave speed was varied . It is worth noting that traveling waves are just one incarnation of activity patterns that exhibit the space-time inseparable correlations necessary for periodic patterning . Our results may be generalized to any reliable temporal sequence of activity , some examples of which include navigation codes in the hippocampus [112] , pre-motor coding in songbirds [113] , and spontaneous ‘synfire chains’ in the cortex [114] .
We model a reduced feedforward network consisting of a presynaptic layer of input neurons and a single postsynaptic output neuron . Traveling waves of activity traverse the input layer and recruit input neurons , which discharge bursts of spikes and provide excitatory synaptic inputs to the output neuron . Each feedforward synapse is modified according to the time delay between spikes from its corresponding input neuron and spikes from the output neuron . A schematic of the network is provided in Fig 1A . Our model is simulated using a time step of 1 ms . During each step , a number of processes are simulated . First , input neurons generate spikes from one of two wave models described below . Each input spike elicits in the output neuron an excitatory postsynaptic potential ( EPSP ) , which is weighted by the synaptic strength . EPSPs modulate the output membrane potential , which determines the generation of output spikes according to either a linear or nonlinear process , described below . Once all spikes have been generated for a given time step , synaptic strengths are modified by a STDP rule . To measure the speed and size of simulated complex wave patterns , we analyzed a 2000 s segment of RGC spiking activity from a simulation of the slow waves only , as these waves were less compressed in time and therefore lasted longer , enabling a more accurate measure of the wave properties . Spike times were first converted into a firing rate movie , M ( x , y , t ) , with dimensions 64 × 64 × 20000 , where the ( i , j , k ) th bin in M contained the number of spikes fired by a single RGC , at location ( xi , yj ) , during a 100 ms period , tk . During the retinal wave simulation , multiple waves were occasionally present at the same time . Concurrent waves would occasionally collide into or split from each other , but they were mostly well separated in space . We sought to isolate concurrent waves to accurately measure their individual properties , such as speed and size , using a center of mass ( COM ) tracking algorithm , of which a schematic is drawn in S7 Fig . First , we computed a smoothed firing rate movie , Ms ( x , y , t ) = ( M*G ) ( x , y , t ) , where G ( x , y , t ) = e x 2 2 σ x 2 + y 2 2 σ y 2 + t 2 2 σ t 2 is a Gaussian filter with a spatial standard deviation ( SD ) , σx = σy = 34 μm , and temporal SD , σt = 100 ms . After smoothing , pixel values below 10 were set to zero . Thus , a single time bin in Ms might have several domains of non-zero pixels . To track the COM of different waves , a boundary was drawn around each domain of activity in time bin k and the COM within the boundary calculated , where mass refers to the firing rates within the same boundary in M ( x , y , t = k ) . Two domains , colored blue and orange , and their COMs ( purple and green , respectively ) are shown in S7 Fig , and the firing rates given in greyscale . Each domain was assigned a wave identification number ( ID ) but , if the COMs at two domains were less than 680 μm apart , they were given the same ID . If the COM of one domain was within a 680 μm radius from the COM of another domain in the previous time bin ( purple lines extending from the COM in S7 Fig ) , it was given the same ID . In this way , wave COMs were tracked according to their ID . The COM trajectory of each wave was zero-padded and smoothed in time by a Gaussian filter with a SD of 0 . 1 s , and the first two and last two COM locations discarded . Thus , wave speeds and sizes were only computed for waves that lasted for more than 0 . 5 s . We used the path length of a COM trajectory as the distance travelled by a wave , and computed the wave speed from this by dividing it by the time taken to travel that trajectory . RGCs near the edges of the model retina received fewer lateral inputs from SACs and were less active than those in the centre . Within the central 44 × 44 RGCs , the total level of spiking activity was comparably uniform . Accordingly , waves with a time-averaged centre of mass ( COM ) that resided in the outer 10 neurons were discarded from any further analysis . Occasionally , small segments of waves were isolated from larger waves . These were discarded from the analysis by removing waves that covered fewer than 1000 space-time bins . The COM tracking algorithm performed well in separating waves that eventually merged with , or split from , other waves , and allowed us to perform analysis on almost every wave that had a unique ID by removing concurrent waves from that analysis . A total of 797 waves were isolated in the 2000 s segment , of which 779 lasted long enough to compute the wave speed and distance travelled . All analysis was performed with Matlab R2012a using built-in and custom built functions . Concordance of the spiking model with analytical results is verified by numerically integrating Eq 6 using the forward Euler method . Synaptic strengths were initialized in the same way as in the spiking simulation . However , as there was no spiking noise when solving Eq 6 numerically , low amplitude Gaussian white noise was added to the initial synaptic strengths . To further align numerical integration with the simulations , nonlinearities that were present in the simulations were also incorporated in the numerical integration by: 1 ) restricting synaptic strengths to the range [0 , 1]; 2 ) alternating the direction of the wave by replacing the wave speed , v , with ( −1 ) n v for the nth iteration; 3 ) applying the arbor function in Eq 23 when the initial condition of w ( x ) supported a RF structure . To prevent numerical solutions from becoming chaotic , the learning rate , η , was varied depending on the values of τ+ and v . To measure the robustness , Ψw , with which a periodic structure emerged in a connectivity pattern , we computed the discrete power spectrum , P ( ki ) , of the connectivity pattern , with the DC component removed , and determined the ratio of the power at the dominant spatial frequency to the total power: Ψw=P ( ki* ) ∑i=−kNkNP ( ki ) , ( 29 ) where kN is the Nyquist limit . As a predictor of robustness in the connectivity patterns , a theoretical robustness measure , Ψκ , was computed for the power spectrum of the real component of κ ~ ( k ) : ∣ Re [ κ ~ ( k ) ] ∣ 2 , which was obtained by first numerically computing κ ( x ) with a spatial resolution of 0 . 6 μm . The theoretical robustness was therefore: Ψκ=|Re[ κ˜ ( ki* ) ]|2∑i=−kNkN| Re[ κ˜ ( ki ) ] |2 . ( 30 ) To set up simulations of RF refinement , we use data from experimental studies regarding receptive fields , or traveling wave speeds , for example , which require conversion between degrees of visual angle and units of distance along the retinal surface . The precise conversion relationship between visual angle and retinal distance depends on the geometry of the eye , which differs for different animals . We therefore use a general conversion factor to provide a useful estimate . We assume that the retina covers two thirds of the circumference of a horizontal section through the center of the eye [116] , and that this maps to 180° of visual angle . This means that distance in millimeters , R , corresponds to visual angle , A , according to R = πrA/135 , where r is the radius of the eyeball . We used the following values for the radii of eyes in different animals at different stages of development: mouse , 1–7 days old: r ≈ 1 . 1 mm ( Table 1 in [117] ) ; mouse , adult: r ≈ 1 . 7 mm ( Table 1 in [117] ) ; ferret , eye opening: r ≈ 2 . 8 mm ( Fig . 8 in [118] ) ; cat , 4 weeks old: r ≈ 5 mm ( Fig . 3 in [119] ) .
|
In several areas of the developing brain , waves of electrical activity trace out distinct patterns across the nervous tissue . These waves are intricately involved in developmental processes that set up the structural connections of the adult brain , but it is unclear what role the wave patterns play . Here , we examine how the strength of connections in these brain areas may change by a process called spike-timing dependent plasticity , which is sensitive to the precise times at which individual neurons become electrically active . We use mathematical models and simulations to show that interactions between waves and plasticity build highly structured patterns into the connections . The results of our model are analogous to many cases of biological pattern formation seen , for example , in zebra stripes , leopard spots and seashells . An important connectivity pattern we consider is the receptive field , which determines to a large extent the specific function of a neuron . We demonstrate how pattern formation can refine the shape of a receptive field and therefore the specificity of a neuron , and explore several ways in which pattern formation may be disrupted , providing clues regarding pathologies in receptive field development . Our theory makes several predictions that may be tested using existing experimental paradigms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Refinement and Pattern Formation in Neural Circuits by the Interaction of Traveling Waves with Spike-Timing Dependent Plasticity
|
Here we report on the identification and functional characterization of the ADAMTS-like homolog lonely heart ( loh ) in Drosophila melanogaster . Loh displays all hallmarks of ADAMTSL proteins including several thrombospondin type 1 repeats ( TSR1 ) , and acts in concert with the collagen Pericardin ( Prc ) . Loss of either loh or prc causes progressive cardiac damage peaking in the abolishment of heart function . We show that both proteins are integral components of the cardiac ECM mediating cellular adhesion between the cardiac tube and the pericardial cells . Loss of ECM integrity leads to an altered myo-fibrillar organization in cardiac cells massively influencing heart beat pattern . We show evidence that Loh acts as a secreted receptor for Prc and works as a crucial determinant to allow the formation of a cell and tissue specific ECM , while it does not influence the accumulation of other matrix proteins like Nidogen or Perlecan . Our findings demonstrate that the function of ADAMTS-like proteins is conserved throughout evolution and reveal a previously unknown interaction of these proteins with collagens .
The establishment and maintenance of extracellular matrices ( ECM ) are important tasks to allow proper organ function in metazoans . Among other factors , changes in ECM composition , turnover and homeostasis are crucial mediators of human cardiovascular disease leading to life threatening conditions and premature death . The ECM allows cells to resist mechanical forces , protects complex tissues from being damaged and promotes specific physical properties like elasticity or stiffness in order to maintain organ functionality . While the composition of the ECM is very complex and extremely variable the basic structural constituents can be grouped as collagens , glycoproteins and proteoglycans , which are highly conserved throughout metazoan species [1] . Consequently , defects in ECM proteins or matrix composition cause major developmental defects and strongly contribute to prevalent human disease like fibroses or cancer [2] . During the last years fibrotic disease and mutations in various ECM proteins were correlated to cardiovascular disease . For example mutations in human Col4a1 cause the weakening of the major vasculature leading to life threatening aneurysms or stroke [3] while mutations in murine Col4a1 and Col4a2 induce vascular defects causing internal bleedings and prenatal lethality [4] . Even more recently ADAMTS-like ( ADAMTSL , A Disintegrin and Metalloprotease with Thrombospondin repeats ) proteins have gained significant importance in the understanding of certain types of fibrillinopathies [5] , [6] . Mutations in human ADAMTSL4 were identified in patients suffering from isolated ectopia lentis ( EL ) , a recessive disorder of the occular lense [7] , [8] and , more severely , aberrations in ADAMTSL2 cause geleophysic dysplasia a syndrome which , amongst others , manifests in the thickening of the vascular valves and progressive cardiac failure causing premature death [9] . Unfortunately , despite the pathological mutations no ADAMTSL alleles in genetically treatable model systems were described so far . In the present study we use Drosophila melanogaster as a model of ECM function in the cardiac system . In Drosophila the maintenance of cardiac integrity is of great importance , since no mechanisms of cardiac cell replacement or tissue repair exist . A variety of mutations in ECM genes have been analyzed with respect to their function in different tissues and processes like neurogenesis , muscle attachment , wing development and others [10]–[12] . Cardiogenesis in the fly embryo depends on several ECM components including the evolutionarily conserved toolkit of proteins forming the basement membrane . The basement membrane constitutes a specialized type of ECM consisting of Laminins , Collagen IV , Perlecan and Nidogen found at the basal side of epithelial cells [13] . The interaction of laminins with cellular receptors like integrins or dystroglycan and its self-assembly into a higher meshwork forms the initial step of basement membrane formation in animals [14] , [15] . Consequently , mutations in any of the four laminin encoding genes in Drosophila lead to severe embryonic cardiac defects . For example loss of lanB1 , encoding the only β-subunit of the laminin trimer , prevents the accumulation of collagen IV and perlecan towards cardiac cells , while mutations in lanA and lanB2 ( encoding the α3 , 5-subunit and the γ-subunit , respectively ) cause the detachment of pericardial cells , a specific type of nephrocytes in arthropods , from the heart tube [14] , [16]–[18] . The highly abundant proteins forming the basement membrane have in common that they are distributed ubiquitously and cover all internal organs of the fly [14] , [19] . Compared to that the cardiac ECM is unique , since it contains the collagen Pericardin ( Prc ) , which is rather specifically decorating the heart tube [20] , [21] . Prc displays certain homologies to mammalian collagen IV and was shown to be crucial for heart morphogenesis and cardiac cell to pericardial cell adhesion [20] , [22] . However , the question of how Prc accumulates in a cell specific manner in the fly embryo or how specific matrices are specified in the rather open body cavity of insects in general was not addressed in detail so far . Here we introduce the gene lonely heart ( loh ) , which is crucial to maintain cardiac integrity during postembryonic developmental stages . We show that Loh is a member of the ADAMTSL protein family and constitutes the essential mediator of Prc accumulation and matrix formation already in embryonic cardiac tissue . ADAMTSL proteins belong to the evolutionary conserved family of ADAM proteases with the exception that these proteins lack a proteolytically active domain in their primary sequence and therefore its function is unclear [5] , [6] . We found evidence that Loh is sufficient to specifically recruit Prc to the ECM of different tissues indicating that Loh regulates the assembly of tissue and organ specific matrices . This is of great interest since the composition of the ECM determines its mechanical properties crucial for correct organ function and cellular behavior [23] . We also address the physiological relevance of cardiac integrity and show that lack of either loh or prc prevents proper blood circulation in the animals and cause a reduction of the fly's life span . The findings presented in here demonstrate that mutations in ADAMTSL proteins lead , like in human disease , to progressive heart failure and premature death in flies , strongly arguing for an evolutionary conserved function .
In order to identify novel mediators of cardiac function we screened a set of pupal lethal EMS induced mutants , known as the Zuker collection , for the presence of postembryonic cardiac malformations [24] . To mark all cells contributing to the mature heart we introduced the previously described handC-GFP reporter into each individual mutant strain [25] . We identified a single allele , lonely heart ( loh1 ) , showing a strong detachment of pericardial cells from the heart tube during larval stages ( Figure 1 and Figure S1A , B ) . To map the mutation to the genome we introduced the loh1 allele to a collection of genomic deficiencies and assayed the progeny for the presence of the pericardial cell detachment phenotype . The allele failed to complement the deficiencies Df ( 2L ) Exel7048 , Df ( 2L ) BSC453 and Df ( 2L ) BSC144 but complements Df ( 2L ) BSC209 ( Figure S1C–F ) . This allowed us to narrow down the location of the mutation to a 14 kb genomic region at band 31E3-4 containing three open reading frames ( Figure S1I ) . Since EMS is known to promote secondary hits on the same chromosome we decided to assay existing alleles of these three genes for the presence of the pericardial cell detachment phenotype . We were able to identify two alleles , MB05750 and MI02765 , that are allelic to loh1 and Df ( 2L ) Exel7048 and produce the heart phenotype in transheterozygous condition ( Figure 1A–D and Figure S1G , H ) . Both mutations were induced by the insertion of minos elements within the locus of the previously uncharacterized gene CG6232 [26] , [27] . Based on sequence predictions CG6232 encodes an ADAMTS-like ( A Disintegrin and Metalloproteinase with Thrombospondin repeats ) protein , containing several Thrombospondin type 1 repeats , a central ADAM-spacer domain and a C-terminal Protease and Lacunin ( PLAC ) domain ( Figure S1J ) . The primary sequence of Loh/CG6232 shows high homologies to mammalian ADAMTSL6 , known to promote the formation of fibrillar matrices in mice [28] . During a parallel reverse genetic approach we also tested transposon induced alleles affecting known ECM genes for the appearance of late cardiac defects . We identified the allele MB03017 carrying a minos element in the pericardin ( prc ) locus . Homozygous prcMB03017 and transheterozygous prcMB03017/Df ( 3L ) vin6 animals display a strong pericardial cell detachment phenotype similar the loh phenotype ( Figure 1E , F and Figure 2E ) . The Prc protein constitutes a rather heart specific collagen , which shows homologies to vertebrate collagen IV [22] . Previous studies implicated Prc to be involved in dorsal closure as well as cardiogenesis [20] . However , no gene specific mutant was available so far . To investigate the adhesion defects arising in both loh and prc mutants in more detail we analyzed the morphology of the heart at different developmental stages . During embryogenesis the heart tube arises from two bilateral primordia and forms a simple tube at the dorsal midline . Determination and migration of heart precursor cells is not affected in either lohMB05750/Df ( 2L ) Exel7048 or prcMB03017/prcMB03017 mutant animals ( Figure 2A , D and G ) . During larval development the pericardial cells irreversibly detach from the heart tube with the phenotype becoming fully visible in third instar larvae ( Figure 2B , E and H ) . The loss of cardiac integrity in both mutants does not constrain the development into adult animals and we could detect the pericardial cell detachment phenotype in pharate adult animals , which further develop into viable and fertile flies ( Figure 2C , F and I ) . These findings show that the phenotype arises progressively during development and indicate that proper heart function is not essential for development into the imago . Of note the alleles loh1 and lohMI02765 cause larval lethality in homozygous condition , while the alleles are viable in transheterozygous combination indicating second site mutations or yet unknown dominant effects of the mutated proteins . Since lohMB05750 and prcMB03017 animals are homozygous viable and show the pericardial cell detachment phenotype all experiments predominantly focus on these two alleles . Postembryonic pericardial cells are enclosed by a dense network of Prc fibers and connected to the alary muscles ( Figure 2J–K ) . Since the heart tube and the alary muscles are not connected via direct cell-to-cell contacts this Prc network is likely to be a fundamental structural component to suspend the heart to the body cavity [29] . To evaluate the adhesion of the heart tube to the alary muscles in more detail we stained transheterozygous lohMB05750/Df ( 2L ) Exel7048 and prcMB03017/Df ( 3L ) vin6 larvae for F-actin and βPS integrin ( Figure 2L–N ) . The detachment of pericardial cells also ruptures the connection between the alary muscles and cardiomyocytes demonstrating that the lack of pericardial cell adhesion consequently lead to a breakdown of the heart's suspension towards the epidermis . Furthermore , the morphology of the cardiomyocytes itself is dramatically altered in lohMB05750/Df ( 2L ) Exel7048 and prcMB03017/Df ( 3L ) vin6 mutants ( Figure 2O–Q ) . While in the wild type cardiomyocytes show a defined arrangement of F-actin fibers in a circular fashion mutant cells exhibit an uncoordinated distribution of actin fibers and an altered cell shape . Since the arrangement of actin fibers might be a secondary effect of a changed cardiac cell polarity we stained mutant embryos for the polarity markers FasIII and αSpectrin ( Figure S2A–L ) . Neither loh nor prc mutant hearts displayed changes in cell polarity proving that the changed actin arrangement is an effect of the defective cellular adhesion . We next elucidated how heart beat is influenced in the mutants . For this purpose the beating pattern of the heart was recorded in semi-dissected third instar larvae ( Movies S1 , S2 , S3 ) [30] . Wild type heart beat follows a very regular pattern and the heart walls display systolic and diastolic movements ( Movie S1 ) . Compared to that the beating pattern in lohMB05750/Df ( 2L ) Exel7048 and prcMB03017/Df ( 3L ) vin6 mutant larvae is dramatically altered . The disorganized actin fibers cause a changed contraction movement of the whole organ along the posterior-anterior axis ( Movie S2 and Movie S3 ) . In addition no systole and diastole are detectable already indicating that the pumping performance of the organ is altered . To evaluate whether the disruption of heart architecture and the changed beating pattern impairs heart functionality we analyzed the capability of mutant hearts to provide circulatory activity . To visualize the hemolymph flow by dye angiography we injected a fluorescent tracer into the abdomen of adult animals shortly before eclosion ( pharate adults ) and semi-quantified the pumping capacity of the dorsal vessel by measuring the tracer accumulation within the head ( Figure 3A–C ) [31] . To verify the reliability of the technique a control strain that does not display any cardiac defects was tested and showed a strong accumulation of the tracer in the head ( Figure 3B–C and Movie S4 ) . In contrast homozygous prcMB03017 and lohMB05750 mutant animals displayed a dramatic reduction or total absence of dye accumulation within the examination time , which proves that the observed disruption of heart integrity directly influences the ability to promote circulatory activity ( Figure 3C ) . Since it is known that heart failure can cause a significant reduction of Drosophila's life span [32] , [33] we tested whether the isolated alleles show a direct effect on adult survival . As a wild type control we used the white1118 strain , because this genotype resembles the genetic background of both minos insertion strains . Wild type flies ( white1118 ) revealed an average life time of 46 days , while the mean life span of homozygous lohMB05750 and prcMB03017 animals was decreased by 26% ( 34 days ) or 46% ( 25 days ) , respectively ( Figure S3 ) . This strongly argues that impaired cardiac function in the mutants reduces the survival of the animals . We investigated the temporal expression pattern of loh and prc by developmental Northern blots . The loh locus encodes two transcripts - a longer isoform A ( 3081 bp predicted ) and a shorter isoform C ( 2131 bp predicted ) ( Figure 4A ) . While isoform A constitutes the major transcript during embryogenesis , isoform C becomes additionally expressed during the first and second larval stage ( L1 and L2 ) . Later on expression declines and becomes weakly re-activated during pupal and adult stages . Compared to loh the temporal expression profile of prc was found to be remarkably similar ( Figure 4A ) . A single transcript ( 5535 bp predicted ) becomes expressed from the embryo to L2 and declines in L3 . During metamorphosis expression re-initiates and lasts until adulthood . In order to reveal if both loh isoforms are essentially needed to ensure proper heart integrity we expressed two independent gene specific hairpins either effecting only isoform A ( loh-IRNIG6232-2 ) or both isoforms ( loh-IRVDRC31020 ) under the control of handC-Gal4 to knock down the gene's expression ( Figure 4B ) . Expression of both hairpins causes a pericardial cell detachment phenotype . However , since expression of the loh-IRNIG6232-2 hairpin , which only targets isoform A , resulted in a detachment phenotype ( Figure S4 ) we concluded that isoform A constitutes the relevant one for the observed adhesion defect . To investigate the effect of the isolated mutations on the expression level we analyzed the total protein amounts by immunoblotting ( Figure 4D , E ) . Therefore we raised a specific peptide antibody recognizing both Loh isoforms . In embryonic extracts the antibody detects a single protein band corresponding to isoform A . The band runs slightly higher compared to the predicted molecular mass of 100 kDa , most likely due to posttranslational modifications ( Figure 4D ) . The protein is absent from extracts of homozygous Df ( 2L ) Exel7048 embryos proving the specificity of the antibody . Significantly , the protein is also undetectable in extracts of homozygous lohMB05750 embryos . RT-PCR analysis proved that lohA transcripts are severely reduced but not absent in these animals ( Figure S4A ) , obviously leading to massively decreased protein levels . Similarly , Prc protein could be detected in extracts of different developmental stages in the control , but is absent from homozygous mutants ( Figure 4E ) . Given the similar phenotypes of the mutants we sought to analyze the spatial expression pattern of both genes . Transcripts of loh and prc can be detected from embryonic stage 13 onwards until the end of embryogenesis in cardioblast and pericardial cell precursors ( Figure 5A–F ) , where loh seems to be more prominently expressed in the ventricle of late stage embryos ( Figure 5C ) . Additionally , loh transcripts were detected in the chordotonal organs , while prc is expressed by the oenocytes . Since it is known that prc is only expressed by a subset of cardiac cells we analyzed the expression of loh mRNA in combination with the cardiac cell markers Tinman and odd skipped-lacZ [20] , [34] , [35] . loh transcripts are expressed by both cell types demonstrating that most cardioblasts and pericardial cells contribute to the gene's expression ( Figure 5G , H ) . As previously reported , Prc protein distributes predominantly along the basal side of the cardiomyocytes where it co-localizes with the collagen IV fusion protein Vkg::GFP ( Figure 5I ) [20] , [36] . Strikingly , Loh co-localizes with Vkg::GFP as well as Prc , demonstrating that it constitutes an integral part of the basal cardiac ECM ( Figure 5J–K ) . The detected signal was considered to be specific since it follows the observed mRNA pattern and is undetectable in homozygous Df ( 2L ) Exel7048 embryos ( Figure S5A–C ) . The expression of Loh and Prc supports a function in mediating the adhesion between pericardial cells and cardiomyocytes in the mature heart , while the observed co-localization throughout the whole embryo indicates a cooperative function ( Figure 5L ) . The data presented so far pointed us to the question if Loh and Prc act cooperatively in the cardiac ECM . To test if the proteins affect each other we analyzed the localization of Prc in loh mutant background and vice versa ( Figure 6A–O ) . In homozygous lohMB05750 , lohMI02765 and loh1 embryos Prc becomes normally secreted but strikingly fails to assemble properly in between the pericardial cells and the heart ( Figure 6A–F and Figure S6A , B ) . While in the wild type Prc organizes into a proteogenic sheet at the basal side of the cardiomyocytes this regular distribution is completely disrupted in loh mutant embryos ( Figure 6A–F ) . We also tested whether impaired loh expression affects other ECM proteins like Laminin , Nidogen or Perlecan ( Figure 6B , E and Figure S6F , G and I , J ) . The expression and distribution of all tested proteins was unchanged in loh mutant animals indicating that Loh specifically regulates the correct accumulation of Prc but is not needed for ECM formation in general . The other way around the lack of Prc in homozygous prcMB03017 embryos does not affect the localization of Loh ( Figure 6J–O ) or any other tested ECM protein demonstrating that the function of both proteins is not mutual ( Figure 6G–N and Figure S6H , K ) . To prove that the phenotypes in lohMB05750 and prcMB03017 definitely arise from the inserted transposons we generated revertants by precise excision of the minos elements [26] , which was verified by PCR and subsequent sequencing ( Figure S6C–E ) . The precise remobilization of both transposons lead to a restored Prc expression and distribution in both revertants demonstrating that the mutations are gene specific . To study the effect of loh and prc mutants on heart cell morphology in more detail we investigated TEM cross sections of wild type and homozygous lohMB05750 and prcMB03017 embryos ( Figure 6P–V and Figure S6L–N ) . Like in wild type the cardiomyocytes are localized along the dorsal midline at the end of embryogenesis in both mutants showing that dorsal closure is not affected ( Figure S6L–N ) . However , frequently the cardiomyocytes in homozygous prcMB03017 mutants fail to seal the lumen properly at the ventral side of the heart tube ( Figure S6N ) . Staining against the ligand Slit , which is involved in heart lumen formation did not reveal any changes in its distribution indicating that the Slit/Robo signaling cascade is not affected ( Figure S6O–Q ) [37] . Most importantly , the luminal and basal membranes of the cardiomyocytes are covered by a distinct basement membrane in both homozygous mutants supporting the immunocytochemical data ( Figure 6P–R ) . Measuring its thickness does not reveal any significant changes ( Figure 6S ) . However , even if the pericardial cells are not fully detached from the embryonic heart , small gaps between the cells and rupture of the connecting ECM are detectable ( Figure 6T–V ) . Taken together these data demonstrate that Prc and Loh are essential to maintain pericardial cell to cardiomyocyte adhesion and heart integrity but are not involved in ECM formation in general . Hypothetically the open circulatory system of insects would allow ECM proteins to be expressed by a certain cell type , then be distributed over the blood flow and finally become recruited by specific receptors expressed on the target cells . The embryonic expression pattern of loh and prc argue that both proteins are primarily produced locally by heart cells and become secreted into the cardiac ECM . To analyze the expression of prc during later stages we used the previously described prc-Gal4 driver to express GFP and found that it exactly mimics the expression pattern of prc in the embryo ( Figure 7A ) [20] . Upon larval hatching the driver becomes strongly activated in the fat body ( Figure 7B ) raising the question , whether the reporter mimics the endogenous prc expression . To test if Prc becomes produced by adipocytes we trapped the protein by inhibiting the protein secretion machinery of the cell by knocking down the expression of the small GTPase Sar1 , which is essential for the establishment of COPII coated vesicles and protein secretion ( Figure 7C , D ) [38] . Compared to wild type , adipocytes of prc>sar1-IR first instar larvae displayed a strong accumulation of intracellular Prc protein unambiguously demonstrating that it becomes expressed by the larval fat body . To estimate the contribution of fat body derived Prc to the total amount of the protein made , we knocked down prc expression either in heart cells alone ( handC-Gal4 ) or in both heart and fat body ( prc-Gal4 ) and detected the protein by immunoblotting ( Figure 7E ) . The specificity of the knock down was ensured by the use of two independent hairpins ( Figure 3C ) . Prc levels are not markedly changed in handC>prc-IR third instar larvae , while the protein is nearly undetectable in extracts of prc>prc-IR animals illustrating that most of the larval Prc protein becomes secreted by adipocytes . Finally , the pericardial cell detachment phenotype could be induced by knocking down prc expression using both drivers ( Figure S7 ) . However , the penetrance of the induced pericardial cell detachment phenotype is strikingly higher if the knock down was mediated via prc-Gal4 ( Figure 7F ) , showing that the protein secreted from adipocytes indeed contributes to pericardial cell adhesion . From these experiments we conclude that the major source of Prc in larvae is non-cardiac tissue . Nevertheless , locally produced Prc contributes to proper heart integrity , since heart specific knock down of Prc expression does induce the detachment phenotype as well . Taken together these experiments prove a developmental switch in Prc expression with embryonic Prc being locally produced by cardiac cells and during later stages becoming mainly secreted by the fat body ( Figure 7G ) . Furthermore , the integration of fat body derived Prc into the cardiac ECM is essential to promote organ integrity . Although Prc is produced by adipocytes , the protein is not incorporated into the ECM of the fat body indicating that these cells lack specific adhesion properties for Prc ( Figure 8A ) . We found that in third instar larvae the protein almost exclusively accumulates around tissues that initially expressed loh during embryogenesis , but is nearly absent from other mesodermal tissues . From these observations we concluded that Loh might act as a mediator or receptor of Prc matrix formation in Drosophila . To test if Loh is indeed sufficient to induce the formation of Prc matrices we expressed the protein ectopically either in adipocytes or myocytes by using prc-Gal4 or mef2-Gal4 , respectively . Even if some sole Prc fibers can be found along both cell types these organs are not naturally covered by a Prc matrix ( Figure 8A ) . Ectopically expressed LohA protein becomes secreted from both cell types and localizes around the cells ( Figure 8C ) . The protein is retained at the cell surface of adipocytes or myocytes indicating proper localization in the ECM . Upon expression in the fat body , LohA distributes along the whole organ showing a higher accumulation at cellular contacts . Similarly , LohA ectopically expressed by myocytes distributes along the whole myotube with higher accumulation at the muscle tendons ( Figure 8C , inset ) . Most importantly , we found that LohA expression strongly induces the formation of an ectopic proteogenic Prc network around both cell types ( Figure 8C ) . Adipocytes and myocytes ectopically expressing LohA are tightly covered by Prc fibers , which are interconnected to each other and form a dense meshwork . Immunoblot analysis on whole extracts revealed that the overall amount of Prc was not changed in these animals ( Figure S8A ) , demonstrating that ectopic LohA expression leads to a re-direction of Prc protein . To evaluate if Loh acts within the ECM we ectopically expressed a secretion defective version of the protein ( Figure 8B ) , lacking the N-terminal signal peptide . The mutated protein localizes to the nuclei of the cells and fails to recruit Prc to the target matrix demonstrating that LohA has to be secreted in order to act as an initiating factor of Prc matrix formation . To evaluate if both proteins co-localize in such artificial matrices we counterstained dissected prc>LohA third instar larvae for Loh and Prc ( Figure 8D , E ) . High resolution images of dissected fat bodies showed that ectopic LohA distributes as a very faint network at the surface of adipocytes and clusters in a pointy fashion along the cell contacts ( Figure 8D ) but does not completely co-localize with the recruited Prc fibers . Single slices and optical cross sections further demonstrate that Loh co-localizes with Prc at the anchoring points of the Prc network ( Figure 8E ) , indicating that Loh might connect the root of each Prc fiber to the cell surface . Eventually , co-immunoprecipitation experiments using protein extracts isolated from prc>LohA adults proved a either direct or indirect biochemical interaction of both proteins ( Figure 8F ) . In the respective experiments Prc co-precipitated if Loh was pulled down and vice versa . Based on these findings we hypothesize that Loh acts as a linker protein allowing Prc to interact with the cell surface , and wondered if Loh co-localizes with specific cell surface receptors . We found that LohA co-localizes to βPS integrin in adipocytes of prc>LohA third instar larvae ( Figure S8B ) tending us to speculate that LohA binds to integrin receptors , which has to be proven by further experiments . In summary we found that LohA is a crucial and sufficient mediator of Prc matrix formation , very likely acting by interconnecting Prc with the cell's ECM .
In this study we demonstrate that the Drosophila ADAMTSL protein Loh constitutes an unique protein of the cardiac ECM , essentially mediating cell adhesion and matrix formation . Loh is the first protein of its family identified and characterized in depth in flies . We isolated three independent alleles of the gene , all displaying the very same phenotype - the detachment of pericardial cells from the contracting heart tube during larval stages . Thus , the gene loh constitutes a novel and essential mediator of heart cell adhesion and cardiac function . Surprisingly , impaired heart function does not hamper proper development into adult animals but significantly reduces life span . This might be explained by the fact that oxygen transport and blood flow is uncoupled in insects and therefore a reduced hemolymph circulation might not immediately result in cytotoxicity . Furthermore , the open body cavity of the larvae might also allow a distribution of hemolymph independently of a pumping organ supporting the finding that larvae seem not to achieve any drawbacks by the loss of heart function . Based on the primary sequence the domain architecture of Loh is extremely similar to that of vertebrate ADAMTSL6 and is likely to be its ortholog . Furthermore , ADAMTSL6 is the only protein of this family known to produce two transcriptional isoforms from one gene locus . In contrast to Loh the shorter ADAMTSL6 isoform was found to be functional in organizing the ECM in mice [28] . Our data demonstrate that LohA , the larger protein , is functional and sufficient to mediate matrix formation in Drosophila while the role of the shorter isoform C remains elusive by now . However , since the lohC transcript is not expressed during embryogenesis , the critical time window of loh function , we exclude any role of LohC in mediating cardiac ECM formation . By testing different ECM proteins we demonstrated that Prc , a collagen with a very restricted distribution in the animal , is particularly affected in all isolated loh mutant alleles , emphasizing the specific function of Loh to promote Prc matrix formation . Consequently , we isolated the first prc mutant allele , which phenocopies the cardiac defects found in loh mutant strains . In loh mutant animals Prc mislocalizes along the heart already during embryogenesis , leading to a progressive loss of tissue integrity , which eventually causes the observed collapse of the heart tube and an abolishment of cardiac activity . The main function of both proteins is therefore the mediation of cellular adhesion between the heart , the pericardial cells and the alary muscles which further connect the whole organ system to the body cavity . In addition to the cell adhesion defects we also found that the process of heart lumen formation was impaired in prc but not loh mutants . Since we have not followed up the details of this phenotype the role of Prc in lumen formation remains elusive for now . However , the data implicates that the presence of Prc is critical to allow cardioblasts to seal the lumen correctly , while the correct localization of Prc into the matrix seems not to be essential for this process . Analyzing the embryonic and larval expression patterns of loh and prc revealed that both genes are predominantly active during the growing stages of the animal and become deactivated after the heart has grown to its final size . In the embryo , both genes are transcribed in either the same or very proximate cells indicating that the proteins are not distributed over longer distances once they are secreted . Importantly , the final localization of Prc therefore mainly follows the expression of loh . This can be seen best in the oenocytes of the embryo , where Prc becomes secreted but later on mainly localizes to the overlying chordotonal organs that in turn express loh . Thus , loh expression is a prerequisite for the successful establishment of a Prc matrix . This local protein distribution changes during larval stages . As demonstrated by an inhibited secretion in adipocytes of prc>sar1-IR animals , Prc becomes strongly expressed by the fat body during early larval stages . Hence , the protein becomes distributed over longer distances in the larva but still decorates organs and tissues that initially expressed loh . Based on these data , we provide a conceptual model ( Figure 7G ) in which Loh predetermines the ECM to allow Prc to become coupled to the cell surface and to be organized into a reticular matrix . Previously it was shown that Collagen IV , the major collagen in the basement membrane , becomes also secreted by adipocytes and distributes through the hemolymph [39] . We can now prove that Prc as a second collagen is also synthesized by the larval fat body , which enhances the importance of this organ for ECM biogenesis . The developmental change in prc expression might therefore be explained by the ongoing differentiation of pericardial cells into mature nephrocytes during larval stages . While embryonic pericardial cells are able to secrete large amounts of protein into the extracellular space , the major function of pericardial nephrocytes is endocytosis [40] , thus requiring adipocytes to take over Prc production . Finally our results show that the cardiac matrix is maintained during larval growing phases presumably by the consecutive incorporation of fat body derived Prc . The ectopic expression of Loh showed that the secreted protein is readily incorporated into different matrices raising the question how Loh itself interacts with the ECM in general . At the moment it is not fully understood if ADAMTSL proteins interact with miscellaneous ECM components or require specific cell surface receptors . Based on the spatial proximity of Loh to βPS integrin we speculate that Loh may interact with integrin receptors and link these to Prc bundles , thereby promoting the connection of the Prc network to the cell surface . This idea is supported by the observed changes in fiber orientation of mutant cardiomyocytes . Since it is known that integrins are connected to the underlying Z-disks of muscle cells by a structure called the costamere [41] we propose that lack of integrin-ECM binding induces the redistribution of myofibrils . However , there is no evidence of an interaction between ADAMTSL proteins and integrins or any other cellular receptor so far . Nevertheless , in such a model Loh would allow the specific binding of specialized ECM molecules to only some unique matrices . Since Drosophila possesses only two β integrin subunits the number of α/β-dimers is limited and the use of Loh as an adapter molecule increases the diversity of matrix composition and opens up the possibility to create sub-functional matrices . Furthermore , integrin mediated binding seems to influence the correct assembly of Prc since previous findings already showed that lack of αPS3- or βPS integrin can interfere with the distribution of Prc and induce pericardial cell detachment phenotypes [42] . In addition to a receptor mediated ECM incorporation of Loh , binding might also be achieved by some or all of the five TSR1 domains found in the primary sequence of the protein . Previously it was demonstrated that ADAMTS ( L ) proteins can bind to the ECM via the various TSR1 motifs that interact with glycosaminoglycans [43] . This would not need special receptors and allow Loh to incorporate into any matrix . The cell specific expression of loh would then mainly decide which matrix will incorporate Prc and this would in turn strongly depend on the cis-regulation of the gene's expression . On the molecular level we propose that Loh basically acts as a linker protein . Based on the ectopic expression of Loh and the co-immunoprecipitation experiments we can demonstrate that Loh and Prc interact in vivo . In our hands Loh behaves like a secreted receptor molecule that specifically recruits Prc to the cell surface . Our findings indicate that the main molecular function might therefore be binding , but does not exclude additional functions of the protein . It was suggested previously that ADAMTSL proteins act as regulators of extracellular proteases and thereby regulate ECM content and composition [6] . For example it was demonstrated that Drosophila Papilin , another member of ADAMTSL related proteins , is sufficient to inhibit a vertebrate procollagen proteinase in vitro [44] . Thus , it is possible that also Loh regulates a so far unknown proteinase that renders the matrix unsuitable for the accumulation of Prc in some way . In such a model the activity of Loh would then influence the pre-existing microenvironment around a cell to allow Prc to assemble into a network . However , there is no evidence for such a function or the involvement of proteinases so far . The observed roles of Loh in Drosophila partially reflect the function of ADAMTSL proteins in vertebrates , which were shown to organize Fibrillin-1 ( FBN1 ) microfibrils in specialized matrices . Genetic and biochemical analyses showed that ADAMTSL4 and ADAMSTL6 are sufficient to mediate the formation of FBN1 fibrils in cultured fibroblasts as well as in vivo [28] , [45] . ADAMTSL4 acts as a FBN1 binding protein that mediates microfibril assembly in the zonule fibers of the human eye leading to isolated ectopia lentis ( IEL ) if mutated . Thus , IEL is caused predominantly by altered mechanical properties of the zonular fibers leading to a progressive dislocation of the lens [45] . In Drosophila , where no FBN1 homolog exists , Loh interacts with Prc and mediates its distribution within the ECM in a very similar manner . Therefore , the correct assembly of Prc between the pericardial cells and the heart tube could promote the mechanical properties needed to sustain the permanent mechanical forces during heartbeat . The clinical phenotypes of geleophysic dysplasia ( GD ) observed in ADAMTSL2 mutant patients exceed a function of simply promoting mechanical stability of the ECM . It was shown that ADAMTSL2 binds to FBN1 but also interacts with LTBP1 , a regulator of TGFβ signaling , and therefore the phenotypes of GD also include growing defects , muscular hypertrophy and thickening of the skin [9] . None of these additional phenotypes were observed in Drosophila loh mutants . Therefore , it is obvious that ADAMTSL proteins developed novel functions during evolution making them essential mediators of ECM development and homeostasis . So far there are no reports of interactions between any ADAMTSL proteins with collagens but the obviously similar functions in flies and vertebrates strongly argue for a conserved function in organizing fibrillar matrix proteins .
Flies were kept under standard conditions at 25°C on cornmeal agar . The following fly stocks were obtained from the Bloomington stock center: w1118; Mi ( ET1 ) prcMB03017/TM6c , Sb1 , w1118; Mi ( ET1 ) lohMB05750 , y1 , w1118; Mi ( MIC ) lohMI02765/SM6a , Df ( 2L ) Exel7048/CyO , Df ( 3L ) vin6/TM3 , Sb1 , Ser1 , w1118; Sco/SM6a , P{hsILMiT}2 . 4 , w1118; UAS-eGFP and balancer stocks KrIf-1/CyO , Kr>GFP and Dr1/TM3 , Kr>GFP . Further fly stocks used are: handC-GFP and handC-Gal4 [25] , oddrk111 ( odd-lacZ ) ( C . Rauskolb ) , vkg::GFP-454 [36] , UAS prc-IR41320 , UAS prc-IR100357 , UAS loh-IR31020 and UAS Sar1-IR34191 [46] , UAS loh-IR6232-2 ( Drosophila Genetic Resource Center , Kyoto ) , mef2-Gal4 ( H . Nguyen ) and prc-Gal4 [20] . Precise excision of minos elements was carried out essentially as described before [26] . Briefly , homozygous w1118; Mi ( ET1 ) lohMB05750 or w1118; Mi ( ET1 ) prcMB03017 males were mated to w1118; Sco/SM6a , P{hsILMiT}2 . 4 “jump starter” females . After two days adults were removed and the F1 progeny was heat shocked each day at 37°C for 1 h until hatching . F1 males , carrying the minos element ( expressing GFP ) and the transposase source ( recognized by the SM6a balancer ) were mated to adequate balancer stocks . In the F2 generation revertant chromosomes were identified by the absence of GFP expression and isolated via backcrossing to the F1 balancer stocks . Revertant lines were established and removal of the minos elements was evaluated by amplifying closely flanking sequences of the transposon by PCR and sequencing . Oligonucleotides ( minos-flank ) used for PCR and sequencing are: loh-fwd GCGGTCAGCTAAATAGCATC , loh-rev GAATTGGTTTGTCCCACAACG , prc-fwd CACACAGTGGAGCGAGATCC and prc-rev CCTTTCGAAGTGTAAAGTGC . Embryos were prepared for staining by chemical or heat fixation as described previously [47] , [48] . Staining of larvae was done on dissected tissue samples , fixed 1 h in 3 . 7% formaldehyde in 1× PBS . Primary antibodies used are: guinea pig anti-Loh ( 1∶500 , heat fixation , this study ) , mouse anti-Prc/EC11 ( 1∶5 , Developmental Studies Hybridoma Bank , DSHB ) , mouse anti-βPS integrin/CF . 6G11 ( 1∶3 , DSHB ) , mouse anti-FasIII/7G10 ( 1∶3 , DSHB ) , mouse anti-αSpectrin/3A9 ( 1∶3 , DSHB ) , mouse anti-Slit/C555 . 6D ( 1∶3 , heat fixation , DSHB ) , rabbit anti-Perlecan/Trol ( 1∶1 . 000 ) [49] , rabbit anti-Nidogen/Entactin ( 1∶1 . 000 , a gift from S . Baumgartner ) , rabbit anti-Laminin ( detects only secreted Laminin trimers; a gift from J . Fessler ) , rabbit anit-Tinman ( 1∶800 ) [34] and rabbit anti-GFP ( 1∶1 . 000 , Abcam ) . Secondary antibodies used are anti-mouse-Cy2/Cy3 ( 1∶100/1∶200 , Dianova ) , anti-rabbit-Cy2/Cy3 ( 1∶100/1∶200 , Dianova ) and anti-guinea pig-Cy2/Cy3/Alexa633 ( 1∶100/1∶200/1∶200 , Dianova and Abcam ) . F-Actin was visualized by staining fixed tissues using TRITC coupled phalloidin ( Sigma ) , at a concentration of 0 . 4 µg/ml in 1× PBS , for 1 h at room temperature . All images were acquired using a Zeiss LSM 5 PASCAL confocal microscope and standard objectives . The ability of insect nephrocytes to sequester colloids from solutions can be used to specifically label living cells . Therefore colloidal toluidine blue was used as vital stain . Third instar larvae were dissected in 1× PBS and incubated in 0 , 1 mg/ml colloidal toluidine blue solution for 1 min . Living nephrocytes specifically take up the dye resulting in a deep blue staining . Unspecific signals were removed by three consecutive washes in 1× PBS and animals were photographed immediately . Embryonic protein extracts were isolated from 20 selected embryos , which were homogenized in 25 µl ECM extraction buffer ( 1 mM EDTA , 1 , 5% Triton-X 100 and 2 M urea ) . Samples were supplemented with 25 µl 2× SDS sample buffer , cooked at 99°C for 2 min and 20 µl were used for SDS-PAGE . Larval and adult extracts were obtained from 10 whole animals homogenized in extraction buffer . Primary antibodies were diluted in 10% dry milk powder ( w/v ) in TBS-T and incubated overnight at 4°C . Antibodies used were guinea pig anti-Loh ( 1∶5 . 000 , this study ) , mouse anti-Prc/EC11 ( 1∶200 , DSHB ) and mouse anti-βTub/E7 ( 1∶5 . 000 , DSHB ) . Alkaline phosphatase coupled secondary antibodies ( Dianova , Germany ) were diluted 1∶10 . 000 and phosphatase activity was visualized by colorimetric NBT/BCIP reaction . Total protein was stained using 0 . 1 µg/ml amido black 10B ( Sigma ) in 7% acetic acid . Animals were equilibrated for 20 min and heart beat was recorded on a Zeiss Axioplan upright microscope equipped with a 10× air objective ( n . a . = 0 . 30 ) . Single pictures were recorded at 80 frames per second ( fps ) using a Hamamatsu EM-CCD C9100 camera . Images were processed using Fiji and transformed into movie files . For dye injections staged pharate adults ( <90 h APF ) were glued on a glass object slide using double sided scotch tape . After 10 min the operculum was removed with fine forceps to allow imaging of dye accumulation . One single injection per animal was carried out , using a glass capillary applied to a micro manipulator and an Eppendorf FemtoJet microinjector . The capillary was filled with 10 µl uranin solution ( 1 µg/µl in PBS ) that was injected laterally into the abdomen of the animal . Dye accumulation was recorded over three minutes using a stereo microscope equipped with an UV lamp , a corresponding filter set and a consumer digital camera ( Canon PowerShot A650 IS ) . Pixel intensities were measured using the “Plot Z-axis profile” tool of Fiji within a region of interest ( R . O . I ) of the head ( excluding the eyes due to different pigmentation ) . Freshly hatched animals were collected and separated according their sex and genotype . The flies were kept in plastic vials filled with standard cornmeal agar in groups of less than 20 animals at 22°C . The number of living animals was evaluated every three to five days and the flies were transferred onto new vials . Late stage embryos were selected according their genotype , judged by balancer expression . Fixation of embryos , sectioning and image acquisition was described previously [47] . The thickness of the basement membrane ( BM ) was investigated in sections of three independent animals ( two sections per animal ) of each genotype using Fiji . Therefore , BM thickness was measured at ten randomly picked positions in each image leading to a total number of 60 values per genotype . Northern blot was done as described previously with 15 µg total RNA loaded per lane [50] . Hybridization was carried out at 66°C for 24 h . The cDNA of lohA was amplified from cDNA clone GM15606 ( BDGP ) . Oligonucleotides used were lohA-EcoRI-F TACTCAGAATTCATGGCGAAGCTGTTGTTAATATTCAG and lohA-KpnI-R TACTCAGGTACCTTAAATGCCACCCGTGCAGGAAAAAC . The lohAΔSP coding DNA was amplified using the modified oligonucleotide lohAΔSP-EcoRI-F TACTCAGAATTCATG GATTTAACAACTAAAGAGCG . The resulting DNA fragments were cloned into the pUAST vector and transgenic flies were established after standard protocols ( TheBestGene Inc . , USA ) . An antiserum against Loh was generated by injecting two guinea pigs with the sequence specific peptide VFDYHRIDGAEDSNGVTEW-C bound to KLH . Harvested antiserum was affinity purified against the peptide . Peptide synthesis , serum production and affinity purification were carried out by a commercial service ( Pineda Antikörperservice , Berlin ) . All steps were carried out at 4°C or on ice . Total protein from 100 mg adult prc-Gal4/+; UAS-LohA/+ flies ( ∼100 flies ) was extracted in 500 µl ECM extraction buffer ( 1 mM EDTA , 1 , 5% Triton-X 100 and 2 M urea ) . Flies were homogenized , pulled 6-times through a syringe ( Ø = 0 , 8 mm ) and debris was spun down at 8 . 000 g for 30 min . The supernatant was centrifuged again at 13 . 100 g for 30 min . The soluble protein fraction was split into four 100 µl aliquots . One aliquot served as input . The other aliquots were supplemented with 10 µl Protein A-Sepahrose 4B ( Sigma ) , 0 , 1% BSA and either 10 µl PBS ( negative control ) , 10 µl anti-Loh or 67 µl anti-Prc antibody and incubated under constant shaking overnight . Protein A slurry was spun down at 13 . 100 g for 10 min and the pellet was washed in 500 µl ice cold 1M NaCl . The washing step was repeated three times , afterwards the pellets were resolved in 60 µl 2× SDS sample buffer and used for Western blotting
|
Cellular adhesion and tissue integrity in multicellular organisms strongly depend on the molecular network of the extracellular matrix ( ECM ) . The number , topology and function of ECM molecules are highly diverse in different species , or even in single matrices in one organism . In our study we focus on the protein class of ADAMTS-like proteins . We identified Lonely heart ( Loh ) a member of this protein family and describe its function using the cardiac system of Drosophila melanogaster as model . Loh constitutes a secreted protein that resides in the ECM of heart cells and mediates the adhesion between different cell types - the pericadial cells and the cardiomyocytes . Lack of Loh function induces the dissociation of these cells and consequently leads to a breakdown of heart function . We found evidence that the major function of Loh is to recruit the collagen Pericardin ( Prc ) to the ECM of the cells and allow the proper organization of Prc into a reticular matrix . Since the function of Loh homologous proteins in other systems is rather elusive , this work provides new important insights into the biology of cell adhesion , matrix formation and indicates that ADAMTS-like proteins might facilitate an evolutionary conserved function .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"animal",
"genetics",
"model",
"organisms",
"zoology",
"cell",
"adhesion",
"adhesion",
"molecules",
"heart",
"development",
"molecular",
"development",
"extracellular",
"matrix",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"morphogenesis"
] |
2013
|
The Conserved ADAMTS-like Protein Lonely heart Mediates Matrix Formation and Cardiac Tissue Integrity
|
Buruli ulcer is a neglected tropical disease of the skin that is caused by infection with Mycobacterium ulcerans . We recently established an experimental pig ( Sus scrofa ) infection model for Buruli ulcer to investigate host-pathogen interactions , the efficacy of candidate vaccines and of new treatment options . Here we have used the model to study pathogenesis and early host-pathogen interactions in the affected porcine skin upon infection with mycolactone-producing and non-producing M . ulcerans strains . Histopathological analyses of nodular lesions in the porcine skin revealed that six weeks after infection with wild-type M . ulcerans bacteria extracellular acid fast bacilli were surrounded by distinct layers of neutrophils , macrophages and lymphocytes . Upon ulceration , the necrotic tissue containing the major bacterial burden was sloughing off , leading to the loss of most of the mycobacteria . Compared to wild-type M . ulcerans bacteria , toxin-deficient mutants caused an increased granulomatous cellular infiltration without massive tissue necrosis , and only smaller clusters of acid fast bacilli . In summary , the present study shows that the pathogenesis and early immune response to M . ulcerans infection in the pig is very well reflecting BU disease in humans , making the pig infection model an excellent tool for the profiling of new therapeutic and prophylactic interventions .
Buruli ulcer ( BU ) is a slow progressing , necrotising disease of the skin that mainly affects rural African communities [1 , 2] and is caused by Mycobacterium ulcerans . In contrast to other closely related mycobacterial pathogens such as M . tuberculosis and M . leprae , M . ulcerans produces a polyketide exotoxin named mycolactone , which is considered the main virulence factor of the bacteria and responsible for the extensive tissue damage seen in BU lesions [3 , 4] . Three distinct non-ulcerative forms of the disease are described ( nodules/papules , plaques and oedema ) , which may all progress to ulceration as soon as the damage of the subcutaneous tissue leads to the collapse of the overlying epidermis and dermis [1] . Extensive histopathological analyses of advanced BU lesions were possible with excised tissue from surgically treated patients and diagnostic punch biopsies . The major hallmarks of M . ulcerans infection , which are also used for histopathological confirmation of clinical diagnosis , are the presence of coagulative necrosis , fat cell ghosts , epidermal hyperplasia and extracellular clusters of acid fast bacilli ( AFB ) in the absence of major inflammatory infiltrates in central parts of the lesions [5 , 6] . Although it was long thought that inflammatory infiltrates were completely absent in BU lesions [7] , more recent studies demonstrated that cellular infiltration occurs at the periphery of lesions , where mycolactone levels are believed to be low [8 , 9] . The distribution of AFB and of cellular infiltrates is very heterogeneous in advanced BU lesions [7 , 10–12] . In the course of antibiotic treatment , massive leukocyte infiltration is observed , which culminates in the development of ectopic lymphoid structures in the lesions [13] . Since 2004 , with the replacement of surgical treatment by the antibiotic combination therapy of rifampicin and streptomycin for eight weeks [14 , 15] , tissue samples are no longer available for histopathological investigation . Additionally , the unknown mode of transmission , the very slow growth rate of M . ulcerans , and late care seeking behaviour of the affected populations are all factors that have made histopathological description of early BU stages difficult . This gap can be filled with the pig ( Sus scrofa ) model for experimental M . ulcerans infection we established recently [16] , enabling the study of early host-pathogen interactions and pathogenesis in BU . Productive M . ulcerans infection in the pig skin leads to the development of lesions that closely resembled human BU lesions in their macroscopic as well as microscopic appearance [16] . All key features of BU pathology in humans were also found in the infected pig skin , which led us to conclude that the pig model is suitable for studying the early pathogenesis of BU and for the evaluation of new treatment and vaccination approaches [16] . In order to further characterize the developing lesions in the pig skin and in particular the role of mycolactone in the pathogenesis of BU , we aimed at determining the different infiltrating cell types by immunohistochemistry ( IHC ) . To this end , protocols were established for different cell markers by IHC on formalin fixed , paraffin-embedded pig skin tissue . These markers were then used to characterize the cellular infiltrates in nodular and ulcerative lesions of the pig skin six weeks after infection . Finally , the lesions caused by wild-type M . ulcerans were compared immunohistochemically with lesions caused by mycolactone non-producing M . ulcerans strains .
All animal experiments were approved by the Animal Welfare Committee of the Canton of Berne under licence number BE92/14 , and conducted in compliance with the Swiss animal protection law ( SR 455 ) . The M . ulcerans strain S1013 was isolated in 2010 from a swab taken from the undermined edges of the ulcerative lesion of a Cameroonian BU patient [17] . Two passages of the strain after isolation were done in BacT/ALERT medium ( MB-251011 , Biomerieux , USA ) at 30°C . The mycolactone-deficient M . ulcerans strain S1228 was kindly provided for this study by Kris Huygen [18 , 19] . For preparation of the infection inocula , bacteria were cultivated in BacT/ALERT medium for nine weeks , recovered by centrifugation and diluted in sterile phosphate-buffered saline ( PBS ) to 375 mg/ml wet weight corresponding to 1 . 3 x 107 colony forming unit ( CFU ) /ml ( S1013 ) and 3 . 1 x 107 CFU/ml ( S1228 ) respectively , as determined by plating serial dilutions on 7H9 agar plates . Three specific pathogen-free 2-month-old pigs ( Large White ) from the in-house breeding unit of the Institute of Virology and Immunology ( IVI ) were kept under Biosafety-level-3 ( BSL3 ) conditions one week prior and during the time of experimental infection . Animals were checked once daily for macroscopic signs of infection , had ad libitum access to water , straw and hay , and were fed daily with complete pelleted food . Pigs were infected subcutaneously on both flanks at four to five infection sites with 1 . 3 x 106 ( six replicates on one pig ) and 1 . 3 x 105 ( 14 replicates on two pigs ) CFU S1013 and 3 . 1 x 106 ( two replicates on one pig ) and 3 . 1 x 105 ( four replicates on two pigs ) CFU S1228 in 100μl PBS . Injection areas were wiped with 70% ethanol and bacterial suspensions injected subcutaneously with a 23G needle . Individual infection sites were encircled with a marker and the labelling renewed at least once a week . The pigs were euthanized at six weeks post-infection and tissue samples taken as described below . Pigs were euthanized by intravenous injection of pentobarbital ( 150 mg/kg bodyweight ) and subsequent exsanguination . Skin tissue at infection sites were extensively excised with a scalpel and scissors , including all layers of the skin , the fascia and the first layer of muscle . The samples were transferred immediately into 10% neutral-buffered formalin solution ( approx . 4% formaldehyde; HT501128-4L , Sigma-Aldrich , USA ) . Additionally to the infected skin tissue a total of 16 lymph nodes were excised . The three main draining lymph nodes ( Nll . cervicales superficiales ventrales and dorsales ) and the Nll . Subiliaci [20] were excised bilaterally from each pig , except for pig number two , for which only the Nll . cervicales superficiales dorsales and Nll . subiliaci were collected . After fixation in formalin for 60 hours , samples were transferred to 70% ethanol for storage and transport , dehydrated and embedded into paraffin . 5 μm thin sections were cut , deparaffinised , rehydrated and directly stained with Haematoxylin/Eosin ( HE; 51275-500ML , Sigma-Aldrich , Switzerland; 3446 , J . T . Baker , Netherlands ) or Ziehl-Neelsen/Methylene blue ( ZN; 21820-1L , Sigma-Aldrich , Switzerland; 03978-250ML , Sigma-Aldrich , Germany ) according to WHO standard protocols [1] . In order to characterize cellular infiltrates occurring in nodular and ulcerative lesions in the pig skin after infection with M . ulcerans , we conducted a search for antibodies to stain neutrophils , macrophages/monocytes , B-cells and T-cells . We preferentially tested monoclonal antibodies ( mAb ) against porcine antigens that were evaluated for IHC on paraffin-embedded tissue . Because of the high similarity of epitopes of human and pig cell surface molecules we also tested mAb against human antigens if no suitable reagents against the porcine antigens were found . Standard IHC staining protocols could be established for macrophages/monocytes , T-cells and neutrophils ( Table 1 ) . Despite major efforts , labelling of B-cells was unsuccessful . Nevertheless , based on labelling of all other markers and exclusion of the non-labelled cells with lymphocyte appearance in HE and Methyleneblue staining , clusters of B-cells could be identified ( Fig 1 ) . For monocytes/macrophages , two markers were detected by IHC ( IBA-1 , CD107a ) , but due to stronger labelling ( Fig 1 ) , IBA-1 staining was preferred . For IHC labelling , endogenous peroxidase was blocked in 0 . 3% H2O2 ( 31642-500ML , Sigma-Aldrich , Germany ) for 20 min and unspecific binding was prevented by incubation with blocking serum matching the secondary antibody host . Subsequently , slides were pre-treated by the Citrate method [21] and incubated at room temperature for 1 hour with monoclonal antibodies in appropriate dilutions ( Table 1 ) . Sections were washed three times in PBS and incubated for 30 min with a biotinylated secondary antibody corresponding the primary antibody host ( Goat-anti-rabbit , BA-1000; rabbit-anti-rat , BA-4001 , both Vector Laboratories; goat-anti-mouse 1038–08 , Southern Biotech ) . Sections were washed again three times in PBS and incubated for additional 30 min with streptavidin-horseradish peroxidase conjugate ( VectastainR Elite ABC KIT; PK-6100 , Vector Laboratories , USA ) . Staining was performed using Vector R NovaRed TM Substrate KIT ( SK-4800 , Vector Laboratories , USA ) as a substrate and Meyer’s Haematoxylin as counterstain ( Sigma ) . Stained sections were mounted with Eukitt mounting medium ( 03989-100ML , Sigma-Aldrich , Germany ) . Pictures were taken with a Leica DM2500B microscope or with an Aperio scanner at 20x magnification .
Pigs infected with a dose of 1 . 3 x 106 CFU M . ulcerans S1013 developed large nodular lesions of which 3/6 had ulcerated after six weeks . The central necrotic core of non-ulcerated lesions contained large clumps of AFB and was surrounded by layers of cellular infiltrates ( Fig 2A1 and 2A2 ) . What we previously defined as rings one , two and three [16] based on the degree of necrosis and cell types visible in Methyleneblue staining ( S1 Fig ) , was now confirmed by IHC to be neutrophilic infiltration ( Fig 2B1 and 2B2 ) . The antibody used for staining of neutrophils does not only stain intact cells but also the remaining target antigen , a 21 kDa protein expressed by porcine neutrophils , in otherwise necrotic tissue . Strong staining of the completely necrotic centre of the lesion containing clumps of AFB ( Fig 2B1 and 2B2 ) was therefore reflecting the presence of large amounts of neutrophilic debris . The next ring ( previously described as ring number two [16] ) consisted of apoptotic vesicles and neutrophils that were mostly necrotic , but retained some cellular appearance . Ring number three consisted mainly of intact neutrophils . These three rings were surrounded by a macrophage belt ( Fig 2C1 and 2C2 ) that was interspersed with T-cells ( Fig 2D1 and 2D2 ) . T-cells appeared regularly distributed around the lesion and were more frequent towards the outside of the macrophage belt . Some of the lesions showed lymphocyte clusters at the border of the outermost infiltration ring that were CD3 negative and shared features with B-cell clusters found in human lesions [8 , 13 , 22] . Overall smaller nodules had developed six weeks after infection with a ten times smaller inoculum ( i . e . 1 . 3 x 105 CFU M . ulcerans S1013 , Fig 3A ) . However , the general organization of the cellular infiltration was comparable to that of the lesions that had developed after inoculation with the higher dose ( Fig 3B ) . Both small and large nodules displayed typical histopathological signs of BU found in human lesions , such as coagulative necrosis , extracellular clumps of AFB , fat cell ghosts and epidermal hyperplasia . While laterally to the main necrotic core and surrounding infiltration the epidermis appeared largely unchanged and showed no hyperplasia , the epidermis located above the lesion displayed clear thickening and elongation of the rete ridges ( Fig 3 ) . Six weeks after infection with 1 . 3 x 106 CFU M . ulcerans S1013 , 3/6 lesions had ulcerated . IHC staining demonstrated that the necrotic core containing neutrophilic debris and the majority of AFB present in nodular lesions had been ejected through the opening in the epidermis ( Fig 4A and 4C ) . Leftovers of the initially high bacterial burden were found in the crust that had remained on the surface of the ulcerative lesion ( Fig 4A1 ) and in the neutrophilic infiltrate directly below the crust ( Fig 4A2 ) , representing also leftovers of the former necrotic core . Besides these residual AFB , colonization of the wound with cocci was found ( Fig 4A1 ) . The cellular infiltration that remained in the dermis between the fibroblasts consisted primarily of macrophages that were heavily interspersed with T-cells ( Fig 4D and 4E ) . In order to assess how the production of the cytotoxic macrolide exotoxin mycolactone affects the local cellular infiltration we infected pigs with a mycolactone-deficient M . ulcerans mutant and compared the resulting lesions to those caused by wild-type M . ulcerans bacteria . In contrast to the single nodular structures with a central AFB-containing necrotic core observed after infection with wild-type bacteria ( Fig 5A1 , 5A2 , and 5A5 ) , the lesions caused by mycolactone-deficient bacteria presented typically as multiple granulomatous structures densely grouped together ( Fig 5B1 ) . Several small central clusters of neutrophils were surrounded by an overall much larger proportion of infiltrating macrophages and T-cells ( Fig 5B1–5B4 and 5B6 ) than in lesions caused by wild-type M . ulcerans . Fat cell ghosts were only observed in lesions caused by wild-type M . ulcerans ( Fig 5A7 ) and the necrotic core was larger in those lesions compared to lesions caused by mycolactone-deficient bacteria , in which single granulomas did not contain a completely necrotic core ( Fig 5A7 and 5B7 ) . Additionally to the local infected skin we removed the three main lymph nodes ( Nll . cervicales superficiales ventrales and dorsales , Nll . subiliaci [20] ) draining the infected skin regions on each side of the pigs . Bacteria were found in one of the dorsal subiliac lymph nodes of the pig that had received injections with 1 . 3 x 106 CFU M . ulcerans ( Fig 6 ) . A small structure of cellular infiltration was seen in the lymph node that consisted of a neutrophilic core without major necrosis . Similar to what we had observed in nodular lesions of the skin , this core was surrounded by a belt of macrophages that was interspersed with T-cells ( Fig 6 ) .
BU is most prevalent in rural regions of West African countries like Benin , Ghana , Cameroon and Ivory Coast , where patients often have only limited access to health care [23] . Together with the painless nature of the disease , patients often display late health care seeking behaviour [1] . In consequence , early non ulcerative stages of BU were rarely described histopathologically . Additionally , the number of tissue samples available for histopathological analyses has decreased tremendously , since the standard treatment consisting of surgical excision of the lesions was replaced with an eight-week antibiotic therapy with rifampicin and streptomycin . To contribute to the understanding of early host-pathogen interactions and to evaluate new treatment approaches and vaccines , we recently established the pig as a novel model for BU [16] . We demonstrated that infection of pigs with M . ulcerans induces lesions comparable to early BU lesions and we described the developing lesions by histopathology based on HE and ZN stainings . In order to study the early local immune responses to M . ulcerans infection in more detail , we established and applied standard immunohistochemistry protocols for staining of porcine neutrophils , macrophages and T-cells in paraffin-embedded pig skin tissue sections . Although we failed to develop a method for staining of B-cells , cluster of B-cells , as also observed in treated human BU lesions [13] , could still be identified . Despite the large number of antibodies specific for pig leukocyte antigens available , the number of antibodies suitable for staining paraffin-embedded tissue samples turned out to be very limited . Due to the fact that in vivo infection experiments with M . ulcerans have to be carried out under BSL-3 conditions in Switzerland , there was no alternative to formalin fixation and paraffin embedding of the tissue samples . With the three different cell types that we were able to stain , the main players in M . ulcerans infection , as known from the analysis of advanced human lesions [13] , could be identified . However , a broader panel of antibodies against leukocyte markers for use in IHC on paraffin-embedded pig tissue would be of high interest . To date only few human nodular lesions excised after antibiotic treatment have been analysed by immunohistochemistry [22] . They showed a similar cellular architecture to what we observed in the pig lesions six weeks after infection , with layers of infiltrating cells surrounding a necrotic core structure . Whether this layered structure can also be observed in human untreated BU nodules needs to be elucidated in future studies . However , the strong leukocyte infiltration at the site of infection in the pig lesions indicates that mycobacterial compounds are readily recognized by the innate immune system . The early immune response may in certain circumstances be capable of clearing the inoculum , a view that is supported by the results of serological studies that demonstrated that only a small proportion of individuals from BU endemic areas exposed to M . ulcerans develop BU disease [24 , 25] . Both the size of the inoculum and the yet not fully understood mode of transmission are likely to be of crucial importance for the outcome of an exposure to M . ulcerans . While subcutaneous inocula of 1 . 3 x 105 and of 1 . 3 x 106 CFU M . ulcerans were basically leading to the same pathogenic processes , we have shown previously , that after infection of pigs with 2 x 104 CFU signs of infection found after 2 . 5 weeks had spontaneously resolved at 6 . 5 weeks [16] . With respect to the site of inoculation it has been shown in a guinea pig model that intradermal injection , but not topical application of M . ulcerans to abrasions leads to an establishment of an infection [26] . Human pre-ulcerative BU lesions may contain enormous amounts of AFB [8] . It is generally assumed that the major bacterial burden of these lesions is lost by ulceration , although this was never directly observed . This is now confirmed by our studies in pigs . While nodular lesions in the pig skin contained large amounts of AFB surrounded by layers of infiltrating leukocytes , the majority of M . ulcerans bacilli was no longer present in the ulcerated lesions . However , remains of AFB were detected in the crust on the ulcer and directly below in the upper dermis . Because we were mainly interested in studying early forms of BU lesions , we have so far not followed the progression of lesions to large ulcers . Therefore , we do not know whether the observed ulceration of the nodular lesions would lead to self-healing as it was recently reported in the guinea pig model [27] or whether similar to human lesions , AFB focally remaining in the undermined edges of the lesions would be sufficient to keep the disease on going . Human and pig skin are strikingly similar [28] and our comparison of pig and human BU lesions supports a similar pathogenesis of the disease in the two species . Our observation in the pig skin that epidermal hyperplasia occurs only in close proximity to the lesion , is important for the analysis of human histopathological samples . Only small excisions or punch biopsies of the diseased tissue are usually available for diagnosis of human BU lesions by histopathology . Presence or absence of epidermal hyperplasia could be used as an estimate for the relative distance of the respective sample to the core of the lesion . The changes in the skin compared to healthy pig skin appear enormous considering an infection time of only six weeks , which might also be due to the high infection dose used . Further studies should focus on earlier infection time points and address the dynamics of the development of the lesions observed . Infection with a mycolactone-negative M . ulcerans mutant led to a granulomatous reaction that was clearly different from the infiltration seen upon infection with wild-type M . ulcerans . The extent of cellular infiltration was larger and the AFB containing foci were not completely necrotic . Multiple small infection foci containing fewer and smaller clusters of AFB developed upon infection with the mycolactone-negative mutant . In contrast , infection with wild-type bacteria typically led to the development of one single infection focus containing multiple large AFB clusters . Despite these differences , the individual mutant-induced granulomas comprised the same sequential infiltration layers with central neutrophils surrounded by macrophages and T-cells . As expected , the mycobacterial antigen reservoir recognized by the innate immune system and leading to infiltration with different leukocytes thus seems to be essentially the same for wild-type and mutants . Mycolactone therefore appears to be the main factor responsible for the development of a central necrotic infection focus surrounded by layers of neutrophils , macrophages and lymphocytes . Our data show that mycolactone-negative M . ulcerans are more effectively cleared from the site of infection than wild-type bacteria . The pathogenesis observed upon infection with mycolactone-negative M . ulcerans resembles that caused by M . marinum infection [29 , 30] , which is primarily described both in fish and humans as a granulomatous disease . M . ulcerans developed from M . marinum by acquisition of a giant plasmid encoding for the enzymes necessary to produce myoclactone and has since drastically reduced its genome [31] . However , the evolutionary history appears to be still reflected in the response to infection , with mycolactone-negative M . ulcerans resembling more closely infection with M . marinum than infection with virulent M . ulcerans . Concerning the type of cellular infiltration occurring upon infection with toxin-negative mutants , results from experimental BU infections in the guinea pig [18] as well as the mouse model [32] are largely in line with what we observed in the pig model , except for the lack of neutrophil infiltration in guinea pig lesions . However , neutrophilic debris is present also in the necrotic centre of human BU lesions [8] , which is indicative for an early wave of neutrophil infiltration in response to M . ulcerans infection . Of the 16 dissected lymph nodes , only one contained AFB that were surrounded by neutrophils and macrophages . However , we cannot rule out , that more such micro-lesions were present in other lymph nodes , since we did not cut thin sections through the entire lymph nodes . It is unknown yet to what extent M . ulcerans is spreading to lymph nodes in BU patients , because lymph nodes adjacent to ulcerative lesions are usually not resected . The fact that the majority of BU patients present only a single lesion indicates that M . ulcerans is not readily spreading throughout the body , which is attributed to its temperature sensitivity [21] . In summary , the pathogenesis and early immune response to M . ulcerans infection in humans seems to be very well mirrored by the pig infection model . This makes the model an excellent tool for the final pre-clinical profiling of new treatment options and candidate vaccines .
|
Buruli ulcer is a necrotizing ulcerative disease of the skin and underlying tissue caused by infection with Mycobacterium ulcerans . Because patients often present late to health facilities , early stages of Buruli ulcer are only insufficiently described by histopathology . To study early host-pathogen interactions , we recently established an experimental pig infection model for Buruli ulcer . Here we used the model to study the pathogenesis and the local cellular immune responses upon infection with mycolactone-producing and non-producing M . ulcerans strains . Infection with toxin-producing bacteria led to the development of nodular lesions six weeks after infection , in which extracellular clumps of acid fast bacilli were surrounded by distinct layers of leukocytes . Ulceration of the nodular lesions subsequently led to the loss of most of the bacterial burden . In contrast , after infection with toxin-deficient M . ulcerans bacteria increased granulomatous cellular infiltration was observed , and massive tissue necrosis was absent . Pathogenesis as well as early immune responses to M . ulcerans infection in the pig is very well reflecting the human disease , making it a good model for the evaluation of the efficacy of new treatment options and candidate vaccines .
|
[
"Abstract",
"Introduction",
"Material",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"dermatology",
"livestock",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pig",
"models",
"immunology",
"vertebrates",
"animals",
"mammals",
"animal",
"models",
"model",
"organisms",
"skin",
"infections",
"signs",
"and",
"symptoms",
"immunologic",
"techniques",
"bacteria",
"neutrophils",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"swine",
"white",
"blood",
"cells",
"animal",
"cells",
"lesions",
"actinobacteria",
"agriculture",
"mycobacterium",
"ulcerans",
"immunohistochemistry",
"techniques",
"diagnostic",
"medicine",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"amniotes",
"histochemistry",
"and",
"cytochemistry",
"techniques",
"organisms"
] |
2016
|
Local Cellular Immune Responses and Pathogenesis of Buruli Ulcer Lesions in the Experimental Mycobacterium Ulcerans Pig Infection Model
|
Over the last three decades , the epidemiological profile of visceral leishmaniasis ( VL ) has changed with epidemics occurring in large urban centers of Brazil , an increase in HIV/AIDS co-infection , and a significant increase in mortality . The objective of this study was to identify the risk factors associated with death among adult patients with VL from an urban endemic area of Brazil . A prospective cohort study included 134 adult patients with VL admitted to the University Hospital of the Federal University of Mato Grosso do Sul between August 2011 and August 2013 . Patients ranged from 18 to 93 years old , with a mean age of 43 . 6 ( ±15 . 7% ) . Of these patients , 36 . 6% were co-infected with HIV/AIDS , and the mortality rate was 21 . 6% . In a multivariate analysis , the risk factors associated with death were secondary bacterial infection ( 42 . 86 , 5 . 05–363 . 85 ) , relapse ( 12 . 17 , 2 . 06–71 . 99 ) , edema ( 7 . 74 , 1 . 33–45 . 05 ) and HIV/AIDS co-infection ( 7 . 33 , 1 . 22–43 . 98 ) . VL has a high mortality rate in adults from endemic urban areas , especially when coinciding with high rates of HIV/AIDS co-infection .
Over 90% of visceral leishmaniasis ( VL ) cases occur in six countries: Bangladesh , Brazil , Ethiopia , India , South Sudan and Sudan [1] . In Brazil , an increase in incidence over the past three decades has coincided with epidemics in large urban centers such as Campo Grande , Teresina , São Luis , Natal and Belo Horizonte [2] and a significant increase in mortality [3] . Despite increased mortality , there are few studies assessing risk factors associated with death among VL patients and the majority involves populations of all age groups in urban and non-urban areas with low rates of HIV co-infection . In non HIV infected patients , jaundice , bleeding , and associated infections were most frequent risk factors [4–8] , while being older than 45 years of age was reported by one study [9] . In addition , over the last three decades , Brazil has had an increase in the HIV/AIDS epidemic [10] . Due to the changes in the epidemiological profile of VL and its expansion into urban areas with higher prevalence of HIV/AIDS , prospective studies are necessary to better understand the determinants associated with death and to propose future interventions . This study aims to identify risk factors associated with death from VL in adult patients from endemic urban areas of Brazil .
This prospective cohort study was conducted between August 2011 and August 2013 with adult patients with VL . Patients were monitored by the Reference Service for Infectious and Parasitic Diseases of the University Hospital in the Federal University of Mato Grosso do Sul , Campo Grande , Mato Grosso do Sul , Brazil . Clinical , sociodemographic and laboratory variables were recorded on a standardized form and . The criteria for inclusion in the study were for patients to be aged 18 years or older and have a confirmed laboratory diagnosis of VL . The VL diagnosis was performed according to the recommendations of the laboratory of the Ministry of Health [11] . Parasitological diagnoses were performed based on detection of amastigotes of the parasite in Giemsa stained bone marrow smears by an experienced microscopist and by isolation of promastigotes from culture media ( Mc Neal , Novy & Nicolle-NNN and Brain Heart Infusion-BHI ) . Immunological tests were performed using an immunochromatographic test ( Kala-azar Detect , INBIOS , WA ) . Cases were considered confirmed if they presented at least one positive laboratory test , parasitological or immunochromatographic test . HIV screening was performed by enzyme-linked immunosorbent assay and confirmed with an immunofluorescence antibody test , or Western Blot assay [11] . The treatment protocol was defined according to the guidelines of the Ministry of Health and considered severity signs ( jaundice , bleeding hemorrhages , generalized edema , signs of toxemia and severe malnutrition ) , co-infection with HIV/AIDS , age , renal failure and heart disease [12 , 13] . Patients with less than 50 years , without HIV infection , without renal failure or heart disease and without severity signs were treated with antimoniate of N-methylglucamine , which contains pentavalent antimony ( Sbv ) , at a dose of 20 mg/kg/day of Sbv for 30 days . Patients with severity signs received amphotericin B deoxycholate at a dose of 1 mg/kg/day for 14 to 20 days [14] . Patients with more than 50 years or with HIV/AIDS or renal or cardiac failure , with or without severity sign , received liposomal amphotericin B at 4 mg/kg/day for 5 days or 3 mg/kg/day for 7 days with 20 mg/kg of total dose [13] . After discharge , patients were instructed to return after 1 , 3 , 6 and 12 months for clinical and laboratory revaluation . After treatment , patients with AIDS received secondary prophylaxis of liposomal amphotericin B , administered at 3 mg/Kg every 15 days to avoid relapse [12 , 15] . Relapse was defined as showing clinical signs and symptoms suggestive of VL within 12 months in successfully treated patients [11] . Successful therapy was achieved if there was improvement of general condition , resolution of fever , regression of splenomegaly and recovery of blood counts [15] . Data were tabulated in a spreadsheet ( S1 Data ) . SAS version 9 . 2 ( SAS Institute , Cary , NC ) and SPSS version 22 . 2 were used to analyze bivariate and multivariable models . Dichotomy or categorical data were analyzed with the chi-square test or Fisher’s exact test . For continuous variables , the t-test or Anova was utilized . Missing data were excluded of analyses . Univariate and multivariate logistic regressions were conducted to identify factors associated with death . Variables were included in the model if they achieved a significance level of p<0 . 20 in the univariate analysis . Correlated variables were tested individually , and the Wald test was used to evaluate the significance level of risk factors in the final model . The results were expressed as relative risk ( RR ) with 95% confidence intervals ( 95% CI ) . A correlation matrix was used to assess confounding variables and correlations between variables . Kaplan-Meier survival curves were constructed for each variable using the long-rank test . Statistical significance was set at p<0 . 05 . The present study was approved by the Ethics Committee of the Federal University of Mato Grosso do Sul under protocol number 2179 and case number 02480049000–11 . All patients voluntarily signed a statement of informed consent for the collection of data .
One hundred thirty-seven patients were eligible for the study , however , three not agreed to participate . Of the 134 participants in the present study , 94 ( 70 . 1% ) were male . The age of participants ranged from 18 to 93 years , with a mean age of 43 . 6 ( SD 41 ± 15 . 7 ) . Most of the individuals ( 95% ) were from urban areas , 107 ( 79 . 8% ) from Campo Grande , and the remainder from 17 other cities . The most common comorbidity in VL patients was AIDS ( 49;36 . 6% ) , followed by tuberculosis ( 3;2 . 3% ) , erythematosus systemic lupus ( 1;0 . 75% ) and chronic myelocytic leukemia ( 1;0 . 75% ) . At the time of diagnosis of VL in patients with HIV/AIDS , the mean T-CD4+ cell count of was 68/mm3 . VL and HIV/AIDS were diagnosed simultaneously in 32 . 7% of the patients . Parasitological examination ( direct and culture ) was performed in 126 ( 94 . 0% ) patients , with 93 positive results ( 73 . 84% ) . Of the 119 ( 88 . 8% ) samples tested by immunochromatographic test , 96 ( 80 . 7% ) were positive . In 57 ( 42 . 5% ) patients , diagnosis was confirmed by two methods ( parasitological and immunochromatographic test ) , only by parasitological in 36 ( 26 . 9% ) , and only by immunochromatographic test in 41 ( 30 . 6% ) patients . The time between the first symptoms and diagnosis of VL ranged from one to 421 days , with a median of 31 days . The period between diagnosis and death ranged from three to 431 days , with a median of 346 days . The mortality rate was 21 . 6% ( 29/134 ) . Differences in demographic , clinical , laboratory and therapeutic relation to the evolution of the disease are shown in Tables 1 and 2 . Patients with VL and HIV/AIDS exhibited a mortality rate of 36 . 7% ( 18/49 ) ( Fig 1 ) . Variables identified as risk factors for mortality were HIV/AIDS , relapse , secondary bacterial infection and edema ( Table 3 ) . Although splenomegaly presented a p<0 . 05 in the univariate analysis , it showed a strong correlation with HIV and when included in the final model and a reduction in the Wald test value when included in the multivariable model . Therefore , we only HIV variable was evaluated in the final model . Among 23 patients who relapsed , 15 ( 65 . 2% ) were HIV-positive . Although there was an association between HIV and relapse ( p<0 . 01 ) , the two variables in the final model had a higher Wald test value than when only one of the variables was included in the model . Of the 8 patients without HIV infection who relapsed , one had systemic erythematous lupus , one had leukemia , one had cirrhosis , and two were older than 85 years of age . Of the 49 patients with AIDS , 7 ( 14 . 3% ) died during hospitalization and only 15/42 ( 35 . 7% ) of them regularly adhered to secondary prophylaxis . No difference in relapse and death rates was observed among of those who adhered and those did not adhere to prophylaxis [33 . 3% versus 29 . 6%; p = 0 . 92 and 40 . 0% versus 22 . 2%; p = 0 . 38 , respectively] .
Most studies that evaluated risk factors for death among VL patients are retrospective design , secondary data analysis and involving different age groups [4 , 6 , 7 , 8] . Our study differs by providing a 12-month follow-up of a cohort of adult patients from an urban center with a high co-infection rate of HIV/AIDS . In this way , relapses and deaths that occurred after the first episode of VL in this particular group of patients could also be documented . The mortality rate ( 21 . 6% ) observed in the present study is higher than the rates normally detected in Brazil [2 , 3 , 5] . This high rate could be associated with the clinical differences of patients included in the study who were mostly over 50 years of age and were co-infected with HIV/AIDS [4 , 16 , 17] . Herein , we identified four variables associated with death among VL patients: secondary bacterial infection , edema , HIV/AIDS co-infection and VL relapse . Secondary bacterial infection is a well-known risk factor related to the severity of sepsis and described in previous studies [7 , 8 , 18 , 19 , 20] . The presence of edema , which may reflect protein malnutrition , liver or renal failure , has also been described as a risk factor for an unfavorable outcome [7 , 8] . Recent studies have also identified that co-infection with HIV/AIDS is a risk factor for death [4 , 5 , 20] . The 36% mortality rate among HIV/AIDS patients , in this study , is much higher than studies previously conducted in Brazil [21 , 22 , 23 , 24] . It is known that patients who are HIV/AIDS-positive are more likely to develop VL due to the depletion of both cellular and humoral immune responses against Leishmania [25] and , moreover , VL/HIV co-infection can accelerate the evolution of both diseases [26] , given that the two agents are located in the same host cell . The co-infection can also enhance pathogenic effects and impair the correct function of macrophages [27 , 28] . This synergism favors a fatal outcome of patients with VL . Liposomal amphotericin B is the first choice of treatment in Brazil for the following individuals: VL positive patients over 50 years of age , patients co-infected with HIV/AIDS , patients with immunosuppressive diseases , and those with renal or cardiac disorders [13 , 29] . Therefore , patients who used this drug were those affected with greater severity and a higher propensity towards death . In the present study , the follow-up period was 12 months , which enabled the detection of relapses , deaths during relapse episodes and a correlation between HIV and relapses of VL . In general , co-infection with HIV/AIDS has a strong association with VL relapses [30] and the frequency of relapses in patients co-infected with VL-HIV/AIDS ranges from 10 to 60% [31] . A systematic review identified that the predictors of VL relapses in HIV-infected patients as: the absence of an increase in CD4+ T cells at follow-up , a lack of secondary prophylaxis , previous history of VL relapses , and CD4+ T cell counts below 100 cells/μl at the time of primary VL diagnosis [32] . Altered immunity may persist in co-infected patients , and the lymphocyte T-CD4+ count can be low , despite the administration of HAART ( Highly Active Antiretroviral Therapy ) and anti-leishmania therapy [33] . The cell activation increases the transcription of the integrated HIV that results in CD4+ cell death . It is induced by the activation of CD4+ and CD8+T cell memory population , resulting in an exhaustion of immune resources . After the treatment some parasites remains inside macrophages and leishmania antigens was thought to be responsible for the cellular activation observed in AIDS patients [27] . This additional monocyte/macrophage activation in VL/AIDS patients has been also associated with increased microbial translocation [34 , 35] . According to the recommendations of the WHO , patients with HIV/AIDS should receive secondary prophylaxis to prevent a relapse of VL [32 , 36 , 37] . Although secondary prophylaxis was indicated , in our study , for all patients with a T-CD4+ count lower than 350 cell/mm3 , adherence to secondary prophylaxis was not associated with protection of relapse or death during the follow-up . Other studies have shown that secondary prophylaxis is not completely effective in prevent relapses [38 , 39] . It is possible that the VL treatment used in patients with HIV infection has been insufficient to reduced parasitic load and cure . During the study period , the recommended liposomal amphotericin B total dose was 20mg/kg . However the current recommendation by the Ministry of Health of Brazil is 40mg/kg , instead of 20mg/kg , in patients co-infected with HIV [15 , 40] . Although the cure criteria are essentially clinical , many clinical and laboratory changes are still present , including splenomegaly , anemia and hypergammaglobulinemia , at the end of treatment in co-infections patients . There is no recommendation to repeat the parasitological examination at the end of treatment [41] and Real-time PCR has been proposed as a suitable tool for monitoring the parasite load during follow-up of co-infected patients and predict the risk of relapses after treatment [42] . In conclusion , this study demonstrated that VL is a serious disease with high mortality rates in adult patients from urban areas , especially when co-infected with HIV/AIDS . Further studies are needed to define the best therapeutic options for effective treatment and prevent VL relapses in these patients .
|
Visceral Leishmaniasis ( VL ) is considered a neglected disease by the World Health Organization ( WHO ) . Over the last two decades , the epidemiological profile of VL has changed with epidemics occurring in large urban centers of Brazil with increased HIV/AIDS co-infection and mortality . Understanding the factors that lead to death in these patients is important to improve public health strategies and reduce mortality . We performed a prospective cohort study between August 2011 and August 2013 of 134 adult patients with VL from Campo Grande urban area . Approximately 40% had AIDS and 22% died during the 12-month follow-up . Relapses , bacterial infection , AIDS and edema were strongly associated with death among VL patients in the Brazilian urban centers . All VL patients should be screened for HIV and bacterial infection as well as to prevent VL relapses .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Risk Factors for Death from Visceral Leishmaniasis in an Urban Area of Brazil
|
The evolution of higher virulence during disease emergence has been predicted by theoretical models , but empirical studies of short-term virulence evolution following pathogen emergence remain rare . Here we examine patterns of short-term virulence evolution using archived isolates of the bacterium Mycoplasma gallisepticum collected during sequential emergence events in two geographically distinct populations of the host , the North American house finch ( Haemorhous [formerly Carpodacus] mexicanus ) . We present results from two complementary experiments , one that examines the trend in pathogen virulence in eastern North American isolates over the course of the eastern epidemic ( 1994–2008 ) , and the other a parallel experiment on Pacific coast isolates of the pathogen collected after M . gallisepticum established itself in western North American house finch populations ( 2006–2010 ) . Consistent with theoretical expectations regarding short-term or dynamic evolution of virulence , we show rapid increases in pathogen virulence on both coasts following the pathogen's establishment in each host population . We also find evidence for positive genetic covariation between virulence and pathogen load , a proxy for transmission potential , among isolates of M . gallisepticum . As predicted by theory , indirect selection for increased transmission likely drove the evolutionary increase in virulence in both geographic locations . Our results provide one of the first empirical examples of rapid changes in virulence following pathogen emergence , and both the detected pattern and mechanism of positive genetic covariation between virulence and pathogen load are consistent with theoretical expectations . Our study provides unique empirical insight into the dynamics of short-term virulence evolution that are likely to operate in other emerging pathogens of wildlife and humans .
The extensive variation in the amount of harm that pathogens cause their hosts has intrigued biologists for centuries and has generated several decades of theoretical work to explain the evolution of pathogen virulence ( e . g . , [1]–[4] ) . Despite this broad and long-standing interest , empirical studies of virulence evolution in non-laboratory systems remain rare ( but see [5]–[10] ) . Furthermore , very few theoretical studies have addressed the short-term dynamics of virulence evolution , which are likely to differ from expected long-term or evolutionarily stable outcomes [4] , [11]–[15] . As new pathogens continue to emerge there is a growing need to understand the short-term dynamics of virulence evolution and the ecological and anthropogenic mechanisms that may influence these dynamics [16] , [17] . The seminal work of theoretical biologists in the 1980s–1990s on the evolution of virulence is regarded as one of the most important recent developments in population and evolutionary biology ( e . g . , [1] , [2] , [18] , [19] ) . This body of theory assumes a virulence “trade-off” and disproved the long-standing hypothesis that evolutionary interactions of parasites and pathogens would necessarily evolve to the lowest possible level of virulence . The trade-off theory's key assumption is that virulence is intimately coupled to transmission because of the dependence of both traits on host exploitation . Thus , the within-host rate of pathogen replication necessary for transmission will also directly ( or indirectly ) reduce host fitness ( i . e . , cause the pathology associated with virulence ) . If increased virulence and transmission to a new host are both associated with higher pathogen load , under certain conditions virulence is expected to evolve toward an intermediate value that is evolutionarily stable [1] , [2] , [4] , [11] , [18]–[21] . The assumptions and predictions of the trade-off model have been tested empirically in a number of systems ( reviewed in [22] ) . However , opportunities to investigate the evolution of virulence during the disease emergence process , when changes in virulence represent short-term rather than equilibrium dynamics , have been rare [5] , [23] , [24] . Arguably the best study of the evolution of virulence following pathogen emergence is of Myxoma virus in rabbits , where the evolution of lower and then intermediate pathogen virulence was documented following the initial introduction of a highly virulent strain into naïve rabbit populations in both Australia and Europe [5] , [25] , [26] . Although the rabbit–Myxoma system suggests that changes in virulence can occur rapidly following pathogen emergence , the initial emergence of myxomatosis resulted from an artificial introduction of a highly virulent strain as a means of rabbit population control . The extent and direction of short-term virulence evolution remain unknown in emerging disease systems more broadly [27] . Theoretical work on nonequilibrium virulence evolution during disease emergence [12] , [13] predicts that , if transmission and virulence positively covary , selection will favor increasing virulence early in an epidemic , while susceptible hosts are common . Furthermore , positive genetic correlations between transmission and virulence among pathogen strains can result during disease emergence from demographic processes even in the absence of the constraints that underlie the trade-off model [12] , and these correlations can alone result in increasing virulence via indirect selection on transmission . One of the best-studied emerging wildlife disease systems is the infection of house finches ( Haemorhous [formerly Carpodacus] mexicanus ) by the bacterial pathogen Mycoplasma gallisepticum ( MG ) [28]–[31] that is spread by direct contact or short-term indirect contact on bird feeders [32] . A novel strain of MG emerged , presumably from poultry , and spread within introduced populations of eastern North American house finches in the mid-1990s [29] , [33] . At the time of MG emergence , eastern North American house finch populations were still expanding westward from an initial small colony of individuals introduced on the east coast from the pet trade in the 1940s [34] . After rapidly spreading through introduced host populations , MG then spread slowly further westward [35] , emerging in the now-contiguous native , western range of house finches in the early 2000s [36]–[38] . Westward spread of MG was slowed by relatively low house finch densities across the Great Plains [39] and the historically separate nature of eastern and western house finch populations that only recently became contiguous [39] . Because western house finches are essentially nonmigratory [40] and show significant population genetic structure from eastern populations at multiple loci [41] , [42] , admixture between eastern and western house finch populations is likely minimal . Infection with MG results in severe conjunctival inflammation in house finches [31] , ultimately reducing overwinter survival rates for affected individuals [43] . Notably , the emergence of MG caused up to 60% declines in previously expanding eastern North American house finch populations [44] . Although MG apparently emerged as a single strain in house finches [33] , [45] , both genotypic [33] , [46]–[48] and phenotypic [49] differentiation among MG isolates from songbirds have now been documented . In particular , the virulence of the earliest detected MG isolate on the west coast of the United States ( CA2006 ) is significantly lower than that of the initially emerged isolate in the eastern United States ( VA1994 ) [49] . Because a phylogeny of archived isolates indicates that a single lineage of MG derived from an eastern isolate colonized the west coast of North America ( W . M . Hochachka , A . A . Dhondt , A . P . Dobson , D . M . Hawley , D . H . Ley , et al . , unpublished data ) , we can consider the emergence of MG in eastern and western house finches as replicate events in order to examine whether phenotypic virulence systematically changed following the initial emergence of MG in geographically separate host populations . Here we show that virulence increased following the sequential emergence of MG in eastern and western North American house finches , a pattern consistent with theoretical predictions for short-term virulence evolution [12] , [13] . Also consistent with theoretical assumptions , we describe positive correlations between indices of transmission and virulence among isolates of MG circulating in North American house finches . We present results from two complementary experiments , one that examines the trend in pathogen virulence in eastern North American MG isolates collected from a limited geographic region over the course of the epidemic ( 1994–2008 ) , and the other a parallel experiment using Pacific coast isolates of the pathogen collected after MG established itself in western house finch populations ( 2006–2010 ) . Our experiments provide important insights into the patterns and mechanisms of virulence evolution during disease emergence that are likely to also apply to emerging pathogens less amenable to detailed experimental study .
Our experimental design minimized potential contributions of host–pathogen coevolution by assaying virulence using finches from a different geographic area than that of the bacterial isolates to which they were exposed ( i . e . , eastern MG isolates were assayed in western finches , and western isolates in eastern finches; Table 1 ) . Our prior work found no evidence of differences in host response to two MG isolates between house finches from eastern ( New York ) and western ( California ) North America [49] . Nevertheless , because we used hosts from two geographic areas in our two experiments , we tested for potential differences due to host origin by infecting finches of both geographic origin with the same reference isolate—the original index MG isolate ( VA1994 ) collected in June 1994 from a Virginian house finch with conjunctivitis [28]—as part of experiment 2 ( Table 1 ) . We then examined the effect of experiment and host origin on virulence for VA1994 because this isolate was the only one used in both experiments and inoculated into hosts of both geographic origin . Because we found significant host and experimental effects for VA1994 ( Figure S1 ) , our quantitative inferences are restricted to within-experiment and within-host-population comparisons only . Qualitative between-experiment inferences are made only with respect to the reference isolate ( VA1994 ) . On both the east and west coasts of North America , MG isolates that were collected in successive years after initial discovery in that region were increasingly more virulent as measured by eye score ( Figures 1 and 2 ) . Consistent with prior results [49] , the earliest detected isolate on the west coast ( CA2006 ) was substantially less virulent than the east coast reference isolate ( VA1994; Figure 2 ) . However , subsequent isolates circulating in the west increased in virulence in successive years after discovery in ways that closely resembled the increase in virulence of successive isolates that circulated on the east coast . After only four years , the most recently collected west coast isolate ( CA2010 ) produced an eye score response similar to that of the original east coast reference isolate ( VA1994; Figure 2 ) . Although we cannot make robust quantitative comparisons , the rate of increase in virulence , as measured by eye score , was approximately 3 . 5 times faster on the west coast . There was a strong positive relationship between average virulence index ( the mean pathogen component of virulence over the sampling period; Equation 5 , Methods S1 ) and average pathogen load ( the mean pathogen load for a given isolate over the sampling period; Equation 5 , Methods S1 ) among the examined isolates ( Figure 3 ) . The posterior mean correlation between our indices of virulence and pathogen load was 0 . 65 ( 95% credible interval , 0 . 42 , 0 . 84 ) when experiment effects were not removed , and 0 . 39 ( 95% credible interval , 0 . 02 , 0 . 70 ) within an experiment after controlling for differences in means between the two experiments ( see Materials and Methods ) . The general among-isolate pattern of covariation ( Figure 3 ) that we detected was underlain by qualitatively different within-host trajectories of virulence and pathogen load over the course of infection between the east and west coast isolates . More specifically , infection by west coast isolates led to lower virulence and pathogen loads ( relative to the reference isolate ) later in infection compared to infection by east coast isolates . For the east coast , all isolates maintained a relatively high eye score until day 56 post-inoculation ( PI ) compared to the reference isolate ( VA1994; Figure 1A ) . For west coast isolates , eye scores were initially similar to or lower than those for the reference isolate ( VA1994 ) and then tended to drop more quickly through time ( Figure 1B ) . For pathogen load , results were qualitatively similar: east coast isolates maintained high observed pathogen load relative to the reference isolate until day 56 PI ( Figure 4A ) , whereas the pathogen load of west coast isolates fell rapidly after day 7 PI in comparison to the reference isolate ( Figure 4B ) . When the two components of observed pathogen load ( probability of observing a non-zero pathogen load ( Figure S3 ) , and expected pathogen load given a non-zero result ( Figure S2; see Methods S1 ) were examined separately , a qualitatively similar pattern was found for expected and observed pathogen load . However , expected pathogen load decreased more slowly for the west coast isolates ( Figure S2B ) than did observed pathogen load ( Figure 4B ) . The probability of observing a non-zero pathogen load was high for east coast isolates across all days PI relative to the reference isolate ( Figure S3A ) . In contrast , these probabilities decreased rapidly with day PI for west coast isolates compared to the reference isolate ( Figure S3B ) .
The strong positive covariation we detected between pathogen load and virulence is consistent with a simple selection process whereby selection for increased transmission results indirectly in higher virulence because of the dependence of both traits on pathogen load . Although there are several potential mechanisms that may have contributed to this process , we hypothesize that two mechanisms may have been particularly important in this system . First , within-host mechanisms such as host immunity , which has been implicated in driving pathogen virulence evolution in both experimental and natural systems [25] , [64] , [65] , may have contributed to the increases in virulence of MG observed on each coast . The role of host immunity in driving pathogen virulence evolution is particularly important when the immune system's ability to reduce pathogen growth , due to immunity from prior exposure or vaccination , is imperfect [16] , [66] . If more virulent isolates of MG ( also known as “virulence mutants”; [67] ) are better able to reproduce in partially immune house finches , then an isolate with optimal virulence in a partially immune host will express higher than optimal virulence when measured in naïve hosts , as we have used in this study . This mechanism has been shown in Marek's disease virus strains in vaccinated chickens [68] . Under this scenario , the increase in the number of partially immune hosts following MG's emergence on each coast may have selected for increasingly virulent isolates . The conditions are present in our system for partial immunity to contribute to virulence evolution: first , house finches recover from MG infection in the wild [43] and in captivity [31] , and individuals that were previously exposed are partially immune upon re-exposure [69]; second , rates of exposure to MG are as high as 50% in some regions [37] , [70] , [71] , providing a strong selective pressure on MG to overcome the partial host immune response . The second mechanism that may have been particularly important in this system involves anthropogenic features such as bird feeders that may have also contributed to the detected increases in virulence of MG on both coasts . During the nonbreeding season , house finches in most of North America feed extensively at bird feeders , where the majority of MG transmission occurs [30] . Feeders are fomites of MG [32] , and the amount of time that susceptible individuals spend on bird feeders predicts the likelihood of transmission in an experimental system ( S . L . States , W . M . Hochachka , and A . A . Dhondt , unpublished data ) . Furthermore , house finches with clinical signs of MG spend significantly more time on bird feeders than healthy conspecifics [55] , thus potentially increasing localized transmission . In this system , increasing virulence may therefore be associated with higher rates of local transmission on bird feeders , contributing to positive among-isolate covariation between virulence and transmission potential . It remains unknown whether that relationship would become saturating at high virulence levels , thus ameliorating the transmission advantage of additional increases in virulence . Understanding the evolution of virulence is central to a deeper understanding of parasite–host coevolution and , more specifically , to the management of infectious diseases [72] , including vaccination strategies [16] , [67] . We have demonstrated parallel patterns of virulence evolution in the same pathogen–host system on two sides of a continent , whereby the pathogen rapidly evolved increasing levels of virulence over the first decade of its interaction with each population of the host species . Both the detected evolutionary pattern and the mechanism are consistent with current theory on short-term virulence evolution . Our study therefore provides one of the first empirical examples of short-term virulence evolution in an emerging pathogen . It will be critical to assay a longer time series of isolates from each location in order to determine whether the virulence of MG converges on an evolutionarily stable value and whether that value is similar across localities . We suspect that similar patterns of changing virulence will be detected in other emerging pathogens of wildlife , domestic animals , and humans whenever positive genetic covariation exists between virulence and transmission .
All procedures for animal care and use were conducted in accordance with national guidelines and were approved by Virginia Tech's Institutional Animal Care and Use Committee . Each of our two experiments used house finches from a different population , such that the finches used in a given experiment came from a different geographic area than the bacterial isolates to which they were exposed ( Table 1 ) . This design minimizes the potential for confounded effects due to host–pathogen coevolution that may have resulted from the population genetic structure between eastern and western North American house finches previously documented at both neutral [41] , [42] and functional [73] genetic markers . First , in experiment 1 , in order to examine how the virulence of MG changed following its initial emergence in eastern North American house finches , we experimentally infected house finches from a single population origin in western North America ( Arizona ) with MG isolates collected from a limited eastern geographic region ( Virginia and North Carolina ) throughout the course of the eastern North American MG epidemic ( 1994–2008 ) . Experiment 2 used an identical approach in order to test how the virulence of MG changed following the establishment of the pathogen on the Pacific coast . Again , we selected isolates from a limited western geographic region ( California ) collected in 2006–2010 following the initial establishment of MG in that region , and assayed these isolates in eastern North American house finches from Alabama . We tested for potential differences in virulence due to host origin and experiment by infecting a small number of finches from Arizona with the reference isolate ( VA1994 ) in experiment 2 ( Table 1 ) . In 2010 , 66 house finches from Maricopa County , Arizona , were captured with cage traps ( permits from the state of Arizona [SP573456] , the US Fish and Wildlife Service [MB158404-1] , and the US Geological Survey Bird Banding Laboratory [23513] ) and shipped via commercial air carrier to Virginia Tech in February 2010 . An additional 64 individuals were captured in Auburn , Alabama , in September 2010 ( state of Alabama permit 5436 ) and transported to Virginia Tech via state vehicle . All house finches were held in individual cages at constant day length ( 12 h light∶ 12 h dark ) and temperature and fed ad libitum pelleted diet ( Daily Maintenance Diet , Roudybush ) throughout the experiment . All individuals were seronegative for MG at capture . We quantified virulence , which we define as the reduction in host fitness ( survival and reproduction ) due to infection , using the severity and duration of inflammatory eye lesions . We used this index of virulence because prior work has shown that the presence of eye lesions is linked with higher mortality in free-living house finches [43] . In contrast , experimental infections in captivity produce little to no mortality [31] , [49] . In combination , these results indicate that infection per se does not directly cause mortality , but that mortality in the wild results from visual impairment associated with the eye lesions that we measure . Disease prevalence is negligible during the breeding season [30] , so direct effects of infection on reproductive rates are not important in this system . Thus , we used eye score in captivity as a relevant proxy for decreased fitness associated with disease severity ( virulence ) in the wild . To score lesions , all birds were examined on day 0 PI , day 7 PI , and then weekly until day 56 PI . Lesions were scored on a 0 to 3 scale ( as per [75] ) : 0 = no detectable swelling or eversion; 1 = minor swelling around the eye ring , 2 = moderate swelling and eversion of the conjunctival tissue , and 3 = the eye nearly hidden by swelling and crusted exudate . MG presence and load were quantified using qPCR as described in [76] . Each bird's eyes were individually swabbed at days 0 , 7 , 14 , 21 , 28 , 42 , and 56 PI . A sterile cotton swab dipped in tryptose phosphate broth was inserted into each conjunctiva . Following 5 s of swabbing in the conjunctiva , the swab was placed directly into 300 µl of sterile tryptose phosphate broth . Swabs were swirled and wrung out on the inside of the tubes to remove liquid from them before they were discarded . Samples were kept on ice and then frozen at −20°C prior to DNA extraction . Genomic DNA was extracted from all conjunctival swabs using Qiagen DNeasy 96 Blood and Tissue kits ( Qiagen ) . As per [76] , we used primers and a probe that target a portion of the mgc2 gene of MG that is conserved across songbird-derived MG isolates ( unpublished data ) . Each 25-µl reaction contained 12 . 5 µl of iQ Supermix ( 2× ) , 0 . 65 µl each of 10 µM forward and reverse primers , 0 . 35 µl of 10 mM probe , 5 . 85 µl DNase-free water , and 5 µl of DNA sample . Cycling was performed using a MyiQ Single Color Real-Time PCR Detection System ( Bio-Rad ) and the following parameters: 95°C for 3 min , and 40 cycles of 95°C for 3 s and 60°C for 30 s , with a ramp rate at 0 . 5 degrees/s . Standard curves were generated for each run . The standard was based on 10-fold serial dilutions of plasmid containing a 303-bp mgc2 insert [76] . The curve was created using 1 . 15×102 to 1 . 15×108 copy numbers . In experiment 1 , two negative control individuals were qPCR positive in one eye on a single day PI ( n = 1 on day 14 and n = 1 on day 21 ) . Because these individuals never developed lesions , they were not positive at any other point throughout the experiment , and the detected pathogen levels were very low ( log 2 . 24–3 . 08 ) and limited to one eye , these results were assumed to reflect sampling or assay error rather than exposure to MG . We used generalized linear mixed models implemented with hierarchical Bayesian methods that accounted for bird-to-bird variation in response to infection and for correlated day-to-day changes in response through time for each combination of host origin , isolate of the pathogen , and experiment ( Methods S1 ) . All models were fit using Markov chain Monte Carlo methods implemented in WinBUGS [77] . Because of the ordinal nature of the eye score response , we used the cumulative logistic link to model eye scores relative to the asymptomatic state ( eye score of zero ) ; this can be thought of as three logistic regressions that predict the probability of a positive eye score ( 1 , 2 , or 3 ) as a function of host individual , pathogen isolate , and day PI . We then used these probabilities to estimate the expected ( average ) eye score as a function of the covariates . For pathogen load data , we assumed that the observed response ( qPCR ) came from a two-step process where zero observations are modeled as originating from a different process than positive observations; these models are often referred to as zero-inflated models and are commonly used for ecological count data [78] . To make the results more interpretable , we calculated derived quantities from the estimated parameters at the mean of the individual bird effects . For eye score and pathogen load , we calculated mean response values for each isolate at each day PI ( Equation 4 , Methods S1 ) ; we refer to these as “average eye score” and “observed pathogen load , ” respectively . We quantified the pathogen component of virulence for each isolate ( “average virulence index” ) by additionally averaging over the eight sampled days PI ( Equation 5 , Methods S1 ) . We quantified an “average pathogen load” similarly for each isolate . Finally , we calculated the correlation between an isolate's average virulence index and average pathogen load in two ways . First , we calculated the correlation by removing the average of the isolates within an experiment/isolate origin ( east or west ) . This provided a minimum , conservative correlation between virulence and pathogen load , in which the variation due both to experiment and isolate/host origin were removed . We also calculated a maximum correlation coefficient as the observed effect without removing the mean of isolate/host origin . We examined the effect of experiment and host origin on virulence only for VA1994 because this isolate was used in both experiments and was therefore inoculated at different doses ( experiment 1 versus 2; Table 1 ) and into hosts of multiple geographic origins within experiment 2 ( Table 1 ) . Because we found significant host and experimental effects for VA1994 ( Figure S1 ) , our quantitative inferences are restricted to within-host-population and within-experiment comparisons only . Qualitative between-experiment inferences are made only with respect to the reference isolate ( VA1994 ) . Statistical methods , including WinBUGS code , are explained fully in Methods S1 .
|
A long-standing paradox in the study of infectious diseases is why pathogens evolve to cause harm to the very hosts they depend on to survive and reproduce . Research over several decades suggests that this harm , or virulence , is an inevitable by-product of the pathogen replication needed to maximize the chance that a given pathogen will be transmitted to another host . Here we demonstrate that a recently emerged bacterial pathogen of a North American songbird species has gradually become more virulent during each of two emergence events in different regions of the host range . This evolution of higher virulence appears to have been driven by selection for high rates of pathogen replication , because bacterial isolates that are more virulent in finches also attain the highest loads in infected host tissues . Overall , our results indicate that emerging pathogens can evolve to become more virulent in their hosts , at least in the short term , when an increase in the pathogen's ability to replicate is linked with higher virulence . Our findings have important implications for understanding and predicting the severity of disease caused by emerging pathogens in wildlife , domestic animals , and humans .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology",
"biology"
] |
2013
|
Parallel Patterns of Increased Virulence in a Recently Emerged Wildlife Pathogen
|
MicroRNAs ( miRNAs ) are small RNAs that regulate diverse biological processes including multiple aspects of the host-pathogen interface . Consequently , miRNAs are commonly encoded by viruses that undergo long-term persistent infection . Papillomaviruses ( PVs ) are capable of undergoing persistent infection , but as yet , no widely-accepted PV-encoded miRNAs have been described . The incomplete understanding of PV-encoded miRNAs is due in part to lack of tractable laboratory models for most PV types . To overcome this , we have developed miRNA Discovery by forced Genome Expression ( miDGE ) , a new wet bench approach to miRNA identification that screens numerous pathogen genomes in parallel . Using miDGE , we screened over 73 different PV genomes for the ability to code for miRNAs . Our results show that most PVs are unlikely to code for miRNAs and we conclusively demonstrate a lack of PV miRNA expression in cancers associated with infections of several high risk HPVs . However , we identified five different high-confidence or highly probable miRNAs encoded by four different PVs ( Human PVs 17 , 37 , 41 and a Fringilla coelebs PV ( FcPV1 ) ) . Extensive in vitro assays confirm the validity of these miRNAs in cell culture and two FcPV1 miRNAs are further confirmed to be expressed in vivo in a natural host . We show that miRNAs from two PVs ( HPV41 & FcPV1 ) are able to regulate viral transcripts corresponding to the early region of the PV genome . Combined , these findings identify the first canonical PV miRNAs and support that miRNAs of either host or viral origin are important regulators of the PV life cycle .
Papillomaviruses ( PVs ) comprise a large family of circular double-stranded DNA viruses . Numerous PV genomes have been described including over 200 human PV ( HPV ) types . A minority of these are known as carcinogenic agents [1–3] , however only a small fraction of hosts infected with these high risk types will go on to develop high grade lesions . It remains incompletely understood what factors dictate whether or not HPV infection will develop into malignant cancer [1 , 3] . Further , it is unclear why HPVs that share a high level of sequence similarity can have stark differences in tropism and infect different regions of the body . Developing a better understanding of PV gene products and their regulation throughout diverse PV lineages provides an evolutionary foundation for deciphering the mechanisms resulting in differential outcomes of infection . MicroRNAs ( miRNAs ) are small regulatory RNAs that are an emerging class of viral gene products found in select virus families [4–7] . miRNAs are approximately 22 nucleotides long and function by docking to specific target mRNAs to repress translation [8 , 9] . miRNAs derive from primary transcripts containing a hairpin structure that is processed by a series of endonucleases ( Drosha in the nucleus , Dicer in the cytosol ) generating the final effector RNA [10–15] . The miRNA then enters a multi-protein complex called the RNA Induced Silencing Complex ( RISC ) where it scans mRNAs and docks to regions of partial sequence complementarity [8] . The so-called seed region of miRNAs , approximately 6 or more nucleotides towards the 5' end , typically binds transcripts with perfect complementarity and this feature can be used to help identify mRNA targets [8 , 16 , 17] . miRNAs of a particular sequence typically function by subtly regulating numerous ( 10 or more ) mRNA transcripts involved in a particular biological outcome . Although individual miRNA regulation of any single transcript can be subtle , the sum of regulation by multiple miRNA-bound-RISC complexes ( miRISC ) can add up to significant biological activity [8] . miRNAs of host or viral origin have been implicated in various processes relevant to virus infection including the control of the immune response , cell death , transformation and virus gene expression [18–29] . Over 300 viral encoded miRNAs have been described , all from viruses able to undergo long term persistent infection [4 , 5 , 7 , 30] . Most of these viruses have DNA genomes including the herpes , polyoma , and anello virus families [5] . However , some retroviruses including the delta retrovirus bovine leukemia virus ( BLV ) and foamy retroviruses also encode miRNAs [21 , 31–33] . One likely role of viral miRNAs is to foster long-term interactions within the host [30] . In this regard , at least some members of the PV family would be expected to encode miRNAs . However , to date , no credible examples of PV canonical miRNAs have been described . Two studies examining fully infectious experimental systems of the high-risk HPV18 & 31 types report that these viruses do not encode miRNAs [34 , 35] . Further , at least two studies examining transformed cell lines report that HPV16 does not encode miRNAs [36 , 37] . There have been reports in transformed cells of small RNAs from high risk PVs such as HPVs 16 & 18 , however these studies did not demonstrate a connection of PV-derived small RNAs to the miRNA biogenesis ( Dicer/Drosha ) or effector ( RISC ) machinery , nor did they confirm a biologically meaningful abundance [38–40] . Consequently , these RNAs are not widely accepted as miRNAs and the signal detected likely represents degradation fragments derived from the turnover of longer transcripts . In contrast , it is well documented that PV infection and individual PV gene products can alter the host miRNA repertoire , likely contributing to the biology of cancer [25 , 35 , 41–44] . Furthermore , at least one PV , HPV31 , utilizes a host miRNA to directly regulate early viral gene expression [25] . Thus , what emerges is that although host miRNAs are involved in the PV life cycle and pathogenesis , of the few PV types that have been examined , no canonical PV miRNAs are yet established . One barrier to discovery of PV miRNAs is the dearth of facile fully-infectious laboratory systems . There are experimental systems established for a few PV types ( approximately < 5 ) [45] , but technological barriers have limited any comprehensive large-scale study of viral miRNAs in the majority of PV types . Here we describe a new wet bench approach for the discovery of miRNAs that assays numerous viruses in parallel for the ability to express miRNAs . We identify bona fide PV miRNAs encoded by divergent PVs , and demonstrate these miRNAs depend on canonical miRNA biogenesis effector machinery . Our analysis also rules out canonical papillomaviral miRNAs in cancers associated with high-risk PV ( HPVs 16 , 18 , 31 , 45 , and 58 ) infection . These results provide further evidence for the relevance of miRNA-mediated regulation of PV transcripts .
To identify miRNA genes in situations where transcripts are not easily obtainable , we developed the approach of miRNA Discovery by forced Genomic Expression ( miDGE ) . miDGE relies on generating a library of numerous overlapping genomic segments of DNA from a particular organism or locus and subcloning them behind a heterologous RNA polymerase ( RNP ) II promoter ( Fig 1 ) . The concept relies on the principle that miRNA genes are compact and should be readily expressed by heterologous upstream RNP II promoters , or in the rare cases that a primary miRNA transcript is driven by RNP III , that these promoters are small and proximal to the miRNA gene so as to be included in miDGE library constructs . The miDGE library is then transfected into mammalian cells and small RNA is harvested and sequenced . Next , we apply computational methods to identify miRNA candidates whose transcripts display the hallmarks of processing by the miRNA biogenesis machinery . Finally , candidate miRNAs are validated via a series of molecular assays to establish biogenesis via the canonical miRNA processing machinery and activity within RISC , the miRNA silencing machinery . To test the effectiveness of the miDGE approach before applying it to PVs , we first focused on a single larger genome virus , the herpesvirus Japanese Macaque Rhadinovirus ( JMRV ) . JMRV is a gamma-2 herpesvirus with genomic sequence similar to the highly related Rhesus Rhadinovirus ( RRV ) . When we initiated these studies , it was not yet known if JMRV encoded miRNAs , although this would be expected since numerous precursor miRNAs ( pre-miRNAs ) had already been identified in RRV , which shares high sequence similarity with JMRV [46] . Indeed , work from Skalsky et al . has now identified 15 novel viral miRNA encoded by JMRV [46] . Thus , JMRV serves as a proof-of-principle “test genomic space” to evaluate miDGE . A cosmid with an approximately 36 kilobase ( kB ) region of the JMRV genome that encompassed the region with positional homology to the RRV miRNA cluster was fragmented by sonication and used to construct a miDGE expression library . DNA-seq analysis confirmed that the library covered the entire region carried by the original cosmid construct ( Fig 2A , top panel ) . We then transfected our library into HEK293T cells and analyzed small RNA ( sRNA ) expression by sRNA-seq . In parallel , we also performed small RNA-sequencing from fibroblasts that had been infected with JMRV in vitro . As shown in Table 1 , approximately 13% of all reads from infected fibroblasts mapped to viral genomes , consistent with the fact that JMRV establishes a productive infection and replicates to high titers in such cultures . As expected , the relative fraction of viral reads was much smaller ( < 0 . 1% ) in cultures transfected with the miDGE library . In both cases , the bulk of host reads originated from bona fide small RNA species , indicating no or little contamination of our small RNA preparation by degradation or breakdown products of longer RNA molecules . As shown in Fig 2A and 2B , small RNA-seq reads mapped across the JMRV genome in patterns that were similar between the miDGE and infection experiments , with the great majority of reads ( 96% and 88% of viral reads , respectively; see Table 1 , S1 Table ) aligning to the genomic location of the 15 pre-miRNAs previously identified by Skalsky and colleagues . Comparison of coverage profiles across individual pre-miRNAs furthermore suggested that most precursors were processed to produce ratios of mature 5p- and 3p-species that were likewise similar ( S1 Fig ) between miDGE and infection experiments . In addition to miRNA reads , infected cells produced a profile of low level seemingly randomly scattered reads ( see log-scale plots in Fig 2B ) , likely originating from breakdown products of viral mRNAs in lytically infected cells . Such background was substantially lower in our miDGE experiments . The data were next used to perform a de-novo prediction of pre-miRNAs with the miRDeep2 algorithm [47] , providing the pipeline only with an annotated reference set of known host ( i . e . , human ) miRNAs deposited in miRBase v21 , but not any miRNAs of viral origin . When processed with the sRNA-seq data from the miDGE library , this analysis readily identified 14 of the 15 known pre-miRNAs in JMRV ( Table 2 , S2 and S3 Datasets ) . Another two predictions mapped to the opposite strand of miRs-jR1-2 and -8 , consistent with the previous observation that some herpesvirus pre-miRNAs can produce mature products when transcribed in the antisense orientation [48] . As our approach is agnostic of transcriptional directionality in the parental viruses , it is plausible that such miRNAs will register in the analysis . A single JMRV miRNA , miR-jR1-7 , evaded detection by miDGE . Inspection of sequencing and mapping data revealed that this was not due to the principal absence of miRNA products , given that 14 reads had been correctly mapped to mature miR-jR1-7 species ( see S1 Table and coverage profiles in Figs 2B , 2C and S1 ) . While this number is relatively low , miRs-jR1-10 and -11 were accurately identified despite read counts that were in a similar range ( see Table 2 ) . Indeed , miRDeep2 analysis of small RNAs from JMRV-infected cells also failed to identify miR-jR1-7 , even though more than 2000 reads had mapped to mature species from this miRNA ( S1 Table and Figs 2B , 2C and S1 ) . Although we do not know the precise reason for the pipeline’s inability to predict miR-jR1-7 , we suspect that the close proximity of miRs-jR1-7 and -6 may have interfered with delineation of candidate precursor sequences that are subjected to structure prediction . Given the above , we conclude that , at least in the context of a herpesviral genome and within the detection limits set by the bioinformatic prediction pipeline , miDGE can successfully identify bona fide miRNAs at sensitivity levels that are on par with those observed in an authentic infection system . Our goal was to screen numerous PV genomes representing diverse clades in the PV family for viral miRNAs . To accomplish this , we collected 113 cloned PV genomes from both human and non-human animal sources ( S2 Table ) . Additionally we included two polyomaviruses ( Simian Virus ( SV40 ) and Merkel cell polyomavirus ( MCPyV ) ) known to express viral miRNAs [23 , 49] , as positive controls in our PV library . Since the PV/PyV constructs were considerably smaller than our JMRV cosmid and thus potentially less sensitive to mechanical shearing , we utilized three different 4-base pair cutter restriction enzymes in addition to the sonication procedure to fractionate the viral DNA ( see material and methods for details ) . High throughput Illumina DNA sequencing revealed that we had high coverage ( S2 Table , S1 Dataset ) of numerous genomes . A total of 73 PV genomes had greater than 95% coverage ( see S2 Table ) . As shown in Fig 3A , this set contained representatives from the majority of known PV clades . We next conducted high throughput sequencing of small RNAs from cells transfected with the PV libraries . As expected based on our previous analysis of the JMRV library , we again observed that only a relatively small fraction ( 0 . 08% ) of reads mapped to viral genomes ( Table 1 , right column ) . Likewise , the majority of host reads were derived from bona fide small RNA species , indicating that the RNA preparations were largely free of contaminating RNA degradation products . We next applied our computational methods based on miRDeep2 to identify small RNA reads consistent with bona fide miRNAs . The algorithmic approach prioritizes those small RNAs of appropriate size ( between 17–24 nucleotides ) that had a read distribution as being plausibly derived from a pre-miRNA hairpin . This analysis identifies signatures of bona fide miRNAs including the presence of a predicted stem loop structure and clearly defined 5’ ends as processed by Dicer/Drosha . Typical read density coverage plots of such signatures exhibit two "plateaus" , where each plateau represents a miRNA derivative from either the 5' or 3' arm of a pre-miRNA hairpin ( 5p or 3p miRNA , respectively ) ( see , for example , Figs 2C and 3B ) . In between each plateau is a trough in density coverage where the terminal loop portion of the pre-miRNA is under-represented in our libraries due to terminal loops of processed pre-miRNAs not being stabilized in RISC as miRNA derivatives are . miDGE readily called miRNAs from the positive control polyomavirus genomes included in the libraries . The vast majority of the greater than 660 , 000 combined nucleotide PV genomic space covered did not produce miRNA candidates , consistent with a low false positive rate for miDGE . However , miDGE did call five high-scoring miRNA candidates which , like our SV40 and MCPyV positive controls , were awarded confidence levels of >95% by the miRDeep2 prediction algorithm ( Table 3 and Fig 3B–3D ) . These five high-scoring candidates originated from four different PV genomes ( one each from HPV41 , HPV17 , HPV37 , and two from FcPV1 ) . MiRDeep2 called an additional 7 candidates with lower scores and/or confidence levels ( see S2 and S3 Datasets for all PV predictions made by the pipeline ) . One of these mapped to the FcPV1 genome , i . e . , the same genome which had also produced two high scoring candidates . We noted that the read sequences for another three of the lower scoring predictions ( HPV types 5 , 49 and 105; marked ‘l . c . ’ in S2 Dataset ) were of low complexity . Indeed , applying an entropy-based filter ( see Material and Methods ) to remove low-complexity reads prior to our bioinformatic analysis eliminated these predictions ( but none of the other candidates ) . We thus deem it likely that the predictions made for HPVs 5 , 49 and 105 result from cross-mapping of reads that may originate from repetitive regions . The predicted hairpin structures ( S4 Dataset ) of the remaining lower-scoring candidates have features that are unusual for canonical miRNAs , for example a very short distance ( 2 to 4 nucleotides ) between 5p and 3p reads ( HPV113 and HPV3 ) or a 5p read that overlaps with much of the predicted terminal loop structure ( PePV ) . We therefore suspect that these candidates are likewise false positives . However , as we have not attempted to experimentally verify these three predictions , we cannot exclude the possibility that one ( or more ) of them may represent authentic miRNAs . Our bioinformatic analysis did not predict any candidates for the negative strand of papillomaviruses . Likewise , although their genomes were fully covered in our library , no predictions were made for any of the plus- or minus-strand miRNAs previously suggested for HPV types 6 , 16 , 18 , 38 or 45 [38–40] . While highly unlikely , we nevertheless considered it formally possible that our bioinformatic prediction pipeline had missed all 9 purported candidates in these genomes . Therefore , we inspected small RNA read coverage across the respective genomic regions . As shown in S2 Fig , our high confidence PV candidates and the PyV positive controls exhibited the typical coverage profiles of bona fide miRNAs . In contrast , only few reads mapped to the previously suggested papillomavirus pre-miRNA regions in HPV types 6 , 16 , 38 and 45 ( S3A Fig ) , and these reads furthermore did not match the distribution as expected for mature miRNAs being derived from the purported hairpin precursors ( S3 Fig ) . While the absence of read coverage argues against the existence of miRNAs in above HPV species , for one of the purported miRNAs ( HPV18- LCR ) , we indeed observed a strong signal in our miDGE analysis ( Fig 4A , top panel; see also S3B Fig for a detailed depiction of the covered region ) . However , we also noticed that the mature miRNA sequences that supposedly derive from this region of the viral genome [39] are of very low complexity ( trinucleotide entropy < 63 ) . In agreement with this observation , when we employed the same complexity filter as used for our miRDeep2-predicted candidates , the peak was completely eliminated ( Fig 4A , panel labeled ‘PV filtered’ ) . Given this and the absence of read coverage in adjoining regions of the viral genome , we suspected that the signal originated from cross-mapping host RNAs . To investigate this hypothesis , we mapped the small RNA-reads from our JMRV experiment to the PV genomes . Indeed , as shown in the panel labeled ‘JMRV’ of Fig 4A , this dataset contains a largely identical set of sequences which produced a peak at the exact same location of the purported miRNA ( see S4 Fig for an enlarged view of the peak ) , even though the JMRV miDGE library does not contain any PV sequences . In contrast , reads originating from all of our high scoring miRNA candidates in FcPV1 , HPV17 , HPV37 and HPV41 were not eliminated by complexity filters , and were furthermore highly specific for the PV libraries ( shown exemplary for HPV17 in the lower panels of Fig 4A ) . Together , these results demonstrate the plausibility of using miDGE to identify miRNAs from complex multi-genome expression libraries and suggest that while some PVs may encode miRNAs , most PVs , including the high-risk types 16 , 18 , and 45 , do not encode canonical miRNAs . To further evaluate whether high risk HPVs code for miRNAs , we analyzed the large and small RNA transcriptomes of 303 cervical carcinomas that are part of The Cancer Genome Atlas ( TCGA ) [50] . We limited our subsequent analysis to 213 solid tumors that had both large and small RNA-seq datasets and at least 50% RNA coverage of a particular HPV genome in the large RNA-seq ( Fig 4B ) . Importantly , we observed 42 samples that had >99% RNA coverage for one of the following HPVs: 16 , 18 , 31 , 45 , or 58 . Because these samples displayed essentially complete RNA coverage throughout the entire relevant HPV genome , it would be expected that any putative miRNA encoded in these genomes would have the ability to be processed and detected in these samples . We applied the miRDeep2 pipeline to each small RNA library to identify putative miRNAs . We observed hundreds of miRBase annotated host miRNAs per library ( 255–454 ) , but no HPV derived candidates ( Fig 4C and 4D ) . However , we were able to identify Epstein-Barr virus ( EBV ) miRNAs in ~14% of the tumor samples , demonstrating that we are able to detect viral miRNAs that are present in only trace amounts , likely from EBV-infected infiltrating lymphocytes ( Fig 4C ) . Thus , despite having transcripts that span the entire HPV viral genomes , tumors associated with high risk HPV types do not express canonical HPV-derived miRNAs . Qian et al . previously reported HPV16-derived viral miRNAs present in small RNA-seq libraries prepared from transformed cell lines as well as cervical tissue and cancer samples [40] . We re-analyzed their deposited small RNA transcriptomic datasets [40] for the expression of HPV derived small RNAs . Similar to the HPV-associated tumors in the TCGA , we observed many miRBase-annotated host miRNAs per library ( 50–100 ) , but no HPV-derived candidates ( Fig 4E and 4F ) . Also in contrast to Qian et al . , we observe few reads mapping to either HPV16 reference sequence ( < 20 total reads per library ) . Furthermore , we do not observe any reads for the purported HPV16 mature miRNAs in these datasets ( Fig 3F ) . Inspection of the alignments presented by Qian et al . [40] suggests that an excessive allowance for sequencing errors and small nucleotide polymorphisms ( SNPs ) combined to produce false read mappings to the purported HPV16 miRNAs . Thus , our re-analysis suggests that the small RNAs described in their study may be artifacts of the bioinformatics analysis and do not represent actual small RNAs derived from HPV16 . Combining these findings with our miDGE results ( Figs 4A and S3 ) , we conclude that it is highly unlikely that high risk HPVs 16 , 18 , 31 , 45 and 58 express canonical viral-encoded miRNAs . The burden of proof for establishing bona fide miRNAs includes evidence of specific processing by the miRNA machinery and silencing activity within RISC . To vet the five high-scoring PV candidate miRNAs , we first conducted northern blot analysis . Northern blot analysis can provide information about the size and processing of pre-miRNAs and derivative miRNAs . We cloned the candidate miRNA genes and flanking regions downstream of an RNP II promoter and transfected these plasmids into HEK293T cells . We harvested total RNA and conducted northern blot analysis . For each candidate , bands migrating at positions consistent with the appropriate size typical of miRNAs were observed . Additionally , on most blots , a clear band consistent with a pre-miRNA was also observable . We note that due to the high sequence similarity of the HPV 17 and 37 miRNAs ( ~91% ) , even though we confirmed the blots were completely stripped of specific signal , we consistently observe cross-reacting signal from both lanes when probed with either probe ( Fig 5A & 5B ) . Overall , this analysis showed that all five high-scoring candidates gave rise to banding patterns consistent with canonically processed miRNAs ( Fig 5A ) . To further validate these candidate miRNAs , we investigated their biogenesis , activity and expression . To determine if the biogenesis of these candidates required canonical miRNA machinery , we transfected our candidate miRNA constructs into cells with Drosha levels knocked down or that had Dicer gene expression knocked out ( Fig 5B and 5C ) . As expected , this analysis showed that positive control SV40 miRNAs were dependent on both Drosha and Dicer ( Fig 5B and 5C , respectively ) . All five candidate PV miRNAs showed reduced expression upon knockdown of Drosha ( Fig 5B ) and reduced ratios of miRNA:pre-miRNA in the absence of Dicer ( Fig 5C ) . These results conclusively demonstrate that the five high-scoring candidate PV miRNAs derive from canonical miRNA biogenesis . We next generated luciferase-based RISC reporters for each viral miRNA candidate to determine if they are active in RISC . Reporters contained two perfectly complementary sequences to respective miRNA candidates in the 3’ untranslated region of Renilla luciferase . Negative control reporters contained the same sequences , with three nucleotides in the miRNA seed complement site substituted to disrupt binding . Transfection of these reporters resulted in specific knockdown of luciferase in each of the five reporters when co-transfected with respective miRNA expression vectors , but not in negative control reporters ( Fig 6 ) . These results suggest that each of the identified miRNA/candidates undergo canonical processing and are active in RISC . A final criterion for confirmation of bona fide viral miRNAs includes detectable expression in infected cells or tissues . To date , we have been unable to obtain published or unpublished datasets to uncover samples infected with HPVs 17 , 37 or 41 . It would be informative in future studies to examine the expression/activity of these miRNAs if relevant human datasets or cell culture models become available . However , FcPV1 infection is recognized as a cause of macroscopic proliferative skin lesions affecting the legs and feet of chaffinches ( Fringilla coelebs ) with deeply fissured papillary growths ( i . e . , papilliferous ) , subsequently referred to here as ‘leg lesions’ [51] ( Fig 7C inset ) . Therefore , we investigated the presence of candidate PV miRNAs in available leg lesion samples from wild chaffinches . We were able to obtain RNA from host tissue which tested positive for FcPV1 DNA by PCR [52] . We sampled both available leg lesion and apparently normal pectoral muscle tissue as a negative control from the same birds . We performed small RNA-seq profiling of the libraries and compared this with the data derived from our miDGE libraries ( Fig 7A and 7B ) . Similar to our miDGE results , in the library prepared from leg lesions , we observed a pattern of reads mapping to the two confirmed pre-miRNAs ( Fig 7B and 7C ) . Further consistent with miDGE , this analysis also identified a lower abundance , low-scoring third candidate miRNA ( provisionally “candidate fcpv1-miR-F3” ) in the L2 genomic region ( Fig 7B and 7C ) . Other small RNAs mapped to the FcPV1 genome but these did not match a pattern consistent with miRNA biogenesis and likely represent nonspecific degradation of larger FcPV1 transcripts . In contrast , few read mappings were observed in the library prepared from negative control pectoral muscle ( Fig 7D ) . Combined with the above biogenesis and activity data , these results demonstrate that FcPV1 infection gives rise to at least two bona fide PV miRNAs . Additionally , of the remaining PV miRNA candidates ( HPVs 17 , 37 & 41 ) , we can say with confidence they are highly probable miRNAs , meeting all criteria of bona fide miRNAs with the exception of detection in infected cells . In keeping with the miRNA naming convention set forth by miRBase [53] , we name these miRNAs/candidate miRNAs: fcpv1-miR-F1 , fcpv1-miR-F2 , hpv17-miR-H1 , hpv37-miR-H1 and hpv41-miR-H1 . Previous studies in the small DNA virus polyomavirus ( PyV ) family demonstrated that PyV miRNAs directly regulate early viral transcripts [19 , 54 , 55] . Further , bandicoot papillomatosis carcinomatosis viruses ( BPCVs ) , that have hybrid PyV-like early genes and genomic organization but PV-like capsid genes , also regulate early viral transcripts via viral miRNAs [56] . Therefore , we performed bioinformatic analysis to examine the possibility that PV miRNAs could regulate early viral gene expression . To identify putative viral target transcripts , we identified seed complementary sites of 7 or more nucleotides for each PV miRNA in its respective genome . This analysis revealed candidate target sites for derivatives from each of the five high confidence PV pre-miRNAs ( S3 Table ) . Notably , both FcPV1 and HPV41 had candidate viral miRNA docking sites within the E1/E2 regions of their genomes , which has previously been demonstrated to include transcripts regulated by a host miRNA for HPV31 [25] ( Fig 3D ) . We therefore tested these possible PV mRNA docking sites by engineering chimeric luciferase reporters containing either the entire E1/E2 region ( termed “Early” ) or sub-portions of the E1 or E2 genetic region encompassing a single predicted site of each respective genome ( Fig 8 ) . Co-transfection of the “Early” reporter plasmids with individual PV miRNA expression vectors revealed that “Early” reporters for the FCPV1 and HPV41 genomes display significantly less expression in the presence of their respective viral miRNAs ( Fig 8 ) . Co-transfection of FcPV1 miR-F1 alone reduced expression of the FcPV1 E1/E2 reporter and this regulation was enhanced by co-transfecting plasmids expressing both FcPV1 miRs-F1 and F2 ( Fig 8A ) . As expected , co-transfection of the respective viral miRNA with the predicted single site reporters demonstrated a significant reduction in luciferase expression . Importantly , negative control reporters containing nucleotide mutations in the seed complement region of each predicted PV miRNA docking site alleviated the repression that we observed . When the same experimental setup was performed with the HPV41 miRNA and reporters , we observed similar results ( Fig 8D ) . Co-transfection of the HPV41 E1/E2 reporter and miRNA expression vectors demonstrated a significant reduction in luciferase expression . Co-transfection of the HPV41 E1 reporter , but not the E2 reporter , resulted in a similar reduction as the full E1/E2 genomic region reporter , suggesting the E1 region contains the most relevant miRNA docking site . In contrast , the negative control HPV41 E1 mutant reporter was not regulated in response to miRNA expression vector transfection ( Fig 8B–8D ) . These results demonstrate that FcPV1 and HPV41 miRNAs/candidates are able to directly regulate transcripts corresponding to the PV early genomic region .
Members of diverse virus families express miRNAs [4 , 5 , 7 , 30] . These include the herpesviruses , polyomaviruses , anelloviruses , and retroviruses [6 , 21 , 31 , 32] . Notably , all of these viruses undergo persistent infection , have access to the nucleus where key miRNA processing machinery resides and are exclusively DNA viruses or have a DNA component to their lifecycle . Viruses with a persistent component to their life cycle may especially benefit from the typically subtle regulation afforded by miRNAs . Based on these characteristics , at least some members of the PV family would be expected to encode miRNAs . Yet , until now , no widely accepted PV miRNAs are known . Here we report the first high confidence papillomavirus-encoded miRNAs from a minor subset of PVs . We identify PV miRNAs from chaffinch leg lesions and highly probable miRNAs from the human PVs 17 , 37 , and 41 . We designate the latter as “highly probable” because although they passed stringent cell culture-based criteria , due to a lack of relevant samples , we have not yet verified their existence in vivo . All five display the hallmarks of canonical miRNAs ( Fig 5 ) , including being processed by Dicer and Drosha and being highly active in RISC [8] . These RNAs derive from three divergent clades of PVs . HPVs 17 and 37 are in the beta clade , some members of which have been proposed to have a role in skin cancers [2] . The pre-miRNA hairpin region of the L2 locus for these viruses appears to have evolved in a common ancestor of these viruses , and may be shared with closely related HPVs 15 & 80 . Sequence-structure alignments suggest the existence of a highly conserved hairpin in these viruses . However , since HPVs 80 and 15 were either not included ( HPV80 ) or only very poorly covered ( HPV15 ) in our library , additional experiments will be required to determine whether they indeed produce conserved miRNAs . HPV41 is notable in that it is the sole member of the Nu clade and is one of the few PVs that have starkly different locations in a PV family phylogenetic tree , depending if the tree is built upon the late ( L1 ) or early proteins ( E1 ) . This implies HPV41 is likely a hybrid virus that arose from recombination [57] and may help to explain its atypical ability to encode a miRNA . FcPV1 is only distantly related to human PVs but its association with highly keratinized hyperplastic lesions allowed us to obtain RNA from PV-associated diseased chaffinch tissue . This confirmed that FcPV1 miRNAs are expressed in vivo ( Fig 7 ) and further confirmed their status as bona fide viral miRNAs . The HPV41 miRNA and one of the FcPV1 pre-miRNAs are found in non-protein-coding genomic locations , a common feature of most miRNAs . Notably , one of the FcPV1 miRNA loci , fcpv1-miR-F2 , and the HPV41 miRNA locus are both found in similar locations downstream of the late genes past the likely late polyadenylation signal sequence . In contrast , the HPV17 and 37 miRNAs , as well as the other abundant FcPV1 miRNA ( as well as the low abundance FcPV1 candidate ) , are found in a similar genomic region overlapping and in the same transcriptional orientation as the L2 locus ( Fig 3D ) . Although these miRNAs could derive from an intronic primary transcript , it is nonetheless atypical for a pre-miRNA gene to overlap a protein-coding gene . Similar genomic arrangements are observed for miR-BHRF1 and the EBNA-LP gene in Epstein Barr Virus [58] and miR-K12 and the Kaposin gene in KSHV [48 , 59] . For KSHV , Drosha can suppress Kaposin expression in latent infection , but its steady state levels and consequent ability to regulate Kaposin levels decrease during stress and at late times of lytic infection [60 , 61] . Therefore , it is conceivable that FcPV1 and HPVs 17 & 37 similarly utilize Drosha to aid in controlling the expression of L2 . HPV17 and 37 are the only PV miRNAs/candidates that share a high degree of sequence identity ( 91% 3P , 91% 5P ) , consistent with them being derived from virus types that are closely related . Except for HPVs17 and 37 , there is no obvious relationship between the miRNA-positive PVs that might account for why they would preferentially encode miRNAs . miDGE covered at least 698 , 000 nucleotides of sequence space ( the sum of viral genome sequences covered by our JMRV and PV libraries ) and called only few candidates other than the positive controls and the five high-confidence PV miRNAs that we validated . Therefore , we conclude that miDGE has an intrinsic low false positive rate . These findings are consistent with the current understanding that pre-miRNA hairpins have specific structural features that are required for efficient processing [62–64] . It should be noted , though , that since miDGE forces ectopic expression of genomic sequences , its predictions require independent validation to ensure that potential candidates are also expressed by authentic viral transcriptional units . Compared to false positive rates , it is more difficult to estimate the frequency of false negatives in our approach . Generally , our analysis of JMRV suggests that the majority of bona-fide miRNAs are readily identified by miDGE . The sole JMRV miRNA that was missed by miDGE also evaded detection when we analyzed material from infected fibroblasts , demonstrating failure to identify jmrv-miR-jR1-7 was due to limitations of the miRDeep2 analysis pipeline rather than the miDGE protocol itself . The miRDeep2 algorithm identifies pre-miRNA candidates based on expected read coverage profiles produced by mature 5p- and 3p-miRNA species . These profiles are used to “excise” sequences for prediction of potential pre-miRNA hairpin structures . Based on the inspection of candidate sequences analyzed by the pipeline , we suspect that the close proximity of clustered miRNAs in the JMRV genome had led to an inaccurate excision of the jmrv-miR-jR1-7 precursor . As closely clustered miRNAs are a typical feature of herpesvirus genomes , such limitations should not severely impede our ability to identify PV miRNAs . Instead , we consider it more likely that some bona fide miRNAs in the 113 PV types that we strived to analyze may have evaded detection due to incomplete coverage of their genomes . In fact , only a subset of PV genomes ( n = 63 ) had near-complete ( >99% ) coverage in our DNA libraries . Therefore , we can only make negative conclusions on this limited set of PVs . Our results show that 59/63 PV genomes with near-complete DNA coverage lack the ability to efficiently give rise to canonical miRNAs . This list includes high-risk HPVs such as HPV16 which our further transcriptomic analysis demonstrated does not encode miRNAs in tumors or cancer cell line settings ( Fig 4A–4G ) . Although we acknowledge that our findings are skewed toward the alpha clade human PVs , we conclude that numerous and diverse PVs lack the ability to encode their own miRNAs . What are the functions of the PV miRNAs ? Our bioinformatic analysis and reporter assays ( Figs 4A–4G and 8A–8D ) suggest that one function of FcPV and HPV41 miRNAs is to regulate viral gene expression . This notion is consistent with the known function of other viral miRNAs , especially those derived from the PyVs and BPCVs [23 , 56] . Although we did not predict high confidence docking sites in the E1/E2 region for the HPVs17/37 miRNAs , until experimentally tested , we cannot rule out regulation of viral gene expression by these miRNAs . Since our results suggest that many PVs do not encode miRNAs , this raises the question of if/how such viruses fine-tune their own gene expression . For at least one high risk PV , HPV31 , the answer seems to be by co-opting host miRNAs [25] . Gunasekharan et al . demonstrated that miR-145 is expressed at higher levels in undifferentiated versus differentiated keratinocytes , displaying an inverse pattern to HPV31 replication levels . miR-145 negatively regulates HPV31 genome replication and gene expression . Interestingly , miR-145 directly docks to and regulates the E1/E2 transcripts in a genomic region similar to that we have uncovered for FcPV1 and HPV41 miRNAs . Our bioinformatic analysis of all human miRNAs and fully sequenced PV genomes ( S9 Dataset ) , similar to the published work of Gunasekharan et al . , suggests many other PVs could utilize a similar host miRNA strategy to regulate the E1/E2 transcripts . Further , we observed a near perfect complementary putative target site for Let7 miRNA in the FcPV1 genome in the late 3' UTR , implying regulation by small interfering ( siRNA ) -like RISC-mediated mRNA cleavage ( S4 Fig ) . These results further support the likelihood that diverse PVs utilize host miRNAs to regulate viral gene expression . Combined , these findings suggest that miRNAs of host and/or viral origin are utilized by PVs to optimize viral gene expression . Our results demonstrate that miDGE can be a fruitful approach when applied to numerous viruses . However , while miDGE allowed us to make reasonable conclusions for ~63 PV genomes , some genomes that we intended to include in our miDGE procedure were under-represented in the final libraries . Although we believe one major reason for this discrepancy is due to incorrect PV genomic plasmids included in our original library pools , library coverage should be optimized in future applications of miDGE . Since miRNAs are generally stable , miDGE could be used to identify biomarkers for gene expression of unculturable pathogens . In addition to pathogen genomes that cannot be grown in culture , miDGE may have utility for identifying miRNAs expressed in only a few rare cells of an organism . For example , miR-Lys6 is only expressed in fewer than 10 cells in Caenorhabditis elegans and had been missed by most standard miRNA biochemical identification procedures [65] . It is likely that similar miRNAs exist in complex multicellular organisms and these could be identified using miDGE . In summary , we have developed wet bench technology that can identify miRNAs from genomes for which complete transcriptomes are not readily available , whether viral or otherwise . This approach opens the possibilities for miRNA discovery to the enormous range of pathogens for which genomic data is available , but are unculturable in a laboratory setting . In this initial study , our approach uncovered five new PV highly probable/bona fide miRNAs . As viral miRNAs often alter host targets conducive to infection , it will be interesting to determine any relevant host targets of these miRNAs . Moreover , as our work lends additional support to the role of miRNAs in control of the PV life cycle , it will be important to determine if variability in miRNA expression or activity can contribute to the differences in tropism and pathogenesis associated with the various PV types .
For the construction of our JMRV miDGE library , a cosmid containing ~38kb of the viral genome ( nts 83 , 148 to 119 , 569 ) was sonicated to produce fragments with an average size of 300-400bp . Sub-genomic fragments in the appropriate size range were purified from agarose gels , blunted and phosphorylated using the Epicentre End-It End-Repair kit and cloned into an pDNA3 vector to produce a miDGE expression library . For the PVs , plasmids containing cloned PV genomes ( S2 Table ) were collected from various labs . We reasoned that , due to their substantially lower size , PV genomes might shear less efficiently compared to the JMRV and therefore utilized restriction enzyme digestion in addition to the sonication protocol to produce the PV miDGE library . For this purpose , the collection of PV plasmids and the positive controls SV40 and MCPyV was divided into four sub-groups containing between 27 and 46 genomes , followed by digestion with three different 4-base pair cutter restriction enzymes ( BsrFI , BstyI and EaeI ) . The resulting fragments were then cloned into compatible sites of the pcDNA3 . 1 vector ( AgeI , BamHI and NotI , respectively ) . Coverage of individual genomes in the JMRV sonication library , or the PV fragment libraries produced by restriction-digestion or sonication was determined via high-throughput sequencing on Illumina HiSeq 2500 and MiSeq systems using the TruSeq DNA Library Prep Kit . Dataset S1 provides coverage profiles of PV genomes in bedgraph format , the percentage of individual genomes covered in our miDGE libraries is listed in S2 Table . We transfected the JMRV and PV expression libraries into HEK293T via lipofection , harvested total RNA using PIG-B [66] , and size fractionated the isolated RNA via excision from a 15% denaturing polyacrylamide gel to enrich for RNA in the size class between approximately 10–35 nucleotides . This RNA was then used to produce small-RNA libraries for Illumina sequencing . In parallel , as a positive control we generated small-RNA-seq libraries with RNA from JMRV-infected fibroblasts , kindly provided by Scott Wong ( Vaccine and Gene Therapy Institute , Oregon Health & Science University , Beaverton , Oregon , USA ) . Since different library preparation methods can potentially result in a bias that can result in underrepresentation of individual miRNA species , we used the TruSeq Small RNA Sample Preparation Kit ( Illumina ) as well as the NEBNext Small RNA Library Kit ( New England Biolabs ) to generate Illumina-compatible sequencing libraries . The libraries were then sequenced on a HiSeq 2500 System ( 50 cycles , single end reads ) producing a total of 30 million raw reads for JMRV-infected fibroblast , or 112 and 712 million reads for HEK293T cells transfected with JMRV or PV miDGE libraries , respectively . After trimming , reads were first mapped with bowtie v1 . 2 . 1 . 1 ( options -n 0 -e 80 -l 18 -a -m 5—best—strata ) to viral genomes ( see S2 Table for PV accession numbers ) to investigate viral read coverage . Non-aligned reads were subsequently aligned to the human transcriptome ( ENSEMBL release-91 GRCh38 cDNA and non-coding RNA sequences ) to elucidate mapping to different RNA species as shown in Table 1 . Viral and host reads were then extracted from the bam alignment files , merged and collapsed using the mapper . pl script of the miRDeep2 v2 . 0 . 0 . 8 package [47] . The collapsed reads were then used to perform a miRDeep2 prediction of novel miRNAs , using default options and , in addition to the collapsed reads and viral genomes , providing the pipeline with the set of human precursor and mature miRNA sequences from miRBase v21 [53] as a reference . To eliminate potential low-complexity reads , prior to prediction of novel miRNAs or mapping to viral genomes , collapsed FASTA reads were optionally filtered using the prinseq-lite package v0 . 20 . 4 [67] , using an entropy cutoff value of 70 . Bam files containing all primary reads aligned to viral genomes or the host transcriptome are available via the European Nucleotide Archive ( ENA , https://www . ebi . ac . uk/ena ) under accession number PRJEB25054 . HEK293 or HEK293T cells were originally obtained from ATCC and maintained in DMEM supplemented with 10% ( vol/vol ) FBS and Pen/Strep ( Cellgro ) . HEK293T Dicer KO cells were obtained from Dr . Bryan Cullen , Duke University ( NoDice , [68] ) . All cells were grown at 37C in the presence of 5% CO2 . Plasmids containing the genomes of the papillomaviruses used in this study were obtained from labs indicated in S5 Dataset . A cosmid containing nts 83 , 148 to 119 , 569 of the JMRV genome ( NC_007016 ) was kindly provided by Scott Wong ( Vaccine and Gene Therapy Institute , Oregon Health & Science University , Beaverton , Oregon , USA ) . The miRNA expression vectors were cloned using PCR amplification of the relevant portions ( generally speaking , this contains the putative miRNA site as well as ~100–150 bp flanking said site ) of the viral genome ( from genomic plasmids ) , followed by restriction enzyme digestion and ligation . Specifically , the indicated regions of each viral genome listed in S6 Dataset were inserted into the XhoI/XbaI sites of pcDNA3 . 1 ( Invitrogen ) . The SV40 miRNA expression construct is as previously described [32] . Luciferase reporter plasmids were constructed using PCR amplification of the relevant portions of the viral genome ( from genomic plasmids ) , followed by restriction enzyme digestion and ligation . In this case , these sequences were cloned into the XhoI/XbaI sites of pcDNA3 . 1dsRluc which places the desired sequences into the 3’UTR of the Renilla Luciferase gene . Seed site mutant constructs were altered via PCR-based site directed mutagenesis at the indicated positions 2 , 3 , and 5 of the seed complement site ( FcPV1 mutant constructs ) , or positions 2 , 3 ( HPV41 mutant construct ) listed in S6 Dataset . To make the perfectly complementary RISC reporters for each miRNA , sequences with two perfectly complementary binding sites for each miRNA with a 12 nucleotide spacer region were synthesized ( Integrated Data Technologies ) and cloned into the Xho1/Xba1 sites of pcDNA3 . 1dsRLuc plasmid as listed in S6 Dataset . Respective mutants were made by mutating three nucleotides in the seed complement sequence of each binding site , also listed in S6 Dataset . Species tree was constructed using the nucleotide sequences of L1 gene region of papillomavirus genomes for all genomes with greater than 95% coverage in miDGE analysis ( 73 viral genomes in total ) . Sequences were aligned using CLUSTALW [69] in Geneious [70] . Initial tree ( s ) for the heuristic search were obtained automatically by applying a Neighbor-Joining algorithm [71] to a matrix of pairwise distances estimated using the Maximum Composite Likelihood ( MCL ) approach [72] , and then selecting the topology with superior log likelihood value . The tree with the highest log likelihood is shown . HEK293 or HEK293T cells ( as designated in figure legends ) were split and plated onto twelve well dishes so that they were approximately 70% confluent the following day . These plates were then co-transfected with five ng of the indicated Renilla Luciferase-based reporter constructs ( pcDNA3 . 1dsRluc [73] ) , five ng of Firefly Luciferase reporter ( pcDNA3 . 1dsLuc2CP [73] ) and one ug of either a control miRNA expression construct ( the SV40 miRNA expression construct ) or the indicated miRNA expression vector using Lipofectamine 2000 according to the manufacturer’s instructions . Transfections were carried out in triplicate for each transfection , which were considered technical replicates . Forty-eight hours later , cells were harvested with 100 uL of 1X Passive Lysis buffer from the Dual-Glo Luciferase Assay System ( Promega ) . 5 uL of lysate from each well was then analyzed in duplicate for Renilla and Firefly luciferase activity with a Luminoskan Ascent luminometer ( Thermo Electronic ) . These experiments were then performed for at least 3 biological replicates ( new transfections on separate days ) , with the exact number noted in the individual figure legend . Data was analyzed by dividing the Renilla luciferase activity value by the Firefly luciferase activity value to obtain a Renilla/Firefly luciferase activity ratio . These ratios were then averaged between the two measures for each well of the twelve well dish . Then the averaged Renilla/Firefly ratio was averaged for each of the technical triplicate wells , with the resulting average used to normalize the final values such that the no miRNA control Renilla/Firefly ratio was set to 1 . The one-sided Student’s t-test was used to assess the statistical significance of observed differences , with a P value of < 0 . 05 considered statistically significant . Primary data and analysis are in S7 Dataset . HEK293T cells were plated into 12-well dishes so that they were approximately 70% confluent the next day . At that point , they were transfected with 1ug of the indicated miRNA expression constructs per well of cells with Lipofectamine 2000 ( Invitrogen ) in accordance with the manufacturer’s instructions . 48 hours post-transfection , total RNA was harvested with PIG-B and Northern blot analysis was performed as described previously [66] . Briefly , following harvest , total RNA was quantitated by NanoDrop . Equal amounts of RNA were then run on a denaturing PAGE gel and transferred to Amersham Hybond–N+ membrane ( GE Healthcare ) . This membrane was then probed with indicated DNA oligonucleotide probe that was radioactively labeled with P32 and visualized through exposure to a phosphor screen and images were captured on the Typhoon ( probe sequences indicated S6 Dataset , uncropped scans in S8 Dataset ) . The blots were stripped of the DNA probe through incubation with boiling water and SDS , and reprobed with different DNA oligonucleotide probes . HEK293T cells were plated into 6-well dishes so that they were approximately 70% confluent the next day . At that point , they were transfected with 20 nM Drosha siRNA ( Sigma Aldrich ) or negative control siRNA ( Sigma Aldrich Mission siRNA SIC001 ) per well of cells using Lipofectamine RNAiMAX ( Invitrogen ) according to the manufacturer’s instructions . Forty-eight hours later these cells were trypsinized and , replated to new wells of a 6 well dish . The following day , these cells were approximately 70% confluent , and were transfected with the respective siRNAs and the indicated miRNA expression constructs or HSUR4 transfection/load control expression vector [21 , 74] ( 2ug/well as in the miRNA Detection assay ) with Lipofectamine 2000 ( Invitrogen ) according to the manufacturer’s instructions . Forty-eight hours later , total RNA was harvested with PIG-B and Northern blot analysis was performed as previously described [66] and in miRNA Detection Assay . The membrane for Northern blot analysis was probed with DNA oligonucleotides as indicated in S6 Dataset . Signal was quantitated using Image Studio Lite software , and ratios of mature miRNA signal in negative control cells to Drosha-knockdown cells ( relative to transfection/load control HSUR4 signal ) were calculated for each PV miRNA ( S7 and S8 Datasets ) . HEK293T and HEK293T Dicer KO cells ( NoDice , [75] ) were plated into 6 well dishes so that they were approximately 70% confluent the next day . At that point , they were transfected with 2ug/well of the indicated miRNA expression vectors using Lipofectamine 2000 ( Invitrogen ) in accordance with the manufacturer’s instructions . Forty-eight hours post-transfection , total RNA was harvested with PIG-B and Northern blot analysis was performed as in the miRNA Detection Assay section . The DNA oligonucleotide probes are listed in S6 Dataset . Signal was quantitated using Image Studio Lite software , and ratios of mature miRNA signal in negative control cells to Dicer KO cells ( relative to the same ratio in HEK293T cells ) were calculated for each PV miRNA ( S7 and S8 Datasets ) . Total RNA was extracted from both leg lesions and pectoral muscle ( control ) tissue from two chaffinches with proliferative leg skin lesions that were PCR-positive for FcPV1 DNA . The affected chaffinches were found dead by members of the public and submitted for post-mortem examination to a national scanning surveillance scheme for wild bird disease in Great Britain ( Lawson et al . in prep . ) . Samples of leg lesions and apparently normal pectoral muscle were collected at post-mortem examination and archived at -80 oC and -20 oC respectively . Samples were finely minced using sterile scalpel blades , and then ~20 mg of tissue was mixed with 350 μl of RTL:β-ME solution ( 1 ml buffer RTL [Qiagen] with 10 μl β-mercaptoethanol ) and homogenized . The lysate was centrifuged at maximum speed in a microcentrifuge for 3 min and transferred to a fresh microcentrifuge tube . Ethanol ( 100% ) was added to the cleared lysate to bring the final concentration up to 60% ethanol . Next , the samples were applied to an RNeasy ( Qiagen ) mini-spin column to purify the total RNA according to the manufacturer's instructions , except that after the final wash step , the samples were stored at approximately 4° C for several days while still on the column . The final elution steps were conducted with one volume of nuclease-free water and then repeated with one volume of nuclease-free Tris-EDTA ( TE ) , pH 7 . Pooled small RNA-seq libraries were prepared from RNA harvested from either foot lesions or pectoral muscle from animals CF180/09 and CF229/12 . Equal volumes of total RNA were combined and ethanol precipitated . Recovered RNA was dissolved in nuclease free water ( Ambion ) and quantitated with a NanoDrop spectrophotometer ( ThermoFisher ) . Libraries were prepared using 180 ng of pooled RNA with the NEBNext Multiplex Small RNA Library Prep Set for Illumina ( E7300 , New England Biolabs ) according to the manufacturer’s instructions with the addition of an additional round of indexing PCR to compensate for low input RNA . Libraries were quantitated with the Qubit dsDNA BR assay ( ThermoFisher ) , QC checked with the Bioanalyzer High Sensitivity assay ( Agilent Technologies ) , combined with other barcoded libraries , and sequenced on a single lane of a SR50 run on a HiSeq 4000 ( Illumina ) by the Genomic Sequencing and Analysis Facility at UT Austin . 106 , 251 , 321 and 122 , 129 , 107 reads were obtained for the foot and pectoral samples respectively ( SRA accession: SRP133175 ) . Small RNA reads were pre-processed by trimming the adaptor sequences and removing trimmed sequences shorter than 18 nucleotides with Cutadapt ( version 1 . 4 . 2 ) [76] . Reads were mapped with SHRiMP2 ( version 2 . 2 . 3 ) [77] to the reference sequences consisting of miRBase release 21 annotated zebra finch miRNAs [53] and FcPV1 reference sequence ( NC_004068 ) . The Cancer Genome Atlas cervical cancer RNA-seq datasets were retrieved from the NCI Genomic Data Commons . From the large RNA-seq data sets , sequenced coverage was calculated for the HPV reference sequence with the greatest number of alignments . Tumors with > = 50% coverage for an HPV were used for subsequent analysis ( 213 small RNA data sets ) . Aligned BAM files were converted to miRDeep2 format and the miRDeep2 pipeline was run with default parameters without miRNA annotations . BEDTools was used to assign the de novo miRDeep2 identified miRNAs to miRBase release 21 annotations . Small RNA data sets associated with NCBI GEO project GSE42380 were retrieved from the NCBI SRA . The SRA files were converted to colorspace FASTQ format using the SRA Toolkit . Adapter sequences were trimmed from reads using Cutadapt ( version 1 . 4 . 2 ) [76] . The trimmed libraries were mapped to miRBase release 21 annotated human miRNA sequences [53] and viral reference sequences using SHRiMP2 ( version 2 . 2 . 3 ) [77] ( reference sequences are listed in S7 Dataset ) . Resulting SAM files were converted to miRDeep2 format and the miRDeep2 pipeline was run with default parameters without miRNA annotations [47] .
|
Papillomaviruses ( PVs ) are causative agents of cancer . Currently , there is an incomplete understanding as to why only some infections lead to cancer . Developing a better comparative evolutionary understanding of PV gene products and their regulation is key to comprehending the life cycle of these pathogens . An emerging concept of viral gene regulation is that many persistent viruses will utilize small regulatory RNAs called miRNAs to optimize host and viral gene expression . Yet despite obvious interest , there have been no credible reports identifying canonical PV-encoded miRNAs . Here we develop new broadly applicable methodology to identify miRNAs from organisms lacking a laboratory culture system . We identify the first examples of bona fide canonical PV miRNAs and provide evidence supporting both host and viral miRNA-mediated regulation as relevant to control of PV gene expression . These findings resolve the issue of PV miRNAs and further the notion of miRNA importance to persistent virus infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"cdna",
"libraries",
"sequencing",
"techniques",
"gene",
"regulation",
"microbiology",
"genomic",
"library",
"construction",
"micrornas",
"viral",
"genome",
"genome",
"analysis",
"forms",
"of",
"dna",
"dna",
"libraries",
"dna",
"molecular",
"biology",
"techniques",
"dna",
"construction",
"rna",
"sequencing",
"microbial",
"genomics",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"genomic",
"libraries",
"viral",
"genomics",
"gene",
"expression",
"complementary",
"dna",
"molecular",
"biology",
"biochemistry",
"rna",
"nucleic",
"acids",
"dna",
"library",
"construction",
"virology",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"non-coding",
"rna",
"computational",
"biology"
] |
2018
|
Identification of virus-encoded microRNAs in divergent Papillomaviruses
|
Contacts of leprosy patients are at increased risk of developing leprosy and need to be targeted for early diagnosis . Seropositivity to the phenolic glycolipid I ( PGL-I ) antigen of Mycobacterium leprae has been used to identify contacts who have an increased risk of developing leprosy . In the present study , we studied the effect of seropositivity in patient contacts , on the risk of developing leprosy , stratified by Bacille Calmette Guerin ( BCG ) vaccination after index case diagnosis . Leprosy contacts were examined as part of the surveillance programme of the Oswaldo Cruz Institute Leprosy Outpatient Clinic in Rio de Janeiro . Demographic , social , epidemiological and clinical data were collected . The presence of IgM antibodies to PGL-I in sera and BCG vaccination status at the time of index case diagnosis were evaluated in 2 , 135 contacts . During follow-up , 60 ( 2 . 8%; 60/2 , 135 ) leprosy cases were diagnosed: 41 among the 1 , 793 PGL-I-negative contacts and 19 among the 342 PGL-I-positive contacts . Among PGL-I-positive contacts , BCG vaccination after index case diagnosis increased the adjusted rate of developing clinical manifestations of leprosy ( Adjusted Rate Ratio ( aRR ) = 4 . 1; 95% CI: 1 . 8–8 . 2 ) compared with the PGL-I-positive unvaccinated contacts ( aRR = 3 . 2; 95% CI: 1 . 2–8 . 1 ) . The incidence density was highest during the first year of follow-up for the PGL-I-positive vaccinated contacts . However , all of those contacts developed PB leprosy , whereas most MB cases ( 4/6 ) occurred in PGL-I-positive unvaccinated contacts . Contact examination combined with PGL-I testing and BCG vaccination remain important strategies for leprosy control . The finding that rates of leprosy cases were highest among seropositive contacts justifies targeting this specific group for close monitoring . Furthermore , it is recommended that PGL-I-positive contacts and contacts with a high familial bacteriological index , regardless of serological response , should be monitored . This group could be considered as a target for chemoprophylaxis .
It was widely expected that the treatment of all newly diagnosed leprosy cases with multidrug therapy ( MDT ) would not only cure the disease but also prevent the further spread of Mycobacterium leprae ( M . leprae ) . In fact , from 2004 to 2010 , the number of newly diagnosed cases worldwide fell by 44% . Nonetheless , incidence rates above 1 case in 10 , 000 remain in a few countries , namely Brazil , Nepal , Liberia , and a few islands in the Western Pacific . Brazil reported the most cases in the Americas and the second most worldwide in 2010 [1] . Of the 34 , 894 new leprosy patients diagnosed in Brazil in 2010 , 5 . 36% were children under 15 , an indication that the transmission of M . leprae is ongoing [2] . Notably , 7 . 2% of the newly detected leprosy cases were of disability grade 2 [3] , demonstrating the heretofore limited effectiveness of case detection and the magnitude of the hidden prevalence of leprosy [4] . Maintenance of poverty and of the intensity of exposure may have contributed to the low effectiveness of leprosy control programs . Even if the effectiveness of case detection and MDT treatment could be improved , additional interventions would be needed , with a special focus on groups at particularly high risk of developing clinical leprosy [5] . It has long been known that contacts of leprosy patients have an increased risk of developing leprosy compared with the general population [6] . The detection of antibodies to the phenolic glycolipid I ( PGL-I ) antigen of M . leprae has been used to understand the epidemiology of subclinical infection , as opposed to active disease . However , this technique has not been proven for the early diagnosis of clinical cases and for predicting who ( either among contacts of known cases or among the general population ) will develop clinical leprosy in the future [7] . The relationship between PGL-I seroprevalence and the leprosy burden depends on the population studied [8] , [9] . Seropositivity has been reported to be higher in contacts of leprosy patients than among the general population and has been associated with the development of leprosy [10] , [11] . Although PGL-I ( − ) based serological tests cannot be used as screening tools in the general population , they have been used to identify contacts of leprosy patients who have a higher risk of developing leprosy [12] . After 7 years of follow-up of a cohort of 559 household contacts of multibacillary ( MB ) patients with a bacteriological index ( BI ) greater than or equal to 2 , Douglas et al . [13] reported that seropositive ( PGL-I ( + ) ) contacts in the Philippines had a 7-fold higher risk of developing leprosy compared with seronegative ( PGL-I ( − ) ) contacts . The Yalisombo Study Group [14] reported a slightly higher proportion of cases among PGL-I ( + ) ( 1/189; 0 . 53% ) compared with PGL-I ( − ) contacts ( 10/3018; 0 . 33% ) in a survey of 4 hyperendemic villages in Zaire . Other studies have not reported an increased risk of developing leprosy among seropositive contacts [15] , [16] . The Brazilian Leprosy Control Program has recommended that all household contacts of leprosy patients be examined and receive Bacille Calmette Guerin ( BCG ) immunization as an additional preventive measure against leprosy [17] . According to Bagshawe et al . [18] , the accelerated manifestations of benign tuberculoid leprosy after BCG vaccination reflect BCG vaccination acceleration of the natural history of M . leprae infection in individuals who were infected prior to or immediately after vaccination . In line with this result , Duppre et al . [19] found that vaccinated contacts contracted leprosy mainly from MB index cases ( ICs ) , suggesting the presence of subclinical infection which becomes overt due to vaccination induced immune response activation . Because previous studies have failed to reach a consensus regarding the effect of seropositivity on the risk of developing leprosy among contacts and the degree of protection conferred by prior BCG vaccination ( BCG scar ) , further investigation seemed necessary . Thus , the effects of simultaneous BCG vaccination and other possible covariates on the diagnosis of overt leprosy were studied in a group of contacts participating in a surveillance program . The effect of PGL-I seropositivity in contacts , adjusted by covariates measured at the first examination , on the risk of developing leprosy was assessed per se and according to BCG vaccination status after IC diagnosis .
This dynamic cohort study was based on the contact surveillance programme of leprosy patients who were diagnosed at the Leprosy Outpatient Clinic of the Oswaldo Cruz Institute , FIOCRUZ , in Rio de Janeiro , RJ , Brazil . Among the 6 , 060 contacts examined between June 1987 and December 2007 , 2 , 135 ( 35 . 2% ) were tested for IgM antibodies to PGL-I . During this period , 2 . 2% ( 46/2 , 135 ) of the contacts were diagnosed at the initial examination ( co-prevalent cases ) , did not receive the BCG vaccine and were excluded from the present study . The subsample of contacts selected for the present study was similar to the contacts not selected in terms of gender ( p = 0 . 61 ) , operational classification of IC ( p = 0 . 87 ) and presence of BCG vaccination scar ( p = 0 . 98 ) . However , the selected group of contacts was significantly older than those contacts who were not selected ( p<0 . 001 ) . The possible selection bias toward older contacts could be due to the difficulties of blood sampling in children . Most parents refused to allow the collection of blood from their children . Blood sampling in young children became part of the protocol after the introduction of the ML Flow test , which uses only one drop of blood for testing . The presence of anti-PGL-I antibodies at the first examination was the primary variable of interest . The modifying effect of BCG vaccination after IC diagnosis was highlighted because of its known association with the study outcome . Household contacts were defined as individuals who lived in the same dwelling ( i . e . , sharing the same kitchen or social/recreational area ) . Non-household contacts were defined as those indicated by the IC as having had other types of associations , such as next-door neighbors , blood relatives , friends and colleagues . The duration of association with the IC was not considered during the selection of contacts . After confirmation of the leprosy diagnosis , patients were given educational information about the disease , and medical visits were scheduled for their close contacts ( within and outside of the household ) . During the initial visit , contacts were interviewed by a social worker to obtain demographic and social information ( e . g . , schooling and individual and family income ) and the degree of closeness to the IC . All contacts received health education on leprosy and were instructed to report to the Leprosy Outpatient Clinic if any clinical signs of leprosy occurred . In addition , contacts were instructed to visit the Center once a year for a period of 3–5 years . In general , contacts were followed for at least 2 years , and the follow-up period ended in December 2009 . An experienced clinical dermatologist examined all of the contacts to identify any leprosy lesions and the typical BCG vaccine scar . In addition , a neurological exam of peripheral nerves was performed by a qualified physiotherapist or neurologist . If a contact presented signs and symptoms suggestive of leprosy , he or she was assessed by bacteriological , histopathological , and immunological tests . If leprosy was diagnosed , the individual was classified according to the Ridley and Jopling scale [20] and grouped for treatment according to the bacteriological index ( BI ) results as either multibacillary ( MB - positive BI ) or paucibacillary ( PB - negative BI ) . Since 1991 , the BCG vaccine has been administered to all healthy contacts , as recommended by the Brazilian Leprosy Control Program [21]; however , 248 ( 12 . 8% ) of the contacts in the sample group were not vaccinated at their first visit due to pregnancy , acute disease or vaccine shortage . These contacts were rescheduled for vaccination , but 179 ( 111 of whom had a BCG vaccine scar and 68 of whom had no visible scar ) failed to return for vaccination . These noncompliant cases , together with 200 contacts examined before 1991 ( 104 with a BCG scar and 96 without ) , were included in the study as part of the unvaccinated group . Thus , among the 2 , 135 contacts included in this study , 1 , 756 received simultaneous BCG vaccine at the time of IC diagnosis , and 379 did not . Before vaccination , blood samples were collected , and the sera were separated into aliquots , followed by storage at −20°C to later determine the presence/absence of anti-PGL-I antibodies ( all contacts were eligible ) . Two different rapid tests were used for evaluation of the presence of antibodies in blood serum . The ML Dipstick assay [22] was used between 1987 and 2002 to test 1 , 050 contacts . Beginning in 2003 , the ML Flow test was implemented as part of the routine contact examination , and 1 , 085 contacts were tested in this manner . The visual readings of both tests were performed as previously described [22] . A reddish-stained antigen band indicated a positive reaction . Both tests presented a high level of agreement in the detection of IgM antibodies to PGL-I using the enzyme-linked immunosorbent assay ( ELISA ) ( 97 . 2% [k = 0 . 92] and 91% [k = 0 . 77] for the ML Dipstick and ML Flow tests , respectively ) [12] . For the purposes of this study , those contacts who did not return for evaluation were considered free of leprosy . However , due to the low participation of contacts in re-examination during the study period ( 29% ) , a complementary strategy was adopted . To ascertain the existence of leprosy contacts who might have moved away or visited another health center , the Brazilian Information System for Notifiable Diseases ( SINAN ) database was searched for new cases . Reporting cases to the SINAN is compulsory for all municipalities in Brazil and is performed on a weekly basis . The data feed is monitored at the state and national levels according to specific parameters . SINAN records published in 2010 ( i . e . , with data on new cases up to 2009 ) were matched to the database of the present study by the contact's full name , date of birth , and mother's full name . As a result of this search , 3 contacts in the SINAN database were included in the sample group as new leprosy cases . After receiving educational information about leprosy , all adult participants and the guardians or parents of the child participants provided written consent . A medical history for each contact was taken from routine care medical records . Data collection , management , and analysis were performed by the study coordinators , and confidentiality was maintained throughout the research . The present study , including the use of patient records , was approved by the Research Ethics Committee of the National School of Public Health ( Document N° . 113/06 ) . The leprosy incidence rate at the contact follow-up was based on person-years ( PYs ) between the first examination of a contact and the date of his or her leprosy diagnosis . Contacts who did not return for follow-up or who were not found in the SINAN database were considered to be free of leprosy at the end of the study . The total familial BI was derived from the sum of all BIs of MB cases in the family at the time of the first examination , which was believed to be a better proxy of disease risk for the contacts who were followed up after the initial examination . Contacts who did not receive the BCG vaccine after the IC's diagnosis were considered unvaccinated , and those who were vaccinated subsequent to the IC's diagnosis were considered vaccinated . Accordingly , based on their vaccination status and serological response to PGL-I , the contacts were grouped into the following categories: PVC , Positive Vaccinated Contacts; NVC , Negative Vaccinated Contacts; PUC , Positive Unvaccinated Contacts; and NUC , Negative Unvaccinated Contacts . The association between covariates and seropositivity was analyzed using univariate and multivariate logistic regression to generate odds ratios ( ORs ) . Crude and adjusted rate ratios ( RRs ) were estimated by Poisson regression to verify the association between seropositivity and the development of leprosy , both overall and stratified according to vaccination status . RR estimates were adjusted for age , gender , presence of BCG scar , type of association with the IC , duration of close association with the IC , and sum of the family BIs . The 95% confidence intervals ( CIs ) were determined for all estimates . Multivariate analyses and CIs were based on robust variance estimators using clusters of contacts . To account for the clustering effect in both types of regressions , the CI was based on robust variance estimators that account for a smaller variance of contacts clustered around the IC . Statistical interaction ( RR test of homogeneity ) was assessed when judged scientifically meaningful according to the Mantel-Haenszel test . Statistical analysis was performed with Stata™ version 8 . 0 ( Stata Corp . , College Station , TX , USA ) and Open Source Epidemiologic Statistics for Public Health version 2 . 3 . 1 ( http://www . openepi . com/OE2 . 3/Menu/OpenEpiMenu . htm ) .
The present study included 2 , 135 contacts of 668 ICs ( 220 PB and 448 MB , with an average of 3 . 5 contacts per MB patient and 2 . 6 per PB patient ) who were tested for the presence of IgM antibodies to PGL-I from 1987 to 2009 . Most of the contacts ( 1 , 253; 59% ) were female . The mean age was 28 . 8 ( SD: 17 . 0 ) years . Most of the contacts ( 64% ) had a low monthly family income ( below four minimum salaries defined by law and adjusted periodically according to inflation . There were no demographic differences between the vaccinated and unvaccinated contacts . In both groups , females predominated ( 58 . 3% of vaccinated contacts and 60 . 4% of unvaccinated contacts ) . The mean age was significantly greater ( t-test = 2 . 04; p = 0 . 042 ) in vaccinated contacts ( 29 . 2±17 . 2 years ) compared with unvaccinated contacts ( 27 . 2±16 . 2 years ) . Overall , the rate of seropositivity to PGL-I at the first evaluation was 16 . 0% among contacts . Adjusting for relevant covariates , seropositivity was more frequent among household contacts , females , contacts aged 15–35 years , and contacts with a high family BI . The presence of a BCG scar from prior vaccination , the duration of close association with the IC and the operational classification of the IC did not appear to influence PGL-I positivity ( Table 1 ) . The contacts were followed for an average of 5 . 1±3 . 98 years ( range: 0 . 21–18 years ) . During the follow-up period , 60 ( 28 . 1/1 , 000 PYs ) new cases of leprosy were diagnosed at an incidence density of 5 . 08/1 , 000 PYs . Most of the cases ( 90%; 54/60 ) were contacts of MB patients . The average latency of detection of the new cases after the initial examination was 2 . 8 years ( range: 3 months-10 . 5 years ) . Only 11 cases were detected during the first year after initial examination . The rate of detection declined steeply between the first and fourth years after the initial examination of contacts ( Figure 1 , solid line ) . The incidence density of leprosy varied according to the BCG vaccination status and serology result ( Figure 1 , broken lines ) . PVCs and NUCs had the highest incidences of leprosy during the first year of follow-up at 17 . 9/1 , 000 PYs and 9 . 9/1 , 000 PYs , respectively . The effectiveness of the BCG vaccination was identified at the 2-year follow-up , rapidly reducing the incidence density to 2 . 5/1 , 000 PYs . However , the incidence density in NUCs did not begin to decrease until the third year . In addition , the incidence density was low during the initial years of follow-up in NVCs and progressively decreased , reaching 0 at 5 years of follow-up . Conversely , no cases of leprosy were diagnosed in the PUC group during the first 2 years of follow-up . However , the incidence density progressively increased in this group of contacts , with the highest values identified in the sixth year of follow-up . All of the groups converged to zero incidence during the 11th year of follow-up . Leprosy diagnosis was strongly associated with PGL-I seropositivity . A significantly higher ( χ2 = 11 . 2; p<0 . 01 ) proportion of incident cases was detected among PGL-I ( + ) contacts ( 5 . 6% , 19/342 ) during the follow-up period compared with PGL-I ( − ) contacts ( 2 . 3%; 41/1 , 793 ) . PGL-I ( + ) contacts presented a 3 . 2-fold ( 95% CI: 1 . 6–6 . 1 ) higher risk for leprosy compared with PGL-I ( − ) contacts . Stratification by vaccination status showed that the rate of developing leprosy was 1 . 8 times higher among unvaccinated than vaccinated contacts ( 8 . 3/4 . 6; p = 0 . 03 ) . Among PGL-I ( + ) contacts , BCG vaccination after IC diagnosis increased the adjusted rate of developing clinical manifestations of leprosy ( aRR = 4 . 1; 95% CI: 1 . 8–8 . 2 ) compared with the PGL-I ( + ) unvaccinated contacts ( aRR = 3 . 2; 95% CI: 1 . 2–8 . 1 ) . Contacts aged 15–35 years showed a significantly ( p<0 . 01 ) higher proportion of seropositivity ( 19 . 0% ) compared with children ( 14 . 8% ) and contacts aged >35 years ( 10 . 6% ) ( Table 1 ) . Interestingly , after BCG vaccination , the 15- to 35-year-old age group presented a significantly ( p = 0 . 02 ) lower rate of leprosy ( 2 . 5/1 , 000 PYs ) compared with vaccinated PGL-I ( + ) children ( 6 . 6/1 , 000 PYs ) and contacts over 35 years of age ( 6 . 8/1 , 000 PYs ) ( Table 2 ) . In unvaccinated contacts , long periods of association with the IC and a high family BI were associated with the development of leprosy ( Table 2 ) . A significantly higher ( p<0 . 01 ) proportion of PB leprosy cases was diagnosed in PVCs ( 4 . 8%; 13/269 ) compared with NVCs ( 1 . 9%; 28/1 , 487 ) . All MB cases occurred among unvaccinated contacts and were diagnosed at a significantly higher rate ( χ2 = 8 . 79; p = 0 . 03 ) in PUCs ( 5 . 5%; 4/73 ) than in NUCs ( 0 . 7%; 2/306 ) ( Figure 2 ) .
The predictive value of PGL-I seropositivity in the development of leprosy in contacts was analyzed as a method of identifying susceptible individuals among contacts of recently diagnosed patients . In addition , the possible interference of PGL-I seropositivity with the protective effect of BCG vaccination against leprosy was investigated . The observed proportion of seropositivity ( 16 . 0% ) was similar to that found in another study performed in Brazil [23] . The prevalence of seropositivity in this study showed associations with age and gender similar to those reported in other studies [6][24] . According to Maddison et al . [25] , females tend to demonstrate higher innate IgM levels than males , which may explain the high female seropositivity rate found in the present study . Independent of gender , seropositivity rates increased until young adulthood ( 15–35 years of age ) and decreased in older adults , which is consistent with the general decrease in overall IgM levels observed with age [25] . It is well known that leprosy does not manifest preferentially in women or children , so these high levels are more likely explained by this common feature of the immune system rather than specifically reflecting differences in anti-PGL-I antibody levels in these groups . The presence of a past BCG vaccination scar was associated with a higher seropositivity , but the association was weak and marginally significant when adjusted for covariates . This result corroborates findings of Baumgart et al . [26] , who argued that BCG vaccination or exposure to tuberculosis or environmental mycobacteria could interfere with serological tests such as the PGL-I assay . In the sample group of this study , PGL-I ( + ) contacts had a clear increased risk of developing leprosy . The independent effect of bacterial load , as measured by the familial BI , on the risk of developing leprosy among contacts is consistent with previous findings [19] . The increased incidence of leprosy observed in PVCs and NUCs during the first year of follow-up suggests subclinical infection . PVCs were partially benefited by BCG vaccination , as observed by Bagshawe et al . [18] in children , because they had insufficient time to build their immune capability to fight M . leprae but managed to avoid MB leprosy infection . BCG vaccination induces an increase in interferon-gamma ( IFN-γ ) production , which is highest among previously vaccinated individuals and those exposed to environmental mycobacteria [27] . Thus , the contacts' immune systems are predisposed to a cellular response that is effective against M . leprae [28] . IFN-γ production in response to M . leprae antigens is a measure of the ability to mount an effective immune response against the pathogen [29] . Thus , the lack of immune response among contacts exposed to the infectious agent could indicate susceptibility , as posited by Sampaio et al . [30] . The applicability of PGL-I testing for early diagnosis of clinical cases thus remains uncertain . The overall decline in the incidence density of leprosy in contacts observed after the first year of diagnosis of the ICs , as observed by other authors [31] , [32] , could result from the treatment of index and co-prevalent cases . MDT seems to decrease in infectiousness over time . However , Groenen et al . [14] observed a mean yearly incidence rate of PB leprosy of 0 . 34% , with little variation during 4 years of follow-up . This difference may be explained by the hyperendemicity of the population studied by the latter authors and the irregular use of treatment by the patients . The variation in the incidence density of leprosy according to BCG vaccination status and serological profile indicates that multiple factors are involved in the development of clinical overt leprosy in contacts . Together with early diagnosis and the treatment of ICs and co-prevalent cases , preventive measures such as contact evaluation , health education and immunization can prevent the transmission of leprosy . Interestingly , the early peak in incidence and reduced infection levels in young adults , which reflect constitutional , age-related changes in the immune system [33] , were only observed in the vaccinated group . Although the highest leprosy rates were expected among contacts aged 15–35 years , i . e . , in the age group with the highest seropositivity , this group of contacts had the lowest incidence rate of leprosy after BCG vaccination . Regardless of their anti-PGL-I serological status , children are more susceptible than adults to acquiring leprosy infection and developing overt leprosy due to their incompletely developed immune systems and close and prolonged contact with possible intra-family sources of infection [34] , [35] . Additionally , because BCG vaccination induces IFN-γ production [27] , the strong immune response in young adults will control subclinical infection if present . The known long incubation period of the disease was confirmed in the present study , as most of the MB cases occurred in PUCs ( 4/6 ) after the second year of follow-up . However , in the Yalisombo Study Group [14] , the only MB case among the 13 incident cases in a 4-year cohort of 3207 contacts was diagnosed during the first year of follow-up . However , because the present study cohort was alerted to early signs of the disease , the contacts' awareness and subsequent detection of leprosy signs may have contributed to the high proportion of PB cases . A major limitation of the present study was the use of a non-probabilistic sample group obtained at a reference leprosy center under routine conditions . In addition , the sample group may have had a selection bias toward older individuals , as children did not usually provide blood samples . However , the group in this study included contacts with a wide range of social and demographic characteristics who lived in a medium endemic region , which is similar to many settings in Brazil . Although it was not possible to ascertain the number of deaths during follow-up , the mortality rate due to leprosy is almost negligible within this age group [36] . In the ML flow test used to evaluate the presence/absence of antibodies against PGL-I , a precise distinction between positivity and negativity is sometimes difficult to ascertain . A misinterpretation of results due to grading from 0–4 could , in part , explain the finding of PB cases among the seronegative contacts . It is well known that contacts of leprosy patients are at higher risk of developing leprosy and may even constitute a source of infection in the community at large [37] . In regions where no interventions are undertaken , contacts producing antibodies against M . leprae ( corresponding to the PUCs in the present study ) can be considered to be the main indicators of the maintenance of leprosy's endemic status . Nevertheless , early and effective interventions for contacts will affect the disease burden , leading to exhaustion of cases after 10 years of IC diagnosis . The present study confirms that contact surveillance and health education combined with BCG vaccination remain important strategies for leprosy control . The fact that the highest rate of leprosy cases was found among PGL-I ( + ) unvaccinated contacts justifies targeting this specific group for close monitoring . Furthermore , it is highly recommended that PGL-I ( + ) contacts and contacts with high familial BIs be monitored regardless of serological response . Targeting these groups for a more focused and specific approach such as chemoprophylaxis could make this intervention strategy more cost-effective .
|
Although leprosy has become a neglected disease , it is an important cause of disability , and 250 , 000 new cases are still diagnosed worldwide every year . The current study was carried out in Brazil , where almost 40 , 000 new cases of leprosy are diagnosed every year . The study targeted contacts of leprosy patients , who are at the highest risk of contracting the disease . We studied 2 , 135 contacts who were diagnosed at the Leprosy Outpatient Clinic at the Oswaldo Cruz Foundation in Rio de Janeiro , RJ , Brazil , between 1987 and 2007 . The presence of antibodies against a specific Mycobacterium leprae antigen ( PGL-I ) at the first examination and BCG vaccination status were evaluated . PGL-I-positive contacts had a higher risk of developing leprosy than PGL-I-negative contacts . Among the former , vaccinated contacts were at higher risk than unvaccinated contacts . Our results indicate that contact examination combined with PGL-I testing and BCG vaccination appears to justify the targeting of PGL-I-positive individuals for enhanced surveillance . Furthermore , it is highly recommended that PGL-I-positive contacts and contacts with a high familial bacterial index ( i . e . , the sum of results from index and co-prevalent cases ) , regardless of serological response , should be monitored . This group could be considered as a target for chemoprophylaxis .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine"
] |
2012
|
Impact of PGL-I Seropositivity on the Protective Effect of BCG Vaccination among Leprosy Contacts: A Cohort Study
|
The bacterial flagellar motor is a highly efficient rotary machine used by many bacteria to propel themselves . It has recently been shown that at low speeds its rotation proceeds in steps . Here we propose a simple physical model , based on the storage of energy in protein springs , that accounts for this stepping behavior as a random walk in a tilted corrugated potential that combines torque and contact forces . We argue that the absolute angular position of the rotor is crucial for understanding step properties and show this hypothesis to be consistent with the available data , in particular the observation that backward steps are smaller on average than forward steps . We also predict a sublinear speed versus torque relationship for fixed load at low torque , and a peak in rotor diffusion as a function of torque . Our model provides a comprehensive framework for understanding and analyzing stepping behavior in the bacterial flagellar motor and proposes novel , testable predictions . More broadly , the storage of energy in protein springs by the flagellar motor may provide useful general insights into the design of highly efficient molecular machines .
Our model for stepping relies on two main assumptions: constant or nearly constant torque between stator and rotor and an approximately 26-fold periodic contact potential . Since the motor operates at the molecular scale , its rotation is intrinsically stochastic as it is subject to random thermal fluctuations . Another potential source of noise is fluctuations of the torque applied by the individual MotA/B stators , due to the discrete nature of the proton flux . However , in presence of multiple independent stator units , this noise averages out and can be neglected . Under the combined influence of the applied torque , the contact potential , and thermal fluctuations the rotor performs a circular and continuous random walk in a tilted , approximately periodic potential , which we model by the following Langevin equation: ( 1 ) where is the drag coefficient , the total torque exerted via protein springs by the stators , and where the potential includes the torque , and the approximately 26-fold periodic contact potential: ( 2 ) The term represents Gaussian white noise and accounts for thermal fluctuations: , where is the rotor diffusion coefficient , related to the temperature and the drag coefficient via Einstein's relation: . In experiments , a load ( usually a polystyrene bead ) is attached to a flagellar stump and this load is largely responsible for the drag . For simplicity , we assume that linkage between motor and load is instantaneous , as the relaxation is rapid compared to the typical stepping time ( see Discussion ) .
Numerical simulation of the model ( Eq . 1 ) shows that rotation proceeds in steps ( Fig . 1C ) . These steps correspond to jumps between adjacent wells of the tilted potential . Jumps/steps are possible thanks to thermal fluctuations which drive the system out of energy minima; without these fluctuations , the system would remain stuck in one well forever . Steps therefore correspond to crossings of the energy barriers separating wells . According to the Arrhenius law , the average time to cross a barrier increases exponentially with the barrier height . Because of the tilt induced by the torque , steps in the forward direction correspond to lower energy barriers than steps in the backward direction ( cf . Fig . 1C ) . Forward steps are therefore more likely to occur than backward steps , so that on average the motor moves forward . To further investigate stepping in our model , we wrote a step detector algorithm similar to that described in [1] , and applied it to a simulation of rotations ( see Materials and Methods ) . Fig . 2A shows the histogram of step sizes . As in the experiment , we find that backward steps are smaller on average than forward steps: the mean forward step is ( ) , against for backward steps . ( The precise values vary with the particular choice of potential and torque , but the mean step size is always larger for forward than backward steps . ) Recognizing steps as barrier crossing events allows us to readily explain the difference between average forward and backward steps sizes using a simple intuitive argument , as illustrated in the inset to Fig . 2A . Backward steps occur infrequently because the energy barriers for these steps are higher than for forward steps . By contrast , all forward steps must occur as the motor moves forward , regardless of the heights of energy barriers . In addition , as shown in the inset to Fig . 2A , barrier heights and step sizes tend to be positively correlated . Roughly speaking , we can assume that higher barriers extend over longer ranges . We confirmed the validity of this assumption by randomly generating many approximately 26-fold periodic potentials with a very general functional form ( results shown in Fig . 3 ) . Therefore , backward steps occur mostly over the lower barriers , and lower barriers correspond to smaller step sizes . This implies that backward steps are on average smaller than forward steps . Note that this argument relies on the fact that the barriers are not all identical . To test the scenario proposed above , we asked whether the size of forward steps immediately preceding or following backward steps differ from the average of . According to our picture , the barrier crossed by these forward steps should be the same as the one crossed by the backward step immediately preceding or following , implying a small barrier and therefore a small forward step size . Fig . 2B–E shows that indeed forward steps preceding of following a backward step are smaller on average , in the experiment , in our simulation , than the mean forward step of . ( Note that even for forward steps over the same barrier , backward steps are still slightly smaller on average in both experiment and simulation . This suggests that the step detection algorithm has a small systematic bias—see Materials and Methods . ) The difference between forward and backward step sizes relies on the 26 barriers around the circle not all being identical . These heterogeneities may exist because the potential is only approximately 26-fold periodic and contains other periodicities as well arising from the filament , hook , or other parts of the rotor . In any event , an essential prediction of our model is that step sizes and backward step frequencies will depend strongly on absolute position ( i . e . modulo ) , reflecting the fixed contact potential . We now examine how step frequencies and sizes depend on the properties of particular barriers , specified by the position of the rotor around the circle , and how this can tell us something about the detailed nature of contact forces . According to our model , backward steps should be much more likely to occur at low barriers . In contrast , where forward steps occur should be much less sensitive to barrier heights . This follows simply because for each complete rotation the number of forward steps over any barrier is one plus the number of backward steps over that same barrier . Therefore , as long as backward steps are rare , the average number of forward steps over each barrier will be close to one per rotation and therefore the frequencies of forward steps will be similar for all barriers . This is illustrated in Fig . 4A , which presents average step frequencies for each of the 26 barriers of a particular potential ( chosen to be the same as in Fig . 2 ) , as calculated from first-passage theory ( see Materials and Methods ) . As expected , there is considerable variation among barriers in backward-step frequencies , but much less variation in forward-step frequencies . To relate average step sizes to absolute position , we examined the sizes of backward and forward steps for each of the barriers . To properly assign steps to barriers , we sorted steps into equal bins according to the angular position of the rotor when the step occurred , and calculated the average backward and forward step sizes in each bin . We applied this procedure to simulations of rotations generated with three different potentials ( see Materials and Methods ) , and to four experimental traces , corresponding to four distinct cells , totaling 700 rotations [1] . According to our model , forward and backward steps across the same barrier should have the same average size . In the simulations ( Fig . 4B ) , mean forward and backward steps across the same barrier were found to be linearly correlated , though with a systematic offset toward smaller backward steps . ( As discussed above , this offset is likely the result of a bias in the step detection algorithm . ) We found that mean forward and backward step sizes across the same barrier are also positively correlated in the experiment ( Fig . 4D ) , in agreement with our prediction ( with the same bias towards smaller backward steps ) . Overall , these results suggest that the absolute position of the rotor accounts for much of the variability observed in step size , and supports our model of a fixed , nearly periodic contact potential . We next show that , both in the simulation and in the experiment , the barriers with a high frequency of backward steps are the same barriers where step sizes are short . To this end we plot the frequency ratio of backward steps to forward steps across each barrier versus the average forward step size ( which we showed in Fig . 4B , D correlates with the backward step size ) , for both simulations ( Fig . 4C ) and experiment ( Fig . 4E ) . In both cases backward-step frequencies fall off sharply with average forward step size . Since barriers where steps are smaller have higher backward-step frequencies than other barriers , they contribute more to the average backward step size . Therefore , the mean backward step is smaller than , which would be the mean backward step size if all barriers contributed equally . In contrast , forward-step frequencies vary little from barrier to barrier , so that all barriers contribute more or less equally to the average forward step size , which is therefore approximately . This explains why backward steps are smaller than forward steps on average . The recognition that steps are barrier-crossing events has a direct implication for how rotation speed depends on torque . In the absence of contact forces ( i . e . ) , the average rotation speed depends linearly on torque: . However , when the contact forces are comparable to the torque , rotation is hindered by barriers , and the system spends much of the time in local energy minima . Rotation is then not only limited by drag , but also by the rate of barrier crossing , leading to lower rotation speeds: . We computed analytically ( see Materials and Methods ) the torque-speed relation for loads with various drag coefficients for a perfectly 26-fold periodic sinusoidal potential with amplitude , as shown in Fig . 5A ( using approximately 26-fold periodic potentials yielded qualitatively identical results ) . At high torques , the linear relation is recovered asymptotically , which follows because increasing torque decreases the barriers to forward rotation , and eventually eliminates them completely , as shown in the insets to Fig . 5A . Fig . 5B shows the effect of the contact potential on the effective long-time diffusion coefficient . At low torques , diffusion is slowed down by barriers , while at high torques one recovers the natural diffusion coefficient . Interestingly , at intermediate torques rotor diffusion is actually enhanced by the contact potential . In this regime , the contact potential is a small but variable correction to the torque . This variability contributes to the variance of the rotation speed , thus effectively enhancing rotor diffusion ( see Materials and Methods ) . At large torques , approaches asymptotically: ( 3 ) To check the consistency of the predicted diffusive behavior against previous results , we compared our model's prediction with experimental measurements of the variance in the rotation time [9] , [11] . In [9] , the rotation time of a tethered cell was measured . Simple diffusion predicts that the variance in rotation time per cycle is: ( 4 ) For a single fully-powered torque-generating unit , torque is estimated to be [7] . The measured speed in [9] was noisy and depended on the particular cell but was about for three torque-generating units , leading to an estimate for the drag coefficient , and for the diffusion constant . Thus the variance in cycle time for three torque-generating units predicted by simple diffusion is , which is consistent with the reported value of [9] . We conclude that it may not be necessary to consider other sources of fluctuations ( e . g . proton translocations ) to explain the observed variance in cycle time . Note that in these experiments the torque was high , and therefore contact forces are not expected to have had a significant effect on diffusion . Our model also predicts a negative feedback reaction from the MotA/B protein springs that in principle could reduce diffusion . Namely , every time the rotor moves forward , the springs relax , causing a transient decrease of torque ( and the opposite every time the motor moves backward ) . To estimate the magnitude of this effect , we model torque dynamics by linking spring elongation to rotor position , and by assuming that the springs “restretch” to their equilibrium position prescribed by the PMF with a characteristic relaxation time ( see Materials and Methods ) . Within this model , the effective diffusion coefficient is found to be , where is the slope of the torque-speed relation of the motor near stall . The value of ranges from depending on the temperature [6] and is therefore much smaller than relevant values of the drag coefficient . We conclude that in the conditions of the discussed experiments , the effect of negative feedback from the springs on diffusion is negligible . According to our model , the distribution of waiting times between steps is expected to be roughly exponential . The distribution of waiting times between steps ( forward or backward ) for an exactly 26-fold periodic potential is indeed exponential ( Fig . 6A ) . When the potential is heterogeneous , the average waiting time depends on the barrier . Even though the distribution of waiting times across each barrier is exponential , the overall waiting-time distribution is not , appearing rather as a “stretched” exponential ( Fig . 6B ) . The experimental distribution also resembles a stretched exponential ( Fig . 6C ) .
Our model explains stepping of the bacterial flagellar motor by interpreting its rotation as a viscously damped random walk driven by a constant torque and by a heterogeneous contact potential caused by the physical irregularities of the rotor . In this picture , steps are recognized as barrier-crossing events between adjacent minima of a tilted and corrugated energy potential . Corrugations are caused by contact between the stators and the protein arrays ( FliG , among others ) making up the rotor structure . Recently a more accurate picture of this structure has emerged , thanks notably to electron microscopy studies [12] . Our model predicts a periodicity of the potential , so that the absolute angular position of the rotor with respect to the stator is an underlying determinant for step statistics , and this prediction is found to be consistent with the available experimental data . In particular , our model offers an explanation for the experimental observation that backward steps are smaller than forward steps on average . Another prediction of the model is that rotor speed grows sublinearly as a function of torque . At low torques rotation is slow because of trapping in local minima , whereas at high torques the barriers between minima are lowered and eventually eliminated . Additionally , we predict that at low torques rotor diffusion is hindered by barriers , while at high torques the variability of the potential actually enhances diffusion . Although in principle other sources of fluctuations , such as ion translocation , could impact rotor diffusion , we showed that in the relevant regimes simple diffusion can account for nearly all of the observed variance in cycle time [9] . In order to verify these predictions experimentally , one would need to simultaneously measure the rotor speed and the proton ( or ) motive force , believed to be proportional to torque at low speeds , in the regime where torque and contact forces are comparable . ( Note that in the stepping data [1] we have analyzed , torque could vary during the course of the experiment as the result of changes in the number of stator units . ) Interestingly , both a stepping behavior and a sublinear speed vs . torque relationship were reported in experiments on flagellar motors in Streptococcus [13] . Cells were starved , and then energized or de-energized to control PMF . Motion was found to have a rotational symmetry of , which corresponds to one revolution of the one-start helix of an axial component ( 11 in two revolutions ) . When energized ( increasing PMF ) , cells displayed a sublinear speed vs . PMF dependence in agreement with our prediction . However , when cells were de-energized ( decreasing PMF ) , this relation became linear . Such history dependence could occur if the stator elements hindering rotation are pushed away as PMF increases , leaving rotation unhindered during PMF decrease . Our model is consistent with other experimental results on the bacterial flagellar motor . Because the model relies on the assumption that the energy from ion translocation is reversibly stored in protein springs [10] , it implies a near-perfect efficiency of the motor at low torques [2] . The same mechanism can account for both clockwise and counterclockwise motor rotation [3]—these two cases simply corresponding to the springs being stretched in opposite directions . If the contact potential stays the same when the motor changes direction , our model predicts that backward steps will occur preferentially at the same absolute angles irrespectively of the direction of rotation . The observation that the duty ratio is very close to one even with a single torque-generating unit [7] , [14] can be encompassed in our model by assuming that each torque-generating unit comprises at least two springs . The advantage of a spring mechanism over other mechanisms is that it naturally entails efficiency , at least at lower speeds . When the system is in thermodynamic equilibrium , which is the case near stall where kinetic rates are much faster than spring relaxation , the average energy provided to the springs by the passage of one proton is exactly equal to the potential energy difference between the exterior and the interior of the cell . This is simply a consequence of reversibility . The key point is that all the energy stored in protein springs is eventually used to move the rotor . This stands in contrast to mechanisms driven by irreversible conformational changes , where some energy is typically wasted because the energy required for the conformational change is less than the energy provided by the source , e . g . ATP hydrolysis for myosin motors . The utilization of springs for the reversible storage of the energy suggests a general mechanism underlying the operation of high-efficiency molecular machines . Our study has focused on the adiabatic regime , where springs are near equilibrium with respect to the PMF , as this is the relevant regime for the stepping experiments . However , our model predicts that when the proton flux becomes kinetically limited , the springs will fail to restretch completely , causing the torque to drop . This observation can explain the observed “knee” in the motor's torque-speed relationship [5] , when coupled to an explicit model of proton translocation [15] . In our analysis we have neglected one effect that is not crucial for our analysis , but which may prove important for inferring the detailed nature of the contact potential . Specifically , we have assumed that equilibration of the elastic linkage between the motor and the load is rapid compared to the waiting time between steps . For a torsion constant [16] , [17] between the rotor and the load , and a drag coefficient , the relaxation time is . In contrast , the typical waiting time between steps ranges from , depending on experimental conditions . If the elastic linkage was too soft , the polystyrene bead would respond to the motion of the rotor with a delay , and steps would be smoothed out . This does not seem to occur in the experiment . Another effect , which we have considered ( see Rotor diffusion ) but did not include in our simulations , is the relaxation of MotA/B protein springs as a rotor step occurs . For example , when the rotor moves forward , the torque decreases because the protein springs relax . Usually these springs are restretched so quickly by ion translocation that the transient decrease of torque can be neglected . However , during a barrier crossing event the rotor motion might be so fast that protons are not able keep up . This would result in a temporary drop in torque and make barrier crossing more difficult . A similar argument applies to backward steps . We have already shown that at the “mean-field” level , where rotation speed and ion flux are time-averaged , this negative feedback has only a small effect . However , the instantaneous rotation speed during a step can be much larger than its average . How fast can a proton translocate through the motor ? The maximum flux of protons through a single motor unit can be estimated by considering the maximum rotation speed before the torque starts dropping ( the “knee” of the torque-speed relationship [5] ) . For a single motor unit in natural conditions , this speed is about 150 Hz for a torque of [7] . The power generated by the motor is then . Each proton provides at most , so that the number of protons per second is at least . The timescale of proton passage is therefore less than . A single rotor step corresponds to the passage of [1] , so restretching the protein springs should take less than , which is below the current experimental time resolution . For comparison , an instanton calculation [18] reveals that the typical time for crossing a barrier is bounded from below by ( for , and ) . In our analysis we have also neglected the effect of the “shot noise” arising from the discrete nature of the proton flux . This shot noise leads to fluctuations in torque , which could in principle affect the stepping behavior as well as the rotor diffusivity . While we have neglected this source of noise on the basis that it is averaged out by the presence of multiple stators , its influence can be significant at very low loads [15] . Other molecular motors have shown stepping behavior , including the actin-myosin motor [19] , the dynein-microtubule motor [20] , and kinesin [21] . In these ATP-powered motors , which are less powerful than the bacterial flagellar motor by orders of magnitude , stepping is a built-in and essential part of motor operation . By contrast , we have argued that in the bacterial flagellar motor the observed stepping arises solely from steric hindrance . Our work leaves open a number of questions . It would be interesting to infer the precise form of the contact potential from rotation data and see how and whether it varies in time and among motors , potentially yielding new insight into the dynamics of motor assembly and reorganization . To this end a more sophisticated approach to learning the potential may be required , e . g . employing maximum likelihood techniques . Lastly , one still needs to understand the mechanism of torque generation , including the role played by the discreteness of ion translocation , the chemical nature of protein springs and their attachment sites , as well as the energy conversion process .
All the experimental data presented in this paper is from [1] and were used with the kind permission of Richard Berry . The simulation data were obtained by numerical integration of Eq . 1 by Euler's method . The same step finding algorithm as the one described in [1] was used on both simulation data and experimental data to extract steps . First , an entire episode ( angular position vs . time ) was fitted by a single step function , and thus divided into two intervals . This procedure was repeated iteratively times—at each iteration the interval for which the data had the largest range of angles was replaced by a best-fit step function . Then , a quality factor was calculated for each assigned step , where ( ) is the angular position left ( right ) of the step , and ( ) is the number of data points on the left ( right ) of the step . Finally , steps with a low quality factor were removed and their adjacent intervals merged until all steps have a quality factor greater than . This step finding algorithm introduces a small bias that tends to underestimate the sizes of backward steps . For each interval , the algorithm finds the mean value of the angular position in this interval . But steps do not occur instantaneously , and the data points leading from one interval to the other are themselves included in the interval means . As a result , these means are biased toward where the rotor is coming from and where it is going . When the rotor is stepping forward , these two biases tend to cancel each other . However , when the rotor steps backward , these steps are usually both preceded and followed by forward steps . Thus , the interval before the backward step is biased toward lower angular positions , and the interval after the backward step is biased toward higher positions . As a result , the biases reinforce each other such that backward steps are estimated to be smaller than forward steps . We have checked that even on a perfectly 26-fold periodic potential , with and a torque , the most probable forward step size according to the algorithm was , which is correct , while the most probable backward step size was , which is to small . The step frequencies presented in Fig . 4A were obtained analytically using first-passage theory [22] , [23]: Label the wells , and denote the local minima of by . Consider three consecutive wells centered at , and respectively . Starting at , call the probability of first jumping forward and the probability of first jumping backward . These probabilities are given by: ( 5 ) Given the transition probabilities and , we write a master equation for the probability of the rotor being in well after steps . ( 6 ) At large a stationary state is reached , and satisfies the conservation equation: ( 7 ) The step frequencies presented in Fig . 4 are then given by and . ( Note that the step frequencies sum to one , . ) The torque-speed relation shown in Fig . 5A was also estimated using first-passage theory . For simplicity we assumed a perfectly 26-fold periodic potential , but the results are qualitatively the same when the periodicity is only approximate . The average time to move from one minimum of the potential to the next is given by: ( 8 ) where and is the torque . This yields a rotation speed: ( 9 ) where and are obtained from Eq . 12 . The asymptotic expansion of Eq . 9 for yields ( 10 ) The effective diffusion coefficient shown in Fig . 5B is estimated using similar techniques ( cf . [22] , [23] and Materials and Methods ) . The effective diffusion coefficient is estimated using the techniques of first-passage theory [22] . For an exactly 26-fold periodic potential , this effective diffusion coefficient is given by [23]: ( 11 ) where is the step size . and are the probabilities of first jumping forward or backward , respectively , and are given by: ( 12 ) is the time before a step occurs ( forward or backward ) . Its first two moments are: ( 13 ) ( 14 ) with ( 15 ) Expanding Eq . 11 for large torques gives: ( 16 ) To estimate the magnitude of the negative feedback of the MotA/B protein springs on diffusion , we model the dynamics of rotor angle and torque by the following mean-field differential equations: ( 17 ) ( 18 ) The first equation is the same as Eq . 1 but with a variable torque and without contact forces . In the second equation , torque is assumed to follow the stretching/unstretching of the springs with rotation , and therefore the rate of change of torque is linearly related to the rate of change of rotor angle through an effective spring constant ( second term of r . h . s . of Eq . 18 ) . At the same time , due to the restretching of springs upon proton passage , torque relaxes toward its equilibrium value ( first term of r . h . s . of Eq . 18 ) . Solving the second equation for steady-state rotation yields the torque-speed relationship for the motor: , with . Solving both equations for the effective diffusion coefficient in the limit , we find . We recorded the distribution of waiting times obtained by the step-finding algorithm , both for the simulation and for the experiment . For the simulation , in the case of a perfectly 26-fold periodic potential ( Fig . 6A ) , there is only one type of barrier , and the waiting-time distribution is approximately exponential . When the potential is only approximately 26-fold periodic ( Fig . 6B ) , the waiting-time distribution is the sum of 26 exponentials , and resembles a stretched exponential . The waiting-time distribution for the experiment ( one cell , Fig . 6C ) is consistent with a sum of exponentials . The stretched appearance of the experimental distribution may be due to non-uniform barriers , as predicted by our model , but may also be due in part to the observed variability in average speed , presumably due to changes in torque , during the course of the experiment . For all numerical simulations and analytic calculations the potential was chosen to be of the form: ( 19 ) Experimentally , there is evidence for components of the contact potential with fold periodicity ( see Ref . [1] , Fig . 3b ) . Except where stated otherwise , we used , and . In Fig . 2 and 4A , we used . In Fig . 4B and 4C we chose three sets of values for , , and , in units of : ( ) , ( ) and ( ) . In Fig . 5 , the torque-speed relation and rotor diffusion were calculated with .
|
Many species of bacteria swim to find food or to avoid toxins . Swimming motility depends on helical flagella that act as propellers . Each flagellum is driven by a rotary molecular engine–the bacterial flagellar motor–which draws its energy from an ion flux entering the cell . Despite much progress , the detailed mechanisms underlying the motor's extraordinary power output , as well as its near 100% efficiency , have yet to be understood . Surprisingly , recent experiments have shown that , at low speeds , the motor proceeds by small steps ( ∼26 per rotation ) , providing new insight into motor operation . Here we show that a simple physical model can quantitatively account for this stepping behavior as well as the motor's near-perfect efficiency and many other known properties of the motor . In our model , torque is generated via protein-springs that pull on the rotor; the steps arise from contact forces between static components of the motor and a 26-fold periodic ring that forms part of the rotor . Our model allows us to explain some curious properties of the motor , including the observation that backward steps are shorter on average than forward steps , and to make novel , experimentally testable predictions on the motor's speed and diffusion properties .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biophysics/macromolecular",
"assemblies",
"and",
"machines",
"biotechnology/bioengineering",
"biophysics/theory",
"and",
"simulation"
] |
2009
|
Steps in the Bacterial Flagellar Motor
|
Glycosylation is the most abundant post-translational polypeptide chain modification in nature . Although carbohydrate modification of protein antigens from many microbial pathogens constitutes important components of B cell epitopes , the role in T cell immunity is not completely understood . Here , using ELISPOT and polychromatic flow cytometry , we show that O-mannosylation of the adhesin , Apa , of Mycobacterium tuberculosis ( Mtb ) is crucial for its T cell antigenicity in humans and mice after infection . However , subunit vaccination with both mannosylated and non-mannosylated Apa induced a comparable magnitude and quality of T cell response and imparted similar levels of protection against Mtb challenge in mice . Both forms equally improved waning BCG vaccine-induced protection in elderly mice after subunit boosting . Thus , O-mannosylation of Apa is required for antigenicity but appears to be dispensable for its immunogenicity and protective efficacy in mice . These results have implications for the development of subunit vaccines using post-translationally modified proteins such as glycoproteins against infectious diseases like tuberculosis .
Mycobacterium tuberculosis ( Mtb ) , the etiologic agent of tuberculosis ( TB ) , produces an array of protein antigens ( Ags ) , many of which are post-translationally modified [1]–[4] , which constitute important determinants of innate and adaptive immune response . As with other pathogens , the post-translational protein modifications influence host interactions . In particular , carbohydrate modification of proteins serves as an efficient ligand for innate C-type lectin receptors ( CLRs ) present on the antigen presenting cells ( APCs ) . Members of this receptor family play an important role in immune response induction , immune evasion , immune regulation and tolerance [5] . In addition to their role in innate immunity , carbohydrate modifications of protein Ags contribute to B cell epitopes , and it has been recently shown that glycopeptides may constitute a T cell epitope and can induce a strong T cell response [6] . Consequently , many glycoproteins or glyco-conjugates are considered Ags of interest in vaccine development . However , little is known about the role of protein glycosylation in T cell immunity in TB , and a better definition of immune responses to glycoproteins may aid in deciphering their role in protection or pathogenesis of Mtb . The 45-47-kDa secretory and cell surface alanine-proline-rich antigen ( Apa; Rv1860 ) is one of the few glycoproteins of Mtb for which the complete glycosylation pattern has been described [7] , [8] . The modifications of native Apa ( nApa ) consist of complex O-mannosylation of specific threonine residues present at the N- and C-terminal domains [8] , [9] . Apa is produced by all members of the Mtb complex including the vaccine strain , M . bovis bacillus Calmette Guerin ( BCG ) . The nApa shares significant amino acid homology with a family of fibronectin attachment proteins found in other mycobacteria such as M . avium , M . marinum and M . leprae and is shown to possess fibronectin-binding activity [10] . In Mtb , mannosylated nApa is a potential adhesin and has a role in host cell attachment , entry , and immune evasion [11] , [12] . The nApa is an immunodominant Ag and has been found to be strongly recognized by serum antibodies ( Abs ) of active TB patients [13] , and the Ab reactivity was mainly directed against mannose residues . The Ag-specific T cell recognition of nApa also requires mannosylation . These mannose residues are absent in the recombinant form of Apa ( rApa ) expressed in E . coli , and the rApa is poorly recognized by T cells of BCG inoculated guinea pigs [14] , [15] . We and others have identified Apa as a possible vaccine candidate against TB [16]–[18] , but the role of glycosylation in inducing protective T cell immunity had not been studied . Here , we investigated the effect of Apa mannosylation on the T cell antigenicity after mycobacterial infections and the immunogenicity and protective efficacy following protein vaccination . Antigenicity is the ability of the molecule to react with a preformed Ab or T cell receptor ( TCR ) while immunogenicity is the ability to elicit a cell-mediated or humoral immune response [19] , which may or may not impart protection against a pathogen . Our results demonstrate that the mannosylation of Apa is critical for its T cell antigenicity in humans and mice after mycobacterial infections and provide evidence that a synthetic Apa glycopeptide constitutes a T cell Ag . We show that glycosylation of Apa is expendable for T cell immunogenicity and protective efficacy when used either as a subunit vaccine or as a BCG-booster vaccine against Mtb infection in BALB/c mice . Our data suggest that comparable immunogenicity and protective efficacy of mannosylated and nonmannosylated Apa forms may not be due to the inability of nApa to induce glycopeptide-specific T cells , as generation of nApa-specific hybridomas following subunit vaccination of mice identified a T cell clone specifically reactive to the N-terminal glycopeptide of nApa . Importantly , our data suggest that Apa may be considered a possible component of future vaccines against TB to boost waning BCG immunity , regardless of glycosylation .
We first determined the role of Apa mannosylation in T cell recognition and recall of cytokine responses in healthy , BCG vaccinated ( BCG+ ) and BCG unvaccinated ( BCG− ) adults . The peripheral blood mononuclear cells ( PBMCs ) from 17/24 ( 70 . 8% ) BCG+ individuals produced more than 50 IFN-γ spot forming units ( SFU ) /106 cells ( a positive response ) after in vitro stimulation with purified nApa , as compared to those from only 3/24 ( 12 . 5% ) individuals after stimulation with rApa ( Figure 1A ) . Among BCG+ individuals , an IFN-γ response was predominantly observed in individuals with positive skin test reactivity to Mtb purified protein derivatives ( PPD+ ) , with 15/16 ( 93 . 7% ) individuals showing nApa-specific positive response . Only 2/8 ( 25% ) BCG+ individuals without PPD reactivity ( PPD− ) and 1/22 ( 4 . 5% ) controls , i . e . BCG−PPD− individuals , responded positively to nApa . This dominant nApa-specific T cell response was dose dependent , as demonstrated by the increased frequency of IFN-γ SFU when PBMCs of 5 healthy PPD+ ( including 3 confirmed Mtb-exposed ) individuals were stimulated with varying amounts of nApa , in contrast to different doses of rApa ( Figure S1A ) . Overall , these results demonstrate that nApa is recognized preferentially over rApa in these individuals . Next , we characterized the Ag-specific IFN-γ , TNF-α and IL-2 cytokine responses of 8 randomly selected BCG+PPD+ subjects by intracellular cytokine staining ( ICS ) and polychromatic flow cytometry and determined the percentages of CD4+ and CD8+ T cells expressing one ( 1+ ) , any combination of two ( 2+ ) , or all three ( 3+ ) cytokines . Significantly higher frequencies of IFN-γ , TNF-α or IL-2 producing CD4+ T cells were observed in nApa stimulated PBMCs than those stimulated with rApa ( Figure 1B and 1C ) and the response was characterized by higher percentages ( but comparable proportions ) of nApa-specific polyfunctional CD4+ T cells in the responding donors ( Figure S1B ) . In contrast , the frequency of total cytokine producing CD8+ T cells was not statistically different after stimulation with the two forms of Apa ( Figure 1C ) , although 4 donors had more nApa-specific cytokine producing CD8+ T cells . Collectively , these results suggest that mannosylation of Apa is important for the Ag-specific T cell recall responses in these individuals . The heightened T cell responses seen with mannosylated Ags are presumably due to efficient uptake through mannose binding CLRs on the APCs such as dendritic cells ( DCs ) and higher efficiency of Ag presentation [20]–[22] . Therefore , we sought to determine whether Apa mannosylation influences its uptake by DCs . We found increased binding/uptake of FITC-labeled nApa as compared to labeled rApa by the blood monocyte derived dendritic cells ( MoDCs ) and by the CD11c+HLA-DR+ myeloid dendritic cells ( mDCs ) in the PBMCs of BCG+ and BCG− individuals ( Figure S1C and D ) . A higher percentage of myeloid ( mDCs ) but not plasmacytoid dendritic cells ( pDCs ) from human PBMCs expressed CD80 following in vitro stimulation with nApa than after rApa ( Figure 1D and 1E ) , suggesting increased activation of mDCs by nApa . It is known that mannose binding CLRs such as DC-specific intracellular adhesion molecule-3 grabbing nonintegrin ( DC-SIGN ) and mannose receptor ( MR ) are highly expressed on mDCs but are absent on pDCs , and play an important role in the uptake and presentation of mannosylated Ags [23] . Corroborating these observations , we found that nApa but not rApa binds to recombinant human MR , DC-SIGN and DC-SIGNR using an in vitro CLR adhesion assay ( Figure 1F ) . To determine whether Ag presentation of nApa to specific T cells requires intracellular processing , APCs were fixed with glutaraldehyde before or after Ag pulsing . The MoDCs generated from the blood monocytes of three nApa-responding individuals ( BCG+PPD+ ) were used as APCs , while T cells were purified from the PBMCs of respective individuals . The IFN-γ ELISPOT was used to investigate Ag presentation and activation of T cells after co-culture . Glutaraldehyde fixation of APCs before addition of nApa abrogated IFN-γ response ( Figure S1E ) , whereas fixation of APCs after pulsing with nApa did not prevent IFN-γ production , although the response ( SFU ) was lower than pulsing of Ag alone without fixation . These results indicate a requirement for intracellular processing of nApa for presentation to T cells . In contrast , fixation of APCs before addition of the superantigen , staphylococcal enterotoxin B ( SEB ) , did not prevent the IFN-γ response of T cells . Therefore , the lack of nApa-specific IFN-γ secretion by specific T cells , if fixation of APCs preceded Ag pulsing , is consistent with the presentation of cognate Ag ( versus superantigen ) [24] , [25] . Next , to determine whether increased binding/uptake and activation of mDCs by nApa influences T cell reactivity and epitope recognition , we screened 32 sequential nonmodified synthetic overlapping peptides of Apa with PBMCs of BCG+PPD+ individuals who recognized both forms of Apa and those who recognized nApa over rApa . Several peptides induced positive IFN-γ responses in PBMCs from individuals who recognized both forms of Apa ( Figure S1F ) , whereas no peptide induced >20 IFN-γ SFU/106 PBMCs from individuals who recognized only nApa . These results suggest that mannosylation of Apa does not induce alternate Ag processing in APCs to produce unique peptide epitopes , but rather carbohydrate contributes to T cell epitopes and recognition of such mannosylated epitopes ( for example glycopeptides ) is probably responsible for elevated T cell responses to nApa in these individuals . To determine whether Apa mannosylation is required for T cell antigenicity in animal models , BALB/c mice were infected with Mtb intranasally ( i . n . ) or with M . bovis BCG subcutaneously ( s . c . ) . At different time points after infection , lung , spleen and draining inguinal lymph node ( ILN ) cells were isolated and stimulated with nApa or rApa . The frequency of IFN-γ , TNF-α and IL-2 producing CD4+ T cells after nApa stimulation of lung and spleen cells from BCG-administered mice was comparable ( IFN-γ , IL-2 ) to 2-fold ( TNF-α ) more than that induced by nAg85B stimulation , whereas stimulation by rApa induced only a marginal increase in the frequency of cytokine secreting CD4+ T cells as compared to controls ( Figure 2A and S2A ) . Similar results were also obtained using ILN cells ( data not shown ) . The time kinetics of CD4+ T cell response confirmed the presence of significantly higher frequency of nApa-specific total cytokine secreting cells in the lung ( p = 0 . 0002 ) and spleen ( p = 0 . 0021 ) as compared to those specific for rApa ( Figure 2B ) and revealed that the Apa responses peaked at 32–52 weeks ( wks ) after BCG administration , with a similar expansion and contraction trend for the nApa- and rApa-specific T cells . During the expansion phase , 3 . 9- and 8 . 6-fold more nApa-specific total cytokine secreting CD4+ T cells were observed at 12 wks in the lung and spleen , while at the peak ( 52 wks ) the increase was 7 . 7- and 2 . 7-fold more in the lung and spleen , respectively . At individual cytokine levels , however , the difference between nApa- and rApa-specific cytokine producing CD4+ T cell frequency was more pronounced for IFN-γ and TNF-α ( Figure S2B ) . These results suggest that mannosylation of Apa strongly influences the magnitude of Apa-specific CD4+ T cell responses in BCG administered mice and indicates that the CD4+ T cells of these mice predominantly recognize major histocompatibility complex-II ( MHC-II ) restricted epitopes present in nApa only . On the contrary , the frequency of nApa- and rApa-specific total cytokine secreting CD8+ T cells was not significantly different at the time points evaluated , except at the 32 wk time point in the lung ( Figure 2B ) . This difference in the lung was mainly due to the presence of a higher frequency of nApa-specific IFN-γ producing CD8+ T cells ( Figure S2C ) . These results indicate that both forms of Apa possess MHC-I restricted epitopes recognized by the CD8+ T cells of BCG administered mice . Next , we compared the cytokine expression profiles of nApa- and rApa-specific CD4+ or CD8+ T cells . The lung CD8+ T cell response was dominated almost exclusively by 1+ cytokine ( IFN-γ ) producing cells ( Figure 2C , histograms ) . In contrast , about 38% and 50% of the nApa-specific cytokine secreting CD4+ T cells produced more than 1 cytokine at the peak ( 52 wk ) time point in the lung and spleen , respectively , and a significantly higher frequency of 2+ and 1+ cytokine producing CD4+ T cells was observed after stimulation with nApa than rApa in the lung ( Figure 2C , histograms ) . When the Ag-specific cytokine producing lung CD4+ T cells were categorized into 7 distinct subpopulations based on cytokine expression profiles , IFN-γ and TNF-α co-producing cells dominated nApa-specific 2+ cytokine producing cells ( Figure 2D ) , while IFN-γ or TNF alone producing cells were predominantly present among nApa-specific 1+ cytokine producers . Despite differences in the magnitudes , the proportions of nApa- and rApa-specific cytokine producing 3+ , 2+ and 1+ cytokine producing CD4+ or CD8+ T cells was not significantly different in both organs at the peak time point ( Figure 2C , pie charts ) . The proportions of nApa- and rApa-specific IFN-γ , IL-4 and IL-17 SFU in cultured ELISPOT assay were also not significantly different when compared at 3 time points; however , statistically higher frequency of IFN-γ or IL-17 SFU was present in the lung and spleen cell cultures after nApa stimulation ( Figure 2E ) . These results collectively suggest that Apa mannosylation may have only minor effect on the polyfunctionality and ratios of specific subpopulations of Apa-specific T cells in BCG inoculated mice . Similar dynamics were observed in Mtb Erdman infected mice ( Figure S2D–F ) . Thirty-two non-modified , synthetic overlapping peptides of Apa were screened for their capacity to induce IFN-γ response in mice at 12 and 32 wks post BCG infection . Only peptide p271-288 induced a positive response which only occurred 32 wks after infection ( Figure 3A ) . These analyses support the indication that heightened T cell responses to nApa are mainly due to glycosylation . We synthesized rApa C-terminal peptide ( residues 281–325; 45-mer ) with di-mannosyl-threonine residue at position 316 ( Fig . S3A ) , akin to that found in nApa , to confirm the role of Apa glycosylation in T cell antigenicity . A higher frequency of IFN-γ SFU was observed in the spleen and lung cell cultures of BCG and Mtb infected mice after stimulation with the synthetic glycopeptide as compared to stimulation with the non-glycosylated control peptide ( Figure 3B ) , confirming the role of Apa glycosylation in T cell antigenicity . A higher frequency of cytokine producing CD4+ T cells , but not CD8+ T cells , was found in the splenocytes of BCG mice after in vitro stimulation with glycopeptide compared to stimulation with control peptide ( Figure 3C ) , suggesting that the carbohydrate modification of nApa C-terminus may constitute a CD4+ T cell epitope . The T cell epitope prediction analyses also indicated probable binding of a 15-mer encompassing Thr316 to MHC-II molecules ( Figure S3B ) . To determine whether T cells recognize carbohydrate ( di-mannose ) in the absence of the proper peptide context , we synthesized rApa C-terminal 45-mer peptide with di-mannosyl-threonine residue at its N-terminus ( at an unnatural position ) linked by the additional glycine residues ( Figure S3C ) . We termed this Apa 45-mer with additional four N-terminal residues as a 49-mer glycopeptide ( i . e . , Apa non-glycopeptide residues 281–325 with Gly-Thr ( di-man ) -Gly-Gly extension ) . Significant cytokine response was observed in ELISPOT when the lung , spleen or ILN cells from the BCG mice were stimulated with the synthetic 45-mer glycopeptide with a di-mannosyl residue at its natural position ( Thr316 ) . On the contrary , no positive cytokine response was found after in vitro stimulation with the 49-mer glycopeptide , 49-mer nonglycopeptide ( control ) or the free di-mannose alone ( Figure S3D ) . These results suggest that the carbohydrate ( di-mannose ) attached to the proper peptide backbone is likely required for the recognition of nApa C-terminal glycopeptide by specific T cells from BCG mice . Further characterization of MHC bound glycopeptide interactions with the TCR and the orientation of carbohydrate residues can be accomplished through crystallography studies . To determine whether mannosylation of Apa also influences its ability to induce a protective immune response , mice were vaccinated s . c . with 3 doses of nApa or rApa ( 1 µg/dose ) in the presence of dimethyl-dioctadecyl ammonium bromide ( DDA ) and monophosphoryl lipid A ( MPL ) adjuvants at 4 wk intervals , and the T cell responses were investigated 1 wk after the last dose . Stimulation of splenocytes or ILN cells of vaccinated mice with either nApa or rApa , resulted in a comparable frequency of individual cytokine producing CD4+ T cells , regardless of which form of the Ag was used for vaccination . This Apa-specific T cell response was characterized by more TNF-α and IL-2 than IFN-γ producing cells in an ICS assay ( Figure 4A , 4B and S4A ) . The frequency of Ag-specific total cytokine producing CD4+ and CD8+ T cells induced in the spleen and lung by either Apa vaccine was also comparable ( Figure 4C ) . When the cytokine expression profiles of CD4+ T cells from the spleen and lung were evaluated in the vaccinated groups , IL-2 and TNF-α co-producing cells dominated 2+ cytokine producing cells , while TNF-α or IL-2 single-producers were predominantly present among 1+ cytokine producing cells ( Figure 4A ) . In contrast , the cytokine expression profiles of CD8+ T cells in the spleen and lung were dominated by TNF-α or IFN-γ secreting 1+ cytokine producers , respectively . Higher proportions of total cytokine producing splenic CD4+ T cells of rApa vaccinated mice consisted of polyfunctional T cells than those of nApa vaccinated mice ( Figure 4C ) . Otherwise , we found no significant difference in the quality of nApa- and rApa-specific response , when the proportions of 3+ , 2+ and 1+ cytokine producing CD4+ and CD8+ T cells were determined in the two vaccinated groups . Significantly more immunogen-specific IL-17 and IFN-γ SFU were observed in the spleens of rApa vaccinated as compared to nApa vaccinated mice in cultured ELISPOT assay ( Figure 4D , histograms ) . However , the proportions of specific IFN-γ , IL-17 and IL-4 SFU constituting the total ELISPOT response were not significantly different in nApa or rApa vaccinated mice ( Figure 4D , pie charts ) . Synthetic peptide screening in nApa and rApa vaccinated mice revealed that T cell cytokine responses were predominantly directed toward non-glycosylated p271-288 and p231-250 peptides ( Figure S4B ) , regardless of the immunogen used , indicating that Apa mannosylation does not induce alternate Ag processing in vivo when administered in DDA-MPL . Next , we evaluated the lung and splenic T cell responses of C-terminal 45-mer glycopeptide ( Gp-281-325 ) in nApa vaccinated mice . Comparable cytokine SFU after in vitro stimulation with 45-mer glycopeptide and its non-glycosylated control were found , regardless of dose of nApa ( 1 or 10 µg ) used for vaccination ( Figure 4E ) , and the p271-288-specific response was significantly more than that induced by Gp-281-325 , collectively suggesting that the differences observed between nApa and rApa induced T-cell responses may be partially overcome by co-administration of appropriate adjuvants . Since nApa is also glycosylated with complex mannose modifications at its N-terminus ( Figure S3B ) , the presence of N-terminal glycopeptide ( s ) -specific T cells in our nApa vaccinated mice samples remained a possibility . Our preliminary attempt to synthesize and evaluate N-terminal glycopeptide ( 39-mer ) was unsuccessful , and whether nApa glycopeptide-specific T cells are generated following subunit vaccination was not clear . Therefore , to identify nApa-specific T cell clones we employed a specialized protocol of T-cell hybridoma generation , using a single dose nApa vaccination of BALB/c mice in Freund's incomplete adjuvant ( FIA ) . Individual T cell clones responding to nApa were purified by serial dilutions and tested for reactivity to nApa or rApa using an IL-2 capture ELISA after co-culture with APCs ( i . e . , syngeneic bone marrow derived dendritic cells ) pulsed with Ag . Of the total 17 Apa-reactive hybridoma clones developed , 7 clones recognized both nApa and rApa while 10 clones recognized only nApa and not rApa ( data not shown ) . Three of the 7 clones that recognized both nApa and rApa responded to a specific synthetic , nonmodified 15-mer peptide ( two reactive to peptides spanning p271-288 , and one to peptide p70-84 , data not shown ) , while none of the 10 nApa reactive clones recognized any of the synthetic , overlapping , nonmodified peptides . Of the 10 clones only reacting to nApa , one clone , 4C3 , reacted with a glycopeptide fraction of trypsin digested nApa fractionated by a reversed phase-HPLC column chromatography . Clone 4C3 produced significant amount of IL-2 when cultured with APCs pulsed with the whole digest of nApa but not with the digest of rApa ( Figure 5A ) , indicating that the Ag presentation of nApa to 4C3 T cell clone was not inhibited by trypsin digestion . The RP-HPLC fractions consisting of the N-terminal 106 amino acid glycopeptide ( residues p40-145 ) of nApa only demonstrated reactivity in IL-2 assay ( Figure 5B , Figure S5A and S5B ) . Glycosylation of the N-terminus of nApa from Mtb likely explains specific recognition of the 4C3 clone to the nApa peptide . The lack of biological activity of T cell clone 4C3 when presented with rApa or synthetic nonglycosylated peptides collectively suggest that N-terminal glycopeptide-specific T cells are generated after nApa subunit vaccination . Further confirmation of these results will require identification of the precise epitope and synthesis of the N-terminal glycopeptide . We investigated the protective efficacy of two Apa forms using DDA-MPL , a known Th1 adjuvant . BALB/c mice were vaccinated s . c . with nApa or rApa 3-times at 4 wk intervals using two different concentrations ( 1 or 10 µg ) , in DDA-MPL adjuvant and challenged with Mtb 4 wks after the last dose to assess their protective potential . Mice vaccinated with nAg85B , a known protective Ag , were included for comparison . Mice immunized once with live BCG at the start of vaccination were used as positive controls , while negative controls received adjuvant alone ( 3-times ) or were left untreated ( naïve ) . Both forms of Apa imparted significant protection compared to naïve and adjuvant control groups at 2 different doses used ( Figure 6A ) , and the bacterial burden among the Apa vaccinated groups was not significantly different . Vaccination with nAg85B also induced protection , but only nAg85B-10 µg was statistically different from the negative control groups at the level of lung and spleen . However , the degree of protection afforded by BCG vaccination was greater than any of the subunit vaccinated groups . Vaccination with either nApa or rApa induces comparable levels of protection in our model and suggests that glycosylation of Apa is not critical for protection . To ascertain whether comparable protective efficacy offered by nApa and rApa vaccination in mice correlated with an equal ability to recall cellular and humoral immune responses after Mtb challenge , T cell responses and serum Ab titers were measured before Mtb challenge ( at 4 wks after last vaccination ) and 6 wks post challenge . All Apa vaccinated groups presented with an equal vaccine immunogen-specific CD4+ T cell recall response in the lungs after challenge , characterized by a significantly higher frequency of total cytokine producing CD4+ T cells in the post-challenge versus pre-challenge group ( Figure 6B ) . Significant recall from spleen derived T cells was only observed in rApa-1 µg group . No significant differences were observed when the immunogen-specific lung or spleen CD8+ T cell responses were compared between vaccine immunogens; however , the CD8+ T cell responses of all vaccinated groups were higher than those of control groups in the lungs . Both nApa and rApa were recognized equally well when used for stimulation of lung and spleen cells of vaccinated and challenged groups ( Figure S6A ) . Comparable patterns of cytokine co-expression profiles were found when T cell responses of respective Apa vaccinated groups were compared before and after challenge ( Figure 6B ) ; characterized by immunogen-specific TNF-α and IL-2 producing 2+ and 1+ CD4+ T cells in the spleen and IFN-γ producing 1+ CD4+ and CD8+ T cells in the lungs . On the contrary , splenic CD4+ T cells responses of the challenged naïve controls were characterized by nApa-specific IFN-γ and TNF-α producing 1+ and 2+ CD4+ T cells only . The cultured ELISPOT assay revealed that the frequency of immunogen-specific total IFN-γ , IL-17 and IL-4 SFU in the lung was significantly lower in vaccinated-challenged groups compared to respective vaccinated group pre-challenge ( Figure 6C ) . These results suggest possible decrease in in vitro proliferation of immunogen-specific lung T cells in ELISPOT cultures after recent in vivo recall or regulatory responses suppressing cytokine release following Mtb challenge . Proof for any of these explanations will require further in-depth investigation . All Apa vaccinated groups produced comparable amounts of immunogen-specific serum IgG1 Abs before and after challenge ( Figure 6D ) , and these Abs recognized nApa and rApa equally well ( Figure S6B ) . However , rApa vaccination induced significantly higher amounts of IgG2a and IgG2b Abs than nApa vaccination , except at 10 µg dose post-challenge . Of importance , when pathogen-specific immune responses were investigated in the lung and spleen of vaccinated and control groups of mice in vitro , higher frequency of Mtb short term culture filtrate ( STCF ) -specific total cytokine producing CD4+ and CD8+ T cells was observed in naïve and adjuvant only mice as compared to vaccinated mice post-challenge ( Figure 7A ) , with a significant difference at the level of lung . As expected , STCF-specific responses of unvaccinated-challenged mice were dominated by IFN-γ and TNF-α 1+ and 2+ CD4+ T cells and IFN-γ 1+ CD8+ T cells . Furthermore , when STCF-specific IFN-γ , IL-17 and IL-4 SFU were compared among challenged groups , responses of unvaccinated mice were found to be highly skewed toward IFN-γ ( 87–96% of total response ) , with significantly more SFU compared to all vaccinated groups ( Figure 7B ) . Considering the fact that most individuals in the world have been vaccinated with BCG and new TB vaccines have to be considered in the context of prior BCG vaccination , development of strategies aimed at boosting and improving the protective immunity induced by BCG is considered to be one of the rational approaches to develop an effective vaccination regimen against TB . Since BCG induced protection wanes significantly by 18 months in mice ( unpublished data ) , we investigated the booster effect of 2 doses of nApa or rApa administered 3 wks apart in DDA-MPL at 16 months after BCG vaccination . Both forms of Apa significantly boosted T cell responses compared to boosting with saline , and a significantly higher frequency of T cells from the Apa-boosted groups recognized nApa over rApa after in vitro stimulation ( Figure 8A and B ) . Nonetheless , comparable nApa-specific responses were found in both Apa-boosted groups , except for splenic IFN-γ SFU . Of importance , boosting with either form of Apa , but not with saline , significantly reduced Mtb CFU in the lung and spleen of BCG mice compared to age-matched controls ( Figure 8C ) . BCG mice boosted with nApa or rApa showed about 2 . 0-log reduction in bacterial counts in the lungs as compared to age-matched naïve mice , while 1 . 7-log to 2 . 0-log reduction was found in the spleens of nApa- and rApa-boosted mice , respectively . This protection is characterized by a significantly lower frequency of ESAT-6+CFP-10-specific total as well as CD4+ T cell IFN-γ responses in the spleens of Apa-boosted groups ( Figure 8D ) .
Glycosylation generates an enormous variety of modified protein-derived epitopes . However , the significance of such modified antigenic epitopes in intracellular bacterial pathogens in shaping the T cell repertoire and determining the outcome of an immune response is poorly understood . Here we demonstrate that mannosylation of Mtb Apa ( also known as MPT-32 or ModD ) is crucial for its T cell antigenicity during mycobacterial infections of humans and mice and provides direct evidence that bacterial mannopeptides are T cell Ags using a synthetic Apa glycopeptide . Previously , glycopeptides containing tumor associated carbohydrates [26] , [27] , glycopeptides of self-Ags in autoimmune diseases [28] , and artificially glycosylated peptides of model Ags [6] have been shown to be recognized by T cells . In antigenic glycopeptides , the peptide backbone usually provides the binding motif that interacts with the MHC molecule , while the glycan provides an important part of the structure ( epitope ) that is recognized by the T cell receptor [29] . Conversely , pure carbohydrates are usually incapable of MHC binding and T cell stimulation , and due to their haptenic nature requires a carrier . The exception to this immunologic paradigm is zwitterionic polysaccharides which are presented in an MHC class-II restricted manner [30] . Because complex carbohydrates are not removed during processing by DCs , it remains possible that mannosylated peptide epitopes within nApa are recognized in vitro by T cells primed in infected mice and humans [31] . Our synthetic peptide screenings suggest that the recognition of carbohydrate containing epitopes ( i . e . , glycopeptides ) rather than increased Ag uptake , mDCs activation or alternate Ag processing leading to enhanced recognition of other peptide epitopes is the major determining factor for heightened T cell responses to nApa during infection . Mapping of the precise mannosylated epitopes will further help to identify whether mannose binds to the MHC groove along with peptide backbone or interacts strictly with TCR and will provide direct insight into the role of mannosylation in Ag presentation of Apa to T cells . Unexpectedly , vaccination of mice with nApa or rApa induced comparable CD4+ or CD8+ T cell frequency , regardless of the form of Apa used for in vitro stimulation . Nonetheless , these data together with post-challenge recall responses revealed that the paucity of anti-rApa T cell responses after primary infection , especially CD4+ T cell responses , is not due to an intrinsic inability of rApa to induce these responses . However , why nApa is recognized highly discriminately by T cells during infection but not after subunit vaccination remains to be answered . Speculatively , mannosylated epitope ( s ) of nApa might be effectively generated in phagosomes during intracellular growth of BCG or Mtb in vivo , but the relative dominance of mannose bearing epitopes versus peptide epitopes may change during processing of nApa in the presence of DDA-MPL in the endo-lysosomes of APCs in vivo following subunit vaccination . It is known that transport to different types of compartments in APC can lead to differential processing and generation of different epitopes within the same protein [32] and that different T cell epitopes in the same protein have been found to be recognized during infection and after subunit vaccination [32] . In our protocol , the majority of specific T cells generated after nApa-DDA-MPL vaccination might have been directed against the dominant peptide epitopes ( such as p271-288 and p231-250 ) and the frequency of glycopeptide-specific T cells , if at all generated , might have been diminished in nApa stimulated in vitro cultures . The carbohydrate or glycopeptide epitope-specific T cells induced by glycoprotein vaccination have been reported to be scarce and highly sensitive to in vitro culture conditions [6] , [33] . The requirement of specialized protocols for their expansion [33] including generation of T cell hybridomas has been suggested . Generation of nApa-specific T cell hybridomas following subunit vaccination of mice using FIA identified a T cell clone ( 4C3 ) specifically responding to the glycosylated N-terminal peptide of nApa . The dominant nature of peptide epitope p271-288 was also confirmed , because two hybridomas ( 3F7 and 2D10 ) that recognized both nApa and rApa strongly recognized synthetic peptides spanning p271-288 in IL-2 assay ( data not shown ) . Therefore , the comparable immunogenicity of nApa and rApa following subunit vaccination may not be due to the inability of nApa to induce glycopeptide-specific T cells . Of paramount importance , both nApa and rApa offered comparable protection against virulent Mtb challenge in mice , regardless of whether the 1 or 10 µg/dose was used for vaccination . Despite higher in vitro immunogen-specific cytokine SFU induced in the lungs by 1 µg than 10 µg/dose vaccination and the subtle differences in the quality of serum Ab response induced by the two forms of Apa vaccination prechallenge , comparable protection was found in all Apa vaccinated groups . Therefore , mannosylation appears to be dispensable for protective efficacy of Apa in DDA-MPL adjuvant in our model of progressive TB . It remains to be seen whether heightened T cell responses can be induced against nApa than its recombinant counterpart by using a different delivery vehicle or adjuvant system and whether the immunogenicity and protective efficacy of nApa differs from rApa in FIA , the adjuvant used for T cell hybridoma generation . However , immunization with a mannosylated model peptide in Freund's complete adjuvant ( FCA ) has been previously shown to induce poor Th1 effector functions despite enhanced Ag presentation [34] . Nonetheless , the ability of rApa to impart protection in this study is consistent with previous observation that i . n . subunit vaccination using rApa protein imparts notable protection in mice [18] . Of note , a comparable protection was offered by nAg85B and nApa vaccination despite the presence of 10 to 30-fold higher magnitude of immunogen-specific lung CD8+ and CD4+ T cells in nAg85B vaccinated mice ( data not shown ) . In addition , WCL-specific T cell responses induced by BCG in the lung prechallenge were 2 to 5-fold lower than the immunogen-specific responses induced by nAg85B vaccination , but , the protection imparted by nAg85B was not better than BCG . These results suggest that no direct correlation exists between the magnitude of immunogen-specific T cell responses induced by subunit vaccination and the degree of protection afforded against Mtb challenge . Previously we have shown that the Mtb proteins selected on the basis of dominant T cell and IFN-γ response during human infection do not necessarily impart stronger protection in vaccination experiments [35]–[36] . Interestingly , the decreased T cell responsiveness with higher dose ( 10 µg ) Apa vaccination did not adversely affect the protection using either form in challenge experiments . Understanding the mechanism of such immune regulation and Apa vaccination induced protection will require further investigation . One of the important findings in this study is the particularly significant protective efficacy of Apa when used as a BCG-booster vaccine against Mtb challenge in the elderly mice with waning BCG induced protective immunity . Stronger and comparable protection observed in the nApa or rApa-boosted BCG mice , whose T cell responses are mainly primed against mannose modifications of nApa , further suggests that mannosylation of Apa is dispensable for BCG-boosting potential . It is conceivable that a similar effect might be seen in humans previously vaccinated with BCG in childhood and whose immunity has waned , although of course , this would require clinical verification . The progress of glycoproteins as vaccine candidates often suffers a serious roadblock during the developmental pipeline . It is usually due to perceived loss of immunogenicity when produced as a recombinant protein , synthetic peptides or DNA vaccine due to complete absence or lack of appropriate glycosylation . Despite effectiveness of several glyco-conjugate vaccines against extracellular pathogens , it has been suggested that artificial glycosylation of peptides and targeting of CLRs may not always lead to generation of protective immunity [37] , as CLRs are also targeted by pathogens to evade host immune responses [5] , [38] . Furthermore , some ligand-MR interactions can result in promotion of Th2 , anti-inflammatory or regulatory responses and may even lead to T cell anergy and tolerance [34] , [39] . Glycosylation can mask protective epitopes in Ag , prevent effective Ag processing , inhibit proteolysis by masking cleavage sites and shift the pattern of peptides processed by DCs [40] , [41] . Contrarily , tolerance induction or CLR engagement can be a disease defense strategy [42] , [43] and it is argued that a mere ‘Th1 vaccine’ employing conserved immunodominant Ags can induce hyper-immune activation and may be advantageous to Mtb [44] and a complex balance between pro- and anti-inflammatory host factors is required to prevent and control Mtb infection [45] . Our results , therefore , underscore the importance of conducting comparative protection studies in animal models with a glycosylated Ag and its non-glycosylated counterpart . This may prevent elimination of poorly antigenic but otherwise protective non-glycosylated candidate during preclinical evaluation and may help save lengthy , costly and complex developmental efforts to artificially glycosylate it , carried out with the aim to improve its immunogenicity . In summary , we provide the evidence that the natural glycosylation of a protein may differentially affect its antigenicity and immunogenicity- the two attributes known to influence the protective efficacy of TB subunit vaccines [19] , [46] . Although Apa mannosylation influences T cell antigenicity during infection , it is expendable for induction of protective immunity following vaccination . Recently , a protein-O-mannosylating enzyme has been found to be required for virulence of Mtb [47] and our results highlight the need for further investigation of the role of increased mannosylated epitope-specific T cells in infection . In view of the finding that Mtb Apa improves waning BCG immunity and imparts significant protection in elderly mice , it makes a strong case for its inclusion as a possible component of future vaccines against TB .
The study was approved by the Institutional Review Board of Centers for Disease Control and Prevention ( CDC ) , Atlanta , Georgia , USA , ( approved protocol number 1652 ) , and informed written consent was obtained from all human participants before withdrawal of venous blood . All animal experiments performed in BSL-II or BSL-III animal facilities were in strict accordance with the guidelines of the U . S . Public Health Service Policy on the Humane Care and Use of Animals and the Guide for the Humane Care and Use of Laboratory Animals . The Institutional Animal Care and Use Committees of Centers for Disease Control and Prevention , Atlanta , Georgia , and the Colorado State University , Fort Collins , Colorado , USA reviewed and approved these animal protocols ( approval numbers SABMOUC1664 , 1847 , SHIMOUC 1490 and CSU 98-026A ) . The nApa was purified from Mtb H37Rv culture filtrate by traditional and reversed-phase chromatography as described previously [8] . The recombinant clone for Apa ( pMRLB . 17 ) , WCL , STCF , nAg85B , rESAT-6 and rCFP-10 of Mtb H37Rv were obtained through the NIH Biodefense and Emerging Infection Research Resources Repository . The rApa was expressed in E . coli BL21 ( DE3 ) and purified from lysates by Nickel chromatography with endotoxin removal using analogous methods described elsewhere [48] followed by DEAE-Sepharose chromatography . The overlapping peptides of Apa and the C-terminal 45-mer and 49-mer ( glyco ) peptides were synthesized by the Fmoc/tbu solid-phase peptide synthesis strategy [49] , [50] . See Supplementary Methods ( Text S1 ) for details . Blood was obtained from 50 healthy adult individuals ( 26 BCG+ and 24 BCG− ) . All donors were HIV negative and were without any clinical signs of TB . All BCG+ donors received BCG as a neonatal vaccine . We excluded responses of 1 BCG− and 2 BCG+ donors who tested positive for hepatitis . All 16 BCG+ donors with positive tuberculin skin test reactivity ( >10 mm induration after intradermal injection ) had normal chest radiographs . However , 1 BCG− and 4/16 BCG+ donors demonstrated positive reactivity to ESAT-6+CFP-10 in PBMC ELISPOT assay . The Mtb exposure of these donors was further confirmed by commercial IFN-γ ELISPOT ( Oxford Immunotech ) and ELISA ( Cellestis ) tests ( IGRAs ) . Remaining 22 BCG− donors had negative tuberculin skin test and no known history of contact with individuals with TB . The exposure of donors to environmental mycobacteria is not known but all live in Atlanta area with very low incidence of atypical mycobacterial infections . All donors demonstrated reactivity to phytohaemagglutinin ( PHA ) in ELISPOT assay and none were anergic . Specific-pathogen-free , 6–8 wk old female BALB/c ( H-2d ) mice ( Harlan Sprague Dawley ) were used in the study . BCG vaccinated mice received 100 µl of 1×106 CFU Copenhagen s . c . above the gluteus superficialis and biceps femoralis muscles of hind legs ( 50 µl/leg ) . For protein vaccination , groups of mice received 200 µl of nApa , rApa or nAg85B ( 1 or 10 µg ) in DDA and MPL ( from Salmonella minnesota Re 595 ) [51] s . c . on hind legs ( 100 µl/leg ) either 3- ( subunit ) or 2-times ( prime-boost vaccination ) . Mtb infections were performed i . n . with 5×104 CFU Erdman strain . At 48 h after Mtb challenge , about 0 . 5% of the total CFU delivered could be cultured from the lungs . The bacterial burden was determined by plating organ homogenates onto Middlebrook 7H10 agar supplemented with OADC and TCH ( 2 µg/ml ) . CFU were enumerated after 4 wks of incubation at 37°C . IFN-γ , IL-4 or IL-17A ELISPOT assay was performed using commercially available human or mouse ELISPOT reagent set ( BD-Biosciences ) according to the manufacturer's protocol as described previously [18] , except that no DCs were added as supplemental APCs and cells were stimulated with 10 µg/ml of purified or complex Mtb Ags , synthetic peptides or di-mannose ( unless mentioned otherwise ) for 40 h in RPMI-1640 . See Supplementary Methods ( Text S1 ) for details . Expression of cell surface markers and intracellular cytokine production by human PBMCs or mouse organ cells after in vitro stimulation with Mtb Ags ( 10 µg/ml ) at 37°C for total 12 h was assessed as described previously [52] . See Supplementary Methods ( Text S1 ) for details . For fluorescent labeling of proteins , commercially available FITC labeling kit ( Pierce Biotechnology ) was used . The protein labeling was carried out in 50 mM sodium borate buffer ( pH 8 . 5 ) using 2 mg/ml protein concentration . Unreacted excess florescent dye was removed using spin columns with purification resin as per the manufactures instructions . Protein concentration was assessed using the BCA assay ( Sigma ) . The efficiency of labeling was determined by Nano Drop Fluorospectrometer and was comparable between nApa and rApa . Protein samples were stored in the dark at −20°C before use . Human blood monocyte derived DCs ( MoDCs ) were generated as described before [53] using recombinant human GM-CSF ( 80 ng/ml ) and IL-4 ( 40 ng/ml ) ( PeproTech ) . Freshly isolated human PBMCs or MoDCs ( day 6 ) were pulsed with FITC-labeled nApa , rApa or WCL ( 20 µg/ml ) and incubated at 4°C or 37°C for the indicated amounts of time . Cells were washed with PBS containing 10% FBS . Samples were then stained with desired Abs to identify DC subsets and analyzed using flow cytometry . Mtb nApa or rApa was coated onto ELISA plates ( Nunc , Maxisorp ) at 5 µg/well in 0 . 1 M carbonate-bicarbonate buffer ( pH 9 . 6 ) ; and coating took place for 18 h at room temperature . Wells coated with 5 µg/well of purified BSA fraction-V ( Fisher Scientific ) were used as negative controls while those coated with Mtb H37Rv WCL or mannose caped LAM served as positive controls . Blocking was carried out with 1% BSA for 30 min at 37°C and additional 1 . 5 h at room temperature in TSM buffer ( 20 mM Tris-HCL ( pH 7 . 4 ) containing 150 mM NaCl , 2 mM CaCl2 and 1 mM MgCl2 ) [54] . Soluble recombinant human DC-SIGN ( CD209 ) -Fc chimera , DC-SIGNR ( CD299 ) -Fc chimera or MR ( CD206 ) ( R&D Systems ) ( 2 . 5 µg/ml in TSM buffer ) was added and the adhesion was performed for 2 h at room temperature . Unbound CLR was washed away and the binding of DC-SIGN-Fc or DC-SIGNR-Fc was determined by an anti-IgG1-Fc ELISA using a peroxidase conjugate of goat anti-human-Fc . Binding of MR was determined by anti-human-MR ( CD206 ) mouse mAb ( clone 685641; R&D Systems ) followed by peroxidase conjugate of anti-mouse IgG . The plates were developed using o-phenylenediamine dihydrochloride ( Sigma-Aldrich ) and absorption was measured at 492 nm . Specificity of CLR binding was determined by blocking the interaction in the presence of either 2 mg/ml mannan or 5 mM EDTA ( OD 492 nm <0 . 1 ) . The PBMCs were isolated from the venous blood of healthy BCG+PPD+ human donors who responded to nApa in IFN-γ ELISPOT assay . Recombinant human GM-CSF and IL-4 developed MoDCs ( day 6 ) were used as APCs while T cells purified from the PBMCs of respective donors using ‘Dynabeads Untouched Human T Cells’ purification kit ( Invitrogen ) and magnetic separation ( >95% pure ) were used as effector cells . APCs ( 2×106 ) were pulsed with either nApa ( 10 µg/ml ) or SEB ( 1 µg/ml ) or without any Ag ( complete RPMI-1640 media only ) and incubated at 37°C . After 4 h , APCs were washed with RPMI-1640 and were either left untreated or fixed with 0 . 05% glutaraldehyde for 5 min at 22°C . APCs were washed and the fixation was stopped by incubation with 0 . 2M glycine in RPMI-1640 at 22°C . After 5 min , cells were washed 4 times with RPMI-1640 . Alternatively , APCs were fixed and Ag pulse was carried out for 4 h . APCs were washed 4 times after Ag pulse and were co-cultured with purified T cells ( 1∶2 ratio ) in ELISPOT plates pre-coated with anti-human IFN-γ capture Abs . After 40 hr of incubation at 37°C in the presence of 5% CO2 , ELISPOT plates were developed and SFU were counted . T-cell hybridomas specific for Mtb nApa were generated as previously described [55] . Briefly , four BALB/c mice were each vaccinated with 40 µg of nApa in FIA by injecting 25 µl into each hind footpad and the remainder at the base of the tail . Five days after the vaccination , the draining lymph nodes ( popliteal , inguinal and periaortic ) were harvested to obtain lymph node ( LN ) cells . The primed LN cells were restimulated in vitro with syngeneic bone marrow derived dendritic cells ( BMDCs ) that had been pulsed with 10 µg/ml nApa per well . Approximately 1×106 primed LN cells were added per well in two 24-well tissue culture plates in a total volume of 1 ml of complete RPMI 1640 medium ( RPMI 1640 supplemented with 10% FCS , 5×10−5M 2-ME ( Sigma ) , plus a nutrient cocktail as described [56] ) . After two days of culture , the cells were harvested and pooled from all 48 wells , washed , fused with the T cell fusion partner BWα−β− [57] , and plated out into ten 96-well plates . Clones that grew in individual wells were screened using either BMDCs , as above , or a BALB/c mouse-derived B cell lymphoma line , A20 [58] , pulsed with 0 . 5 µg/well nApa . The selection of responding T cell hybridomas was performed by assaying 24 hr culture supernatants using paired rat monoclonal Abs specific for mouse IL-2 ( Pharmingen/BD-Biosciences ) in a capture ELISA . For Ag presentation studies , microculture wells were prepared containing 250 µl of culture medium , 5×104 each of T cell hybridoma and APC , and a known amount of nApa or rApa ( intact or trypsin digested ) , in flat-bottomed 96-well microtiter wells . Dose response curves with native and recombinant Ags were performed to determine which T cell hybridomas responded nApa alone , or to both the native and recombinant form of the protein . T cell hybridomas were further tested with a panel of synthetic , overlapping , nonglycosylated Apa peptides or trypsin digested and RP-HPLC separated nApa fractions to determine the peptide epitopes that were recognized . For details regarding trypsin digestion of Apa and characterization of RP-HPLC fractions please refer to Supplementary Methods ( Text S1 ) . ELISA assays of mouse sera were performed as described previously [46] , using 2 µg/ml of nApa or rApa for coating and HRP-conjugated anti-mouse IgG1 , IgG2a ( BD-Pharmingen ) and IgG2b ( Santa Cruz Biotech ) Abs for detection . Differences between groups were assessed by the parametric Student's t test or 1-way ANOVA followed by Bonferroni's test and the nonparametric 2-tailed Mann-Whitney U-test ( the Wilcoxon test for matched pairs ) or Kruskal-Wallis followed by the Dunn's post-test ( GraphPad Prism program ) . Unless indicated , all immune response data are presented after subtracting no Ag control values . A value of p<0 . 05 was considered to be significant and * <0 . 05; ** <0 . 01; *** <0 . 001 . The accession/identification numbers of Mtb proteins used in the study are Mtb ( H37Rv ) alanine and proline-rich antigen ( Apa ) , Rv1860 , NCBI reference sequence: YP_177849 . 1; Mtb ( H37Rv ) Ag85B , Rv1886c , NCBI reference sequence: NP_216402 . 1; Mtb ( H37Rv ) 6 kDa early secretory antigenic target ( ESAT-6 ) , Rv3875 , NCBI reference sequence: YP_178023 . 1; Mtb ( H37Rv ) 10 kDa culture filtrate protein ( CFP-10 ) , Rv3874 , UniProtKB/Swiss-Prot reference sequence: P0A566 . 2 .
|
Mycobacterium tuberculosis ( Mtb ) is the most devastating bacterial pathogen of all time that kills approximately 1 . 4 million people each year . Mtb modifies several of its proteins with sugar residues that influence many biological events . However , the significance of such sugar decorations and resulting carbohydrate and glycopeptide epitopes in shaping the T cell response during infection or after vaccination is insufficiently understood . Here , we show that the carbohydrate modifications of the Mtb Apa protein strikingly influence the magnitude of specific T cell responses in humans and mice after infection , but have only minor effect on the polyfunctionality and quality of T cell responses . The glycosylation of Apa was , however , expendable for T cell immunogenicity and protective efficacy when used either as a subunit vaccine or as a BCG-booster vaccine in dimethyl-dioctadecyl ammonium bromide ( DDA ) -monophosphoryl lipid A ( MPL ) adjuvant against virulent Mtb infection in mice . Our results suggest that the carbohydrate modification of microbial protein antigens may not always be critical for protection as our unmodified recombinant protein was sufficient for subunit vaccination . Together , our data underline the need to understand the role of heightened Apa glycoprotein-specific T cell responses in infection processes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
O-mannosylation of the Mycobacterium tuberculosis Adhesin Apa Is Crucial for T Cell Antigenicity during Infection but Is Expendable for Protection
|
Dengue virus is one of the most important arboviral pathogens and the causative agent of dengue fever , dengue hemorrhagic fever , and dengue shock syndrome . It is transmitted between humans by the mosquitoes Aedes aegypti and Aedes albopictus , and at least 2 . 5 billion people are at daily risk of infection . During their lifecycle , mosquitoes are exposed to a variety of microbes , some of which are needed for their successful development into adulthood . However , recent studies have suggested that the adult mosquito's midgut microflora is critical in influencing the transmission of human pathogens . In this study we assessed the reciprocal interactions between the mosquito's midgut microbiota and dengue virus infection that are , to a large extent , mediated by the mosquito's innate immune system . We observed a marked decrease in susceptibility to dengue virus infection when mosquitoes harbored certain field-derived bacterial isolates in their midgut . Transcript abundance analysis of selected antimicrobial peptide genes suggested that the mosquito's microbiota elicits a basal immune activity that appears to act against dengue virus infection . Conversely , the elicitation of the mosquito immune response by dengue virus infection itself influences the microbial load of the mosquito midgut . In sum , we show that the mosquito's microbiota influences dengue virus infection of the mosquito , which in turn activates its antibacterial responses .
Dengue has become one of the most important arboviral diseases , with infections rising at an alarming rate [1] . The dengue virus is transmitted by two highly anthropophilic mosquitoes , Aedes aegypti and Ae . albopictus . Although advances have been made toward the development of a vaccine , no cure for dengue is currently available [1] . Current methods are aimed at lowering the vector population through insecticide use , but there are concerns about the environmental impact of this approach as well as the rapid development of resistance in mosquitoes [2] . These setbacks have underscored the need for the development of additional methods to control dengue transmission . In the past decade , there has been a notable increase in research aiming at the potential application of microbes to control the transmission of vector-borne pathogens [3] . These studies have been encouraged by the fact that pathogens and microbes inhabit the same environment prior to infection ( the arthropod midgut ) and on the observation that pathogen infection is decreased in vectors harboring particular bacterial symbionts . In fact , the midgut is the site of multi-taxon interactions that include the arthropod vector ( host ) , vertebrate blood factors , the pathogen ( virus or parasite ) , and other symbiotic microbes . Although there is growing interest in these associations , our understanding of how these interactions at the molecular level and how they affect vector physiology and influence vector competence is still very basic . It has been shown that some of these interactions involve insect immune factors such as lectins , antimicrobial peptides , digestive enzymes , nitric oxide , and the prophenoloxidase complex [4]–[6] . Other factors and mechanisms that have been suggested to contribute to these interactions and to modulate vector competence include: bacteria-derived cytolisins ( hemolysins ) , siderophores , proteases , anti-parasitic factors , and secondary metabolites [4] . The purpose of the present study was to analyze the cultivable endogenous microbial flora of field mosquitoes collected from dengue-endemic areas in Panama and to assess their influence on the mosquito immune system and dengue virus infection . The incidence of dengue in Panama is the fifth-highest in Central America , and all four dengue virus serotypes are currently present in the country [7] . Molecular and infection assays have revealed intricate reciprocal interactions among the mosquito , the dengue virus , and its microbiota , with some bacterial isolates significantly affecting vector competence by reducing dengue virus infection of the midgut . In turn , the activation of the mosquito immune system by dengue virus infection alters the mosquito's immune homeostasis in the midgut , thereby affecting its microbiota .
The mosquito Ae . aegypti Rockefeller strain used in this study was maintained on a 10% sugar solution at 27°C and 95% humidity with a 12-h light/dark cycle according to standard procedures . Sterile cotton , filter paper , and sterilized nets were used to maintain maximum sterility of the cages . The Ae . aegypti mosquitoes for this study were collected outdoors with BG-sentinel mosquito traps and indoors with mosquito aspirators from three regions: Panama Centro ( Panama City , Felipillo ) , Panama Oeste ( Chorrera ) , and Chiriquí ( David ) . These sites were chosen on the basis of their prevalence of dengue fever and dengue hemorrhagic fever cases in the last 3 years and on mosquito surveys conducted by the Center for Mosquito Surveillance , Ministry of Health ( MINSA , from its Spanish acronym ) . Peridomestic collection of mosquitoes in selected areas was conducted in the early hours of the morning ( 5:30 to 6:30am ) and late afternoon ( 6:00 to 7:30pm ) . At least 10 mosquitoes per site were collected and processed . The collected mosquitoes were transported back to the laboratory , chilled on ice , and identified at the species level using a stereoscope and the taxonomic keys of Galindo and Adames [8] and Rueda [9] . Following species confirmation , mosquitoes were surfaced-sterilized by dipping and shaking them in 75% ethanol for 2 min and rinsing them with 1× PBS twice for 1 min each . Midguts were then dissected from each individual mosquito over a sterile glass slide containing a drop of 1× PBS , then transferred to a microcentrifuge tube containing 150 µl of sterile PBS and macerated for 30 sec . Three 10-fold serial dilutions were then plated on LB agar and kept at room temperature for 48 h . Initial isolation was based on morphology , color , and size of colony ( Figure S1 ) , and then followed by molecular identification via 16s rRNA gene sequencing . The primers used to amplify the 16s rRNA gene were those reported by Cirimotich et al [10] : forward , AGAGTTTGATCCTGGCTCAG; and reverse ( degenerate ) , TACGGYTACGCTTGTTACGACT . PCR conditions were used according to the Platinum Pfx DNA Polymerase ( Invitrogen ) protocol . PCR amplification was done with an initial denaturation of 2 minutes at 94°C , and 40 cycles with a denaturation step at 94°C for 30 seconds , an annealing step at 58°C for 30 seconds and an extension step at 72°C for 1 minute . Bacterial 16s rRNA gene sequences were manually curated and assembled from forward and reverse primer-generated sequences . Curated sequences were then aligned and compared to available bacterial sequences in GenBank and in the Ribosomal Database Project ( RDP Release 10 , http://rdp . cme . msu . edu/ ) . A bacterial phylogenetic tree was constructed using the Ribosomal Database Project “Tree Builder” program , which uses bootstrap sampling and the Weighbor weighted neighbor-joining tree-building algorithm to best estimate the phylogenetic position of a sequence . Mosquitoes were rendered free of cultivable bacteria ( designated as aseptic ) by maintaining them on a 10% sucrose solution with 20 units of penicillin and 20 µg of streptomycin from the first day post-eclosion until 2 days prior to challenge . They were then maintained for 1 day on sterile water and starved for 24 h prior to dengue virus infection . Effectiveness of the antimicrobial treatment was confirmed by colony forming unit ( CFU ) assays prior to blood-feeding or bacterial challenge . Two types of bacterial reintroduction were tested: via blood meal and via sugar meal . Reintroduction of bacteria through the blood meal was accomplished by first treating the mosquitoes with antibiotics and then providing them with cotton balls moistened with sterile water for 24 h post-antibiotic treatment . Mosquitoes were starved overnight and fed on a mixture containing 50% of a given bacterium suspended in 1× PBS ( final concentration: OD600 = 1 , for controls only 1× PBS was added ) , 25% of MEM ( devoid of any antibiotics ) , 25% human commercial blood , and 10% human serum . Mosquitoes were cold-anesthetized , and the fully fed mosquitoes were separated and provided with a dengue virus-infectious blood meal 4 days after bacterial reintroduction . Infection phenotype assays were performed as previously reported [11] and as described below . Following the bacterial reintroduction via blood meal , a subset of bacteria showing an effect on dengue virus infection was further tested through reintroduction via a sugar meal , which would more closely resemble natural bacterial acquisition . The bacteria were reintroduced through a sugar meal by first treating mosquitoes with antibiotics for the first 2–3 days after emergence and then providing them with a sterile 10% sugar meal for 24 h after antibiotic treatment . Mosquitoes were then starved overnight and fed on cotton strips moistened with a bacterial suspension diluted in 3% sucrose solution and suspended in a 1 . 5-ml microcentrifuge tube . Proteus sp . and Pantoea sp . were used at an OD600 of 1 . 00 . Bacterial concentrations used to infect mosquitoes were determined on the basis of the average bacterial load for each bacterial strain found in the midgut of field-collected mosquitoes . Initial assessment of sugar meal acquisition and the location of the sugar meal following ingestion were made by providing a group of mosquitoes with a sugar solution dyed with blue food colorant . Midguts and crops of exposed mosquitoes were dissected at 6 and 24 h . Dengue virus serotype 2 ( New Guinea C strain , DENV-2 ) was propagated in the C6/36 cell line according to standard conditions [11] . In brief , 0 . 5 ml of virus stock was used to infect a 75-cm2 flask of C6/36 cells at 80% confluence . Infection was allowed to proceed for 5–7 days , at which time the cells were harvested with a cell scraper and lysed by freezing and thawing in dry CO2 and a 37°C water bath , centrifuged at 800 g for 10 min , and mixed 1∶1 with commercial human blood . The infectious blood meal was maintained at 37°C for 30 min prior to feeding 5- to 7-day-old mosquitoes . Infected mosquitoes were collected at 7 days post-infection and surface-sterilized by dipping them in 70% ethanol for 1 min , then rinsing them twice in 1× PBS for 2 min each . Midgut dissection was performed in one drop of 1× PBS under sterile conditions , and the midgut was transferred to a microcentrifuge tube containing 150 µl of MEM . Midguts were homogenized using a Kontes pellet pestle motor and stored at −80°C until used for virus titration . Dengue virus titration of infected midguts was done as previously reported [11] , [12] . The infected midgut homogenates were serially diluted and inoculated into C6/36 cells in 24-well plates . After an incubation of 5 days at 32°C and 5% CO2 , the plates were fixed with 50%/50% methanol/acetone , and plaques were assayed by peroxidase immunostaining using mouse hyperimmune ascitic fluid specific for DENV-2 as the primary antibody and a goat anti-mouse HRP conjugate as the secondary antibody . Also , where indicated , dengue virus titration of infected midguts was conducted in BHK-21 cells . At 5 days post-infection , the 24-well plates were fixed and stained with crystal violet . Plaques ( formed by cells with cytopathic effect , CPE ) were counted and analyzed . Real-time PCR assays were conducted by first treating the RNA samples with Turbo DNase ( Ambion , Austin , Texas , United States ) ; they were then reverse-transcribed using M-MLV reverse transcriptase ( Promega , USA ) . The real-time PCR assays were performed using the SYBR Green PCR Master Mix kit ( Applied Biosystems , Foster City , California , USA ) in a 20-µl reaction volume , and all samples were tested in duplicate . The ribosomal protein S7 gene was used for normalization of the cDNA templates . The primer sequences used in these assays are listed in Table S1 . RNA interference assays ( RNAi-based gene silencing ) were conducted as previously reported [11] . In brief , 69 nl of dsRNA ( 3 ug/µl ) re-suspended in water was injected into the thorax of cold-anesthetized 3- to 4-day-old female mosquitoes using a nano-injector . Three days after injection and gene-silencing validation , the mosquitoes were allowed to feed on a dengue virus-laden blood meal . Dissection of midguts and virus titration were carried out as described above . The primer sequences used are listed in Table S2 . Real-time PCR assays were normalized and standardized according to Willems et al . [13] . Mann-Whitney U-tests and one-way ANOVA with Dunnett's post-test were used when appropriate . Statistical analyses were conducted using the GraphPad Prism statistical software package ( Prism 5 . 05; GraphPad Software , Inc . , San Diego , CA ) . Statistical significance is indicated with asterisks: * , p<0 . 05; ** , p<0 . 01; *** , p<0 . 001 .
To investigate the cultivable bacterial species composition of midguts from field-caught adult female Ae . aegypti , we conducted mosquito collections in dengue-endemic areas of Panama . The field-captured mosquitoes were surfaced-sterilized and dissected , and their midguts were homogenized and plated on rich culture medium . We isolated 40 distinct bacterial isolates on the basis of colony morphology and successfully characterized 34 of them . The bacteria isolated from the midguts of the field-collected mosquitoes were mostly Gram-negative , with no overrepresentation of a single genus ( Table 1 , Figure 1 ) . Six bacterial genera have been previously isolated from mosquitoes , Asaia spp . [14] , [15] , Aeromonas spp . , Enterobacter spp . [16] , Paenibacillus spp . [16] , Proteus spp . [17] , and Comamonas spp . [18] . The isolated bacteria belonged to six phylogenetic classes , with the most dominant being the Gammaproteobacteria , the Betaproteobacteria , the Bacilli , and the Alphaproteobacteria ( Figure 2A ) . To investigate whether certain bacteria isolated from field mosquitoes might influence dengue virus infection of the midgut , we conducted bacterial reintroduction assays through a blood meal or sugar meal ( Figure 2B and Figure 2C ) prior to dengue virus infection . Recolonization of mosquito midguts , previously rendered aseptic through antibiotic treatment , with single-isolate bacteria through a blood meal led to a marked decrease in viral titers in the midgut at 7 days post-bloodmeal ( PBM ) . Introduction of two bacteria species ( Proteus sp . Prpsp_P and Paenibacillus sp Pnsp_P ) separately into the mosquito midguts resulted in a significantly lower level of dengue virus infection , while introduction of other species ( among them Pantoea sp . Pasp_P and Comamonas sp . Cosp_P ) produced no significant difference in dengue virus titer from that of control group mosquitoes ( Figure 3A ) . Next we wanted to assess the impact of selected bacteria on dengue virus infection when introduced through a nectar meal , since this would be the most likely route of introduction in the field and exposure in a potential future symbiotic biocontrol strategy . The current perception is that the ingested nectar meal is stored in the mosquito crop and then relocated to the midgut for digestion [19] , [20] . To determine the location of the ingested sugar meal in the mosquito's digestive system , we exposed mosquitoes to a food color-dyed sugar meal . Following a 6-h exposure to the dyed-sugar meal , the blue sugar meal could be observed in the crop and midgut of some mosquitoes , while the remaining mosquitoes showed the presence of the sugar meal only in the midgut ( Figure S2 ) . At the end of a 24-h exposure , all mosquitoes were found to have food color-dyed sugar meal in both the midgut and crop . To assess the successful colonization of the mosquito midgut by the reintroduced bacteria , mosquito midguts were dissected , homogenized , and plated on LB agar at 3 days post-bacterial acquisition and prior to the time point at which dengue virus infection normally occurs . We observed a high prevalence of Proteus sp . Prsp_P ( 100% ) and a somewhat lower prevalence ( 69% ) of Pantoea sp . Pasp_P in the midgut of the mosquitoes ( Figure 2B and Figure 2C ) . Reintroduction of Proteus sp Prsp_P into the midgut through a sugar meal led to a significant decrease in dengue virus titers , but no significant effect on dengue virus infection was observed in mosquitoes colonized by Pantoea sp . Pasp_P ( Figures 3B ) . Reintroduction of isolated bacteria into the antibiotic-treated ( aseptic ) mosquitoes' midguts elicited changes in transcript abundace of a number of antimicrobial peptide genes , including cecropin , gambicin and attacin in the midgut ( Figure 4A ) and the abdominal fat body tissue ( Figure 4B ) . This result suggests that modulation of immune gene transcript abundance by the reintroduced bacteria could have a detrimental effect on dengue virus infection . Dengue virus infection of the mosquito's midgut led to significant decrease in the overall bacterial load ( as assessed by 16s rRNA transcript levels ) at 24 h , 7 days , and 14 days after ingestion of a dengue virus-supplemented blood meal . Interestingly , the difference in the bacterial 16s rRNA transcript levels between dengue virus-infected and uninfected mosquitoes was less prominent at 3 days post-infection ( Figure 5A ) . Analysis of the relative transcript abundance of the antimicrobial peptide genes lysozyme C , and cecropin G revealed that cecropin G transcripts were significantly elevated in dengue-infected mosquitoes at 7 days post-infection but showed no difference from control levels at 10 days post-infection . Lysozyme C also showed a transient changes in transcript abundace , with no difference from control levels at 7 days but significant changes at 10 days post-infection ( Figure 5B ) . To assess the involvement of antimicrobial effector genes in regulating the midgut microbiota , we employed an RNAi-based gene silencing approach in conjunction with CFU assays . Although not statistically significant , silencing of several effector genes led to changes in the growth of the midgut bacterial populations compared to the control group ( GFP dsRNA-injected mosquitoes ) ( Figure 6 ) . This suggests that one function of these immune factors is to maintain a basal level of immunity to control microbial proliferation . Interestingly , we did not observe a significant increase in the midgut bacterial load after silencing the cecropin G and lysozyme C genes , suggesting that these factors may play more specialized roles in immunity ( Figure 7A and 7B ) . We used a RNAi-based gene silencing approach to assess the effect of selected antimicrobial peptide genes on dengue virus infection , some of which are known to be regulated by our field-derived bacteria . This treatment led to an overall increase in dengue virus titers in the mosquito midgut especially for lysozyme C , suggesting that this gene might exert a significant inhibitory effect , on dengue virus infectivity ( Figure 8A ) . However , this effect was lost when the mosquitoes were maintained aseptically with antibiotics prior to receiving an infectious blood meal ( Figure 8B ) . This might indicate that the infection phenotype observed upon lysozyme C–silencing reflects an indirect effect . It is possible that lysozyme inhibit the growth of bacteria that are beneficial to the virus , or , alternatively lysozyme may act against bacteria that compete with other bacteria that have a detrimental effect on the virus . The current analysis does not allow for a detailed mechanistic insight on this .
During their life span , insects harbor a variety of microbes in their intestine , some of which are needed for successful growth to adulthood , and some as aids in digestion , nutrition , and reproduction [21] as well as protection against pathogens [10] , [22]–[25] . This situation is especially true for mosquitoes that , as larvae , develop in stagnant microbe-rich water , feeding on various bacteria and fungi , and that as adults are exposed to microbes , parasites , and viruses through plant nectars and ingested blood . For example , it has been shown that antibiotic treated aseptic Anopheles gambiae mosquitoes are more susceptible to Plasmodium infection and possess a lower basal level of immune gene transcripts than do An . gambiae with a normal microbial population [26] , [27] . The basal level of immune activity appears to be critical in defining the level of susceptibility to Plasmodium infection [28] . With regard to virus-mosquito interactions , the intracellular bacterium Wolbachia spp . has been shown in several studies to affect dengue virus infection in Ae . aegypti mosquitoes [29] and infection with the Japanese encephalitis virus in aseptic Culex bitaeniorhyncus [30] . We have previously shown that mosquitoes with a reduced midgut bacterial load ( as a result of antibiotic treatment ) can support higher dengue infection levels than can septic mosquitoes [11] . Furthermore , the antibiotic-treated aseptic mosquitoes display a lower basal level of several Toll pathway-related genes transcripts . We have shown that the Toll pathway is involved in the anti-dengue defense [11] . We cannot , however , exclude other possible mechanisms by which the bacteria may hinder virus infection in the mosquito . In order to assess this phenomenon in greater detail and select bacteria that can mediate potent anti-dengue activity and meet other criteria ( easily cultivable and major representation in the midgut microbiome ) for the development of dengue biocontrol strategies , we have now isolated and characterized cultivable bacteria from the midguts of field mosquitoes collected in dengue-endemic areas of Panama . Bacterial isolates from field collections belonged to several phylogenetic classes , but no predominant genus was observed . Many of these bacterial species have been previously isolated from mosquitoes and may be better adapted to the mosquito midgut environment . The diversity of microbes isolated from field mosquitoes suggests a complex mosquito midgut microbiome that is likely to affect the outcome of infection and the mosquito's midgut immune homeostasis . Our midgut bacteria discovery method identified only live , replicating bacteria that could grow aerobically on a rich culture medium , and this approach likely explains some of the discrepancies between our results and those of studies that have employed PCR-based amplification of bacterial DNA , much of which may have been derived from dead , minor , and/or transient microbial constituents of the midgut microflora [15] , [31] . Reintroduction of some of these bacterial species through a blood meal led to changes in susceptibility of the midgut tissue to dengue infection . Furthermore , reintroduction of bacterial isolates via a sugar meal into the midgut of Ae . aegypti mosquitoes resulted in a significant decrease in dengue virus infection in the case of one bacterial isolate , Proteus sp . Prsp_P . These bacteria may either indirectly exert an anti-dengue effect by boosting basal immunity or may directly influence the virus' infectivity . The bacteria could , for example , act prior to dengue virus infection of the midgut via bacterial metabolites that are detrimental to the dengue virus , or act as a barrier for the virus via steric hindrance , by growing along the midgut epithelium [15] . In contrast to the effects produced by Proteus sp . Prsp_P , reintroduction of Pantoea sp . Pasp_P had no effect on dengue virus infection , perhaps because of the inability of this bacterium to effectively colonize the mosquito's midgut . This could partially offset the anti-dengue effects that derive from the elicitation of the mosquito's immune system by this bacterium . Alternatively , although Pasp_P shows a slightly higher immune induction than Prsp_P , our gene expression assays only addressed one time point of amp transcript abundance , and it is quite likely that Prsp_P may elicit an overall stronger induction of these genes over an extended time period . It is also possible that Prsp_P induces some other unknown anti-viral factor stronger than Pasp_P . Furthermore , given that our introduction of bacteria was performed with a single bacterial species at a time , it is possible that lack of effect on dengue virus infection was because this bacterium needs to act in synergy with other microbes of the midgut . This type of synergistic effects may also alter some of our observed ant-dengue activities for the other studied bacteria , when combined with multiple bacterial species . Our analyses of immune gene expression in mosquitoes exposed to the studied bacteria revealed responses that were similar in their direction of regulation but different in their magnitude . We observed elevated immune gene transcripts in both the midgut and fat body tissues , thus pointing to a local as well as a systemic immune response . These two compartment-specific responses could act in concert to limit dengue virus infection and dissemination in the mosquito host . The transcript abundance of the antimicrobial peptides we assayed has been shown to be regulated by the immune signaling pathways that govern the defense against dengue virus infection [11] , [32] , [33] . Thus , it is possible that mosquito immune responseselicited by the bacteria play a significant role in reducing the level of dengue infection in the mosquito midgut . In fact , recently , a cecropin-like peptide with anti-dengue virus properties was found to be elicited in the salivary gland of dengue virus-infected mosquitoes [34] and cecropin-D and defensin-C peptides have been shown to have anti-dengue activity in the mosquito midgut [35] . The mosquito can be considered a holobiont unit , in which the mosquito , its midgut microflora , and the dengue virus are involved in complex reciprocal tripartite interactions . Our analysis of these interactions has indicated that dengue infection in the mosquito is able to elicit an immune response involving the elevated transcript abundance of antimicrobial peptide genes such as cecropin , attacin , and lysozyme C [11] , [32] , [33] . Even though the antiviral activity of the mosquito's antimicrobial peptides have yet to be characterized , a cecropin-like peptide was recently found to have anti-dengue virus activity [34] . In addition , antimicrobial peptides are effective in controlling bacteria [36]–[38] , and their elicitation by dengue virus infection can therefore modulate the mosquito's midgut microflora . Our observations agree with this assertion , in that dengue virus-infected mosquito midguts displayed a lower bacterial load ( as measured by 16s rRNA ) than did those of uninfected mosquitoes . In summary , our analysis of the reciprocal interactions between the dengue virus , mosquito immune system , and bacteria isolated from midguts of field mosquitoes collected in Panama has revealed a marked decrease in viral load in mosquitoes infected with certain natural bacterial isolates . Transcript abundance analysis of selected antimicrobial peptide genes suggested that the mosquito's microbiota elicits an immune response that appears to act in part to control dengue infection . In turn , the activation of the immune system by dengue virus infection potentiates the mosquito's immune homeostasis and suppresses the microbiota of its midgut . A better understanding of these complex reciprocal interactions may facilitate the development of novel biocontrol strategies for dengue transmission .
|
Dengue virus is transmitted by Aedes mosquitoes . During their lifecycle , mosquitoes are exposed to a variety of microbes , and many of them inhabit the mosquito midgut , thereby sharing the same environment with ingested pathogens . The mosquito midgut is the site of multiple reciprocal interactions between the mosquito , its commensal bacteria , and ingested pathogens that will ultimately influence the level of pathogen infection and transmission . In this study the authors addressed the reciprocal interactions between the Aedes immune system , dengue virus and mosquito midgut microbiota using molecular and microbiological assays . The study showed that certain field-derived bacterial isolates of the mosquito midgut exert a detrimental effect on dengue virus infection . This effect is at least partly manifested through the action of the mosquito immune system which is activated by microbes . Conversely , dengue virus infection induces immune responses in the mosquito midgut tissue that act against the natural mosquito midgut microbiota . This study contributes to our understanding of dengue virus infection in Aedes mosquitoes , which may aid towards the development of novel biocontrol strategies to halt dengue transmission .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"immunity",
"vector",
"biology",
"immunology",
"biology",
"microbiology"
] |
2012
|
Reciprocal Tripartite Interactions between the Aedes aegypti Midgut Microbiota, Innate Immune System and Dengue Virus Influences Vector Competence
|
Age-related decline in the integrity of mitochondria is an important contributor to the human ageing process . In a number of ageing stem cell populations , this decline in mitochondrial function is due to clonal expansion of individual mitochondrial DNA ( mtDNA ) point mutations within single cells . However the dynamics of this process and when these mtDNA mutations occur initially are poorly understood . Using human colorectal epithelium as an exemplar tissue with a well-defined stem cell population , we analysed samples from 207 healthy participants aged 17–78 years using a combination of techniques ( Random Mutation Capture , Next Generation Sequencing and mitochondrial enzyme histochemistry ) , and show that: 1 ) non-pathogenic mtDNA mutations are present from early embryogenesis or may be transmitted through the germline , whereas pathogenic mtDNA mutations are detected in the somatic cells , providing evidence for purifying selection in humans , 2 ) pathogenic mtDNA mutations are present from early adulthood ( <20 years of age ) , at both low levels and as clonal expansions , 3 ) low level mtDNA mutation frequency does not change significantly with age , suggesting that mtDNA mutation rate does not increase significantly with age , and 4 ) clonally expanded mtDNA mutations increase dramatically with age . These data confirm that clonal expansion of mtDNA mutations , some of which are generated very early in life , is the major driving force behind the mitochondrial dysfunction associated with ageing of the human colorectal epithelium .
Mutations of mitochondrial DNA ( mtDNA ) have been implicated in the ageing process [1] . As humans age , multiple different mutations arise somatically in individual cells and some of these expand clonally to high levels over time , resulting in focal respiratory chain deficiencies [2]–[6] . To date there is poor understanding of the dynamics of mutation accumulation during ageing and of when in the life-course the majority of these clonally expanded somatic mtDNA mutations initially occur [7] . For example , there have been suggestions that the underlying mechanism involves either an accelerating mtDNA mutation rate over time [8] , or clonal expansion of mtDNA mutations which have occurred in early life [7] . Resolving these questions about the dynamics of mtDNA point mutations is important because of their accumulation in human stem cell populations with age , which results in respiratory chain dysfunction [5] , [9]–[14] and consequent reductions in cell function [15] . Age-related dysfunction of somatic stem cells has been proposed to lead to the decreased ability of tissues to regenerate [16] . Mice with increased mtDNA mutagenesis ( mutator mice ) caused by a defect in the proof-reading ability of the mitochondrial DNA polymerase gamma also show a premature ageing phenotype [17] , [18] which has been attributed largely to somatic stem cell dysfunction resulting from high mtDNA point mutation loads [19]–[22] . In mutator mice the majority of the mutational burden leading to a cellular phenotype occurs during embryogenesis [19] . However , in human tissues there have been no comprehensive studies examining mtDNA point mutation occurrence and accumulation over the life-course . Here , we apply different validated techniques to investigate the frequency of both low level ( as an indirect measure of mutation rate ) and clonally expanded mtDNA mutations , using the human colonic epithelium as an exemplar tissue with a well-characterised stem cell population . We found no evidence of a significant increase in the frequency of low level mtDNA mutations with age , but there was a significant increase in the frequency of clonally expanded mtDNA mutations with age . We provide robust evidence that mtDNA mutations occur early in life and that a substantial mtDNA point mutation burden exists within the human colorectal epithelium before the age of 20 .
Colorectal biopsies from 207 subjects aged 17–78 years with no evidence of bowel pathology at endoscopy were collected . Low level mtDNA mutation frequencies were quantified in these biopsy samples using a highly sensitive Random Mutation Capture ( RMC ) assay [23] , [24] ( Figure 1A ) . Approximately 200 million base pairs of mtDNA sequence were screened and a total of 803 mutations were detected . All mutations and full details of the number of bases investigated per individual are shown in Table S1 . Examination of the types of mutational events detected by RMC in our cohort showed that 60% of all mtDNA mutations were G>A or C>T transitions , 24% were T>C or A>G transitions , and the remainder were transversions and small insertions and deletions . We also noted an uneven distribution of mutations across the four base pair TCGA TaqIα site; 63% of changes were at the third base pair , with the remaining 37% spread fairly evenly across the other 3 bases . There was no significant correlation between low level mtDNA mutation frequency and age ( Pearson correlation = 0 . 127 ( p = 0 . 07 , Figures 1B and 1C ) ) . The RMC assay can be intrinsically noisy due to random sampling statistics [25] , with inter-individual variation often observed in studies of ageing populations [26] . Therefore to maximise our chance of detecting a relationship between low level mtDNA mutation frequency and age , we pooled the data by decade of participant age . Although there was a modest increase in low-level mutation frequency with age ( Figure 1D ) , this was not statistically significant ( p = 0 . 343 , One Way ANOVA ) . A Tukey post-hoc comparison revealed no significant differences even for the comparison between the first and the last decade of participant age . In a number of subjects we were not able to detect any mtDNA mutations in the base pairs screened . To ensure that these zero values were not having a significant effect on the data we re-ran the analyses excluding the zero values . There was still no significant association with age in either the individual data points ( p = 0 . 07 , Pearson correlation = 0 . 136 ) or the grouped data ( p = 0 . 46 , One Way ANOVA ) ( Figure S1 ) . These data highlight that even the youngest person studied ( aged 17 years ) had an appreciable mtDNA mutation frequency of 2 . 5 mutations per 106 base pairs . Collectively these analyses demonstrate no significant increase in low level mtDNA mutation frequency with age . There was no significant difference in the frequency of synonymous and non-synonymous mtDNA mutations in older ( >46 years ) vs younger participants ( <46 years ) ( p = 0 . 665 , Fisher's exact test ) . This confirms that there are no selective pressures acting on mtDNA point mutation occurrence with age and that mtDNA mutations present from early adulthood in the human colon could be pathogenic later in life if they were to clonally expand to high levels over time . Due to the small size of the region of the genome under investigation by the RMC assay , and the relative rarity of clonally expanded mutations across the entire mtDNA molecule , the RMC will not detect most of the clonally expanded mtDNA mutations . Indeed , in our RMC mutation dataset there were no mutations expanded to more than the equivalent of 1/100 of a crypt size . Therefore , we employed a next generation sequencing ( NGS ) approach to examine all mtDNA sites and gain information about the age dependent dynamics of clonally expanded mutations . Whole mtDNA Ion Torrent NGS was carried out on DNA extracted from the biopsies from a representative subset of the youngest ( <26 years of age n = 8 ) and oldest participants ( >70 years of age , n = 8 ) investigated by RMC . Our stringent quality control criteria ( see Materials and Methods ) set the heteroplasmy threshold for calling mtDNA mutations at 0 . 8% . Based on this figure and the average number of crypts per biopsy ( ∼200 ) , our NGS assay could detect homoplasmic clonal expansions within individual crypts or clusters of clonal crypts [14] , as well as low levels of heteroplasmic mtDNA mutations ( >0 . 8% ) present throughout the whole tissue [27] , [28] . Figure 2 details the various techniques employed throughout the study and the limits of detection for each technique . We sequenced a total of 556 Mb of mtDNA ( average of 35 Mb per subject ) and detected 109 mutations present at >0 . 8% heteroplasmy . All detected mtDNA variants are detailed in Table S2 . All participants showed some variants at >0 . 8% heteroplasmy , but the frequency of mtDNA mutations was more than 8-fold higher in the older than in the younger group ( Figure 3A , p = 0 . 036 , unpaired t-test ) . There was no significant difference in the types ( transitions/transversions ) of mutations observed between the two age groups ( Figure 3B , p = 1 . 00 , Fisher's exact test ) , with single nucleotide transitions being by far the major mutation type . A recent study has shown that recurrent tissue-specific mtDNA mutations are present in unrelated individuals [29] . We investigated our NGS data to determine whether recurrent mtDNA mutations were also present within our dataset . In our colon data we detected eight unique mtDNA mutations that were present in two individuals; however unlike the data from Samuels et al , they were not restricted to the non-coding regions , but appeared to be located randomly; three were in the non-coding region , three were in protein encoding genes and two were in RNA genes . Five of these were previously reported polymorphic variants [30] , the remaining three were previously unreported . There is no evidence of contamination as they do not fit a haplogroup specific pattern with multiple markers seen in individual subjects [31] , they were present in random pairs of subjects , and they were not observed in the yeast plasmid control . It has recently been shown that mtDNA mutations in adult tissues can originate in embryonic development or even in the germline [27] , [28] , inferring that some of the mutations that we detected by NGS could have occurred during this period . To investigate this possibility , matched buccal scrape samples were collected at the same time as the colorectal mucosal biopsies from the same 16 participants from whom we had carried out NGS ( data in Figure 3 ) , and NGS was performed on DNA from the buccal cells . Identification of the same mtDNA mutations in two different tissues would support the hypothesis that such mutations occurred prior to tissue differentiation during embryogenesis . It should be noted that buccal and colonic epithelial cells both arise from the endoderm with the fore and hind gut becoming separate tissues by weeks 3–4 of gestation [32] . In DNA from the 16 participants investigated , we detected a total of 16 mtDNA mutations that were present at low levels of heteroplasmy in both tissues ( Table S2 and Table S3 ) and these occurred in 10 participants . Five of these people were in the >70 year age group , five were in the <26 year age group . There was no significant difference in the frequency of germline or embryological mtDNA mutations between the <26 year and >70 year age group ( p = 0 . 176 , unpaired t-test ) confirming that there was no age-effect and that these mtDNA mutations were most likely of germline or embryological origin . The frequency of somatic mtDNA mutations in the colon samples was then analysed by subtracting the germline or embryological mtDNA mutations from the total mtDNA mutation frequency . This revealed a significant 10-fold increase in the frequency of clonally expanded somatic mtDNA mutations in those aged >70 years compared with <26 years ( Figure 4A , p = 0 . 035 , unpaired t-test ) . From here on the mtDNA mutations detected in colorectal epithelium only will be referred to as somatic mtDNA mutations and those present in both buccal and colorectal epithelium as germline or early embryological mtDNA mutations . The pattern of somatic mtDNA mutations detected in the buccal epithelium was similar to those in the colonic epithelium . The mtDNA mutations detected were base transitions and were randomly located throughout the genome . We did observe a higher number of somatic mtDNA mutations in the colonic epithelial samples compared to the buccal samples . We have previously shown that there are tissue specific differences in the frequency of clonally expanded mtDNA mutations , with the colon being one of the most highly affected [10] and believe that this could explain these differences . Next we compared the ratio of synonymous or polymorphic protein encoding mtDNA mutations to non-synonymous mtDNA mutations in the somatic and germline or early embryological data sets to see if there were any differences between the two , and therefore any evidence for purifying selection . There was a significantly higher proportion of non-synonymous mtDNA mutations in the somatic data set compared with the germline or early embryological data set ( Fisher's exact test with , p = 0 . 041 , Figure 4B ) ; in fact only one of the germline or early embryological mtDNA mutations was non-synonymous and therefore potentially pathogenic . These data suggest that the mtDNA mutations likely to contribute to the mitochondrial ageing phenotype begin to occur sometime after 3–4 weeks gestation ( 1–2 weeks post-conception ) , which coincides with the resumption of mtDNA replication which is thought to occur post-embryo implantation [33] . We have previously shown that the frequency of crypts deficient in cytochrome c oxidase activity ( complex IV of the respiratory chain ) increases with age in the apparently normal mucosa taken from patients with a colorectal tumour [5] . We also demonstrated that in the vast majority of these crypts there is an intracellular clonally expanded mtDNA point mutation [5] , [34] . The present study provided an opportunity to determine whether there was a similar age-related increase in the frequency of crypts deficient in cytochrome c oxidase activity in healthy participants in whom there was no evidence of mucosal dysplasia . In addition , as cytochrome c oxidase ( COX ) /succinate dehydrogenase ( SDH ) histochemistry is an excellent surrogate marker for mid-high level intracellular clonally expanded mtDNA point mutations , which both RMC and NGS are quite likely to miss , this assay gives an indication of the frequency of such mutations ( Figure 2 ) . Colorectal mucosal biopsies collected by endoscopy from the same 207 subjects investigated by the RMC assay were subjected to sequential COX/SDH histochemistry ( Figure 5A ) and the percentage of COX deficient colonic crypts calculated . As expected , there was a significant increase in the percentage of COX deficient crypts in individuals with age ( Figure 5B , Pearson correlation 0 . 603 ( p<0 . 001 ) ) . The somatic mtDNA mutations detected using the Ion Torrent NGS platform could be clonally expanded mtDNA mutations in individual colonic crypts or low level clonally expanded mtDNA mutations present throughout the whole tissue . Therefore we investigated a possible correlation between the percentage of COX deficient crypts ( known to be a good marker of clonally expanded mtDNA point mutations [5] ) and the somatic mtDNA mutation frequency measured by NGS . This showed that there was a significant correlation between mtDNA mutation frequency as measured by NGS and COX deficient crypts ( Figure 5C , Pearson correlation = 0 . 511 ( p = 0 . 043 ) ) . When we compared the germline mtDNA mutation frequency with the percentage of COX deficient crypts , there was no significant correlation ( Figure 5D , Pearson correlation = 0 . 369 , p = 0 . 176 ) . There was no significant correlation between the mtDNA mutation frequency detected by RMC and COX deficiency ( Figure 5E , Pearson correlation = 0 . 007 , p = 0 . 918 ) , suggesting that the majority of pathogenic mtDNA mutations detected by NGS are somatic clonally expanded variants . In addition when we compared the RMC and NGS data from the same subjects side by side , there was no significant correlation confirming that the two assays were measuring different classes of mtDNA mutations i . e . low level vs clonally expanded ( Figure 5F , Pearson correlation = 0 . 381 , p = 0 . 145 , Figure 2 ) .
In this study we have examined the timing of occurrence and frequency of mtDNA mutations during ageing in human colorectal epithelium . We have employed a range of methodologies to provide accurate assessment of low level mtDNA mutation frequency , germline mtDNA heteroplasmy , and both high and low level intracellular clonal expansion . For low level mtDNA mutation frequency , we have shown previously that the most accurate method to use is the RMC assay [25] , which provides the best available measure of mtDNA mutation rate [23] . An advantage of this assay is that it is not affected by false positive mutational calls caused by either PCR or sequencing errors; however a limitation is that , because only a small mtDNA domain is interrogated , it will miss the majority if not all intracellular clonally expanded mtDNA mutations . The NGS approach we have used is , in our hands , sensitive down to 0 . 8% heteroplasmy which correlates to either low-level germline mtDNA heteroplasmy or homoplasmic intracellular mtDNA clonal expansions in one or more crypts [14] ( based on there being ∼200 crypts in a colonoscopic biopsy sample ) . Sequencing of individual laser micro-dissected crypts , which we have done extensively in our previous work [5] , [34] , is required to detect the remaining mtDNA mutations , i . e . mid-high level intracellular clonal expansions; these can also be detected using COX/SDH histochemistry as a surrogate marker [5] , [14] , [34] , [35] as we have done here ( Figure 2 ) . There is the possibility that there are effectors of COX deficiency other than clonally expanded mtDNA point mutations , such as nuclear DNA mutations or changes in global gene expression; however we have previously found evidence of a clonally expanded mtDNA point mutation in >75% of crypts sequenced [5] , [14] , [34] , [35] and therefore believe that this is an excellent surrogate marker . We have not previously detected any large-scale mtDNA deletions in individual COX deficient crypts [5] . The RMC assay revealed that there was no significant increase in low level mtDNA mutation frequency with age . This is supported by a published study which used RMC to measure mtDNA mutation frequency in colonic epithelium from a much smaller cohort of individuals covering a narrower age range ( ∼50–90 years of age ( n = 20 ) ) [23] . An important result of our study is a demonstration that the lack of an increase in low level RMC-detected mtDNA mutations ( i . e . no increase in mtDNA mutation rate ) does not mean that there is no increase in total mtDNA mutation load with age . Instead , we show that clonally expanded mutations , as measured by NGS , increase very dramatically . This is the first demonstration that mtDNA mutation rate and clonal expansion may follow very different age dynamics . This is in direct contradiction of the mitochondrial vicious cycle hypothesis of ageing [36] which suggests that mtDNA mutations occur during ageing leading to dysfunctional proteins in the oxidative phosphorylation system , precipitating increased mutation i . e . an accelerating mtDNA mutation rate over time . Our results also demonstrate that caution should be exercised in interpreting mutation analysis results , which may be limited to only a portion of mutations depending on the technique used . Together the NGS and RMC datasets suggest that mtDNA mutation rate does not change significantly with age , but that clonal expansion of mtDNA mutations occurs over time . Mathematical modelling studies have suggested that clonal expansion of mtDNA mutations within an individual cell is likely to be due to random genetic drift and predict that it can take at least 20 years for an mtDNA mutation to clonally expand to high levels sufficient to cause COX deficiency [37]–[39] . Indeed , the youngest participant in this cohort in whom COX deficiency was detected was 21 years of age , therefore the initial mutational event ( s ) in this case must have occurred very early in life . Interestingly , the mtDNA mutation frequency data by NGS from our human samples were different to those obtained from a similar NGS study carried out in mice in which no significant increase in mtDNA mutation frequency with age was noted [40] . Previously we examined colonic epithelial tissue from a similar ageing mouse colony and showed that clonal expansion of mtDNA mutations was a very rare event in these animals compared with aged humans [41] . This may explain the species differences in these data , consistent with modelling studies that emphasise the difficulty of generating clonal expansion through random drift in short-lived animals [39] . Whilst our data imply that mutations of very early origin contribute to mitochondrial dysfunction in old age , it does not mean that mutations occurring in adult life play no role . In fact , the number of different clonally expanded mutations per sample detected by NGS robustly increased with age ( 3-fold from <26 years to >70 years group , p = 0 . 001 ) . Such an increase in mtDNA point mutation diversity can be explained with a scenario whereby de novo mtDNA point mutations occurring during adult life , perhaps up to middle-age , are able to clonally expand and join the set of expanded mutations detected in old age . However mtDNA mutations which occur late in life will not have time to expand to high levels . Analysis of the RMC data showed that even in the youngest participants , we observed a substantial load of mtDNA mutations in the colonic epithelium . NGS analysis of a subset of our youngest participants , all of whom were <26 years of age ( n = 8 ) , confirmed the RMC data; young adults have a significant mtDNA point mutation load . This has been previously shown to be the case in DNA extracted from young brain [42] , [43] , where both point mutations and mtDNA deletions have been detected . Our RMC data have now shown that this is also the case for a mitotic tissue , the colonic epithelium . In addition we have clearly demonstrated that the same type of mtDNA mutations ( point mutations which are predominantly transitions ) are present in young individuals as those detected in our previous studies of clonally expanded mtDNA mutations from aged respiratory chain deficient individual crypts [34] , i . e . the seed mutations for clonal expansion can be laid down at an early age . Although this has been predicted by modelling simulations [37] , [38] this is the first experimental evidence to show this definitively . Our NGS analysis has shown that low level heteroplasmic mutations are present in multiple tissues from the same individual . This supports previous studies showing that mtDNA mutations in adult tissues can originate in germline or very early development [27] , [28] . Indeed , our data from the colonic and buccal epithelium show that mtDNA mutations present in both tissues must have occurred prior to the fore and hind guts becoming separate which is thought to occur 1–2 weeks post-conception [32] . It is possible that due to our conservative cut off of 0 . 8% heteroplasmy , there may be additional germline mtDNA mutations which have undergone less drastic clonal expansion in one of the two tissues studied than the other , and therefore are below the threshold of detection; this is one of the limitations of the available technology . These data do show that 95% of the heteroplasmic mutations detected in both tissues were non-pathogenic polymorphic variants , thus suggesting that pathogenic mtDNA mutations which occur in the germline or early development are selected against , and these non-pathogenic mtDNA mutations may make little contribution to the ageing phenotype . This demonstrates purifying selection in the human germline . Previous studies have shown this in mice by looking at transmission of mtDNA mutations through multiple generations [44] , [45]; here we show purifying selection in humans by comparing the germline and somatic mtDNA mutations in different tissues from the same subjects . Recent evidence from the mouse has suggested that transmitted germline mtDNA mutations can cause premature ageing , perhaps by clonal expansion of these germline mtDNA mutations over time [46] . In our dataset 95% of the germline mtDNA mutations are benign and are unlikely to cause mitochondrial dysfunction and premature ageing , suggesting that there are differences in the dynamics of mtDNA transmission between these mutation prone mice and humans . The somatic mtDNA mutations detected by NGS in the colonic epithelium only , are a combination of benign synonymous and polymorphic variants and non-synonymous potentially pathogenic variants , which we believe may begin to occur when mtDNA replication is re-initiated after the embryo has implanted into the uterine wall [33] . There was a significantly higher frequency of non-synonymous coding region somatic mtDNA mutations compared to the germline or early embryological mtDNA mutations which is evidence in support of the hypothesis that the somatic mutations occurred beyond any selective checkpoints , before expanding clonally to detectable levels . The observations in this study are in agreement with evidence from epidemiological studies which suggest that damage arising early in human life can be an important modulator of outcomes in later life [47] , [48] . Due to the time taken for clonal expansion of mtDNA mutations to occur in human cells , we hypothesise that late life de novo mtDNA mutational events make negligible contribution to the ageing phenotype and that early to mid-life mtDNA mutations are likely to be much more important .
Colorectal mucosal samples were collected from the same anatomical site ( 10 cm from the anal verge ) from participants ( n = 207 , age range 17–78 years ) undergoing colonoscopy for disturbed bowel function in whom no evidence of bowel disease was identified ( BORICC 1 Study ) . Buccal epithelial scrapes were also collected concurrently from these subjects . The following subjects were also used in our previous work: BCC010 , BCC011 , BCC017 , BCC022 , BCC028 , BCC085 , BCC087 , BCC088 [25] . Ethical approval was obtained from the Northumbria NHS Trust Local Research Ethics Committee ( Project reference NLREC2/2001 ) . All participants were fully informed and written consent obtained from them . RMC was carried out essentially as previously described [25] . Briefly , mtDNA was extracted from colorectal mucosal biopsies and drop-dialysed using membrane filters ( 0 . 025 µm , Millipore ) to extract any excess salts . One microlitre of mtDNA was digested with 100 U of TaqIα ( New England Biolabs ) for 10 hours with the addition of 100 U every hour . MtDNA copy number was quantified by SYBR Green real-time PCR ( Roche ) targeting a template outside of a TaqIα restriction site in MTND5 ( primers L12473–L12492 and H12573-H12554 ) Absolute quantification was carried out using the standard curve method . PCR was then carried out across a TaqIα restriction site within the MTCOI gene ( bp 6562–6565 , primers L4636–L6455 and H6851–H6870 ) . An average of 2500 copies of the target sequence ( a total of 10000 target bases ) was added to each PCR reaction . Following PCR each product was digested with 50 units of TaqIα for 1 hour at 65°C , followed by 10 minutes at 80°C to inactivate the enzyme . Products were then subjected to electrophoresis through a 1 . 5% agarose gel for 1 hour at 200 V . All full length ( 488 bp ) products were excised from the gel using a QIAquick Gel Extraction kit ( Qiagen ) . These products were sequenced using ABI BigDye chemistries per standard manufacturer's protocols and analysed on an ABI3100 Genetic Analyser ( Applied Biosystems ) . Sequences obtained were compared with the revised Cambridge Reference Sequence ( GenBank accession number: NC_012920 . 1 ) using SeqScape software ( Applied Biosystems ) . Mutation load was calculated by dividing the number of confirmed mutants by the total number of base pairs investigated . To investigate the sensitivity and specificity of the RMC assay in our hands we generated a PCR product which contained a mtDNA mutation in the RMC site and one which was wild-type in the RMC site . A pCR-scriptTM Amp SK ( + ) cloning Kit ( Stratagene ) was used to clone the products following the manufacturer's protocol . Recombinant plasmids were identified by blue–white colour selection and the cultures grown up using a Qiaquick miniprep kit ( Qiagen ) . The DNA was then extracted and quantified and the wild-type and mutant PCR products mixed at concentrations ranging from 100% wild-type to 100% mutant . The RMC assay was then carried out as above and the observed mutant fractions calculated and compared to the expected fractions ( Table S4 ) . There was no difference between observed and expected fractions , confirming the RMC assay to be both highly sensitive and specific . Next generation sequencing ( NGS ) was carried out using an Ion Torrent Personal Genome Machine ( Life Technologies , Paisley , UK ) on whole mtDNA from the same colonic biopsies investigated by RMC and from buccal epithelia from the 8 youngest ( <26 years ) and 8 oldest ( >70 years ) subjects . To exclude the possibility of nuclear pseudogene amplification , extracted DNA was amplified in two overlapping 9 kb fragments using primers L2091–L2111 and H10649-H10629 ( primer set 1 ) , and L10085–L10104 and H2644-2625 ( primer set 2 ) , the specificity of which was established after observing no amplification from Rho Zero cells , cells depleted of their mtDNA by ethidium bromide treatment . Long-range PCR amplicons were quantified on an Agilent 2100 Bioanalyzer with an Agilent DNA 12 , 000 kit ( Agilent Technologies , Stockport , UK ) . Overlapping PCR fragments for each sample were combined in equimolar concentrations . Pooled amplicons ( 100 ng ) were then fragmented , barcoded , size-selected and amplified using the IonXpress Plus Fragment Library kit , Ion Xpress Barcode Adapters and E-Gel SizeSelect 2% agarose gels ( Life Technologies ) , according to the manufacturer's recommendations . Barcoded libraries were quantified with an Agilent Bioanalyzer DNA High Sensitivity kit then pooled ( n = 16 ) in equimolar concentrations and diluted to 26 pM , prior to clonal amplification onto Ion Sphere Particles using the Ion OneTouch 1 System and the Ion OneTouch 200 Template kit v2 ( Life Technologies ) , as per the manufacturer's instructions . Coated spheres were enriched on the Ion Torrent ES ( Life Technologies ) before loading onto Ion 318 sequencing chips ( Life Technologies ) . Next-generation semiconductor sequencing was performed on an Ion Torrent Personal Genome Machine ( Life Technologies ) . Fastq data files downloaded from the Torrent Server ( version 3 . 6 . 2 , Life Technologies ) were analysed using NextGENe software ( v2 . 3 . 0; SoftGenetics , State College , PA , USA ) . The background noise on the Ion Torrent PGM platform was quantified by extraction of DNA from a yeast clone which had been transfected with a plasmid ( pRShmt ) containing the entire human mtDNA [49] ( kindly donated by Dr Brian Bigger ( University of Manchester , UK ) ) which was then subjected to an identical PCR amplification and NGS protocol as the colonic biopsy and buccal scrape samples . On this basis any low level mtDNA variants detected in the cloned mtDNA are likely to be technical artefacts arising from the PCR and sequencing process and we could quantify the level of background noise and exclude this from the sample analysis . In addition the Ion Torrent platform may be prone to base calling errors in polynucleotide tracts , most often calling them as small insertions or deletions; therefore we restricted the analysis to base-pair substitutions only . Further quality control steps taken were; ( 1 ) only base substitutions with a quality score >20 were included in order to be confident that the calls were genuine , ( 2 ) observed variants had to be present in both forward and reverse reads at comparable frequencies with a 3-fold difference permitted to allow for the effects of a binomial sampling distribution at very low variant levels [28] , ( 3 ) at least 3 reads were required for each variant , with a minimum total coverage of 600 reads per site . There was an unstable tract between base pairs 3902 and 3908 which repeatedly showed heteroplasmy levels between 1% and 5% in all of the samples and the plasmid control , as did a recognised variant at base pair 750 . These mtDNA variants were deemed artefactual and removed from the analysis . Using these stringent criteria , there were no variants present at >0 . 65% in the cloned DNA template ( Table S5 ) . We took a conservative approach and only recorded mutations present at >0 . 8% . This approach ensured that any variants detected in the samples at levels of >0 . 8% are likely to be generated in vivo and be of biological origin . The published base-substitution error rate for mtDNA on the Ion Torrent PGM is 0 . 12% [50] . Colon samples were mounted for sectioning and frozen in isopentane previously cooled to −190°C in liquid nitrogen . Cryostat sections ( 12 µm ) were cut onto glass slides and incubated in COX medium ( 100 µM cytochrome c , 4 mM diaminobenzidine tetrahydrochloride and 20 µg . ml−1 catalase in 0 . 2 M phosphate buffer pH 7 . 0 ) at 37°C for 50 minutes . Sections were washed in phosphate buffered saline , pH 7 . 4 ( 3×5 minutes ) and incubated in SDH medium ( 130 mM sodium succinate , 200 µM phenazine methosulphate , 1 mM sodium azide , 1 . 5 mM nitroblue tetrazolium in 0 . 2 M phosphate buffer pH 7 . 0 ) at 37°C for 45 minutes . Finally , sections were washed in phosphate buffered saline , pH 7 . 4 ( 3×5 minutes ) , dehydrated in a graded ethanol series ( 70% , 95% , 2×100% ) , cleared in Histoclear ( National Diagnostics , Atlanta , USA ) and mounted in DPX .
|
Mitochondrial DNA ( mtDNA ) mutations have been shown to accumulate with age in a number of human stem cell populations and cause mitochondrial dysfunction within individual cells resulting in a cellular energy deficit . The dynamics by which mtDNA mutations occur and accumulate within individual cells ( known as clonal expansion ) is poorly understood . In particular we do not know when in the life-course these mtDNA mutations occur . Here we have measured mtDNA mutation frequency using three different techniques; Random Mutation Capture , which measures low level mutation frequency as an indirect measure of mutation rate , Next Generation Sequencing , which measures clonally expanded mtDNA mutation frequency , and mitochondrial enzyme histochemistry as a marker of clonally expanded mtDNA mutations , on colorectal mucosal biopsies obtained from 207 healthy participants aged 17–78 years . We show that , by 17 years of age , there is a substantial mtDNA point mutation burden and that clonal expansion of early to mid-life mtDNA mutations is likely to be the cause of mitochondrial dysfunction associated with ageing in the human colon .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"mitochondrial",
"genetics",
"mitochondrial",
"dna",
"forms",
"of",
"dna",
"genetics",
"mitochondria",
"biology",
"and",
"life",
"sciences",
"bioenergetics",
"dna",
"energy-producing",
"organelles"
] |
2014
|
Clonal Expansion of Early to Mid-Life Mitochondrial DNA Point Mutations Drives Mitochondrial Dysfunction during Human Ageing
|
Rift Valley fever ( RVF ) is an emerging mosquito-borne viral hemorrhagic fever in Africa and the Arabian Peninsula , affecting humans and livestock . For spread of infectious diseases , including RVF , knowledge , attitude and practices play an important role , and the understanding of the influence of behavior is crucial to improve prevention and control efforts . The objective of the study was to assess RVF exposure , in a multiethnic region in Kenya known to experience RVF outbreaks , from the behavior perspective . We investigated how communities in Isiolo County , Kenya were affected , in relation to their knowledge , attitude and practices , by the RVF outbreak of 2006/2007 . A cross-sectional study was conducted involving 698 households selected randomly from three different ethnic communities . Data were collected using a structured questionnaire regarding knowledge , attitudes and practices that could affect the spread of RVF . In addition , information was collected from the communities regarding the number of humans and livestock affected during the RVF outbreak . This study found that better knowledge about a specific disease does not always translate to better practices to avoid exposure to the disease . However , the high knowledge , attitude and practice score measured as a single index of the Maasai community may explain why they were less affected , compared to other investigated communities ( Borana and Turkana ) , by RVF during the 2006/2007 outbreak . We conclude that RVF exposure in Isiolo County , Kenya during the outbreak was likely determined by the behavioral differences of different resident community groups . We then recommend that strategies to combat RVF should take into consideration behavioral differences among communities .
Rift Valley fever ( RVF ) is a mosquito-borne viral disease which affects humans , livestock and other mammals [1 , 2] . The disease is caused by Rift Valley fever virus ( RVFV ) , an important hemorrhagic fever virus that occurs in Africa and the Arabic Peninsula [3 , 4 , 5] . Since 1930 when the first cases were diagnosed during an epizootic among sheep in the Rift Valley of Kenya , mitigation measures have often put emphasis on the veterinary and human disease perspectives . Often , the focus on managing RVF has been on monitoring and reporting of cases and death incidences to veterinary and public health authorities , managing human cases and deploying veterinary vaccines when available [6] . Although , the importance of knowledge , attitude and practices as drivers of infectious disease occurrence and spread among humans and animals has been established [7 , 8] , the contexts within which RVF occurs remain a nearly neglected research area [9 , 10] . People behavior and practices play an important role in the spread of infectious diseases , and understanding the influence of behavior and practices on the spread of diseases can be crucial in improving prevention and control efforts [11] . Muga et al , [9] in a comprehensive literature review have shown that livestock sacrificial rituals , food preparation and consumption practices , gender roles are among the key factors that influence the transmission of RVF . Health outcomes are generally influenced by the social and cultural variables including socio-economic status ( SES ) , such as educational level , income , and occupational status , ethnicity , gender , poverty and deprivation , in addition to aggregate characteristics of the social environments , such as the distribution of income , social cohesion and social capital etc . [12 , 13] . Ethnicity is a complex trait that is particularly useful and important because it possesses the social dimension necessary for understanding its impact on health outcomes . According to Shields et al , [14] , ethnicity can be a powerful predictor for disease risk . The challenge is to demonstrate how health outcomes are influenced by many factors while recognizing that ethnicity and behavior differences may play an important role in their own right [15] . The objective of this study was to assess RVF exposure in Isiolo County in Kenya from the perspective of people behavior by comparing knowledge , attitudes and practices of three different communities ( ethnic groups ) and attempt to relate this to how they were affected by the RVF outbreak of 2006/2007 .
We sought and obtained the necessary approval to conduct the study from the Ethical Review Committee of the Kenya Medical Research Institute ( KEMRI ) ( Non-SSC protocol No . 2346 ) , to ensure adherence to Kenyan and international ethical guidelines ( and regulations ) governing research . We explained the purpose of the study to the research participants , local community and their leaders . During the data collection stage , all the respondents gave verbal consent . We ensured strict confidentiality in data handling and storage . To account for a set of attributes ( socioeconomic status , levels of knowledge , attitude and practices ) associated with the consequences of affecting exposure to RVF , the conceptual framework of multi-attribute utility theory is appropriate . The framework predicts: ( i ) behavior directly from an individual’s evaluation of consequences or ( ii ) outcomes associated with having or not having a given behavior [16] . The underlying assumption , having a given behavior such as refraining from RVF risk factors , would lead to a reduction in the number of infected people and animals . Therefore , the prediction of outcomes associated with having or not having a given behavior was considering two household health objectives regarding RVF: the reduction of the number of people and the number of animals affected by the disease . We hypothesized that the levels of the attributes may explain the health outcome of each household during the RVF outbreak of 2006/2007 in Kenya based on the behavioral differences assessed through the knowledge , attitude and practices among community groups . The study was conducted in Isiolo County , northern Kenya , one of the counties affected by the RVF outbreak of 2006/2007 and now classified as a medium risk county for RVF outbreak [17] . Isiolo County covers an area of 25 , 366 square kilometers between longitude 36°50’ and 39°30’ East and latitudes 0°5’ and 2° North ( Fig 1 ) . The county of Isiolo is characterized by the very arid , arid and semi-arid agro-ecological zones with an average annual rainfall of 350 mm and temperature of 29°C . The vegetation is comprised of shrubs and acacia trees , which supports rearing of camels , goats , sheep and cattle . Isiolo County is inhabited by five ethnic communities namely Maasai , Borana , Somali , Meru , and Turkana . According to the 2009 population and housing census , the county had a population of 143 , 294 people with a population density of about 6 people per square kilometers comprising of nomads and transhumants . In the present study , four divisions ( Kinna , Merti , Ngaremara and Oldonyiro ) were selected purposively due to the pastoralism practiced by residents . Kinna and Merti are inhabited mainly by the Borana ethnic group , Oldonyiro by the Maasai and Ngaremara by the Turkana . All three ethnic communities derive their livelihood from pastoralism; the Maasai and Borana are transhumants while the Turkana are nomads [18 , 19] . We conducted a cross-sectional study which allowed a single point data collection for each household head in the three communities in April 2014 . A total of 698 households were selected randomly using sample frames provided by the veterinary officers of the locality . One hundred sixty ( 160 ) respondents were selected from the Maasai community , 175 from Turkana and 363 from Borana . The Borana community had the highest number of respondents because they are the majority in Isiolo County and occupy two of the selected divisions ( Kinna and Merti ) . Household heads were interviewed with a structured questionnaire that had four sections which gathered various information such as household demography; knowledge about RVF disease in humans and animals , attitude and perception , and practices that would promote or prevent spread of the disease . Also , information was collected from the communities about the number of people and animals that were affected during the RVF outbreak of 2006/2007 as mentioned by the interviewees . Data were entered in Microsoft Excel Sheet and later cleaned and transferred to STATA software version 13 ( StataCorp LP , TX , USA ) for analysis . One-way ANOVA was used to compare ethnic groups’ characteristics and means were separated using Bonferroni adjustment at 95% confidence interval [20] . Spearman correlation test was used for the correlation between two variables . Composite indices were computed for knowledge , attitude and practice ( KAP ) for each household head . Generally , information collected in social science to describe perceptions and attitudes of people involves the use of scales either binary or multiple . We aimed at quantifying constructs such as KAP which are not directly measurable , by using multiple-item scales and summated weighted ratings to quantify the constructs of interest . Principal Component Analysis ( PCA ) was used to generate the composite indices for Knowledge score , Attitude score , Practice score and a composite of the three denoted KAP-score [21] . Thirty-five knowledge questions , eight attitude and nine practice questions ( see supplementary document ) were used in the computation for the indices . Answers to the questions were coded 1 for correct and 0 for incorrect based on established risk factors that are known to facilitate the spread of RVF [9 , 22 , 23] . The communities are pastoralists and their main livelihood is livestock keeping . The number of animals owned was measured through tropical livestock unit ( TLU ) . TLU allows for computation of different livestock types into one standard measure where one TLU is equivalent to an animal of 250 kg live weight [24] . In addition , computation of a household dependency ratio which expresses the economic burden on the working age population , involved computation of dependents of the percentage of the working-age population [25] .
Most of the household heads interviewed in the three ethnic groups ( Maasai , Borana and Turkana ) were male with the Turkana community having the largest number ( 40% ) of female headed households ( Table 1 ) . The average age of the household head was 45 . 5±0 . 5 years , however , those from the Borana community were significantly older ( p < 0 . 01 ) compared to their counterparts of Turkana and Maasai communities . The average years of formal education of the household heads among the three communities was 3 . 1 with the least and most educated household head having 0 and 20 years of schooling respectively . The average number of people living in the household was 7 . 17±0 . 12 with the Turkana and Maasai having significantly more ( p < 0 . 01 ) persons per household compared to the Borana community ( Table 1 ) . A comparison of the economic dependants showed that the average dependency ratio for the three communities was 142 . 6% with the Maasai and Turkana having significantly more ( p < 0 . 01 ) dependants than the Borana community . The mean TLU for the three communities was 29 . 2 units with the Maasai having significantly more ( p < 0 . 01 ) animals than the Borana and Turkana communities . In terms of the religion , results show that the majority ( 97 . 5% ) of the Borana are Muslim , while the majority of the Turkana ( 98 . 3% ) and the Maasai ( 98 . 75% ) are Christian . Table 2 presents a summary of the risky behavior observed in the three community groups . The average RVF knowledge score among the three community groups was 65 . 2±0 . 6 units ( Table 3 ) . However , the Maasai had significantly more knowledge ( p < 0 . 01 ) about RVF compared to the Borana and the Turkana ( Table 3 ) . The attitude score that was comprised of the pastoralist’s perspective towards the sick animals , and people sick with RVF , showed that the Borana had a significantly higher ( p < 0 . 01 ) attitude score compared to the Turkana and the Maasai . The practice score that was comprised of the recommended practices safeguarding a household from RVF , was statistically higher for the Borana compared to the Turkana . However , no significant difference could be observed between the Borana and the Maasai , while the difference was significant between the Maasai and the Turkana ( Table 3 ) . The KAP score of the Maasai was significantly higher ( p < 0 . 01 ) compared to that of the Borana and the Turkana but no significant difference could be observed between the Borana and the Turkana . People and livestock affected by RVF , according to the communities , was used as proxy for the burden of the disease . The Borana and the Maasai communities suffered relatively less compared to the Turkana in terms of the proportion of livestock ( per 1 , 000 livestock heads ) affected by RVF during the 2006/2007 outbreak . However , results showed that there was no significant difference between the Borana and the Maasai ( Table 4 ) . The proportion of persons ( per 1 , 000 people ) affected by the disease during the 2006/2007 outbreak among the three community groups was not statistically significantly different , though the Turkana indicated the highest proportion of people affected followed by the Borana and the Maasai in the decreasing order ( Table 4 ) . The knowledge score ( knowledge about RVF ) of the respondents had significant and negative association with the number of animal affected in the Turkana ( p < 0 . 05 ) and the Maasai ( p < 0 . 1 ) groups ( Table 5 ) meaning the less the community has knowledge about RVF , the more animal are affected during the outbreak . However , though the association was negative with the Borana , it was not statistically significant . For the number of people affected there was no significant difference between the Borana and the Maasai ( Table 4 ) ; however , the association between the knowledge score and the number of people affected was positive and significant for Borana and negative and non-significant for the Maasai ( Table 5 ) . The attitude score of the Turkana has a negative and significant correlation with the number of animals and the number of people affected ( Table 5 ) . This community group was the most affected compared to the Borana and Maasai ( Table 4 ) . For the Borana the association between the attitude score and the number of animals and the number of people affected was positive but not statistically significant . For the Maasai , the association between the attitude score and the number of animals affected was positive but not statistically significant . However , the association between the attitude score and the number of people affected was negative and significant ( p < 0 . 05 ) . The Maasai are more knowledgeable ( Table 3 ) and their attitude score was also negatively associated with the number of people affected; they were less likely at risk as demonstrated by the number of animals and people affected in this community ( Table 4 ) . The association between the practice score and the number of the animals affected during the outbreak was positive but not statistically significant for the three community groups meaning that they are all vulnerable to RVF in terms of risky practices . For the number of people affected the association was negative for Turkana and the Maasai and positive but not statistically significant and also positive for the Borana but only significant at 10% ( Table 4 ) . Results of the study revealed a negative and significant association between total KAP score and the number of animals and people affected among the Turkana who suffered more during the outbreak compared to the Borana and the Maasai . The Turkana had less KAP score ( Table 3 ) and this may explain why they were more affected in terms of the number of sick animals and people affected compared to the Borana and the Maasai ( Table 4 ) . The Maasai had significantly more KAP score ( Table 3 ) and were less affected . However , for this less affected community , the correlation between total KAP score and the number of animals and people affected among Maasai was negative but not statistically significant .
The importance of behavioral factors as determinants of RVF disease occurrence and spread has previously been documented [7 , 9 , 26 , 27 , 28] . However , little is known on the effect of specific ethnic groups’ behavior on RVF exposure . Results from this study indicated that the Turkana were more affected followed by the Borana and the Maasai in terms of the proportion of people in the household affected by RVF during the 2006/2007 outbreak , but the difference between the three community groups was not statistically significant . A single death in a household can be devastating for a family so the occurrence and the loss of persons , one or many , is what matters . However , as far as the number of animals affected is concerned , the Maasai and the Borana were less impacted in terms of the proportion of livestock affected by RVF . This can be explained by the fact that Maasai and the Borana were significantly more educated compared to the Turkana , as measured by the average number of years of schooling of the household heads . An association between higher educational attainment and better health status has been repeatedly reported in literature . Kawachi et al , [29] have demonstrated that there is evidence to suggest that schooling is causally related to improvements in health outcomes . However , they pointed out that much remains to be known for example , what type of education matters for health . Schooling helps people choose healthier life-styles by improving their knowledge of the relationships between health behaviors and health outcomes [30] . At the time of the study , the three community groups had reasonable knowledge about RVF as expressed by their knowledge score ( Table 3 ) . However , the Maasai that have significantly higher knowledge score were the least affected in terms of the proportion of livestock sick of the RVF disease . The results of the study have shown that religion did not influence the burden of RVF outbreak on the affected population , although the majority of the Maasai and the Turkana are Catholic , they were affected differently by the disease . However , a study by Gray demonstrated that among 38 sub-Saharan African countries , the percentage of Muslims within countries negatively predicted HIV prevalence [31] . Also , every year , because of religion , millions of small ruminants are slaughtered during the religious festivals at Mecca . The risk of infection is high at the time of slaughtering , when aerosols of infected blood may be generated , particularly by traditional sacrificial slaughtering practices [32] . This sacrificial slaughtering may represent a risk of infection , but no RVF disease incident has been reported . However , it was then strongly recommended by the author that the movement of sheep and goats to Mecca for the religious festivals should be strictly prohibited from any area in which epizootic RVF virus has occurred in the previous three to six months [32] . Close contact and handling of sick animals is a potential risk factor for contracting RVF . More people in the Borana community group applied good practices , such as wearing gloves when handling sick animals , taking care of aborted fetuses , and helping animals to deliver as compared to the Turkana who were more affected by RVF . Many emerging diseases are zoonotic infectious diseases transmitted between animals and humans; examples include RVF [26 , 33 , 34 , 35] . Applying good practices by wearing protective gloves can significantly reduce the risk of being infected through transmission of the disease between animals and humans . It has been shown in China that the interaction of people with animals favors the emergence and the spread of new microbial threats [36] . The proportion of female headed households was significantly higher in the Turkana community compared to the Borana and the Maasai . In sub-Saharan Africa , women frequently spend more time than men in animal care [10 , 37] . This may explain why Turkana were more affected in terms of the number of people sick of the disease during the outbreak . However , it is not clear how gender differentiation influences the spread of RVF [10 , 37] . Responsibilities of veterinary services include epidemiological surveillance of animal diseases and ensuring the safety and suitability of meat for consumption [38 , 39 , 40] . The Borana community group who slaughtered more animals under the veterinary inspection was the less affected group in terms of the proportion of animals sick of RVF during the 2006/2007 outbreak , compared to the Turkana and the Maasai , although the difference between the later was not statistically significant . Also , the proportion of people in the household sick of RVF was less in the Borana community group compared to the Turkana , though the difference was not statistically significant . The objectives of meat inspection are twofold; first , meat inspection ensures that only healthy physiologically normal animals are slaughtered for human consumption and secondly is to ensure that meat from animals is free from disease and represent no risk to human health [38 , 39 , 40] . This study showed that the Maasai had a higher KAP score compared to the Boran and the Turkana who were more affected in terms of the proportion of livestock sick of RVF . Though the attitude score was not statistically significantly different between the Maasai and the Turkana , they were differently affected by the disease . This can be explained by the fact that the Maasai had significantly higher knowledge score compared to the Turkana . Also , the Maasai had significantly higher practice score compared to the Turkana . The Borana community group was not different from the Maasai in terms of practice score , though the Maasai were more knowledgeable than them . Moreover , the Borana community group had significantly better attitude compared to the Maasai as shown by their higher attitude score ( Table 3 ) . This study showed that to better describe what is important for the disease burden of the affected community is the combination of knowledge , attitude and practices . Generally , higher knowledge score about a specific disease is not always translated into better practices . For instance , farmers’ practical ability to diagnose African animal trypanosomiasis was higher than suggested by their knowledge about the disease [41] . The high KAP score of the Maasai may explain why they were less affected by the RVF disease during the 2006/2007 outbreak . Behavioral differences were important in explaining why various communities were affected differently by the RVF outbreak in Isiolo County , Kenya in 2006/2007 .
Better knowledge about a specific disease is not always translated into appropriate practices , but rather the application of good practices together with the right attitude , as applied by one of the communities in our study , may explain why this community was less affected by RVF disease . Also , the combination of people´s knowledge , attitude and practices in a single index is more appropriate to explain disease burden rather than the single element taken separately . We conclude that RVF exposure in Isiolo County , Kenya during the RVF outbreak , was to a large extent determined by the behavioral differences of different community groups . Therefore , we recommend that strategies to combat RVF should take into consideration sociocultural and behavioral differences among communities .
|
The Rift Valley fever ( RVF ) outbreak of 2006/2007 affected many regions in Kenya in varied degrees . The number of reported human cases and affected livestock herds varied between regions , but the reason for this variation has not been studied . We have investigated knowledge , attitude and practices differences between three ethnic communities in Isiolo County with different experiences of the 2006/2007 RVF outbreak . The pastoralist communities all had a relatively good knowledge regarding RVF , however the necessary preventive measures against RVF were not always practiced and the attitude towards RVF prevention were sometimes not good . When these results were analyzed on a community level , we found that the community with the best attributes had good preventive practices , positive attitude towards RVF prevention with less people and livestock affected by the disease . However , the combination of knowledge , attitude and practices was the determinant of avoiding RVF by the ethnic groups studied . These results indicate that understanding the behavior of the local communities could improve the preventive strategies to mitigate future RVF outbreaks .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Conclusion"
] |
[
"livestock",
"medicine",
"and",
"health",
"sciences",
"behavioral",
"and",
"social",
"aspects",
"of",
"health",
"tropical",
"diseases",
"geographical",
"locations",
"diet",
"animal",
"products",
"rift",
"valley",
"fever",
"nutrition",
"meat",
"neglected",
"tropical",
"diseases",
"africa",
"veterinary",
"science",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"food",
"veterinary",
"diseases",
"zoonoses",
"behavior",
"epidemiology",
"agriculture",
"people",
"and",
"places",
"kenya",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"ethnic",
"epidemiology"
] |
2017
|
Ethnic groups’ knowledge, attitude and practices and Rift Valley fever exposure in Isiolo County of Kenya
|
Epidemiological studies have shown an inverse correlation between the incidence of lymphatic filariasis ( LF ) and the incidence of allergies and autoimmunity . However , the interrelationship between LF and type-2 diabetes is not known and hence , a cross sectional study to assess the baseline prevalence and the correlates of sero-positivity of LF among diabetic subjects was carried out ( n = 1416 ) as part of the CURES study . There was a significant decrease in the prevalence of LF among diabetic subjects ( both newly diagnosed [5 . 7%] and those under treatment [4 . 3%] ) compared to pre-diabetic subjects [9 . 1%] ( p = 0 . 0095 ) and non-diabetic subjects [10 . 4%] ( p = 0 . 0463 ) . A significant decrease in filarial antigen load ( p = 0 . 04 ) was also seen among diabetic subjects . Serum cytokine levels of the pro-inflammatory cytokines—IL-6 and GM-CSF—were significantly lower in diabetic subjects who were LF positive , compared to those who were LF negative . There were , however , no significant differences in the levels of anti-inflammatory cytokines—IL-10 , IL-13 and TGF-β—between the two groups . Although a direct causal link has yet to be shown , there appears to be a striking inverse relationship between the prevalence of LF and diabetes , which is reflected by a diminished pro-inflammatory cytokine response in Asian Indians with diabetes and concomitant LF .
Global epidemiological studies have shown a marked increase in the incidence of diabetes worldwide . India leads the world in absolute numbers of diabetic subjects [1] . Type-2 diabetes mellitus constitutes about ∼90% of the entire diabetic population . The association between diabetes mellitus and increased susceptibility to infections is well known . Many diseases such as tuberculosis and candidiasis are more common in diabetic patients , while some such as invasive otitis externa and rhinocerebral mucomycosis occur almost exclusively in people with diabetes [2] . In addition , infections with group B streptococcus and Klebsiella spp . occur with increased severity in patients with diabetes and may be associated with an increased risk of complications [2] . Infection with systemic helminths , in addition to causing morbidity by themselves , may contribute to increased morbidity due to diabetes . But , there is very little data available on the prevalence of lymphatic filariasis ( LF ) among people with diabetes , although studies have examined the coexistence of LF with HIV [3] , malaria [4] and tuberculosis [5] . Current estimates suggest that 129 million persons worldwide are infected with one of the three lymph-dwelling filariae ( Wuchereria bancrofti , Brugia malayi or B . timori ) , the major causative agents of LF . The disease burden from LF is concentrated in tropical and sub-tropical countries ( such as India ) where the prevalence of type-2 diabetes is greatest [6] . This is particularly true in South India where the prevalence of LF caused by W . bancrofti is between 6–20% based on circulating filarial antigenemia [5] . Thus , in the present study , the influence of LF on diabetes was examined as part of an ongoing , prospective epidemiological study in Chennai , Southern India .
Institutional ethical committee approval from the Madras Diabetes Research Foundation Ethics Committee was obtained ( Ref No-MDRF-EC/SOC/2009//05 ) and written informed consent was obtained from all the study subjects . Study subjects were recruited from the Chennai Urban Rural Epidemiology Study ( CURES ) , an ongoing epidemiological study conducted on a representative population of Chennai ( formerly Madras ) , the fourth largest city in India . The methodology of the study and the prevalence of diabetes in Chennai have been published elsewhere [7] , [8] . Briefly , in Phase 1 of the urban component of CURES , 26 , 001 individuals were recruited based on a systematic sampling technique with random start . Fasting capillary blood glucose was determined using the OneTouch Basic glucometer ( Lifescan , Johnson & Johnson , Milpitas , CA ) in all subjects . Details of the sampling are described on our website ( http://www . drmohansdiabetes . com/bio/WORLD/pages/pages/chennai . html ) . In Phase 2 , detailed studies of diabetic complications , including nephropathy and retinopathy , were performed , and in Phase 3 , every 10th individual in Phase 1 was invited to participate in more detailed studies . As part of the questionnaire , the socio-economic details of the study participants was collected and recorded . For the present study , the following groups were randomly selected from Phase 3 of CURES , Group 1- 943 normal glucose tolerance subjects ( NGT ) ; Group 2- 154 subjects with impaired glucose tolerance ( IGT ) ; Group 3- 158 newly diagnosed type-2 diabetes subjects ( ND-DM ) and Group 4- 161 known type-2 diabetes subjects under treatment ( KDM ) . A larger NGT group was included based on sample size calculations and to obtain baseline values for the normal population . The filarial status of these individuals was not known at the time of recruitment into the study . Anthropometric measurements , including height , weight , and waist circumference , were obtained using standardized techniques . The body mass index ( BMI ) was calculated as the weight in kilograms divided by the square of height in meters . Fasting plasma glucose ( FPG ) ( glucose oxidase-peroxidase method ) , serum cholesterol ( cholesterol oxidase-peroxidase- amidopyrine method ) , serum triglycerides ( glycerol phosphate oxidase-peroxidase-amidopyrine method ) , high density lipoprotein cholesterol ( HDL-C ) ( direct method-polyethylene glycol-pretreated enzymes ) , and creatinine ( Jaffe's method ) were measured using a Hitachi-912 Autoanalyser ( Hitachi , Mannheim , Germany ) . The intra- and inter assay coefficient of variation for the biochemical assays ranged between 3 . 1% and 5 . 6% . Glycated hemoglobin ( HbA1c ) was estimated by high pressure liquid chromatography using a variant machine ( Bio-Rad , Hercules , CA ) . The intra- and inter-assay coefficient of variation of HbA1c was less than 5% . To quantify the filarial antigen levels and prevalence , sera were analyzed using the W . bancrofti Og4C3 antigen-capture enzyme-linked immunosorbent assay ( Tropbio , James Cook University , Townsville , Queensland , Australia ) according to the manufacturer's instructions . The serum antibody ( IgG and IgG4 ) titer against Brugia malayi antigen ( BmA ) was determined by ELISA as described previously [9] . The levels of cytokines ( TNF-α , IL-6 , IL-1β , GM-CSF , IFN-γ , IL-13 and IL-10 ) in the undiluted serum were measured using a Bioplex multiplex cytokine assay system ( Biorad , Hercules , CA ) . The lowest detection limit for the various cytokines were IL-1β -2 . 7pg/ml , IL-2-1 . 16 pg/ml , IL-4-0 . 3 pg/ml , IL-5- 2 . 08 pg/ml , IL-6- 2 . 31 pg/ml , IL-10- 2 . 2 pg/ml , IL-12- 2 . 78 pg/ml , IL-13- 2 . 22 pg/ml , IL-17 2 . 57 , GM-CSF- 0 . 67 pg/ml , IFN-γ- 2 . 14 pg/ml and TNF-a- 4 . 89 pg/ml . TGF-β was estimated by conventional ELISA following manufacturer's protocol ( R&D , Minneapolis , MN ) . The lowest detection limit for TGF-β was 7 . 8 pg/ml . All statistical analyses were performed using SPSS software ( Version 15 . 0 . 0 , Chicago ) . The prevalence of filarial infections among the different groups was analyzed by Pearson's Chi-Square test . The antigenic load and antibody titers were analyzed by Mann Whitney U test . The clinical and biochemical characteristics of the study subjects were compared using one-way ANOVA analysis . To account for multiple comparisons , the cytokine levels in controls ( DM−LF− ) , DM+LF+ and DM+LF− groups were compared by multinomial logistic regression analysis . P values<than 0 . 05 were considered significant .
The baseline characteristics including demographics , clinical and biochemical features of the study population are shown in Table 1 . As can be seen in the table , compared to subjects with normal glucose tolerance ( NGT ) , those with glucose intolerance ( i . e . IGT , ND-DM , KDM ) had higher BMIs , systolic and diastolic blood pressure , serum cholesterol , LDL and triglycerides levels but lower HDL cholesterol levels . Since , socio-economic differences could be a confounding factor for the prevalence of both diabetes [10] and LF [11] , the average monthly income of study subjects was recorded ( Table 2 ) . We also examined the occupation profile of the study subjects ( Table S1 ) . As can be seen , there were no significant differences in the socio-economic status between the various groups . The prevalence of LF among the various groups was determined by quantitative TropBio ELISA ( >128 U/ml ) and differences in the prevalence of LF among the groups were observed . The prevalence of LF was found to be 10 . 4% in the NGT , 9 . 1% in IGT , 5 . 7% in ND-DM and 4 . 3% in KDM respectively . The differences in the prevalence rate between NGT and KDM ( p = 0 . 0463 ) and NGT and ND-DM ( p = 0 . 0095 ) were significant ( Fig 1a ) . To examine more quantitatively the association of type-2 diabetes and LF , we next quantified the serum circulating filarial antigen ( CFA ) levels among the filarial positive subjects . Not only was there a difference in prevalence rates between those with NGT and those with glucose intolerance but there was also a clear difference in the absolute levels of CFA in the LF-infected individuals , with CFA levels being lower among the diabetic groups compared to the NGT group ( Fig 1b ) . The geometric mean ( range ) of CFA levels in the four groups were: NGT-1 , 594 ( 127–32 , 768 ) , IGT-1 , 520 ( 209–16 , 345 ) , ND-DM-929 ( 129–32 , 768 ) and KDM-351 ( 163–1 , 126 ) with the differences in the antigen levels between the KDM and the NGT ( p = 0 . 04 ) and the KDM and the IGT ( p = 0 . 04 ) being statistically significant . We next quantified the serum anti-filarial antibody levels among those with LF in the four groups ( Fig 2 ) . Even though the mean IgG4 levels were not different among the four groups , the mean IgG levels were significantly lower in the KDM compared to NGT ( p<0 . 0023 ) , We next quantified pro- and anti-inflammatory cytokine levels among control ( DM−LF− ) , diabetic only ( DM+LF− ) or both LF and diabetic ( DM+LF+ ) subjects ( Fig 3 ) . In comparison to controls , the diabetes only ( DM+LF− ) group had high levels of IL-6 and GM-CSF , which was significant ( p<0 . 01 ) ( Fig 3b and c ) . There was no significant difference in the levels of TNF- α ( Fig 3a ) . In DM+LF+ subjects , there was a significant reduction in the levels of IL-6 and GM-CSF compared to the diabetic only group ( DM+LF− ) ( geometric mean ( GM ) of 13 . 57 pg/ml versus 45 . 13 pg/ml for IL-6 , p<0 . 05; and GM of 0 . 81 pg/ml versus 2 . 24 pg/ml for GM-CSF , p<0 . 05 ) . TNF- α levels were not significantly different between the two groups . IL-1β levels were not statistically different among the three groups ( data not shown ) . IFN-γ levels were significantly elevated in diabetic only group compared to controls but was unaltered by the LF status ( Fig 3d ) . When the levels of anti-inflammatory cytokines were measured , both the diabetic groups ( DM−LF+ and DM+LF+ ) had comparable levels of IL-13 and TGF-β ( Fig 3e and f ) , but TGF-β was significantly higher in both the diabetic groups compared to controls ( p<0 . 01 ) . No significant difference was seen in IL-10 levels among the three groups ( Fig 3g ) .
It is well known that individuals with diabetes are at increased risk of susceptibility to several infectious diseases including tuberculosis [12] , urinary tract infection [13] and mucormycosis [14] . It is generally assumed that diabetes increases the susceptibility to all infections [2] . But in this study , we clearly demonstrate that the prevalence of LF is lower in subjects with type-2 diabetes . A serial decline in the prevalence of LF was seen as individuals progressed from NGT to IGT to ND-DM to KDM with the significance being maintained even after adjusting for age and gender . The decrease in prevalence was associated with decreased antigen load and anti-filarial IgG antibody titer but the anti-filarial IgG4 titer was unaffected . The reduced IgG levels in diabetic subjects was expected since previous reports have shown reduced levels of IgG and increased levels of IgA among diabetic subjects [15] , [16] . In terms of humoral responses , both subjects with active filarial infection and those who are exposed but resistant to infection mount vigorous antibody responses to parasite antigen , most specifically , IgG4 [9] , [17] . Thus , BmA- specific IgG4 can be considered to be a good surrogate marker for exposure . The fact that IgG4 levels were not significantly different between the four groups , is further evidence to show that , differences in exposure to infection is less likely a reason for the differences in prevalence between the diabetic and non-diabetic groups . It is very unlikely that the decreased prevalence of LF among diabetic subjects was due to LF-mediated mortality , as LF is a chronic , non-lethal disease . Since , differences in socio-economic status could be a confounding factor; the average monthly income and occupation of the study subjects were analyzed . No significant difference was seen between the study groups , indicating that , the socio-economic status is unlikely to be a potential confounding factor for the differences seen in the prevalence of LF . Another confounding factor that could be of significance is the nutritional differences [18] that could arise due to calorie restriction among diabetic subjects . But , there was no difference with respect to average protein intake ( approximately 12% ) among different groups , again suggesting that nutritional differences or dietary intake are unlikely to be potential confounding variables in the study . More likely is the interplay between two chronic conditions in which the longstanding regulatory environment seen in LF may play a role in conferring resistance to type-2 diabetes . There are some reports that have documented a similar inverse relationship between diabetes and hepatitis C infection [19] , but other studies failed to reproduce the association [20] . More studies are thus needed on the coincidence of diabetes and other infectious diseases to have a better understanding about the interplay of infection/inflammation and diabetes . In mice , there is evidence to show that filarial infection can prevent type-1 diabetes ( “hygiene hypothesis” ) [21] , but whether the same immunomodulatory effect can dampen inflammation and protect against type-2 diabetes is currently not known . To better understand the mechanism associated with the decreased prevalence of LF among diabetic subjects , we studied the serum cytokine levels . The main focus was to determine the effect of LF on diabetes in terms of the serum cytokine profile . Although filarial parasites elicit a broad spectrum of inflammatory and regulatory responses mediated by cytokines , whether this type of immunomodulation occurs in co-incident diabetes has not been well-studied . The serum cytokine profile of LF only patients has already been reported by Satapathy et al . [22] and us , previously [23] . Diabetes in conjunction with LF had decreased levels of TNF-α and IL-6 ( cytokines that have already been associated with insulin resistance ( IR ) [24] ) , compared to those with DM alone . Diabetic subjects ( without LF ) had a typical pro-inflammatory phenotype with high levels of TNF-α , IL-6 and GM-CSF . IL-1β , however , was not elevated in these subjects , although it has been previously been shown to act synergistically with TNF-α and IL-6 in inducing IR [24] . The contribution of GM-CSF ( seen to be elevated in diabetic subjects ) in mediating IR is currently not known . Interestingly , in those with both diabetes and LF , the pro-inflammatory cytokines - TNF-α , IL-6 and GM-CSF were reduced , compared to those without LF , suggesting that LF-mediated reduction of pro-inflammatory cytokines negatively influences the development of IR . Although pro-inflammatory cytokine levels are typically elevated in diabetic subjects , the data on the balance of effector T cell phenotypes in this condition has been confusing with some studies reporting Th1 polarization [25] , others reporting Th2 skewing [26] , and still others reporting a balanced response [27] . In the present study , diabetic subjects ( who were LF negative ) had very high levels of IFN-γ and IL-13 suggestive of a mixed ( relatively non-polarized ) phenotype . Whether this immune phenotype is the cause or effect of IR remains to be established . Although the anti-inflammatory cytokines - IL-10 and TGF-β - have largely been associated with immune-regulation in LF ( with IL-10 playing the major role ) , the down regulation of pro-inflammatory cytokines in type-2 diabetic subjects with LF seems to be due to TGF-β and not IL-10 . However , as TGF-β is often elevated in diabetic subjects , other immunomodulatory molecules such as CTLA-4 , PD-1 , IDO might act in concert with TGF-β in the modulation of the immune responses in these groups [28] . Our study suggests that one mechanism by which LF can potentially protect against type-2 diabetes , is by modulating the pro-inflammatory environment seen in diabetes . Helminth infections , such as LF could modulate diabetes by inducing a chronic , non-specific , low-grade , immune regulation mediated by Th2/Tregs ( modified Th2 response ) which in turn can suppress the pro-inflammatory responses [29] . One can speculate that , childhood filarial infection reduces TNF-α , IL-6 and GM-CSF levels thereby conferring protection against type-2 diabetes . An alternate hypothesis could be that the inflammation associated with diabetes may also promote anti-filarial immunity by enhancing anti-filarial antibody responses that could mediate parasite clearance [30] . Although this study suffers from the limitation of being a cross-sectional study and therefore not providing a direct causative mechanism for the decreased prevalence of LF among diabetic subjects , the study highlights the importance of the need to understand the complex interactions between an infectious disease ( LF ) and a metabolic disorder ( Type-2 diabetes ) . In addition , the immune as well as non-immune mechanisms by which the interplay between LF and DM occurs needs to be explored in detail . Finally , the decreasing incidence of LF ( due to mass eradication programs ) could have an impact on the incidence of type-2 diabetes in the future and obviously this is an important area for future study .
|
Childhood helminth infections can reduce the risk and severity of allergies and autoimmune diseases , by means of immunomodulation , and a decrease in helminth infections could potentially account for the increased prevalence of these diseases in the western world ( hygiene hypothesis ) . We hypothesized that the same immunomodulatory effect can have an impact on metabolic diseases like obesity , diabetes , hypertension and atherosclerosis , wherein inflammation plays a crucial role ( extended hygiene hypothesis ) . To test this hypothesis , we examined the prevalence of lymphatic filariasis ( LF ) among diabetic , pre-diabetic and non-diabetic subjects who were part of the CURES ( Chennai Urban Rural Epidemiology Study ) study . In accordance with our hypothesis , we found reduced prevalence of LF among diabetic subjects compared to non-diabetic and pre-diabetic subjects . This was associated with decreased filarial antigen load and anti-filarial antibody levels . The association remained significant even after adjusting for socioeconomic status , age and gender . Interestingly , within the diabetic subjects , those who were filarial positive had reduced levels of pro-inflammatory markers ( TNF-α , IL-6 and GM-CSF ) compared to those who were filarial negative . In light of these findings , the decreasing incidence of filarial infection due to mass drug administration could potentially have an unexpected adverse impact on the prevalence of diabetes in India .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"immunology/cellular",
"microbiology",
"and",
"pathogenesis",
"immunology/immunomodulation",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/helminth",
"infections",
"immunology/immunity",
"to",
"infections",
"diabetes",
"and",
"endocrinology/type",
"2",
"diabetes"
] |
2010
|
Decreased Prevalence of Lymphatic Filariasis among Diabetic Subjects Associated with a Diminished Pro-Inflammatory Cytokine Response (CURES 83)
|
Hantaviruses are important emerging human pathogens and are the causative agents of serious diseases in humans with high mortality rates . Like other members in the Bunyaviridae family their M segment encodes two glycoproteins , GN and GC , which are responsible for the early events of infection . Hantaviruses deliver their tripartite genome into the cytoplasm by fusion of the viral and endosomal membranes in response to the reduced pH of the endosome . Unlike phleboviruses ( e . g . Rift valley fever virus ) , that have an icosahedral glycoprotein envelope , hantaviruses display a pleomorphic virion morphology as GN and GC assemble into spikes with apparent four-fold symmetry organized in a grid-like pattern on the viral membrane . Here we present the crystal structure of glycoprotein C ( GC ) from Puumala virus ( PUUV ) , a representative member of the Hantavirus genus . The crystal structure shows GC as the membrane fusion effector of PUUV and it presents a class II membrane fusion protein fold . Furthermore , GC was crystallized in its post-fusion trimeric conformation that until now had been observed only in Flavi- and Togaviridae family members . The PUUV GC structure together with our functional data provides intriguing evolutionary and mechanistic insights into class II membrane fusion proteins and reveals new targets for membrane fusion inhibitors against these important pathogens .
The Bunyaviridae is a large and diverse virus family of human , animal and plant pathogens that encompasses five genera; Phlebovirus , Orthobunyavirus , Hantavirus , Nairovirus and Tospovirus . Members of the Hantavirus genus are rodent-borne zoonotic viruses and are important human pathogens responsible for severe illnesses such as hemorrhagic fever with renal syndrome ( HFRS ) , and hantavirus pulmonary syndrome ( HPS ) [1–4] . Puumala virus ( PUUV ) , the causative agent of a mild form of HFRS was first isolated in Finland [5] . In humans , PUUV infection is mostly asymptomatic or manifested with minor symptoms . However , outbreaks were recently reported in central Europe with growing numbers of affected patients [6–8] . The bank vole ( Myodes glareolus ) is the main reservoir of the virus and transmission to humans occurs typically via aerosols of the rodent excreta with no role for arthropod vectors . Hantaviruses encompass a tripartite , negative sense ssRNA genome . The viral medium ( M ) segment encodes the two glycoproteins , GN and GC , originating from a glycoprotein precursor ( GPC ) that is cleaved into N- and C-terminal fragments [9–11] . GN and GC assemble into a lipid bilayer envelope to form an outer protein shell . The non-continuous , pleomorphic envelope projects GN and GC as a spike complex bearing an apparent four-fold symmetry [12] . Recently , the atomic resolution structure of GN was published and together with electron cryo-tomography data it was proposed to be located at the membrane distal part of the spike complex [13] . However the structure , orientation and stoichiometry of GC within the spikes remain unclear . To deliver their RNA genome into the host cell cytoplasm , hantaviruses must fuse their envelope with a cellular membrane . Like other enveloped viruses , hantaviruses rely on their glycoproteins to induce membrane fusion [14] . Following attachment to the host cell , hantaviruses usually undergo clathrin-mediated endocytosis ( CME ) . Interestingly , clathrin-independent endocytosis was reported for some hantaviruses [15 , 16] , implying that different routes may be involved in these viruses entry . In both routes , however , the virus is directed to an endosomal compartment where the glycoproteins respond to the reduced pH of the compartment with a sequence of conformational changes [17] . These conformational changes expose a hydrophobic motif , which is inserted into the endosomal membrane [18 , 19] . The glycoprotein then folds back on itself , forcing the cell membrane ( held by the fusion motif ) and the viral membrane ( held by a transmembrane anchor ) to proximity , inducing the viral and endosomal membranes to fuse [20–22] . Based on bioinformatic studies and in vitro experiments using synthetic peptides it was postulated that hantavirus GC adopts a class II membrane fusion protein fold [23 , 24] . Until recently , viral class II fusion proteins were thought to be restricted to members of the Flavivirus genus ( family: Flaviviridae ) and the Togaviridae . However , the crystal structure of GC from Rift Valley fever virus ( RVFV—family Bunyaviridae , genus: Phlebovirus ) showed that the class II fold extends beyond these two families [25] . Interestingly , not all Flaviviridae members contain a class II membrane fusion protein as bovine viral diarrhea virus ( BVDV , genus: Pestivirus ) E2 protein and hepatitis C virus E2 ( HCV , genus: Hepacivirus ) exhibit completely different folds in their proposed fusion proteins [26 , 27] . In the absence of high-resolution structures for the complete E1 proteins from these viruses this data suggests that BVDV and HCV ( flavivirus ) fusion proteins do not adopt a class II fold . The transition of class II membrane fusion proteins from their pre-fusion homo- or heterodimers on the virus surface to a post-fusion homotrimer has been shown to depend on the acidification of the virus’ environment [21 , 28–30] . Recently , Acuña and colleagues have shown that GC from Andes virus ( ANDV , genus: Hantavirus ) forms trimers in response to acidic environment at pH 5 . 5 [17] . Hantavirus fusion activity was also demonstrated by syncytia formation upon low pH treatment of Vero E6 cells expressing GN and GC glycoproteins [14 , 31] . In this cellular context , a pH of 5 . 9 was found to activate fusion of Andes virus while a pH of 6 . 3 was reported as the activation threshold for Hantaan virus [14 , 32] . In the absence of experimental high-resolution structural data for GC , the molecular basis of membrane fusion in hantaviruses remains obscure . Here we present the first high-resolution structure of a fusogen from the hantavirus genus .
The ectodomain of PUUV GC spans residues 659–1114 ( GPC numbering , 1–456 in GC numbering ) . To obtain soluble protein for structural studies , we expressed only PUUV GC residues 659–1106 ( 1–448 , soluble GC or sGC ) using baculovirus expression system and purified it to homogeneity ( see material and methods ) . During the elution step of ion exchange ( IEX ) chromatography we obtained two populations ( termed sGCXF1 and sGCXF2 ) that each crystallized in a distinct crystal form . We then determined the crystal structures of sGCXF1 and sGCXF2 to 1 . 8 Å and 2 . 5 Å resolution , respectively , with excellent crystallographic statistics ( Table 1 ) . Although sGCXF1 crystals appeared in pH 6 . 0 and sGCXF2 in pH 8 . 0 , in both crystal forms PUUV sGC adopts the three-domain architecture of the post-fusion conformation of class II viral fusion proteins . It is not unprecedented that some class II membrane fusion proteins were crystallized in their post-fusion conformation without low pH triggering [33 , 34] , however we cannot exclude that for sGCXF2 the pH was not changed during the crystallization period . The overall structure in both crystal forms is similar so to simplify our discussion we will refer mainly to the sGCXF1 unless mentioned otherwise . Viral class II membrane fusion proteins were found previously only in flaviviruses , alphaviruses , rubivirus and more recently in a phlebovirus [25 , 35–37] ( Figs 1 , S1 ) . The crystal structure of PUUV sGC spans residues 666–1076 ( GPC numbering ) , lacking seven N-terminal and 30 C-terminal residues of the expressed ectodomain . Domain I , an eight-stranded β-sandwich ( with strands termed B0-I0 ) , is the center of the structure that arranges domain II and III around it ( Fig 1 ) . Two insertions in domain I between strands D0-E0 and strands H0-I0 form the elongated , mostly β-stranded domain II . The putative fusion loop , the endosomal membrane anchor , is located on the part of domain II that is distal to domain I . Domain III is an IgC-like module with six β-strands and is followed by a segment of eight amino acids of the so-called stem region . Compared to fusion proteins from flaviviruses and alphaviruses , PUUV GC has a longer stem region connecting domain III to the transmembrane ( TM ) domain . The stem region of PUUV GC spans approximately 44 residues , including two conserved cysteines ( S2 Fig ) . Due to its disordered nature we could not detect electron density for most of this region . However , the first eight residues of the stem ( 1068–1076 ) could be modeled in both , sGCXF1 and sGCXF2 . The last residue visible in both of our structures is T1076 , which lays ~30 Å from the fusion loop ( Fig 1 ) . The remaining 38 residues connecting to the TM anchor can easily cover the distance to the fusion loop . The overall domain organization ( in particular the position of domain III ) , the parallel trimeric assembly and the stem peptide directionality imply that our structure represents sGC in its post-fusion conformation , or at least in the final stages of the fusion between the viral and the host-cell membranes . PUUV sGC from both preparations ( sGCXF1 and sGCXF2 ) is a monomer in solution as determined by size exclusion chromatography ( SEC ) ( S3A Fig ) . To investigate the oligomeric state of sGC at different pHs , we used size exclusion chromatography combined with multiangle light scattering ( SEC-MALS ) at pH 8 . 0 and pH 5 . 0 . Unexpectedly , we found that low pH does not trigger sGC trimerization in solution as sGC scatters as a monomer even at pH 5 . 0 ( S3B Fig ) . Elution of sGC was significantly retarded at pH 5 . 0 compared to pH 8 . 0 , most likely due to non-specific interaction of the protein with the dextran resin [38] . The same effect was reported also for RVFV GC ectodomain [25] . Nevertheless , in both PUUV sGc structures , one molecule in the asymmetric unit assembles into a homotrimer around the crystallographic three-fold axis ( Fig 1 ) . The protomers adopt the post-fusion domain arrangement , resembling other class II post-fusion structures [14 , 20 , 21 , 33 , 34] ( S1 Fig ) . They associate in a parallel arrangement with the fusion loop placed at the same end of a stable elongated molecule . The C-terminal stem region is pointing towards the target membrane ( Fig 1 ) . PUUV GC and RVFV GC ( Genus: Phlebovirus ) , both members of the Bunyaviridae family , share some structural features that are different from other class II proteins . Similar to phleboviruses , GC from hantaviruses has a high cysteine content , with 26 cysteine residues . In our structure we located 24 cysteines involved in 12 disulfide bonds . Electron density for the remaining cysteines ( C1094 and C1098 ) , at the C-terminal end of the protein , could not be detected . It was suggested before that the 787C-X-X-C790 motif , mapped to domain II , might be involved in disulfide rearrangement to prevent hantavirus inactivation under conditions of low-pH treatment [39] . In our structure , C787 and C790 are located at the membrane proximal region of domain II and are involved in two different disulfide bonds ( with C749 and C913 , respectively ) . From the only other Bunyaviridae fusogen structures ( RVFV GC in its pre-fusion and pre-hairpin conformations , PDB ID 4HJ1 and 4HJC , respectively ) , the analogous cysteines have a similar arrangement [25] despite the hinge motions between the two conformations . Therefore , from comparing these two structures with the post-fusion structure of PUUV sGC we conclude that in contrast to the fusogen activation in some class I membrane fusion proteins , where disulfide rearrangement is essential for preventing a premature fusion [40] , these disulfides do not reorganize . Instead , they are responsible to rigidify the structure and stabilize the orientation of the putative fusion loop . Our structure provides a direct view on the putative endosomal membrane anchor of GC known as the fusion loop and contained between β strands c and b ( Fig 2 ) . It was previously demonstrated for Andes virus ( ANDV ) GC , a member of the Hantavirus genus , that single mutations in the conserved residues W773 , N776 and D779 ( W115 , N118 and D121 in GC numbering ) located at the fusion loop eliminate cell-cell fusion activity and ANDV pseudotyped particles infectivity [32] . From our structure it is apparent that W773 and P781 form a conserved hydrophobic surface ( Fig 2B and 2C ) , exposed towards the target membrane . The N-H group of the W773 side chain forms a hydrogen bond with the carbonyl oxygen of P781 , reducing its hydrophilicity and thereby favors the penetration of the fusion loop into the outer leaflet of the endosomal membrane ( Fig 2D ) . This interaction was reported also for dengue virus E trimer where W101 is interacting in the same way with the carbonyl group of G106 [21] . Notably , the side chain of the charged D779 , also located in the fusion loop , is pointing to the opposite direction , away from the purportedly membrane plane . Notably , the essential residue N776 maintains a network of hydrogen bonds principally with the main chain carbonyls of residues C780 , G782 and with the amine group of residue G785 . Therefore , N776 stabilizes the architecture of the fusion loop , thus explaining its importance for fusion . The fusion loop of PUUV GC contains other genus-specific features . It has a three-residue insertion ( 777P-X-D779 ) conserved among hantaviruses ( Figs 2A , S2 ) where X is typically a proline but can be replaced by serine or glycine ( S2 Fig ) . Unlike post-fusion trimers from the Flavivirus , Alphavirus and Phlebovirus genera , the hydrophobic surface at the tip of domain II is extended by the conserved F907 positioned at the loop connecting strands i and j ( Fig 2B ) . Even though it is less conserved , Y746 located at a third loop contained between strand a and αA helix , might participate in the membrane anchoring as its side chain directing towards the target membrane and is nearly at the same plane of the other hydrophobic side chains of the fusion loop ( Fig 2B and 2C ) . PUUV sGC trimerizes through central interactions in domain I and in the domain-I proximal half of domain II . The total surface buried in trimer interfaces is 5850 Å2 ( 1950 Å2 per monomer ) , 17% larger than in DENV2 E trimer ( PDB code 1OK8 ) , but only 3% larger than in the Semliki forest virus ( SFV ) E1 trimer ( PDB code 1RER ) . In addition to the extensive trimerization interface , there are few elements that are exclusive to the PUUV sGC trimer: unlike other class II members , PUUV GC has an N-terminal extension of domain I that donates a strand , A0 , to the B0-I0-H0-G0 β-sheet from the neighboring protomer , creating an intermolecular continuous beta sheet ( Figs 1 and 3A ) . This N-terminal extension has not been found in structures from the well characterized class II fusion proteins , including that of phlebovirus GC [25] , and therefore it seems to be a unique feature of hantaviruses . Cross-protomer interactions are not common in class II trimers . Typically , the protomers are packed against one another making interactions between secondary structure elements in adjacent protomers . A cross-protomer swap was reported only in Rubella virus E1 protein where the C-terminal stem region donates two strands to two different β-sheets of a neighboring protomer [34] . Additionally , there are few cross-protomer salt bridges in the PUUV sGC trimer . The most notable one is at the membrane proximal part of domain II , close to the fusion loop , where E770 forms a salt-bridge with R902 from the neighboring molecule ( Fig 3B ) , thereby stabilizing the trimer in the membrane-proximal region . To functionally test the significance of this salt-bridge in a hantavirus glycoprotein-mediated cell-cell fusion assay ( 14 , 32 ) we introduced an alanine substitution of R902 ( R244 in GC numbering ) into PUUV GPC . In addition , the same mutation was also introduced to GPC of ANDV to exploit several approaches that have been established for this virus . The GC sequence of hantaviruses is highly conserved and amounts in the case of PUUV and ANDV to 76% of identity and 89% of similarity ( S2 Fig ) . When cells expressed the wild type and R902A constructs of PUUV and ANDV , GC localized efficiently at the cell surface ( S4A Fig ) . Upon acid-induced incubation , the PUUV and ANDV GC R902A mutants induced syncytia as the wild type proteins ( S4B Fig ) , indicating that the inter-protomer salt bridge may have a less crucial role for fusion activity ( see discussion below ) . PUUV GC is predicted to have two glycosylation sites , N898 and N937 . In our crystal structure we observed N-linked glycans only on N937 whereas N898 is buried in the trimer interface with no available space to accommodate a glycan chain . We therefore conclude that N898 is not glycosylated . In contrast to other class II post-fusion trimers , where the glycans decorate the perimeter of the trimer assembly , in PUUV GC the glycans linked to N937 are tightly packed between domain II of one protomer and domain III of the neighboring protomer ( Fig 3C ) . Except one hydrogen bond between N999 ( domain III ) and the first N-acetylglucosamine residue , all contacts with the glycans are via hydrophobic interactions . Indeed , it was previously reported that eliminating the glycosylation on N928 in Hantaan virus GC ( analogous to PUUV GC N937 ) is sufficient to prevent cell fusion [41] . Based on our structure and the previous biochemical data , we conclude that the contribution of the glycans to the PUUV GC trimer interface is a key element in stabilizing trimer assembly in hantaviruses . Previous studies on Hantaan virus ( HNTV ) neutralizing monoclonal antibodies ( MAb ) against GC showed sequence dependent reactivity . While the antibodies cross-reacted with other hantaviruses ( SEOV , DOBV ) , they failed to neutralize PUUV [42] . In addition , binding of neutralizing and non-neutralizing MAbs to HNTV GC was mapped to a region that include most of domain III but no specific epitope was determined [43] . Several neutralizing MAb against PUUV have been selected [44–46] , two of which were shown to recognize GC ( human MAb 1C9 and bank vole MAb 4G2 ) . A peptide scan assay was used to identify the linear epitopes for these MAb [39 , 47 , 48] . The epitopes for 1C9 and 4G2 MAb map to domain I and II , respectively , and both epitopes contribute to the trimer interface ( Fig 4 ) . The GN-GC dissociation at pH 6 . 2–6 . 4 [39] implies exposure of epitopes in GC that were previously buried or partially exposed in the assembled virion . However it seems that each antibody targets a different stage in the membrane fusion process . In class II proteins the major conformational change within a protomer during membrane fusion is the relocation of domain III [20 , 21 , 33] . Our structural overlay analysis shows that PUUV sGC is more structurally related to alphaviruses then to phleboviruses ( S5A Fig ) . Furthermore , previous homology modeling studies used various alphavirus E1 proteins as a template for hantavirus GC [39] . To generate a pre-fusion model for PUUV sGC monomer , we therefore used SFV E1 protein as our reference model . Interestingly , the 1C9 epitope is exposed in our pre-fusion model while in the post-fusion structure it is protected by domain III ( Fig 4 ) . This suggests that binding of MAb 1C9 restricts domain III relocation and thus inhibits the fusion process . However , the multimerization arrangement of GC on the virus envelope needs to be taken into account as this epitope might be partially or completely buried in the context of the mature virion . In contrast , the epitope of MAb 4G2 maps to domain II in proximity to the fusion loop . It was shown for PUUV that the neutralizing MAb 4G2 binds to GC at neutral pH , however 4G2 does not recognized GC that was exposed to low pH [39] . The 4G2 epitope was narrowed down to five residues that are sufficient for the antibody to bind and neutralize ( Fig 4 , dark yellow surface ) [42] . Although this region of the epitope barely makes contacts with the neighboring protomer , the presence of a bound antibody will sterically hinder the formation of the trimer and thereby is expected to prevent fusion . Once a trimer is formed , the 4G2 antibody can no longer bind this epitope and therefore will not be reactive . Taken together , our structural epitope analysis and the disappearance of the 4G2 epitope below pH 6 . 2 [39] propose that the 4G2 MAb inhibit membrane fusion through interfering in trimer formation . It has been shown that in class II membrane fusion proteins there is a hinge motion between domain I and II [reviewed in [19]] , and mutations at that region affect the pH threshold for fusion . However it seems that in phlebovirus GC this region is more rigid [25] . As mentioned before , we also obtained crystals of PUUV sGC at pH 8 . 0 ( sGCXF2 , see Table 1 ) . Intriguingly , despite the slightly basic pH of the crystallization condition , sGC still adopted the post-fusion conformation and assembles as trimers around the crystallographic 3-fold axis , however in a different space group lattice ( Table 1 ) . Although individual domains superposition did not reveal significant differences ( Fig 5A ) we still observed some noteworthy differences in the post-fusion structure of PUUV sGCXF2 , particularly in the membrane proximal part of domain II including the fusion loop . In sGCXF2 this region has higher B-factor values than in the crystal form obtained at pH 6 . 0 ( Fig 5B and 5C ) . Domain II undergoes a hinge motion of 4 . 5° away from the three-fold axis in the Cα backbone with respect to the sGCXF1 structure , increasing the distances between the fusion loops by approximately 35% ( Fig 5D ) . Intriguingly , in sGCXF2 , E770 and R902 adopt different rotamers that do not allow the salt bridge to form that is in contrast to the β-barrel at the domain I-II interface which limits the hinge motion at that region , unlike other class II membrane fusion proteins , but similar to RVFV GC ( Fig 5A ) [25] . The absence of this inter-protomer salt-bridge plausibly contributes to the flexibility of the trimer at domain II membrane proximal region in the sGCXF2 structure ( Fig 5E ) . However the unaffected fusion activity of the R902A in our functional assay suggests that it is not mandatory for fusion activity ( S4B Fig ) . Indeed it was suggested before that there is no preferred distance between fusion loops of class II proteins required for fusion activity [49] . Finally , it was postulated that histidine residues function as pH sensors in class II membrane fusion proteins from flaviviruses [50–53] . We did not observe any significant differences in rotamers of histidine residues between the low and high pH crystal form . Furthermore , the poor sequence similarity between PUUV GC ( post fusion ) and RVFV GC ( pre-fusion ) shows no conserved histidines neither in sequence nor in three-dimension position ( S5B Fig ) implying that pH-sensing mechanism might be different in these two viruses . The total length of the stem region connecting between domain III and the TM domain is 46 residues ( 1069–1114 ) ( S2 Fig ) . To maximize solubility we included in our expression construct just the first 38 residues of the stem . However , in both our crystal structures ( sGCXF1 and sGCXF2 ) only the first eight residues of the stem ( 1069–1076 ) are visible in the electron density map , indicating either major flexibility or proteolytic cleavage at the C-terminus of sGC during preparation . Unlike flaviviruses , in which the stem has an α-helical structure [54] , or rubella virus in which the stem has a mixed α/β secondary structure content [34] , secondary structure prediction of the stem region from PUUV GC shows mostly random coil structure with a few residues predicted to be in β-strand conformation towards the TM domain ( S2 Fig , pink/gray arrow ) . This might resemble the rubella E1 C-terminal β-strand ‘n’ as it joins the i-j β-sheet of a neighboring protomer [34] . It is possible that the C-terminal part of the stem region of PUUV GC might extend the i-j β-sheet from domain II of the adjacent protomer and thereby might enhance the stability of the trimer . Most of the inter- and intramolecular contacts at that region of the stem of PUUV sGC are either main-chain/main-chain or main-chain/side-chain interactions ( Fig 6A ) . Interestingly , R1074 side chain at the N-terminal of the stem is inserted into a negatively charged cavity at the same protomer ( Fig 6A ) . The main-chain carbonyls of G883 , D884 , K893 and C894 create the cavity’s negative charge and lead the stem to a canyon formed by two adjacent protomers ( Fig 6A ) . In flaviviruses the domain III-proximal part of the stem participates in both , intramolecular contacts with domain II and intermolecular interactions with the adjacent protomer , in what that appears to be a late-stage fusion intermediate [55] . The resemblance of our stem region orientation to flavivirus E stem implies the same for PUUV GC . The stem region’s sequence is conserved among hantaviruses ( S2 Fig ) . A recently published work exploring the stem region characteristics in ANDV showed inhibition of fusion activity for stem peptides derived from the C-terminal half of the stem region but not for peptides that were derived from the N-terminal half ( domain III-proximal ) [56] . The nature of the stem interactions with domain II observed in our structure might explain a weak binding of such exogenous peptides . Nevertheless , the zipper-like contact that we observed for residues 1069–1076 is evidently strong enough to immobilize a covalently attached stem segment but apparently not to bind a soluble peptide . In Semliki forest virus ( genus: Alphavirus ) it was shown that no specific sequence of the stem region was required for membrane fusion [57] . R1074 ( R417 in Gc numbering ) is the only residue at the base of PUUV sGC stem that maintains side chain intramolecular contacts with domain II and it is highly conserved among hantaviruses ( except HNTV and SEOV where it is substituted with lysine of similar properties , S2 Fig ) . To investigate the role of R1074 in membrane fusion , we introduced an alanine substitution of R1074 in both , PUUV and ANDV GPC in order to test their activity in the available in vitro systems established mostly for ANDV [17 , 58] . The R1074A mutants of ANDV and PUUV GC were expressed as the wild type proteins , localized on the cell surface and assembled into virus like particles ( VLPs ) ( S4A Fig ) . However , despite being present on the cell surface , we found that the fusion index of the R1074 mutants from PUUV and ANDV dropped below 0 . 2 , indicating a strong impairment of the acid pH-triggered syncytia formation activity ( Fig 6B and 6C ) . The fact that the mutation of a conserved residue such as R1074 in hantavirus GC from PUUV and ANDV led to equivalent fusion activity results provides a direct proof for its high conservation among hantaviruses in both , structure and function . Therefore , this data imply that the PUUV GC structure can be used for rational design and characterization of mutations in different hantaviruses . In this context , and to further assess mechanistically the stage in which the R1074A mutant was arrested in the fusion process , we used the ANDV system to test acid-induced trimerization . Therefore , the wild type or R1074A mutant GC from ANDV was incorporated together with wild type GN into VLPs , that were collected and concentrated from the supernatants of cells expressing ANDV wild type or R1074A mutant GPC ( S4 Fig ) . The concentrated VLPs were then treated at pH 7 . 4 or pH 5 . 5 and the glycoproteins subsequently extracted by non-ionic detergent . Their sedimentation on sucrose gradients revealed that the R1074A mutant underwent trimerization at pH 5 . 5 as efficient as the wild type control ( Fig 6D ) . However , when the resistance of the trimer was tested for its stability by trypsin digestion , not only the neutral pH form , but also the acid-treated R1074A mutant was readily degraded by trypsin , in contrast to the low pH form of wild type GC ( Fig 6E ) . From these data it can be concluded that the R1074A mutant underwent acid-induced trimerization , but this trimer did not reach a stable post-fusion conformation . This difference in stability may be related to an incomplete fold-back of the stem region against the trimeric core . Combining our structural and functional data we conclude that the ‘base’ of the stem region in hantaviruses is essential for fusion through the formation of a stable post-fusion trimer . It was shown previously for class II membrane fusion proteins that the activity of small molecule inhibitors in an assay for infectivity correlates well with their capacity to compete with stem-derived peptides [59] . Schmidt and co-workers suggested that the conformational transition from a pre-fusion arrangement to a post-fusion trimer will require removal of the ligand , imposing a barrier to completion of the fusion process . For this reason , in silico screens found potential pocket-binding compounds , that in some cases yielded active inhibitors [60–63] . Thus the electrostatic interaction of R1074 in a well-defined cavity at the base of the stem region and our functional data showing its role in trimer stabilization and membrane fusion activity suggest that this cavity might be a target for small molecule fusion inhibitors . The existence of a class II fold in a virus family other than Flaviviridae and Togaviridae was already suggested to diverge either from a viral or a common cellular class II ancestor [25 , 35 , 64 , 65] . What are the driving forces that shaped the evolution of class II membrane fusion proteins ? To address this question we computed structure-based sequence alignment based on both , the full-length ectodomains and the individual domains of various class II membrane fusion protein structures and calculated the corresponding cladograms ( Fig 7 ) . As expected , cladograms based on the structures of the individual domains do not show significant difference in topology compared to the full-length-based cladogram . Despite the structural similarities of phlebovirus GC to flavivirus E proteins [25] , PUUV GC seems to be more structurally related to alphavirus E1 proteins ( S5A Fig ) . On the other hand , rubella virus ( RV ) E1 and PUUV GC appear to be more related in terms of the particle arrangement . Both assemble into pleomorphic virions with a non-continuous protein envelope with local symmetry properties in contrast to other viral class II membrane fusion proteins that are part of an icosahedral envelope arrangement [12 , 34] . Furthermore , while the viruses containing class II membrane fusion proteins assembled into icosahedral symmetry are all arthropod-borne , hantaviruses and RV are transmitted among mammals ( rodent-to-human and human-to-human , respectively ) . As suggested before for RV , a human-restricted infection cycle forced the virus to evolve unique structural features for its fusogen [34] . It is possible that hantaviruses followed a similar evolutionary path in mammals and further diverged to an additional branch separated from arboviruses containing class II membrane fusion proteins ( Fig 7 ) . Nonetheless , other evolutionary mechanisms such as convergent evolution , cannot be ruled out for this observation . Hopefully with the determination of more fusogens structures from the Bunyaviridae family the molecular basis for these proteins evolution will be more comprehensively studied .
The open reading frame encoding the ectodomain of GC ( sGC ) from PUUV ( M segment residues 659–1106 ) were amplified from the M segment cDNA of Puumala virus P360 strain ( GenBank accession code P41266 . 1 ) and subcloned into the pAcGP67 vector ( BD Biosciences ) in frame with the baculovirus gp67 signal sequence and a C-terminal eight-histidine purification tag . Sf9 insect cells ( Expression Systems ) were co-transfected with sGC expression constructs and linearized baculovirus genomic DNA ( Expression Systems ) to produce recombinant baculoviruses expressing sGC . Virus stocks were amplified with three sequential infections of Sf9 cells . For sGC expression , Tni insect cells ( Expression Systems ) grown at 27°C were infected at a density of 2 × 106 cells/ml with 1% ( v/v ) of third-passage ( P3 ) baculovirus stock . After culture in suspension for 96–108 h at 20°C the culture media was collected and its pH was adjusted with addition of Tris pH 8 . 0 to final concentration of 20 mM . Following media concentration , secreted sGC was purified by nickel affinity chromatography ( Ni-NTA agarose , QIAgen ) . A subsequent anion-exchange chromatography purification step ( monoQ , GE Healthcare ) resulted in two populations of sGC eluting in different salt concentrations . From this point on the two populations ( termed sGCXF1 and sGCXF2 ) were separated and further went through the same steps . The His-tag was subsequently removed with carboxypeptidase A ( CPA ) treatment at 4°C for 16 h ( 1 mU CPA per microgram of sGC ) . CPA was then inhibited with 1 mM EDTA and 1 mM 1 , 10-phenanthroline and separated from sGC by size-exclusion chromatography ( Superdex 200 10/300 GL , GE Healthcare ) . Protein samples were concentrated to 2 . 5–3 . 5 g/l , frozen in liquid nitrogen and stored at -80°C in 10 mM Tris pH 8 , 0 . 1 M NaCl . Crystals of sGCXF1 ( eluted from the mono-Q at low salt concentration ) were grown by hanging drop vapor diffusion at 16°C . sGCXF1 at 2 . 4 g/l in 10 mM Tris pH 8 . 0 , 0 . 1 M NaCl was mixed in 2:1 protein to reservoir containing 12% ( w/v ) polyethylene glycol 2000 mono-methyl ether ( PEG 2000 MME ) , 0 . 1 M MES pH 6 . 0 and 0 . 2 M ammonium sulfate . Multi-crystals clusters appeared after 3–5 weeks and very few single crystals were observed after 6–8 weeks . A single crystal was then crushed and used as microseeds in drops pre-equilibrated for 24 h prior to seeding . Rhombohedron shaped crystals reached a size of 150 × 70 × 70 μm 7–10 days post-seeding and belonged to space group R32 . Crystals were frozen in liquid nitrogen in reservoir solution supplemented with 30% PEG 400 as a cryoprotectant . Derivative sGCXF1 crystals were obtained by soaking in reservoir solution plus 1 mM methyl mercury phosphate ( Hampton Research ) for one week . sGCXF2 crystals appeared in 40% ( w/v ) polyethylene glycol 400 ( PEG 400 ) , 0 . 1 M Tris pH 8 . 0 , 0 . 2 M lithium sulfate ( 1:1 protein to reservoir ratio ) . After 12 weeks sharp-edges cubic crystals were observed and reached a size of 60 × 60 × 60 μm . Upon optimization , crystals with cubic morphology at the size of 75 × 75 × 75 μm appeared after 4–6 weeks and belonged to space group I213 . Data were collected at 100 K on a PILATUS detector ( Dectris ) and processed with XDS [66] . The structure of sGCXF1 was determined by single isomorphous replacement with anomalous signal ( SIRAS ) with PHENIX [67] . Initial atomic coordinates for sGCXF1 built with PHENIX were used as starting model in refinement and building cycles with the highest resolution ( 1 . 84 Å ) native data set . The atomic model was completed with COOT [68] and refined to an Rfree of 21% with PHENIX and REFMAC [69] . The structure of sGCXF2 was determined by molecular replacement using domains I+III and domain II of sGCXF1 as separate search models . Atomic coordinates and structure factors for sGCXF1 and sGCXF2 have been deposited in the Protein Data Bank ( ID codes 5J81and 5J9H , respectively ) . See Table 1 for data collection and refinement statistics . All molecular graphics were produced using PyMol ( PyMOL Molecular Graphics System , Version 1 . 8 Schrödinger , LLC ) . Molecular surface calculations were performed using UCSF Chimera [70] . Surface electrostatic potential was calculated with APBS [71] . B-factor analysis was calculated using baverage module in the CCP4 suite [72] . Surface conservation was calculated using CONSURF server ( http://consurf . tau . ac . il/ ) [73] . PUUV sGC structure was superimposed on the crystal structure of Semliki forest virus E1 in its pre-fusion state ( PDB ID codes 2ALA ) using domains I+II and domain III as two separate rigid bodies . The flexible linker between domain I and domain III was eliminated from the model . Analytical size-exclusion chromatography and multiangle light scattering ( MALS ) experiments were performed in 20 mM sodium acetate pH 5 . 0 , or Tris⋅HCl pH 8 . 0 and 0 . 1 M NaCl . A total of 0 . 2 mL sGC at 2 . 5 g/L was loaded onto a Superdex 200 ( 10/300 ) column coupled to mini DAWN TREOS spectrometer and Optilab T-rEX ( Wyatt technology ) refractometer at a flow rate of 0 . 7 mL/min . PUUV sGC was detected as it eluted from the column with a UV detector at 280 nm , a light scattering detector at 690 nm , and a refractive index detector . The molar mass of PUUV sGC was determined from the Debye plot of light scattering intensity versus scattering angle . Data processing was performed with ASTRA software ( Wyatt Technology ) . Mutations were introduced into the expression vectors pI . 18/ANDV-GPC [74] and pWRG/PUUV-M ( s2 ) ( kindly provided by Jay Hooper , USAMRIID , USA ) [75] coding for GPC from ANDV strain Chi-7913 and PUUV strain K27 ( GenBank accession numbers AAO86638 and L08754 ) , respectively , by using DNA synthesis and sub-cloning into the corresponding expression vectors ( Genscript ) . For expression and localization analysis , 8 μg of plasmids were calcium-transfected into 293FT cells ( Invitrogen ) grown on 100 mm plates and 48 hrs later , proteins located on the cell surface were biotinylated using a cell-surface protein isolation kit ( Pierce ) , and the fractions corresponding to intracellular and surface proteins separated on a neutravidin resin . The presence of Gc and β-actin in each fraction were analyzed by western blot using anti-Gc 2H4/F6 [76] and anti-β-actin ( Sigma ) MAb at a 1:2 , 500 dilution . Primary antibodies were detected by chemiluminsecence using anti-mouse immunoglobulin horseradish peroxidase conjugate ( Thermo Fisher Scientific ) . To prepare VLPs , a previously established protocol was used [17] . Briefly , 48 hrs post-transfection the supernatant of 293FT cells transfected with wild type or mutant pI . 18/ANDV-GPC or pWRG/PUUV-M ( s2 ) constructs was collected and VLPs concentrated by ultracentrifugation for 1 hr at 135 , 000 g . The presence of VLPs was assayed by western blot analysis as described above . A fluorescence-based syncytia assay was performed as reported before ( 32 ) . Vero E6 cells ( ATTC ) seeded in 16-well chamber slides were transfected with 0 . 5 μg of wild type or mutant pI . 18/ANDV-GPC or pWRG/PUUV-M ( s2 ) constructs using lipofectamin 2000 ( Invitrogen ) . 48 hrs later , the cells were incubated for 5 min at 37°C with MEM culture media adjusted to the corresponding pH . Next , incubation of cells was continued for 3 hrs at 37°C in neutral pH MEM culture media . To label the cell cytoplasm , cells were subsequently incubated for one hr with 1 μM 5-chloromethylfluorescein diacetate ( Cell Tracker CMFDA , Molecular Probes ) . Subsequently , cells were fixed with 4% ( w/v ) paraformaldehyde , permeabilized with 0 . 1% ( v/v ) Triton X-100 and Gc detected with anti-Gc 2H4/F6 MAb and anti-mouse immunoglobulin MAb Alexa555 conjugate ( Invitrogen ) . Cell nuclei were stained with DAPI 1 ng/μl in PBS . To visualize syncytia samples were examined under a fluorescence microscope ( BMAX51; Olympus ) and pictures taken for quantification ( ProgRes C3; Jenoptics ) . The fusion index of Gc-expressing cells was calculated using the formula: 1- [number of cells/number of nuclei] . For each sample approximately 200 nuclei per field were counted ( 200 x magnification ) and the mean fusion index of five fields calculated from at least two independent experiments . Acid-induced Gc trimerization was tested by sucrose sedimentation using a previous protocol [17] . Briefly , VLPs were incubated for 30 min at the indicated pH to allow for Gc conformational changes . The pH back-neutralized , and Triton X-100 ( 0 , 5%; v/v ) -extracted glycoproteins were subsequently loaded to the top of a sucrose step gradient ( 7–15% , w/v ) . After 16 hrs of centrifugation at 150 , 000 g , fractions were collected and the presence of Gc in each fractions tested by western blot analysis . The stability of neutral pH and acid pH conformation of a Gc mutant was assayed by its resistance to trypsin as shown for wild type Gc previously [17] . In brief , VLPs including wild type or mutant Gc were incubated at the indicated pH and presence of Gc assessed by western blot as described above . A set of structures of class II fusion proteins in their post fusion conformation was obtained from the DALI server [77] with the atomic coordinates of PUUV sGC as the query . Structures of viral class II fusion proteins and of C . elegans EFF-1 were aligned with the MUSTANG server [78] . The resulting structure-based sequence alignment was used for the estimation of the cladogram by the neighbor-joining method with the BLOSUM62 substitution matrix using Jalview [79 , 80] . The same process was further executed on individual domains .
|
Hantaviruses ( family: Bunyaviridae ) encompass pathogens responsible to serious human diseases and economic burden worldwide . Following endocytosis , these enveloped RNA viruses are directed to an endosomal compartment where a sequence of pH-dependent conformational changes of the viral envelope glycoproteins mediates the fusion between the viral and endosomal membranes . The lack of high-resolution structural information for the entry of hantaviruses impair our ability to rationalize new treatments and prevention strategies . We determined the three-dimensional structure of a glycoprotein C from Puumala virus ( PUUV ) using X-ray crystallography . The two structures ( at pH 6 . 0 and 8 . 0 ) were determined to 1 . 8 Å and 2 . 3 Å resolutions , respectively . Both structures reveal a class II membrane fusion protein in its post-fusion trimeric conformation with novel structural features in the trimer assembly and stabilization . Our structures suggest that neutralizing antibodies against GC target its conformational changes as inhibition mechanism and highlight new molecular targets for hantavirus-specific membrane fusion inhibitors . Furthermore , combined with the available structures of other class II proteins , we remodeled the evolutionary relationships between virus families encompassing these proteins .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"crystal",
"structure",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"condensed",
"matter",
"physics",
"microbiology",
"viral",
"structure",
"viruses",
"membrane",
"fusion",
"hantavirus",
"rna",
"viruses",
"andes",
"virus",
"protein",
"structure",
"crystallography",
"cellular",
"structures",
"and",
"organelles",
"bunyaviruses",
"solid",
"state",
"physics",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"cell",
"membranes",
"physics",
"protein",
"structure",
"comparison",
"biochemistry",
"macromolecular",
"structure",
"analysis",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms",
"cell",
"fusion"
] |
2016
|
Crystal Structure of Glycoprotein C from a Hantavirus in the Post-fusion Conformation
|
Schistosomiasis affects over 200 million people and there are concerns whether the current chemotherapeutic control strategy ( periodic mass drug administration with praziquantel ( PZQ ) —the only licenced anti-schistosome compound ) is sustainable , necessitating the development of new drugs . We investigated the anti-schistosome efficacy of polypyridylruthenium ( II ) complexes and showed they were active against all intra-mammalian stages of S . mansoni . Two compounds , Rubb12-tri and Rubb7-tnl , which were among the most potent in their ability to kill schistosomula and adult worms and inhibit egg hatching in vitro , were assessed for their efficacy in a mouse model of schistosomiasis using 5 consecutive daily i . v . doses of 2 mg/kg ( Rubb12-tri ) and 10 mg/kg ( Rubb7-tnl ) . Mice treated with Rubb12-tri showed an average 42% reduction ( P = 0 . 009 ) , over two independent trials , in adult worm burden . Liver egg burdens were not significantly decreased in either drug-treated group but ova from both of these groups showed significant decreases in hatching ability ( Rubb12-tri—68% , Rubb7-tnl—56% ) and were significantly morphologically altered ( Rubb12-tri—62% abnormal , Rubb7-tnl—35% abnormal ) . We hypothesize that the drugs exerted their activity , at least partially , through inhibition of both neuronal and tegumental acetylcholinesterases ( AChEs ) , as worms treated in vitro showed significant decreases in activity of these enzymes . Further , treated parasites exhibited a significantly decreased ability to uptake glucose , significantly depleted glycogen stores and withered tubercules ( a site of glycogen storage ) , implying drug-mediated interference in this nutrient acquisition pathway . Our data provide compelling evidence that ruthenium complexes are effective against all intra-mammalian stages of schistosomes , including schistosomula ( refractory to PZQ ) and eggs ( agents of disease transmissibility ) . Further , the results of this study suggest that schistosome AChE is a target of ruthenium drugs , a finding that can inform modification of current compounds to identify analogues which are even more effective and selective against schistosomes .
More than half a billion people are at risk of contracting schistosomiasis ( bilharzia ) , a neglected tropical parasitic disease that is endemic in more than 75 countries in Africa , Asia , and South America , where over 200 million individuals are infected and approximately 280 , 000 die every year [1 , 2] . Schistosomiasis is caused by infection with blood flukes of the genus Schistosoma , and disease results from the deposition of eggs in host tissues , leading to granuloma formation that can cause fibrosis , portal hypertension , and even death [3] . Despite this considerable disease burden , to date there are no vaccines against schistosomiasis [4] . Praziquantel ( PZQ ) remains the only effective frontline drug to treat the parasite despite its shortcomings , which include ineffectiveness against juvenile stages of the worm [5] , poor efficacy in treating pre-patent infections [6] , reports of reduced efficacy in field studies [7] and the inevitable risk of the development of resistant strains in response to periodic mass drug administration [8] . Since there are no alternative effective drug treatments for these parasites , new therapies are urgently required . Among potential targets for chemotherapy are acetylcholinesterases ( AChEs ) . These enzymes catalyze the rapid breakdown of the neurotransmitter acetylcholine ( ACh ) in both central and peripheral nervous systems of eukaryotic organisms , and so control neuronal function [9] . In addition to controlling cholinergic synapses , the enzyme is present in large amounts on the tegument of schistosomes [10] , where it has been implicated to play a role in the regulation of host glucose uptake by the parasite by limiting the interaction of host ACh with tegumental nicotinic ACh receptors ( nAChRs ) [11] . These receptors are associated both spatially and temporally with surface AChE expression and are concentrated on the tegument [12] , the major site of glucose uptake [13] . Evidence for this relationship is shown by the ablation of glucose uptake with a membrane-impermeable inhibitor of AChE ( which has the same result as the administration of an excess of ACh ) or specific agonists of nAChRs . The interaction of ACh with tegumental nAChRs is thought to decrease the amount of glucose uptake through surface glucose transporters but the specific mechanism for this is not known [11] . With respect to its termination of synaptic transmission , inhibition of AChE produces an excess accumulation of ACh and overstimulation of its receptors , causing uncoordinated neuromuscular function that often results in death due to respiratory paralysis [14] . As such , AChE inhibitors are widely used as pesticides [15] and anthelmintics [16] . Indeed , metrifonate , an organophosphorus AChE inhibitor originally used as an insecticide has also been used for the treatment of schistosomiasis until it was withdrawn from the market and further development because of off-target toxicity [17] . In addition to organophosphates , mono-nucleated chemical complexes of the transition metal ruthenium have been shown to target and inhibit enzymes such as AChE [18] , and there are numerous recent studies documenting the efficacy of polypyridylruthenium ( II ) complexes against a variety of different microbial pathogens [19–21] . Unlike their organophosphorus counterparts , ruthenium complexes are speculated to exert their inhibitory effects through a combination of electrostatic and hydrophobic interactions at the peripheral anionic ( PAS ) site of AChE , which is located at the gate of the enzyme’s catalytic gorge [22] , and not through direct interaction with the active site . Ruthenium complexes are thought to be less toxic to human cells than small-molecule inhibitors because of this mode of inhibition and also because the overall neutral charge in the outer membrane leaflet of eukaryotic cells [23] creates a greatly reduced capacity for electrostatic interaction with the metal compounds [24] . Herein , we demonstrate the AChE-inhibitory action of two mononuclear and a series of di- , tri- and tetra-nuclear polypyridylruthenium ( II ) complexes linked by the bis[4 ( 4’-methyl-2 , 2’-bipyridyl ) ]-1 , n-alkane ligand ( “bbn”; n = 7 , 10 , 12 and 16 ) against extracts of Schistosoma mansoni and Schistosoma haematobium and both adult and juvenile S . mansoni parasites in vitro . We also provide evidence consistent with the capacity of these complexes to disrupt the parasite’s glucose uptake ability through the inhibition of tegumental AChE , a cholinergic pathway unique to schistosomes [11] . Finally , we show the in vivo efficacy of two ruthenium complexes in a mouse model of schistosomiasis , providing evidence that drugs based on these compounds could be a valuable addition to the chemotherapeutic arsenal against this debilitating disease .
[Ru ( phen ) 2 ( Me2bpy ) ]2+ and the mononuclear ( Rubbn-mono ) , dinuclear ( Rubbn-di ) , trinuclear ( Rubbn-tri ) , tetranuclear linear ( Rubbn-tl ) and tetranuclear non-linear ( Rubbn-tnl ) polypyridylruthenium ( II ) complexes ( Fig 1 ) were synthesised using the appropriate bis[4′- ( 4-methyl-2 , 2′-bipyridyl ) ]-1 , n-alkane bridging ligand ( bbn ) as previously described [19] . Compounds were dissolved in H2O at stock concentrations of 1 mM . S . mansoni cercariae were shed from infected Biomphalaria glabrata snails ( Biomedical Research Facility , MD , USA ) by exposure to light at 28°C for 2 hours . Cercariae were used to infect 6–8 week old male BALB/c mice ( Animal Resources Centre , WA , Australia ) by tail penetration ( 180 S . mansoni cercariae/mouse ) and adults were harvested by vascular perfusion at 7 weeks post-infection [25] . S . mansoni eggs were purified from infected mouse livers as previously described [26] and were used in the xWORM egg hatching assay or to make soluble egg antigen ( SEA ) [27] . For experiments involving schistosomula , S . mansoni cercariae were mechanically transformed as previously described [27] . To make PBS-soluble extracts , S . mansoni adult worms were homogenized in PBS ( 50 μl/worm pair ) at 4°C using a TissueLyser II ( Qiagen ) and the supernatant collected by centrifugation at 15 , 000 g for 60 mins at 4°C . Triton X-100-soluble extracts of S . mansoni and S . haematobium were made in the same way except worms were lysed in buffer containing 1% Triton X-100 , 40 mM Tris-HCl , pH 7 . 4 . Protein concentration was determined using the Pierce BCA Protein Assay kit ( Thermofisher ) , aliquoted and stored at -80°C until use . AChE , nucleotide pyrophosphatase-phosphodiesterase 5 ( SmNPP-5 ) and alkaline phosphatase ( AP ) activity in Triton X-100-soluble adult worm extracts and AChE activity in SEA were determined in a Polarstar Omega microplate reader ( 200 μl final volume in 96-well plates ) . AChE activity was determined using the Ellman method [28]; extracts were serially diluted ( 20–5 μg ) in AChE assay buffer ( 0 . 1 M sodium phosphate , pH 7 . 4 ) , 2 mM acetylthiocholine ( AcSCh ) and 0 . 5 mM 5 , 5’-dithio-bis ( 2-nitrobenzoic acid ) ( DTNB ) were added and absorbance was measured at 405 nm every 10 min for 5 h at 37°C . Specific activity was calculated using the initial velocity of the reaction . For AChE inhibition assays , parasite extracts equal to a specific activity of 0 . 55 nmol/min/well were diluted in AChE assay buffer to a final volume of 170 μl and pre-incubated with ruthenium complexes ( 10 nM– 100 μM ) for 20 min at RT . AcSCh and DTNB were added at 2 mM and 0 . 5 mM , respectively and absorbance was measured at 405 nm every 10 min for 5 h at 37°C . SmNPP-5 activity [29] was measured by serially diluting extracts in SmNPP-5 assay buffer ( 50 mM Tris-HCl , pH 8 . 9 , 120 mM NaCl , 5 mM KCl , 60 mM glucose ) , adding 0 . 5 mM p-nitrophenyl thymidine 5′-monophosphate ( p-Nph-5’-TMP ) and reading the absorbance ( 405 nm ) every 10 min for 5 h at 37°C . Specific activity was calculated using the initial velocity of the reaction . For SmNPP-5 inhibition assays , parasite extract equal to 32 nmol/min/well was diluted in SmNPP-5 assay buffer to a final volume of 180 μl and pre-incubated with ruthenium complexes ( 10 nM– 100 μM ) for 20 min at RT . Substrate ( p-Nph-5’-TMP ) was added to 0 . 5 mM and absorbance was measured at 405 nm every 10 min for 5 h at 37°C . AP activity [30] was measured by serially diluting extracts in AP assay buffer ( 0 . 1 M glycine , pH 10 . 4 , 1 mM MgCl2 , 1 mM ZnCl2 ) , adding 2 mM p-nitrophenyl phosphate ( pNPP ) and measuring absorbance ( λ = 405 nm ) every 2 min for 1 h at 37°C . Specific activity was calculated using the initial velocity of the reaction . For AP inhibition assays , parasite extract equal to 1 nmol/min/well was diluted in AP assay buffer to a final volume of 180 μl and pre-incubated with ruthenium complexes ( 10 nM– 100 μM ) for 20 min at RT . Substrate ( pNPP ) was added to 2 mM and absorbance was measured at 405 nm every 2 min for 1 h at 37°C . For all assays , inhibition for each sample was calculated relative to the negative control ( reactions without ruthenium complexes ) and reactions were performed in duplicate with data presented as the average ± SEM . Newly-transformed schistosomula ( 100 parasites ) were cultured ( 37°C , 5% CO2 ) in 200 μl of Basch medium supplemented with 4× antibiotic/antimycotic ( AA—200 units/ml penicillin , 200 μg/ml streptomycin and 0 . 5 μg/ml amphotericin B ) in a 96 well plate in the presence of a series of serially-diluted ruthenium complexes ( 100 μM– 10 nM ) . After 48 h , parasite viability was assessed microscopically by trypan blue exclusion staining as previously described [31] . Six of the most effective compounds were tested again for their larvacidal efficacy ( 100 schistosomula per treatment ) ; this time at concentrations of 200 , 100 , 50 , 25 , 12 . 5 and 6 . 25 μM . Experiments were performed in duplicate with IC50 data presented as the average ± SE . Five pairs of adult worms were cultured in 2 ml of Basch medium supplemented with 4x AA in a 24 well plate in the presence of ruthenium complexes ( 50 μM ) . Control worms were treated with an equal amount of H2O . Worms were cultured at 37°C and 5% CO2 for 7 days , monitored every 24 h for motility by microscopic examination and considered dead if no movement was seen . The most effective ruthenium complexes ( 5 compounds ) were tested in duplicate ( five pairs of worms each ) at 10 , 50 and 100 μM . Data is presented as the average of each duplicate experiment ± SE . Egg hatching was evaluated by the xWORM egg hatching assay [32] . Ova ( 5 , 000 per well , 200 μl reaction volume ) were incubated in 0 . 1x PBS , pH 7 . 2 , containing ruthenium complexes ( 50 μM ) and induced to hatch under bright light at RT for 16 h . The motility of the miracidia released from the hatched eggs was monitored every 15 s and the motility index was calculated as described [32] . Experiments were performed in triplicate with control reactions containing no ruthenium compounds . To investigate the effects of ruthenium complexes on egg development , triplicate sets of five pairs of adult S . mansoni worms were cultured in Basch media with or without 5 μM Rubb12-tri , for 72 h . The eggs released into the media were counted and misshapen , malformed or immature eggs [33] were scored as “abnormally developed” . Egg hatching and morphology data are presented as the average of each triplicate experiment ± SE . Freshly perfused worms were cultured in the presence of sub-lethal concentrations ( 5 μM ) of Rubb12-tri or Rubb16-tnl—two ruthenium compounds determined to be most effective in terms of their combined activity against schistosomula , adult worms and eggs—in Basch medium at 37°C and 5% CO2 . To measure surface AChE or AP activity , 5 pairs of worms ( preliminary experiments by us showed this number of parasites was sufficient to accurately measure surface enzyme activity ) were transferred to either AChE assay buffer ( 0 . 1 M sodium phosphate , pH 7 . 4 , 2 mM AcSCh , 0 . 5 mM DTNB ) or AP assay buffer ( 0 . 1 M glycine , pH 10 . 4 , 1 mM MgCl2 , 1 mM ZnCl2 , 2mM p-nitrophenyl phosphate ) . Surface enzyme activities were quantified by measuring the absorbance ( λ = 405 nm ) after incubation for 60 min ( AChE ) or 30 min ( AP ) . Activity was measured from triplicate sets of parasites ( 5 pairs of worms ) and each assay was technically replicated three times . For each enzyme assay , activities of drug-treated parasites were expressed relative to worms cultured without ruthenium complexes ( negative controls ) . To measure somatic AChE activity , PBS-soluble extracts were made from worms used for surface AChE activity ( triplicate sets of five pairs of worms ) and then assayed in triplicate as described above . Data is presented as the average of each triplicate biological and technical experiment ± SE . Five pairs of freshly-perfused worms were cultured in the presence of sub-lethal concentrations ( 5 μM ) of Rubb12-tri or Rubb16-tnl in DMEM ( 1000 mg/l glucose ) . Media ( 20 μl ) from each experiment was collected after 24 h and the amount of glucose was quantified using a colorimetric glucose assay kit ( Sigma ) according to the manufacturer’s instructions . Media was collected from triplicate sets of parasites ( five pairs of worms ) and each assay was replicated 3 times . Glucose levels were expressed relative to media collected from worms which received no drug ( negative control ) . Data is presented as the average of each triplicate biological and technical experiment ± SE . To measure the glycogen content of worms treated with ruthenium drugs , Triton X-100-soluble extracts were made and assayed for glycogen in a modified procedure described by Gomez- Lechon et al . [34] . Briefly , 0 . 2 M sodium acetate , pH 4 . 8 , was added to 30 μg parasite extract and 50 μl glucoamylase ( 10 U/ml ) to make a reaction volume of 150 μl . The mixture was incubated at 40°C for 2 h with shaking at 100 rpm , 40 μl added to a new microplate with 10 μl 0 . 25 M NaOH and the amount of glucose quantified using the colorimetric glucose assay kit . Extracts were made from triplicate sets of parasites ( five pairs of worms ) and assays were performed three times . Data is presented as the average of each triplicate biological and technical experiment ± SE . Control parasites and worms treated with 5 μM Rubb12-tri were prepared for scanning electron microscopy by fixation in 3% glutaraldehyde followed by successive dehydration for 15 mins each in a graded ethanol series ( 100% , 90% , 80% , 70% , 60% , 50% ) , 1:1 ethanol:hexamethyldisilizane ( HMDS ) and , finally , 100% HMDS . Dehydrated worms were covered and left overnight in a fume hood to allow the HMDS to evaporate then mounted on an aluminium stub , sputtered with gold and visualized using a JEOL JSM scanning electron microscope operating at 10 kV . Images were acquired digitally using Semaphore software . Cytotoxicity assays were performed using the mitochondrial-dependent reduction of 3- ( 3 , 4-dimethylthiazol-2yl ) -5-diphenyl tetrazolium bromide ( MTT ) to formazan as described by Pandrala et al . [35] . The human bile duct cell line H69 was cultured in 96-well microtiter plates containing 0 . 1 ml of growth factor-supplemented media ( DMEM/F12 with high glucose ( 4 mg/ml ) , 10% FCS , 1×AA , 25 μg/ml adenine , 5 μg/ml insulin , 1 μg/ml epinephrine , 8 . 3 μg/ml holo-transferrin , 0 . 62 μg/ml hydrocortisone , 13 . 6 ng/ml triiodo-1-thyronine ( T3 ) and 10 ng/ml epithelial growth factor ( EGF ) ) [36] to a cell density of 5 , 000 per well at 37°C in 5% CO2 . Cell viability was assessed after continuous exposure to a concentration series ( 50 , 25 , 10 , 5 , 1 , 0 . 5 and 0 . 1 μM ) of Rubb12-tri , Rubb16-tnl , PZQ or dichlorvos—a metabolite of the anti-schistosome AChE inhibitor metrifonate—for 72 h . The amount of reduced MTT to formazan within the cells was quantified by measuring the absorbance at λ = 550 nm . Data are the average of six replicate experiments ± SE . In order to determine the maximum tolerated dose of Rubb12-tri and Rubb7-tnl to be administered to mice infected with S . mansoni , five intravenous ( i . v . ) doses ( tail vein ) of Rubb12-tri and Rubb7-tnl were given to groups of three male BALB/c mice ( 6–8 weeks ) daily for five consecutive days . The doses ranged from 0 . 25 to 10 mg/kg in PBS and were administered in a volume of 30 μl . Animals were closely monitored for adverse clinical signs throughout the study and mice showing adverse effects were euthanized using CO2 asphyxiation . The highest dose that did not cause any adverse clinical signs was considered to be the maximum tolerated dose ( MTD ) for five consecutive daily doses . Groups of 8 male BALB/c mice ( 6–8 weeks ) were infected with S . mansoni cercariae as described above . At 35 days post-infection , groups were given five consecutive daily i . v . doses ( tail vein—30 μl ) of the MTD of either Rubb12-tri ( 2 mg/kg in PBS ) or Rubb7-tnl ( 10 mg/kg in PBS ) . PBS was similarly administered to the control group . Two independent trials were performed . Parasites were harvested by vascular perfusion at 49 days post-infection and the average worm burden per mouse for each group of mice ( trial 1 PBS control—n = 8 mice , trial 1 Rubb12-tri-treated—n = 8 mice , trial 1 Rubb7-tnl-treated—n = 6 mice , trial 2 PBS control—n = 7 mice , trial 2 Rubb12-tri-treated—n = 7 mice , trial 2 Rubb7-tnl-treated—n = 8 mice ) was calculated . Data is presented as a combination of the two independent trials ± SE . Livers from each group ( trial 1 PBS control—n = 8 mice , trial 1 Rubb12-tri-treated—n = 8 mice , trial 1 Rubb7-tnl-treated—n = 6 mice , trial 2 PBS control—n = 7 mice , trial 2 Rubb12-tri-treated—n = 7 mice , trial 2 Rubb7-tnl-treated—n = 8 mice ) were collected , halved and weighed , with one half digested with 5% KOH to determine liver eggs per gram of tissue ( epg ) as previously described [37] . The other half of each liver was pooled according to group , homogenized in H2O and placed in identical foil-covered volumetric flasks under bright light to hatch eggs released from the livers . After 1 h , the number of miracidia in 10 x 50 μl aliquots of H2O ( sampled from the extreme top of each flask ) were counted . The amount of eggs in each flask at the start of the hatching experiment was determined by liver epg calculations , allowing the egg hatching index of each group to be calculated by expressing the hatched eggs ( miracidia ) as a percentage of the total eggs . Data presented is for trial 1 only and represents the average of ten counts ± SE . To assess fitness of parasites recovered from mice treated with ruthenium complexes compared to controls , worms recovered from each group were pooled and five pairs were assayed for surface enzyme activity ( Sm-AChE , SmNPP-5 and Sm-AP ) or glucose uptake ability as described above . Somatic Sm-AChE activity was determined from homogenates made from the worms assayed for surface Sm-AChE activity , also as described above . Each assay was technically replicated three times and data is presented as the average of triplicate technical replicates of both trials combined ± SE . To compare eggs released from parasites recovered from mice treated with ruthenium complexes and controls , triplicate sets of worms ( five pairs ) from each pool were incubated in Basch media at 37°C in 5% CO2 for 72 h . The number of eggs released were counted and scored on the basis of development and morphology [33] by visualization under a FITC filter on a Zeiss AxioImager-M1 fluorescent microscope . Data presented is for trial 2 and is the average of triplicate experiments ± SE . Statistical analyses were performed using Graphpad Prism 7 . Inhibition curves and IC50 values were generated using sigmoidal dose-response ( variable slope ) with a non-linear fit model . One-way ANOVA with Dunn’s multiple comparison was used to determine significance ( p ) , which was set at 0 . 05 . In the case where only two groups were compared , student’s t test was used . The James Cook University ( JCU ) animal ethics committee approved all experimental work involving animals ( ethics approval number A2271 ) . Mice were raised in cages in the JCU quarantine facility for the duration of the experiments in compliance with the 2007 Australian Code of Practice for the Care and use of Animals for Scientific Purposes and the 2001 Queensland Animal Care and Protection Act . All reasonable efforts were made to minimise animal suffering .
A series of ruthenium complexes of different nuclearity ( mono- , di- , tri- and tetra-linear and tetra-nonlinear ) and with different chain lengths in the linking ligand ( bb7 , bb10 , bb12 , bb16 ) were screened ( 1 μM ) for AChE inhibitory activity in Triton X-100-soluble extracts from adult S . mansoni and S . haematobium and S . mansoni soluble egg antigen ( SEA ) ( Table 1 ) . In S . mansoni extracts , all tri- and tetra-nuclear complexes inhibited AChE activity by 70–90% and 7 of the 13 compounds had IC50 values ≤ 1 μM . A dose-response curve and Lineweaver-Burk plot is shown for the most potent of these complexes ( Rubb12-tri , IC50 = 0 . 3 μM ) ( Fig 2 ) . Interestingly , a different pattern of inhibition by the ruthenium complexes was observed against AChE activity in S . haematobium extracts with the IC50 values being considerably more varied than was observed for AChE activity in S . mansoni adult extracts , and not all compounds showed correlated potency between the two species . In addition , there was more variability among the tri- and tetra-nuclear complexes with the more lipophilic complexes ( e . g . Rubb10-tri and Rubb12-tri ) having stronger inhibitory activity . Rubb12-mono , Rubb12-tri and Rubb12-tnl showed greater activity and the IC50 values of these complexes were less than 1 μM . Overall , ruthenium compounds displayed a similar pattern of inhibition against AChE activity in S . mansoni egg versus adult extracts , although most complexes showed a stronger inhibitory capacity towards AChE activity in SEA with three compounds ( Rubb12-tri , Rubb12-tl and Rubb7-tnl ) achieving >95% inhibition . In order to examine the selectivity for AChE , the series of ruthenium complexes was screened ( 10 μM ) against S . mansoni extract for inhibition of two major tegumental enzymes—the phosphodiesterase SmNPP-5 and alkaline phosphatase ( AP ) . None of the compounds strongly inhibited activity of either enzyme at 10 μM , a tenfold higher concentration than was used for the AChE inhibition assays ( S1 Table ) . The entire series of Ru complexes were screened for their larvacidal activity against S . mansoni schistosomula with IC50 values calculated for the most effective compounds in a separate experiment . ( Table 2 ) . S . mansoni worms were cultured in the presence of 50 μM of each ruthenium complex to investigate their effectiveness in killing adult parasites , which was assessed by lack of motility . The effects of selected compounds on worm survival is shown in Fig 3A . Similar to the enzyme inhibition in parasite extracts , the tri-and tetra-nuclear complexes were the most effective compared to the mono- and di-nuclear complexes . The killing ability increased with the increasing number of ruthenium centres in the complex , and the tetra-nonlinear complexes were more active in comparison with their linear counterparts . As with the schistosomula killing experiment , the most effective compounds ( five ) were tested again , this time at concentrations of 100 μM , 50 μM and 10 μM ( Fig 3B ) . All tri-and tetra-nuclear complexes ( 10 μM ) killed 100% of the parasites in six days . Treatment with the ruthenium complexes induced significant changes in the gross morphology of the parasites ( Fig 3C ) . In particular , a tight coiling of the treated worms was observed . S . mansoni egg hatching in the presence ( 50 μM ) of ruthenium complexes was investigated by measuring the motility index of hatched miracidia from eggs using the xWORM assay ( Fig 4A ) . Significant reduction in hatching/motility was observed for 9/13 compounds tested . Rubb12-tnl was the most effective , reducing the motility index by 67% ( P < 0 . 0001 ) . To analyze the effect of ruthenium complexes on egg development , the eggs released from worms incubated for 3 days in the presence of 5 μM Rubb12-tri were scored for morphology . Eggs released from treated worms were abnormally developed ( immature or misshapen ) compared to controls ( Fig 4B; P < 0 . 01 ) . Adult worms were cultured in the presence of sub-lethal concentrations ( 5 μM ) of Rubb12-tri or Rubb16-tnl—the two ruthenium compounds deemed to be most effective at killing adult parasites—for 24 h and then examined for changes in surface and somatic AChE activity and glucose uptake ( given this pathway can be ablated by an organophosphorus AChE inhibitor ) . Treated worms showed significantly decreased levels of surface and somatic AChE activity in the presence of each complex ( Fig 5A and 5B ) however , consistent with enzyme inhibition experiments using parasite extracts , AP activity was not significantly affected ( Fig 5C ) . The amount of glucose in the media of parasites treated with either complex was significantly higher than control worms ( Fig 6A ) , suggestive of impaired glucose uptake in the presence of ruthenium compounds . Consistent with these results , extracts of treated parasites had a significantly lower glycogen content compared to controls ( Fig 6B ) . Moreover , scanning electron micrographs of the tegument of male parasites treated with Rubb12-tri or Rubb16-tnl showed the dorsal tubercules ( a site of glycogen storage [38] ) to be withered and flattened ( Fig 6C ) . The toxicity of Rubb12-tri and Rubb7-tnl , two of the ruthenium complexes shown to have high in vitro efficacy against all schistosome stages tested , was assessed against human bile duct cells and in male BALB/c mice ( 6–8 weeks ) before investigating their in vivo efficacy in a mouse model of schistosomiasis . PZQ and dichlorvos—a metabolite of the anti-schistosome AChE inhibitor metrifonate—were included in the study for comparison . When Rubb12-tri and Rubb7-tnl were used at 5 μM , a concentration where they significantly inhibited surface and somatic Sm-AChE activity and glucose uptake in adult worms , cell viability was 42% and 73% for Rubb12-tri and Rubb7-tnl , respectively . The EC50 values of Rubb12-tri and Rubb7-tnl were calculated as 3 . 489 ± 0 . 532 μM and 6 . 829 ± 0 . 625 μM , respectively . Dichlorvos was highly toxic to the cells , killing 100% of the cells even at 0 . 1 μM ( Fig 7 ) . To determine the MTD in mice , Rubb12-tri or Rubb7-tnl was administered to groups of male BALB/c mice ( 6–8 weeks ) . Rubb7-tnl did not show any toxicity even after five consecutive daily injections ( the proposed dosage frequency of the in vivo drug efficacy study ) of 10 mg/kg ( mice were adversely affected at doses of 20 mg/kg ) and so the MTD of Rubb7-tnl was considered to be at least 10 mg/kg . The MTD of Rubb12-tri , using the same dosage frequency , was determined to be 2 mg/kg ( mice were adversely affected at doses of 4 mg/kg ) . If 100% bioavailability is assumed due to i . v . administration , the host bloodstream concentrations of Rubb7-tnl and Rubb12-tri at the MTD can be approximated at 12 μM and 49 μM , respectively . Over two independent trials , a significant reduction in worm burden ( 42% , P = 0 . 009 ) was seen in mice treated with Rubb12-tri compared to controls whereas a non-significant trend towards decreased worm burden was observed in Rubb7-tnl-treated mice ( Fig 8A ) . Surface AChE activity was decreased in worms collected from mice treated with ruthenium complexes but only reached significance for Rubb12-tri-treated animals ( Fig 8B ) . Surface SmNPP-5 , surface Sm-AP , somatic AChE and glucose uptake activity was not significantly different . Although there was no decrease in parasite egg burden ( as determined by recovery of ova from the liver ) , the viability of these eggs from trial 1 was examined and highly significant ( P < 0 . 0001 ) reductions in hatching capability ( 68% and 56% ) were observed for the Rubb12-tri- and Rubb7-tnl-treated mice , respectively ( Fig 8C ) . Egg hatching viability was not determined for trial 2 . Moreover , eggs released from worms recovered from treated mice ( trial 2 ) were significantly different ( P < 0 . 0001 ) in terms of their development ( immature , misshapen , eggshell malformation ) compared to those from parasites recovered from control mice ( Fig 8D and 8E ) . Morphology of eggs released from worms recovered from mice in trial 1 was not determined .
Control of schistosomiasis , a neglected tropical disease which affects over 200 million people , relies on periodic treatment with single drug , PZQ , a strategy that is unsustainable in its current form [5–8] . As no new anti-schistosome drug ( or any anti-parasitic ) has been registered in the last decade [39] , the need for additional therapeutic compounds has become unquestionable and has driven research efforts towards the discovery of alternative anti-schistosome chemotherapies , including those derived from natural products [40] and metal-based compounds [41] . Accordingly , this study has described the anti-schistosome efficacy of a series of mono- and multi-nucleated metal-based compounds ( ruthenium complexes ) which exert their action through the inhibition of AChE , an enzyme pivotal to the control of worm neuromuscular function and implicated in the mediation of host glucose scavenging [11 , 42] . It has been previously shown that mononuclear ruthenium complexes inhibit AChE ( E . electricus ) by a non-competitive or mixed mode of inhibition [18] . However , in the present study , the trinuclear complex ( Rubb12-tri ) displayed a competitive mode of inhibition in the kinetic experiments . The mononuclear polypyridylruthenium ( II ) complexes are thought to interact with the peripheral anionic site ( PAS ) of AChE located at the rim of the active-centre gorge through a combination of electrostatic and hydrophobic interactions [43] . The tri- and tetra-nuclear complexes showed greater activity compared to mononuclear complexes , presumably due to the presence of the flexibly-linked multiple metal centres which may provide more interactions ( electrostatic and hydrophobic ) with the PAS , or each individual centre may contribute nonspecific additional points of contact . The activity of the ruthenium complexes varied in extracts made from different life stages ( adult extracts and SEA ) and various species of the parasite which is most likely due to differences in enzyme orthologues and the existence of multiple isoforms of AChE which are present in different life stages and species [42] . Encouraged by the activity of ruthenium complexes against parasite extracts , we tested the compounds against all three intra-mammalian stages of the parasite in vitro and found similar trends in anti-parasite activity as was seen for extracts; i . e . , the tri- and tetra-nucleated complexes were more effective against each stage of the parasite than the mono- and di-nuclear compounds . Three of the most effective compounds in terms of their combined activity against S . mansoni extracts and all intra-mammalian stages were Rubb7-tnl , Rubb12-tri and Rubb16-tnl . Further , Rubb12-tri was also the most effective at inhibiting AChE activity in S . haematobium extracts; availability of material prevented us from doing any experiments on live parasites or eggs but we believe that that the similar trends observed between S . mansoni anti-parasitic activity and extract activity hold true for S . haematobium and ruthenium complexes such as Rubb12-tri would display potent anti-schistosome activity against this species . Further , S . haematobium has higher levels of tegumental AChE than S . mansoni , which makes the parasite more sensitive to AChE inhibitors [44] and might render this species more vulnerable to ruthenium drugs . Any differences observed between the AChE-inhibitory ability and anti-schistosome effect of ruthenium complexes could be due to the target of these drugs not solely being AChE . While we showed that ruthenium complexes did not have any inhibitory effects on the major tegumental enzymes SmNPP-5 and AP , these drugs have been documented to act as dual inhibitors of telomerase and topoisomerase [45] , thioredoxin reductase [46] and protein and lipid kinases [43] . There are numerous reports in the literature documenting the development of drug resistance in parasites due to mutation ( for example , benzamidazole resistance in nematodes due to single nucleotide polymorphisms in β-tubulin [47] and mutation of a schistosome sulfotransferase resulting in resistance to oxamniquine [48] , and so the use of a drug that is directed against multiple molecular targets may decrease the chance of resistance evolving . Visually , the effects of the compounds were most pronounced against adult worms , which became immobile and coiled when incubated with ruthenium complexes , possibly due to paralysis induced by AChE inhibition and cholinergic accumulation , effects similarly seen in schistosomes treated with other inhibitors of AChE [49–51] . This observation should be treated with caution , however , as other drugs , such as PZQ ( which is not a cholinesterase inhibitor ) , induce the same morphological changes . Additional evidence of the mechanistic effects of ruthenium complexes on schistosomes manifested in the reduced glucose uptake observed in drug-treated worms , potentially a consequence of inhibiting the tegumental AChE-mediated regulation of host glucose scavenging , a pathway unique to schistosomes [11] . Further confirmation of this inhibition was evidenced by significantly depleted glycogen stores ( quantified in parasite extracts and observed by the withering of male tubercules—a site of glycogen storage [38] ) in these parasites , an effect seen in worms recovered from mice treated with carbamate-based AChE inhibitors [38] . In another example of tegument-mediated glucose regulation , previous work by You and colleagues [52] has shown that inhibition of schistosome insulin receptor activity significantly decreased glucose uptake by the parasite . It would be interesting to explore any relationship that existed between these two regulatory mechanisms , a possibility given the alternative is to imagine the evolution of two mechanistically distinct pathways of glucose regulation . Additionally , a combination chemotherapeutic strategy could be developed using drugs which target different aspects of schistosome glucose uptake . Two of the ruthenium complexes which we considered most effective in vitro ( Rubb12-tri and Rubb7-tnl ) were tested for cytotoxicity before assessment of their in vivo efficacy in a mouse model of schistosomiasis . Rubb16-tnl , even though effective in vitro , was not included in the cytotoxicity assay or the in vivo study as earlier work by us has shown that ruthenium complexes with longer chain lengths are more toxic to eukaryotic cells [24] . Both ruthenium complexes tested , Rubb7-tnl and Rubb12-tri , exhibited lower cytotoxicity against eukaryotic cells ( H69 ) compared to dichlorvos , an organophosphorus AChE inhibitor and previously licensed , but now withdrawn , anti-schistosome drug . Further , studies comparing AChE from schistosomes and higher eukaryotes [53 , 54] reveal differences in functionally important amino acid residues ( human AChE shares 33–36% primary sequence homology across all schistosome AChEs ) with the active site serine conserved across species . It was previously shown that dichlorvos covalently binds to the active site serine and reduces the AChE activity in eukaryotes ( e . g . in rat forebrain , erythrocytes and plasma ) [55 , 56] . By contrast , it was considered that the ruthenium complexes may be relatively less toxic to mice than dichlorvos due to the differential binding to AChE . The results of the MTD study , where both the ruthenium complexes were well tolerated by mice , supported this argument . Rubb12-tri and Rubb7-tnl both showed promising in vivo efficacy at doses which were equivalent to lethal in vitro concentrations yet well tolerated in mice , with Rubb12-tri-treated mice showing a significant reduction in worm burden and recovered worms displaying a small but significant decrease in tegumental AChE activity , providing evidence that the anti-schistosome effect may be partially due to AChE inhibition . Studies by us on di-nuclear ruthenium complexes have shown the compounds to have a short serum half-life [57] , and the different in vivo efficacies of each complex in this study may be attributed to differences in the pharmacokinetic/pharmacodynamic ( PK/PD ) properties of Rubb12-tri and Rubb7-tnl . Although these experiments have yet to be performed , the differences exhibited between these two compounds in the cytotoxicity assay and tolerability study suggests that their PK/PD are not the same . Despite no significant reduction in egg burden in both trials , ova recovered from both Rubb12-tri- and Rubb7-tnl-treated mice had significantly reduced hatching ability and were morphologically abnormal compared to controls , in agreement with in vitro data . That there was no decrease in egg burden in light of a reduced worm load in either treated group was surprising , but this did result in an increase in the number of eggs per female in these groups , compared to controls . To explain these observations , we postulate that treatment with ruthenium drugs may stimulate schistosome reproductive tract motility ( AChE inhibitors have been shown to stimulate gastrointestinal motility in various organisms [58 , 59] ) , resulting in premature release of under-developed eggs , and have a direct effect on egg formation , resulting in abnormally developed eggs ( studies in ticks have shown that treatment with AChE inhibitors effect ova development [60 , 61] ) . Another possible factor contributing to abnormal egg development is that the worms are under-nourished due an impaired glucose uptake ability ( albeit an effect we could only measure in vitro ) resulting from ruthenium drug treatment and unable to meet the energetically demanding task of producing normally developed ova . There is considerable interest in the use of agents that show ovicidal activity or affect oviposition to control schistosomiasis due to their ability to block transmission of the disease . In this regard , ruthenium complexes offer a potential advantage over PZQ in that it is only effective against mature worms and so cannot be used to interrupt disease transmission , as evidenced by high rates of re-infection in PZQ-treated endemic populations [62] . To our knowledge , this is the first report detailing the anti-parasitic activity of ruthenium complexes and this work has identified some lead anti-schistosome compounds . The modular nature of ruthenium complexes makes it possible to synthesize these compounds to target specific enzymes , so future work will involve tailoring Ru complexes to increase their selectivity and potency . Finally , these complexes could be administered in combination with PZQ , overcoming the limitations of current monotherapy and augmenting existing schistosomiasis control initiatives .
|
Schistosomiasis is a neglected tropical disease which affects over 200 million people and there is only one licensed drug , praziquantel , currently available for treatment . In a search for new drugs to control schistosomiasis , we tested the anti-schistosome efficacy of a series of ruthenium compounds and found that a number of them were able to inhibit parasite eggs from hatching and kill adult worms and praziquantel-refractory juvenile worms in vitro . We demonstrated that the compounds inhibit schistosome acetylcholinesterase ( the enzyme that breaks down the neurotransmitter acetylcholine ) , which could potentially result in paralysis of the parasite , likely due to uncontrolled neuromuscular function caused by acetylcholine excess . Moreover , we showed that drug-treated worms had a significantly reduced ability to uptake exogenous glucose and markedly depleted glycogen stores , presumably through inhibition of the acetylcholinesterase-mediated glucose scavenging pathway . Lastly , we found that two of the drugs—Rubb12-tri and Rubb7-tnl—when used to treat schistosome-infected mice , were able to reduce worm burdens and significantly affect the viability of parasite eggs in vivo , which would have a marked impact on disease transmission . We believe that these complexes are desirable drug lead scaffolds which could be used to develop effective and selective compounds to control and treat schistosomiasis and , potentially , other parasitic diseases .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"helminths",
"tropical",
"diseases",
"enzymology",
"carbohydrates",
"parasitic",
"diseases",
"organic",
"compounds",
"glucose",
"animals",
"animal",
"models",
"model",
"organisms",
"pharmaceutics",
"experimental",
"organism",
"systems",
"neglected",
"tropical",
"diseases",
"enzyme",
"inhibitors",
"ruthenium",
"research",
"and",
"analysis",
"methods",
"chemistry",
"mouse",
"models",
"biochemistry",
"helminth",
"infections",
"chemical",
"elements",
"schistosomiasis",
"eukaryota",
"organic",
"chemistry",
"monosaccharides",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"drug",
"therapy",
"organisms"
] |
2017
|
Polypyridylruthenium(II) complexes exert anti-schistosome activity and inhibit parasite acetylcholinesterases
|
One to two percent of all children are born with a developmental disorder requiring pediatric hospital admissions . For many such syndromes , the molecular pathogenesis remains poorly characterized . Parallel developmental disorders in other species could provide complementary models for human rare diseases by uncovering new candidate genes , improving the understanding of the molecular mechanisms and opening possibilities for therapeutic trials . We performed various experiments , e . g . combined genome-wide association and next generation sequencing , to investigate the clinico-pathological features and genetic causes of three developmental syndromes in dogs , including craniomandibular osteopathy ( CMO ) , a previously undescribed skeletal syndrome , and dental hypomineralization , for which we identified pathogenic variants in the canine SLC37A2 ( truncating splicing enhancer variant ) , SCARF2 ( truncating 2-bp deletion ) and FAM20C ( missense variant ) genes , respectively . CMO is a clinical equivalent to an infantile cortical hyperostosis ( Caffey disease ) , for which SLC37A2 is a new candidate gene . SLC37A2 is a poorly characterized member of a glucose-phosphate transporter family without previous disease associations . It is expressed in many tissues , including cells of the macrophage lineage , e . g . osteoclasts , and suggests a disease mechanism , in which an impaired glucose homeostasis in osteoclasts compromises their function in the developing bone , leading to hyperostosis . Mutations in SCARF2 and FAM20C have been associated with the human van den Ende-Gupta and Raine syndromes that include numerous features similar to the affected dogs . Given the growing interest in the molecular characterization and treatment of human rare diseases , our study presents three novel physiologically relevant models for further research and therapy approaches , while providing the molecular identity for the canine conditions .
One to two percent of all children are born with a developmental disorder , such as a heart defect , skeletal abnormality , or mental retardation as a result of errors in embryogenesis and early neurodevelopment . These disorders make a major contribution to pediatric hospital admissions and mortality [1] . Rare pediatric disorders are typically with homozygous , compound heterozygous , or de novo pathogenic variants . The advent of cost-efficient next generation sequencing ( NGS ) technologies drives gene discovery in many disorders without a cognate gene [2] and thousands of rare variants have been described ( www . orpha . net ) . There is also a growing interest in the development of therapeutics for rare diseases , which requires the identification of the genetic defects , comprehensive understanding of the molecular pathology and access to physiologically relevant animal models . Developmental disorders are frequent also in other species , including dogs , which as large animals bear very close physiologic and genetic resemblance with us . Dogs give birth to litters with multiple puppies . However , often some of the littermates are affected by developmental or other abnormalities , and perinatal mortality ( stillbirth , fetal and neonatal death ) is common with prevalence ranging from 5 to 35% [3] . The causes include respiratory distress syndrome/hypoxia , infectious diseases , severe malformations and suspected hereditary diseases . Culling is a common practice among dog breeders and the deceased or abnormal puppies are not always presented to the veterinarian . Therefore , numerous syndromes with likely genetic origin remain unknown . It would be important to investigate the extent of this common phenomenon at the clinical and molecular level to better understand the diverse causes of the morbidity and to better manage it through advised breeding programs . At the same time , the identification of the causative gene could inform gene functions , disease etiology , molecular pathology and phenotypic overlap across species . Importantly , physiological similarity of dogs with human would establish relevant therapeutic models to human rare disorders . Online Mendelian Inheritance in Animals ( OMIA ) , a catalogue of inherited disorders and associated genes in animals , reports more than 350 inherited diseases in dogs as potential models for human disease ( http://omia . angis . org . au/ ) . NGS approaches are rapidly changing the diagnostic landscape in veterinary medicine in companion animals and enable now a feasible approach to tackle the molecular background of developmental conditions in small pedigrees with translational potential to human rare disease . For example , we have previously discovered a new gene ( ATG4D ) responsible for a neurodegenerative vacuolar storage disease in Lagotto Romagnolos [4] and a missense variant in canine FAM83G causing palmoplantar hyperkeratosis and demonstrating its role in maintaining the integrity of the palmoplantar epidermis [5] . A CNGB1 frameshift variant has been identified to cause a progressive retinal atrophy in dogs [6] and the same gene has been associated with retinal degeneration in human as well [7] . Similarly , we have found that a congenital skeletal disease in Brazilian Terriers is caused by a pathogenic variant in the GUSB orthologue responsible for human pediatric disorder , mucopolysaccharidosis type VII [8] . This study addressed the clinical and genetic background of three developmental disorders in dogs; craniomandibular osteopathy ( CMO; OMIA: 000236–9615 ) in West Highland White Terriers ( WHWT ) , Cairn Terriers and Scottish Terriers , a previously undescribed developmental syndrome in Wire Fox Terriers , and dental hypomineralization in Border Collies . CMO is a self-limiting proliferative bone disease seen in young dogs [9] . It manifests between 4 to 8 months of age with typical signs including swelling of the jaw , periodical fever , lack of appetite , pain , difficulty opening the mouth and dysphagia . The excessive proliferation causes bony lesions primarily on the skull bones , especially on the mandible and tympanic bulla , but occasionally also on the metaphyses of long bones . Signs of the disease usually resolve with time , when the growth period is finished . CMO exists in several breeds with the highest frequency in WHWT and has been suggested to be an autosomal recessive trait [10–12] . Canine CMO corresponds to human Caffey disease [MIM: 114000] and its genetic characterization might reveal insights into similar painful human swelling disorders [10] . We identified a novel CMO gene that represents a candidate gene for human Caffey disease . The two other syndromes have not previously been reported in dogs . We describe the detailed clinical features for them in our study . Moreover , we demonstrate their shared genetic etiology with the corresponding human syndromes , van den Ende-Gupta ( VDEGS [MIM: 600920] ) and Raine syndromes [MIM: 259775] .
We performed a genome-wide association study ( GWAS ) to map the CMO locus with Illumina’s 22K canine SNP chips in a cohort of 51 WHWTs , including 10 cases ( diagnosed by radiography; Fig 1A ) and 41 controls . A case-control association test revealed significant association on CFA5 with the best SNP ( BICF2S23544899 ) at 8 , 953 , 507 Mb ( praw = 1 . 2 x 10−7 , pgenome = 0 . 02 ) ( genomic inflation factor λ = 1 . 17 ) ( Fig 1B ) . Manual assessment of genotypes at the CFA5 locus revealed a shared 1 . 9-Mb homozygosity block in all affected dogs spanning from 7 , 764 , 955 bp to 9 , 707 , 794 bp . The same homozygosity block was seen in five controls as well . The associated region was replicated and fine-mapped using 105 additional SNPs in 88 samples from three related breeds , WHWT , Cairn Terriers and Scottish Terriers . Fine mapping confirmed the association in all three breeds with the best SNP BICF2S23134295 at 8 , 183 , 669 ( p = 2 . 09x10-15 ) . To identify candidate variants , the associated and fine mapped region ( 1 . 8 Mb ) was captured from two affected and two healthy WHWT with opposite haplotypes followed by a paired-end NGS by HiSeq2000 . The shared variants identified in the two affected WHWT dogs were filtered against the two WHWT controls and 32 additional healthy dogs from four different breeds . We found altogether three homozygous variants shared in cases ( S1 Table ) ; two intergenic indels and a synonymous variant in exon 15 of solute carrier family 37 member 2 gene ( SLC37A2 c . 1332 C>T ) ( Fig 1C ) . As an additional independent verification to avoid potential targeted capture biases , we performed whole genome sequencing in one CMO-affected WHWT and 188 control dogs from other breeds ( S2 Table ) to compare the variants in the associating region . This analysis yielded a single case-specific variant ( chr5 g . 9 , 387 , 327G>A ) that was the same as the one identified in the capture experiment . Although the variant , c . 1332C>T in exon 15 of SLC37A2 is synonymous , it was predicted to affect a splicing enhancer element based on ESEfinder analysis ( Fig 1D ) . The mutant T allele eliminates a potential binding site for the splicing factor ASF/SF-2 . To confirm the predicted effect on splicing , we amplified the region between exons 7 and 18 in lymphocyte mRNAs from three cases , six carriers and seven controls . RT-PCR experiments showed that two alternatively spliced SLC37A2 transcripts were expressed in all dogs that carried one or two copies of the mutant T allele at the SLC37A2 SNP , regardless of disease status ( Fig 1E ) . Sequencing of the RT-PCR products indicated that the smaller band corresponded to a mutant SLC37A2 transcript lacking 79 bp from exon 15 . The splice variant resulted in a frameshift and premature stop codon at the beginning of exon 16 . This altered splicing was predicted to lead to a C-terminally truncated protein lacking 75 amino acids compared to the wild-type SLC37A2 protein ( Fig 1F ) . The RT-PCR indicated that both wild-type and mutant transcripts were expressed in even the CMO affected dogs , although the expression of the wild-type transcript was significantly reduced in the affected homozygous dogs ( Fig 1E ) . The reduction of the wild type transcript level was more moderate in healthy heterozygous carrier dogs . None of the examined wild-type dogs expressed the mutant transcript . Overall , these results demonstrate the leaky nature of the splice site mutation . To investigate the segregation and frequency of the variant across the CMO affected breeds , we performed a large variant screening by genotyping the c . 1332C>T variant altogether in 1052 dogs , including 695 WHWT , 249 Scottish Terriers and 108 Cairn Terriers ( S3 Table ) . We found 123 homozygous dogs in the WHWT breed , of which 48% had been reported with CMO . About 40% of WHWT ( 275 dogs ) carried the pathogenic variant , of which 10 dogs ( 3 , 6% ) were reported with CMO . In Scottish Terriers , 10 dogs ( 4% ) were homozygous and all were reported with CMO , and 43 dogs ( 17% ) were carriers , of which 3 dogs ( 7% ) were reported with CMO . In Cairn Terriers , 9 dogs ( 8% ) were homozygous and reported with CMO , and 15 dogs ( 14% ) carried the pathogenic variant , from which 3 dogs ( 20% ) were reported with CMO . We found one wild-type dog in both Scottish and Cairn Terriers with CMO , and screened the coding regions of the entire SLC37A2 gene in these two dogs for possible other pathogenic variants , but did not find any . This suggests phenocopies , misdiagnoses or genetic heterogeneity . The analysis of the pathogenic variant in the three main breeds with 96 cases resulted in a highly significant association ( p = 6 . 62x10-303 ) with CMO . In addition to the above three breeds , we screened the c . 1332C>T variant in 458 dogs in 124 breeds , but found only a single heterozygous carrier dog in Jack Russell Terrier breed ( S3 Table ) . The phenotype information for this dog was not available . The variant was also screened in the known CMO cases from seven breeds ( two Bull Terriers , one Curly Coated Retriever , two Border Collies , one Australian Terrier , one Basset , one German Wirehaired Pointer , one Old English Sheepdog ) , but they did not have it . Collectively , our results suggest that CMO is inherited as dominant disease with incomplete penetrance . Canine CMO is equivalent to Caffey disease and our data reveals a novel candidate gene , SLC37A2 , for the syndrome . Wire Fox Terrier breeders contacted us for help in the characterization of an unknown congenital syndrome with severe mandibular prognathia and other skeletal features , mainly severe patellar luxation , in the breed . Two affected 7-week-old puppies from different litters , two unaffected littermates and two affected adult Wire Fox Terriers were examined by radiography . Additionally , a computed tomography ( CT ) study of the skull was made on two affected dogs ( one adult and one puppy ) , and three dogs were studied for general clinical characteristics and neurological examination . A prominent underbite with short maxilla ( brachygnathia superior ) was evident in all affected Wire Fox Terriers except one adult dog ( Fig 2A and 2B ) . The caudodorsal border of the maxilla was slightly convex in all affected animals . In CT images , the nasal septum deviated prominently to the left at the level of the dorsocaudal frontal bone in both examined dogs ( Fig 2C ) . The number and position of the vertebrae were normal , but the mid-thoracic spinous processes were thinner , longer and more horizontally aligned in the affected than in the normal dogs . One adult dog had an abnormally wide second rib . The adult dog had marked spondylosis of the spine . One affected puppy had unilateral congenital elbow luxation ( Fig 2D ) , and in the other the secondary ossification centers of the olecranon were non-mineralized . The secondary ossification centers of the tibial tuberosities were small in both affected puppies , when compared to a healthy puppy ( Fig 2E and 2F ) . The proximal epiphyses of the fibulae were not mineralized in the affected puppies and the patellae were medially luxated . The femurs of the affected dogs had medial bowing of mid-shafts of the bone . In an eye examination of a puppy and two affected adult dogs , the eyes appeared small and the sclera thinner than normal . Clinical examination of three affected dogs indicated swollen knee joints and patellar luxation . Neurologically , all the examined dogs were alert and exhibited no remarkable neurological deficits . Postmortem examination of two puppies ( one newborn and one 7 weeks old ) did not reveal any additional gross abnormalities . To identify the genetic cause of the syndrome , we performed GWAS with Illumina’s CanineHD array in a cohort of 12 Wire Fox Terriers including 4 cases and 8 controls . A case-control association test revealed association on chromosome 26 with seven nearby SNPs at 29 , 607 , 333 to 31 , 863 , 083 Mb ( praw = 7 . 74x10-6 , pgenome = 0 . 05 ) ( genomic inflation factor λ = 1 . 10 ) ( Fig 3A ) . Manual assessment of genotypes at the CFA5 locus revealed a shared homozygosity segment of ~3 Mb in the affected dogs spanning from 29 , 176 , 909 to 32 , 226 , 403 bp ( Fig 3B ) . The associated region was captured and resequenced from five samples including two affected , two healthy and one obligate carrier Wire Fox Terrier ( S4 Table ) . We identified a 2-bp homozygous deletion in exon 6 of the SCARF2 gene in the affected dogs after filtering the data according to an autosomal recessive model and against additional 169 unaffected control dogs from different breeds ( S2 Table ) . The identified SCARF2 c . 865_866delTC variant results in a frameshift and a premature stop codon , ( p . S289Gfs*15 ) , leading to a truncated protein in the first half of the coding region ( Fig 3C ) . Screening of the variant within the Wire Fox Terrier breed ( 57 dogs ) confirmed full segregation with the disease ( S5 Table ) . The four cases from the GWA study and an additional case were homozygous for the variant , while obligate carriers were heterozygous . The larger screening for the mutation revealed one homozygous dog that was a littermate of one of the genotyped affected dogs . This dog was then clinically examined , including neurological examination , radiography and CT scanning . The clinical findings confirmed the disease , including mandibular prognathia and patellar luxation . The carrier frequency among the population controls ( n = 45 ) was 22% . The effect of the 2-bp deletion on the stability of the SCARF2 mRNA was investigated by RT-PCR in postmortem skin samples from one affected and one unaffected dog ( Fig 3D ) . The SCARF2 mRNA was detected both in affected and unaffected samples suggesting that the mutated transcript is not affected by nonsense-mediated RNA decay in the studied tissue . SCARF2 defects have been reported in the rare human bone disease van den Ende-Gupta syndrome . Our results thus established an orthologous canine model with clinical similarity . We were approached by a Border Collie breeder with a family of several affected dogs that suffered from severe tooth wear resulting in pulpitis and requiring extraction of those teeth . Further inspection of the tooth problem in the breed identified additional related cases , suggesting an autosomal recessive mode of inheritance ( S1 Fig ) . Two affected dogs were subjected to a clinical study , including dental examination and radiography , as well as to histology of the extracted teeth and were regularly followed up in the next years . In addition , dental radiographs were available from two other cases . Dental examination of a neutered 9-year-old female Border Collie revealed that all remaining teeth had significant wear . The previous dental treatment was performed 2 years earlier and multiple teeth were extracted . The length of the crowns was reduced . Lower incisor teeth were worn close to gingival margin . The enamel appeared dull and had light brown discoloration . The worn occlusal surfaces were discolored dark brown and there was reparative dentin formation . There were five teeth that had pulp exposure and pulpitis as a result of the wear ( Fig 4A ) . The dog’s occlusion was normal and , therefore , the dental wear was not caused by abnormal tooth-to-tooth contact ( attrition ) . Calcitriol ( 1 , 25 ( OH ) 2D3 ) , phosphate and alkaline phosphatase levels in blood were normal . The other dental examination was performed for a neutered 10-year-old male Border Collie , revealing similar findings ( Fig 4B and 4C ) . Other external causes such as abrasive hard chews were excluded as a cause of dental wear in both affected dogs . Extracted teeth from two affected dogs , a 9-year-old female Border Collie described above and its female littermate , were submitted to histopathological examinations . The analysis of ground sections did not reveal structural aberrations but the enamel of the incisor was smooth and slightly hypoplastic as compared to the unaffected control dog ( Fig 4D and 4E ) . The enamel of the premolar had largely worn and cracked , but the cervical enamel , which was preserved , showed no structural defects . Coronal dentin of both teeth comprised three distinct , circumferential zones . The tubular pattern and the structure of the matrix in the thick peripheral zone subjacent to the enamel were regular . The middlemost zone was pronouncedly globular ( Fig 4E and 4F ) . The neighboring globules largely adapted to each other and the tubules ran uninterrupted . Wider interglobular spaces were filled with air . The globules diminished and disappeared in the central direction . The central dentin zone next to the reduced pulp showed no matrix defects , but the tubular pattern was slightly irregular ( Fig 4F ) . The proportion of globular dentin gradually reduced in the apical direction . The analysis of paraffin sections demonstrated significant wear of the dentin . The structure of the peripheral zone was regular , whereas the middlemost zone was globular . Between the globules there were wide , contiguous defects with an angular contour , void of matrix and tubules and filled with amorphous , barely detectable material ( Fig 4G ) . A pulp chamber did not become visible at the level of the sections . Tubules in the central dentin zone were slightly irregular . Pulp tissue in the root canal was necrotic . A patchy , chronic inflammatory cell infiltrate with plasma cells predominating was present apically . The structure of the acellular cementum was regular . Lacunae in the cellular cementum and at the periphery of the alveolar bone trabeculae were in places obliterated . Demarcated areas in the periodontal ligament facing the alveolar bone showed an amorphous , fibrotic texture . Overall , clinical and histopathological analyses indicate severe hypomineralization of teeth in the affected dogs . To identify the cause of the mineralization defect we sequenced the whole genomes of three affected dogs . The variants of the affected dogs were filtered against two unaffected obligate carriers and fourteen other unaffected Border Collie genomes assuming recessive transmission , resulting in the identification of a case-specific non-synonymous homozygous variant , c . 899C>T , in the FAM20C gene . This leads to a missense change , p . A300V , in a highly conserved position in the kinase domain of the FAM20C protein ( Fig 5 ) . Bioinformatic predictions by SIFT ( with a score of 0 . 00 ) and Polyphen2 ( with a HumVar score of 0 . 992 ) suggested pathogenicity . Genotyping the pedigree and additional Border Collies ( 191 dogs ) demonstrated complete segregation of the variant with the disease phenotype ( S1 Fig and S6 Table ) and showed 11% carrier frequency in the breed . We also genotyped 186 dogs from 20 additional breeds ( S6 Table ) , but did not find any carriers in the other breeds suggesting that this pathogenic variant is specific for Border Collies . Defects in FAM20C have been associated with a rare human disease , Raine syndrome/FGF23-related hypophosphatemia characterized by dental and bone hypomineralization . Our results indicated a causative variant in the kinase domain of FAM20C and established a canine model for human Raine syndrome .
Our study unraveled a physiological function of SLC37A2 and provided new insights into infantile swelling diseases , which may be related to disturbances in the intracellular glucose homeostasis during bone development . We identified an unusual leaky splicing defect in the SLC37A2 gene in CMO-affected dogs . CMO is a self-limiting hyperostosis in multiple bones in young dogs . The most prominent sign in the affected puppies included painful swelling of the jaw , leading to dysphagia and difficulty in opening the mouth . CMO is clinically equivalent to human infantile cortical hyperostosis [10 , 15] . A common missense variant in COL1A1 has been found in several patients with an autosomal dominant condition with incomplete penetrance [16 , 17] . Interestingly , the mechanism how the defected collagen leads to self-limiting hyperostotic bone lesions is still unknown . Early molecular diagnosis of Caffey patients would avoid invasive procedures , however , the molecular etiology remains unknown in many cases . SLC37A2 represents a new functional candidate gene . It belongs to the SLC37 family of four ER-associated glucose-phosphate transporters [18] . SLC37A2 is ubiquitously expressed , but transcript and protein levels are particularly high in bone-related tissues such as bone marrow and hematopoietic cell linages such as osteoclasts and macrophages [19 , 20] . Murine Slc37a2 was shown to be one of the genes strongly involved in the osteoclast differentiation , suggesting that it plays a role in osteoclast function and differentiation [20] . Therefore , SLC37A2 may play a central role in glucose homeostasis in the key cell types that participate in osteogenesis . For example , an impaired function of SLC37A2 due to a truncating splice variant might disturb proper glucose supply in the osteoclasts , decreasing their overall activity , which in turn would result in an imbalance between osteoblastic and osteoclastic functions in the developing bones eventually leading to hyperostosis . Defects in SLC37A4 , glucose-6-phosphate transporter ( G6PT ) , have been associated with glycogen storage disease 1b and 1c , characterized by recurrent infections and neutropenia due to disturbed blood glucose metabolism [21–23] . Our results link SLC37A2 to bone physiology and disease , and we propose SLC37A2 as an excellent candidate for genetic screening in Caffey patients . Meanwhile , the affected dogs provide unique resources for future experiments to address SLC37A2-related mechanisms in osteogenesis biology . A recent study in hematopoietic cells identified SLC37A2 as a primary vitamin D target with a conserved vitamin D receptor-binding site [24] . This may open investigations to study the opportunity to use vitamin D as a therapeutic booster to regulate diminished expression of wild type expression of SLC37A2 in the affected dogs to alleviate clinical signs . Caffey disease is an autosomal dominant disease with incomplete penetrance , although rare cases of recessively inherited Caffey disease have also been reported [25] . Corresponding canine diseases exist in several terrier breeds with the highest frequency in West Highland White Terriers . The determination of the exact mode of inheritance in dogs is not straightforward due to the nature of the leaky splice variant and mild self-limiting phenotype that may remain unobserved and prevent retrospective diagnosis . We found some dogs that were homozygous for the variant but had no reported clinical signs . However , we observed a considerable level of the wild-type SLC37A2 transcript in homozygous dogs in the peripheral blood due to the splicing leakage , suggesting that the leaky expression is sufficient to avoid a clinical phenotype in some cases . We also found several heterozygous dogs that had developed CMO . We found that heterozygous dogs had lower levels of wild-type SLC37A2 transcript compared to the unaffected dogs with individual variation of expression between dogs . This result suggested a dominant disease with incomplete penetrance that could help to explain the reported differences in the severity and duration of CMO among the affected dogs , although alternative models of inheritance cannot be completely ruled out yet . The dominant phenotype could be due to a dominant-negative effect , but this hypothesis requires further experimental validation to better understand the details of the gene , its regulation and protein function , including potential pairing with other proteins as described for SLC37A4/G6Pase complexes [18] . The in vivo function of SLC37A4 has been shown to depend upon its ability to couple functionally with either G6Pase-a or G6Pase-b [18 , 26] . We identified a 2-bp deletion in SCARF2 in dogs with severe mandibular prognathia and other skeletal abnormalities and established a canine model for van den Ende-Gupta syndrome ( VDEGS ) . VDEGS is a very rare disease with less than 30 reported cases [27] . It is characterized by a heterogeneous variety of craniofacial and skeletal abnormalities including blepharophimosis , a flat and wide nasal bridge , narrow and beaked nose , hypoplastic maxilla with or without cleft palate and everted lower lip , prominent deformed ears , down-slanting eyes , arachnodactyly , and camptodactyly . Patients may present congenital joint contractures that improve without intervention , and have normal growth and development . Enlarged cerebellum is an infrequent finding yet intelligence is normal . Some patients experience respiratory problems due to laryngeal abnormalities . Human and canine VDEGS patients share many similarities including hypoplastic maxilla , dislocated radial head , patellar dislocation , and deviated nasal septum . Both have small eyes . It remains unknown how the loss of function of SCARF2 leads to VDEGS . SCARF2 is a poorly characterized member of the scavenger receptor type F family [28] . Besides epidermis , Scarf2 is expressed in branchial arches , mandible , maxilla and urogenital ridge tissue of developing mouse embryos [29 , 30] . It is a single-pass transmembrane protein with homology to calmodulin ( CaM ) -like Ca2+-binding protein genes . The extracellular domain contains several putative epidermal growth factor-like ( EGF ) domains , and it has a number of positively charged residues within the intracellular domain , suggesting a role in intracellular signaling . The 2-bp deletion of the canine SCARF2 gene in one of the extracellular EGF domains leads to a severely truncated protein that completely lacks the transmembrane and intracellular domains . The lack of a transgenic mouse model and scarcity of human patients highlight the role of affected dogs as a novel resource to understand SCARF2 functions and molecular pathology . As some of the affected dogs survive past 10 years , they could potentially serve also as preclinical models . Whole genome sequencing of a family of several affected dogs that suffered from a severe dental wear and loss of teeth revealed a recessive missense variant in the kinase domain of the FAM20C gene . FAM20C defects cause autosomal recessive osteosclerotic bone dysplasia ( Raine syndrome ) in humans . This rare syndrome with less than 40 reported cases was originally described to be neonatal lethal , but recently there have been several reports of cases surviving into childhood with variable severity and clinical heterogeneity [31–38] . Typical characteristics in Raine syndrome include craniofacial anomalies , such as exophthalmos , abnormal and hypomineralized teeth , midface hypoplasia , microcephaly and cleft palate , as well as gingival hyperplasia , generalized osteosclerosis and intracerebral calcifications . Variable extent of hypophosphatemia has been observed , sometimes as the primary diagnosis [32 , 35 , 38] . Canine findings in our cohorts were more limited to severe hypomineralization of teeth , leading to extensive wear and inflammation as prominent features . We did not observe some of the typical gross changes described in Raine patients such as hypophosphatemia and craniofacial anomalies . However , there is a significant clinical heterogeneity in the symptoms between human patients and more detailed radiographic analyses should be performed in dogs to observe potential mild changes outside the dental phenotype . FAM20C is a Golgi casein kinase that phosphorylates secretory proteins such as FGF23 and SIBLING ( Small Integrin-Binding Ligand , N-linked Glycoprotein ) family [39 , 40] . Fam20c is significantly expressed in mouse teeth and bone and transgenic mice studies have indicated a role in differentiation and mineralization of odontoblasts , ameloblasts , osteoblasts and osteocytes during tooth and bone development . FAM20C-deficient mice also have a prominent dental phenotype [41–43] . Ablation of the Fam20c gene in conditional knockout mice affects tooth and bone development by downregulation of SIBLING family of proteins such as DMP1 and DSPP and by increasing FGF23 in serum and promoting phosphate excretion and hypophosphatemia [44] . Our FAM20C-deficient dogs have a dental phenotype similar to mice and humans and provide a new research and preclinical model for this rare human bone disease . Unlike rodents , dogs have dental physiology similar to human , having both deciduous and permanent dentition . In summary , we describe here the clinical and genetic characteristics of three new canine models for rare human bone disorders . While highlighting clinical and genetic similarities between canine and human conditions , our study have several implications; it indicates new physiological functions for the identified genes and provides new candidate genes to rare human diseases , establishes potential preclinical models , and finally enables the development of genetic tests for veterinary diagnostics and breeding purposes .
Sample collection in Finland was ethically approved by the Animal Ethics Committee of State Provincial Office of Southern Finland ( Finland , ESAVI/6054/04 . 10 . 03/ 2012 ) . The collection of blood samples in Switzerland was approved by the “Cantonal Committee For Animal Experiments” ( Canton of Bern; permit 23/10 ) . EDTA-blood and tissue samples were collected from privately owned dogs in Finland , US and Switzerland . The samples were stored at -20°C until genomic DNA was extracted using the semi-automated Chemagen extraction robot ( PerkinElmer Chemagen Technologie GmbH ) . DNA concentration was determined either with the NanoDrop ND-1000 UV/Vis Spectrophotometer or Qubit 3 . 0 Fluorometer ( Thermo Fisher Scientific Inc . ) . Pedigrees were drawn by the GenoPro genealogy software ( http://www . genopro . com/ ) , and utilizing the public dog registry by the Finnish Kennel Club ( http://jalostus . kennelliitto . fi ) . Clinical examinations for each condition are described in detail in the results section . CMO cases were diagnosed by radiography by local veterinarians . The developmental disorder in Wire Fox Terriers was investigated by radiography and CT . Neurological examination was performed for three and ophthalmoscopy for two of the affected dogs . A specialized dental veterinarian examined two of the affected Border Collies with dental hypomineralization , while the others were examined by local veterinarians . Clinical phenotype information was not available for dogs used as population controls . Genome-wide association studies were performed in the CMO and VDEGS projects . For CMO , altogether 51 dogs including 10 affected and 41 control dogs , were genotyped using Illumina’s CanineSNP20 BeadChip of 22 , 362 validated SNPs . For VDEGS , a total of 15 dogs including 4 affected and 11 unaffected Wire Fox Terriers , were genotyped using Illumina’s HD array . The genotype data in both projects was filtered with a SNP call rate of >95% , array call rate of >95% and minor allele frequency of >5% . No individual dogs were removed for low genotyping and no SNPs were removed because of significant deviations from the Hardy-Weinberg equilibrium ( p ≤ 0 . 0001 ) . After frequency and genotyping pruning , 14 , 835 and 69 , 694 SNPs remained for analyses for CMO and VDEGS data , respectively . Basic case-control association test was performed by PLINK [45] . Genome-wide significance was ascertained with phenotype permutation testing ( n = 10 , 000 for CMO and n = 100 , 000 for VDEGS ) . Fine mapping of the identified CMO locus was performed with 105 selected SNPs from a 1 . 9-Mb region ( 7 , 764 , 955–9 , 707 , 794 bp ) on CFA5 ( based on CanFam3 . 1 ) . The SNPs were selected using Broad Institute SNP collection CanFam2 . 0 . Genotyping was performed using the Sequenom ( San Diego , CA , USA ) iPLEX methodology at our local core facility in the FIMM Technology Centre , University of Helsinki , Finland . A total of 88 samples were genotyped including 8 cases and 8 controls in Cairn Terriers , 9 cases and 8 controls in Scottish Terriers and 29 cases and 26 controls in WHWTs . Association analysis was performed with PLINK using a single-marker association analysis . We performed a targeted sequence capture and next generation sequencing to identify the pathogenic variants . We used NimbleGen’s in-solution capture technology to enrich the target regions for sequencing ( Roche NimbleGen , Madison , WI , USA ) . We captured 1 . 8-Mb region of CMO associated locus at position CFA5: 10 , 750 , 000–12 , 550 , 000 using two WHWT cases and controls with opposite haplotypes . The haplotypes were assessed manually using SNP genotype data . The same targeting experiment also contained samples from our other targeting projects including 8 Border Terriers , 12 Duck Tolling Retrievers , 8 Schipperkes and 4 Brazilian Terriers and these samples were used as additional controls . For target enrichment and sequencing of associated locus in Wire Fox Terriers , we captured a 3 . 3-Mb region at CFA26: 29 , 030 , 700–32 , 328 , 700 using two affected and two controls with opposite haplotypes and one obligate carrier . The filtered case-specific variants were further checked from 169 additional dogs in our variant database . Probes in the target regions were designed by Roche NimbleGen ( Roche NimbleGen ) . Target enrichment , alignment and variant calling pipeline were performed as previously described [8] . Further data analysis was performed using open source R language and environment ( http://www . r-project . org ) . Canine genome build CanFam3 . 1 was used as a reference sequence . The genetic causes of CMO in WHWTs and tooth attrition in Border Collies were studied by whole genome sequencing . In the CMO study , we performed whole genome sequencing of one affected WHWT dog and used 188 other available whole genomes as controls ( S2 Table ) . A fragment library was prepared with a 290 bp insert size and collected a single lane of Illumina HiSeq2000 paired-end reads ( 2 x 100 bp ) . The reads were mapped to the dog reference genome using the Burrows-Wheeler Aligner ( BWA ) version 0 . 5 . 9-r16 with default settings . The Picard tools ( http://sourceforge . net/projects/picard/ ) were used to sort the mapped reads by the sequence coordinates and to label the PCR duplicates . The Genome Analysis Tool Kit ( GATK version v2 . 3–6 ) was used to perform local realignment and to produce a cleaned BAM file . Variant calls were then made using the unified genotyper module of GATK . Variant data was obtained in variant call format ( version 4 . 0 ) as raw calls for all samples and sites flagged using the variant filtration module of GATK . Variant calls that failed to pass the following filters were labeled accordingly in the call set: ( i ) Hard to Validate MQ0 ≥ 4 & ( ( MQ0 / ( 1 . 0 * DP ) ) > 0 . 1 ) ; ( ii ) strand bias ( low Quality scores ) QUAL < 30 . 0 || ( Quality by depth ) QD < 5 . 0 || ( homopolymer runs ) HRun > 5 || ( strand bias ) SB > 0 . 00; ( iii ) SNP cluster window size 10 . The SnpEff software together with the CanFam 3 . 1 annotation was used to predict the functional effects of detected variants . In addition to the SNP and short indel variant calling , large deletions contained in the candidate region were searched by visual inspection of the BAM file using the Integrative Genomics Viewer ( IGV ) . In the Border Collie study , we whole genome sequenced altogether nineteen dogs , including three affected dogs , two carriers and fourteen unaffected Border Collies , in the Science for Life Laboratory in Stockholm , Sweden . The reads were processed using speedseq align module available in SpeedSeq suite to produce a duplicate-marked , sorted and indexed BAM file . The Genome Analysis Tool kit ( version = 3 . 3 . 0-g37228af ) was used to perform realignment around potential indel sites and base quality score recalibration using the known SNP variation available at the Broad Institute ( https://www . broadinstitute . org/ftp/pub/vgb/dog/trackHub/canFam3/variation ) . Dual algorithms , Samtools mpileup ( version samtools-1 . 2 ) and GATK haplotype caller module were used to detect variants and the variants from both algorithms were merged into variant call format ( VCFv4 . 1 ) . Annovar and SnpEff tools were used to annotate the variants to Ensembl , NCBI and Broad annotation databases to predict the functional effects of the variants . We identified on average ~6 million variants per sample and the sequencing coverage varied between 22-49x . The three affected dogs shared ~1 . 5 million homozygous variants . Filtering under recessive model against two carriers and fourteen controls left 2690 homozygous variants in total , of which five variants were in the predicted coding regions ( 1 indel , 2 non-synonymous , 1 synonymous ) . Large numbers of dogs were genotyped for the identified variants using various protocols . Genotyping of individual dogs for the CMO variant was performed either by TaqMan assay ( Applied Biosystems ) or by sequencing a 786-bp PCR product using a forward primer ( 5-GGCTCCAGTCTAAGCCAGGT-3 ) and a reverse primer ( 5-AAGGAGTGCGCTCAAGACAG-3 ) flanking the SLC37A2 SNP . The PCR products were amplified with AmpliTaqGold360Mastermix ( Life Technologies ) , and the products were directly sequenced using the PCR primers on an ABI 3730 capillary sequencer ( Life Technologies ) after treatment with exonuclease I ( New England Biolabs ) and rapid alkaline phosphatase ( Roche ) . The sequence data were analyzed using Sequencher 5 . 1 ( GeneCodes ) . Potential exonic splice enhancer ( ESE ) motifs were detected with ESEfinder 3 . 0 [46 , 47] . Pathogenicity of the FAM20C c . 899C>T variant was evaluated using web-based bioinformatic prediction tools SIFT [48] and PolyPhen-2 ( genetics . bwh . harvard . edu/pph2 ) [49] was applied to evaluate the pathogenic effect of the mutation . SIFT score ranges from 0 to 1 . The amino acid substitution is predicted to be damaging if the score is smaller than 0 . 05 . PolyPhen-2 score ranges from 0 to 1 . The amino acid substitution is predicted to be damaging if the score is bigger than 0 . 85 . Genotyping of FAM20C and SCARF2 variants was performed by standard PCR with the following primers: FAM20C: 5-GCTTCTATGGCGAGTGTTCC-3 and 5-CCGGGATGTCTGAGTAAGGA-3; SCARF2:5-CAATCCCCGAGTGCTCTCC-3 and 5-AGGAAACTGCCCCCAAAGAG-3 . Expression analyses were performed for the CMO and VDEGS projects . Blood samples were collected into PAXgene tubes ( PreAnalytix ) and RNA was isolated using PAXgene Blood RNA Kit ( Qiagen ) . The cDNA synthesis was performed using SuperScriptIII enzyme ( Invitrogen ) with an oligo d ( T ) 24V primer according to manufacturer’s instructions . To investigate possible aberrant splicing events , we amplified cDNAs from the junction of exons 7 and 8 to the junction of exons 17 and 18 of the SLC37A2 gene ( 810 bp ) using SequalPrep Long Polymerase ( Invitrogen ) with a forward primer GAATACCCAGAAGACGTGGAC and a reverse primer CCTCTGTCTCTGTTCAGGAATG in 16 WHWTs . B2M was used as a loading control . The identity of the amplicons was confirmed by Sanger sequencing . In the VDEGS project , the possible effect of the 2-bp deletion on the stability of the SCARF2 transcript was investigated by RT-PCR . Total RNAs were isolated from skin samples from one affected and unaffected dog obtained in the postmortem autopsies . The cDNA synthesis was performed as described above . Forward 5-CAACCACGTCACTGGCAAGT-3 and reverse 5-TTACAGTGGGG CCCGTGG-3 primers were designed to amplify a 188-bp region between exon 6 and exon 8 of SLC37A2 . Semi-quantitative analysis of the expression was determined and visualized by electrophoresis . Permanent and deciduous incisor and premolar teeth were removed for therapeutic reasons from two Border Collie bitches with clinically affected teeth . The dogs were from the same litter . Three teeth , two incisors and one premolar , were obtained from one dog . One incisor and the premolar were processed to ground sections and the other incisor to paraffin sections . From the other dog , four teeth , two incisors and two premolars , were obtained . One incisor and one premolar were processed to ground sections and the other incisor and the other premolar to paraffin sections . For comparison , deciduous teeth obtained from a healthy Border Collie were studied . One incisor was processed to ground sections and one tooth to paraffin sections . The teeth to be processed to ground sections ( a procedure preserving the enamel with a proportionally high mineral content and sparse organic matrix ) were fixed with 10% neutral buffered formalin , dehydrated and embedded in liquid methylmethacrylate monomer . After complete polymerization , started up by benzoylperoxide ( 2 g/l ) , the teeth were serially cut to 100–150 μm thick longitudinal sections with a rotating diamond-coated saw microtome , let dry and mounted unstained with DePex ( Gurr , BDH , Poole , UK ) . For preparation to paraffin sections , the teeth were fixed with formalin , demineralized with EDTA ( 0 . 33 mol/l , pH 7 . 2 ) , which leads to the loss of the enamel , and embedded in paraffin . A representative series of sections were longitudinally cut at 7 μm and stained with haematoxylin and eosin ( HE ) .
|
Rare developmental disorders make a major contribution to pediatric hospital admissions and mortality . There is a growing interest in the development of therapeutics for these conditions , but that requires understanding of the genetic cause and pathology . Research can be facilitated by physiologically relevant models , such as dogs with corresponding disorders . We have characterized the clinical features and genetic causes of three developmental syndromes in dogs , including craniomandibular osteopathy ( CMO ) , a previously undescribed skeletal syndrome , and dental hypomineralization , for which we identified mutations in the canine SLC37A2 , SCARF2 and FAM20C genes , respectively . CMO is a clinical equivalent to an infantile cortical hyperostosis ( Caffey disease ) for which SLC37A2 is a new candidate gene . SLC37A2 is a glucose-phosphate transporter in osteoclasts , and its defect suggests an impaired glucose homeostasis in developing bone , leading to hyperostosis . Mutations in the SCARF2 and FAM20C genes have been associated with the human van den Ende-Gupta and Raine syndromes . Our study provides molecular identity for the canine conditions and presents three novel physiologically relevant models of human rare diseases .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome-wide",
"association",
"studies",
"animal",
"types",
"medicine",
"and",
"health",
"sciences",
"dentin",
"vertebrates",
"pets",
"and",
"companion",
"animals",
"dogs",
"animals",
"mammals",
"incisors",
"genome",
"analysis",
"mammalian",
"genomics",
"zoology",
"veterinary",
"science",
"digestive",
"system",
"veterinary",
"diseases",
"genomics",
"head",
"animal",
"genomics",
"teeth",
"anatomy",
"jaw",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"amniotes",
"organisms",
"human",
"genetics"
] |
2016
|
Molecular Characterization of Three Canine Models of Human Rare Bone Diseases: Caffey, van den Ende-Gupta, and Raine Syndromes
|
Bats are reservoirs for a wide range of zoonotic agents including lyssa- , henipah- , SARS-like corona- , Marburg- , Ebola- , and astroviruses . In an effort to survey for the presence of other infectious agents , known and unknown , we screened sera from 16 Pteropus giganteus bats from Faridpur , Bangladesh , using high-throughput pyrosequencing . Sequence analyses indicated the presence of a previously undescribed virus that has approximately 50% identity at the amino acid level to GB virus A and C ( GBV-A and -C ) . Viral nucleic acid was present in 5 of 98 sera ( 5% ) from a single colony of free-ranging bats . Infection was not associated with evidence of hepatitis or hepatic dysfunction . Phylogenetic analysis indicates that this first GBV-like flavivirus reported in bats constitutes a distinct species within the Flaviviridae family and is ancestral to the GBV-A and -C virus clades .
Bats ( order Chiroptera ) , after rodents , comprise the most diverse group of mammals with more than 1 , 100 species . They are present on six continents , often have substantial habitat overlap with humans [1] and harbor several zoonotic viruses causing significant human morbidity and mortality , including Ebola- and Marburgvirus , Nipah virus ( NiV ) , and SARS-like coronaviruses [2]–[5] . Proximity of bats to human populations may facilitate the zoonotic transmission of viruses either through direct contact , via amplifying domestic animal hosts , or through food-borne routes [6]–[8] . The current study was set up as part of a viral discovery effort to target key wildlife reservoirs in emerging disease hotspots . Bangladesh is a ‘hotspot’ for emerging zoonotic diseases [9] , with a relatively high diversity of wildlife that likely harbors new zoonotic pathogens , one of the densest human populations on the planet , and a high level of connectivity between people , domestic animals and wildlife . In Bangladesh and India , frugivorous Pteropus giganteus bats have been identified as a reservoir for NiV [10] , [11] , which has been recognized as the cause of several outbreaks of encephalitis [12]–[14] . Pteropus giganteus bats are common throughout the Indian subcontinent , living in close association with humans and feeding on cultivated fruit [14] . NiV transmission from bats to humans has been linked with the harvest and consumption of raw date palm sap , which becomes contaminated with bat feces , urine or saliva overnight when bats such as P . giganteus come to feed from the collecting pots [14] , [15] . Date palm sap or other foods eaten by both bats and people , may also serve as a vehicle for transmission of other bat-borne agents . Several zoonotic flaviviruses , including Japanese encephalitis virus , West Nile virus , and Kyasanur forest virus have been identified in bats; however , to date , GB viruses have not [1] . GB viruses A and C ( GBV-A and -C ) represent two recently identified species that are currently unassigned members of the family Flaviviridae [16] . GBV-A viruses have been described in New World primates and are not known to infect humans [17]–[19] , while GBV-C ( also known as Hepatitis G virus ( HGV ) ) have frequently been isolated from humans in many regions of the World , including India and Bangladesh [19]–[23] , and from wild chimpanzees ( Pan troglodytes ) in Africa [24] , [25] . Here we describe discovery of a virus in the serum of healthy bats in Bangladesh , tentatively named GB virus D ( GBV-D ) , that is distantly related to GBV-A and -C and represents a new member of the family Flaviviridae .
Every effort was made to minimize bat stress and avoid injury during capture , restraint , and sampling procedures . This study was conducted following Wildlife Trust institutional guidelines under IACUC approval G2907 issued by Tufts New England Medical Center , Boston , Massachusetts . As part of a longitudinal surveillance study of Nipah virus in bats , 98 free-ranging P . giganteus bats were caught from a colony of approximately 1800 individuals in the Faridpur district of Bangladesh in December 2007 ( Figure 1 ) . Each bat was anesthetized using isoflurane gas; morphometric measurements ( weight , forearm length , head length , and body condition ) were taken and bats were aged [10] . Each bat was marked for future identification using an RFID microchip ( AVID corp , www . avidid . com ) implanted subcutaneously between the scapulae . Three mL of blood were collected and placed into serum separator tubes ( vacutainer; Becton Dickinson , Franklin Lakes , NJ , USA ) . Serum was allowed to separate overnight at 4°C then drawn off without centrifugation and immediately frozen using a liquid nitrogen dry shipper . To inactivate potentially infectious agents , serum samples were heat-treated at 56°C for 30 min and then stored at −70°C . For RNA extraction , 250 µL of serum was added to 750 µL Tri-Reagent LS ( Molecular Research Center , Cincinnati , OH , USA ) . Saliva was collected from the bat's throat using a sterile cotton swab . Urine was collected either by catching urine in a 1 . 0 mL sterile cryovial while the bat was urinating , or by urethral swab . Urine and saliva swabs were immediately placed into 1 mL Tri-Reagent LS and frozen in liquid nitrogen . Total RNA from serum was extracted for UHTS analysis to screen for the presence of microorganisms . Five microliters of total RNA from each bat were combined into 4 pools: 4 pregnant bats; 4 non-pregnant female bats , and 2 pools of 4 adult male bats , respectively . Reverse transcription ( RT ) was performed on DNase I-treated ( DNA-free , Ambion Inc . , Austin , TX , USA ) RNA pools to generate cDNA using Superscript II RT ( Invitrogen , Carlsbad , CA , USA ) and random octamers linked to a defined arbitrary , 17-mer primer sequence tail ( MWG , Huntsville , AL , USA ) [26] . After RNase H treatment cDNA was amplified by the polymerase chain reaction ( PCR ) , applying a 9∶1 mixture of the defined 17-mer primer sequence and the random octamer-linked 17-mer primer sequence , respectively [27] . Products of >70 base pairs ( bp ) were selected by column purification ( MinElute , Qiagen , Hilden , Germany ) and ligated to specific linkers for sequencing on the 454 Genome Sequencer FLX ( 454 Life Sciences , Branford , CT , USA ) without DNA fragmentation [28] , [29] . Sequences were analyzed using software applications implemented at the GreenePortal website ( http://tako . cpmc . columbia . edu/Tools/ ) . Multiple forward and reverse primers for RT-PCR ( available upon request ) were designed using the sequences obtained by UHTS in order to fill gaps between fragments . Amplifications were performed with Bio-X-act ( Bioline , London , UK ) according to manufacturer's protocols . Products were size fractionated by electrophoresis and directly sequenced in both directions with ABI PRISM Big Dye Terminator 1 . 1 Cycle Sequencing kits ( Perkin-Elmer Applied Biosystems , Foster City , CA , USA ) at a commercial facility ( Genewiz , South Plainfield , NJ , USA ) . Additional methods applied to obtain the genome sequence included touch-down PCR [30] , 2-step walking PCR [31] , and 3′- and 5′- RACE ( Invitrogen ) . A real time Taqman PCR assay was developed to screen bat samples for GBV-D . Reactions were performed in a 25 µL volume by using commercial Taqman Universal Master Mix ( Applied Biosystems , Foster City , CA , USA ) . Primers and probe were designed to target a 60 nt region in the NS4A gene region: Fadi-forward , 5′- gCAgCTgCgTgTgCCA; Fadi-reverse , 5′- ACACCCATgATgTTACCACgAC; Fadi-probe , 5′- FAM- AggACCCggTCgCTCCAgCA-T-BQX ( TIB Molbiol , Adelphia , NJ , USA ) . Cycling conditions were: 50°C for 2 min , and 95°C for 10 min , followed by 45 cycles at 95°C for 15 sec and 60°C for 1 min . Thermal cycling was performed in an ABI 7300 real-time PCR system ( Applied Biosystems ) . A liver function panel was conducted at the International Center for Diarrheal Disease Research ( Dhaka , Bangaldesh ) using non heat-treated bat sera ( Automated Chemistry Analyzer AU 640 , Olympus Corporation , Tokyo , Japan ) . The following parameters were analyzed: total protein , albumin , globulin , albumin∶globulin ratio , total cholesterol , total bilirubin , alkaline phosphatase , alanine transferase , aspartate aminotransferase , gamma glutamyltransferase , and lactate dehydrogenase . Sequence alignments were generated with ClustalW software [32] and phylogenetic relationships deduced using Geneious software [33] . Statistical significance was assessed by bootstrap re-sampling of 1000 pseudoreplicate data sets . Sequence relations were determined from p-distance matrices calculated with pairwise deletion for missing data and homogeneous patterns among lineages based on ClustalW alignments as implemented in MEGA software [34] . Sliding window similarity analysis was performed using SimPlot [35] . Potential signalase cleavage sites , glycosylation sites , and phosphorylation sites were analyzed using the respective prediction servers available at the Center for Biological Sequence Analysis ( http://www . cbs . dtu . dk/services/ ) .
Total RNA from the serum of healthy bats captured at a roost in the Faridpur district of Bangladesh was extracted for UHTS analysis . Extracts of 16 individual bats were combined into 4 pools consisting of 4 pregnant adult bats , 4 non-pregnant adult female bats , or 2×4 adult male bats . Each pool yielded between 1 , 400 and 2 , 000 assembled contigs or singlton reads ( representing 50 , 000–75 , 000 reads ranging in size from 31–328 nt ) . Two reads of 238 and 215 nucleotides ( nt ) derived from the pregnant bat pool had distant homology to GBV-A sequences at the deduced amino acid ( aa ) level in the E2 and NS4A gene regions respectively ( BLASTX ) ; no homology was detected by searches at the nt level ( BLASTN; local copy of the executables with standard settings except that the reward for a nucleotide match was set to 2 instead of 1 ) . No viral sequences were detected in other pools at the nt or aa levels . Screening of the individual RNA preparations from the pregnant bat pool using primers derived from the UHTS reads confirmed the presence of the GBV-like sequence in the serum of bat 93 . A quantitative real time PCR assay indicated a load of approximately 30 000 RNA copies in bat-93 serum extract , and identified an additional 4 positive bat sera from the original 98 samples ( 5/98; 5% ) , indicating serum loads ranging from 350 to 70 , 000 RNA copies per assay . These positive samples came from male bats that were not included in the initial UHTS pools . Extracts of saliva from the five positive bats indicated a load of approximately 200 RNA copies in bat 93; no signal was obtained with urine extracts from the five positive bats . Near full-length genome sequence was generated from bat-93 and a second positive serum ( bat 68 ) , applying primers crossing gaps between UHTS reads as well as touch-down PCR [30] , 2-step walking PCR [31] , and 3′- and 5′-RACE ( Invitrogen ) protocols . The two genome sequences were 96% identical at the nt level ( GenBank Accession nos . GU566734 and GU566735 ) , indicating two strains of the same virus . Comparison of deduced polyprotein sequence to other GBV and hepaciviruses indicated highest nt and aa sequence identities to GBV-A and -C ( Table 1 , Figure 2 ) . The genomic sequence of the GBV-like virus identified in P . giganteus bats , tentatively named GBV-D , comprises 9 , 633 nt with 52 nt of potentially 5′-untranslated region ( UTR ) , one continuous open reading frame ( ORF ) of 9318 nt ( 3106 aa ) and 265 nt of 3′-UTR ( Figure 3 ) . Mature structural proteins in GB viruses , as well as other flaviviruses , are the product of cleavage by host signal peptidase [36] . In GBV-D the first potential signal sequence cleavage site is present after a stretch of 57 , largely basic aa ( 6 kDa , pI = 12 ) , followed by sequence homologous to E1 ( pfam 01539 , http://pfam . sanger . ac . uk/ ) ( Figure 3 ) . The single glycosylation site N177IT present in that sequence is located in a position comparable to GBV-C , -A , -B and HCV glycosylation sites . Identification of the downstream E2 termini is less apparent as the next 580 aa contain multiple potential signal sequences and 10 potential glycosylation sites that indicate no homology to hepaciviral E2/NS1 ( pfam 01560 ) , until the sequence aligns with N-terminal NS2 motifs ( pfam 01538 ) ( Figure 2 , Figure 3 ) . However , despite similarity to pfam 01538 no signal sequence compatible with cleavage at A759/A was found; cleavage may occur at G826/R , which combined with potential signalase cleavage at A584/F may indicate the existence of a heavily glycosylated potential 26 kDa product instead of the p7 trans-membrane protein identified in HCV [37]–[39] or the 13 kDa variant described in GBV-B [40] , [41] . Conserved C-terminal motifs of the autocatalytic NS2/NS3 endoprotease domain are compatible with NS2/NS3 cleavage at S1067/A and comparable to other GBV and HCV [42] . Figure 3 indicates potential cleavage sites for NS3 ( peptidase S29 , pfam 02907; DEAD box helicase , pfam 07652; helicase C , pfam 00271 ) , NS4A ( pfam 01006 ) , NS4B ( pfam 01001 ) , NS5A ( domain-1a zinc finger , pfam 08300; domain-1b , pfam 08301 ) , and NS5B ( pfam 00998 ) . Conserved aa motifs were recognized in NS proteins . RNA-dependent RNA polymerase ( RdRp ) motifs in RdRp block III that are conserved with respect to other GBV and hepaciviruses were identified in NS5B ( Figure 3 ) [43]–[46] . Potential phosphorylation sites are present at multiple serine ( 9 ) , threonine ( 14 ) and tyrosine ( 4 ) residues in NS5A , compatible with its possible function as a phosphorylation-regulated mediator of viral replication [47] . However , significant conservation of primary sequence is not obvious for phosphorylation sites , proline-rich , or interferon-sensitivity determining region motifs [48]–[50] . The C-terminal portion of NS3 has homology to conserved NTPase/helicase motifs [51]; the N-terminal portion includes conserved active triad residues H1123 , D1147 , S1204 of serine protease [52] , the viral protease responsible for cleavage of mature non-structural proteins [53] . Likewise , the active triad H991 , E1011 , C1032 of the cis-acting protease activity in the C-terminal portion of NS2 is conserved with respect to other GBV and HCV [42] . The only other discernable motif identified was a well-conserved N75 C/D C motif at the N-terminus of E1 ( Figure 3 ) [54] . Phylogenetic analysis of GBV-D was performed in comparison to selected representatives of GBV-A , GBV-B , GBV-C and HCV . Analysis of NS5B aa sequence ( Figure 4A ) confirmed a closer relationship of GBV-D to GBV-A and -C than to GBV-B or HCV as also indicated by pairwise sequence comparisons ( Table 1 ) . The same relationships were also apparent when NS3 , or the complete polyprotein sequence were analyzed ( Figure 4B and C , respectively ) . All three trees show GBV-D consistently at the root of the GBV-A/-C viruses , indicating an independent phylogenetic clade compatible with a separate species distinct from the recently created genus Hepacivirus [16] . A liver serum chemistry panel was conducted on sera from 15 bats , the five GBV-D infected and 10 non-infected animals . Standard assays to detect hepatitis and/or impaired liver function were performed [55] . Levels of total protein , alanine transferase , aspartate aminotransferase and total cholesterol were within published ranges reported for P . giganteus , except for bat 33 ( infected ) and bat 73 ( uninfected ) , which had modest elevation in aspartate aminotransferase . Reference values for albumin , globulin , albumin∶globulin ratio , total bilirubin , alkaline phosphatase , gamma glutamyltransferase and lactate dehydrogenase are not available for P . giganteus , however , values were comparable to those reported for other Pteropus species [56] . Mean values did not significantly differ between infected and uninfected bats ( Table 2 ) .
Molecular analyses of sera from Pteropus giganteus bats from Faridpur , Bangladesh led to the identification of a 9 , 633 nt sequence consistent in genomic organization with known GBV and other species within the family Flaviviridae [16] . Whereas previous studies of bats have employed assays that test for known pathogens , ours is the first report of an unbiased molecular approach to pathogen discovery in this important reservoir of emerging infectious diseases . The modest yield of novel microbial sequences may reflect the choice of sample ( e . g . , serum vs feces , tissue or another specimen ) , competition between host and microbial template during unbiased amplification , or both . Efforts to address template competition are under way that include subtraction of host nucleic acids or the use of semi-random primers that do not amplify host sequences . Such efforts will likely enhance the sensitivity and throughput of unbiased sequencing technologies for pathogen discovery . The discovery of this chiropteran flavivirus broadens both the taxonomical and geographical distribution of GB-like viruses . Three types of GB viruses have been described: GBV-A , -B and -C [18] , [19] , [24] , [25] , [54] , [57] . GBV-B , which has never been found in humans and was only reported in captive tamarins after serial passage of the original human GB serum [58] , is most closely related to HCV and was recently classified together with HCV into a new genus , Hepacivirus , within the family Flaviviridae [16] . GBV-A and -C remain unclassified members of the family . GBV-A have been isolated from several New World monkeys . Different genotypes appear to be associated with specific monkey species of the genera Saguinus , Callithrix ( Callitrichidae family ) and Aotus ( Aotidae family ) , without any clinical signs associated with infection [24] , [54] , [57] . GBV-C have been isolated from humans with non-A-E hepatitis; however , its pathogenicity is unknown and the virus is widespread in the human population [21] , [59]–[61] . Population studies showed that GB viruses are enzoonotic and species-specific within both Old and New World nonhuman primates as well as humans , and have likely co-evolved with their hosts over long periods of time [62] . Previously , the only GBV found in the Old world was GBV-C from chimpanzees ( in Africa ) and humans . Although GBV-C were found in humans , GB viruses have not been previously reported in primates or other animals on the Indian subcontinent . GBV-C and -A are remarkable for a truncated or missing capsid ( C ) protein [18] , [19] . Due to exhaustion of our samples we were unable to complete assessment of the 5′-terminal sequence; nonetheless , RACE experiments suggest that GBV-D likely codes for a short basic peptide , instead of a full-length C protein . The first methionine ( M1 ) predicts a peptide of 57 aa ( pI = 12 ) ; however , the more favorable Kozak context [63] of M3 indicates a 55 aa peptide . After signalase cleavage from the polyprotein precursor , this peptide may be functional , possibly influencing maturation of , or directly binding to , the E1 and/or E2 glycoproteins . Phylogenetic analyses of NS5B , NS3 and complete polyprotein sequence place GBV-D at the root of the GBV-A and -C clades and are consistent with a model wherein GBV-D is ancestral to GBV-A and -C clades . Mixed relationships indicative of recombination events [64] were not evident ( Figure 2 , Figure 4 ) . Both pteropid bats and chimpanzees are restricted to the Old World . While the range of chimpanzees ( Africa ) and P . giganteus ( the Indian subcontinent ) do not overlap , it is possible that other primate species in Bangladesh or India , such as macaques , or other fruit bats in Africa such as Eidelon spp . , whose range overlaps that of chimpanzees , may carry related viruses . While GBV-A is only known from primates of the New World , an African origin has been suggested for GBV-C based on a 12-aa indel sequence in NS5A [65] . Although the NS5A sequence of GBV-D , similar to that of GBV-A , appears elongated in the indel region , compatible with their respective earlier phylogenetic branching compared to GBV-C , little sequence conservation is observed in that region . The bats in this study , like primates infected with their associated GBV [66] , all appeared to be healthy . The lack of chemical evidence of hepatic inflammation or dysfunction suggests that this virus may not target hepatic cells in bats . This is consistent with the behavior of GBV-A in its natural primate hosts [54] . In contrast , elevated alanine transferase levels and mild hepatitis are observed in experimental infections of macaques with GBV-C isolates from humans [67] . Five percent of the bats we studied were infected with one of at least two different strains of GBV-D , which suggests widespread viral circulation within this species . The observation that bats are asymptomatically infected with diverse strains that constitute a distinct phylogenetic clade is compatible with a co-evolutionary relationship between GBV and their hosts [57] , [62] , and supports the hypothesis that P . giganteus bats may be a natural reservoir for GBV-D . In one case we were able to detect GBV-D nucleic acid in saliva . This suggests a potential route for viral transmission via fighting or grooming behavior , or via food shared by bats . Pteropus giganteus is a frugivorous bat species that carries NiV , a zoonotic paramyxovirus [10] , [11] . This species lives in close association with humans in Bangladesh and bats have been observed drinking from ( and urinating into ) date palm sap collecting pots [14] . Human consumption of contaminated palm juice is proposed to be a major route of NiV transmission [68] . Although it is unclear whether infectious virus was present in bat saliva , the observation that saliva can contain GBV-D nucleic acids provides a biologically plausible mechanism for transmission from infected bats to other hosts . While it is currently unknown whether GBV-D virus occurs in humans , up to 20% of non-A-E hepatitis cases remain unexplained [19] .
|
Bats are important reservoirs for emerging zoonotic viruses with significant impact on human health including lyssaviruses , filoviruses , henipaviruses and coronaviruses . Opportunities for transmission to humans are particularly prominent in countries like Bangladesh , where people live in close association with bats . Whereas previous studies of bats have employed assays that test for known pathogens , we present the first application of an unbiased molecular approach to pathogen discovery in this reservoir for emerging zoonotic disease . Unbiased pyrosequencing of serum from Pteropus giganteus bats enabled identification of a novel flavivirus related to Hepatitis C and GB viruses . Viral nucleic acid was present in 5 of 98 ( 5% ) sera , and in the saliva of one animal . Sequence identification of two strains of the virus , tentatively named GBV-D , suggests P . giganteus as a natural reservoir . Detection of viral nucleic acid in saliva provides a plausible route for zoonotic transmission . Phylogenetic analysis indicates that GBV-D is ancestral to GBV-A and -C , and separate from the recently classified genus Hepacivirus . Our findings provide new insight into the range of known hosts for GB-like viruses and demonstrate the power of unbiased sequencing to characterize the diversity of potentially zoonotic pathogens carried by bats and other reservoirs .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"ecology/evolutionary",
"ecology",
"genetics",
"and",
"genomics/comparative",
"genomics",
"computational",
"biology/sequence",
"motif",
"analysis",
"computational",
"biology/comparative",
"sequence",
"analysis",
"computational",
"biology/molecular",
"genetics",
"virology/emerging",
"viral",
"diseases",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"infectious",
"diseases/viral",
"infections",
"physiology/gastroenterology",
"and",
"hepatology",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"pathology/clinical",
"chemistry",
"genetics",
"and",
"genomics/bioinformatics"
] |
2010
|
Identification of GBV-D, a Novel GB-like Flavivirus from Old World Frugivorous Bats (Pteropus giganteus) in Bangladesh
|
Prion diseases are characterised by the accumulation of PrPSc , an abnormally folded isoform of the cellular prion protein ( PrPC ) , in affected tissues . Following peripheral exposure high levels of prion-specific PrPSc accumulate first upon follicular dendritic cells ( FDC ) in lymphoid tissues before spreading to the CNS . Expression of PrPC is mandatory for cells to sustain prion infection and FDC appear to express high levels . However , whether FDC actively replicate prions or simply acquire them from other infected cells is uncertain . In the attempts to-date to establish the role of FDC in prion pathogenesis it was not possible to dissociate the Prnp expression of FDC from that of the nervous system and all other non-haematopoietic lineages . This is important as FDC may simply acquire prions after synthesis by other infected cells . To establish the role of FDC in prion pathogenesis transgenic mice were created in which PrPC expression was specifically “switched on” or “off” only on FDC . We show that PrPC-expression only on FDC is sufficient to sustain prion replication in the spleen . Furthermore , prion replication is blocked in the spleen when PrPC-expression is specifically ablated only on FDC . These data definitively demonstrate that FDC are the essential sites of prion replication in lymphoid tissues . The demonstration that Prnp-ablation only on FDC blocked splenic prion accumulation without apparent consequences for FDC status represents a novel opportunity to prevent neuroinvasion by modulation of PrPC expression on FDC .
Prion diseases ( Transmissible spongiform encephalopathies; TSE ) are sub-acute neurodegenerative diseases that affect both humans and animals . Many prion diseases , including natural sheep scrapie , bovine spongiform encephalopathy , chronic wasting disease in mule deer and elk , and kuru and variant Creutzfeldt-Jakob disease in humans , are acquired by peripheral exposure ( eg: orally or via lesions to skin or mucous membranes ) . After peripheral exposure prions accumulate first upon follicular dendritic cells ( FDC ) as they make their journey from the site of infection to the CNS ( a process termed , neuroinvasion ) [1]–[7] . FDC are a unique subset of stromal cells resident within the primary B cell follicles and germinal centres of lymphoid tissues [8] . Prion accumulation upon FDC is critical for efficient disease pathogenesis as in their absence neuroinvasion are impaired [1]–[4] . From the lymphoid tissues prions invade the CNS via the peripheral nervous system [9] . During prion disease aggregations of PrPSc , an abnormally folded isoform of the cellular prion protein ( PrPC ) accumulate in affected tissues . Prion infectivity co-purifies with PrPSc [10] and is considered to constitute the major , if not sole , component of infectious agent [11] . Host cells must express cellular PrPC to sustain prion infection [12] and FDC appear to express high levels of PrPC on the cell membrane in uninfected mice [13] , [14] . Although prion neuroinvasion from peripheral sites of exposure is dependent upon the presence of FDC in lymphoid tissues , it is not known whether FDC actually replicate prions themselves . FDC characteristically trap and retain native antigen on their surfaces for long periods in the form of immune complexes , consisting of antigen-antibody and/or complement components . Prions are also considered to be acquired by FDC as complement-opsonized immune complexes [15]–[18] . Thus , during prion infection FDC might simply trap and retain PrPSc-containing immune complexes on their surfaces following synthesis by other infected cells such as neurones . Many cell types including classical DC , lymphocytes , mast cells , platelets , reticulocytes and epithelial cells secrete membrane vesicles termed exosomes that are enriched in cell-specific protein [19] , [20] . Although the functions of exosomes are uncertain FDC can bind them on their surfaces . These microvesicles permit FDC to passively acquire and display proteins on their surfaces that they do not express at the mRNA level [21] . Studies have shown that prions only accumulated in the spleens of mice in which the FDC-containing stromal compartment expressed PrPC [13] , [14] . However , in each of those studies it was not possible to dissociate the Prnp expression status of the FDC from that of the nervous system and all other host-derived non-haematopoietic and stromal cell populations [13] , [14] , [22] . This is important as prion infection can occur within inflammatory PrPC-expressing stromal cells that are distinct from FDC [23] . Furthermore , as both PrPC and PrPSc can be released from cells in association with exosomes [20] FDC may passively acquire PrPC and prions after release in exosomes from other infected cells [24] , [25] . No therapies are available to treat prions diseases . A thorough characterization of the host cells that are infected by prions is imperative for the identification of candidate molecular targets for therapeutic intervention , the development of useful pre-clinical diagnostics and to aid our understanding of the risk of transmission . To definitively determine the role of FDC in prion pathogenesis , two unique compound transgenic mouse models were created in which PrPC expression was specifically “switched on” or “switched off” only on FDC . These mice were then used to establish: i ) whether FDC express PrPC or simply acquire it from other host cells; and ii ) whether FDC amplify prions , or simply acquire them from other infected host cells . Our data clearly show that PrPC-expressing FDC alone are sufficient to sustain prion replication in the spleen . Furthermore , prion replication in the spleen is blocked in mice in which PrPC-expression is specifically ablated only on FDC .
To study FDC-specific gene function transgenic mice were used that expressed Cre recombinase under the control of the Cr2 locus ( CD21-Cre mice ) which directs expression in FDC and mature B cells [26] , [27] . First the cellular specificity of the Cre recombination was assessed by crossing the CD21-Cre mice with the ROSA26flox/flox reporter strain [28] . Histological analysis showed efficient LacZ expression indicative of Cre-mediated gene recombination in FDC and B cell follicles in the spleens , lymph nodes and Peyer's patches of CD21-Cre ROSA26flox/flox mice ( Figure 1A , B ) . No recombination was observed in FDC and mature B cells in the spleens of ROSA26flox/flox reporter mice that lacked Cre expression ( Figure 1B ) . Unlike lymphocytes , FDC do not derive from bone marrow precursors [29] . As a consequence , it is possible to mix-and-match the genotype of FDC and lymphocytes by grafting bone marrow cells from donor mice into recipients of a different genetic background [13] , [14] , [22] . To restrict Cre-expression to FDC , adult CD21-Cre ROSA26flox/flox mice were lethally γ-irradiated and 24 h later reconstituted with bone marrow from Cre-deficient C57BL/6 wild-type ( WT ) mice ( termed WT→CD21-Cre ROSA26flox/flox mice ) and tissues from six mice from each group analysed 100 days after transfusion . Using this approach , in these mice all B cells lack Cre-expression as they derive from the WT donor bone marrow , whereas the FDC express Cre as they are host-derived . Analysis of the cellular sites of LacZ expression in WT→CD21-Cre ROSA26flox/flox mice confirmed that Cre-mediated recombination was associated with FDC ( Figure 1B ) . No other cellular sites of Cre-mediated recombination were observed in the spleens of WT→CD21-Cre ROSA26flox/flox mice . Furthermore , no other cellular sites of Cre-mediated recombination were observed in a wide range of non-lymphoid peripheral tissues from CD21-Cre ROSA26flox/flox and WT→CD21-Cre ROSA26flox/flox ( heart , liver , kidney , pancreas , ear , tongue , skeletal , muscle , ovary , uterus , bladder , testes , epididymis , sciatic nerve and spinal cord; data not shown ) . These data clearly demonstrate that CD21-Cre mice are a useful tool to study FDC-specific gene expression and function . Cre toxicity can occur in some Cre transgenic mouse lines whereby Cre recombinase causes mis-recombination , DNA damage and death of Cre-expressing cells [30] . However , immunohistochemical ( IHC ) analysis of spleens from CD21-Cre ROSA26flox/flox mice and WT→CD21-Cre ROSA26flox/flox mice showed no significant effect of Cre-expression on the status of FDC networks and B cell follicles when compared to spleens from WT control mice and ROSA26flox/flox mice that lacked Cre expression ( Figure 1C ) . Furthermore , the expression of Cre recombinase under the control of the Cr2 locus had no observable effect on CD21/35 expression ( Figure 1C ) . Next , mice were created in which Prnp expression ( which encodes PrPC ) was restricted only to FDC . To do so , CD21-Cre mice were first bred onto a PrPC-deficient ( Prnp-/- ) background . The resulting CD21-Cre Prnp-/- mice were then crossed with Prnpstop/- mice in which a floxed β-geo stop cassette was inserted into intron 2 of the Prnp gene upstream of exon 3 [31] . In the progeny CD21-Cre Prnpstop/- mice , PrPC is only expressed in cells expressing Cre recombinase ( CD21-expressing FDC and mature B cells ) . To restrict the Prnp-expression to FDC , CD21-Cre Prnpstop/- mice were lethally γ-irradiated and grafted with bone marrow from Cre-deficient Prnpstop/- mice ( Prnpstop/-→CD21-Cre Prnpstop/- mice ) . We also performed bone marrow transfers from CD21-Cre Prnpstop/- donors into CD21-Cre Prnpstop/- recipients ( CD21-Cre Prnpstop/-→CD21-Cre Prnpstop/- mice ) , CD21-Cre Prnpstop/- donors into Cre-deficient Prnpstop/- mice ( CD21-Cre Prnpstop/-→Prnpstop/- mice ) and Prnp+/- donors into Prnp+/- recipients ( Prnp+/-→Prnp+/- mice ) as controls ( Figure 2A ) . Spleens , tails and blood from six mice from each group were examined 100 days after bone marrow transfusion . PCR analysis of DNA isolated from the tails , blood and spleens of mice in each group was used to confirm the presence of Cre ( Figure 2B , upper panel ) and Cre-mediated DNA recombination ( Figure 2B , lower panel ) within the stromal , haematopoietic or both compartments ( respectively ) . The detection of Cre in the tail and spleen but not blood of the Prnpstop/-→CD21-Cre Prnpstop/- mice confirmed the restriction of the Cre-expression to the stromal but not haematopoietic compartments of these mice . In addition , PCR analysis also confirmed that in these mice efficient Cre-mediated recombination of the Prnpstop/- allele was restricted to the FDC-containing stromal compartment of the spleen ( Figure 2B ) . In Prnpstop/-→CD21-Cre Prnpstop/- mice the recombined Prnpstop/- allele ( Prnpstop ( R ) ) was detected in the spleen , but not blood and tail . Thus these data indicate that in the spleens of Prnpstop/-→CD21-Cre Prnpstop/- mice Cre-mediated recombination is restricted to FDC and not B cells . As anticipated , in the spleens of Prnp+/-→Prnp+/- control mice high levels of PrPC expression were observed upon FDC and tyrosine hydroxylase ( TH ) -positive sympathetic nerves ( Figure 2C ) . In contrast , in the spleens of Prnpstop/-→CD21-Cre Prnpstop/- mice PrPC was only expressed on FDC ( Figure 2C ) . In the absence of Cre-recombinase expression by FDC and peripheral nerves in CD21-Cre Prnpstop/-→Prnpstop/- mice , PrPC expression was not expressed by either cell population ( Figure 2C ) . Morphometric analysis confirmed that the amount of the PrPC expression co-localized upon the surfaces of FDC in the spleens of Prnpstop/-→CD21-Cre Prnpstop/- mice was not significantly different from that observed upon FDC in spleens from Prnp+/-→Prnp+/- control mice ( P<0 . 69 , n = 48 FDC/group; Figure 2D ) . In contrast , in the absence of Cre-recombinase expression by FDC in CD21-Cre Prnpstop/-→Prnpstop/- mice , PrPC expression was substantially lower than that observed upon FDC in spleens from Prnp+/-→Prnp+/- control mice ( P<1×10-25 , n = 48; Figure 2D ) . Morphometric analysis also confirmed that PrPC expression upon the surfaces of sympathetic nerves in the spleens of Prnpstop/-→CD21-Cre Prnpstop/- , CD21-Cre Prnpstop/-→ CD21-Cre Prnpstop/- and CD21-Cre Prnpstop/-→Prnpstop/- mice was significantly ablated when compared to that observed upon sympathetic nerves in spleens from Prnp+/-→Prnp+/- control mice ( p<1×10−25 , n = 48 sympathetic nerves/group ) . Together , these data confirm that in the spleens of Prnpstop/-→CD21-Cre Prnpstop/- mice PrPC expression is specifically restricted to FDC , whereas in spleens from CD21-Cre Prnpstop/-→Prnpstop/- mice , FDC lack PrPC expression . FDC can passively acquire the expression of some surface molecules including MHC class II and complement component C4 [21] , [32] . However , these data confirm that FDC express high levels of cellular PrPC on their surfaces and do not simply acquire it from neighbouring cells . IHC analysis confirmed that the microarchitecture ( Figure 3A ) , size ( P = 0 . 755 , n = 32; Figure 3B ) and number ( P = 0 . 249 , n = 32; Figure 3C ) of the FDC networks in spleens from mice with Prnp-expression restricted to FDC ( Prnpstop/-→CD21-Cre Prnpstop/- mice ) were normal when compared to control mice . Other studies have shown that the density of sympathetic nerves can significantly influence the amount of prion accumulation within in the spleen [33] . Quantitative analysis of the relative positioning of FDC and sympathetic nerves showed there were no significant differences in average distance between these cell populations in spleens from each mouse group ( Figure 3D & E; P<0 . 932 , n = 48 ) . Next , we determined the effect of FDC-restricted Prnp-expression on prion replication in the spleen . In this study , the normal cellular form of the prion protein is referred to as PrPC , and two distinct terms ( PrPSc or PrPd ) are used to describe the disease-specific , abnormal accumulations of PrP that are characteristically found only in prion-affected tissues and considered a reliable biochemical marker for the presence of infectious prions [10] . Disease-specific PrP ( PrPd ) accumulations are relatively resistant to proteinase K ( PK ) digestion , whereas cellular PrPC is destroyed . Where we were able to confirm this resistance by treatment of samples with PK and subsequent paraffin-embedded tissue ( PET ) immunoblot analysis [34] , PrPSc is used as a biochemical marker for the presence of prions . Unfortunately , treatment of tissue sections with PK destroys the microarchitecture . Therefore , for IHC analysis tissue sections were fixed and pre-treated to enhance the detection of the disease-specific abnormal accumulations of PrP ( PrPd ) , whereas cellular PrPC is denatured by these treatments [4] . We have repeatedly shown in a series of studies that these PrPd-accumulations occur only in prion-infected tissues , and correlate closely with the presence of ME7 scrapie prions [1] , [4] , [13] , [35]–[37] . Within weeks after i . p . exposure of WT mice to ME7 scrapie prions , strong accumulations of prion-specific PrPSc occur upon FDCs within the spleen and are sustained until the terminal stages of disease [1] , [13] , [35] . Here , mice were injected i . p . with ME7 scrapie prions and spleens from 4 mice from each group collected 35 , 70 and 105 days after exposure . In spleens from control mice ( Prnp+/-→Prnp+/- mice ) heavy PrPd accumulations , consistent with localisation upon FDC , were detected at 70 days after i . p . injection with the scrapie agent and had increased in intensity by 105 days after infection ( Figure 4A & B ) . PET immunoblot confirmed the presence of PrPSc upon the surfaces of the FDC in spleens from control mice ( Figure 4C ) . Furthermore , in the spleens of Prnpstop/-→CD21-Cre Prnpstop/- mice in which cellular PrPC was expressed only on FDC , heavy PrPSc accumulations were likewise maintained upon FDC ( Figure 4A & B ) . In contrast , in the absence of PrPC expression by FDC in the spleens of CD21-Cre Prnpstop/-→Prnpstop/- mice , no PrPSc accumulations were observed upon FDC . In the spleens of mice with PrPC-deficient FDC , if PrP was detected at all , it was only occasionally observed within tingible body macrophages ( Figure 4A and B , arrowheads; Figure S1 ) . We also analysed prion infectivity levels in spleens collected 70 days after infection from control mice ( Prnp+/-→Prnp+/- mice ) and Prnpstop/-→CD21-Cre Prnpstop/- mice in which cellular PrPC was expressed only on FDC ( Figure S2; n = 3/group ) . As anticipated high levels of prion infectivity were observed in each control spleen . Furthermore , consistent with data above our analysis showed that PrPc expression only of FDC was sufficient to sustain high levels of prion infectivity within the spleen ( Figure S2 ) . These data demonstrate that PrPC expression only on FDC is sufficient to sustain prion replication in the spleen . In the absence of PrPC expression on FDC the prions appeared to be scavenged by tingible body macrophages resident within the B cell follicles . Next , mice were created in which Prnp expression was specifically ablated in FDC . To do so , CD21-Cre Prnp-/- mice were crossed with mice carrying a “floxed” Prnp gene ( Prnpflox/flox mice; [31] ) . In the progeny CD21-Cre Prnpflox/- mice , Prnp expression is conditionally ablated in cells expressing Cre recombinase ( CD21-expressing FDC and mature B cells ) . To restrict the Prnp-ablation to FDC , CD21-Cre Prnpflox/- mice were lethally γ-irradiated and grafted with bone marrow from Cre-deficient Prnpflox/- mice ( Prnpflox/-→CD21-Cre Prnpflox/- mice ) . We also performed bone marrow transfers from CD21-Cre Prnpflox/- donors into CD21-Cre Prnpflox/- recipients ( CD21-Cre Prnpflox/- mice→CD21-Cre Prnpflox/- mice ) , CD21-Cre Prnpflox/- donors into Cre-deficient Prnpflox/- mice ( CD21-Cre Prnpflox/-→Prnpflox/- mice ) , and Prnp+/- donors into Prnp+/- recipients ( Prnp+/-→Prnp+/- mice ) as controls ( Figure 5A ) . Spleens , tails and blood from 6 mice from each group were examined 100 days after bone marrow transfusion . PCR analysis of DNA isolated from the spleens , blood and tails of Prnpflox/-→CD21-Cre Prnpflox/- mice confirmed that efficient Cre-mediated DNA recombination and Prnp-ablation was restricted to the FDC-containing stromal compartment of the spleen ( Figure 5B ) . In Prnpflox/-→CD21-Cre Prnpflox/- mice the recombined Prnpstop/- allele ( Prnpdeflox ) was detected in the spleen , but not blood and tail . Thus these data indicate that in the spleens of Prnpstop/-→CD21-Cre Prnpstop/- mice Cre-mediated recombination and Prnp-ablation is restricted to FDC and not B cells . IHC analysis showed that in the spleens of Prnpflox/-→CD21-Cre Prnpflox/- mice and CD21-Cre Prnpflox/- mice→CD21-Cre Prnpflox/- mice FDC did not express PrPC whereas high levels were associated with TH-positive sympathetic nerves ( Figure 5C ) . In the absence of Cre-recombinase expression by FDC in CD21-Cre Prnpflox/-→Prnpflox/- mice , high levels of PrPC were expressed by FDC and sympathetic nerves ( Figure 5C ) . Morphometric analysis confirmed that the magnitude of the PrPC expression co-localized upon the surfaces of FDC in the spleens of Prnpflox/-→CD21-Cre Prnpflox/- mice and CD21-Cre Prnpflox/- mice→CD21-Cre Prnpflox/- mice was substantially lower than that observed upon FDC in spleens from Prnp+/-→Prnp+/- control mice ( P<1×10-24 and P<1×10−23 , respectively , n = 48 FDC/group ) and not significantly different when compared to background levels ( Figure 5D ) . In contrast , in the absence of Cre-recombinase expression by FDC in CD21-Cre Prnpflox/-→Prnpflox/- mice , PrPC expression was not significantly different from the level observed upon FDC in spleens from Prnp+/-→Prnp+/- control mice ( P<0 . 106; Figure 5D ) . In contrast , morphometric analysis showed that the magnitude of the PrPC expression co-localized upon the surfaces sympathetic nerves in the spleens of Prnpflox/-→CD21-Cre Prnpflox/- , CD21-Cre Prnpflox/-→ CD21-Cre Prnpflox/- and CD21-Cre Prnpflox/-→Prnpflox/- mice was similar to that observed upon sympathetic nerves in spleens from Prnp+/-→Prnp+/- control mice ( p = 0 . 400 , n = 48 sympathetic nerves/group ) . Together , these data confirm that in the spleens of Prnpflox/-→CD21-Cre Prnpflox/- mice the Prnp ablation is specifically restricted to FDC . Data in the current study definitively demonstrate that FDC express high levels of PrPC but the role PrPC plays in FDC function and homeostasis is not known . IHC analysis showed that the microarchitecture of the FDC networks from Prnp-ablated Prnpflox/-→CD21-Cre Prnpflox/- mice were normal when compared to control mice ( Figure 6A ) . Furthermore , no significant difference was observed in the size ( P = 0 . 750 , n = 32 ) and number ( P = 0 . 713 , n = 32 of the FDC networks in spleens from each mouse group ( Figure 6B & C , respectively ) . The relative positioning of the FDC and sympathetic nerves was likewise similar in spleens from each mouse group ( Figure 6D & E; P<0 . 765 , n = 48 ) . FDC characteristically trap and retain native antigen on their surfaces in the form of immune complexes , consisting of antigen-antibody and/or complement components . Antigens trapped on the surface of FDC are considered to promote immunoglobulin-isotype class switching , affinity maturation of naïve IgM+ B cells and the maintenance of immunological memory [38]–[42] . Indeed , prions are also considered to be acquired by FDC as complement-opsonized immune complexes [15]–[18] . To determine whether antigen retention by Prnp-ablated FDC was affected six mice from each group were passively immunized with preformed PAP immune complexes , and 24 h later , the presence of FDC-associated immune complexes identified by IHC ( Figure 7 ) and the presence of peroxidase activity ( data not shown ) . No significant difference in the magnitude of immune complex trapping could be detected between FDC from Prnp-ablated Prnpflox/-→CD21-Cre Prnpflox/- mice and control mice ( Figure 7; P = 0 . 85 , n = 40/group ) . Together , these data demonstrate that Prnp-ablation does not impair FDC status or their ability to trap and retain immune complexes . Next , the effect of FDC-specific Prnp-ablation on prion replication by FDC was determined . Mice were injected i . p . with ME7 scrapie prions and spleens from 4 mice from each group collected 70 days after exposure . As anticipated , heavy PrPd ( Figure 8A and B ) and PrPSc ( Figure 8C ) accumulations consistent with localisation upon FDC were detected in spleens from control mice ( Prnp+/-→Prnp+/- mice ) and mice in which Prnp was ablated only in mature B cells ( CD21-Cre Prnpflox/-→Prnpflox/- mice ) . In the spleens in which cellular PrPC was ablated only on FDC ( Prnpflox/-→CD21-Cre Prnpflox/- mice ) , or FDC and mature B cells ( CD21-Cre Prnpflox/-→CD21-Cre Prnpflox/- mice ) , no PrP accumulations were observed upon FDC ( Figure 8A–C ) . Consistent with data above ( Figure 4 ) , in spleens of mice with PrPC-deficient FDC PrP accumulations were only occasionally observed within tingible body macrophages ( Figure 8A and B , arrowheads; Figure S1 ) . We also analysed prion infectivity levels in spleens from Prnpflox/-→CD21-Cre Prnpflox/- mice in which cellular PrPC expression was ablated only on FDC ( Figure S2; n = 3 ) . Consistent with data above this analysis showed that in the absence of PrPc expression only on FDC the accumulation of high levels of prion infectivity in the spleen was blocked ( Figure S2 ) . Taken together , these data show that in the specific absence of PrPC expression FDC are unable to sustain prion replication upon their surfaces and as a consequence the agent is scavenged by tingible body macrophages . When mice with PrPC-ablated FDC ( Prnpflox/-→CD21-Cre Prnpflox/- mice ) were injected intracerebrally ( i . c . ) with the ME7 scrapie agent strain directly into the CNS all mice succumbed to clinical signs of scrapie approximately 300 days after exposure with incubation periods indistinguishable from those of Prnp+/- control mice [43] ( Prnpflox/-→CD21-Cre Prnpflox/- , 297±4 days , n = 4; Prnp+/- , 290±4 days , n = 5; P = 0 . 386 ) . Histopathological analysis showed that brains from all clinically-affected mice from each group displayed the characteristic spongiform pathology , astrogliosis , microgliosis and PrPd accumulation typically associated with terminal infection with the ME7 scrapie agent ( Figure 9A ) . Following i . c . -injection with the ME7 scrapie agent , high levels of PrPSc accumulate upon FDC and are maintained for the duration of the incubation period [13] ( Figure 9B and C ) . However , FDC are not critical for ME7 scrapie pathogenesis when infection is established directly within the CNS [4] , [13] , [35] , [44] . In the spleens from clinically-scrapie affected mice in which PrPC expression was specifically ablated only on FDC ( Prnpflox/-→CD21-Cre Prnpflox/- mice ) , PrPSc replication upon FDC was also blocked . These data how that FDC do not simply trap and retain prions after their release from infected neurones in the CNS . These data also confirm that the Prnp-ablation in Prnpflox/-→CD21-Cre Prnpflox/- mice was specific to FDC and had no effect on prion neuropathogenesis and disease susceptibility when the infection was established directly in the CNS . Studies in mice show that efficient prion neuroinvasion from peripheral sites of exposure is dependent upon the presence of FDC in lymphoid tissues [1] , [4] , [35] , [44]–[46] . Next , the effect of FDC-specific Prnp-ablation on prion neuroinvasion via the peritoneal route was determined . Unfortunately , due to the advanced ages of the mice in this experiment , some succumbed to ageing-related inter-current illness . As there was a 100 days interval between the time of lethal γ-irradiation/bone marrow reconstitution and prion infection , many mice were approximately 500–600 days old when culled . However , most mice with PrPC-expressing FDC in their spleens succumbed to clinical prion disease after i . p . injection ( Prnp+/-→ Prnp+/- control mice , n = 5/7; CD21-Cre Prnpflox/-→Prnpflox/- mice , n = 3/6; Table S1 ) . Histopathological analysis showed that brains from all clinically-affected mice from these groups displayed the characteristic spongiform pathology , astrogliosis , microgliosis and PrPd accumulation typically associated with terminal infection with the ME7 scrapie agent ( Figure S3 , third and fourth columns ) . In contrast , none of the mice with PrPC-ablated FDC ( Prnpflox/-→CD21-Cre Prnpflox/- mice , n = 0/6; CD21-Cre Prnpflox/-→CD21-Cre Prnpflox/- mice , n = 0/7 ) succumbed to clinical prion disease during their life-spans ( Table S1 ) . Although we cannot exclude the possibility that if the clinically-negative mice with PrPC-ablated FDC mice had lived longer some may have succumbed to clinical prion disease after substantially extended incubation periods , no PrPd or other characteristic histopathological hallmarks of prion disease were detected in their brains ( Figure S3 , first two columns ) . Together , these data suggest that in the specific absence of PrPC expression on FDC neuroinvasion following peripheral exposure is impaired .
These data definitively demonstrate that FDC are essential sites of prion replication in lymphoid tissues . In order to precisely establish the role of FDC in prion pathogenesis two unique compound transgenic mouse models were created in which PrPC expression was specifically “switched on” or “off” only on FDC . Our data confirm that FDC express high levels of PrPC and do not simply acquire it from other host cells . Furthermore , we show that following peripheral exposure PrPC-expressing FDC alone are sufficient to sustain high levels of prion replication in the spleen . Accordingly , when PrPC-expression was specifically ablated only on FDC prion replication in the spleen was blocked . These data likewise demonstrate that FDC do not simply acquire prions after their release from other infected host cells . Our analysis showed that the effects of Prnp-ablation on prion replication in the spleen were specific to FDC and had no effect on prion neuropathogenesis when the infection was established directly in the CNS . In the absence of PrPC expression on FDC the PrPSc from the initial inoculum appeared to be scavenged by tingible body macrophages resident within the B cell follicles . Together , these data definitively demonstrate that FDC are the critical early sites of prion replication in lymphoid tissues . This study is the first to demonstrate that the specific ablation of a cellular protein only on FDC , without apparent consequences for FDC status and function , blocks the replication of an important pathogen in the spleen . FDC reside in the primary B cell follicles and germinal centres of lymphoid tissues and are a completely distinct cell lineage from bone-marrow-derived classical dendritic cells [47]–[49] . FDC possess many slender and convoluted dendritic processes which provide the FDC with an extremely large surface area . This helps the FDC to efficiently trap and retain large amounts of native antigen in the form of immune complexes , consisting of antigen-antibody and/or complement components . The longevity of FDC ensures that antigen is retained upon their surfaces for long periods [50] , [51] . Antigens trapped on the surface of FDC are considered to promote immunoglobulin-isotype class switching , affinity maturation of naïve IgM+ B cells and the maintenance of immunological memory [38]–[42] . FDC are also considered to aid the clearance of apoptotic B lymphocytes [52] , and play a role in infection with human immunodeficiency virus [53] and the pathogenesis of chronic inflammatory and autoimmune diseases [54] and peripherally-acquired prion infections . A number of studies have addressed the role of FDC in prion pathogenesis . They show that prion replication in the spleen and subsequent neuroinvasion are both impaired in immunodeficient mice that lack FDC [4] , [44] , [45] , or following their temporary de-differentiation [1] , [35] , [46] . Although the precise identity of FDC precursor cells is unknown , other studies have exploited their non-haematopoietic-origin to address their role in prion pathogenesis . In these bone marrow chimera studies , mismatches were created in Prnp expression between the FDC-containing stromal and haematopoietic compartments by grafting bone marrow cells from PrP-deficient ( Prnp-/- ) mice into PrP-expressing wild-type mice , and vice versa [13] , [14] . Using this approach FDC and all other stromal cells were derived from the recipient , whereas lymphocytes and other haematopoietic lineages were derived from the donor cells . Following peripheral exposure prion accumulation upon FDC was only detected in the spleens of mice with a Prnp-expressing stromal compartment . While the above studies clearly show that the presence of FDC is important for prion replication in the spleen , it was not possible to dissociate the Prnp expression status of FDC from that of the nervous system and all other non-haematopoietic host-cell populations and therefore precisely characterise the role of FDC in prion neuroinvasion [13] , [14] . This is important for a number of reasons . Firstly , prion infection can occur within inflammatory PrPC-expressing stromal cells that are distinct from FDC [23] . Secondly , the FDC's ability to bind exosomes may have lead to the wrong interpretation to be made in earlier studies describing their ontology [55] . Most evidence indicates that FDC do not derive from haematopoietic precursors [29] , [49] . However , the detection of donor bone marrow derived MHC class-I molecules , and other donor-derived antigens , on the surface of FDC in recipient mice was considered evidence of FDC precursor cells within bone marrow [55] . With hindsight these observations are most likely due to the FDC's capacity to acquire exosome-associated antigens from other cell types [21] . Both PrPC and PrPSc can be released from cells in association with exosomes [20] . The possibility , therefore , cannot be excluded that FDC passively acquire prions after their release in exosomes from other infected non-haematopoietic cell populations . Finally , FDC characteristically trap and retain immune complexes on their surfaces . FDC express negligible levels of complement component C4 at the mRNA level but the detection of abundant activated C4 on their surfaces by IHC using mAb FDC-M2 ( as used in this study ) is indicative of the capture and retention of immune complexes by FDC [32] . Opsonising complement components and cellular CR are likewise considered to play an important role in the retention of prions by FDC [15] , [16] , [18] . Thus FDC may simply act as concentrating depots for prion-containing complement-opsonized immune complexes . The practical hurdles that are encountered when attempting to isolate highly purified FDC from lymphoid tissues have made detailed analysis of their pathobiological functions extremely difficult . The main issues include: contamination with other cell types such as B cells and tingible body macrophages which express MFGE8 ( FDC-M1 ) , a common marker used to identify FDC [52] , [56] , low yield , and their dependence on constitutive lymphotoxin β receptor-stimulation to maintain their differentiated state [57] . FDC and mature B cells express high levels of Cr2 which encodes the complement receptors CR2/CR1 ( CD21/35 ) [18] , [27] . A previous study used CD21-cre mice to study FDC-specific gene function [27] . In the current study , our data confirm that Cre/loxP-mediated DNA recombination was specific to FDC and mature B cells in CD21-cre mice , and could be restricted to FDC by transfusing the mice with Cre-deficient bone marrow . In some Cre transgenic mouse lines Cre-toxicity is encountered whereby Cre recombinase causes mis-recombination , DNA damage and death of Cre-expressing cells [30] . However , our analysis suggested no significant effect of Cre-expression on the number , size and status of FDC networks and B cell follicles . CD21-Cre mice are therefore a powerful in vivo tool in which to study FDC-specific gene expression and function . Expression of PrPC is mandatory for host cells to sustain prion infection [43] . In the current study to establish whether FDC actively amplify prions a compound transgenic mouse model was created using the CD21-cre mouse line to specifically “switch on” PrPC expression only on FDC ( Prnpstop/-→CD21-Cre Prnpstop/- mice ) . As a consequence , only FDC in these mice had the potential to be actively infected with and replicate prions . Our analysis showed that expression of PrPC only on FDC was sufficient to sustain high levels of PrPSc accumulation upon FDC in the spleen after peripheral prion exposure . These data definitively demonstrate that FDC are the critical sites of prion replication in lymphoid tissues . Ultrastructural analysis of the cellular compartments within which PrPd localizes upon/within FDC has failed to show any intracellular accumulation . Instead the PrPd appears to be restricted to the plasmalemma of their dendritic processes [58] . This implies that early de novo PrPSc conversion occurs upon the surface of FDC . A second compound transgenic mouse model was created in which PrPC expression was specifically “switched off” only on FDC ( Prnpflox/-→CD21-Cre Prnpflox/- mice ) . If , as shown above , FDC do actively amplify prions , then one would also expect the specific ablation of PrPC expression only on FDC to block prion replication in the spleen . Our data confirmed this to be the case . As PrPC expression in all other host cells ( eg: neurones ) in these mice was unaffected , these data clearly show that FDC do not simply acquire prions following release from other infected host cells , even in mice with clinical prion disease in the brain . IHC analysis implied that in the spleens of mice with PrP-deficient FDC the prions appeared to be scavenged by tingible body macrophages resident within the B cell follicles . The lack of detection of PrPd within tingible body macrophages in the spleens of clinically-affected mice with PrP-deficient FDC ( Figure 9 ) clearly demonstrates that these cells are not alternative sites of replication of ME7 scrapie prions . High levels of prions rapidly accumulate within the spleen and other lymphoid tissues within weeks of peripheral exposure . The magnitude of the prion accumulation within the spleen rapidly reaches a plateau level which is maintained for the duration of the disease [13] , [44] . The maintenance of this plateau may be the consequence of a competitive state whereby FDC act to amplify prions above the threshold required to achieve neuroinvasion , whereas phagocytic cells such as macrophages act to destroy them [59] , [60] . Indeed increased numbers of PrPd-containing tingible body macrophages are found within the B cell follicles of TSE-affected animals [58] . Thus , our data suggest that in the specific absence of PrPC expression by FDC the initial inoculum is phagocytosed and gradually degraded by mononuclear phagocytes such as tingible body macrophages [59] , [60] . These data are congruent with data from our earlier study which likewise occasionally detected trace levels of prions from the initial inoculum within tingible body macrophages in the spleens of mice with a PrPC-deficient FDC-containing stromal compartment [13] . The density of sympathetic nerves can significantly influence the amount of prion accumulation in the spleen [33] . In the current study the distribution of TH-positive sympathetic nerves in the spleens of the FDC-specific gene targeted mouse lines was not adversely affected . Furthermore , when prions were injected directly to the brain , FDC-specific Prnp ablation had no influence on the onset of clinical disease or the neuropathology . These data provide strong evidence that the effects of Cre-mediated Prnp ablation on prion replication in the spleen were specific to FDC and not due to unregulated ablation of PrPC expression within the nervous system . In the current study PrPSc accumulation upon PrPC-ablated FDC ( Prnpflox/-→CD21-Cre Prnpflox/- mice ) was blocked even in spleens from i . c . injected clinically-scrapie affected mice . These data contrast those reported by Crozet and colleagues [61] which used Tg ( OvPrP4 ) mice that express the ovine PRNP gene under the control of the neuron-specific enolase promoter on a murine Prnp-/- background . As a consequence ovine PrPC is expressed only in neurones . In contrast to data in the current study , when Tg ( OvPrP4 ) mice were injected i . c . with a high dose of natural sheep scrapie PrPSc was detected in the germinal centres of their spleens . The reasons for this discrepancy are uncertain . However , the expression of PrPC in the neuronal compartment of Tg ( OvPrP4 ) mice is 2-4X higher than in controls . In the current study in mice in which PrPC was ablated only on FDC ( Prnpflox/-→CD21-Cre Prnpflox/- mice ) the expression of murine Prnp in Cre-deficient cells such as neurones is controlled by the endogenous Prnp promoter and expressed at similar levels to controls ( Figure 5E ) . In the presence of increased PrPC expression on neurones it is plausible that greater prion replication occurred within the peripheral nervous system , which may have been subsequently trapped on the surface of the FDC and scavenged by macrophages as the prion burden increased . Similarly , hyper-innervation of the spleen likewise leads to increase prion burden in this tissue [33] . In conclusion , our data demonstrate that PrPC-expressing FDC are the essential sites of prion replication in lymphoid tissues . Indeed , PrPC-expression on FDC alone was sufficient to sustain high levels of prion replication . In contrast , the specific ablation of PrPC expression on FDC blocked prion replication . Although FDC have the capacity to bind exosomes and immune complexes which may contain PrPSc , this finding clearly demonstrates that FDC do not simply passively acquire prions from other infected cell populations such as neurones . Previous data show treatments which impair the status or immune complex-trapping function of FDC reduce prion susceptibility after peripheral exposure [1] , [16] , [35] , [46] , [62] . The demonstration that Prnp-ablation only on FDC blocked splenic prion replication without apparent consequences for FDC status represents a novel opportunity to prevent neuroinvasion by modulation of PrPC expression on FDC .
All studies using experimental mice and regulatory licences were approved by both The Roslin Institute's and University of Edinburgh's Protocols and Ethics Committees . All animal experiments were carried out under the authority of a UK Home Office Project Licence within the terms and conditions of the strict regulations of the UK Home Office ‘Animals ( scientific procedures ) Act 1986' . Where necessary , anaesthesia appropriate for the procedure was administered , and all efforts were made to minimize harm and suffering . Mice were humanely culled using by a UK Home Office Schedule One method . The CD21-Cre [26] , ROSA26flox/flox reporter strain [28] , Prnp-/- [12] mice and tga20 mice over-expressing PrPc [63] were generated as described previously . Prnpflox/flox mice have loxP sites flanking exon 3 of the Prnp gene [31] . Prnpstop/- mice have a floxed β-geo cassette inserted into intron 2 of the Prnp gene upstream of exon 3 [31] . Mice were maintained under SPF conditions . Prior to their use in experiments , the genotype of each mouse was confirmed by PCR analysis . DNA was prepared from tails , blood and spleens using the DNeasy blood and tissue kit ( Qiagen , Crawley , UK ) according to the manufacturer's instructions . Where indicated DNA samples were analysed for presence of Cre , LacZ , Prnp+/+ , Prnp-/- , Prnpflox , recombined Prnpflox ( Prnpde-flox ) , Prnpstop and recombined Prnpstop ( Prnpstop ( R ) ) using the primers listed in Table 1 . PCR products were resolved by electrophoresis through a 1 . 0% agarose gel containing 0 . 002% GelRed ( Biotium , Cambridge Biosciences Ltd , Cambridge , UK ) . Bone-marrow from the femurs and tibias of donor mice was prepared as single-cell suspensions ( 3×107–4×107 viable cells/ml ) in HBSS ( Invitrogen , Paisley , UK ) . Recipient adult ( 6–8 weeks old ) mice were γ-irradiated ( 950 rad ) and 24 h later reconstituted with 100 µl bone-marrow by injection into the tail vein . Recipient mice were used in subsequent experiments as described 100 days after bone marrow reconstitution to allow sufficient time for removal of long-lived B lymphocyte populations and their replacement from the donor bone marrow . Tissues were first immersed in LacZ fixative [PBS ( pH 7 . 4 ) containing 2% paraformaldehyde , 0 . 2% gluteraldehyde , 0 . 02% Nonidet P40 , 0 . 01% sodium deoxycholate , 5 mM EGTA , 2 mM MgCl2] and washed in LacZ wash buffer [PBS ( pH 7 . 4 ) containing 0 . 02% Nonidet P40 , 0 . 01% sodium deoxycholate , 2 mM MgCL2] . Tissues were subsequently incubated in 15% ( wt/vol ) sucrose in PBS overnight followed by a further overnight incubation in 30% ( wt/vol ) sucrose in PBS and embedded in Tissue-Tek O . C . T . compound ( Bayer PLC , Newbury , UK ) . Serial sections ( thickness 8 µm ) were cut on cryostat and stained overnight at 37°C with LacZ staining solution ( Glycosynth , Warrington , UK ) . Staining reaction was stopped by washing in LacZ wash buffer followed by dH2O . Sections were counterstained with nuclear fast red ( Vector Laboratories , Peterborough , UK ) . For i . c . or i . p . exposure mice were injected with 20 µl of a 1% ( v/w ) scrapie brain homogenate prepared from mice terminally-affected with ME7 scrapie prions ( containing approximately 1×104 i . c . ID50 units ) . Following exposure , mice were coded and assessed blindly for signs of clinical disease and culled at a standard clinical endpoint [64] . Survival times were recorded for mice that did not develop clinical signs of disease and were culled when they showed signs of intercurrent disease . Scrapie diagnosis was confirmed blindly on coded sections by histopathological assessment of vacuolation in the brain . For the construction of lesion profiles , vacuolar changes were scored in nine grey-matter areas of brain as described [65] . Where indicated , some four mice from each group were culled at the times indicated post injection with scrapie and tissues taken for further analysis . For bioassay of scrapie agent infectivity , individual half spleens were prepared as 10% ( wt/vol ) homogenates in physiological saline . Groups of four tga20 indicator mice were injected i . c . with 20 µl of each homogenate . The scrapie titre in each sample was determined from the mean incubation period in the indicator mice , by reference to a dose/incubation period response curve for ME7 scrapie-infected spleen tissue serially titrated in tga20 mice using the relationship: y = 9 . 4533–0 . 0595x ( y , = log ID50 U/20 µl of homogenate; x , incubation period; R2 = 0 . 9562 ) . As the expression level of cellular PrPc controls the prion disease incubation period , tga20 mice overexpressing PrPc are extremely useful as indicator mice in prion infectivity bioassays as they succumb to disease with much shorter incubation times than conventional mouse strains [63] . Spleens were removed and snap-frozen at the temperature of liquid nitrogen . Serial frozen sections ( 10 µm in thickness ) were cut on a cryostat and immunostained with the following antibodies: FDCs were visualized by staining with mAb 7G6 to detect CR2/CR1 ( CD21/CD35; BD Biosciences PharMingen ) , mAb FDC-M2 to detect C4 ( AMS Biotechnology , Oxon , UK ) or mAb 8C12 to detect CR1 ( CD35; BD Biosciences PharMingen ) . Cellular PrPC was detected using PrP-specific polyclonal antibody ( pAb ) 1B3 [66] . B cells were detected using mAb B220 to detect CD45R ( Caltag , Towcester , UK ) , or anti-CD19 ( BD biosciences PharMingen ) . Marginal zone B cells were detected using mAb 1B1 to detect CD1d ( BD Biosciences PharMingen ) . Sympathetic nerves were detected using tyrosine hydroxylase ( TH ) -specific pAb ( Chemicon Europe ) . For the detection of disease-specific PrP ( PrPd ) in spleens and brains , tissues were fixed in periodate-lysine-paraformaldehyde fixative and embedded in paraffin wax . Sections ( thickness , 6 µm ) were deparaffinised , and pre-treated to enhance the detection of PrPd by hydrated autoclaving ( 15 min , 121°C , hydration ) and subsequent immersion formic acid ( 98% ) for 5 min [67] . Sections were then immunostained with 1B3 PrP-specific pAb . For the detection of EGF-like module-containing mucin-like hormone receptor-like 1 ( EMR1 ) -expressing macrophages , paraffin-embedded spleen sections were micro-waved in citric acid buffer ( pH 6 . 0 ) for 10 min . Endogenous peroxidase activity was blocked using 1% hydrogen peroxidase in methanol , and macrophages detected using rat mAb F4/80 to detect EMR1 ( clone CI:A3-1 , AbD Serotec ) . For the detection of astrocytes , brain sections were immunostained with anti-glial fibrillary acidic protein ( GFAP; DAKO , Ely , UK ) . For the detection of microglia , deparaffinised brain sections were first pre-treated with Target Retrieval Solution ( DAKO ) and subsequently immunostained with anti-ionized calcium-binding adaptor molecule 1 ( Iba-1; Wako Chemicals GmbH , Neuss , Germany ) . Immunolabelling was revealed using HRP-conjugated to the avidin-biotin complex ( Novared kit , Vector laboratories , Peterborough , UK ) . Paraffin-embedded tissue ( PET ) immunoblot analysis was used to confirm the PrPd detected by immunohistochemistry was proteinase K ( PK ) -resistant PrPSc [34] . Membranes were subsequently immunostained with 1B3 PrP-specific pAb . For light microscopy , following the addition of primary antibodies , biotin-conjugated species-specific secondary antibodies ( Stratech , Soham , UK ) were applied followed by alkaline phosphatase or HRP coupled to the avidin/biotin complex ( Vector Laboratories ) . Vector Red ( Vector Laboratories ) and diaminobenzidine ( DAB; Sigma Aldrich , Dorset , UK ) were used as substrates , respectively , and sections were counterstained with haematoxylin to distinguish cell nuclei . For fluorescent microscopy , following the addition of primary antibody , species-specific secondary antibodies coupled to Alexa Fluor 488 ( green ) , Alexa Fluor 594 ( red ) dyes or Alexa Fluor 647 ( blue ) dyes ( Invitrogen , Paisley , UK ) were used . Sections were mounted in fluorescent mounting medium ( DakoCytomation ) and examined using a Zeiss LSM5 confocal microscope ( Zeiss , Welwyn Garden City , UK ) . Digital microscopy images were analyzed using ImageJ software ( http://rsbweb . nih . gov/ij/ ) as described [68] . Intensity thresholds were first applied and then the number of pixels of each colour ( black , red , green , yellow ) were then automatically counted and presented as a proportion of the total number of pixels in each area under analysis . The preferential co-localisation of fluorochromes was determined by comparisons of the observed distribution of colours with those predicted by the null hypothesis that each element of positive staining was randomly and independently distributed . Values found to be significantly greater than the null hypothesis confirm significant co-localisation of fluorochromes . Spleens from 6 mice from each group were analyzed . From each spleen , 2 sections were studied and on each section data from 4 individual FDC networks collected . Thus , for each mouse group data from a total of 48 individual FDC were analysed . Similarly , data from 48 images from each group were analyzed to determine the preferential co-localisation of fluorochromes upon TH-positive sympathetic nerves within the spleen . A one-way ANOVA test was then used to compare the null hypothesis ( that the pixels were randomly distributed ) to the observed levels of co-localisation . To assess antigen trapping by FDC in vivo , mice were passively immunized by intravenous injection with 100 µl preformed PAP immune complexes ( Sigma ) . Spleens were removed 24 h later and the presence of FDC-associated immune complexes identified by IHC . Data are presented as mean ± SE . Unless indicated otherwise , significant differences between samples in different groups were sought by one-way ANOVA . Values of P<0 . 05 were accepted as significant .
|
Prion diseases are infectious neurological disorders and are considered to be caused by an abnormally folded infectious protein termed PrPSc . Soon after infection prions accumulate first upon follicular dendritic cells ( FDC ) in lymphoid tissues before spreading to the brain where they cause damage to nerve cells . Cells must express the normal cellular prion protein PrPC to become infected with prions . However , whether FDC are infected with prions or simply acquire them from other infected cells is unknown . To establish the role of FDC in prion disease PrPC expression was specifically “switched on” or “off” only on FDC . We show that PrPC-expressing FDC alone are sufficient to sustain prion replication in the spleen . Furthermore , prion replication is blocked in the spleen when PrPC-expression is switched off only on FDC . These data definitively demonstrate that FDC are the essential sites of prion replication in lymphoid tissues .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"veterinary",
"prion",
"diseases",
"infectious",
"diseases",
"infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"veterinary",
"diseases",
"immune",
"cells",
"immunology",
"biology",
"veterinary",
"science",
"prion",
"diseases"
] |
2011
|
Follicular Dendritic Cell-Specific Prion Protein (PrPc) Expression Alone Is Sufficient to Sustain Prion Infection in the Spleen
|
A number of studies on visceral leishmaniasis ( VL ) vector control have been conducted during the past decade , sometimes came to very different conclusion . The present study on a large sample investigated different options which are partially unexplored including: ( 1 ) indoor residual spraying ( IRS ) with alpha cypermethrin 5WP; ( 2 ) long lasting insecticide impregnated bed-net ( LLIN ) ; ( 3 ) impregnation of local bed-nets with slow release insecticide K-O TAB 1-2-3 ( KOTAB ) ; ( 4 ) insecticide spraying in potential breeding sites outside of house using chlorpyrifos 20EC ( OUT ) and different combinations of the above . The study was a cluster randomized controlled trial where 3089 houses from 11 villages were divided into 10 sections , each section with 6 clusters and each cluster having approximately 50 houses . Based on vector density ( males plus females ) during baseline survey , the 60 clusters were categorized into 3 groups: ( 1 ) high , ( 2 ) medium and ( 3 ) low . Each group had 20 clusters . From these three groups , 6 clusters ( about 300 households ) were randomly selected for each type of intervention and control arms . Vector density was measured before and 2 , 4 , 5 , 7 , 11 , 14 , 15 , 18 and 22 months after intervention using CDC light traps . The impact of interventions was measured by using the difference-in-differences regression model . A total of 17 , 434 sand flies were collected at baseline and during the surveys conducted over 9 months following the baseline measurements . At baseline , the average P . argentipes density per household was 10 . 6 ( SD = 11 . 5 ) in the control arm and 7 . 3 ( SD = 8 . 46 ) to 11 . 5 ( SD = 20 . 2 ) in intervention arms . The intervention results presented as the range of percent reductions of sand flies ( males plus females ) and rate ratios in 9 measurements over 22 months . Among single type interventions , the effect of IRS with 2 rounds of spraying ( applied by the research team ) ranged from 13% to 75% reduction of P . argentipes density compared to the control arm ( rate-ratio [RR] ranged from 0 . 25 to 0 . 87 ) . LLINs caused a vector reduction of 9% to 78% ( RR , 0 . 22 to 0 . 91 ) . KOTAB reduced vectors by 4% to 73% ( RR , 0 . 27 to 0 . 96 ) . The combination of LLIN and OUT led to a vector reduction of 26% to 86% ( RR , 0 . 14 to 0 . 74 ) . The reduction for the combination of IRS and OUT was 8% to 88% ( RR , 0 . 12 to 0 . 92 ) . IRS and LLIN combined resulted in a vector reduction of 13% to 85% ( RR , 0 . 15 to 0 . 77 ) . The IRS and KOTAB combination reduced vector densities by 16% to 86% ( RR , 0 . 14 to 0 . 84 ) . Some intermediate measurements for KOTAB alone and for IRS plus LLIN; and IRS plus KOTAB were not statistically significant . The bioassays on sprayed surfaces or netting materials showed favourable results ( >80% mortality ) for 22 months ( IRS tested for 12 months ) . In the KOTAB , a gradual decline was observed after 6 months . LLIN and OUT was the best combination to reduce VL vector densities for 22 months or longer . Operationally , this is much easier to apply than IRS . A cost analysis of the preferred tools will follow . The relationship between vector density ( males plus females ) and leishmaniasis incidence should be investigated , and this will require estimates of the Entomological Inoculation Rate .
Visceral leishmaniasis ( VL ) [known as kala-azar in the Indian sub-continent] is a parasitic disease present in South-East Asia since before the early 1800’s [1] . VL appears to have spread along the Ganges and the Brahmaputra rivers , the major transport routs of Bengal and Bangladesh . In this area , VL was first described in 1824 in the Jessore district where about 75 , 000 people died [2] . An intensive control programme aimed at the eradication of malaria was mounted in the late 1950s and early 1960s throughout the South Asian sub-continent with the main effort based on indoor residual spraying ( IRS ) of DDT ( Dichlorodiphenyltrichloroethane ) . During the malaria eradication programme the incidence of VL dropped dramatically as a collateral benefit with DDT spraying [3] . However , within a few years after the end of the Malaria eradication effort , VL returned to Bihar and Bengal on both sides of the borders of India and Bangladesh [4] . In Bangladesh , five districts , Sirajgang , Pabna , Mymensingh , Rajshahi and Tangail were more affected following the end of the malaria eradication programme to 1980s [5] . These districts continue to have the highest number of cases along with other districts reporting few cases . The Malaria and Vector Borne Disease Control Unit , Directorate General of Health Service ( DGHS ) , Government of Bangladesh has reported 109 , 226 VL cases including 329 VL related deaths from 1994 to 2013 [6] . Mymensingh district alone contributed about 50% of total reported cases and the highest number of cases was reported from the Fulbaria upazila ( sub-district ) with several other upazilas in Mymensingh reporting cases [6] . To combat VL in the Indian Sub-continent , a common platform was developed through signing a memorandum of understanding ( MoU ) by the health ministers from the three affected countries ( Bangladesh , India and Nepal ) in 2005 [7] . In the MoU , a target was set to reduce the VL incidence to less than 1 case per 10 , 000 population at the sub-district ( upazila in Bangladesh ) level by 2015 [7] . This MoU has been extended up to 2017 with inclusion of Bhutan and Thailand in the group [8] . Integrated vector management is one of the most important pillars in the elimination strategy , however virtually no vector control activities were under taken in Bangladesh for VL vector control until 2010 [6 , 9] . Since 2011 , the National Kala-azar Elimination Programme ( NKEP ) in Bangladesh conducted IRS for vector control using deltamethrin 5 WP in the affected communities . In addition to IRS , two commercially manufactured long lasting impregnated bed-nets ( LLIN ) were given to each patient who was treated in the government hospitals for the last three years ( 2011 to 2013 ) . The IRS ( using deltamethrin 5WP ) and LLIN were associated to decrease the level of the Phlebotomus argentipes sand fly by 70–80% in Bangladesh [10] . In India and Nepal , deficiencies in the quality of IRS was observed when carried out by the national programme [11] . Mosquito nets impregnated with a slow release insecticide ( K-O TAB 1-2-3; deltamethrin with a binder ) resulted in a 65% reduction in sand fly levels in Bangladesh [12] . None of the studies however determined the effect of both LLIN and insecticide treated local nets in a single study . VL vectors usually breed in the shady places with loose soil where enough moisture is available around the houses [13] . There is lack of evidence in Bangladesh for controlling immature stages of P . argentipes sand fly by applying insecticide spraying on their breeding habitat around the houses . Hypothetically , it is believed that the combined application of two vector control methods will increase the effect on VL vectors but currently there is no evidence for this in South-East Asia . The present study on single and combined vector control interventions was conducted to assist the NKEP by identifying suitable VL vector control method ( s ) to achieve the elimination target within the set time frame and maintain a low vector density during the maintenance phase of the VL elimination programme .
The sample size estimation was based on the vector densities ( female and male P . argentipes sand fly counts per household ) , variations and distributions documented in previous entomological studies and sand fly reduction rates in similar intervention studies performed in Venezuela; Bangladesh , India and Nepal [19 , 20] . Based on our previous experience , male:female ratio was always about 50:50 throughout the year [10 , 20] , and so we assume that joint counts will not affect the outcome of the analysis , but will produce more accurate estimate through regression model . We assumed that the distribution of sand fly counts would follow a negative binomial distribution with a dispersion coefficient of k = 0 . 05 and an intra-cluster coefficient of 0 . 03 , a reduction from 20 to 5 vectors per trap/ night , and an average of 50 households per cluster . The minimum sample size was found to be 6 clusters per intervention arm , with a total of 60 clusters in the study to achieve 80% power and a significance level of 5% . VL surveillance data for three years ( 2009–2011 ) was collected from Upazila Health Complex ( UHC ) , Fulbaria . Based on the passive surveillance reports , endemic villages were identified ( containing 300 to 600 households [HH] ) and were selected where the national programme was not conducting routine vector control activities . The following villages were selected ( Fig 1 ) : Mahespur in Bhabanipur union ( 311 HHs ) ; Mandolbari and Chalkgarbajail in Balian union ( 319 HHs ) ; Patira in Kaladah union ( 323 HHs ) ; Hurbari in Kaladah union ( 297 + 312 = 609 HHs ) ; Dulma in Enayetpur union ( 309 HHs ) ; Kathgarh in Naogaon union ( 307 HHs ) ; Bisania and Natuapara in Kushmail union ( 300 HHs ) ; Anuhadi in Rangamatia union ( 322 HHs ) ; and Haripur in Rangamatia union ( 302 HHs ) , in total 3079 HHs ) . We had no indication or evidence that the study villages were atypical . The study was conducted from September 2012 to October 2014 . Based on their high endemicity levels , 11 villages were selected from seven unions . Eleven villages were divided into 10 sections with a minimum of 300 HHs each ( Fig 2 ) . Each section with at least 300 HHs was divided into six clusters , each with about 50 HHs . In each study cluster , all HHs were numbered using enamel paint on the front door . There was a minimum of 50 meters distance between two clusters to avoid cross-contamination of the interventions . The total number of 60 clusters ( 10 sections x 6 clusters each ) was assigned for implementation of interventions ( Fig 2 ) . Five HHs from each cluster ( 5 HHs x 60 clusters = 300 HHs ) were selected by simple random sampling for measuring sand fly densities at 3 weeks before intervention ( baseline survey ) and at 2 , 4 , 5 , 7 , 11 , 14 , 15 , 18 and 22 months after intervention ( follow-up surveys ) . In order to achieve homogenous and comparable groups of clusters , the following approach was undertaken . Based on vector ( P . argentipes ) density during the baseline collection , the 60 study clusters were stratified into 3 groups: ( i ) high [range of P . argentipes– 59 to 143] , ( ii ) medium [range of P . argentipes– 31 to 57] and ( iii ) low [range of P . argentipes– 1 to 30] vector density . Each group ( vector density stratum ) had 20 clusters . From these 60 clusters ( 20 clusters in each group ) , the different intervention and control arms were randomly selected and each arm had 6 clusters [about 300 HHs each] ( Fig 2 ) . After the formation of the clusters , all HH heads were interviewed by trained field staff ( Research Assistants [RA] ) using a structured questionnaire to record the HH information , socioeconomic characteristics and some epidemiological information . A standard data entry interface was designed using Microsoft Office Access for entering the study data . Data were checked and cleaned before analysis . Descriptive analysis was performed to determine the nature of the data . The main analysis was based on P . argentipes sand fly counts per HH collected by CDC light traps over two nights . No zero count was removed . All CDC traps worked perfectly . The average sand fly density among different interventions and control groups at baseline as well as follow-up time points was determined . Mean P . argentipes sand fly count between control and intervention areas were compared using a non-parametric approach ( Mann Whitney U test ) . It was found that the negative binomial distribution fitted the data and all analyses were performed under that assumption . A generalized estimating equation ( GEE ) modelling technique was used to adjust for data correlations due to the longitudinal/ repeated measurements in cluster sampling . An interaction term for the intervention arm at follow-up was included in this model in order to estimate the effect of the intervention . Technically , the regression model had the following structure: Count=Intercept+a*Treatment+b*Time+c*Interaction+error where treatment is one if it is the intervention and zero if it is the control; where time is one if follow up and zero if baseline; and where interaction is one if the intervention group at follow up . Intervention effect was measured using incidence rate ratio ( IRR ) of P . argentines sand fly count generated from the exponent of c-coefficient in the model . In the tables , IRR represented as the rate ratio ( RR ) and its p-value are given . Significances stated at 5% level and 95% confidence intervals are given . The main outcome variable was “P . argentipes sand flies per household” at before and 2 , 4 , 5 , 7 , 11 , 14 , 15 , 18 and 22 months after the intervention . The following variables were controlled for in the full model: cattle shed , family members sleep in the bed room , number of bed nets in the house , type of house wall , presence of crack in wall and socio economic status . Economic status of the household was measured through the HH asset index . Household asset index was generated by the principal component method in factor analysis using the following variables: electricity , radio , television , mattress , bed net , motor cycle , bicycle , van , power tiller , shallow machine , chair/table , mobile phone , clock , sewing machine , and fishery . We categorized the index as low ( score less than 33th percentile ) , medium ( score between 33th to 66th percentile ) and high ( score greater than 66th percentile ) . We did not perform any sensitivity analysis as the study objective was to compare the efficacy of different vector control interventions against control arm . All analyses were performed by using STATA 10 . 1 . Investigators and external experts conducted all training activities to ensure the quality of the training . All study activities were monitored by the investigators to maintain the quality . The study was approved by Bangladesh Medical Research Council ( BMRC ) . Written informed voluntary consent was obtained from the HH heads/responsible family members before conducting any study related activities . All control ( no intervention ) arm HHs were donated one LLIN per family after completing the study .
The study was conducted in 3079 HHs with a population of 13 , 406 inhabitants in 10 sections of 60 clusters ( Fig 2 ) . Nine types of interventions were tested and one control arm . Within the study population , 49 . 4% were female and 39 . 6% were below 17 years of age ( Table 1 ) . About 5% and 0 . 6% of the total population had a past history of VL and PKDL respectively . The proportion of VL and PKDL varied from 3 . 6% to 5 . 8% and 0 . 1% to 1 . 1% respectively among the different study arms ( Table 1 ) . About 60% ( 1846/3079 ) of HHs had a cattle shed . The percentage of the population with cattle sheds varied from 53 . 5% ( 161/301 ) to 69 . 5% ( 214/308 ) among the different study arms . Almost all HHs had a non-impregnated bed net ( 99 . 1% ) with an average of 2 . 33 per HH ( SD = 1 . 35 ) . About 30% of houses had precarious walls and about 15 . 0% had cracks in their walls . The study HHs were almost equally distributed among low ( 33 . 4% ) medium ( 30 . 7% ) and high ( 35 . 6% ) asset index groups ( Table 1 ) . A total 17 , 434 sand flies were collected during the entire study period including baseline and the 9 follow-up surveys ( Table 2 ) . Of all sand flies , 53 . 75% were P . argentipes and the remainder other species . Among P . argentipes , 4 , 443 ( 47 . 42% ) were female including 16 . 01% gravid . At baseline ( before implementation of intervention ) , 3 , 616 sand flies were captured of which 78 . 32% were P . argentipes . There were 83 ( 18 . 12% ) , 412 ( 36 . 52% ) , 1315 ( 42 . 57% ) , 1443 ( 58 . 52% ) , 383 ( 55 . 59% ) , 67 ( 35 . 26% ) , 331 ( 55 . 72% ) , 1419 ( 37 . 92% ) and 1085 ( 74 . 21% ) P . argentipes sand flies collected respectively in the first to ninth follow up measurements . There is no significant difference between male and female ratio of collected P . argentipes sand flies throughout the study period ( S1 Table ) . At baseline , the average P . argentipes density per household was 10 . 57 ( SD = 11 . 51 ) in the control arm and 7 . 3 ( SD = 8 . 46 ) to 11 . 53 ( SD = 20 . 17 ) in the different intervention arms ( Table 3 ) . The difference of P . argentipes sand fly densities among intervention arms and control arm were not statistically significant at baseline except for the KOTAB intervention arm ( p = 0 . 032 ) . However , P . argentipes sand fly densities in most of the intervention arms were significantly lower than the control arm at different follow-up measurements except OUT ( Table 3 ) . Fig 3 shows that the mean P . argentipes sand fly density was always below the values in the control arm throughout all the measurements . The efficacy of the interventions was measured through the reduction of P . argentipes sand fly densities in intervention HHs compared to the control HHs . The bioavailability of insecticides on treated surfaces ( indicating how long the insecticide was capable of killing insect vectors ) was determined using bioassay tests . The Abbot corrected sand fly mortality at 24 hours of exposure on treated surfaces was as follows: in both cycles of IRS , the mortality was above 80% ( which is the threshold level ) even after 5 months following spraying ( Fig 5 ) . The mortality for LLIN was 82 . 59% at 20 months of use . The mortality on K-O TAB 1-2-3 impregnated nets dropped from 88 . 37% at 3 months to 69 . 12% at 20 months after use .
Our cluster randomized controlled trials on VL vector interventions are to our knowledge the largest ever conducted in the South-East Asia Region . In the present study a large number of sand flies was captured of which 45 . 9% were other than P . argentipes species which is three times higher than in a previous study conducted in the same sub-district [10] . We have tested four individual types of interventions and five combinations against VL vectors . In the present study , we used alpha cypermethrin 5WP for IRS as it is less expensive than deltamethrin and as efficacious as other pyrethroids [10 , 20] . IRS is however challenging in terms of operational complexity and cost and it is difficult to maintain a uniform quality of spraying . It was observed in India and Nepal that when IRS was applied by the research team under well controlled conditions , it was found to be very effective against VL vectors but when it was delivered by the national programme the efficacy dropped significantly [11] . At the current stage of the VL elimination initiative in the Indian subcontinent , Nepal has reached the target of less than one case per 10 , 000 population , Bangladesh has only a few sub-districts ( upazilas ) above this threshold and India has reached the goal in many areas but is still facing elevated VL endemicity in a number of districts [30] . In sub-districts where the final push towards elimination is still required , IRS alone or in combination with other measures are still needed . In sub-districts where VL/PKDL cases appear sporadically and case numbers are below the thresholds , new ways of active case detection ( to reduce the transmission ) and vector control ( to prevent transmission ) are required . Regarding vector management in the post-elimination phase , recent studies have shown the potential of different vector control tools , including insecticide treated durable wall lining ( DWL ) , commercially impregnated long lasting insecticidal nets ( LLIN ) , slow release insecticides ( K-O TAB 1-2-3 , ) treatment of existing bed nets ( ITNs ) , as well as insecticidal paint ( Inesfly company , Valencia , Spain ) . Prospects and limitations of these products include the following: Our study contributes important information to what is known already . ( i ) The combination of different approaches leads to better results than single approaches in reducing the vector population . ( ii ) The combination of a chemical intervention in breeding and larval-development sites in and around rural houses together with measures against adult vectors using LLINs was particularly successful in significantly reducing vector populations for at least 22 months . Furthermore , this measure can be used in remote areas where sporadic cases appear as it lends itself to community actions . The use of ITNs would be even more feasible as it is independent of LLIN donations . Local bed nets to prevent mosquito bites are common in rural Bangladeshi communities and over 90% of HHs have nets [31 , 35] . Slow release insecticide impregnated bed nets might be a good alternative to prevent sand fly bites but the effect on the vector population was shorter and less marked compared to LLINs . It will be a remarkable innovation if applications of insecticides in breeding places of sand flies around houses in endemic communities are able to reduce the vector density . In the current study , we tested chlorpyrifos 20EC to control immature stages of sand flies because this insecticide has no/very limited side effects on the environment or on human health [36] . The result following the first round of chlorpyrifos spraying was not promising but after the second round of spraying , it was effective in reducing the vector densities . This effect was considerably enhanced when combined with the treatment of bed nets . In conclusion: The combination of LLIN and OUT ( outdoor spraying of vector breeding sites ) was the most efficacious measure among the different tools tested . This combination measure could play an important role during the maintenance phase of the VL elimination programme to maintain a low vector density particularly in remote areas where the community can take care of the measure . The relationship between vector density ( males plus females ) and leishmaniasis incidence should be investigated , and this will require estimates of the Entomological Inoculation Rate .
|
Integrated vector management ( IVM ) is of the important elements in the Regional Visceral Leishmaniasis ( VL ) elimination strategy . VL supposed to be eliminated from the Region ( Bangladesh , India and Nepal ) by 2015 which is extended up to 2017 . There are several factors are responsible for not achieving the elimination goal and IVM is one of them . In Bangladesh , currently VL vector control activity is confined to indoor residual spraying with deltamethrin . Therefore , it is crucial to identify alternative VL vector control method ( s ) to enhance VL elimination programme . In the present study we have investigated the efficacy of ( 1 ) indoor residual spraying ( IRS ) with alpha cypermethrin 5WP; ( 2 ) long lasting insecticide impregnated bed-net ( LLIN ) ; ( 3 ) impregnation of local bed-nets with slow release insecticide K-O TAB 1-2-3 ( KOTAB ) ; ( 4 ) insecticide spraying in potential breeding sites using chlorpyrifos 20EC ( OUT ) and different combinations of the above . Study findings showed that the combination of LLIN and OUT was the most effective method against VL vector and among single interventions , IRS with two rounds of spraying found efficacious . In addition to IRS , VL elimination programme should consider LLIN plus OUT for vector control in relation to IVM .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"kala-azar",
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"geographical",
"locations",
"india",
"sand",
"flies",
"parasitic",
"diseases",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"insect",
"vectors",
"zoology",
"bangladesh",
"infectious",
"diseases",
"agrochemicals",
"zoonoses",
"protozoan",
"infections",
"disease",
"vectors",
"agriculture",
"insecticides",
"people",
"and",
"places",
"asia",
"entomology",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"species",
"interactions"
] |
2017
|
Control of Phlebotomus argentipes (Diptera: Psychodidae) sand fly in Bangladesh: A cluster randomized controlled trial
|
The mouse polyoma virus induces a broad array of solid tumors in mice of many inbred strains . In most strains tumors grow rapidly but fail to metastasize . An exception has been found in the Czech-II/Ei mouse in which bone tumors metastasize regularly to the lung . These tumors resemble human osteosarcoma in their propensity for pulmonary metastasis . Cell lines established from these metastatic tumors have been compared with ones from non-metastatic osteosarcomas arising in C3H/BiDa mice . Osteopontin , a chemokine implicated in migration and metastasis , is known to be transcriptionally induced by the viral middle T antigen . Czech-II/Ei and C3H/BiDa tumor cells expressed middle T and secreted osteopontin at comparable levels as the major chemoattractant . The tumor cell lines migrated equally well in response to recombinant osteopontin as the sole attractant . An important difference emerged in assays for invasion in which tumor cells from Czech-II/Ei mice were able to invade across an extracellular matrix barrier while those from C3H/BiDa mice were unable to invade . Invasive behavior was linked to elevated levels of the metalloproteinase MMP-2 and of the transcription factor NFAT . Inhibition of either MMP-2 or NFAT inhibited invasion by Czech-II/Ei osteosarcoma cells . The metastatic phenotype is dominant in F1 mice . Osteosarcoma cell lines from F1 mice expressed intermediate levels of MMP-2 and NFAT and were invasive . Osteosarcomas in Czech-II/Ei mice retain functional p53 . This virus-host model of metastasis differs from engineered models targeting p53 or pRb and provides a system for investigating the genetic and molecular basis of bone tumor metastasis in the absence of p53 loss .
Invasion and metastasis are major factors underlying cancer morbidity and mortality [1] . Mouse models have been developed for many types of human cancer though not all present the same biological behavior as human tumors with respect to invasion and metastasis . Some models are of spontaneous origin while most are based on genetically engineered animals . Transgenic mice , mice with germline or conditional knockouts , targeted mutations [2] , or ones emerging from screens following germline mutagenesis [3] have all been used to establish experimental models of specific types of cancer . Mice genetically engineered to develop cancer are typically derived from relatively few inbred strains thus reflecting a limited range of effects imposed by the host genetic background . Crosses of genetically engineered mice to other strains have been useful for identifying tumor modifiers [4] , [5] . The mouse polyoma virus ( Py ) is a small DNA virus that rapidly induces a variety of solid tumors in its natural host under laboratory conditions [6] . The virus establishes a disseminated infection using specific gangliosides as broadly expressed cell receptors [7] . Features of the major viral capsid protein VP1 are important in receptor discrimination and polymorphisms in VP1 are major determinants of pathogenicity [8] , [9] . The replication and transforming functions of the virus reside in the T ( tumor ) antigens . These have been extensively characterized using cell culture systems . Expression of the T antigens leads to alterations in the regulation of growth and apoptotic pathways of the cell . These alterations involve both protooncogene activation and tumor suppressor gene inactivation . Well characterized wild type and mutant virus strains have been used to assess the roles of specific viral determinants in tumorigenesis using a standard inbred mouse strain as the common host [10]–[13] . The virus can be introduced into any of a large number of inbred strains of mice to investigate effects of the host genetic background on various aspects of tumor development . Inbred mice vary greatly with respect to overall tumor frequency , spectrum of tumor types , and individual tumor behavior [6] . Py-induced tumors typically remain at the site of origin enclosed within a basement membrane . Susceptible hosts develop multiple tumors and may carry a cumulative tumor load in excess of 25% of total body weight while showing no evidence of metastasis [14] . The middle T antigen , the major viral transforming protein , activates multiple signal transduction pathways essential for tumor induction [15] . Transgenic mice expressing middle T under control of the mouse mammary tumor virus regulatory region develop mammary tumors which metastasize to the lung [16] . In this transgenic mouse system , secretion of osteopontin ( OPN ) was found to be necessary but not sufficient for metastasis [17] . In a survey of over thirty inbred mouse strains inoculated with Py [6] , the Czech II/Ei mouse ( CZ ) emerged with unique and interesting properties . In addition to a typical array of polyoma tumors [18] , CZ mice occasionally developed small nodules on the lung , normally not a site of tumor induction by the virus . These nodules proved to be metastatic lesions derived from osteosarcomas . The metastatic behavior of bone tumors in the Py-infected CZ mouse matches important clinical and pathological features of human osteosarcoma with respect to lung metastases [19] . C3H/BiDa mice ( C3 ) , a standard susceptible strain [20] , develop osteosarcomas that fail to metastasize . Here we begin to explore the CZ mouse as a model of metastatic osteosarcoma using the C3 mouse as a control . A molecular pathway has been identified that correlates with properties of invasion in vitro and metastasis in vivo .
Osteosarcomas have previously been noted in Py-infected mice of several inbred strains but with no evidence of metastasis . In CZ mice , however , these tumors have been found to metastasize regularly to the lung . Osteosarcomas develop most frequently on the femur , skull , ribs or other long bones , and occasionally on the spine and sternum . A typical case in a CZ mouse is shown in Figure 1 . The primary tumor arose on the head of the femur ( panel A ) . This tumor showed evidence of invasion of adjacent muscle and destruction of bone ( panel B ) . A metastatic lesion was evident as a smooth shiny nodule on the surface of the lung ( panel C ) . This nodule showed abundant tumor cells mixed with deposits of osteoid ( panel D ) . Lung metastases were seen in a high percentage of CZ mice by routine step-sectioning and histological examination . These lesions consistently showed production of osteoid ( panel E ) . Metastases were noted rarely in the liver . To confirm that the liver lesions in Py-infected CZ mice also derived from osteosarcomas , cells from a primary bone tumor were inoculated subcutaneously into an uninfected CZ mouse . This animal showed growth of tumor cells in the liver with deposits of osteoid ( panel F ) . The presence of osteoid in metastastic lesions in CZ mice confirms their origins from osteosarcomas and not from other tumor types arising in the same animals . Further support derives from the fact that no lung metastases were noted in CZ mice that failed to develop primary osteosarcoma while developing other kinds of tumors . Osteosarcomas in C3 mice resembled those in CZ mice grossly and histologically . However , these tumors showed no evidence of lung metastasis using the same search criteria as for CZ mice . A direct comparison of the frequencies of primary and metastatic osteosarcoma in CZ and C3 mice highlights important differences between these strains ( Table 1 ) . A high proportion of CZ mice developed osteosarcomas and 90% of these were found to have metastasized to the lung . The actual frequency of lung metastasis in CZ mice may approach 100% given the likelihood that not all lesions will be found in the step sectioning process . Osteosarcomas in C3 mice were found at a lower frequency due in part to higher frequencies of other tumor types that develop early and require the animals to be sacrificed . Nevertheless , of 18 cases of osteosarcoma in C3 mice , none were found to have metastasized to the lung . The sharp difference in frequency of metastasis between hosts is not due to differences in the time allowed for tumor development . The average ‘time to necropsy’ was 190 days for CZ and 188 for C3 , with ranges of 104–514 days for CZ and 70 – 655 for C3 . [CZ x C3] F1 mice were infected and followed for the development of osteosarcoma and then screened for lung metastases . These mice were found to resemble their CZ parent with respect to the high frequency and metastatic behavior of osteosarcomas ( Table 1 ) . The dominance of the metastatic phenotype suggests that CZ mice express a heritable factor ( s ) which promotes bone tumor metastasis and which may be lacking or insufficient in C3 mice . To investigate the molecular basis of the metastatic phenotype , cell lines were established from primary osteosarcomas , two from CZ , three from C3 and two from F1 mice . Two additional lines were derived from lung metastatic lesions , one from a CZ and one from an F1 animal . It should be noted that no osteosarcomas were found in uninfected mice of either strain and expression of the Py T antigens was confirmed in each of the cell lines from infected mice ( data not shown ) . The abilities of the tumor cells to migrate in response to a source of chemoattractant were tested in a transwell assay using a Boyden chamber with polycarbonate filter ( 8 µm pore size ) . Conditioned medium from a C3 osteosarcoma line was used as the attractant and serum-free medium as a control . Migration was stimulated 3–4 fold over a one hour period in response to the conditioned medium . No significant differences were noted among the lines ( Figure 2A ) . To identify the chemoattractant ( s ) , the conditioned medium was first treated with antibodies to OPN or to vascular endothelial growth factor ( VEGF ) . These factors , well recognized for their chemoattractant as well as mitogenic activities , play roles in normal bone development as well as tumor cell invasion and metastasis . OPN is of particular interest because expression of middle T has been shown to lead to transcriptional activation of OPN via two distinct pathways [21] . Furthermore , in an in vitro ‘wound healing’ assay , blocking induction of OPN with anti-sense RNA inhibited cell migration without interfering with transformed cell growth [21] . Pretreatment of conditioned medium with anti-OPN resulted in significant reduction in the migration of CZ-I cells while anti-VEGF had no effect ( Figure 2B , left panel ) . Results were the same using the other lines as responding cells ( data not shown ) . To determine whether the pathways from middle T to OPN operated equally in the two host backgrounds , levels of OPN were compared in conditioned media prepared from CZ and C3 bone tumor lines . Levels were found to be high and comparable between the two groups: 1975±262 pg/ml for the two CZ lines and 1592±38 pg/ml for the three C3 lines . Similarly , levels of VEGF , though much lower , were comparable: 153±12 pg/ml for CZ and 225±13 pg/ml for C3 . Additional experiments were carried out to test the specificity of migration in relation to these factors and also with respect to tumor type . Conditioned medium from a Py-induced hemangioma in a C3 mouse contained a higher level of VEGF ( 792±38 pg/ml ) and a similar level of OPN ( 1622±20 pg/ml ) compared to osteosarcoma conditioned media . Migration of CZ cells was again inhibited only by anti-OPN and not by anti-VEGF ( Figure 2B , middle panel ) . In contrast , migration of the hemangioma cells in response to its own conditioned medium was inhibited by anti-VEGF as well as anti-OPN ( Figure 2B , right panel ) . These results confirm OPN as the major attractant for osteosarcoma cells . They also demonstrate specificity in migratory behavior based on tumor type as well as attractant . The osteosarcoma lines were compared for their abilities to migrate in response to recombinant OPN added to serum-free medium as the sole attractant . All nine lines responded well with no significant differences among the lines ( Figure 2C ) . We conclude that the difference in metastatic behavior between CZ and C3 osteosarcomas is not due to a difference either in production of OPN or in ability to migrate in response to this factor . This conclusion is consistent with studies of mammary tumor metastasis in Py middle T transgenic mice which indicated that factor ( s ) in addition to OPN are required for metastasis [17] . An important difference between CZ and C3 osteosarcoma cell lines was seen using a ‘Matrigel’ invasion assay in Boyden chambers with coated membranes [22] . CZ osteosarcoma cells readily invaded across the extracellular matrix barrier in response to C3 bone tumor cell conditioned medium ( Figure 3A ) . Migration of these cells through the barrier was stimulated 6–8 fold over a 4 hour period . Tumor cells from F1 mice behaved like those from CZ . In contrast , C3 tumor lines showed almost no ability to penetrate the coated membranes under the same conditions . The abilities of osteosarcoma cells to invade in this assay match their metastatic behavior in vivo . Tumor lines established directly from metastatic lesions in lungs of CZ and F1 mice behaved similarly to those from primary tumors with respect to their properties of migration ( Figure 2A and C ) and invasion ( Figure 3A ) . The abilities of CZ and F1 tumor cells to penetrate ‘Matrigel’ suggests the involvement of one or more matrix metalloproteinases ( MMPs ) . A zymograph assay was used to identify MMPs and estimate their levels of secretion . Conditioned medium from each of the tumor cell lines was concentrated as sources of enzyme ( s ) . Gelatin was used as a substrate in the ‘in gel’ assay for detection of MMP-2 or MMP-9 , two metalloproteinases linked to invasion and metastasis by bone tumor cells [23] . Clearing of the gel was seen at a single position of ∼70kD corresponding in molecular weight to MMP-2 ( Figure 3B , top ) . Clearing was strongest for the three CZ tumors and somewhat weaker for the F1 tumors . Conditioned media from the C3 cell lines produced little or no clearing of the gel . Tumor cell extracts were tested to confirm the presence of MMP-2 and to compare levels among the lines . Extracts were separated by SDS gel electrophoresis and immunoblotted with anti-MMP-2 antibody . MMP-2 was detected in each of the cell lines but levels were higher in CZ and F1 compared to C3 tumor cells ( Figure 3B , bottom ) . Levels of endogenous MMP-2 were estimated by densitometry . Values were normalized to loading controls and averaged for the cell lines based on host of origin . The values ( arbitrary units ) were 1 . 14±0 . 22 for the three C3 lines , 3 . 79±0 . 27 for the F1 lines and 5 . 18±1 . 37 for the CZ lines . The tumor cells from CZ thus expresses 4 to 5 fold higher levels of the protease than those from C3 . Comparison of the zymograph with the immunoblot ( Figure 3B ) suggests that secreted levels of MMP-2 may be even more disproportionate ( CZ > C3 ) . This would indicate that differences in the secretory pathway or extracellular activation of the enzyme may also contribute to the ability to invade . Other MMPs not directly tested for here may also be involved . The finding of higher levels of MMP-2 secreted by CZ and F1 compared to C3 tumor cells raises the possibility of involvement of the transcription factor NFAT [24] . NFAT is known to regulate the expression of MMPs and to be involved in tumor cell invasion [25] , [26] . To test this possibility , tumor cell extracts were separated and blotted with antibody to NFATc1 . This antibody detects three alternatively spliced species among which there are no well established functional differences [27] , [28] . The levels of all three were elevated in tumor cells from CZ and F1 compared to C3 ( Figure 4A ) . Relative levels of NFAT based on scanning across all isoforms were as follows ( arbitrary units ) : for the three C3 lines 384 . 3±68 . 4 , for F1 676 . 7±58 . 5 and for CZ 1038 . 7±157 . 3 . To measure levels of activated NFAT , tumor cells were transfected with a luciferase reporter responsive to NFAT [29] . This reporter carries elements from the IL-2 promoter and is known to be responsive to NFAT [30] . CZ and F1 cells showed 3–4 fold activation of the reporter compared to C3 cells ( Figure 4B ) . These results demonstrate higher levels of expression and activation NFATc1 in the metastatic compared to the non-metastatic osteosarcoma cell lines . To test for the functional importance of MMP-2 and NFAT , ‘Matrigel’ invasion assays were carried out with the CZ II line in the presence of TIMP-2 as a specific inhibitor of MMP-2 [31] , [32] , or following transfection with siRNAs to suppress NFAT expression ( Figure 4C ) . Addition of TIMP-2 resulted in a 3 to 4 fold inhibition of invasion . Two siRNAs were used to directly inhibit expression of NFAT . Both were effective in reducing NFAT levels with concomitant effects on MMP-2 . Importantly , both were effective in inhibiting invasion . siRNA2 was particularly effective , reducing MMP-2 levels 4 to 5 fold and blocking invasion 6 to 7 fold . Similar results were obtained with all three CZ and all three F1 lines . Thus , treatments targeting either the protease or its upstream regulator result in inhibition of invasion . These results suggest that operation of an NFAT → MMP-2 pathway at elevated levels contributes to the metastatic behavior of osteosarcomas in CZ and F1 mice . Several mouse models of osteosarcoma have been developed based on loss or alteration of p53 or pRb . These reflect the occurrence of osteosarcoma in families with germline mutations in these tumor suppressor genes and with their frequent somatic alterations in sporadic cases of the disease . Mice with germline loss of p53 or ones expressing certain gain-of-function mutants develop osteosarcoma along with a variety of other tumors [33]–[35] . Loss of pRb by itself has not been noted to give rise to osteosarcoma , but when coupled to p53-deficiency or mutant p53 , pRb deficiency potentiates the development of the disease [36] , [37] . Variable rates of metastasis are found in these models . Interestingly , the frequency of p53 loss in osteosarcoma patients with localized disease was found to be the same as in patients with metastatic disease [38] , suggesting that events beyond loss of p53 are important in metastasis . The events that give rise to metastasis following disruption of the transcriptional regulatory functions of these tumor suppressors are not known . The predisposition to bone tumor metastasis in Py-infected CZ mice is dominantly inherited and presumed to be independent of p53 or pRb loss . Neither CZ nor C3 mice develop spontaneous tumors at a rate that would suggest the absence or altered function of p53 or pRb . Polyoma is unusual among the oncogenic DNA tumor viruses in its failure to inhibit or degrade p53 in tumors [39] . The possibility of spontaneous loss of p53 in CZ osteosarcomas was nevertheless investigated . Tumor cell lines from CZ , C3 and F1 mice were examined for p53 expression and for their response to DNA damage ( Figure 5 ) . Cells were exposed to actinomycin D ( 10–50 nM ) for 24 hours . This led to phosphorylation on serine-18 ( serine-15 in humans ) and accumulation of p53 and to induction of p21Cip1/Waf1 , as previously shown for cells in culture productively infected by the virus [40] . Similar results were found for the other lines ( data not shown ) . pRb function is disrupted by binding to Py large T antigen [41] . However , the ability of the virus to induce tumors does not depend on this interaction [11] , and metastatic mammary tumors develop without large T in middle T transgenic mice [16] . The large T-pRb interaction is expected to occur equally in tumors of CZ , C3 and F1 mice . While disruption of pRb by large T may be necessary for metastatic behavior , it is not sufficient . It thus appears that the difference between CZ and C3 mice with respect to bone tumor metastasis is based on host genetic factor ( s ) unrelated to p53 loss or to the action of the virus on pRb . As in the human disease , the downstream effectors of metastasis in the engineered mouse models remain largely unknown . Elevated expression of MMP-2 and NFAT are important factors in the invasive behavior of CZ osteosarcoma cells in vitro . Enhanced operation of a pathway linking these factors provides a plausible mechanism contributing to the metastatic behavior of osteosarcomas in CZ mice . The genetic determinant ( s ) of metastasis in this system nevertheless remain unknown . Factors that impinge on calcineurin activation of NFAT [29] , on secretion or activation of MMPs , expression of specific integrins involved in tumor cell invasion [42] , and microenvironmental influences in the lung are among many additional factors that may be involved . Mapping and identifying genetic determinants in this system will be necessary to confirm the importance of the NFAT→MMP-2 pathway and to fully establish the molecular basis of metastasis .
Czech II/E mice ( Jackson Laboratory , Bar Harbor , ME ) and C3H/BiDa mice ( National Cancer Institute , Frederick , MD ) were maintained in our SPF barrier animal facility . No spontaneous tumors were noted . Newborn mice were inoculated i . p . with ∼50 µl of a crude virus suspension ( RA strain , 2−5×106 PFU ) , transferred to a separate facility and monitored for tumor development as described [14] . The methods for mice use and care were approved by the Harvard Medical Area Standing Committee on Animals ( “HMA IACUC” ) , and are in accordance with PHS policy on Care and Use of Laboratory Animals under the guidance of the Office of Laboratory Animal Welfare ( OLAW ) within the NIH . Histology was carried out through the Rodent Histopathology Core of the Dana Farber-Harvard Cancer Center . Screening for lung metastases was carried out by step-sectioning and microscopic examination . A total of five sections ( 5 µ ) taken 100 µ apart were examined . Osteosarcomas were removed at necropsy , cut into small fragments , washed and digested at 37° C overnight in medium containing 200 U/ml Collagenase , Type 1 ( Worthington ) . Cells were spun out , resuspended and plated in DMEM containing 10% FBS . Cells were expanded initially to five 10 cm dishes and frozen . All experiments were carried out within 2–5 further passages . CZ-I and CZ-II are from osteosarcomas in two individual CZ mice . CZ-Mx is a lung metastasis from the same mouse as CZ-I . C3-I , C3-II and C3-III are non-metastatic osteosarcma lines from three individual C3 mice . F1-I and F1-II are from two F1 mice . F1-Mx is a lung metastasis from the same mouse as F1-I . Conditioned medium was prepared from confluent cultures of cells grown in DMEM with 10% FBS . Cultures were washed and incubated at 37°C overnight in serum-free DMEM . Media were concentrated ten fold using centrifugal filter devices ( Millipore Corporation ) . Concentrates were filter sterilized and stored at −80°C . Levels of OPN and VEGF in conditioned media were determined by double sandwich ELISA . Tumor cell migration and invasion were assayed using Boyden chambers [22] with polycarbonate membranes ( 8 µm pore size; 6 . 5 mm diameter from Corning Inc . , Corning , NY ) . Tumor cells ( 3×105/well in DMEM with 0 . 1% BSA ) were placed in the upper compartment and either conditioned medium or recombinant osteopontin ( 4 µg/well; R&D Systems , Inc . , Minneapolis , MN ) in the lower compartment . Incubation was at 37°C for 1 h . Where indicated , conditioned medium was neutralized with anti-OPN or anti-VEGF antibody ( 5 µg/well; R&D Systems , Inc ) . After removal of cells from the upper side of the filter , cells on the bottom side were stained by crystal violet and counted over 7 randomly chosen fields . Invasion assays were performed in the same manner except that the polycarbonate filters were coated with ECM gel ( 20 µg/well; Sigma ) . Tumor cells ( 5×105 ) were added to the upper chamber and incubated for 4 h at 37°C . Procedures were essentially as described [43] . Acetone precipitates of concentrated conditioned media were resuspended and separated on 10% polyacrylamide gel containing 1 mg/mL gelatin . Cell extracts were separated by SDS-PAGE , transferred and blotted with antibody against NFAT c1 ( Santa Cruz Biotech , Santa Cruz , CA ) , MMP-2 ( Santa Cruz Biotech ) , phospho-p53 ( ser-15 ) ( Cell Signaling , Beverly , MA ) , or p21 ( C-19 ) ( Santa Cruz Biotech ) . Membranes were washed and incubated with Alexa fluor 680 anti-mouse and 800 anti-rabbit IgG antibodies ( Invitrogen , Carlsbad , CA ) . Odyssey infrared imaging system ( LI-COR Biosciences , Lincoln , NE ) was used to reveal band intensities and integrated intensity values were obtained using LI-COR Odyssey software ( Li-COR Biosciences ) . Cells were cotransfected with pTL-Renilla vector and pGL3 basic or pGL3 containing IL2 promoter by using Lipofectamin 2000 ( Invitrogen ) according to the manufacturer's protocol . At 24 hr , cells were harvested , washed and lysed with 100 µl of Passive lysis buffer ( Promega , Madison , WI ) at 4°C for 15 min . Twenty microliters of the cell lysate and a Dual assay kit ( Promega ) were used to measure fluorescence intensities . Firefly luciferase activities were normalized to those of renilla luciferase . Cells were transfected with NFATc1 siRNA ( NFATC1MSS275981 and NFATC1MSS275982; Invitrogen ) or Stealth RNAi Negative Control Duplex ( Invitrogen ) using Oligofectamin and Opti-MEM medium ( Invitrogen ) . Levels of NFAT at 24 hr post- transfection , were determined by immunoblotting . TIMP-2 ( 400 ng/ml , Sigma ) was used to pretreat CZ and F1 bone tumor cells for 30 min and was added to the upper and lower chambers during the assay . 72 kDa type IV collagenase : P33434 ( MMP2_MOUSE ) Nuclear factor of activated T-cells , cytoplasmic 1 : O88942 ( NFATc1_MOUSE ) Osteopontin : P10923 ( Osteopontin_MOUSE ) Cellular tumor antigen p53 : P02340 ( P53_MOUSE ) Metalloproteinase inhibitor 2 : P25785 ( TIMP2_MOUSE )
|
The oncogenic mouse polyoma virus and its mutants have previously been used to investigate viral determinants of tumor induction using a standard inbred mouse strain as a common host . Here we use wild type virus to investigate the role of the host genetic background , focusing on two host strains that differ with respect to bone tumor metastasis . Comparing osteosarcoma cell lines from these mice , we have identified a molecular pathway that underlies invasive behavior in vitro and correlates with metastasis in vivo . The pathway involves secretion of the metalloproteinase MMP-2 under partial control of NFAT as a transcriptional regulator . This virus-host system reflects an important feature of human osteosarcoma with respect to pulmonary metastasis . Based on naturally occurring differences among inbred mice , the model differs from genetically engineered models targeting p53 or pRb as known risk factors in the human disease . Here , metastatic osteosarcomas retain functional p53 . As noted by others , the frequency of p53 loss in patients with localized versus metastatic disease is the same , suggesting that events beyond p53 loss are important in metastasis . While the downstream effectors of metastasis in the genetically engineered models remain unknown , evidence presented here implicates upregulation of an NFAT → MMP-2 pathway in the development of metastatic osteosarcoma .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"oncology/sarcomas",
"virology/animal",
"models",
"of",
"infection",
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"infectious",
"diseases/viral",
"infections",
"virology/viruses",
"and",
"cancer"
] |
2010
|
Polyoma Virus-Induced Osteosarcomas in Inbred Strains of Mice: Host Determinants of Metastasis
|
HIV persists in a small pool of latently infected cells despite antiretroviral therapy ( ART ) . Identifying cellular markers expressed at the surface of these cells may lead to novel therapeutic strategies to reduce the size of the HIV reservoir . We hypothesized that CD4+ T cells expressing immune checkpoint molecules would be enriched in HIV-infected cells in individuals receiving suppressive ART . Expression levels of 7 immune checkpoint molecules ( PD-1 , CTLA-4 , LAG-3 , TIGIT , TIM-3 , CD160 and 2B4 ) as well as 4 markers of HIV persistence ( integrated and total HIV DNA , 2-LTR circles and cell-associated unspliced HIV RNA ) were measured in PBMCs from 48 virally suppressed individuals . Using negative binomial regression models , we identified PD-1 , TIGIT and LAG-3 as immune checkpoint molecules positively associated with the frequency of CD4+ T cells harboring integrated HIV DNA . The frequency of CD4+ T cells co-expressing PD-1 , TIGIT and LAG-3 independently predicted the frequency of cells harboring integrated HIV DNA . Quantification of HIV genomes in highly purified cell subsets from blood further revealed that expressions of PD-1 , TIGIT and LAG-3 were associated with HIV-infected cells in distinct memory CD4+ T cell subsets . CD4+ T cells co-expressing the three markers were highly enriched for integrated viral genomes ( median of 8 . 2 fold compared to total CD4+ T cells ) . Importantly , most cells carrying inducible HIV genomes expressed at least one of these markers ( median contribution of cells expressing LAG-3 , PD-1 or TIGIT to the inducible reservoir = 76% ) . Our data provide evidence that CD4+ T cells expressing PD-1 , TIGIT and LAG-3 alone or in combination are enriched for persistent HIV during ART and suggest that immune checkpoint blockers directed against these receptors may represent valuable tools to target latently infected cells in virally suppressed individuals .
Although antiretroviral therapy ( ART ) is highly effective at suppressing HIV replication , viral reservoirs persist despite treatment and lead to rapid viral rebound when ART is interrupted [1–4] . A major step to achieve natural control of HIV replication after ART cessation would be to eliminate , or at least reduce , the number of long-lived infected cells from which HIV reignite infection . The characterization of cell surface markers that could identify HIV-infected cells persisting during ART is a research priority towards an HIV cure [5] as it could lead to the development of novel eradication strategies . Several subsets of CD4+ T cells harbor replication-competent HIV during ART . These CD4+ T cells are usually defined on the basis of their differentiation stage [6–8] , functionality or homing potential [9 , 10] . Central memory ( TCM ) and transitional memory ( TTM ) CD4+ T cells were identified as the major cellular reservoirs for HIV during ART [6] . More recently , a less differentiated subset of long-lived cells with high self-renewal capacity , the stem-cell memory CD4+ T cells ( TSCM ) , has been identified as a main contributor to long-term HIV persistence [7 , 8] . The functional and homing capacities of CD4+ T cells also dictate their capacity to serve as persistent reservoirs for HIV: Th17 and Th1/Th17 CD4+ T cells as well as cells expressing CCR6 and CXCR3 show increasing contribution to the viral reservoir with duration of ART [11 , 12] . Immune checkpoint molecules ( ICs ) are co-inhibitory receptors which down-modulate immune responses to prevent hyper-immune activation , minimize collateral damage , and maintain peripheral self-tolerance [13] . ICs are up regulated upon T-cell activation and constrain the effector response through feedback inhibition . Overexpression of these molecules is associated with T-cell exhaustion and dysfunction in cancer and chronic viral infections , including HIV [14–17] . We hypothesized that ICs , through their ability to inhibit T-cell activation , will favour HIV latency during ART , and that CD4+ T cells expressing ICs would be enriched for persistent HIV in individuals receiving ART . We focused our analysis on 7 ICs , namely PD-1 ( programmed cell death-1 ) , CTLA-4 ( cytotoxic T-lymphocyte-associated protein 4 ) , LAG-3 ( lymphocyte activation gene 3 ) , TIGIT ( T-cell immunoglobulin and ITIM domain ) , TIM-3 ( T cell immunoglobulin and mucin 3 ) , CD160 and 2B4 ( CD244 ) . PD-1 , a member of the B7-CD28 superfamily , enforces an inhibitory program that blocks further TCR-induced T-cell proliferation and cytokine production [18 , 19] . In HIV infection , high levels of PD-1 are associated with T cell exhaustion [14–16 , 20] and incomplete immunological response to ART [21] . CTLA-4 , a CD28 homolog , regulates the amplitude of T-cell activation by both outcompeting CD28 in binding CD80 and CD86 , as well as actively delivering inhibitory signals to T cells [13] . TIGIT , which also belongs to the B7/CD28 superfamily , acts as a co-inhibitory molecule by directly down regulating proliferation of human T cells [22] , but also by modulating cytokine secretion of DCs , decreasing IL-12 and enhancing IL-10 productions [23] . TIGIT has been recently associated with CD8+ T-cell dysfunction during HIV infection [24] . The expression of 2B4 ( CD244 ) , a member of the signalling lymphocyte activation molecule ( SLAM ) is also modulated on T cells during HIV infection [17 , 25] . LAG-3 , a member of the immunoglobulin superfamily , is structurally highly homologous to the CD4 receptor and share MHC-II as a ligand [26] . Its expression on T regulatory cells plays a role in the modulation of T cell homeostasis and effector T cell responses [27 , 28] . TIM-3 is also an immunoglobulin superfamily member and its expression is increased on HIV-specific CD8+ and CD4+ T cells [29 , 30] . Finally , CD160 , through its binding to its ligand Herpes Virus Entry Mediator ( HVEM ) , an atypical member of TNF-receptor superfamily , delivers a co-inhibitory signalling to CD4+ T cells or CD8+ T cells dampening their activation in HIV-infected individuals [31 , 32] . To assess the relationship between the expression of these ICs and HIV persistence , we analysed the association between their levels of expression on CD4+ T cells and the size of the HIV reservoir in individuals receiving ART for at least 3 years .
Forty-eight HIV-infected participants receiving suppressive ART were recruited at the University of California San Francisco ( UCSF ) for this cross-sectional study . Participants were receiving ART for >3 years , had CD4+ T-cell count >350 cells/μl and HIV RNA <40 copies/mL as measured by the Abbott real time HIV-1 PCR for at least 3 years . Whole blood ( 50mL ) was collected by regular blood draw . For cell sorting experiments , 27 HIV-infected individuals were enrolled at UCSF and at VGTIFL and underwent leukapheresis . All subjects signed informed consent forms approved by the UCSF and Martin Memorial Health Systems review boards ( IRB #10–1320 , Ref # 068192 and FWA #00004139 , respectively ) . PBMCs were isolated from peripheral blood and leukapheresis using previously described methods [6 , 33] . Cryopreserved PBMCs were thawed , washed and stained for phenotyping or cell sorting . Two antibody panels were used to measure the expression of IC in subsets of memory CD4+ T cells . The same antibody backbone was used in the two panels: CD3-Alexa700 ( clone UCHT1 , BD#557943 ) , CD4-QDot605 ( clone S3 . 5 , Invitrogen#Q10008 ) , CD8-PB ( clone RPA-T8 , BD#558207 ) , CD14-V500 ( clone M5E2 , BD#561391 ) , CD19-AmCyan ( clone SJ25C1 , BD#339190 ) , LIVE/DEAD Aqua marker ( Invitrogen#L34957 ) , CD45RA-APC-H7 ( clone HI100 , BD#560674 ) , CD27-BV650 ( clone O323 , Biolegend#302828 ) and CCR7-PE-Cy7 ( clone 3D12 , BD#557648 ) . The following antibodies were added to this backbone: PD-1-AF647 ( clone EH12 . 1 , BD#560838 ) , CTLA-4-PE ( clone BNI3 , BD#555853 ) , LAG-3-FITC ( R&D#FAB2319F ) , TIGIT-PerCP-eF710 ( clone MBSA43 , eBioscience#46-9500-41 ) , TIM-3-PE ( clone F38-2E2 , Biolegend#345006 ) , CD160-AF488 ( clone By55 , eBioscience#53–1609 ) , 2B4-PerCP-Cy5 . 5 ( clone C1 . 7 , Biolegend#329515 ) . For expression of all ICs , gates were defined using fluorescence minus one controls . CD4+ T-cell subsets were identified by CD27 , CD45RA , and CCR7 expression on CD4+ T cells after exclusion of dump positive cells ( LIVE/DEAD , CD14 and CD19 ) . ICs were measured in gated CD4+ T-cell subsets including naïve CD4+ T cells ( CD3+CD8-CD4+CD45RA+CCR7+CD27+ ) , central memory CD4+ T cells ( CD3+CD8-CD4+CD45RA-CCR7+CD27+ ) , transitional memory CD4+ T cells ( CD3+CD8-CD4+CD45RA-CCR7-CD27+ ) , effector memory CD4+ T cells ( CD3+CD8-CD4+CD45RA-CCR7-CD27- ) and terminally differentiated CD4+ T cells ( CD3+CD8-CD4+CD45RA+CCR7-CD27- ) . Data was acquired on a BD LSR II flow cytometer using the FACSDiva software ( Becton Dickinson ) and analysed using Flow Jo version 9 ( Treestar ) . Central , transitional and effector memory CD4+ T cells were sorted based on their expression of PD-1 , TIGIT or LAG-3 . The antibodies used for sorting were similar than those used for phenotyping with the exception of CD27-QDot655 ( clone CLB-27/1 , Invitrogen#Q10066 ) . In a second set of experiments , total memory CD4+ T cells ( CD3+CD4+CD45RA- ) were sorted based on their expression of PD-1 , TIGIT and LAG-3 . Sorted cells were collected using an ARIA FACS sorter ( Becton Dickinson ) . Total CD4+ T cells were isolated from cryopreserved PBMCs using magnetic depletion as per the manufacturer’s protocol ( Stem Cell Technologies , Vancouver , Canada ) . Total CD4+ T cells or sorted CD4+ T cell subsets were used to measure the frequency of cells harboring HIV DNA ( total , integrated and 2-LTR circles ) by real time nested PCR as previously described [34] ( S1A Text ) . The CA-US RNA was measured by real time nested PCR as previously described [35] . The frequency of CD4+ T cells with inducible multiply spliced HIV RNA was determined using Tat/rev inducible limiting dilution assay ( TILDA ) [36] . Data distributions were assessed through descriptive statistics and scatter plots . Negative binomial regression models were run for each set of comparisons with the percentage of CD4+ T cells expressing ICs being the predictor and the measure of HIV persistence the outcome . We chose this approach for reasons described previously [12 , 35] , and for consistency with those previous publications ( S1B Text ) . The approach allowed us to fit models adjusting for the effects of absolute current or nadir CD4+ T-cell , which were examined for all combinations of IC predictors and HIV persistence outcome measures . In addition , the negative binomial regression models take into account that copies/input is measured with less precision when the number of copies is lower and when the amount of input is lower . The methods also permit proper quantitative use of instances where zero copies were present in the specimen assayed , without a need for ad hoc modifications to permit taking logarithms . We did not evaluate the results of alternative analysis methods and did not choose the methods post-hoc based on the results that they produced . Analyses were run in Stata version 13 . 1 ( Stata Corp , College Station , TX ) . For TILDA results analysis , we estimated the within-person fold difference in TILDA between the 2 cell subsets analyzed ( memory CD4+ T cells expressing any versus none of the ICs ) by fitting a maximum likelihood model to the raw data on numbers of positive and negative wells at each dilution ( S1C Text ) .
To determine the relationship between ICs and HIV persistence , 48 HIV-infected participants on suppressive ART for a median time ( IQR ) of 8 . 5 years ( 5 . 0–12 . 4 ) and a median CD4+ T-cell count ( IQR ) of 684 cells/μL ( 533–858 ) were recruited ( Table 1 ) . The expressions of 7 ICs on CD4+ T cells ( PD-1 , CTLA-4 , LAG-3 , TIGIT , TIM-3 , CD160 and 2B4 ) were measured by multiparametric flow cytometry ( S1 Fig ) . The frequencies of CD4+ T cells expressing these ICs were variable ( median ( IQR ) of 16 . 7% ( 13 . 2–22 . 7 ) , 12 . 2% ( 8 . 8–16 . 4 ) , 12 . 0% ( 8 . 9–16 . 1 ) , 9 . 5% ( 3 . 5–18 . 5 ) , 1 . 1% ( 0 . 8–2 . 5 ) , 0 . 8% ( 0 . 6–1 . 5 ) and 0 . 7% ( 0 . 6–1 . 0 ) for TIGIT , PD-1 , LAG-3 , 2B4 , CD160 , TIM-3 and CTLA-4 respectively ) ( Fig 1A ) . The size of the HIV reservoir was determined by measuring the frequencies of CD4+ T cells harboring integrated HIV DNA , total HIV DNA and 2-LTR circles as well as cell-associated unspliced ( CA-US ) HIV RNA ( Fig 1B and S1 Table ) . Total HIV DNA and cell-associated US HIV RNA were detected in all samples tested , whereas integrated HIV DNA and 2-LTR circles were detected in 98% , and 80% of the samples , respectively . We evaluated the association between markers of HIV persistence and the frequencies of CD4+ T cells expressing ICs using a negative binomial regression model that was adjusted for current and nadir CD4+ T-cell counts when indicated ( Table 2 and S2–S4 Tables ) . Using these tailored analytical methods for HIV reservoir measurements , we identified 3 ICs for which the expression on CD4+ T cells was statistically significantly associated with the frequency of CD4+ T cells harboring integrated HIV DNA , namely PD-1 , TIGIT and LAG-3 ( Fig 1C–1E and Table 2 ) . These correlations persisted after adjusting for nadir CD4+ T-cell counts but were no longer significant after adjusting for current CD4+ T-cell count , a clinical parameter strongly associated with the size of the reservoir during ART [6 , 37 , 38] . The frequency of PD-1 expressing CD4+ T cells was also associated with the frequency of CD4+ T cells harboring total HIV DNA ( S3 Table ) , but only marginally ( 1 . 23-fold effect , p = 0 . 07 ) when the model was adjusted for current CD4+ T-cell count . CA-US HIV RNA and 2-LTR circles did not show statistically significant correlation with any IC expression levels , with the exception of a negative association between the frequency of CD160+ CD4+ T cells and 2-LTR circles that remained statistically significant after adjusting for current and nadir CD4+ T-cell count ( S3 and S4 Tables ) . ICs are co-expressed on exhausted CD4+ and CD8+ T cells during untreated HIV infection [39] . Using a Boolean gating strategy , we determined the frequency of CD4+ T cells co-expressing PD-1 and/or TIGIT and/or LAG-3 in our cohort of 48 HIV-infected participants receiving suppressive ART ( Fig 2A and 2B ) . The majority of CD4+ T cells did not express any of these markers ( median ( IQR ) of 65 . 8% ( 59 . 0–72 . 4 ) ) ( S6 Table ) . Less than 10% ( 8 . 5% ) of CD4+ T cells expressed more than one of these markers and 0 . 9% simultaneously expressed PD-1 , TIGIT and LAG-3 . We further assessed if the frequency of these discrete CD4+ T-cell subsets was associated with markers of HIV persistence . Using the negative binomial regression model , we found that the frequency of CD4+ T cells not expressing PD-1 , TIGIT and LAG-3 was strongly and negatively correlated to the frequency of CD4+ T cells harboring integrated HIV DNA ( p = 0 . 002 , Table 3 and Fig 2C ) . Conversely , the frequency of CD4+ T cell co-expressing PD-1 , TIGIT and LAG-3 was strongly and positively associated with the frequency of CD4+ T cells harboring integrated HIV DNA ( p = 0 . 001 , Table 3 and Fig 2F ) . Interestingly , the frequencies of CD4+ T cells co-expressing TIGIT with either PD-1 or LAG-3 were also positively associated with the frequency of CD4+ T cells harboring integrated HIV DNA ( p = 0 . 002 and p = 0 . 029 respectively , Table 3 and Fig 2D and 2E ) . Although several of these associations were less or no longer statistically significant when the model was adjusted for the current CD4+ T-cell count , the association between the size of the HIV reservoir and the frequency of triple positive cells ( PD-1+ , LAG-3+ and TIGIT+ ) remained statistically significant after adjustment ( p = 0 . 038 ) . Adjusting for duration of ART did not produce any substantial changes to the results from the unadjusted analysis ( all fold-effects adjusted for ART duration within 7% of those unadjusted ) . All together , these results indicate that the co-expression of PD-1 , TIGIT and LAG-3 identifies a unique subset of CD4+ T cells that strongly predicts the frequency of cells harboring integrated HIV DNA during ART . HIV persists preferentially in memory CD4+ T-cell subsets [6–8] . To determine the role played by ICs in each individual CD4+ T-cell memory subset , we first analyzed the expression of PD-1 , TIGIT and LAG-3 , the 3 ICs we identified to be associated with HIV persistence , on naïve ( TN ) , central memory ( TCM ) , transitional memory ( TTM ) , effector memory ( TEM ) and terminally differentiated ( TTD ) cells in 48 HIV-infected participants ( Cohort 1: clinical characteristics in Table 1 ) ( Fig 3A , 3B and 3C respectively ) . As expected , TN cells expressed low levels of these ICs . The frequency of CD4+ T cells expressing PD-1 or LAG-3 increased with differentiation , with TEM cells displaying the highest levels of expression of these markers . The highest frequency of TIGIT+ cells was found within the TTM subset . These results demonstrated that the subsets of memory cells that were previously shown to harbor persistent HIV during ART express PD-1 , TIGIT and LAG-3 . To determine whether PD-1 , TIGIT and LAG-3 identify cells more likely to carry persistent HIV in virally suppressed participants , individual memory CD4+ T-cell subsets were sorted based on their expression of PD-1 , TIGIT or LAG-3 in a subset of subjects who underwent leukapheresis ( Cohort 2: clinical characteristics in Table 1 ) and the results were analyzed by negative binomial regression model ( S5 Table ) . The frequency of cells harboring integrated HIV DNA was moderately higher in PD-1 expressing TTM when compared to their PD-1 negative counterparts ( p = 0 . 053 , fold-difference = 1 . 5 ) ( Fig 3D ) . TEM cells expressing TIGIT were enriched for integrated genomes when compared to their TIGIT- counterparts ( p = 0 . 001 , fold-difference = 2 . 7 ) ( Fig 3E ) . Finally , all the memory CD4+ T-cell subsets ( TCM , TTM and TEM cells ) expressing LAG-3 were enriched for integrated HIV DNA when compared to their negative counterparts ( p<0 . 0001 , fold-difference = 1 . 9 , p = 0 . 003 , fold-difference = 1 . 8 and p = 0 . 030 , fold-difference = 2 . 5 respectively ) ( Fig 3F ) . All together these results indicate that PD-1 , TIGIT and LAG-3 enrich for infected cells in distinct memory CD4+ T-cell subsets in individuals on ART . We calculated the contribution of cells expressing PD-1 , TIGIT or LAG-3 to the total reservoir by taking into account the frequency of these subsets within the CD4 compartment and their relative infection frequencies . The mean contributions of CD4+ T cells expressing PD-1 , TIGIT and LAG-3 were 29% , 34% and 31% , respectively ( S2 Fig ) . As a comparator , TCM , TTM and TEM cells contributed 43% , 27% and 24% to the pool of infected cells in these same virally suppressed individuals . These data indicate that a third of the reservoir is encompassed in cells expressing each individual marker . We then determined if the combination of PD-1 , TIGIT and LAG-3 would further enrich memory CD4+ T cells for HIV-infected cells during ART . The average frequency of cells expressing 0 , 1 , 2 or 3 of these markers in the memory CD4+ T compartment ( CD45RA- ) from our cohort of 48 individuals ( Table 1 ) indicated that an average of 33% of memory CD4+ T cells expressed one of the 3 IC only , 12% expressed 2 and 2% expressed the 3 markers simultaneously . Large numbers of memory CD4+ T cells were sorted based on their expression of PD-1 , TIGIT and LAG-3 from 5 individuals . The combination of these 3 markers allowed us to sort eight subsets of cells to high purity , namely PD-1/TIGIT/LAG-3 triple - , PD-1 single + , TIGIT single + , LAG-3 single + , PD-1/TIGIT double + , TIGIT/LAG-3 double + , PD-1/LAG-3 double + and PD-1/TIGIT/LAG-3 triple + cells . The frequency of cells harboring integrated HIV DNA was measured by qPCR in each sorted subset ( S3 Fig ) and the mean frequency for each category was calculated relative to total CD4+ T cells ( Fig 4B ) . Memory CD4+ T cells showed a gradual enrichment in HIV-infected cells when expressing an increasing number of ICs . Memory CD4+ T cells expressing simultaneously PD-1 , TIGIT and LAG-3 were enriched for HIV-infected cells up to 10 times more when compared to total CD4+ T cells , with a median fold increase ( IQR ) of 8 . 15 ( 4 . 92–9 . 59 ) . These results demonstrated that memory CD4+ T cells expressing a combination of ICs were highly enriched for integrated HIV DNA during ART . As the majority of HIV genomes , even when integrated , are defective [40] , we assessed if PD-1 , TIGIT and LAG-3 identify cells in which HIV production can be induced . As the frequency of triple positive cells was too low to perform this experiment , we sorted memory CD4+ T cells ( CD45RA- ) expressing any ( i . e . at least one ) versus none of PD-1 , TIGIT or LAG-3 ( mLPT+ and mLPT- respectively ) . We measured the frequency of cells in each population that transcribe multiply spliced HIV RNA molecules upon induction with PMA/ionomycin using the Tat/rev induced limiting dilution assay ( TILDA ) [36] . Tat/rev transcripts were detectable by TILDA in both cell subsets from all of the 8 individuals tested . The rate of inducible virus per million cells was estimated in our maximum likelihood model to average 3 . 0-fold higher in mLPT+ cells than in mLPT- cells from the same participant ( 95% CI 1 . 0 to 9 . 0 , p = 0 . 049 , S1C Text ) ( Fig 4C ) . Taking into account the frequency of these cell subsets , the contribution of cells expressing at least one of these markers to the total pool of memory CD4+ T cells infected with inducible HIV genomes was calculated . This contribution ranged from 30 to 98% ( median of 76% ) , indicating that the majority of inducible HIV genomes were found in memory CD4+ T cells expressing at least one of these markers . These experiments provide evidence that memory CD4+ T cells expressing PD-1 , TIGIT and/or LAG-3 are enriched for HIV-infected CD4+ T cells harboring inducible proviruses during ART .
In this study , we identified PD-1 , TIGIT and LAG-3 as novel markers of cells that are more frequently infected in HIV-infected individuals receiving suppressive ART . Co-expression of the 3 ICs identified a unique subset of CD4+ T cells that was strongly associated with the size of the HIV reservoir and that was highly enriched for integrated HIV DNA . Finally , our results provide evidence that memory CD4+ T cells expressing at least one of these markers are the major contributors to the pool of inducible HIV genomes during ART . The frequencies of CD4+ T cells expressing PD-1 , CTLA-4 , LAG-3 and TIM-3 were similar to those reported by other groups [41–44] , indicating that the cohort of participants used for this study is likely to be representative of the HIV population receiving suppressive ART . In addition , the association between PD-1 and TIGIT expression on CD4+ T cells and the frequency of CD4+ T cells carrying HIV proviruses was in agreement with previously reported findings [6 , 24 , 41] . We found positive associations between the expression of PD-1 , LAG-3 and TIGIT in CD4+ T cells and the frequency of cells harboring integrated HIV DNA . Of note , these three markers showed the strongest inverse associations with CD4+ T cell counts among the 7 markers we examined , suggesting a link between T cell homeostasis and HIV persistence ( S4A , S4B and S4C Fig ) . The associations between individual IC expression and HIV persistence marker were substantially smaller and no longer statistically significant after adjusting for current CD4+ T-cell count . These findings from the negative binomial regression models suggest that the current CD4+ T-cell count is an important predictor of the size of the HIV reservoir when measured as the frequency of cells harboring proviral genomes [6 , 37] . Importantly , and in contrast to cells expressing a single marker , the frequency of cells co-expressing simultaneously PD-1 , TIGIT and LAG-3 was strongly associated with the size of the reservoir and remained after adjusting for nadir and current CD4+ T-cell counts . This result reinforces the possibility of a direct—and maybe synergistic—role for these molecules in HIV persistence during ART . In addition , the frequency of CD4+ T cells co-expressing simultaneously PD-1 , TIGIT and LAG-3 positively correlated with the frequency of CD4+ T cells expressing HLADR/CD38 ( p = 0 . 003 , r = 0 . 42 ) and Ki67 ( p = 0 . 022 , r = 0 . 33 ) ( S1D and S1E Text and S4D and S4E Fig ) . These associations suggest that the persistence of the small pool of cells expressing the 3 markers is associated with T cell activation and proliferation . Interestingly , we observed a strong negative association between CD160 expression and 2-LTR circles . Notably , this correlation remained after adjusting for current and nadir CD4+ T cell counts . A possible explanation for these findings is that CD160+ cells may be preferential targets for infection and depletion during ART , which would explain the strong negative association between the frequency of CD160+ CD4+ T cells and a putative marker of persistent viral replication . By sorting TCM , TTM and TEM cells expressing PD-1 , TIGIT and LAG-3 , we observed that cells expressing these markers were enriched for HIV-infected cells in different memory CD4+ T-cells subsets during ART . While LAG-3 enriched for integrated HIV DNA in all memory subsets ( TCM , TTM , and TEM ) , PD-1 and TIGIT enriched for HIV genomes exclusively in TTM and TEM cells , respectively . These observations suggest that ICs may exert different pro-latency effects in subsets endowed with distinct proliferative and activation status . One may hypothesize that different ICs provide infected cells with different selective advantage to persist by counteracting distinct stimuli specific to an individual memory cell subset . Further investigations will be needed to characterize the mechanisms by which these ICs may specifically contribute to HIV persistence within these distinct subsets . Overall , the majority of inducible HIV genomes were found in memory CD4+ T cells expressing at least one of these markers ( median of 76% ) . Although triple negative cells also contain inducible HIV genomes , our data provide evidence that there is an enrichment for inducible viral genomes in CD4+ T cells expressing these markers . Importantly , we found a gradual enrichment in integrated HIV DNA in cells that express multiple ICs simultaneously . This observation mirrors the synergistic mechanisms of action of these receptors to dampen T cell functions . Indeed , LAG-3 and PD-1 are commonly co-expressed on exhausted or dysfunctional T cells in models of chronic infections [45] , autoimmune diseases [46] , and cancers [47 , 48] . Potential synergistic functions were highlighted in murine models of autoimmune diseases [49] . Anti-LAG-3 blocking mAb has recently entered clinical testing in cancer in monotherapy or in combination therapy with anti-PD-1 ( NCT01968109 ) . Additionally , TIGIT is co-expressed with PD-1 on activated CD8+ tumor-infiltrating lymphocytes from patients with melanoma [50] . Blockages of TIGIT and PD-1 synergize to improve T cell proliferation , cytokines production and degranulation in vivo in melanoma treatment and in vitro in HIV infection [24 , 50] . All together , these studies indicate that ICs can synergize to repress T cell functions and suggest that these synergies may also play a role in HIV persistence during ART . The expression of PD-1 , LAG-3 and TIM-3 on CD4+ and CD8+ T cells prior to ART was recently identified as a strong predictor of time to viral rebound after treatment interruption in the SPARTAC study [43] . It is possible that CD4+ T cells expressing these markers before ART represent a preferential niche for the establishment of a stable reservoir for HIV and that latently infected cells expressing these markers preferentially persist during ART , as suggested by our observations . In our study , we identified a discrete subset of CD4+ T cells co-expressing PD-1 , TIGIT and LAG-3 as an important predictor of the frequency of cells harboring integrated HIV DNA during ART . Of note , the expression of TIGIT before ART initiation was not measured in the SPARTAC study and further studies will be needed to determine if this IC could also represent a pre-ART predictor of viral rebound . Our data provide a rationale for the use of immune checkpoint blockers ( ICBs ) to target latently infected cells during ART . Targeting ICs by ICBs , a novel class of molecules in development in oncology , may have a double benefit in the context of HIV remission by both targeting latently infected cells and restoring HIV-specific T cell immunity . By enhancing T cell activation and increasing viral transcription , ICBs may facilitate HIV reactivation in latently infected cells when used alone or in combination with latency reversing agents . The anti-CTLA-4 antibody iplimumab was recently shown to significantly increase CA-US HIV RNA in an HIV-infected individual on ART , consistent with latency reversal [51] . An alternative mechanism of action of some ICBs would be to directly deplete cells expressing these markers , as observed with the anti-CTLA-4 ipilumimab , which induces direct elimination of CTLA-4+ regulatory T cells in tumor tissue in patients with melanoma [52] . Our results suggest that the administration of antibodies with effector functions targeting PD-1 , LAG-3 and TIGIT may significantly reduce the size of the latent HIV reservoir during ART by targeting cells in which HIV persists . Several limitations are associated with our study . We have not adjusted p-values for multiple comparisons , because such adjustment would neglect the biological relationships among our positive results and would require that each analysis detract from the others , rather than reinforcing one another when there is biological coherence [53] ( S1F Text ) . Nevertheless , our evidence may be weaker than if it had arisen from a narrower set of analyses , and , in any case , additional studies will be needed to confirm the hypotheses supported by our results . Most of our analyses were performed using integrated HIV DNA as a marker of HIV persistence . We chose this readout as it was applicable to small subsets of CD4+ T cells on which measures of replication competent HIV cannot be performed . The majority of viral genomes persisting during ART are known to be defective [40 , 54] , and although our experiments indicate that cells that express ICs can produce multiply spliced RNA upon activation ( TILDA ) , they do not demonstrate that replication competent virus persists in these cells . In addition , our results are limited to circulating T cells . It is possible and indeed likely that the biology of ICs expression and HIV persistence will differ in tissues , particularly in secondary lymphoid tissues where many of the ligands for these receptors are likely to be expressed . A better understanding of the nature of the cells that encompass the latent HIV reservoir is a prerequisite to the development of novel curative strategies . Despite similarities in their mechanisms of action , PD-1 , TIGIT and LAG-3 are likely to be non-redundant in their functions . Blocking these pathways simultaneously may show synergies in latency reversal , as suggested by their synergistic activities in the restoration of T cell immunity .
|
The persistence of HIV in a small pool of long-lived latently infected resting CD4+ T cells is a major barrier to viral eradication . Identifying cellular markers that are preferentially expressed at the surface of latently infected cells may lead to novel therapeutic strategies to cure HIV infection . We identified PD-1 , TIGIT and LAG-3 as markers preferentially expressed at the surface of infected cells in individuals receiving ART . CD4+ T cells co-expressing these markers were highly enriched for cells carrying HIV . Our results suggest that PD-1 , TIGIT and LAG-3 may represent new molecular targets to interfere with HIV persistence during ART .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"hiv",
"infections",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"antiviral",
"therapy",
"pathogens",
"immunology",
"microbiology",
"cloning",
"neuroscience",
"learning",
"and",
"memory",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"preventive",
"medicine",
"rna",
"viruses",
"antiretroviral",
"therapy",
"cognition",
"memory",
"molecular",
"biology",
"techniques",
"vaccination",
"and",
"immunization",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"white",
"blood",
"cells",
"memory",
"t",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"hiv",
"t",
"cells",
"microbial",
"pathogens",
"molecular",
"biology",
"cell",
"biology",
"viral",
"persistence",
"and",
"latency",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"viral",
"diseases",
"lentivirus",
"cognitive",
"science",
"organisms"
] |
2016
|
CD4+ T Cells Expressing PD-1, TIGIT and LAG-3 Contribute to HIV Persistence during ART
|
Burkina Faso is endemic with soil-transmitted helminth infections . Over a decade of preventive chemotherapy has been implemented through annual lymphatic filariasis ( LF ) mass drug administration ( MDA ) for population aged five years and over , biennial treatment of school age children with albendazole together with schistosomiasis MDA and biannual treatment of pre-school age children through Child Health Days . Assessments were conducted to evaluate the current situation and to determine the treatment strategy for the future . A cross-sectional assessment was conducted in 22 sentinel sites across the country in 2013 . In total , 3 , 514 school age children ( 1 , 748 boys and 1 , 766 girls ) were examined by the Kato-Katz method . Overall , soil-transmitted helminth prevalence was 1 . 3% ( 95% CI: 1 . 0–1 . 8% ) in children examined . Hookworm was the main species detected , with prevalence of 1 . 2% ( 95% CI: 0 . 9–1 . 6% ) and mean egg counts of 2 . 1 epg ( 95% CI: 0–4 . 2 epg ) . Among regions , the Centre Ouest region had the highest hookworm prevalence of 3 . 4% ( 95% CI: 1 . 9–6 . 1% ) and mean egg counts of 14 . 9 epg ( 95% CI: 3 . 3–26 . 6 epg ) . A separate assessment was conducted in the Centre Nord region in 2014 using community-based cluster survey design during an LF transmission assessment survey ( TAS ) . In this assessment , 351 children aged 6–7 years and 345 children aged 10–14 years were examined , with two cases ( 0 . 6% ( 95% CI: 0 . 2–2 . 1% ) ) and seven cases ( 2 . 0% ( 95% CI: 1 . 0–4 . 1% ) ) of hookworm infection was identified respectively . The results using both age groups categorized the region to be 2% to <10% in STH prevalence according to the pre-defined cut-off values . Through large-scale preventive chemotherapy , Burkina Faso has effectively controlled STH in school age children in the country . Research should be conducted on future strategies to consolidate the gain and to interrupt STH transmission in Burkina Faso . It is also demonstrated that LF TAS provides one feasible and efficient platform to assess the STH situation for post LF MDA decision making .
Soil-transmitted helminthiasis ( STH ) , one of the major neglected tropical diseases ( NTDs ) , is caused by a group of nematodes , namely hookworms ( Ancylostoma duodenale and Necator americanus ) , Ascaris lumbricoides and Trichuris trchiura . Chronic infection with these parasites can cause malnutrition , iron deficiency , anemia and impairment of physical and intellectual development in school age children [1–3] . The transmission of STH is linked to poverty and lack of health practice , and associated with low parental literacy rates , poor hygiene and sanitation , and lack of access to safe and clean water [4–7] . The disease is widely endemic in the developing countries in the world and causes the highest burden of the NTDs among the poorest populations [8 , 9] . Worldwide , it is estimated that about two billion people in developing countries are infected with one or more species of helminths [10 , 11] , with approximately 300 million people suffering from severe morbidity resulting in 10 , 000–135 , 000 deaths annually [12] . STH infections are normally treated by a single dose of albendazole or mebendazole [13] . World Health Organization ( WHO ) recommends controlling morbidity caused by STH infections through preventive chemotherapy with anthelmintic drugs in pre-school age children , school age children , as well as adolescent girls , women of reproductive age and pregnant women ( second and third trimester ) [11 , 14] . The current global objective is to attain regular treatment of 75% of pre-school age children and school age children in all endemic countries by the year 2020 [15] . Burkina Faso is a West African country that is divided into 13 health regions with 63 health districts . The country has three sub climate zones: north-Sudanese in the south , sub-Sahelian in the middle and Sahelian in the north , with an annual rainfall between 400 and 1 , 200 mm [16] . The country is known to be endemic with STH according to historic data , but published literatures on population-based surveys is scarce [17–20] . A 1984 survey in two villages in Kaya in Centre Nord region showed that hookworm ( N . americanus ) prevalence was 14 . 7% in Louda and 9 . 3% in Damesma , while prevalence of A . lumbricoides or T . trichiura was found to be below 0 . 5% [17] . During the baseline data collection for schistosomiasis and STH in 2004–05 in four regions ( Boucle du Mouhoun , Nord , Sahel and Sud Ouest ) highly endemic with schistosomiasis , it was shown that hookworm infection among school age children was 6 . 3% , T . trichiura prevalence was 1 . 1% and no A . lumbricoides infection was found [21] . A 2015 systematic review and geostatistical meta-analysis in sub-Saharan Africa showed 9 . 9% prevalence for hookworm , 0 . 4% prevalence for both A . lumbricoides and T . trichiura and 10 . 7% prevalence for overall STH in Burkina Faso , from 2000 onwards [22] . Since the early 2000s , deworming activities in Burkina Faso have been implemented through a number of different platforms as shown in Table 1 . Firstly , the national lymphatic filariasis ( LF ) elimination program initiated annual mass drug administration ( MDA ) in 2001 with albendazole and ivermectin for LF elimination , targeting all individuals aged 5 years or older , and reached national coverage in 2005 . Secondly , the national schistosomiasis and STH control program was established in 2004 . Albendazole tablets were added to praziquantel distribution to treat school age children [23 , 24] . The MDA was conducted once every two years in all 63 health districts reaching approximately 90% coverage in school age children [23] . In 2007 , the national schistosomiasis and STH control program became part of the national integrated NTD program for the five major NTDs targeted by preventive chemotherapy [25] . Thirdly , deworming has also been implemented in pre-school age children ( 12–59 months old ) together with vitamin A supplementation through Child Health Days in the country with support from HKI and UNICEF . With the LF MDA being the largest deworming program and being gradually stopped in more and more districts since 2012 in Burkina Faso after achieving the LF program objectives , it is essential to assess the STH situation in order to plan for the STH-specific treatment strategies in the post-LF MDA setting . In 2013 , STH infections were assessed in school age children during a schistosomiasis sentinel site survey in 22 sentinel sites in 11 regions across the country . LF transmission assessment surveys ( TAS ) have been used to include STH assessment elsewhere through school-based surveys [26 , 27] . In 2014 , an assessment of STH infections during LF TAS was conducted in the Centre Nord region in Burkina Faso to test the feasibility of assessing STH through community-based LF TAS design , compared with the conventional school-based surveys . The current paper presents a full picture of current STH prevalence and distribution in the country , discusses the current STH situation and the future need for STH control in Burkina Faso , and demonstrates the feasibility of STH assessment during LF TAS at community level .
The survey was part of the monitoring and evaluation activities of the national integrated NTD program and was authorized by the Ethics Committee of the Ministry of Health of Burkina Faso . The surveys were conducted by the Ministry of Health monitoring and evaluation team . The populations were informed of the background of the surveys through the community health workers and town criers . Administrative and local authorities and community leaders were involved in the surveys . Parents were informed about the purpose and objectives of the survey through community meetings . They were also informed that they had the right to withdraw their children at any time of the survey . Verbal consent was given by parents for all children selected for the survey and recorded on survey forms and this was approved by the Ministry of Health as the literacy rate was low in rural areas in Burkina Faso . Written informed consent was obtained , before the survey started , from the head teachers of the schools as the legal guardian of all children in schools . Any children who did not want to participate were free to leave . Survey results were used for decision making for national strategy of STH control . At the beginning of the integrated national NTD program , 22 sentinel sites ( schools ) were purposefully selected in 2008 for schistosomiasis impact assessment based on prior knowledge for schistosomiasis . The 22 sites were located across 11 health regions ( Boucle du Mouhoun , Cascades , Centre-Est , Centre Nord , Centre Ouest , Centre Sud , Est , Hauts Bassins , Nord , Sahel and Sud Ouest ) with relatively even geographical distribution in the country [28] . Cross-sectional surveys in these sentinel sites for schistosomiasis were conducted in 2008 and 2013 . At the same time , STH infections were also examined in selected school age children . Within each school , 16 boys and 16 girls from each of the 7–11 age groups ( Classes 1–5 ) , totalling approximately 160 children per school , were systemically selected and examined by parasitological examinations . If there were fewer children than required sample size in a school , additional children were selected from a neighbouring school within a five kilometer radius . LF TAS have been used as a platform to assess the impact of MDA on STH in school-based surveys and to determine the treatment strategy for STH after community-wide LF MDA has been stopped [26 , 27] . An STH assessment survey during LF TAS was conducted in 2014 in the Centre Nord evaluation unit ( EU ) to test the feasibility of STH assessment during community-based LF TAS survey . The survey design for LF TAS followed the WHO TAS guidelines [29] and the concurrent STH survey followed the then-draft and now-published WHO TAS-STH survey guidelines [30] , assisted by the Survey Sample Builder tool developed by the Task Force for Global Health ( http://www . ntdsupport . org/resources/transmission-assessment-survey-sample-builder ) . The Centre Nord EU consists of four health districts ( implementation units ) . The EU , comprising a total of 1285 enumeration areas , had an estimated population of 1 . 5 million people and 60 , 120 children of 6–7 year-old in 2014 . The primary school enrolment rate was 67% . Therefore , a community-based cluster survey was conducted in accordance with the WHO guidelines [29 , 30] . In total , 42 clusters ( villages ) were selected and surveyed . For the concurrent STH survey , a subset of 336 children of 6–7 years old was sampled in the same clusters ( villages ) as TAS . This gave rise to eight children per cluster . After the LF team selected the 6–7 year-old children for LF tests , additional STH technicians identified those children in the sample who would also be assessed for STH . In order to compare any differences with older age groups and also to facilitate the comparison with the conventional school-based STH survey , an additional group of 336 children aged 10–14 years old were sampled in the survey . These 10–14 year-old children were selected from the same households as the selected 6–7 year-old children , i . e . when one 6–7 year-old child was selected , one 10–14 year-old child from the same household was also selected . If there were more than one 10–14 year-old children in the household , a random selection was used . If there were no 10–14 year-old children in the household , the missing number of 10–14 year-old children was made up by random selection from other households with 6–7 year-old children being selected for LF tests . The TAS-STH survey methodology uses the same critical cutoff values for decision-making as the standalone LF TAS [29 , 30] . The critical cutoff values indicate the maximum number of positive cases that can be found in a given EU for clarifying the EU to a certain prescribed prevalence threshold , so that corresponding treatment strategies can be applied to the whole EU . For comparison , a conventional school-based STH survey was conducted separately in the same EU according to the WHO recommendations [31] . Five primary schools were randomly selected in the EU . 50 children aged 10–14 years old from each school were selected and examined . If there were fewer than 50 children aged 10–14 years old in a school , additional children were selected from a neighbouring school within a five kilometer radius . One stool sample was collected from each of the selected children in containers which were labeled with unique identification numbers . As described elsewhere [28] , the samples were sent back to a local laboratory for examinations on the same day . The sample processing and examination methods were as described previously [32] . The Kato-Katz method was used to determine STH infections ( hookworms , A . lumbricoides and T . trichiura ) . Two slides were prepared from each sample and examined on the same day . Eggs from each of these parasites were counted and individual egg counts were calculated and expressed as eggs per gram of faeces ( epg ) . The data collected were entered into an Excel spreadsheet and double checked by biomedical technicians . The SPSS software ( IBM , version 19 ) was used for statistical analysis . When calculating the overall prevalence in the country , the samples were weighted according to the proportion of the regional population among the total population and the Complex Samples module was used taking into consideration the cluster nature of school children using region as strata and school as clusters . The 95% confidence intervals ( CIs ) for prevalence were calculated using the CI calculator ( available: http://vl . academicdirect . org/applied_statistics/binomial_distribution/ref/CIcalculator . xls ) . Arithmetic mean egg count from all subjects examined ( including both positive and negative ) was calculated . Individual egg count was categorized as light , moderate or heavy infection according to the WHO recommendations [33] . The Chi-squared test was used to compare differences in prevalence and the Kruskal-Wallis test was used to compare differences in mean egg counts . The dataset from the 2008 survey of the same 22 sentinel sites was not available for statistical comparison , therefore the STH prevalence data from the national survey report were used for descriptive comparison with the 2013 data [34] . For STH data from assessment during the LF TAS , the number of positive cases identified was used to clarify the prevalence threshold in the EU , against the pre-defined critical cut-off values in the WHO guidelines [30] . The coordinates of survey sites were collected using a handheld GPS device . Where there was an error , the location was estimated on the google map . The site location map was drawn in ArcMap version 10 ( ESRI , Redlands , CA ) .
Table 2 summarizes the 2013 survey results from the 22 sentinel sites . In total , 3 , 514 school age children ( 1 , 748 boys and 1 , 766 girls ) were examined . Overall STH prevalence in children tested was very low at 1 . 3% ( 95% CI: 1 . 0–1 . 8% ) , ranging from 0% to 6 . 9% among 22 sites ( median 0 . 6% ) . Among three major STH species , hookworm was the main species detected . A . lumbricoides and T . trichiura were only detected in four and two cases respectively , and therefore prevalence and mean egg counts of these two parasites were not calculated separately . As in Table 2 , overall hookworm prevalence was 1 . 2% ( 95% CI: 0 . 9–1 . 6% ) , ranging from 0% to 6 . 3% ( median 0 . 6% ) . All individual hookworm infections were light infections with mean egg count of 2 . 1 epg ( 95% CI: 0–4 . 2 epg ) . The prevalence at each of the 22 sentinel sites in 2013 and the prevalence at each site in 2008 from these same sites taken from the national survey report are shown in Fig 1 . It showed a general reduction in the STH prevalence in 2013 compared with the STH prevalence in 2008 across the 22 sentinel sites in the country . But statistical comparison was not possible due to the lack of availability of the 2008 dataset . There were significant differences in hookworm prevalence between regions ( Chi-square test , χ2 = 36 . 447 , P<0 . 001 ) . Centre Ouest region had the highest prevalence of 3 . 4% ( 95% CI: 1 . 9–6 . 1% ) and mean egg count of 14 . 9 epg ( 95% CI: 3 . 3–26 . 6 epg ) , followed by Hauts Bassins region of 2 . 1% ( 95% CI: 1 . 1–3 . 8% ) and 2 . 2 epg ( 95% CI: 0–4 . 5 epg ) respectively ( Table 2 ) . There was no significant difference in hookworm infection between boys , with prevalence of 1 . 0% ( 95% CI: 0 . 6–1 . 6% ) and mean egg count of 1 . 0 epg ( 95% CI: 0 . 2–1 . 8 epg ) , and girls , with prevalence of 1 . 3% ( 95% CI: 0 . 9–2 . 0% ) and mean egg count of 3 . 1 epg ( 95% CI: 0–7 . 3 epg ) , ( Chi-square test for prevalence , χ2 = 0 . 695 , P>0 . 05; Kruskal Wallis test for mean egg count , H = 0 . 409 , P>0 . 05 ) . Table 3 summarizes the results of STH assessment during the LF TAS in 2014 ( LF results are not presented in this paper ) . In total , 351 children ( 184 boys and 167 girls ) aged 6–7 years old were examined for STH infection during the TAS in the EU . Two cases of STH infection were identified with estimated prevalence of 0 . 6% ( 95% CI: 0 . 2–2 . 1% ) in children tested . Similarly , 345 children ( 164 boys and 181 girls ) aged 10–14 years old were examined for STH infection during the TAS , and seven cases of STH infection was identified with estimated prevalence of 2 . 0% ( 95% CI: 1 . 0–4 . 1% ) in children tested . All identified positive cases were hookworm infections and no infection with A . lumbricoides or T . trichiura was found . In both age groups , there was no difference in STH infections between boys and girls ( Chi-square test , χ6−72 = 1 . 826 , χ10−142 = 0 . 264 , P>0 . 05 ) . Although there were more STH cases identified in the 10–14 year-old group , the difference was not statistically significant ( Chi-square test , χ2 = 2 . 902 , P>0 . 05 ) . Both age groups categorized the EU to be within 2% to <10% in STH prevalence according to the threshold cut-off values ( Table 3 ) . The general distribution of the positive cases in the EU is shown in Fig 2 . The positive cases were aggregated in the north part of the region , particularly in the Kongoussi health district . In the conventional school-based survey , a total of 250 school children aged 10–14 years ( 124 boys and 126 girls ) were examined , and no STH infection was identified . The estimated STH prevalence was 0% ( 95% CI: 0–1 . 5% ) . When compared with the two TAS survey groups , there was no significant difference with the 6–7 year-old group ( Chi-square test , χ2 = 1 . 429 , P>0 . 05 ) , but there was a significant difference with the 10–14 year-old group ( Chi-square test , χ2 = 5 . 133 , P<0 . 05 ) .
Survey of the 22 sentinel sites across 11 regions showed that STH infections in school age children in Burkina Faso were at a low level in the majority of the country’s endemic districts . The residual infections were mainly hookworm infections , while Ascaris and Trichuris infections were very rarely seen . No moderately or heavily infected cases were found during the survey . This is in line with the TAS-STH results in the Centre Nord region . It is suggested that over a decade of large-scale preventive chemotherapy targeting different age groups through various program platforms implemented in Burkina Faso has effectively controlled STH in the country . Burkina Faso was one of the first countries in sub-Saharan Africa to start national NTD programs with large-scale preventive chemotherapy with external financial and technical support . The national schistosomiasis and STH program established in 2004 was the first national STH program in the country . The results from the schistosomiasis baseline survey at the time showed that STH , particularly hookworm , was endemic in Burkina Faso , but with a relatively low prevalence [21] . Regardless of the low prevalence , the national program decided to conduct the large scale MDA intervention together with the schistosomiasis MDA . This decision was based on: 1 ) Burkina Faso was and still is among the poorest countries in the world according to the Human Development Index [35]; 2 ) there was high prevalence of anemia in the country [36 , 37] , and hookworm infection is a risk factor for anemia in women and children [3 , 37–39]; 3 ) adding albendazole to praziquantel distribution to treat school age children does not incur extra cost for drug delivery; and 4 ) deworming is among the most cost-effective investments in global health and benefits of deworming were demonstrated [39–41] . With all deworming activities through community-wide LF MDA , school-based and community-based schistosomiasis and STH MDA , and Child Health Days , the program rapidly achieved national coverage with all endemic health districts targeted , with good treatment coverage in pre-school age children and school age children . Burkina Faso successfully achieved and maintained the target of at least 75% of national coverage as recommended by WHO [11 , 15] . The current results of sentinel site survey suggest that STH have been successfully controlled as a public health problem in school age children in Burkina Faso as STH infection of moderate or high intensity from the surveys was below the threshold of 1% as defined by WHO [11] . Integrated , community-wide MDA programs for schistosomiasis and STH can be highly cost effective , even in communities with low disease burden in any helminth group [42] . Multi rounds of large scale community-wide LF MDA may have helped to achieve successful control of STH in Burkina Faso . However , lack of baseline data prior to the commencement of LF MDA for statistical comparison makes it difficult to attribute conclusively the low STH prevalence to the impact of such MDA activities . Despite the achievements in Burkina Faso as shown by the data , it is however noted that there are still some hot spot infections as there were three sentinel sites showing hookworm prevalence being 6 . 9% , 5 . 6% and 3 . 8% respectively from the survey , particularly with one site in Centre Ouest showing increased prevalence from 2008 ( Fig 1 ) . Single dose of albendazole has 87 . 8% cure rate and >90% fecal egg count reduction rate for hookworm [43] , and annual community-wide mass treatment such as LF MDA is expected to reduce hookworm infection to ground level within a few years [44] . The fact that such hot spots for hookworm infection still existed after many years of large scale MDA intervention suggests that some focal factors may have affected the impact of the treatment . While research is needed on possible local factors that contributed to the persistence or increase of hookworm infection in these locations , potential reasons may include: 1 ) poor focal coverage–the overall national treatment coverage may have been high , but treatment coverage at some communities may not have been satisfactory , and 2 ) there may be some particular local factors , such as lack of clean water , poor hygiene and sanitation [7] . The national program needs to pay special attention to such hot spot communities , i . e . providing supervision and monitoring in future MDAs to ensure the high treatment coverage . There are other limitations in this study . Firstly , the sentinel sites were selected for schistosomiasis impact assessment according to the endemicity of schistosomiasis . Therefore the results from these sites may not represent the true STH situation across all the communities in the country . Secondly , the survey was conducted among school age children attending schools . Given the low school enrolment rate in Burkina Faso , a third of school age children were in communities who were not subject to sampling and who may be more disadvantaged and prone to STH infection . The current results from the community-based STH assessment during LF TAS versus conventional school-based survey may reflect this . On the other hand , hookworm is the main STH species in Burkina Faso and adult population harbors significant worm load [44–46] . The survey results in school age children may represent better the Ascaris and Trichuris situation , but may not represent the hookworm situation in the whole communities . Taken together , it is advisable that the national program should consider community-based assessment and include survey in adult population to confirm the current STH , particularly hookworm situation in the country . LF MDA with albendazole and ivermectin is the largest community-wide deworming activity in Burkina Faso , and this has been stopped in more and more districts and is expected to stop in all districts soon due to meeting stopping LF MDA criteria . To consolidate the impact already achieved and to avoid recrudescence , the national program should continue to implement STH treatment through 1 ) LF MDA where LF MDA has not been stopped; 2 ) adding albendazole to schistosomiasis MDA where LF MDA has been stopped; and 3 ) Child Health Days . Given the low level of STH infection from the current results , the country may be in a good position to pursue interruption of STH transmission [44] . Research is needed on what strategies are required to sustain the gain and to interrupt STH transmission in such settings as in Burkina Faso , particularly after community-wide LF MDA has stopped . Where hookworm is the dominant STH species , mass treatment of all age groups is recommended [44] . Bearing in mind that chemotherapy alone may not be enough to interrupt STH transmission [47] , other strategies would be needed , i . e . provision of clean water , hygiene and sanitation . Funding should be sourced to conduct such research for cost-effective strategies in Burkina Faso to interrupt STH transmission . LF TAS provides a perfect timing and platform to assess the STH situation to determine the STH-specific deworming strategy after LF MDA is stopped . In the current study , a community-based cluster survey was tested in one EU ( Centre Nord region ) , and we assessed the feasibility of integrating STH assessment with community-based cluster survey for LF TAS . The process of the survey suggests that the joint assessment in community-based surveys is indeed feasible and efficient , testing either 6–7 years old ( same as LF TAS target ) or 10–14 years old . Sampling 10–14 year-olds detected more STH cases , but the result was not significantly different from that of sampling 6–7 years old . The results from both groups classified the EU as between 2% to <10% STH prevalence . The TAS design for STH assessment assumed a design effect of 2 . 0 for the cluster sampling . In our survey , the actual design effect was 0 . 943 for 6–7 years old group and 1 . 165 for 10–14 years old group . Therefore the survey was sufficiently powered for estimating the STH situation . Comparing with the conventional school-based surveys , community-based testing of 10–14 years old using the LF TAS platform did show significantly higher STH prevalence . It is noted that the separate conventional school-based survey had to be conducted shortly after MDA that had not allowed sufficient time for re-infection . Therefore , the results of the conventional school-based survey may represent an underestimate of the true prevalence in school age children and account for the difference with the TAS-STH results . However , considering the results from two sentinel sites in the same region surveyed in 2013 with larger sample size per school that did not detect any STH infection either ( Fig 1 ) , the results from the conventional school-based survey may have been a true representation of prevalence using such a survey methodology . This suggests that community-based TAS-STH may give a better estimate of STH situation in low prevalence areas with low school enrollment rate such as in Burkina Faso . In particular , the TAS-STH results provide a clear indication of geographical aggregation of clusters with STH infections . This provides national program managers a powerful tool for program decision . Although conventional school-based survey is easier to organize , LF TAS-based assessment provides better estimate of STH situation and is integrated with the LF TAS , therefore the national program will continue to conduct such assessments in other regions . In conclusion , through large-scale preventive chemotherapy Burkina Faso have successfully controlled STH in school age children in the country . Research is needed on potential reasons and factors that hookworm infection persists in some locations after many rounds of MDA and for future strategies to consolidate the gains made from interventions to date and to target interruption of STH transmission in Burkina Faso . LF TAS provides one feasible and efficient platform to assess the STH situation for post LF MDA decision making and should be further examined and implemented as a monitoring and evaluation tool .
|
Burkina Faso is a West African country endemic with soil-transmitted helminth infections ( STH ) . Mass treatment with albendazole has been implemented for over a decade in the country through annual mass drug administration ( MDA ) for lymphatic filariasis for population aged five years and over , biennial treatment of school age children with albendazole together with schistosomiasis MDA , and biannual treatment of pre-school age children through Child Health Days . A sentinel site survey in 2013 showed that STH infection in Burkina Faso was very low at 1 . 3% . Hookworm was the main species detected , but infection was low and light . An assessment was also conducted with two age groups in Centre Nord region through lymphatic filariasis transmission assessment survey in 2014 . The results confirmed the low level of infection with soil-transmitted helminths in the region . Through large-scale preventive chemotherapy , Burkina Faso has effectively controlled STH in school age children in the country . The future STH deworming strategy may focus on consolidating the gain and interrupting the STH transmission in the country .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"children",
"medicine",
"and",
"health",
"sciences",
"education",
"cancer",
"treatment",
"sociology",
"geographical",
"locations",
"tropical",
"diseases",
"social",
"sciences",
"clinical",
"oncology",
"parasitic",
"diseases",
"oncology",
"age",
"groups",
"clinical",
"medicine",
"pharmaceutics",
"neglected",
"tropical",
"diseases",
"africa",
"families",
"schools",
"people",
"and",
"places",
"helminth",
"infections",
"schistosomiasis",
"chemotherapy",
"population",
"groupings",
"drug",
"therapy",
"soil-transmitted",
"helminthiases",
"burkina",
"faso"
] |
2016
|
Successful Control of Soil-Transmitted Helminthiasis in School Age Children in Burkina Faso and an Example of Community-Based Assessment via Lymphatic Filariasis Transmission Assessment Survey
|
Although extensive data exist on avian influenza in wild birds in North America , limited information is available from elsewhere , including Europe . Here , molecular diagnostic tools were employed for high-throughput surveillance of migratory birds , as an alternative to classical labor-intensive methods of virus isolation in eggs . This study included 36 , 809 samples from 323 bird species belonging to 18 orders , of which only 25 species of three orders were positive for influenza A virus . Information on species , locations , and timing is provided for all samples tested . Seven previously unknown host species for avian influenza virus were identified: barnacle goose , bean goose , brent goose , pink-footed goose , bewick's swan , common gull , and guillemot . Dabbling ducks were more frequently infected than other ducks and Anseriformes; this distinction was probably related to bird behavior rather than population sizes . Waders did not appear to play a role in the epidemiology of avian influenza in Europe , in contrast to the Americas . The high virus prevalence in ducks in Europe in spring as compared with North America could explain the differences in virus–host ecology between these continents . Most influenza A virus subtypes were detected in ducks , but H13 and H16 subtypes were detected primarily in gulls . Viruses of subtype H6 were more promiscuous in host range than other subtypes . Temporal and spatial variation in influenza virus prevalence in wild birds was observed , with influenza A virus prevalence varying by sampling location; this is probably related to migration patterns from northeast to southwest and a higher prevalence farther north along the flyways . We discuss the ecology and epidemiology of avian influenza A virus in wild birds in relation to host ecology and compare our results with published studies . These data are useful for designing new surveillance programs and are particularly relevant due to increased interest in avian influenza in wild birds .
Birds of wetlands and aquatic environments such as the Anseriformes ( particularly ducks , geese , and swans ) and Charadriiformes ( particularly gulls , terns , and shorebirds ) are thought to constitute the major natural reservoir for avian influenza A virus [1 , 2] . Influenza A viruses of all hemagglutinin ( HA ) and neuraminidase ( NA ) subtypes ( H1–H16 and N1–N9 ) and most HA/NA combinations have been identified in the wild bird reservoir [3 , 4] . Anseriformes and Charadriiformes are distributed globally , except for the most arid regions of the world , and represent an almost global coverage of influenza A virus host species [1 , 2] . In birds , influenza viruses preferentially infect cells lining the intestinal tract and are excreted in high concentrations in their feces . Transmission is thought to be achieved primarily via the fecal–oral route [1] , which likely represents an efficient way to transmit viruses between waterfowl , by shedding the virus via feces into the surface water [1] . The prevalence of avian influenza A viruses in their natural hosts depends on geographical location , seasonality , and species . For instance , the prevalence of avian influenza A viruses in ducks in North America varies from less than 1% during spring migration to 30% prior to and during fall migration , but large variations in virus prevalence have been observed in different surveillance studies [1 , 4 , 5] . The peak in prevalence during fall migration is believed to be related to the large number of young immunologically naïve birds of that breeding season [1 , 2 , 6 , 7] . Although extensive data exist on surveillance studies of influenza A viruses in ducks and shorebirds in North America [4 , 5] , limited up-to-date information is available for Eurasia , Africa , South America , and Oceania , and only for limited numbers of species [8–11] . Because of the apparent species-specific niches of certain HA subtypes such as H13 and the recently discovered H16 [3 , 12–14] , as yet unidentified influenza A viruses may exist in nature . Information about influenza A viruses in Eurasia and North America is of particular interest because the influenza A viruses found in Eurasian wild birds are genetically distinct from those of wild birds in the Americas [1 , 2] . The direct zoonotic potential of several Eurasian lineage avian influenza A viruses is currently the cause of serious concern [15–18] . The increasing problems with outbreaks of highly pathogenic avian influenza ( HPAI ) , the potential spread of HPAI H5N1 by wild birds over large geographic areas , and the threat certain avian influenza A viruses pose to public and animal health emphasize the need for more information on the ecology of avian influenza A viruses circulating in the wild bird reservoir . Our current knowledge of the epidemiology of avian influenza A viruses , virus ecology in relation to host ecology , the temporal and spatial patterns of avian influenza A viruses in their natural hosts , the role of potential new hosts in the influenza A virus ecology , and the interaction between wild birds and poultry are still very limited . Traditionally , influenza A virus surveillance studies in wild birds have been performed by direct virus isolation from fecal samples or cloacal swabs in embryonated hen's eggs [19] . This method is labor intensive due to the handling time of each of the individual samples , and is quite sensitive to laboratory contaminations , in particular if blind passage is used routinely during virus isolation attempts . Currently this diagnostic method is being replaced in many laboratories by molecular diagnostic tests , such as conventional or real-time RT-PCR methods targeting highly conserved gene segments of the influenza A virus . Such molecular methods allow the rapid identification of influenza A virus positive specimens from large collections of samples , which can then be used for targeted virus isolation attempts [20–22] . In this study , we present data on the prevalence of influenza A viruses from our ongoing wild bird surveillance studies . From 1998 to 2006 , we screened more than 36 , 000 samples collected from 323 bird species using molecular diagnostics . The sample collection includes many bird species reported to be permissive to avian influenza A virus in earlier influenza A virus surveillance studies [1 , 11] . To obtain more detailed information on potential host species , large numbers of samples of birds from different bird families and geographical locations were included . This was done in part because earlier studies relied solely on virus isolation in embryonated hen's eggs as a screening method for investigating whether molecular detection methods would yield different results . Of all samples , 90% were from The Netherlands and Sweden , 4 . 5% from elsewhere in Northern Europe ( seven countries , multiple sites ) , and 5 . 5% from other parts of the world , including Africa ( Nigeria , Ghana ) , North America ( United States , Canada ) , South America ( Argentina ) , Asia ( Kazakhstan , South Korea ) , the Arctic ( Norway , Iceland ) , and the Antarctic Peninsula . All samples were taken from healthy birds . We used this data to describe temporal and spatial patterns in influenza A virus prevalence in different wild migratory bird species .
From 1998 to 2006 we sampled 36 , 809 birds belonging to 323 species of 18 orders ( Table 1 ) . Lists of species , sample numbers , and locations are included in Tables S1 and S2 . All influenza A virus positive bird species were obtained in Northern Europe , unless mentioned otherwise . Of the 992 RT-PCR positive samples , 332 virus isolates were recovered , yielding an overall recovery rate of 33 . 5% . The majority of the samples from which we were unable to isolate virus had threshold cycle values above 35 , which indicates a low viral load . In addition , a small subset of the samples was initially stored at −20 °C , which may have a negative effect on the virus isolation rate . All influenza A virus isolates were obtained from birds belonging to the orders of Anseriformes and Charadriiformes migrating along the East Atlantic flyway [1 , 2] . The prevalence of influenza A virus in the different duck , goose , and swan species is presented in Table 2 . The prevalence in dabbling ducks was 6 . 1% . Mallards and teals had a higher prevalence of the virus than wigeons , pintails , gadwalls , and shovelers combined ( 7 . 2% versus 3 . 0% , Pearson X2-test , p < 0 . 001 ) . The sampled dabbling ducks all migrate along the East Atlantic flyway and were sampled during fall migration ( Sweden ) or upon arrival and stay at their wintering grounds ( The Netherlands ) ( Table S1 ) . Figure 1 shows the ring recovery for mallards ringed at Ottenby Bird Observatory ( Öland , Sweden ) in 2002 and 2003 , and mallards , Eurasian wigeons , and common teals ringed in The Netherlands from 1998 to 2005 . Influenza A viruses were occasionally detected in common eiders , common shelducks , and tufted ducks , which belong to the guilds of stifftails , shelducks , and pochards , respectively . Influenza A viruses were not detected in 20 other ducks belonging to eight additional species . Influenza A viruses of subtypes H1–H13 were obtained from mallards; H1 , H4 , H6 , and H9 from Eurasian wigeons; H1 , H3 , H6 , and H8 from common teals; H9 from gadwalls; H2 from northern pintails; and H11 from northern shovelers ( Figure 2 ) . The HA subtype distribution in mallards was different from that in all other ducks ( Pearson χ2-test , p < 0 . 001 ) . Note , however , that relatively few virus isolates were obtained from other duck species ( n = 26 ) . Samples were obtained from eight goose and three swan species ( Tables 2 and S2 ) . Influenza A viruses were detected in white-fronted , barnacle , greylag , brent , bean , and pink-footed geese , as well as bewick's and mute swans . HA subtypes detected in geese and swans were H1 ( 9 . 5% ) , H2 ( 4 . 8% ) , H6 ( 81% ) , and H9 ( 4 . 8% ) . Thus , the vast majority of influenza A virus isolates obtained from geese and swans were of the H6 subtype: H6N1 ( 18% ) , H6N2 ( 35% ) , and H6N8 ( 47% ) . Within the Laridae family , a total of 4 , 099 samples were obtained from nine gull and two tern species that originated predominantly from Northern Europe ( Tables 1 , S1 , and S2 ) . Influenza A viruses were detected in black-headed , common , herring , and greater black-backed gulls ( Table 2 ) , but not in five other gull species and two tern species ( Table S2 ) . Virus prevalence varied greatly with respect to colonies sampled , timing , and geography . Prevalence of 60% was detected in juvenile black-headed gulls during fall migration in Öland , Sweden [3] , while the overall prevalence was only 0 . 8% ( Table 2 ) . Influenza A virus was undetectable in many different colonies during breeding season over multiple years in The Netherlands and Sweden . Positive samples were predominantly obtained in June , July , and August . Influenza A virus subtypes isolated from gulls were H6N8 ( 10% ) , H13N6 ( 10% ) , H13N8 ( 40% ) , and H16N3 ( 40% ) . We obtained 3 , 159 samples from 47 wader species in a variety of sampling sites in Europe , 60% of which were taken during fall migration , 35% during spring migration , and 5% at the breeding grounds . We obtained one positive sample from a red knot out of 230 birds caught at Delaware Bay , United States , in early May 2005 and one from a red-necked stint out of five sampled in South Korea ( Table 2 ) . All other waders were negative for influenza A virus . Within the Alcidae family , we obtained 907 samples from four bird species . Three influenza A virus positive samples were obtained from 817 guillemots; all were H6N2 viruses [23] . An influenza A virus was detected in one out of 237 common coots sampled . More than 10 , 000 samples were collected from wild birds in 15 orders other than the Anseriformes , Charadriiformes , and Gruiformes , but no influenza A viruses were detected in those samples ( Table S2 ) . For mallards , we compared the prevalence of influenza A virus from 1999 to 2005 in The Netherlands . The sample size was approximately evenly distributed from 1999 to 2004 , but increased in 2005 in response to the HPAI H5N1 threat . The peak prevalence varied from 0 . 93% in September 2002 to 20 . 76% in October 2001 . The peak prevalence for most years was in September and October , with the exception of 2000 ( January; see Figure 3 ) . Similar fluctuations in peak prevalence were observed in Eurasian wigeons ( 0 . 83% in December 2002; 20% in September 2005 ) and common teals ( 4% in November 2000; 30% in November 2005 ) ( unpublished data ) . Although sample size may vary somewhat between years and peak prevalence may vary considerably , we calculated a generalized trend line for influenza A virus in mallards in The Netherlands and Sweden . The winter and summer distribution of these mallard populations is shown in Figure 1 . Virus prevalence in mallards in Sweden was ∼3-fold higher as compared with The Netherlands ( Figure 4 ) . For both countries , influenza A virus prevalence was already high upon arrival on the sampling sites in August , only to drop after November . The prevalence of influenza A viruses with respect to age and sex was determined for mallards and Eurasian wigeons . The prevalence was different between juveniles ( year 1 ) and adults ( consecutive years ) . Influenza A virus prevalence was 6 . 8% for juvenile ducks ( n = 2038 ) and 2 . 8% for adults ( n = 895 ) . Juvenile ducks thus had a greater chance to be influenza A virus positive than adults ( RR: 2 . 24 , 95% CI: 1 . 61 to 3 . 71 ) . No apparent differences in prevalence were observed between male ( n = 4737 ) and female ( n = 3114 ) ducks ( RR: 1 . 13 , 95% CI: 0 . 944 to 1 . 35 ) . H6 ( 17 . 8% ) and H4 ( 16% ) were the most abundantly detected HA subtypes , followed by H7 ( 11 . 1% ) , H3 ( 9 . 6% ) , H11 ( 8 . 7% ) , H1 ( 8 . 1% ) , H2 ( 7 . 8% ) , H5 ( 7 . 5% ) , H10 ( 4 . 8% ) , H12 ( 2 . 1% ) , H8 ( 1 . 8% ) , H13 ( 1 . 8% ) , H9 ( 1 . 5% ) , and H16 ( 1 . 2% ) . H14 and H15 were never detected . Viruses of the H13 and H16 subtypes were primarily obtained from Charadriiformes ( Figure 2 ) . Viruses of subtype H6 were obtained relatively frequently from wild birds other than mallards . All H5 and H7 isolates were low pathogenic avian influenza viruses [24] . The most frequently detected NA subtype was N2 ( 19 . 9% ) , followed by N6 ( 17 . 8% ) , N8 ( 14 . 8% ) , N7 ( 13% ) , N9 ( 10 . 8% ) , N3 ( 10 . 2% ) , N1 ( 8 . 7% ) , N5 ( 2 . 7% ) , and N4 ( 2 . 1% ) ( Table 3 ) . Subtypes N5 and N7 were only found in viruses isolated from mallards . From Charadriiformes we only obtained viruses of the N3 , N6 , and N8 subtypes and from geese and swans only of the N1 , N2 , and N8 subtypes ( Figure 2 ) . In total , 55 different HA/NA subtype combinations were detected ( Table 3 ) . The most frequently detected subtype combination was H4N6 , comprising 13 . 6% of all isolated influenza A viruses , followed by H7N7 ( 10 . 5% ) and H6N2 ( 9 . 9% ) . Viruses containing H8 matched only with N4 and viruses containing H16 only with N3 ( Table 3 ) .
Recent improvements in molecular diagnostic tests have facilitated high-throughput screening of wild birds for influenza A virus . Despite the introduction of the molecular tests and the wide range of bird species tested , only a few “new” influenza A virus hosts were identified: barnacle , bean , brent , and pink-footed goose , bewick's swan , common gull , and guillemot . It is thus reassuring that the use of classical methods for virus detection in numerous surveillance studies has not resulted in an apparent biased detection toward viruses that can be isolated easily in embryonated hens' eggs; therefore , it remains a viable approach for virus detection . However , in our hands , virus detection by RT-PCR was more sensitive than using classical tests , since viruses were isolated from only one-third of the RT-PCR positive samples . Even under ideal conditions of transport , storage , and processing , not all RT-PCR positive samples yielded virus isolates . We confirmed the high virus prevalence of dabbling ducks in fall as observed in previous studies in the Northern Hemisphere [1] . Our data indicate that timing relative to migration , instead of the absolute time point , is a determinant of virus prevalence . High virus prevalence early in fall migration likely declines gradually as the migration proceeds , thus forming a north–south gradient of virus prevalence even within single species . This explains prevalence differences in earlier surveillance studies [4 , 5] . Influenza A virus prevalence was generally higher in juvenile ducks as compared with adults , as reported for North America [5–7] . The estimated yearly turnover of mallards in Northern Europe is roughly one-third; 56% of the juvenile mallards die during their first year and the mortality in adult birds is ∼40% [24] . Thus , one-third of the mallard population consists of juvenile birds , which are immunologically naïve and therefore probably more susceptible to influenza A virus[1 , 6] . The influenza A virus prevalence in mallards was comparable to that of other dabbling ducks . Viruses were also detected in ducks belonging to other guilds , but the prevalence was lower . Influenza A virus was detected in 811 of 13 , 297 dabbling ducks , but in only six of 440 other ducks ( Pearson χ2-test , p = < 0 . 001 ) . Analysis of 7 , 130 samples from 11 goose and swan species revealed that virus prevalence was also low , ranging from 0 . 7% to 2 . 4% . Several factors could contribute to the high virus prevalence in dabbling ducks as compared with other species . The dabbling behavior itself is likely an important factor; virus excreted in surface waters via feces may efficiently transmit viruses to other ducks that feed on the same waters . Influenza A virus can remain infectious for prolonged periods in surface water depending on temperature , salinity , and pH [25] . The prolonged presence of influenza A viruses in surface water may enable the spread of viruses in different host sub-populations that otherwise would be separated in time and space . In contrast , diving ducks forage deeper under the surface and more often in marine habitats , and most goose and swan species graze in pastures and agricultural fields . Such differences in feeding behavior could lead to less-efficient virus transmission and thus account for the differences in prevalence . Population size and age structure could be additional important factors enabling the annual co-circulation of multiple virus subtypes within the same ( meta- ) populations [26] . The dabbling duck populations are estimated at between 5 , 000 , 000 and 10 , 000 , 000 in Northern Europe alone [27] . Mallards are the most abundant species , followed by Eurasian wigeon and common teal [27] . We observed the highest virus prevalence in two of these species: mallards and common teals . The population estimates for goose species in Northern Europe are significantly lower as compared with the dabbling ducks [27] . The smaller population sizes could limit in general the potential of perpetuation of influenza A virus in these species and in particular the continuous co-circulation of multiple virus subtypes . When we plotted the influenza A virus prevalence in duck , goose , and gull species as a function of population size ( Figure 5 ) , population size did not appear to be the main correlate of virus prevalence ( R2 = 0 . 0001 ) . The relative clustering of the data points from the duck , goose , and gull species ( Figure 5 ) suggests that other factors ( taxonomy , behavior , etc . ) could determine virus prevalence . The population size of 2 , 000 , 000 black-headed gulls in Northern Europe seems to be sufficiently large for the continuous circulation of influenza A viruses . Behavioral factors influencing influenza A virus ecology in gulls could include colony breeding , gregariousness during migration and wintering , feeding patterns , and the mixing of different populations of birds . From our data , it appears that virus prevalence in gulls peaks shortly after they have left their breeding grounds . Surveillance studies performed along the East Coast of North America suggested a distinct role for wader species in the perpetuation and maintenance of certain influenza A virus subtypes [4 , 12] . It was suggested that different families of wetland birds are involved in perpetuating influenza viruses and that waders may carry the virus north to the duck breeding grounds in spring . We detected influenza A virus in one shorebird sample obtained from Delaware Bay ( United States ) and one from South Korea . The differences in prevalence between our American wader data and those described by others [4 , 12] could be due to sampled species , sampling procedures , and timing . Within our European surveillance study , not a single influenza A virus was detected in waders . Although the majority of our wader samples were collected during fall migration , a reasonable sample size was collected during spring migration . Thus , there is no evidence that waders play a role in the perpetuation of influenza A virus in Europe . The recently intensified surveillance in waders , including serological data collection , may allow a definitive conclusion about their role in the influenza virus ecology in Europe . Although historically influenza A viruses have been obtained from more than 105 species of 26 different families [2] , we did not detect significant influenza virus prevalence in species other than those belonging to the orders Anseriformes and Charadriiformes . We therefore suggest that although multiple bird species can be infected , their contribution to the overall virus ecology could be limited; infections of these hosts , although potentially with high peak prevalence , may be only transient . Influenza A virus subtypes H1–H12 were isolated frequently from mallards , and several of these subtypes were also detected in Eurasian wigeons , common teals , gadwalls , and northern shovelers . The absence of subtypes H14 and H15 in our collection was probably due to the geographical separation of virus hosts [28 , 29] . Because all HA subtypes isolated from Anseriformes were also isolated from mallards , it is likely that mallards play a pivotal role in the perpetuation of influenza A virus subtypes H1–H12 in Europe . Subtypes H5 and H7 were rarely detected in longitudinal studies of ducks in Canada , whereas in this European study , they represent 7 . 5% and 11 . 1% of viruses obtained from ducks . The most common virus subtypes in Europe—H3 , H4 , H6—were also common subtypes in Canada . The predominant isolation of H13 and H16 viruses from gull species confirms the common notion that these viruses belong to the influenza A virus “gull lineage . ” H13 and H16 viruses are genetically distinct from viruses from other hosts and seem to have adapted to replication in gull hosts in particular [3 , 12] . Interestingly , viruses of the H6 subtype seem to have a broader host range compared with that of other virus subtypes in our study . Of the influenza viruses obtained from birds other than dabbling ducks and gulls , 79% were H6 viruses . H6 viruses were isolated from gulls , auks , swans , and geese . The relative abundance of this subtype in ducks does not explain the large variety of species from which these viruses were isolated; H4 and H7 viruses were also detected frequently in ducks , but rarely in other birds . H6 viruses have been transmitted from wild birds to poultry on several occasions [30 , 31] , providing further evidence for the ability of these viruses to be transmitted between different bird species . HPAI H5N1 viruses have caused large-scale outbreaks in poultry in Southeast Asia since 1997 and have also been transmitted to a variety of mammalian species , including humans [17 , 18 , 32] . Until 2005 , wild migratory birds probably did not play a significant role in the epidemiology and spread of HPAI H5N1 , although the virus was detected sporadically in wild birds . A large-scale outbreak in wild migratory birds occurred in April–June 2005 at Lake Qinghai in China [33–35] , after which the HPAI H5N1 virus rapidly spread westward across Asia , Europe , the Middle East , and Africa . Since then , affected wild birds have been reported in several countries [36] , but even in areas with significant outbreaks in poultry , the virus prevalence in wild birds is low and their role in spreading the disease is unclear . It is likely that the influenza A surveillance studies in wild birds such as those presented here could provide “early warning” signals for the introduction of HPAI H5N1 into new regions [37] . The current increased interest in influenza virus surveillance in wild and domestic birds provides a unique opportunity to increase our understanding not only of HPAI epidemiology but also of the ecology of low pathogenic avian influenza viruses in their natural hosts .
Birds were trapped by expert ornithologists using duck decoys , duck traps , wader funnel traps , mist nets , clap nets , cannon nets , or Helgoland traps . Cloacal swabs were collected using sterile cotton swabs of two different sizes depending on the size of the birds . The cloacal swabs were stored in transport medium ( Hank's balanced salt solution containing 0 . 5% lactalbumin , 10% glycerol , 200 U/ml penicillin , 200 μg/ml strepromycin , 100 U/ml polymyxin B sulfate , 250 μg/ml gentamycin , and 50 U/ml nystatin [ICN Pharmaceuticals , http://www . icnpharm . com] ) and shipped to the laboratory where they were stored at −80 °C upon analysis . Before shipment , the samples were stored at 4 °C for less than a week , at −80 °C if such freezers were available nearby the sampling site , and at −20 °C only if rapid transport or storage at −80 °C was practically impossible . Samples were obtained from 323 different bird species belonging to 18 different orders and a wide variety of sampling locations . The majority of samples were obtained consistently from the same sites in The Netherlands ( Krimpen a/d Lek , 51°54′N 4°41′E; and Lekkerkerk , 51°54′N 4°38′E ) and Sweden ( Ottenby Bird Observatory , Öland , 56°12′N 16°24′E ) . Samples were also collected during short-term sampling expeditions at different sites in Europe , Asia , Africa , North America , South America , Antarctic , and the Arctic . In 2005 , numerous sampling sites were added in response to potential HPAI H5N1 threats . RNA isolation and RT-PCR were performed as described previously for samples obtained until 2002 [22] . From 2003 onward , RNA was isolated using a MagnaPure LC system with the MagnaPure LC Total nucleic acid isolation kit ( Roche Diagnostics , http://www . roche-diagnostics . nl ) and influenza A virus was detected using a real-time RT-PCR assay targeting the matrix gene [21] . To ensure efficient influenza A virus detection , the published probe sequence was changed to 6-FAM-TTT-GTG-TTC-ACG-CTC-ACC-GTG-CC-TAMRA-3′ , based on the avian influenza A virus sequences available from public databases . Amplification and detection was performed on an ABI7700 with the TaqMan EZ RT-PCR Core Reagents kit ( Applied Biosystems , http://www . appliedbiosystems . com ) using 20 μl of eluate in an end volume of 50 μl . Pools of five individual samples were prepared and processed in parallel with several negative ( three negative controls on 32 samples ) and positive ( one positive control on 32 samples ) control samples in each run . Upon identification of influenza A virus positive pools , the RNA isolation and RT-PCR procedures were repeated for the individual samples within each positive pool ( again processed in parallel with three negative controls and one positive control per 32 samples ) , and individual RT-PCR positive samples were subsequently used for virus isolation . RNA isolation and real-time RT-PCR were performed by the diagnostic facility of the Erasmus MC Department of Virology . For influenza A virus RT-PCR positive samples , 200 μl of the original material was inoculated into the allantoic cavity of 11-d-old embryonated hens' eggs . The allantoic fluid was harvested 2 d after inoculation and influenza A virus was detected using hemagglutination assays with turkey erythrocytes . When no influenza A virus was detected upon the initial virus isolation attempt , the allantoic fluid was passaged once more in embryonated hens' eggs . The HA subtype of virus isolates were characterized using a hemagglutination inhibition assay with turkey erythrocytes and subtype-specific hyperimmune rabbit antisera raised against all HA subtypes [3] . The NA subtype of virus isolates was characterized by RT-PCR and sequencing . RT-PCR was performed using primers specific for the conserved non-coding regions of NA , essentially as described by others [38] . PCR products were purified from agarose gels using the Qiaquick Gel Extraction kit ( Qiagen , http://www . qiagen . com ) and sequenced . Sequencing was performed using the Big Dye terminator sequencing kit version 3 . 0 ( Amersham Pharmacia Biotech , http://www . gelifesciences . com ) and a 3100 genetic analyzer ( Applied Biosystems ) , according to the instructions of the manufacturer . Nucleotide sequences were aligned using the Clustal W program running within the BioEdit software package , version 5 . 0 . 9 . NA nucleotide sequences were analyzed with the basic local alignment search tool available from GenBank [39] to identify the NA subtype . The 95% confidence interval analysis and the Pearson χ2-test were used for analysis of the dataset used in this study .
|
Significant gaps in our knowledge of the ecology of avian influenza in wild migratory birds have become apparent during recent outbreaks of H5N1 highly pathogenic avian influenza , in particular in relation to the risk of virus spread by wild birds . An eight-year surveillance study , which included more than 36 , 000 wild birds tested for low pathogenic avian influenza , provides new information on host species , prevalence , and temporal and geographical variation of avian influenza in wild migratory birds in Europe . Dabbling ducks harbored nearly all known influenza virus subtypes , with the exception of H13 and H16 , which were found primarily in gulls . In contrast to American studies , waders did not play a role in the epidemiology of avian influenza in Europe . This study provides important information on the ecology and epidemiology of avian influenza A virus and could assist in the design of new surveillance studies for high and low pathogenic avian influenza in wild birds .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"viruses",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"birds",
"ecology",
"virology",
"microbiology",
"evolutionary",
"biology"
] |
2007
|
Spatial, Temporal, and Species Variation in Prevalence of Influenza A Viruses in Wild Migratory Birds
|
Bacteria from the genus Bartonella are emerging blood-borne bacteria , capable of causing long-lasting infection in marine and terrestrial mammals , including humans . Bartonella are generally well adapted to their main host , causing persistent infection without clinical manifestation . However , these organisms may cause severe disease in natural or accidental hosts . In humans , Bartonella species have been detected from sick patients presented with diverse disease manifestations , including cat scratch disease , trench fever , bacillary angiomatosis , endocarditis , polyarthritis , or granulomatous inflammatory disease . However , with the advances in diagnostic methods , subclinical bloodstream infection in humans has been reported , with the potential for transmission through blood transfusion been recently investigated by our group . The objective of this study was to determine the risk factors associated with Bartonella species infection in asymptomatic blood donors presented at a major blood bank in Southeastern Brazil . Five hundred blood donors were randomly enrolled and tested for Bartonella species infection by specialized blood cultured coupled with high-sensitive PCR assays . Epidemiological questionnaires were designed to cover major potential risk factors , such as age , gender , ethnicity , contact with companion animals , livestock , or wild animals , bites from insects or animal , economical status , among other factors . Based on multivariate logistic regression , bloodstream infection with B . henselae or B . clarridgeiae was associated with cat contact ( adjusted OR: 3 . 4 , 95% CI: 1 . 1–9 . 6 ) or history of tick bite ( adjusted OR: 3 . 7 , 95% CI: 1 . 3–13 . 4 ) . These risk factors should be considered during donor screening , as bacteremia by these Bartonella species may not be detected by traditional laboratory screening methods , and it may be transmitted by blood transfusion .
Bartonella species are fastidious alpha-proteobacteria with worldwide distribution . To date , at least 15 species have been associated with human infections , with at least eight species also capable of infecting dogs and cats . Several blood-sucking arthropods have been suggested or confirmed as vectors for this genus , including sandflies , body lice , fleas , ticks , and keds [1] . In humans , select species of Bartonella were confirmed as etiologic agents of cat scratch disease ( CSD ) , trench fever , bacillary angiomatosis and Oroya fever [2] . Moreover , Bartonella infections have also been documented by culture or molecular methods in human cases of endocarditis , myocarditis , polyarthritis and granulomatous inflammatory disease [1 , 2] . In countries throughout the world , most diseases associated with Bartonella species infection are not reportable; therefore , incidence data is scarce . One study performed at the end of the 20th century estimated that 22 , 000 new cases of cat scratch disease appear every year in the United States , and roughly 10% of these infections were thought to require hospitalization [3] . Bartonella species cause chronic and intermittent intra-erythrocytic bacteremia and infect endothelial cells of both incidental and natural reservoir hosts . The establishment of chronic , stealth infection is achieved by evasion of innate immune responses . These include resistance to complement activation , antigenic variation of surface proteins , and inhibition of host cell apoptosis [4] . Consequently , subclinical bloodstream infection in humans has been reported [5] , supporting the fact that these bacteria will not follow Koch’s postulates for disease causation [6] . Asymptomatic Bartonella species infection poses a hazard for blood recipients because fast , sensitive and specific diagnostic tests are not currently available for donor screening . To the authors’ knowledge , no prophylactic measures are currently in place to prevent collection and transfusion of blood and blood products contaminated with species of Bartonella worldwide . Recently , using a combination of culture methods and PCR assays , we documented Bartonella henselae and Bartonella clarridgeiae infection of blood samples from 16 asymptomatic blood donors at a large blood bank center in Southeast Brazil [5] . Risk factors for Bartonella species exposure in blood donors were initially evaluated in one study from Sweden using serology methods [7] . However , antibody detection has limited predictive value in confirming or excluding Bartonella species bacteremia in humans and animals [6 , 8–10] . Furthermore , higher vector activity is expected in tropical and sub-tropical regions of the world . Therefore , this study objective was to determine which risk factors are associated with Bartonella species infection in asymptomatic blood donors from a population in Campinas , Sao Paulo State , Brazil .
The Institutional Review Board of the State University of Campinas , Brazil approved this study under protocol number CEP 122/2005 . The voluntary blood donors presented at the Blood Bank ( HEMOCENTRO ) of the State University of Campinas ( UNICAMP ) , Brazil , and were enrolled after inform written consent was obtained . We collected blood from five hundred apparently healthy voluntary blood donors in this cross sectional study . Sample size was determined with an alpha of 0 . 05 , power of 0 . 8 , and estimated prevalence of 5% , as previously reported in a serology-based study [11] . Donors were selected through convenience sampling , with inclusion and exclusion criteria following the current international standards for blood bank donor selection [12] . Bartonella species infection from the bloodstream was detected based on enrichment blood culture in a liquid growth medium ( Bartonella alpha-Proteobacteria growth medium-BAPGM ) , coupled with isolation in solid medium and Bartonella-specific DNA amplification by PCR , followed by DNA sequencing to confirm species identification [8] . A standardized epidemiological questionnaire was delivered to each blood donor participant . The interviewer had no knowledge of the diagnostic test results , as Bartonella species testing was performed subsequent to interviews . The following information was captured: gender; self-reported ethnicity as African-American , native Indian , Caucasian , Asian or multi-racial; average monthly income represented by multiples of Brazilian monthly minimal wage; occupational animal exposure ( veterinary professionals and others with direct animal exposure such as veterinary assistants , ranchers , biologists , and volunteers at shelters ) ; contact with cats , dogs , other companion animals , livestock or wildlife ( including handling animals , bedding , waste or sharing the same environment ) ; bites from dogs , cats , and other animals; arthropod bites caused by ticks , fleas , or other insects; previous blood transfusion; previous history of blood donation; and presence of tattoos . Past history of contact with animals , animal bites , and/or arthropod bites ( defined as more than one year after contact ) were also recorded . Potential risk factors were first compared in a univariate analysis using Fisher’s exact test or the Fisher-Freeman-Halton test . All risk factors significant at the p< 0 . 25 level were entered into a stepwise logistic regression model , and variables significant to p< 0 . 05 were retained . Univariate odds ratios ( OR ) , adjusted odds ratios ( aOR ) , and 95% confidence intervals ( CI ) were calculated . Variables with collinearity were removed from the multivariate analysis . Statistical analyses were performed using JMP Pro 10 for Windows ( SAS Institute Inc . , Cary , NC ) .
Bartonella species bloodstream infection was detected in 16/500 blood donors ( 3 . 2% ) . DNA amplification and sequencing identified B . henselae in 15 blood donors ( 3% ) and B . clarridgeiae in one donor ( 0 . 2% ) , which was previously reported [5 , 13] . B . henselae bacteremia was also confirmed in six donors by bacterial isolation . The univariate and multivariate analyses of risk factors between blood donors infected with Bartonella species and uninfected subjects are provided in Tables 1 and 2 . With univariate analysis , a professional with animal exposure was seven times more likely to be infected with Bartonella species than blood donors working in all other professions . When significant variables were entered into the multivariate logistic regression model , collinearity between animal-related professions and cat contact was documented; therefore , profession was not maintained in the final model . Adjusted odds ratio indicated that subjects with cat contact , or past history of tick bite , were approximately 3 to 4 times more likely to have a Bartonella species infection than donors without cat contact or lack of history of tick bite ( Table 2 ) .
This study identified two risk factors associated with subclinical Bartonella species bloodstream infections in blood donors . Bartonella species infection was three times more likely to be diagnosed in blood donors who had contact with cats compared to blood donors with no contact . Similarly , blood donors working in animal-related professions were seven times more likely to be infected with these pathogenic bacteria , although this variable was not included in the multivariate model due to collinearity . In Swedish blood donors , contact with cats was also a risk factor for B . elizabethae seropositivity [7] . Our findings indicate that cat contact also increases the risk of subclinical bloodstream infections with B . henselae or B . clarridgeiae . Previously , 24% of 192 non-immunocompromised Americans with frequent exposure to cats , cat scratches or fleas had detectable Bartonella species DNA in blood specimens using the same diagnostic approach as used in our study [9] . Cats are natural reservoir hosts for B . henselae , B . clarridgeiae , and Bartonella koehlerae , all of which are important zoonotic species [2] . After flea-transmitted infection with most Bartonella species strains , cats rarely develop clinical manifestations but remain persistently infected with high levels of intravascular bacteria , which facilitates pathogen acquisition by blood-sucking vectors [10] . Recently , artificial feeding of Ctenocephalides felis with B . henselae- or B . clarridgeiae- infected blood demonstrated that these pathogens can persist in C . felis and be excreted in flea feces [14] . Cat nails can be contaminated with infected flea feces , where B . henselae can survive for several days [15] . Needle stick transmission of Bartonella species has also been reported [16] . Bacterial transmission is less likely to occur by cat bite , since shedding of Bartonella species in cat saliva has not been clearly documented [17] . Interestingly , a history of cat bite , cat scratches , or flea bite was not significantly associated with Bartonella species infection in our study . Possible explanations include donors avoiding cat bites and scratches during “rough play” , lack of identifying fleas as the source of an insect bite , cat confinement to the house , and routine use of parasiticides . In Southeast Brazil , the combined Bartonella prevalence reported in domestic and stray cats by five reports was 38 . 5% ( 112/291 ) , but ranged from 4 . 3% to 97% among feline populations tested [18 , 19 , 20 , 21 , 22] . Such wide variation may be associated with level of flea infestation or analytical sensitivity of PCR assays used . B . henselae and B . clarridgeiae were the only two species detected by molecular methods from cats in these studies . Furthermore , we have tested 112 cats from the same county of the blood donors ( Campinas , Brazil ) for Bartonella bacteremia , using the same diagnostic approach used in the present report ( BAPGM culture and PCR ) , where we detected bacteremia in 90% ( 101/112 ) of cats and B . henselae was confirmed by DNA sequencing and BLAST analysis from three culture isolates [23] . Therefore , similar to reports from other countries , cats may be the main reservoirs of B . henselae and B . clarridgeiae in Southeast Brazil , and pose as a risk factor for subclinical infection in humans in this region . A previous history of tick bite was also determined to be a risk factor for Bartonella species infection in this human population . While cat fleas are a well-established vector , the capability of ticks to transmit these organisms has been the subject of substantial debate . It has been demonstrated that seven Bartonella species are capable of growing in an Amblyomma americanum tick cell line [24] . Recently , the vector competency of Ixodes ricinus for transmission of Bartonella birtlesii was confirmed using a mouse model [25] . Although . Ixodes ricinus and Amblyomma americanum are not present in Brazil , other ticks such as Ixodes loricatus , Ixodes didelphidis , Amblyomma cajennense , Amblyomma aureolatum , Amblyomma ovale , and Rhipicephalus sanguineus have been isolated from marsupials and rodents living in a locality where other tick-borne organisms had been previously identified [26] . Therefore , while the mode of transmission of any Bartonella species to humans is still unclear , the potential role of tick transmission should not be ignored . Cat-scratch disease , bacillary angiomatosis and endocarditis are the most frequently reported manifestation of Bartonella infection in both immunocompetent or immunocompromised Brazilian populations [27–29] , although ocular , neurologic and dermatologic abnormalities were also reported [30] . In addition , Correa et al . reported that the risk of Bartonella DNA bloodstream infection was 45 times higher in arrhythmic patients from Brazil or Argentina when compared to controls [31] . Our results expand the current understanding about Bartonella species infection in humans in Brazil by reinforcing the ecological role of cats and ectoparasites in the transmission of Bartonella . Risk of human infection can be minimized by implementing year-round ectoparasite control in domestic animals , by avoiding cat bites and scratches , and by keeping cats indoors to minimize exposure to vectors [32] . The medical impact of Bartonella species occurrence in the blood supply is still unknown and should be carefully investigated .
|
Bacteria from the genus Bartonella are capable of causing long-lasting infection . Despite the fact that these bacteria may cause several diseases such as cat scratch disease , trench fever , and infection of cardiac valves , which can be fatal , they may also cause asymptomatic infection in humans . Several blood-sucking arthropods have been suggested or confirmed as responsible for transmitting these bacteria , including sandflies , body lice , fleas , ticks , and keds . In this study , 500 asymptomatic human blood donors from Brazil were screened for infection with species of Bartonella by blood culture coupled with molecular detection and genetic sequencing , and risk factors associated with such infection were identified . In this population , contact with cats and history of tick bite were significantly associated with human infection by Bartonella species . Since laboratory screening of donated blood for the presence of Bartonella species is not generally performed by blood banks , these risk factors should be should be considered during donor screening in order to avoid transmission of Bartonella species by blood transfusion .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"ixodes",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"geographical",
"locations",
"microbiology",
"vertebrates",
"animals",
"mammals",
"health",
"care",
"fleas",
"ticks",
"bacteria",
"bacterial",
"pathogens",
"south",
"america",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"hematology",
"insects",
"brazil",
"disease",
"vectors",
"bartonella",
"arthropoda",
"people",
"and",
"places",
"arachnida",
"blood",
"anatomy",
"cats",
"bloodstream",
"infections",
"physiology",
"biology",
"and",
"life",
"sciences",
"blood",
"donors",
"organisms"
] |
2016
|
Risk Factors for Bartonella species Infection in Blood Donors from Southeast Brazil
|
Foamy viruses ( FV ) belong to the genus Spumavirus , which forms a distinct lineage in the Retroviridae family . Although the infection in natural hosts and zoonotic transmission to humans is asymptomatic , FVs can replicate well in human cells making it an attractive gene therapy vector candidate . Here we present cryo-electron microscopy and ( cryo- ) electron tomography ultrastructural data on purified prototype FV ( PFV ) and PFV infected cells . Mature PFV particles have a distinct morphology with a capsid of constant dimension as well as a less ordered shell of density between the capsid and the membrane likely formed by the Gag N-terminal domain and the cytoplasmic part of the Env leader peptide gp18LP . The viral membrane contains trimeric Env glycoproteins partly arranged in interlocked hexagonal assemblies . In situ 3D reconstruction by subtomogram averaging of wild type Env and of a Env gp48TM- gp80SU cleavage site mutant showed a similar spike architecture as well as stabilization of the hexagonal lattice by clear connections between lower densities of neighboring trimers . Cryo-EM was employed to obtain a 9 Å resolution map of the glycoprotein in its pre-fusion state , which revealed extensive trimer interactions by the receptor binding subunit gp80SU at the top of the spike and three central helices derived from the fusion protein subunit gp48TM . The lower part of Env , presumably composed of interlaced parts of gp48TM , gp80SU and gp18LP anchors the spike at the membrane . We propose that the gp48TM density continues into three central transmembrane helices , which interact with three outer transmembrane helices derived from gp18LP . Our ultrastructural data and 9 Å resolution glycoprotein structure provide important new insights into the molecular architecture of PFV and its distinct evolutionary relationship with other members of the Retroviridae .
Spuma or foamy viruses ( FV ) are the only members of the Spumaretrovirinae subfamily of the Retroviridae . As such they share many similarities in their life cycle with the Orthoretrovirinae as well as some features with the more distant Hepadnaviridae [1] . FVs infect nearly all mammals and the best-studied member is the Prototype FV ( PFV ) previously called Human FV ( HFV ) , which has been isolated from infected human cells [2] . FV infection in humans is asymptomatic [3] but the virus can replicate very efficiently in human cell lines and is therefore an elegant model system for the study of more hazardous orthoretroviruses and a promising candidate for gene transfer therapy [4] . The main PFV structural protein is Gag ( 648 amino acids ( aa ) ) , which is encoded as a single protein and unlike orthoretroviruses does not exist as a Gag/Pol fusion variant . PFV Gag is also unusual in that it is not processed by the viral protease into canonical Matrix ( MA ) , Capsid ( CA ) and Nucleocapsid ( NC ) domains , like other retroviral Gag polyproteins . The 71 kDa PFV Gag ( pr71Gag ) precursor is only partially proteolysed at its C-terminus to yield a 68 kDa ( p68Gag ) and a 3 kDa ( p3Gag ) peptide [5 , 6] . In addition , three less systematic secondary cleavage sites are located in the middle of the Gag sequence ( residues 311 , 339 and 352 ) and have been proposed to serve capsid disassembly during virus entry [5 , 7] . The C-terminus of PFV Gag contains a Glycine–Arginine Rich ( GR ) region , which is important for interaction with the viral RNA genome and is the functional equivalent to the Cys-His motif found in orthoretroviruses [8] . PFV Gag lacks a membrane-binding domain and instead virus egress relies on a physical interaction between Gag and the Leader Peptide ( LP ) domain of the Envelope ( Env ) protein [9 , 10] . The PFV Env precursor comprises 988 residues and undergoes post-translational processing by cellular proteases resulting in three distinct domains: Surface ( SU ) gp80SU , TransMembrane ( TM ) gp48TM and LP gp18LP [11] . Mutants defective in the SU/TM cleavage produce non-infectious virions [11 , 12] while an inactive SU/LP processing inhibits budding [13] . Env forms trimers of heterotrimers ( gp80-gp48-gp18 ) anchored by two transmembrane regions ( per monomer ) in the virus membrane , where it can further assemble into hexameric lattices [14] . PFV capsid assembly commences at the centrosome or microtubule organizing center similar to type B/D orthoretrovirus assembly [15] . Capsids are then transported to the secretory pathway either the ER or the Golgi or directly to the plasma membrane where Env LP interacts with the N-terminal Gag region [9 , 10 , 16] . Budding into intracellular compartments or at the plasma membrane depends on the recruitment of the ESCRT machinery [17 , 18] , which completes budding by membrane scission [19] . Intracellular virions are most likely transported in vesicles for release at the plasma membrane ( reviewed in [1] ) . In order to obtain structural insight into PFV organization at medium resolution we used cryo-electron tomography ( cryo-ET ) and microscopy ( cryo-EM ) to analyze isolated wild-type particles and variants with mutations in PFV Gag and Env proteins . These data are correlated to Gag assemblies analyzed in infected cells by employing high pressure freezing , cryo-substitution and electron tomography . We further present a 9 Å resolution structure of the glycoprotein in situ , which shows unprecedented molecular details of its membrane-anchored organization and higher order assemblies stabilized most likely by the leader peptide gp18LP .
In order to probe the various structures of PFV found in vivo and correlate the results with the studies on purified particles analyzed by cryo-ET ( see below ) , we prepared samples of HT1080 cells infected with replication competent wt PFV by high pressure freezing and cryo-substitution 24 to 48h post infection ( see Material and Methods ) . Viruses with clear capsid at their center were observed budding from the plasma membrane ( Fig 1A and 1C ) . Viruses were often found as well into large vacuolar compartments which can either be related to plasma membrane budding or constitute an alternative budding site ( Fig 1B and 1D ) . Naked capsids were also observed in the cytoplasm ( Fig 1B ) . All these observations agree well with previous results obtained in [15] . Electron tomography of 100 to 300 nm thick sections of viruses budding from the plasma membrane ( Fig 1E ) show that the virions are composed of the capsid spaced from the viral membrane by an additional fainter intermediate shell of density ( Fig 1E , arrowheads ) . In addition to the previous observations , we found on several occasions cells with an unusually high concentration of round objects in the cytoplasm ( S1 Fig ) , which are rather regular in size ( r = 28 nm ) similar to the capsid size of released virions and the one observed at the plasma membrane ( S1 Fig ) . They are also often aligned in membrane delimited tubes ( S1 Fig and S1 Movie ) . Because , we did never observe such assemblies in non-infected HT1080 cells , we speculate that they constitute assembled capsids , which accumulate in tubular membrane compartments . Three different PFV viruses , wild type virus ( wt ) , a Gag mutant impaired in RNA binding ( iNAB ) and an Env mutant ( iFuse ) were purified from the supernatant of cells expressing different combinations of PFV proteins from a replication-deficient PFV vector system . In the iNAB mutant , 23 arginines in the glycine/arginine rich ( GR ) region in the C-terminus of Gag have been replaced by alanine , which results in a Gag protein unable to bind nucleic acid [20] ( S2 Fig ) . Virus particles are still released from cells , although less efficiently , but are non-infectious and display capsid assembly defects . The iFuse mutant is a variant of Env where the furine cleavage site between the SU ( gp80SU ) and TM ( gp48TM ) domains of Env has been mutated [11 , 12] ( S2 Fig ) . This results in a partially processed glycoprotein as the cleavage between the LP and SU domains is preserved . Particles are released at nearly wild type level from cells but are non-infectious . The presence of viral proteins of wild type and mutant viruses ( pr71Gag , p68Gag , gp130Env for the iFuse mutant , gp80Env and gp18LP for wt and iNAB mutant ) were confirmed by Western blot analysis ( S2 Fig ) . When observed by cryo-ET , PFV wt forms mainly near spherical particles ( Fig 2 ) as previously reported in [16] . The Env glycoprotein projections on the virus surface are ~14 nm in length . Underneath the membrane a fuzzy density not directly attached to the membrane connects to the capsid . Rare oval shaped viruses also exist but no tubular , elongated or more irregular structures were observed ( Fig 2 ) . Although the majority of viruses are spherical , their overall sizes are variable , ranging from 28 to 63 nm in radius measured from the center of the particle to the outer margin of the viral membrane ( rmean = 45 . 6 ± 8 . 2 nm ( n = 141 ) ) confirming previous observations [16] . Apart from their dimensions , PFV particles differ with respect to their internal structures . We classified PFV wt particles into four main classes ( A to D ) according to the variation in shape and morphology of their interior ( Fig 2 ) . Class A virus ( 15% of total particles ) contains an intact core at the center of the particle ( rmean = 30 . 0 ± 0 . 6 nm ( n = 25 ) ) consistent with previous measurement [16] . The capsid is surrounded by a second shell of density ( width~ 5–6 nm ) , which most likely corresponds to the Gag N-terminal domain that interacts with Env LP [10 , 16] . Class B ( 20% of total particles ) is similar to A except that the central Gag core is either incomplete or disrupted resulting in an open Gag structure , which generally merges with the intermediate shell . Class C ( 45% of total particles ) contains spherical particles with no regular , ordered internal structures and class D ( 20% of total particles ) miscellaneous particles having non spherical shape or an interior morphology significantly different from class A and B . We interpret classes A , B and ( most of ) D as particles containing capsids . Such viruses have been observed by electron microscopy of infected cells ( Fig 1 ) [15 , 21] and by cryo-EM [16 , 22] . They likely represent the released , mature and infectious form of PFV . Whether any of the virions of class C are infectious or not remains to be determined [23 , 24] . For near spherical viruses ( classes A to C ) , the distribution of particle dimensions within each group is different . Class A virions are less variable in dimensions ( rmean = 54 . 1 ± 2 . 9 nm , n = 25 ) than the other classes ( rmean = 50 . 3 ± 5 . 2 nm , n = 35 and 40 . 7 ± 6 . 7 nm , n = 81 for class B and C respectively ) ( S3 Fig ) . This implies that a complete capsid with the intermediate shell leads to particles with more homogeneous dimensions . Although the viral membrane , capsid and intermediate shell appear as distinct ( separated ) substructures in tomograms , there are nevertheless clear interactions between them ( capsid/intermediate shell and intermediate shell/membrane bilayer ) ( Fig 3A ) as indicated previously [16] , which probably contribute to the stability of the virus and the narrower size distribution of Class A viruses . When complete , the capsid at the center of particles has a relatively uniform mean radius ( r ~ 30 nm ) ( Fig 2 , class A–radial plots ) but few exceptions with larger and less uniform shapes exist as well ( Fig 2 , class D ) . The population with a constant radius often has a hexagonal outline with more or less sharp vertices by cryo-ET . Ab initio 2D class averages of only the capsid , calculated from images of iFuse mutant ( identical to wt in term of Gag sequence ) acquired by cryo-EM , reveal a variety of capsid shapes ranging from hexagonal to near round ( Fig 3D ) . This could be an indication of a regular , ( pseudo- ) symmetrical arrangement of the capsid . The capsid shell is relatively thick ( ~120 Å ) as there are strong densities associated with the capsid outer wall while its center appears comparatively less dense ( Fig 2 , plots and Fig 3D ) . The intermediate shell , which strictly follows the contour of the capsid , precludes a direct interaction of the capsid with the viral membrane . This 5–6 nm thick shell is less polyhedral ( hexagonal ) and appears discontinuous and structurally variable . Despite that , local regular organization ( spacing ~ 8–9 nm ) are visible in some tomographic slices ( Fig 3A ) [22] . The iNAB mutant can't bind the viral RNA or any nucleic acid . We previously showed by cryo-EM that this mutant displays severe assembly defects [20] . In the present study , we extended the analysis of iNAB mutant to cryo-ET to unambiguously visualize the interior of the viral particles ( Fig 3C ) . As wt PFV , the iNAB mutant forms a majority of near spherical particles on which the glycoproteins are clearly visible . Their dimensions ( rmean = 53 . 1 ± 10 . 3 nm , n = 42 ) are larger than wt PFV ( rmean = 45 . 6 ± 8 . 2 nm ) . On the average , more deformed and irregular particles are observed compared to wt PFV . We also confirm our previous result that without the RNA binding motif in Gag , and hence no RNA interaction , no regular capsid are assembled inside the near spherical particles ( Fig 3C ) . However , these particles are not empty and some diffuse densities are detected often in the vicinity of the viral membrane ( Fig 3C–arrows ) but they are distinct from the intermediate shell visualized in wt PFV . We conclude that the capsid and the intermediate shell observed for wt and iFuse PFV ( S3 Fig ) are mainly resulting from a co-assembly of Gag with viral RNA to form the nucleocapsid . The identification in some wt PFV tomogram of sections that reveal a capsid plus the intermediate shell and no glycoprotein on the surface strengthens this conclusion ( Fig 3B ) as it rules out the possibility that the intermediate shell is made only of the Env gp18LP cytoplasmic domain . Nevertheless , the Env gp18LP cytoplasmic domain ( 67 residues long ) has been shown to interact with the N-terminal domain of Gag [10] and therefore it most likely contributes as well to the overall intermediate shell density of mature PFV particles ( Class A , B and D of Fig 2 ) . Wt PFV and the iNAB mutant particles should have an identical Env structure ( same amino acid sequence , same processing by proteases ) , at least for the extracellular domain . The iFuse mutant lacks the gp80SU-gp48TM cleavage , which could influence the position of the fusion peptide region , but should have otherwise a similar structure than wt . Central slices through tomograms of the samples show that the glycoprotein extends around 14 nm away from the membrane while sections perpendicular to the glycoprotein's long axis clearly confirm that they are trimeric ( S3 Fig ) [14] . The 3D reconstruction of PFV wt , iNAB and iFuse glycoproteins by subtomogram averaging at ~3 nm resolution ( S3 Fig ) demonstrates that all three structures have a similar knob-like shape at this resolution ( Fig 4D–4F ) . They are arranged such that each trimer can interact with up to three other trimers to form a network of interlocked hexagons ( Fig 4A–4C and 4G–4I ) . This hexagonal network organization described previously in [14] is not obligatory as the viral membrane is not fully covered with glycoproteins . Isolated trimers were hardly found but incomplete hexagonal networks ( with less than six trimers ) were observed . Less ordered areas are also possible as well as pentagonal arrangements ( Fig 4A and S3 Fig ) . From cryo-ET reconstructions of a network of three adjacent and interlocked hexagons for wt PFV ( Fig 4G ) , it appears that the trimers at the periphery ( arrows in Fig 4G ) are less well defined than the central one , which indicates that the occupancy rate of the trimers within hexagons is quite low or that not all glycoproteins arrange in a hexagonal fashion or that the hexagonal packing is flexible . On the contrary , the same reconstructions of iNAB and iFuse mutants ( Fig 4H and 4I ) show stronger resolved features for all trimers , including the one at the periphery . This was also confirmed directly on tomograms where slices orthogonal to the mutant's glycoprotein long axis often display more regular hexagonal lattices than the wt ( Fig 4B and 4C ) . Thus , it appears that the two mutants have a stronger propensity to form a regular hexagonal network suitable for 3D reconstruction than the wt . It is not clear if this observation results directly from the mutations of the iNAB and iFuse samples or if the observed difference simply results from variability between virus preparations . To gain more insights into the structure of the PFV glycoprotein , we use 3D reconstructions of iNAB and iFuse Env hexagonal assemblies of six trimers determined by subtomogram averaging as initial references for automatic particle selection of micrographs acquired by cryo-EM ( S4 Fig ) ( see Material and Methods ) . The data were refined to higher resolution imposing C6 symmetry . Once the refinement converged , the six equivalent trimers of the hexagonal assembly were 3-fold symmetrized to yield the final single spike structure . The iNAB Env ( a fully processed Env as in wt virus ) and the iFuse mutant ( a gp80SU-gp48TM cleavage site mutant ) were computationally processed independently but yielded virtually identical structures at ~9 Å resolution at FSC = 0 . 143 ( S4 Fig ) . In the density map of the hexagonal assembly ( Fig 5A and 5B ) , the neighboring trimers are spaced by ~110 Å and are interacting with each other approximately 45 Å above the membrane level . This interaction seems to serve as a spacer for hexagonal lattice formation ( Fig 5A and 5B ) . There is no sign of ordered densities extending from the inner leaflet of the membrane towards the virus interior , which can be attributed to the cytoplasmic domain of Env gp18LP or the N-terminal region of Gag . The PFV Env trimer can be decomposed in an upper and lower region followed by the transmembrane ( TM ) region ( Fig 5C–5E ) . The upper part likely composed of gp80SU forms three arch-like structures , which join at the top and delineate a less dense area in its center ( Fig 5E ) . We propose that the three central short rods of density visible at the top of the lower part and surrounded by extra density are three α-helical regions derived from the fusion protein subunit gp48TM ( Fig 5C–5F ) . Gp48TM contains two potential coiled coil regions from residues 664 to 685 and 881 to 902 [25] or residues 664 to 691 and 880 to 908 [26] , which can correspond to heptad repeat region 1 and 2 , respectively , present in retroviral class I fusion proteins [27] . Based on the cryo-EM map , the length of the predicted helices is approximately 32 ± 5 Å ( Fig 5C–5E ) , which is consistent with 6 to 7 helical turns as predicted by sequence analysis . However , due to the limited resolution of the reconstruction , the actual helices may be slightly longer . The lower part of the spike is likely composed of parts of gp80SU and gp48TM , which anchor the extracellular domain in the membrane ( Fig 5C and 5D ) . The interaction on the top via the coiled coil and the one close to the membrane generate a central open space ( Fig 5C–5E and S5 Fig ) . Each Env monomer is predicted to have two transmembrane helices ( TMH ) : one in the gp18LP domain between aa 68–89 and one in the gp48TM domain between aa 961–980 . We propose that the gp48TM density extends into three central TMHs , which interact with three outer TMHs derived from the gp18LP ( Fig 5C , 5D and 5G ) . Notably , the three central TMHs of gp48TM are positioned such that they can interact with each other . Furthermore each TMH derived from gp18LP is in close contact to one gp48TM TMH ( Fig 5C , 5D and 5G ) . However , at the current resolution of the map , we cannot exclude the possibility that the three central helices are derived from gp18LP and the external ones from gp48TM
The majority ( ~80% ) of wt PFV virions have a nearly spherical shape with an average radius of ~54 nm as indicated previously in [16] which is smaller than mature particles formed by other orthoretroviruses such as HTLV-1 ( 57 nm ) [28] , HIV-1 ( 73 nm ) [29 , 30] and Rous Sarcoma Virus ( 63 nm ) [31] . Moreover , PFV viruses containing an intact capsid ( class A ) have an even more uniform radius . PFV budding is strictly dependent on the physical interaction between Env and Gag [21] . As Gag with RNA forms internal structures ( nucleocapsid with a surrounding shell ) of homogeneous size , it is not surprising that the viral membrane that contains inserted Env trimers ends up near spherical as well and follows closely the contour of the Gag assembly . Therefore , the narrower size distribution of PFV compared to other retroviruses could be the direct consequence of the interaction of Env with a preformed nucleocapsid . We interpret the viruses containing well-defined internal Gag structures ( class A , B and D of Fig 2 ) as the mature PFV particle . This is supported by in vivo observation of the same viral layout in virus infected cells . This organization is quite different from the mature particles of other retroviruses , which consist of a polyhedral core of CA containing the NC/RNA complex [28 , 30 , 31] as well as from their immature forms [32–35] . These differences in virus assembly likely relate directly to differences in posttranslational processing . Gag from most retroviruses is cleaved after the release of immature virions from the cell at various positions , which triggers the reorganization of the MA , CA and NC domains inside the virus [36] . For PFV , there is only one primary cleavage of Gag by the viral protease occurring in the course of virus assembly in infected cells , which produces the two Gag products p68Gag and p3Gag [5 , 6] . PFV virions contain an extra layer of intermediate density between the viral membrane and the capsid which we believe is derived from the Gag N-terminal domain and the cytoplasmic domain of the gp18LP as suggested previously [16] . This also indicates that in mature virions , the Gag N-terminal and gp18LP domains are flexibly linked to the capsid core and the glycoprotein respectively . Secondary cleavages have been also observed at three different positions of Gag [5] , however , their relevance for the virus life cycle needs to be established . Foamy viruses also share similarities with Hepatitis B virus . Notably HBV contains a capsid with icosahedral symmetry twice smaller ( r = ~14 nm ) than PFV [37] , buds intracellularly in the ER and forms large amounts of capsidless subviral particles used as a decoy for the host immune system [38] . Although PFV capsids can be variable in morphology , the majority has a uniform mean radius of 30 nm with a round to hexagonal outline . The latter are compatible with 2D projections of a symmetrical object including icosahedra . However , the analysis of intact capsids by either cryo-ET or cryo-EM did not allow defining a clear repetitive unit . The same conclusion was obtained by analyzing capsid images acquired by cryo-EM of complete virions in an earlier study [16] . Therefore , more studies are required to firmly establish whether the capsid has a defined ( pseudo- ) symmetry or not . Among the various morphology of PFV viruses , one large population ( Class C in Fig 2 ) representing 57% of the spherical particles has no sign of regular internal ordering and it is also the most variable class in terms of dimensions . The smallest members of this class cannot accommodate the 30 nm capsid of the mature particles and may constitute Env-only virus-like particles consistent with the role of Env in budding [23] . We used high pressure freezing and cryo substitution to analyze thin sections of PFV infected cells by ET . We observed viruses in various states and cellular localization . The landmark for each virus is the capsid , which appears as a dense round material in the cell . Capsids were present with or without viral membrane as reported before [15 , 21] . In addition , we found viruses budding from the plasma membrane , which have an intermediate shell of density between the capsid and the viral membrane . The same topology is found in released particles by cryo-ET and cryo-EM confirming that these particles are the mature form of PFV and that maturation occurs in the cell before virus egress . We also observed capsids accumulating in intracellular membrane tubes . Although PFV capsid assembly was reported to take place at the pericentriolar region [15] , we provide evidence that capsids accumulate in a membrane compartment whose origin is currently unknown . We suggest that they are immature unprocessed capsids , because of their slightly larger radius compared to the capsid at the plasma membrane ( S1 Fig ) . Notably the intermediate shell of density present in mature capsids [16] is absent and a closer interaction of the N-terminal domain of Gag with the core capsid could explain the larger capsid radius . Because they lack glycoprotein they are most likely not related to budding into intracellular vacuoles as observed before [21] . However , the nature of these potentially capsid-like structures has to be confirmed by further experiments as it cannot be ruled out that they are only indirectly related or completely unrelated to PFV . For instance , virus-related small vesicles have been recently reported during Tick-Borne Encephalitis Virus ( TEBV ) infection [39] . In contrast , these TEBV related vesicles were variable in size and not arranged regularly as in the case of PFV infection . The glycoprotein plays a central role in entry and virus budding [1] . Cryo-ET combined with sub-tomogram averaging is a well tailored method to study glycoprotein in situ ( embedded in the viral membrane ) . Several such structures have been determined to medium resolution for different enveloped viruses ( up to 20Å for HIV , SIV Env [40] ) . Here we used a different strategy and acquired 2D images on a direct electron detector by cryo-EM . We took advantage of the high density of glycoprotein at the surface of PFV as well as its symmetrical organization [14] and advanced automatic particle picking to reconstruct in situ PFV glycoprotein to ~9 Å resolution . We propose that PFV Env is a class I glycoprotein due to the two predicted coiled coil regions in the gp48TM fusion protein subunit and consistent with this prediction , we identify three rods of density centrally located in the structure which we interpret as a coiled coil region , a hall mark of class I fusion glycoproteins [41–48] . Furthermore , the maps show clear density for three TMHs derived from gp48TM and three from gp18LP . The central gp48TM TMHs are in close proximity to interact with each other and can be therefore considered to exert the role of a trimerization domain . Because of the importance of the transmembrane region for membrane fusion [49] , it will be interesting to study PFV Env gp48TM induced fusion due to the presence of the gp18LP transmembrane region . Current models of membrane fusion suggest that the trimeric TM region has to come apart to rearrange the fusion protein to catalyze membrane fusion [50 , 51] . The extra gp18LP TMHs may prevent the dissociation of the gp48TM TMHs and the TM “complex” may therefore move as a block during fusion . In line with the hypothesis that gp18LP affects fusion , mutations within its cytoplasmic region strongly enhance the fusogenic activity of such mutant glycoproteins [9] . An unusual feature of PFV Env is the physical interaction of glycoproteins with each other on the surface of the viral membrane to form intertwined hexagonal networks [14] . The interaction might be mediated by the extracellular part of gp18LP and/or parts of gp48TM , which could act as a spacer between glycoproteins . While this topology is not systematic , it seems to be the preferred one . Clustering of the glycoproteins on the surface has been reported for a number of other viruses having class I spike-like glycoproteins such as ASLV Env [52] , Influenza hemagglutinin [53] , and class III HSV gB [54] . In all these examples , the glycoproteins rather form patches or clusters arranged without symmetry or regular pattern . It has been suggested that the concentration of glycoprotein on the virus surface may confer an advantage for the virus to more efficiently recognize and attach to cellular receptors . Bunyaviruses , having class II glycoproteins [55] , are known to have local or global symmetrical arrangement of glycoprotein on the viral membrane [56–58]: trimeric , tetrameric and icosahedral organization have been reported but none of these resembles PFV's . Clusters of Env glycoproteins have been shown to be the preferred site of host cell interaction and thus entry for HIV-1 [59] and it is therefore likely that PFV Env arrangement in hexameric clusters confers also an advantage for host cell entry . The cellular receptor ( s ) essential for PFV Env-mediated membrane fusion is/are currently unknown but must be ubiquitously expressed molecule ( s ) due to the large range of permissive cells . Therefore the hexagonal arrangement of Env could enhance low affinity receptor interactions by increased avidity . In summary our structural analysis of isolated PFV and of PFV infected cells provides new medium resolution insights into the structural proteins Gag and Env , two important players in the PFV life cycle and will thus help to further develop PFV as a gene therapy transfer vector .
For production of replication-deficient , wild type PFV vector supernatant a 4 component PFV vector system , consisting of the expression-optimized packaging constructs pcoPG4 ( PFV Gag ) , pcoPE ( PFV Env ) , pcoPP ( Pol ) , and the enhanced green fluorescent protein ( eGFP ) -expressing PFV transfer vector puc2MD9 was used , which has been described previously [8 , 60 , 61] . Mutant PFV vector particles deficient in nucleic acid incorporation ( iNAB ) were generated using the PFV Gag packaging construct pcoPG4 GR R/A whereas particles deficient in PFV Env fusion ( iFuse ) were generated using the PFV Env packaging construct pcoPE32 instead of the respective wild type packaging constructs [12 , 62] . The CMV-driven proviral expression vector pczHSRV2 ( wt ) , described previously [63] , was used for production of replication-competent PFV supernatants . The human kidney cell line 293T [64] was cultivated in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% heat-inactivated fetal calf serum and antibiotics . Cell culture supernatants containing recombinant viral particles were generated by transfection of 293T cells with the corresponding plasmids using polyethyleneimine ( PEI ) as described previously [7 , 8] . For subsequent Western blot analysis the supernatant generated by transient transfection was harvested , passed through a 0 . 45-μm filter and centrifuged at 4°C and 25 , 000 rpm for 3 h in a SW32Ti rotor ( Beckman ) through a 20% sucrose cushion . The particulate material was resuspended in phosphate-buffered saline ( PBS ) . For cryo-EM analysis , viral particles were produced in serum-free medium and a further concentration step using Amicon Ultra 0 . 5 ml 100K concentrators was included following the first concentration by ultracentrifugation through 20% sucrose similar as described recently [20] . Cells from a single transfected 100-mm cell culture dish were lysed in detergent-containing buffer and the lysates were subsequently centrifuged through a QIAshredder column ( QIAGEN ) . Protein samples from cellular lysates or purified particulate material were separated by SDS-PAGE on a 10% polyacrylamide gel and analyzed by immunoblotting as described previously [9] . Hybridoma supernatants specific for PFV Env LP ( Env LP , clone P3B8-B7 ) , PFV Env SU ( Env SU , clone P3E10 ) or PFV Gag ( Gag , clone SGG-1 ) were employed [13 , 15 , 61] . After incubation with a horseradish peroxidase ( HRP ) -conjugated secondary antibody , the blots were developed with Immobilon Western HRP substrate . The chemiluminescence signal was digitally recorded using a LAS-3000 ( Fujifilm ) imager and quantified using ImageGauge ( Fujifilm ) . Wt PFV , iNAB and iFuse mutants were all prepared following the same procedure except that wt PFV particles were first inactivated for at least 1 h in 4% paraformaldehyde before being processed . 4 μl of sample containing 10 nm gold beads was applied to 2:1 Quantifoil holey carbon grid ( Quantifoil Micro Tools GmbH , Germany ) and the grid was plunge frozen in liquid ethane with a Vitrobot Mark II ( FEI , the Netherlands ) . For cryo-ET , the frozen grid was transferred to a FEI F20 FEG cryo electron microscope . Tilt series were recorded at 200kV with FEI tomography at a nominal magnification of 29 , 000 on a 4 k by 4 k Eagle CCD camera from -60 to +60° in 2° steps for a total dose of ~60 e-/Å2 . The defocus was set between -2 and -8 μm . Tilt series were aligned using gold particles as fiducials , binned two times ( final sampling of 7 . 6 Å/pixel ) and tomograms were calculated with IMOD [65] . For visualization , the tomograms were denoised by anisotropic diffusion in IMOD . For cryo-EM , samples were observed with a FEI Polara at 300 kV . Images were recorded on a K2 summit direct detector ( Gatan Inc . , USA ) in super resolution counting mode . For the iNAB sample , movies were recorded at a nominal magnification of 15 , 500 ( 1 . 32 Å/pixel at the camera level ) for a total exposure of 4s and 100 ms per frame resulting in 40 frames movies with a total dose of ~20 e-/Å2 . For iFuse , the magnification was 20 , 000 ( 0 . 97 Å/pixel at the camera level ) , the total exposure was 6s with frames of 0 . 2 ms resulting in 30 frames movies with a total dose of ~25 e-/Å2 . 230 and 297 movies were manually recorded with Digital Micrograph ( Gatan Inc . , USA ) for iNAB and iFuse PFV respectively . Data interpretation and processing were done with IMOD and PEET [66 , 67] . The protocol described below was applied for subtomogram averaging of glycoproteins of wt PFV , iNAB and iFuse mutants . The center and radius ( defined as the distance from the center to the outer margin of the viral membrane ) of each virus was defined in 3dmod . Initial centers for the glycoprotein sub volumes were defined on each virus surface on a regular grid ( defined by the spacing between the centers of the neighbor subvolumes and the distance between the subvolume and virus centers ) with seedSpikes in PEET . Two of the three Euler angles were estimated based on the spike orientation on the virus surface with spikeInit in PEET . Sub tomogram averaging was done in PEET . The initial reference was an average of all the sub volumes . In the first iterations , no symmetry was enforced but a cylindrical mask was applied to eliminate contributions from the neighbor spikes . The resulting asymmetric reconstruction clearly shows 3-fold symmetry ( which was also seen in raw tomograms ) . C3 was then enforced in the subsequent steps . At that stage , the subtomogram positions were checked manually in 3dmod and sub volumes that converged at positions where there was no glycoprotein or at the same position as other subvolumes were discarded . A final 3D reconstruction of 403 voxels was calculated with this new set of “clean” subvolumes using a soft edged mask around the reference . To obtain reconstruction of the hexagonal network , few more iterations were done without any masking and by increasing the reconstructed volume to 703 voxels . The final 3D models include 2000 , 3000 and 3000 subvolumes for wt , iNAB and iFuse PFV respectively . The final resolutions calculated by Fourier Shell Correlation between two half reconstructions were 3 . 2 , 2 . 9 and 2 . 8 nm at FSC = 0 . 5 for respectively wt , iNAB and iFuse PFV . Images were first motion corrected with unblur [68] . From the 3D reconstruction of a glycoprotein hexagonal network obtained by subtomogram averaging ( see above ) , we extracted one hexagon of six trimers and centered it on its 6-fold axis . The resulting reconstruction was used as a template for automatic particle picking using the Fast Projection Matching ( FPM ) method developed in the lab [69] . To speed up calculation , the raw micrographs were binned by a factor of 4 ( 5 . 28 Å/pixel ) and 6 ( 5 . 82 Å/pixel ) for iNAB and iFuse samples respectively . Masks were designed manually around each viral particle to further restrict the area where the automatic picking of glycoproteins was performed . For this initial automatic picking by template matching , the resolution was limited to 3 nm . For each micrograph , only particles having cross correlation with the reference higher that the mean correlation of all the particles were kept . The output coordinate files from FPM were converted to boxer [70] one ( . box ) and imported in RELION [71] where all subsequent steps were done with two-times binned images ( final sampling of 2 . 64 and 1 . 94 Å/pixel for iNAB and iFuse respectively ) . CTF estimation , particle extraction ( in boxes of 160 and 218 pixels ) for iNAB and iFuse respectively ) and preprocessing were done in RELION . The data sets were first cleaned by 2D classification and only classes showing clear glycoprotein network were kept . A first 3D autorefine was then done using the model obtained by subtomogram averaging as an initial reference ( low pass filtered to 60 Å ) and imposing C6 symmetry . The obtained 3D model was used as input for a 3D classification with 5 classes and C6 symmetry . The particles belonging to the classes showing the best resolved features were used to do another 3D autorefine . The resulting 3D reconstruction showed much improved features than the subtomogram 3D model . Therefore , we did another round of automatic picking with FPM using the latter model calculated with RELION as a reference and extending the resolution limit from 3 to 2 nm . The new coordinate files were imported again in RELION and the same procedure described above was repeated ( 2D classification , 3Dautorefine , 3D classification and 3D autorefine ) leading to new improved maps . We noticed that masking away the viral membrane by keeping only the densities corresponding to the extracelllular domains of the glycoprotein improved the alignment and quality of the 3D reconstruction . Following the last 3D autorefine , we therefore decided to do a focused 3D classification within a mask containing only the extracellular domains of the spikes and their TM regions ( 3 classes total without alignment ) . This leads to one class having much sharper details . We used the particles from this class to do a last 3D autorefine . The final reconstructions of hexagonal assemblies include 5541 out of 29659 and 5543 out of 25680 particles for respectively iNAB and iFuse and have resolutions of 11 . 7 and 9 . 8 Å at FSC = 0 . 143 . Each hexagonal assembly consists of six identical trimers and each one of them has an additional 3-fold symmetry that was not yet applied . The center of one trimer was defined in 3dmod and volumes of 1003 , 1363 voxels for iNAB and iFuse dataset respectively were extracted from the two half unfiltered maps calculated by RELION in the last 3Dautorefine run . The two volumes were then aligned such as their 3fold axis lies parallel to the Z axis with e2align3D . py from EMAN2 [69] . The two aligned volumes were then 3-fold symmetrized before being post processed in RELION with a mask to remove the contribution from neighbor spikes . This leads to an improvement of the density maps to 9 . 1 and 8 . 8 Å for iNAB and iFuse dataset respectively at FSC = 0 . 143 . Intact capsids from the iFuse mutant were picked manually from the same cryo-EM micrographs ( binned 6 times to 5 . 82 Å/pixel ) used for the glycoprotein reconstruction with boxer [70] . Particles were extracted in boxes of 130 pixels with RELION and 2D classification in RELION was performed which resulted in the isolation of a subset of 481 intact and regular particles . These particles were assembled in a stack , low pass filtered to 30 Å and further 2D classified in 5 classes by Multi variate Statistical Analysis in IMAGIC [72] . The 3D reconstructions of single trimer were sharpened with Bfactors of -780 Å2and -730 Å2 for iNAB and iFuse dataset respectively in RELION . Density maps visualization , segmentation and figure preparation were done with CHIMERA [73] . PFV Env amino acid sequence was submitted to the GeneSilico Metaserver ( https://genesilico . pl/meta2 [74] ) to predict the number and position of transmembrane helices . No structural homologues was found by the metaserver . Coiled coil prediction in PFV gp48 was done either with Coils [26] or Multicoils [25] . Human fibrosarcoma cells HT1080 were cultivated in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% heat-inactivated fetal calf serum and antibiotics . After 48 to 72h , cells were infected at a MOI of 10 and harvested 24 to 48 hours post infection . The cells were fixed and the virus inactivated with 4% paraformaldehyde . The cells were gently detached , pelleted at 500 g and most of the supernatant removed to obtain a thick cell slurry . The cell suspension was loaded in 100 μm deep gold carrier and cryo immobilized by high pressure freezing with a HPM100 ( Leica ) . The carriers were then transferred in an AFS2 device ( Leica ) maintained at -120°C . The temperature was slowly raised to -90°C before starting the cryo-substitution protocol ( adapted for Hawes et al . [75] ) . The freeze substitution cocktail ( 0 . 5% UrAc , 5% EtOH , 5% H20 in acetone ) was added for 4h at -90°C . The temperature was then raised at 20°C/h to -50°C . Samples were washed with Ethanol ( 1h ) and then infiltrated with increasing concentrations ( 33 , 50 , 66%; 1 h each ) of Lowicryl ( HM20 ) in ethanol followed by three incubation in 100% HM20 ( 2x1h + 18h ) . UV polymerization was initiated for 48h at -50°C , the temperature was then raised to 18°C ( 10°C/hour ) and polymerization continued for 40 h . Sections ( 100 to 300nm thick ) were obtained with a UC7 ( Leica ) ultramicrotome and post-stained with 5% UrAc and 1% Lead Citrate . For electron tomography , 10 nm gold beads were added to both sides of the sections . The grids were loaded in a F20 ( FEI , the Netherlands ) FEG electron microscope . Tilt series were recorded at 200kV with FEI tomography at a nominal magnification of 29 , 000 on a 4 k by 4 k Eagle CCD camera from -60 to 60° in 3° steps . The defocus was set around -1μm . Tilt series were aligned using gold particles as fiducials , binned two times ( final sampling of 7 . 6 Å/pixel ) and tomograms were calculated with IMOD [65] . Segmentation was done in 3dmod as well .
|
Foamy viruses ( FVs ) , which belong to the retroviral genus Spumavirus , are endemic to non-human primates and can be transmitted to humans . They are considered as potential vectors for gene therapy due to their broad cell tropism and their apparent apathogenicity in natural hosts and humans . In order to gain more insight into the ultrastructure of the prototype FV ( PFV ) we performed ( cryo- ) electron tomography and microscopy of infected cells and of isolated virions . We find that PFV contains a nucleocapsid of constant dimensions at its center , an intermediate shell of protein positioned between the core capsid and the viral membrane and glycoprotein that arranges into regular hexagonal lattices on the virus membrane . Structural analysis of the glycoprotein was performed in situ to a resolution of 9Å , which shows regular helical features such as a trimeric coiled coil of the fusion protein subunit , a hallmark of class I fusion proteins , spacer arms between the glycoprotein trimers and the arrangement of six transmembrane helices , a characteristic feature of the PFV Env glycoprotein . We discuss our results in light of the evolutionary relationship of PFV with other retroviruses as well as the role of the unique glycoprotein architecture on the virus life cycle .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"microbiology",
"viral",
"structure",
"electron",
"cryo-microscopy",
"membrane",
"fusion",
"microscopy",
"glycoproteins",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"viral",
"packaging",
"viral",
"replication",
"virions",
"cell",
"membranes",
"biochemistry",
"cell",
"biology",
"virology",
"virus",
"glycoproteins",
"electron",
"microscopy",
"biology",
"and",
"life",
"sciences",
"glycobiology"
] |
2016
|
Cryo-electron Microscopy Structure of the Native Prototype Foamy Virus Glycoprotein and Virus Architecture
|
We describe a statistical framework for QTL mapping using bulk segregant analysis ( BSA ) based on high throughput , short-read sequencing . Our proposed approach is based on a smoothed version of the standard statistic , and takes into account variation in allele frequency estimates due to sampling of segregants to form bulks as well as variation introduced during the sequencing of bulks . Using simulation , we explore the impact of key experimental variables such as bulk size and sequencing coverage on the ability to detect QTLs . Counterintuitively , we find that relatively large bulks maximize the power to detect QTLs even though this implies weaker selection and less extreme allele frequency differences . Our simulation studies suggest that with large bulks and sufficient sequencing depth , the methods we propose can be used to detect even weak effect QTLs and we demonstrate the utility of this framework by application to a BSA experiment in the budding yeast Saccharomyces cerevisiae .
Bulk segregant analysis ( BSA; [1] ) is a QTL mapping technique for identifying genomic regions containing genetic loci affecting a trait of interest . Starting with a segregating population from a genetic cross , individuals are assayed for the focal trait and two pools ( bulks ) of segregants are created by selecting individuals from the tails of the phenotypic distribution ( other sampling designs can also be used as discussed below ) . Genotype frequencies are estimated for the two bulks , either via genotyping of individuals or via the creation of pooled DNA samples from which allele frequencies are estimated . Allele frequencies should be approximately equal between the two bulks in genomic regions without loci affecting the trait . Regions of the genome containing causal loci should exhibit allele frequency differences between bulks . BSA is most effective with high marker density and accurate allele frequency estimation within bulks [2] . The former was effectively addressed with the application of microarray based genotyping to BSA [3]–[8] . More recently , investigators have begun to use massively parallel sequencing methods to estimate allele frequencies for BSA studies [9]–[11] , which has a number of advantages . For organisms with moderately sized genomes , next generation sequencing can provide essentially single base-pair resolution . In such cases rather than simply observing markers in linkage with causal loci the BSA-sequencing approach should allow one to observe allelic biases at the causal loci themselves . For larger genomes where high coverage of the entire genome is less practical , BSA-sequencing still has many potential advantages . For example , it does not require the design of new genotyping arrays for new crosses and may provide greater resolution than array based genotyping . Furthermore , sequencing data yields counts of alleles at polymorphic loci and thus provides a simple and intuitive way of estimating allele frequencies . In bulk segregant studies based on high-throughput sequencing there are two sources of variation that affect allele frequency estimates . The first is variation due to the sampling of segregants that constitute the bulks themselves . This source of variation can be minimized by increasing both the size of the segregant population and the size of the bulk samples . The second source of variation is a consequence of the measurement technique used to estimate allele frequencies in the bulks . In the case of sequencing of pooled DNA samples , the sources of variation of this second type include , but are not limited to , library preparation , sequencing chemistry , sequencing coverage , post-sequencing alignment of reads , and base/allele calling algorithms . Here again , some of these sources of variation can be minimized by standardization of experimental protocols and analysis pipelines . However some of these sources of variation , particularly stochasticity in sequencing coverage , are an inherent property of short-read sequencing methods . In this paper , we develop explicit statistical models to describe the sources of variation that should be considered in the analysis of BSA-sequencing data . We first develop test statistics based on the classic -statistic accounting for the two phase sampling inherent to BSA . We then propose an analysis pipeline for whole-genome studies and present a proof-of-concept example with data from yeast . A combination of simulation and empirical application demonstrate the utility of this analytical framework .
Based on the arguments developed above , we propose the following analytical pipeline for the analysis of BSA-sequencing data sets . We assume that sequencing reads have been aligned to a reference genome where physical distances between polymorphic sites and ( approximate ) rates of recombination are known . We assume that all sites are biallelic . Following alignment of reads to a reference genome , per site counts of each allele are generated from the reads . Our recommended analysis pipeline for estimating QTLs is as follows: We used simulations to conduct a simple power analysis of our proposed methodology . In this analysis we used the mean at a causal site as measure of power for given values of , , , window size ( ) , SNP density , and for different magnitudes of QTL effect on phenotype . Figure 2 summarizes results for two different values of , corresponding to large ( ) and very large ( ) F populations . We find that increasing coverage , , is advantageous until , but has minimal effect beyond that . A somewhat counterintuitive result is that larger bulk size , , is generally beneficial as long as sequencing coverage is modest to high . This is despite the fact that larger bulks imply weaker selection for a given ( and hence a smaller allele frequency divergence among bulks ) . Based on these findings we recommend bulks consisting of at least 10% and as perhaps as high as 20% of the F segregant population in order to maximize power to detect QTLs . To demonstrate the correspondence between theory and data we here draw on a BSA-sequencing data set generated to identify loci that contribute to variation in colony morphology in the budding yeast Saccharomyces cerevisiae [27] . A full description and analysis of these data will appear elsewhere ( Granek et al . , in prep ) . Here , these data serve to illustrate the utility of both our theoretical framework and the associated robust estimators for data analysis . The yeast data consist of a low and high bulk , each composed of 288 homozygous diploid segregants drawn from an F population of size generated by sporulating a naturally heterozygous diploid strain [28] . The low bulk consists of segregants with simple colony morphology , while the high bulk consists of segregants with complex colony morphology ( see [27] for a description of morphology scoring ) . Creation of DNA pools , sequencing , and mapping of reads is described in the Methods section . Because each segregant is homozygous , the effective number of alleles sampled for each bulk is instead of 2 . In total 44 , 066 polymorphic sites were analyzed with a mean interval between sites of approximately 280 bp . Below we refer to the two sequencing runs for the low bulks as and , and those for the high bulks as and . The coverage per SNP ( ) for each sequencing run was as follows: , , , and . For each of the analyses below , we used a smoothing window width of ( 30 cM ) , and took the average coverage of each bulk being compared as the estimate of coverage , . Because there are two sequencing runs per DNA pool , variation in allele frequency estimates between sequencing runs from the same segregant bulk should be exclusively due to stochastic aspects of the sequencing reaction and primary bioinformatics analyses ( base calling , read alignment ) . The structure of this data set is thus useful for dissecting the impact of sequencing variation on estimates of and , and the subsequent impact of this variability on the inference of QTL regions and peaks . We use these data to explore both the null model ( no QTL; by analyzing the low-vs-low and high-vs-high comparisons ) as well as the case where QTLs are expected ( comparing low-vs-high bulks ) . In the null case , the differences in allele frequencies are subject to only one source of variation because the bulks are fixed but sequencing is variable . The non-null analyses are individually affected by both sources of variation ( bulking and sequencing ) , but when comparing the results from comparable analyses ( e . g . comparing QTL peak locations between the -vs- and -vs- analyses ) , the differences are again simply a function of sequencing variation .
Our simulations suggest that for the experimental design considered here using bulk sizes as large as 15–20% of the phenotyped segregant population increases power to detect causal QTLs despite the fact that this means relatively smaller allele frequency differences between bulks . This is due to tradeoffs between bulk-size , selection intensity , and the variance of allele frequencies under the hierarchical sampling . Consider , for example , a single locus with alleles and , where the effect of is additive and the two homozygotes differ by units on average . Assuming no segregation distortion , and an population generated from inbred lines , the change in the allele frequency of in the high bulk after truncation selection is approximately [30] , [31] where is the intensity of selection , and is the ‘standardized effect of the locus’ ( these quantities can be related to the selection coefficient , , by ) . Given truncation selection on a normal distribution , the intensity of selection is given by where is the proportion of selected individuals and is the probability density function at the truncation point [31] . Since the intensity of selection increases at a rate much less than ( e . g . see [31] , Fig . 11 . 3 ) , an -fold decrease in results in a much less than -fold change in the intensity of selection . For example , let and consider truncation on the upper 20% , 10% , and 1% , of the phenotypic distribution . The increase in the frequency of in the high bulk given these truncation points is approximately 3 . 5% , 4 . 4% , and 6 . 7% respectively ( translating to allele frequency differences of 7% , 8 . 8% , and 13 . 4% in the two-bulk case ) . On the other hand , the variance of the realized frequencies of the alleles in each bulk is inversely proportional to bulk size ( ) . Thus , a twenty-fold decrease in bulk size translates to less than a two-fold increase in allele frequency divergence , but a twenty-fold increase in the variance of allele frequencies . As long as average coverage , , is moderate to large , the benefit of increasing offsets the relatively smaller penalty resulting from a decrease in selection intensity . However , there is little benefit to increasing sequencing coverage beyond the size of the bulks . Sequencing can introduce complications such as biases toward particular nucleotide calls; however in general this should effect both segregant bulks in the same direction . Due to the averaging affect of , unless such biased sites are common over very large map distances they are unlikely to have substantial affects on results derived under our proposed framework . Similarly , a low percentage of mismapped reads or miscalled SNP calling are unlikely to be problematic for our framework , again because of the averaging affect of . However caution should be exercised in genomic regions that are particularly problematic in this regard , such as repeat rich regions . In this paper we have focused on QTL mapping with an F experimental design , but clearly our framework can be extended to other designs . Common alternatives include mapping populations produced by imposing one or more generations of inbreeding on an F , such as Recombinant Inbred Lines ( RILs ) . The increased homozygosity of such populations should also be taken into consideration , as it increases the expected change in allele frequency due to selection but it also decreases the number of independent chromosomes that are sampled for a given number of selected individuals . Chromosomes in such RILs experience as much as twice the number of crossovers as do F populations so the physical size of the smoothing window should be reduced to take this reduced linkage disequilibrium into account . Even greater reductions of linkage disequilibrium can be accomplished by an alternative design that imposes additional generations of random mating , rather than inbreeding , on an F , resulting in more precise localization of QTLs . Additional generations of outcrossing ( beyond the F ) will likely magnify deviations of the null allele frequency from 0 . 5 owing to segregation distortion and/or inadvertent selection . This can be accommodated by application of formulas in Text S1 with estimated from all sites within a genomic window . Other experimental designs , such as backcrosses , will not have allele frequencies of 0 . 5 . For these situations the null expected distributions of and can be approximated using the equations presented in Text S1 , although in this case it will be necessary to know the parental origin of the SNP alleles . Similarly , since can be generalized to an arbitrary number of classes [12] , one-tailed scenarios ( e . g . [9] ) involving comparison to either a theoeretical population or a random sampling of segregants can be addressed in this framework .
To create the bulked DNA pools each segregant was grown overnight in liquid medium to saturation ( cells/ml ) and equal volumes of each culture were mixed to form cell bulks . Genomic DNA was isolated from the cell bulks and single Illumina DNA sequencing libraries were prepared from each bulk , using standard protocols as described in [28] . Each bulk DNA pool was sequenced twice using 50 bp reads on an Illumina GAII sequencing instrument . Approximately 15 M reads were generated in each sequencing run . Reads were aligned to the yeast reference genome ( obtained from the Saccharomyces Genome Database , January 2010 ) using the program BWA [32] and polymorphic sites were called using SAMtools [33] . For each sequencing run , SAMtools was used to create a pileup file giving the alleles at each polymorphic site , from which allele counts were derived using scripts written in Python .
|
Quantitative or complex phenotypes are traits that are under the control of multiple genes and environmental factors . Identifying the parts of the genome that contribute to variation in complex traits ( Quantitative Trait Loci or QTLs ) , and ultimately the genes and alleles that are mechanistically responsible for trait variation , is a primary challenge in animal and plant breeding , population studies of human health and disease , and evolutionary genetics . In this study we describe an analytical framework that allows investigators to marry a QTL mapping approach called “bulk segregant analysis” ( BSA ) with high-throughput genome sequencing methodologies in order to map traits quickly , efficiently , and in a relatively inexpensive manner . This framework provides a statistical basis for analyzing BSA experiments that use next-generation sequencing and will help to accelerate the identification of QTLs in both model and non-model organisms .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genome",
"sequencing",
"genomics",
"mathematics",
"statistics",
"genetics",
"statistical",
"methods",
"population",
"genetics",
"biology",
"evolutionary",
"biology",
"evolutionary",
"genetics",
"genetics",
"and",
"genomics"
] |
2011
|
The Statistics of Bulk Segregant Analysis Using Next Generation Sequencing
|
We have performed the first extensive profiling of Epstein-Barr virus ( EBV ) miRNAs on in vivo derived normal and neoplastic infected tissues . We describe a unique pattern of viral miRNA expression by normal infected cells in vivo expressing restricted viral latency programs ( germinal center: Latency II and memory B: Latency I/0 ) . This includes the complete absence of 15 of the 34 miRNAs profiled . These consist of 12 BART miRNAs ( including approximately half of Cluster 2 ) and 3 of the 4 BHRF1 miRNAs . All but 2 of these absent miRNAs become expressed during EBV driven growth ( Latency III ) . Furthermore , EBV driven growth is accompanied by a 5–10 fold down regulation in the level of the BART miRNAs expressed in germinal center and memory B cells . Therefore , Latency III also expresses a unique pattern of viral miRNAs . We refer to the miRNAs that are specifically expressed in EBV driven growth as the Latency III associated miRNAs . In EBV associated tumors that employ Latency I or II ( Burkitt's lymphoma , Hodgkin's disease , nasopharyngeal carcinoma and gastric carcinoma ) , the Latency III associated BART but not BHRF1 miRNAs are up regulated . Thus BART miRNA expression is deregulated in the EBV associated tumors . This is the first demonstration that Latency III specific genes ( the Latency III associated BARTs ) can be expressed in these tumors . The EBV associated tumors demonstrate very similar patterns of miRNA expression yet were readily distinguished when the expression data were analyzed either by heat-map/clustering or principal component analysis . Systematic analysis revealed that the information distinguishing the tumor types was redundant and distributed across all the miRNAs . This resembles “secret sharing” algorithms where information can be distributed among a large number of recipients in such a way that any combination of a small number of recipients is able to understand the message . Biologically , this may be a consequence of functional redundancy between the miRNAs .
Epstein-Barr virus ( EBV ) , a member of the gamma-herpesvirus family , is the most common virus in the human population[1] . It infects nearly 95% of adults and persists latently for the lifetime of healthy hosts . In the generally accepted model of EBV persistence , the virus initiates infection by crossing the epithelium of the oro-pharynx and infecting resting naïve B cells in Waldeyer's ring [2] , [3] . The establishment of persistent infection is characterized by the sequential employment of a series of latency transcription programs that allow the virus to drive the newly infected naïve B cell into the memory B cell compartment . Initially newly infected cells express all of the nine known latent proteins ( Latency III ) whose function is to cause the resting B cell to become an activated lymphoblast . This program may be important for cancer development , because it is capable of initiating the activation of B cells in vitro into continuously proliferating lymphoblastoid cell lines ( LCL ) . Furthermore , some of the latent proteins have been shown to possess oncogenic , pro-proliferation and/or pro-survival functions that could contribute to the development of malignancy [4] . The activated naïve B lymphoblasts in vivo rapidly migrate to the follicle to participate in a germinal center reaction [5] , [6] . Here they continue to proliferate but , unlike in vitro , they switch to a more restricted latency program ( Latency II ) where only three of the latent proteins are expressed . Ultimately the cells leave the germinal center as resting memory B cells ( MemB ) – the site of long term latent persistent infection . In MemB cells , all viral protein expression is extinguished ( Latency 0 ) except when the cells divide and express EBNA1 ( Latency 1 ) , the protein required for replication of the viral genome . This mechanism is thought to allow EBV infected cells to escape immune surveillance , enabling lifelong persistence . EBV was discovered in Burkitt's lymphoma ( BL ) and was the first human tumor virus identified . Subsequent studies revealed that the virus is associated with several other lymphoid and epithelial malignancies including Hodgkin's disease ( HD ) , nasopharyngeal ( NPC ) and gastric carcinomas ( GaCa ) ( reviewed in [1] , [7] ) . Surprisingly , none of the tumors that arise in immunocompetent individuals uses the viral growth promoting program Latency III . Instead they use Latency I ( BL ) and II ( HD , NPC and GaCa ) , Latency III specific transcription is not detected . Several lines of evidence have led to the suggestion that certain stages of the EBV life cycle are linked to the pathogenesis of distinct cancer types expressing the equivalent latency program . Thus EBV-positive BL cell has been linked with MemB cells and HD with germinal center B cells ( GCB ) [7] . Numerous studies have focused on analyzing and interpreting the function of the viral latent proteins in order to better understand their contribution to tumorigenesis . Recently , attention has been directed toward EBV microRNAs ( miRNAs ) that are expressed in latently infected cells . miRNAs are short noncoding RNAs , average length 22 nucleotides , that post transcriptionally regulate gene expression [8] , [9] . They function either by repressing translation or inducing mRNA degradation . An increasing body of literature suggests that miRNAs are involved in a wide array of biological events . With studies using molecular biology , computational analysis and newly emerging deep sequencing techniques , 44 mature EBV miRNAs derived from 25 precursors have been described [10]–[13] . EBV miRNAs are mainly encoded from two regions: BHRF1 ( Bam HI fragment H rightward open reading frame 1 ) and BART ( Bam HI-A region rightward transcript ) ( Figure 1 ) . miRNAs are the only known functional products of the BART transcripts [14] . BHRF1 derived miRNAs were reported to be highly expressed in LCL ( Latency III ) [10] , [13] , whereas BART miRNAs have been found in all EBV-infected cell lines tested including LCL , BL and NPC and tumor biopsies from NPC , GaCa and DLBCL [10] , [11] , [13] , [15]–[18] . However a comprehensive comparative accounting is lacking since most studies only examined a limited repertoire of miRNAs ( frequently with non- or semi quantitative techniques ) , a limited range of tissues was studied ( frequently employing cell lines instead of fresh infected tissue ) and appropriate computational methods for data mining were not employed . Using quantitative multiplex RT-PCR with specific 6′FAM-probes and primers for each miRNA , we have reported previously an EBV miRNA profile for NPC tissues [16] however a comparable profile does not currently exist for the other EBV associated tumors including BL , HD and GaCa . Therefore it is unknown if there is tumor specific variation in the patterns of EBV miRNA expression . A number of investigations have previously focused on the profiling of human cellular miRNAs in B-cell subpopulations and B cell associated lymphomas ( for example , see [19]–[24] ) . However , nothing is known about the expression of EBV miRNAs in normal infected B cells in vivo and consequently it is unknown if there are specific changes in their expression associated with tumorigenesis . Several groups have identified potential functions or targets for EBV miRNAs . It has been reported that the BHRF1 miRNAs are associated with Latency III and viral replication [15] , [25]–[27] , chemokine modulation [28] and cell cycle progression and proliferation [26] . There are conflicting reports about the possible role of the BARTs . They are suggested to regulate both viral [29]–[31] and cellular proteins associated with apoptosis , survival and immune evasion [32]–[34] . However , they are dispensable for infection and immortalization of B cells in vitro [35] and their absence had no reported effect on susceptibility to apoptosis of infected B cells [26] . This raises the possibility that the BART miRNAs may have an important role to play during normal infection of B cells in vivo that is not required in vitro . This parallels the behavior of LMP2 for example which is believed to play an important signaling and survival role in vivo as a B cell receptor surrogate but is completely dispensable for immortalization in vitro [36] , [37] . Therefore there is an immediate need to characterize and understand the expression profiles of EBV miRNAs in normal infected B cell populations and tumors in vivo . In the present study , we aimed to quantitatively assess and computationally analyze the miRNA expression profiles in EBV-associated tumor biopsies ( NPC , GaCa , BL , HD ) and EBV-infected LCL , GCB and MemB cells from normal populations . The goal was to discover if subsets of miRNAs are associated with specific latency programs and if these expression profiles are disrupted during tumor development . We present the first demonstration of deregulation of EBV miRNA expression associated with tumorigenesis . Specifically , we identify a subset of BART miRNAs that are restricted to Latency III in normal infection but are up regulated in tumors that express Latency I and II .
We have previously described a technique that allows the profiling of EBV miRNAs in small amounts of EBV positive tissue [16] . We have now used this technique to profile EBV miRNA expression in primary infected normal tissues and tumor biopsies . The tissues tested are listed in Table 1 . For the normal tissues , we profiled latently infected germinal center B cells ( GCB , Latency II ) from the tonsils ( n = 5 ) and latently infected memory B cells ( MemB , Latency I/0 ) from the peripheral blood ( n = 4 ) all derived from normal , persistently infected individuals . For the tumors , we profiled four types of primary biopsies including Burkitt's lymphoma ( BL: n = 6 , Latency I ) , gastric carcinoma ( GaCa: n = 6 , Latency II ) , nasopharyngeal carcinoma ( NPC: n = 5 , Latency II ) and Hodgkin's disease ( HD: n = 3 Latency II ) ( Figure 2 ) . For the tumor biopsies , we could either normalize the results to the cellular small RNA U6 or by fraction of total viral miRNAs . However , normalization to U6 was not meaningful for the GCB and MemB cells because the fraction of infected cells in these samples was both very low and variable . Therefore to compare tumor biopsies with normal infected tissue , we expressed each miRNA as a fraction of total EBV miRNAs . The results for expression of the BART miRNAs are shown in Figure 3 . There were two striking findings . The first was that despite the disparate tissue origins of the biopsies and the viral latency programs they represent , the profiles for all four tumor types were remarkably similar ( Figure 3A ) . Second , the profiles for the two normal infected tissues ( GCB and MemB ) were markedly different from the tumors but similar to each other ( Figure 3B ) , despite again originating from different tissues and employing different latency program . Therefore the similarity of the profiles was determined by whether or not the tissue of origin was neoplastic not on the latency program or the tissue of origin . The most striking difference was the absence of 11 BART miRNAs from the normal tissues that are highly expressed in the tumors . These included a large subset of the Cluster 2 BART miRNAs . Of the 18 Cluster 2 miRNAs tested , all were present in the tumor biopsies but only 8/18 ( 44% ) were found in the GCB and MemB samples . By comparison , all 10 Cluster 1 BART miRNAs were present in the tumor biopsies but only 1 was absent from the GCB and MemB samples . We conclude that EBV associated tumors expressing Latency I and II up regulate a subset of BART miRNAs that are silenced in their normal infected counterparts in vivo . One possible explanation for the different patterns described above is that the absent 11 BART miRNAs are associated with cellular proliferation and become down regulated when the cells enter a resting state . However , we can exclude this possibility because infected GCB are proliferating[6] . We therefore investigated an alternate hypothesis namely that these miRNAs are normally specifically expressed only in Latency III ( infected lymphoblasts ) i . e . with virus driven proliferation , and that their presence in tumors represents aberrant expression . To test this hypothesis we could apply a more quantitative approach since we are able to estimate absolute copy numbers per cell of the miRNAs in all three tissue types namely B cells driven to proliferate by EBV ( spontaneous lymphoblastoid cell lines - LCL ) derived from infected individuals ( Latency III ) , GCB ( Latency II ) and MemB ( Latency1/0 ) . This was possible for the LCL because they are homogeneous cell lines . For the GCB and MemB samples we could estimate miRNA copy number per cell by first measuring the number of infected cells in the samples to be profiled and then dividing the total copy number of each miRNA by this value . The result for the BART miRNAs is shown as a bar graph in Figure 4 and the actual values are tallied in Table 2 . The results for GCB and MemB were again very similar indicating that the profiles were essentially the same both in terms of relative representation ( Figure 3B ) and absolute copy number ( Figure 4B and Table 2 ) for the BART miRNAs . One notable exception was 17-5p which was almost undetectable in GCB ( average 2 copies/cell ) but present at a copy number almost 2 logs higher in MemB ( average 110 copies/cell ) . Of the 18 Cluster 2 miRNAs profiled 2 were undetected in all three tissues ( LCL , GCB and MemB ) . The remaining 16 were all present in LCL . As with the tumors this included the one Cluster 1 and eight Cluster 2 BART miRNAs that were absent from GCB and MemB samples . A second feature to emerge from this comparison is that the BART miRNAs that were detected in GCB and MemB were present at copy numbers that averaged 5–10 fold higher than in LCL . We have investigated whether our failure to detect BART miRNAs from in vivo samples was due to lack of sensitivity of the PCR reaction rather than true absence . Of the 12 BART miRNAs that were not detected in GCB and MemB cells 4 were also either absent or present at marginal levels ( ∼1 copy /cell ) in the LCL . Therefore the failure to detect these in the GCB and MemB may not be meaningful . However , the remaining 8 absent miRNAs ( highlighted in red in Figure 4 ) were present in LCL at copy numbers ranging between 5 and 327 , identical to the range in LCL for the BART miRNAs that were found in GCB and MemB ( range 5–250/LCL cell ) . Therefore the undetected miRNAs were not consistently those present at low copy numbers in the LCL . Furthermore , when BART miRNAs were detected in GCB and MemB cells they were typically detected at higher levels/cell than in the LCL suggesting that we are not under representing or under detecting miRNAs in these samples . We also performed profiles on samples of 106–107 EBV negative tonsils into which had been spiked various numbers of LCL cells and on GCB and MemB samples with low numbers of infected cells . We determined that we could quantitatively profile most of the miRNAs in samples that contained ≥1000 infected cells in a population of 106–107 uninfected cells ( data not shown ) . As the number of infected cells dropped from 1 , 000 failure of profiling tended to be associated with drop out of most or all of the miRNAs rather than selective gradual disappearance . In particular we did not see preferential drop out of the miRNAs absent from GCB and MemB . This suggests we may be approaching a general rather than miRNA specific threshold for profiling . The exceptions were mirBARTs 9 , 12 , 16 and 19-3p . The PCR for these miRNAs consistently gave significant signals due to cross reaction with RNA from the large number of uninfected cells and were therefore excluded from all analysis . With the exception of one GC sample ( 600 infected cells ) all of the GCB and MemB samples assayed in our study contained >1000 infected cells ( Table 1 ) . We conclude that there is a subset of BART miRNAs that reside predominantly in Cluster 2 and are specifically expressed only during Latency III in normal infected B cells . We refer to these as Latency III associated BARTs . Furthermore there is co-ordinate regulation of the BART miRNAs where approximately one third are extinguished and two thirds are up regulated as the cells traverse out of Latency III into Latency I and II . We have established that there is a subset of Latency III associated BART miRNAs that are also expressed in tumors , irrespective of latency type . We wished to investigate therefore whether the absolute level of expression of BART miRNAs in the tumors also matched those in the LCL i . e . lower compared to normal tissue . To gain insight into this we compared levels of BART miRNA expression in LCL and tumor biopsies after normalization to the ubiquitous small cellular RNA U6 . The results are summarized in Figure 5 . Overall , the pattern of miRNA expression was similar for all of the tumors ( as already shown in Figure 3A when normalized by fraction ) and for the LCL . The LCL , HD and BL were closely matched ( Figure 5A ) both in overall pattern and copy number with the exception that the LCL lacked mirBART 9* and mirBARTs 15 and 18-3p were present at a level more than tenfold higher in the LCL . Also HD lacked mirBARTs 2-3p , 18-3p and 20-3p which were among the group of miRNAs absent from GCB and MemB . However , when LCL were compared with NPC and GaCa ( Figure 5B ) it was apparent that the epithelial tumors had significantly higher overall expression of the BART miRNAs . When we estimated the average fold increase of the BART miRNAs in the tumors relative to LCL , we found BL 1: NPC 13: HD 0 . 34: GaCa 8 . Note that all of the tumors had some levels of infiltrating non-tumor cells ( Figure 2 ) that would lower the estimates of miRNA expression . However , with the exception of HD , these constituted a small ( less than half ) fraction of the tumors and would not significantly affect the estimates . Therefore , the much lower levels of BART miRNAs in HD can be explained at least in part from the low abundance of tumor cells in these biopsies . Overall though it appears that the levels of BART miRNAs in the B cell tumors are comparable to those in the LCL but 5–10 fold lower than the epithelial tumors . We conclude therefore that the BART miRNAs are expressed in all four tumors and that this represents deregulated expression of the Latency III associated BARTs . HD might represent an intermediary state where most but not all of the Latency III associated BART miRNAs are expressed . Previous studies have demonstrated that the BHRF1 miRNAs are expressed in Latency III where they are reported to play an anti-apoptotic role [15] , [25]–[27] . This result was confirmed in our profiling ( see Table 2 and Figure 6 ) . All four BHRF1 miRNAs were readily detected in the LCL with copy numbers per cell ranging from 10–2000 depending on the miRNA and the cell line tested ( Table 2 and not shown ) . However , they were all absent from GCB ( Latency II ) and only one ( BHRF1-1 ) was found in MemB ( Latency0/1 ) . We have demonstrated above that the Latency III associated pattern of BART miRNA expression is deregulated in all four tumor types we have studied . To test if this was also true for the BHRF1 miRNAs , we profiled their expression in all of our tumor samples and the result , compared to LCL , is shown in Figure 6 . The BHRF1 miRNAs were not detected at all in GaCa and only sporadically and at low levels in the other tumor biopsies . Using the LCL values as a standard it is possible to estimate that , with the exception of BHRF1-1 in HD which was present in ∼ 5 copies , all of the rest were present at ≤1 copy per cells . This means that in the tumor samples the levels of all of the BHRF1 miRNAs were at least 10 fold lower in expression than in Latency III ( LCL ) . Thus BHRF1 expression was not consistently deregulated in the tumors . We have described a pattern of deregulated BART miRNA expression in EBV associated tumors however at the crude level of our analysis we did not detect tumor type specific miRNA expression . To investigate this more rigorously we have performed clustering analysis on our data sets using heat maps and principal component analysis . Figure 7 shows a heat map with clustering dendrograms of EBV miRNA expression for all of the normal and tumor tissue samples we have tested normalized by expressing each miRNA as a fraction of the total EBV miRNAs . The ordering of the samples across the heat map coincided exactly with the three major branches of the miRNA dendrograms which in turn were associated with a functionally distinct group of samples namely normal tissue ( GCB and MemB ) , LCL and the tumor biopsies . This analysis did not distinguish between GCB and MemB cells and the miRNAs responsible for resolving the GCB+MemB from the LCL ( Table 3 ) coincided with those already identified above . This served as validation for the groupings indicated by the heat map . This was important because the heat map revealed new information namely that the tumors all formed a discrete branch on both the sample and miRNA dendrograms ( blue box and Table 3 ) and that within this group the tumor samples were ordered by tumor type . This was unexpected given the similarity of the profiles as shown in Figures 3A and 5 . This result was confirmed when we repeated the analysis on a second set of biopsies ( not shown ) . This implies that there are tumor specific patterns of EBV miRNA expression . Curiously though , inspection of the heat maps , failed to identify specific subsets of miRNAs that could account for this resolution . In an attempt to understand the basis for this and possibly identify tumor specific patterns of miRNAs , we analyzed the data by principal component analysis ( PCA ) . Figures 8 and S2 and Video S1 show the result from the same data set used for the heat map in Figure 7 . Output for the first three principal components , which account for 55% of all the variation within the normalized data , is shown and confirms that the different tissue types do indeed cluster discreetly . When we examined the contributions of different miRNAs to the first 3 principal components we reproduced the findings from the heat map ( Table 3 , Figure S2 ) . Of greatest interest was the third principal component which resolved all four tumor types . Resolution of the four tumor types could be shown even more clearly when PCA was performed on the data from the biopsies alone ( Figure 9A and Video S2 ) and this result was confirmed when we analyzed data from a second completely independent set of biopsies ( not shown ) . We performed two tests of the statistical significance of these results ( see Materials and Methods ) . Both of these tests showed that successfully separating the four cancer types has a p-value of approximately 0 . 001 . We developed a number of analytical tools to try and extract information about the miRNAs responsible for this effect to no avail . The reason for this became apparent when we attempted to identify which miRNAs were essential and which dispensable for resolving the four tumor types by PCA . To do this we randomly generated subsets of miRNAs and asked if they were capable of resolving all four tumor types by PCA . The surprising result , shown in Figure 9B , was that 10% of such subsets that contained just 3 miRNAs and 60% that contained 5 miRNAs could resolve the tumors . When we then looked at which miRNAs were present in these subsets we found that all of the miRNAs were represented – there was no subset of miRNAs that was uniquely responsible for distinguishing the tumors . Similarly when we asked the question: which miRNAs were dispensable in a given sub-set , i . e . could be removed without affecting resolution of the tumors , we again found that all miRNAs were essential in certain subsets ( not shown ) . This means that the information about the EBV miRNAs which varies between the tumor types and allows their resolution is contained in part by all of the miRNAs such that when combined multiple different subsets of miRNAs contain sufficient information to distinguish the tumors . This phenomenon is reminiscent of a behavior that is well known in computer science referred to a “secret sharing” [38] , [39] and represents perhaps a novel and first description of a biological system where information is so distributed across a population . We have profiled EBV miRNAs from a number of tumor derived cells lines including 5 BL derived lines and one each of gastric , nasopharyngeal and Hodgkin's origin . When this data was analyzed in a heat map with the data from Figure 7 , all of the cell lines clustered with the LCL not with the tumor biopsies they originated from ( Figure S1 ) . This was most striking for the BL lines when analyzed by PCA . As shown in Figure 10 and Video S3 the cluster of BL lines completely intersects the cluster of LCL whereas the non BL lines lie just outside . The tendency of the tumor cell lines to drift towards an LCL phenotype was further confirmed when we analyzed expression of the BHRF1 miRNAs ( [16] and Table 4 ) . As discussed above these miRNAs are associated with Latency III , the LCL phenotype , and only one is expressed at a significant level in normal infected tissue or tumor biopsies using Latency II or I . As shown in the Table they were all expressed at significant levels in the tumor derived cell lines tested with the exception of the NPC derived C666-1 cells . We conclude therefore that miRNA expression in the cell lines is not fully representative of their tumors of origin .
In this paper we have reported on the expression profiling of EBV miRNAs in a wide variety of infected normal and neoplastic tissues that express all of the known EBV associated latency transcription programs . We have shown that there are distinct patterns of miRNA expression associated with Latency III and the restricted forms of latency ( Latency II/I/0 ) and that these patterns are deregulated in EBV associated tumors . GCB ( Latency II ) and MemB ( Latency I/0 ) express the same unique , restricted pattern of miRNAs with the exception that mirBART 17-5p is preferentially expressed in MemB cells . This unique pattern includes the absence of 12 BART miRNAs that include approximately half of Cluster 2 . At least 8 of these absent miRNAs are expressed in B cells expressing Latency III ( LCL ) and in all of the tumor biopsies we have tested although none of the tumors uses Latency III . Thus this is a Latency III restricted pattern of miRNA expression , which we refer to as Latency III associated BARTs , that is deregulated in tumors . Interestingly another group of miRNAs that are associated with Latency III , the BHRF1s [15] , [25]–[27] , are not deregulated in the tumors suggesting that viral miRNA deregulation in the tumors is specifically targeted at the BARTs . This is an important conclusion since it represents the first demonstration of Latency III specific gene expression in tumors that are otherwise expressing restricted ( Latency I/II ) forms of latency[1] , [7] and raises the possibility that BART miRNAs may contribute to oncogenesis . The observation of latency program specific miRNA expression could only have been made by studying in vivo derived infected material since the Latency III associated BARTs are expressed in all tumor biopsies and cell lines we have tested . In the generally accepted model of EBV persistence , newly infected naïve B lymphoblasts ( LCL ) expressing Latency III , switch to Latency II when transiting the germinal center ( GCB ) to become resting memory B ( MemB ) cells , the site of long term persistence where viral latent protein expression is extinguished [2] , [3] , [7] . Our results here suggest that the transit from EBV driven growth into more restricted forms of latency in GCB and MemB cells is associated with turning off expression of the Latency III associated BARTs and up regulation of the remaining BART miRNAs by 5–10 fold . This is not simply related to the cessation of proliferation because latently infected GCB are , like LCL cells , proliferating [6] . Since proliferation in the GC is not driven by Latency III , the viral growth program , we may conclude that the Latency III BART profile ( presence of the Latency III associated BARTs and reduced expression of the remaining BARTs ) is specifically associated with EBV driven growth . One caveat to this conclusion is that it would have been desirable to confirm the Latency III pattern of miRNA expression on in vivo infected cells rather than spontaneous LCL . Unfortunately newly infected tonsil naïve B cells in vivo are present at a level 5–10 fold lower than infected GCB [40] . This puts them below the threshold of sensitivity and reliability for our profiling and therefore was technically not feasible . Several studies have reported on potential roles for EBV miRNAs . There is no striking correlation between these reports and the BART miRNAs we have as the Latency III associated BARTs . Marquitz et al [41] have suggested that Cluster 1 and 2 BART miRNAs interact in apoptosis resistance by targeting BIM . However , their observations are not consistent with those of Seto et al , who have reported that BART miRNAs have no impact on LCL growth or survival in vitro[26] , or that the entire BART region can be deleted without impacting the transforming capacity of the virus[35] or that the prototypical laboratory strain B95-8 has most of the BART region , including most of Cluster 2 , deleted yet is unimpaired in its transforming ability . The explanation for this discrepancy may lie in the fact that Marquitz et al performed their studies in an epithelial cell line not in B cells . Taken together these results suggest that the Latency III associated miRNAs we have identified , may play a crucial survival role in vivo for newly infected naïve B lymphoblasts activated by the EBV Latency III program but that this role is dispensable for in vitro growth much as has been shown for LMP2 [36] , [37] . We assume that the specific up regulation of this group of miRNAs in tumors implies they could play a similar survival role in tumor development . Our results suggest that expression of the Latency III associated BARTS is coordinately regulated . It seems unlikely that this is occurring at the level of transcription/splicing . It is known that the BART miRNAs are derived from the first four introns of the BART transcript prior to the splicing event [42] and the miRNAs absent from GCB and MemB cells are not contiguous but randomly distributed among these introns . For examples , Bart 15 is located in the region between exon 1a and 1b whereas Bart 10 and Bart 20 are in the junction of exon 2 and 3 . Therefore , it is unlikely that the differential miRNA expression we have described is related to the selection of splicing patterns . Other possible mechanisms that are known to regulate miRNA expression are differential DNA methylation [43]–[45] and RNA editing both of which have been shown to function on BART miRNAs [46] , [47] . However , these mechanisms defer rather than answer the question as to why or how this particular subset of miRNAs is targeted for coordinate expression . A mechanisms that we favor is based on the observation that the stability of miRNAs is dependent on the presence of their target mRNA [48] . In this case the absence of miRNAs in GCB and MemB that are present in LCL and the tumors would arise because the mRNAs targeted by those miRNAs were only present in LCL and the cancers . We were surprised to find that the four tumor types clustered together in the heat map . This was irrespective of the tumor type , tissue of origin or the EBV transcription program that they employed . Perhaps more unexpected was our finding that despite the very similar patterns of miRNA expression the different tumor types were nevertheless clearly distinguished in two separate assays ( heat map/clustering and PCA ) applied to two completely separate data sets ( p in both cases = 0 . 001 ) . The basis for this resolution is less clear since it is not associated with any particular subset of miRNAs . Rather we discovered that a majority of all subsets of five or more miRNAs and some subsets as small as three were capable of distinguishing all four cancer types . Even more curious was the finding that every miRNA is capable of contributing to one of these subsets . Taken together , these results mean that the signal distinguishing these cancers is highly redundantly encoded across the miRNA expression profiles . Such a distribution of information is uncommon indeed may never have been reported before for a biological system . However , this type of behavior is well known in physics and computer science where there is a close analogy to secret sharing algorithms [38] , [39] . For example , it is possible to share a secret message among any number ( in our case ∼40 ) people in such a way that if any 5 of them divulge their information to each other , the message can be read . It is interesting to speculate that the signal in our assays , i . e . , the causes of or responses to each cancer type is in some similar manner parceled out among the miRNAs . How this might work at the molecular level is unclear but we assume it must reflect extensive redundancy in the miRNAs both in the number that target the same gene and in targeting genes that lie in the same or parallel signaling pathways related to tumor development . Evidence for such layers of redundancy in miRNA function is well known [49]–[52]and has recently been reported for both the Cluster 1 and 2 BART miRNAs in reducing apoptosis susceptibility in an epithelial cell line [41] . It has been suggested previously , based on cell lines , that the copy number of BART miRNAs is higher in epithelial cells than B cells . This is consistent with our estimates of the relative abundance of the miRNAs in our tumor biopsy samples versus the LCL . However , the correlation of high expression with epithelial tissue is confounded by our measurement of relative abundance in the MemB and GCB populations where the expressed BART miRNAs are present at comparable levels to the epithelial tumors . The exact meaning of these variable levels of expression is therefore now unclear . The only EBV latency transcription program that we found to be associated with a specific pattern of miRNA expression was Latency III characterized by up regulation of the Latency III associated BART and BHRF1 miRNAs . The later serves as validation for our approach since it has been reported [27] and confirmed [15] previously . We found BHRF1 expression in some of the tumor biopsies notably BL and HD although they were present at a very low level which we estimate to be generally less than one copy per cell . The absence of BHRF1 miRNAs from EBV associated tumors is consistent with previous findings that BHRF1 miRNAs were not found in biopsies from GaCa [53] and DLBCL tissues [17] . By contrast we found abundant BHRF1 miRNAs in the tumor derived cell lines . This , together with our finding that the BART miRNA expression profile in these lines tended to resemble LCL more closely than the originating tumor biopsies , casts doubt on the value of using such lines to study EBV miRNAs . In conclusion , we have presented the first comprehensive profiling of EBV miRNAs from in vivo derived normal and neoplastic tissue . These results demonstrate specific patterns of expression in Latency III versus more restricted forms of latency and deregulation of miRNA expression in tumors .
The EBV-positive lymphoblastoid cell line ( LCL ) IB4 ( gift of Dr . Elliot Kieff ) and the murine CB59 T cell line ( gift of Dr . Miguel Stadecker ) were used respectively as the positive and negative control for EBV . For miRNA profiling , six spontaneously EBV-infected B LCLs ( gift of Dr . Alan Rickinson ) with different EBV strains ( type 1 or type 2 ) , five EBV-positive Burkitt's lymphoma ( BL ) cell lines Rael ( gift of Dr . Sam Speck ) , Jijoye , BL36 , Raji and Akata 2A8 . 1 ( gift of Dr . Jeff Sample ) , the gastric carcinoma ( GaCa ) line AGS/BX1 ( gift of Dr Lindsey Hutt-Fletcher ) , the nasopharyngeal carcinoma ( NPC ) line C666-1 and the Hodgkin's disease ( HD ) -derived cell line L591 ( gift of Dr . Paul Murray ) were included in this study ( Table 1 ) . The EBV-negative BL lines Akata , BJAB , DG75 , BL2 and BL31 , the NPC line HONE-1 ( gift of Dr . Ronald Glaser ) , and the GaCa line AGS were used as the negative controls . The GaCa cell lines were grown in Ham's F-12 medium containing 10% fetal bovine serum ( FBS ) , 2 mM sodium pyruvate , 2 mM glutamine , and 100 IU of penicillin-streptomycin . All other cell lines were maintained in RPMI 1640 medium with the same supplements . The research described herein was approved by the Tufts University Institutional Review Board and our collaborating institutions . Peripheral blood mononuclear cells ( PBMCs ) of whole blood samples were provided by the University of Massachusetts at Amherst Student Health Service as previously described . Adolescents ( ages 17 to 24 years ) presenting to the clinic at the University of Massachusetts at Amherst Student Health Service ( Amherst ) with clinical symptoms consistent with AIM were recruited for this study . Blood was collected following the obtainment of written informed consent . These studies were approved by the Human Studies Committee at the University of Massachusetts Medical School ( Worcester ) . Tonsils were collected from patients 18 years of age or younger receiving routine tonsillectomies at the Tufts Medical Center at Boston , MA . Informed consent was not obtained since this was deidentified , discarded material and was deemed exempt by the Tufts University Institutional Review Board . Tumor biopsy samples for this study were obtained from the archives of the Vrije Universiteit , Amsterdam ( VU ) medical center . Consent was not obtained because we used left over archival material from earlier studies ( listed below ) . This was approved by the Medical Ethical Committee of the VU University medical center , Amsterdam , The Netherlands according to the code for proper secondary use of human tissue of the Dutch Federation of Biomedical Scientific Societies ( http://www . federa . org ) . Burkitt's lymphoma ( BL ) samples were collected in Malawi from 1996–1998 under study nr . IC19-CT96-0132 . Hodgkin's disease ( HD ) samples ( all of nodular sclerosing subtype ) were collected in Amsterdam from 1994–2002 under studies nr . KWF-VU1994-749 and 2001–2511 Gastric carcinoma ( GaCa ) samples were collected in Amsterdam from 1999–2004 under study nr . KWF1999–1990 . Nasopharyngeal carcinoma ( NPC ) samples were collected in Indonesia during 2001–2005 in studies KWF-IN2000-02/03 . A single case of HD was obtained from the Children's Hospital , Birmingham , UK , with permission from the Childhood Cancer and Leukaemia Group ( CCLG ) of the United Kingdom . This sample was used in accordance with Trent Research Ethics Committee REC reference number 05/MRE04/ . Written consent was obtained and the sample taken under ethical approval obtained from South Birmingham Research Ethics Committee . EBV status of tumor biopsies was assessed based on EBER1/2 in situ hybridization using commercial PNA-based hybridization probes ( Dakocytomation , Glostrup , Denmark ) and immunohistochemical staining for EBNA1 and LMP1 using previously described monoclonal antibodies[54] , [55] . Tonsils were cut into small fragments in phosphate-buffered saline with 1% bovine serum albumin ( PBSA ) with razor blades . Cells were obtained by filtering the fragment suspensions through a 70 µm mesh size cell strainer . Mononuclear cells were isolated from buffy coats using the standard Ficoll-Paque Plus ( Fisher Scientific ) centrifugation method and saved for analysis . Germinal center B ( GCB ) and memory B ( MemB ) cell populations were purified by fluorescence-activated cell sorting ( FACS ) . Surface staining for FACS analysis was performed by standard procedures . Monoclonal antibodies against specific cell surface markers including allophycocyanin ( APC ) -labeled anti-CD19 , phycoerythrin ( PE ) -anti-CD10 , and fluorescein isothiocyanate ( FITC ) -anti-CD27 were used . GCB cells ( CD19+CD10+ ) and MemB cells ( CD19+CD27+ ) were sorted from tonsil and blood PBMCs , respectively as previously reported [5] , [56] . For miRNA studies , 5×106 to 107 GCB cells were sorted for RNA isolation . Since the blood samples were small and MemB cells ( CD19+CD27+ ) only account for 0 . 5-2 % of PBMCs , we sorted all the memory B cells into 106 CB59 cells prior to RNA isolation . EBV miRNA profiling of CB59 cells indicated that it is appropriate to use them as filler cells since they do not generate detectable signals for any EBV miRNAs ( data not shown ) . Determination of the EBV frequency of infected cells in purified GCB and MemB cells was done by limiting dilution and DNA real-time PCR as described previously [12] . Briefly , FACS-gated cells were sorted onto a 96-well plate with 10 replicates each of serially diluted cells . Genomic DNA was isolated by Proteinase K digestion , followed by real time Taqman DNA PCR specific for the W-repeat genome of EBV . The fraction of EBV-negative wells was calculated and the frequency of infected cells was estimated using the Poisson distribution . Total RNA was extracted from frozen sections or sorted cell populations with Trizol ( Invitrogen ) . The quantity and integrity of the RNA was assessed by nanodrop and by the Agilent Bioanalyzer . EBV miRNA expression was assessed by real-time multiplex reverse transcript ( RT ) -PCR as previously detailed [16] . Briefly , multiple stem-loop RT primers specific for the 3′ end of each mature miRNA were mixed and applied to RT reaction , followed by Taqman PCR using primers and probes specifically assigned to each miRNA . Synthetic oligonucleotides representing all of the miRNAs were employed to generate the standard curves . The small cellular nuclear RNA U6 was used as the internal control for normalization . EBV-negative materials were used for negative controls . Profiles were performed in duplicate and repeated three times . Since , in our to-be-profiled sample pool the frequency of infected cells in 5×106 to 107 GCB ranged between 6–130 per 105 GCB , most of the RNA was derived from cellular rather than viral origin . To test if this compromised any of the PCR assays we profiled 5×106 and 107 EBV negative cells . Of the 38 miRNAs tested only four ( BARTs 9 , 12 , 16and 19-3p ) gave detectable and significant signals with EBV negative cells so these were excluded from further analysis involving these cell types . The data we have addressed here consist of copy numbers for 34 EBV miRNAs ( 38 for the cell lines and biopsies ) from each of 43 samples . These include biopsies from four tumor types , EBV+ MemB-cells and GCB-cells from normal carriers , lymphoblastoid cell lines and tumor derived cell lines . We can consider these as a collection of 43 points in 34-dimensional space , i . e . , . In order to perform specific comparisons , we have analyzed subsets of these points , which we will refer to generically as . We normalized each sample to the total EBV miRNA count of that sample . That is , given a sample we normalized this to: We normalized each component of the vectors by taking the Z-score for that component across all samples . That is , we took where is the Z-score of among the values . Heat Maps We performed heat maps on using MATLAB's clustergram function . This produces a rectangular array of squares whose rows correspond to miRNAs and whose columns correspond to samples . The color of the square denotes the relative up- or down-regulation of the miRNA in that sample . In addition , it produces dendrograms for the rows and columns which are computed using hierarchical clustering . The ordering of the rows and columns is the one most compatible with the dendrograms . PCA . We performed PCA on using MATLAB's pca function . We then produced 3-dimensional plots ( convex hulls ) by using the first 3 principal components . Resolution and convex hulls . We are able to detect if the convex bodies are resolved or disjoint by the following computation . Given a set of points , their convex hull is the set of points of the form where for each i , ai ≥0 and Put differently the convex hull of a set of points is the set of all possible weighted averages of these points . Given two sets of points , and , we can determine whether their convex hulls are disjoint by attempting to minimize the distance between and subject to the appropriate constraints . The minimum value is 0 if and only if the two convex hulls meet . We performed this constrained minimization using MATLAB's fmincon function . The solid bodies shown in Figures 8–10 are the convex hulls of the data points grouped according to cell type . In Figure 9A they resolve the four cancer types , that is to say , these four convex hulls are disjoint . We performed two in silico experiments to establish a p-value for this result . Both of these involve examining large numbers of cases and determining whether the resulting convex hulls are disjoint . In the first in silico experiment we considered the null hypothesis that the points are randomly located in the cube . We generated sets of 20 random points in the unit 3-dimensional cube . We grouped these points into groups of 6 , 6 , 3 and 5 ( similar to our cancer types ) and determined whether the resulting convex hulls are disjoint . Note that the probability of these convex hulls being disjoint is independent of the size of the cube . Out of 10 , 000 trials , 12 were disjoint giving an estimated p-value of ∼0 . 001 . In the second in silico experiment we tested the null hypothesis that it is some hidden feature of the points themselves and not their grouping into the four cancer types that is responsible for the separation of the convex hulls . To test this we used MATLAB's randperm function to randomly assign the points of Figure 9A to groups of 6 , 6 , 3 and 5 and tested whether the resulting convex hulls were disjoint . Out of 10 , 000 trials , 7 were disjoint giving an estimated p-value of ∼0 . 001 .
|
miRNAs are small ( ∼22 bp ) RNAs . They play central roles in many cellular processes . Epstein-Barr virus ( EBV ) is an important human pathogen that establishes persistent infection in nearly all humans and is associated with several common forms of cancer . To achieve persistent infection , the virus infects B cells and uses a series of discrete transcription programs to drive these B cells to become memory B cells – the site of long term persistent infection . It was the first human virus found to express miRNAs of which there are at least 40 . The functions of a few of these miRNAs are known but their expression in latently infected normal and neoplastic tissues in vivo have not been described . Here we have profiled EBV miRNAs in a wide range of infected normal and neoplastic tissue . We demonstrate that there are indeed latency program specific patterns of viral miRNA expression and that these patterns are disrupted in EBV associated tumors implicating EBV miRNAs both in long term persistence and in oncogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"oncology",
"medicine",
"infectious",
"diseases",
"viral",
"persistence",
"and",
"latency",
"virology",
"viruses",
"and",
"cancer",
"oncology",
"agents",
"biology",
"microbiology",
"viral",
"diseases",
"epstein-barr",
"virus",
"infectious",
"mononucleosis"
] |
2011
|
A Novel Persistence Associated EBV miRNA Expression Profile Is Disrupted in Neoplasia
|
The innate immune system , acting as the first line of host defense , senses and adapts to foreign challenges through complex intracellular and intercellular signaling networks . Endotoxin tolerance and priming elicited by macrophages are classic examples of the complex adaptation of innate immune cells . Upon repetitive exposures to different doses of bacterial endotoxin ( lipopolysaccharide ) or other stimulants , macrophages show either suppressed or augmented inflammatory responses compared to a single exposure to the stimulant . Endotoxin tolerance and priming are critically involved in both immune homeostasis and the pathogenesis of diverse inflammatory diseases . However , the underlying molecular mechanisms are not well understood . By means of a computational search through the parameter space of a coarse-grained three-node network with a two-stage Metropolis sampling approach , we enumerated all the network topologies that can generate priming or tolerance . We discovered three major mechanisms for priming ( pathway synergy , suppressor deactivation , activator induction ) and one for tolerance ( inhibitor persistence ) . These results not only explain existing experimental observations , but also reveal intriguing test scenarios for future experimental studies to clarify mechanisms of endotoxin priming and tolerance .
Innate immune cells such as macrophages and dendritic cells constitute the first layer of host defense . Like policemen constantly patrolling the streets for criminal activity , these cells are responsible for initiating the first attack against invading pathogens [1] , [2] . For example , using Toll-like receptor 4 ( TLR4 ) , macrophages recognize lipopolysaccharide ( LPS , also called endotoxin ) , a pathogen-associated molecular pattern ( PAMP ) that is expressed on the outer membrane of gram-negative bacteria . Within hours of stimulation , hundreds of regulatory genes , kinases , cytokines , and chemokines are activated in sequential waves , leading to a profound inflammatory and anti-microbial response in macrophages [3] . Although effective levels of inflammation require potent cytokine production , excessive or prolonged expression can be detrimental , resulting in various immune diseases , such as autoimmunity , atherosclerosis , sepsis shock and cancers [3] , [4] . Owing to this double-edged nature of innate immunity , living organisms have evolved a highly complex signaling network to fine-tune the expression of cytokines [5] . A fundamental question in this field is what kinds of network topologies and dynamics in the signaling network ensure the appropriate expression of cytokines . This question is part of a larger current theme in systems biology of the design principles of biological networks . Are there small network motifs that serve as building blocks to perform complex “information processing” functions in biological signaling networks [6]–[12] ? In this context , a systems and computational biology approach may greatly deepen our understanding in innate immunity [13]–[17] . Here we focus on the signaling motifs responsible for endotoxin priming and tolerance of macrophages . The interaction between host macrophages and bacterial endotoxin is arguably one of the most ancient and highly conserved phenomena in multi-cellular eukaryotic organisms [5] . Through TLR4 , LPS activates MyD88-dependent and MyD88-independent pathways , which eventually lead to the regulation of a number of downstream genes and pathways , including the mitogen-activated protein kinase ( MAPK ) , phosphoinositide 3-kinase ( PI3K ) , and nuclear factor κB ( NFκB ) . The integration of these intracellular pathways leads to measured induction of pro-inflammatory mediators . Intriguingly , the induction of inflammatory mediators is also finely controlled by the quantities and prior history of LPS challenges . The latter is physiologically relevant since cells are likely repetitively exposed to stimulants in their natural environment . For example , numerous in vitro studies have found that significant induction of cytokine TNF-α and IL-6 requires at least 10 ng/mL LPS in mouse peritoneal macrophages [18] , [19] and macrophage cell lines [20] , and a high dose of LPS ( 100 ng/mL ) is sufficient to trigger a catastrophic “cytokine storm” . Strikingly , however , the dose-response relationship can be reprogrammed by two successive treatments with LPS , to give either a reduced or an augmented expression of cytokines ( Figure 1A ) . In vitro , preconditioning macrophages with a high dose ( HD ) of LPS ( 10–100 ng/mL ) renders the cells much less responsive to a subsequent HD stimulation in terms of pro-inflammatory cytokine expression . This phenomenon , known as “endotoxin tolerance” or “LPS tolerance” [21] , is reported to last up to 3 weeks in vivo [22] . On the other hand , macrophages primed by a low dose ( LD ) of LPS ( 0 . 05–1 ng/mL ) show an augmented production of cytokine in response to a subsequent HD challenge , a phenomenon known as “LPS priming” [18] , [19] , [23]–[25] . Both priming and tolerance are present in other cells of the innate immune system including monocytes and fibroblasts , and are highly conserved from mice to humans . Our own studies on murine macrophages show both effects ( Figure 1B ) . Endotoxin priming and tolerance may confer significant survival advantages to higher eukaryotes . Priming of innate immune cells may enable robust and expedient defense against invading pathogens , a mechanism crudely analogous to vaccination of the adaptive immune system . On the other hand , tolerance may promote proper homeostasis following robust innate immune responses . However , despite these survival advantages , endotoxin priming and tolerance are also closely associated with the pathogenesis of both chronic and acute human diseases . For example , despite the potential ability to limit pro-inflammatory cytokine production , endotoxin tolerance is responsible for the induction of immunosuppression in patients with sepsis shock , and this suppression leads to increased incidence to secondary infections and mortality [22] . Endotoxin priming , on the other hand , reprograms macrophages to super-induction of proinflammatory cytokines . Increasing evidence relates this phenomenon to low-grade metabolic endotoxemia , where an elevated but physiological level of LPS in the host's bloodstream results in a higher incidence of insulin resistance , diabetes and atherosclerosis [26]–[29] . Augmented IL-6 expression has also been observed in human blood cells that were primed by LD and challenged by HD LPS [30] . Despite the significance and intense research efforts , molecular mechanisms responsible for endotoxin priming and tolerance are not well understood , apparently due to the complex nature of intracellular signaling networks . Tolerance has been attributed to the negative regulators at multiple levels of the TLR4 signaling pathway . These include signaling molecules ( e . g . SHIP , ST2 , induction of IRAK-M and suppression of IRAK-1 ) , transcriptional modulators ( e . g . ATF3 , p50/p50 homodimers ) , soluble factors ( e . g . IL-10 and TGFβ ) , and gene-specific chromatin modifications [21] , [31]–[38] . These negative regulators are likely to work together to drive macrophages into a transient refractory state for cytokine expression after LPS pretreatment [33] . Molecular mechanisms for priming are rarely studied and even less well understood than tolerance . Early studies suggest that like endotoxin tolerance , both intra- and inter-cellular events may be involved in LPS priming [24] . Morrison and coworkers first revealed that LPS priming of cytokine TNF-α production is induced , at least in part , by a reprogrammed counterbalance between endogenous IL-10 and IL-12 in an autocrine fashion [19] . However , it is still elusive exactly how the change in two counteracting soluble secretory products can contribute to the priming effect , and whether LPS priming is exclusively an intercellular event or it takes place at both intra- and inter-cellular levels . These published observations and our own new experimental results have inspired us to look for all possible mechanisms for LPS priming and tolerance . To do this , we computationally searched the high-dimensional parameter space associated with a generic mathematical model of a three-node regulatory network . The search reveals only three mechanisms accounting for priming ( pathway synergy , suppressor deactivation , activator induction ) and one for tolerance ( inhibitor persistence ) . Existing experimental results support these mechanisms . In summary , our approach provides a systematic , quantitative framework for understanding numerous experimental observations , and it suggests new experimental procedures to identify the players and investigate the dynamics of priming and tolerance . Our analysis suggests that endotoxin tolerance and priming are rooted in the basic structure of the immune regulatory network: a signal often triggers synergizing pathways to ensure that sufficient responses can be elicited efficiently , as well as opposing pathways to ensure that the responses can be resolved eventually [2] . Therefore , in addition to shedding light on LPS-induced tolerance and priming , our approach is applicable in the more general context of cross-priming and cross-talk in the signal transduction mechanisms of the innate immune system 39–41 .
Although separate experimental studies of priming and tolerance have been carried out in many laboratories , no systematic study of both effects has been performed in the same setting . Thus , we first set out to measure priming and tolerance in the same experimental system . We used murine bone marrow derived macrophages ( BMDM ) , which are widely used for measuring LPS responses . BMDM were treated with various combinations of LD ( 50 pg/mL ) and HD ( 100 ng/mL ) LPS for times indicated in Figure 1B . Cells were washed with PBS and fresh medium between consecutive treatments . Figure 1B shows that 50 pg/mL LPS induced negligible IL-6 , while 100 ng/mL LPS induced robust expression of IL-6 in BMDM ( ∼3300 fold ) . Consistent with previous findings , cells pre-treated for 4 h with 50 pg/mL LPS exhibited ∼4500 fold induction of IL-6 when challenged with 100 ng/mL LPS , a ∼36% augmentation as compared to cells treated with 100 ng/mL LPS alone ( p<0 . 05 ) . In contrast , cells pretreated for 4 h with 100 ng/mL LPS exhibited only ∼700 fold induction of IL-6 when re-challenged with 100 ng/mL LPS , a ∼80% reduction as compared to cells treated with 100 ng/mL LPS alone ( p<0 . 05 ) . Figure 1C shows that LPS binding to TLR4 triggers two groups of parallel pathways: MyD88-dependent and ( several ) MyD88-independent pathways . Together , these pathways control the expression of different but overlapping inflammatory mediators in a delicate time-dependent and dose-dependent manner . Based on these parallel pathways , we proposed a three-node model in Figure 1C as a minimal abstraction of the system . Each node can positively or negatively regulate the activity of itself and the other two nodes . The interactions are governed , we assume , by a standardized set of nonlinear ordinary differential equations ( Figure 1C ) for xj = activity of the jth node ( 0≤xj≤1 , j = 1 , 2 , 3 ) . For a complete description of the mathematical model , see the section on Materials and Methods . The “network topology” of the model is determined by the sign pattern of the nine interaction coefficients ( −1≤ωji≤1 , j , i = 1 , 2 , 3 ) which express the magnitude and direction of the effect of node i on node j . This is a coarse-grained model , with no distinction between intra- and inter-cellular events . For example , in a real cell the self-regulation of a node may correspond to a feedback loop involving many intermediates , including extracellular cytokines . The simplicity of the model allows full search of the 14-dimensional parameter space ( although there are 18 parameters in Table 1 , four of them are held constant , as explained in Materials and Methods ) . Similar three-node models have been studied in other contexts [6] , [42] , [43] . We searched the 14-dimensional parameter space of the model for priming and then for tolerance . The behavior of the model is defined as “priming” if the maximum level of the output variable x3 under the priming dose ( step 3 in Figure 1A ) is small ( x3<0 . 3 ) , but with the subsequent high dose ( step 4 in Figure 1A ) x3 is at least 50% higher than the level reached without priming ( step 1 in Figure 1A ) . Similarly , for “tolerance” the maximum level of x3 must be high enough under the first HD exposure ( x3>0 . 3 ) but less intense by at least 50% under the second HD challenge ( step 2 in Figure 1A ) . Precise criteria for priming and tolerance are provided in Table S1 . Brute force search of the parameter space is impractical . Unbiased searching results in <1000 parameter sets exhibiting priming after 108 Monte Carlo steps . Noticing that parameter sets giving priming or tolerance ( called “good sets” for convenience ) are clustered into a small number of isolated regions in parameter space , we designed a two-stage sampling procedure . First we perform a Metropolis search slightly biased for good sets . Next , to identify any isolated regions of parameter space where good sets are clustered , we analyzed the good sets using K-means clustering and Principal Component Analysis ( see Text S1 ) . The good sets then serve as seeds in the second stage of sampling , which restricts Metropolis searching to each local region of good sets . This two-stage procedure allows us to search the parameter space thoroughly and to obtain good-set samples that are large enough for statistical analysis . The overall procedure is illustrated schematically in Figure S1 and discussed in Text S1 . By trial-and-error , we found that the two experimentally measurable quantities , Δx1 and Δx2 ( see Figure 2A ) , are effective in dividing the “good” parameter sets into three regions ( see Figure 2B ) . Here Δx1 = maximum difference between x1 during the LD priming stage and the steady state value of x1 in the absence of any stimulus , and Δx2 = difference between the maximum values of x2 during the HD period with and without the priming pretreatment ( Figure 2A ) . Further analysis ( discussed below ) revealed that the three groups correspond to three distinct priming mechanisms: “Pathway Synergy” ( PS ) , “Activator Induction” ( AI ) , and “Suppressor Deactivation” ( SD ) . All AI and PS parameter sets show considerable increase in x2 ( >0 . 1 ) after the priming stage , while SD does not ( Figure S2 ) . To characterize these priming mechanisms , we next examined the parameter sets within each group for shared topological features . The topology of a regulatory motif is defined as the sign pattern ( + , − or 0 ) of the nine interaction coefficients , ωji , with the proviso that ωji's in the interval [−0 . 1 , 0 . 1] are set = 0 . We define a backbone motif as the simplest network topology that is shared by most of the good priming sets in each group and that is able to generate a priming effect on its own . Therefore , a backbone motif represents a core network structure in each group . Figure 3A shows that each group has its unique backbone motif ( s ) , directly revealing different priming mechanisms in each group . Figure S3 and Text S1 provide detailed statistical methods used to identify the backbone motifs . The two-dimensional parameter histograms in Figure S4 provide further support for the backbone motifs we have identified . Figure 3B–D shows typical time-courses and state-space trajectories for the three priming mechanisms ( see Table S2 for the parameter values used to generate this figure ) . Pathway Synergy ( PS ) : As shown in the upper left panel of Figure 3A , the backbone motif of PS mechanism contains both pathways through x1 and x2 activating x3 . Under a single HD , the faster pathway through x1 prevents activation of x2 , either directly or through x3 . Consequently there is no synergy between the two pathways after a single HD . With LD pretreatment , however , x2 is partially activated . During the following HD treatment , this partial activation allows x2 to increase significantly , either transiently ( Figure 3B left panel , called “monostable” ) or persistently ( Figure 3B right panel , called “bistable” ) , despite inhibition from x1 and/or x3 . Simultaneous activation of both pathways leads to synergy between them and a priming effect for x3 . Activator Induction ( AI ) : In the backbone motif ( see upper right panel of Figure 3A ) , the pathway through x1 ( with high activation threshold ) inhibits x3 , whereas the pathway through x2 ( with a low activation threshold ) activates x3 . Consequently , under a single HD , the two pathways work against each other to prevent full activation of x3 . A LD pretreatment partially activates x2 without significantly affecting x1 . Then , during the following HD treatment , x2 gets a head start on x1 to induce greater activation of x3 than observed under a single HD . The activation of x3 can be either transient ( monostable ) or persistent ( bistable ) , as illustrated in Figure 3C and Figure S5A . Suppressor Deactivation ( SD ) : In this case there are two backbone motifs slightly different from each other ( the lower panel of Figure 3A ) . Both motifs contain an inhibition pathway ( x1 ―| x3 ) with slow dynamics and low sensitivity to LPS , and an activation pathway ( x2→x3 ) with fast dynamics and high sensitivity to LPS . The basal level of the suppressor x1 is relatively high , which is typical of some suppressors ( e . g . TOLLIP , TRAILR , PI3K and nuclear receptors ) that are constitutively expressed in macrophages to prevent unwanted expression of downstream pro-inflammatory genes under non-stimulated conditions [44] , [45] . Compared to AI , in this case the LD pretreatment decreases the level of suppressor x1 , through direct inhibition of x1 by x2 . The basic SD effect is amplified either by x2 self-activation ( backbone motif I ) or by negative feedback from x3 to x1 ( backbone motif II ) . As before , the activation of x3 can be either transient ( monostable ) or persistent ( bistable ) , as illustrated in Figure 3B and Figure S5B . Each of these groups contains many different network topologies ( 187 in PS , 139 in SD , and 82 in AI ) . Taking SD as an example , Figure 4A shows the sorted density distribution of the 139 unique topologies represented by the SD parameter sets . The top 7 of these topologies ( Figure 4B ) comprise 31% of all the SD parameter sets . Consistent with other studies [6] , [43] , the most highly represented topologies contain more links than the corresponding backbone motif , indicating that additional links may increase the robustness of a network . While the two backbone motifs rank Top 27 and Top 10 respectively ( Figure 4B ) , their combination ranks Top 4 . The Venn diagram in Figure 4C shows that of the 93% of SD parameter sets that contain at least one of the two backbone motifs , 64% contain both . Notice that the two backbone motifs use different helpers to deactivate the suppressor ( x1 ) under LD , the combination of motifs ( Top 4 ) integrates both helpers so that deactivation of the suppressor can be enhanced ( Figure 4C ) . The results of a similar analysis applied to PS and AI mechanisms are given in Figure S7 . Additionally , in the Figure S8 and Text S1 , we discuss a parameter compensation effect that further expands the priming region in the parameter space . We used the 3-node model to search for endotoxin-tolerance motifs . The tolerance effect requires that pro-inflammatory cytokine expression ( x3 ) is markedly reduced ( by at least 1 . 5 fold ) under two sequential HD treatments with LPS , compared to the level induced by a single HD ( see Table S1 for details ) . Over 1660 unique topologies are found to give a tolerance effect ( Figure 5A ) , indicating that the requirements for tolerance are much lower than for priming . A typical time course ( Figure 5B , left panel ) highlights the essential dynamical requirement for tolerance — to sustain a sufficiently high level of inhibitor ( x1 in this case ) after the first HD of LPS so that x3 is less responsive to the second HD stimulus . The effect is transient: if the second HD stimulus is delayed long enough for the suppressor to return to its basal level , then the tolerance effect is lost ( Figure 5B , right panel ) . This “memory” effect has been noticed in other modeling studies [46]–[49] and is consistent with experimental observations . For example , the tolerance status of IL-6 is reported to persist for 48 h after the initial HD of LPS , but beyond this time a re-challenge started to recover the expression of IL-6 [34] . Figure 5C shows two backbone motifs that support temporary persistence of the inhibitor: by slow removal or by positive auto-regulation of the inhibitor . It is of interest to ask whether priming and tolerance can be observed in a single 3-node network given the corresponding dosing conditions . It turns out that about 11% of the priming motifs exhibit tolerance as well , and most of them belong to the SD or the AI mechanism . Figure 6A shows qualitatively the dose-response relationship for priming and tolerance in a typical network motif . First , both priming and tolerance require a relatively large second dose ( >0 . 5 ) . Second , the dosing regions for priming and tolerance are well separated . A low first dose ( 0 . 1–0 . 4 ) leads to priming while a higher one ( 0 . 5–1 ) leads to tolerance . There exists a range separating the priming and the tolerance region where neither are observed . Most experimental studies of priming and tolerance are performed with fixed durations of the three time periods ( T1 , T2 , and T3 in Figure 1A ) . Time-course measurements are rarely reported . The phase diagrams in Figure 6B & C show how varying each time period can affect the induction of priming and tolerance in a typical network motif . Altogether , these results reveal important dynamical requirement in priming and tolerance and suggest systematic studies in real biological experiments . The left panel of Figure 6B shows the effects of varying stimulus durations ( T1 and T3 ) at fixed gap duration ( T2 ) . To generate priming , T1 must be sufficiently long , while T3 can be relatively short ( left panel of Figure 6B ) . A sufficient priming duration is crucial because the system utilizes this time to activate/deactivate the regulatory pathway with slower dynamics , i . e . , the synergizing pathway in PS and the suppressor pathway in SD . Therefore , if T1 is too short , one may erroneously conclude that priming does not exist in the system . On the other hand , tolerance is less dependent on T1 ( right panel of Figure 6B ) . Figure 6C shows results when all durations are varied under the constraint T1 = T3 . In this case , both priming and tolerance require that T2 is sufficiently short compared to the time required for the system to relax to its basal state after the first stimulus . This result reveals priming and tolerance as essentially the result of cellular memory of the first stimulation .
Our in silico analysis identifies three basic mechanisms for priming ( Figure 7 ) . In these mechanisms two pathways interact either constructively ( pathway synergy–PS ) or destructively ( activator induction–AI , suppressor deactivation–SD ) . Compared to the response of these systems to a single high dose ( HD ) of LPS , a priming dose of LPS modifies the relative phases of the two pathways so as to strengthen pathway synergy ( for PS mechanism ) or weaken pathway interference ( for SD and AI mechanisms ) . In this work we define the priming effect as a response of x3 that is at least 50% higher with priming than without . The threshold of 50% is consistent with experimental observations [23] , [25] , but to be sure that our conclusions are robust , we also performed the computational analysis at two other thresholds: 30% augmentation or 70% augmentation ( i . e . , λ = 1 . 3 or λ = 1 . 7 in Table S1 ) . In both cases we obtained results similar to those shown in Figure 2B , corresponding to the three priming mechanisms , although the exact percentage of each priming mechanism among the data sets varies with the priming threshold . The priming effect may be viewed as a primitive counterpart of the more sophisticated memory mechanisms of the adaptive immune system . For a limited period of time after exposure to a weak stimulus , the system is prepared to launch a stronger response to a second exposure to the ( same or another ) stimulus [39] , [50] . On the other hand , tolerance reflects a transient refractory status to produce inflammatory cytokines due to the memory of an earlier exposure . The actual molecular and cellular networks responsible for endotoxin priming and tolerance are highly complex , involving both intra- and inter-cellular signaling modalities . A combination of priming/tolerance motifs most likely coexist in real signaling networks , and their interactions will determine the specific properties of the priming/tolerance effect in vivo . LPS is known to activate multiple intracellular pathways through TLR4 , including MyD88-dependent , TRIF-dependent pathways [51] . Cross-talk among these pathways may be differentially modulated by low vs . high dosages of LPS , and thus contribute to differential priming and tolerance [37] , [52] , [53] . Endotoxin tolerance has drawn significant attention in the past due to its relevance to septic shock . Existing literature reveals the involvement of multiple negative regulators ( SHIP , ST2 , IL-10 , IRAK-M , SOCS1 ) at either intracellular or intercellular levels . Many of them are shown to be persistently elevated during endotoxin tolerance , a key feature ( confirmed by our systems analysis ) creating a refractory state that suppresses the expression of pro-inflammatory mediators ( see Table 2 ) . For example , SHIP and ST2 are documented to have very slow degradation rates . On the other hand , negative regulators with faster turn-over rates , such as A20 and MKP1 ( induced between 2–4 h by LPS ) , are known not to be required for LPS tolerance [21] , [54] . In terms of priming , our in silico results are consistent with limited experimental data regarding potential molecular mechanisms . For example ( Figure 8A ) , IL-12 and IL-10 are differentially induced by low vs . high dose LPS , and subsequently serve as autocrine mediators to modulate LPS priming [19] . Figure 8B provides a second example . Low dose LPS ( 50 pg/mL ) can selectively activate transcription factor C/EBPδ , yet fails to activate the classic NFκB pathway [53] . Hence , by a pathway synergy motif , the selective activation of C/EBPδ by low dose LPS may synergize with NFκB under the subsequent high dose to induce the priming effect . While the removal of nuclear repressor by low dose LPS is reported [53] , further evidence for the predicted suppressor deactivation mechanism awaits additional , targeted experimentation . In this context , one needs to be aware that our predicted network motifs are simple topologies that have the potential to generate priming or tolerance , within proper parameter ranges . Our predictions warrant further experimental studies to determine the physiologically relevant ranges of signaling parameters required for priming and tolerance . Our analysis of priming and tolerance is not limited to LPS . Bagchi et al . showed that cross-priming may happen between specific TLRs [41] . Ivashkiv and coworkers reported that IFN-γ can prime macrophage for an augmented response to a variety of stimulants , including bacterial LPS , virus , IFN-α/β and IFN-γ itself [39] , [40] . IFN-γ self-priming is similar to LPS self-priming: a low dose can prime for boosted expression of interferon-responsive genes . The priming mechanism as reported by Hu et al . resembles the AI strategy [55] . Interferon-responsive genes such as IRF1 and IP-10 are transcriptionally induced by transcription factor STAT1 , and are inhibited by SOCS1 through a negative feedback mechanism . Low dose IFN-γ ( 1 U/ml ) is able to elevate the expression level of STAT1 , preparing macrophage for a boosted activation of STAT1 ( through phosphorylation and dimerization of STAT1 ) under the high dose IFN-γ stimulation . With STAT1 being active , however , the inhibitor SOCS1 cannot be expressed during the priming stage , resulting in an augmented expression of IRF-1 and IP-10 ( Figure 8C ) . Furthermore , Figure 8C suggests a possible cross-priming between IFN-γ and TLR4 via a PS mechanism . Priming of macrophage by a low dose IFN-γ promotes STAT1 expression , which may synergistically cooperate with NFκB to give boosted cytokine expression to secondary stimulation by LPS [55] , [56] . Further experimental studies are needed to confirm the prediction . Three-node models have been used to analyze functional network motifs in several contexts [6] , [7] , [43] . The simplicity of three-node models allows a thorough search of the parameter space . However , the model should be viewed as a minimal system . A typical biochemical network surely has more than three nodes . Therefore each node or link in the three-node model is normally coarse-grained from more complex networks . The model parameters are also composite quantities . Three-node models are limited in their ability to generate certain dynamic features such as time delays . Figure 3A shows the backbone motifs of the three mechanisms we have identified . Further studies of models with additional nodes will be necessary to determine whether all of the links are necessary . For example , in Figure 8B , we cannot find evidence for IL-6 inhibiting C/EBPδ ( either by direct or indirect links ) . This lack of evidence may indicate a missing link waiting for experimental confirmation , or it may indicate a limitation of the three-node model . The parameter search algorithm developed in this work can be applied to models with 4 or more nodes , although the search space grows rapidly with the number of nodes . Despite the above-mentioned limitations , we expect that the three priming mechanisms and the one tolerance mechanism discovered here are quite general , holding beyond the three-node model . We expect that the present work can serve as a basis for analyzing larger networks with more mechanistic details . As illustrated in Figure 8 , motifs can be combined together in series or in parallel , and these combined structures may lead to new dynamic properties of functional importance . Our analysis in Figure 6 suggests that systematic studies of signal durations ( T1 , T2 and T3 ) may reveal important details of the dynamics of priming and tolerance . For example , both relatively short ( 4 h , as the experiment in this paper ) and longer priming duration ( ≥20 h ) are exhibit priming effects in macrophages [25] . Relatively fast transcriptional regulators like NFκB and AP-1 , as well as numerous signaling repressors such as PI3K and nuclear receptors , may be involved in intracellular priming motifs , inducing priming in response to short pretreatments . On the other hand , a longer pretreatment orchestrates more complex intercellular pathways whereby autocrine or paracrine signaling of cytokines ( e . g . IL-10 , IL-12 and type I IFNs ) might dominate the induction of priming effects [19] . Therefore , measurements of the full time spectrum are necessary to reveal different parts of the network contributing to priming/tolerance . Furthermore , our analysis predicts that priming networks may respond in two distinct fashions: monostable ( transient super-induction of cytokine ) or bistable ( sustained super-induction of cytokines ) . Time-course measurements can distinguish between these two responses , keeping in mind that the bistable behavior predicted here is relative to the effective time-scale of the model . Each motif considered here is embedded in a larger network . Eventually , in a healthy organism pro-inflammatory cytokines have to be cleared out by some other slow processes that resolve the inflammation . On this longer time scale , the sustained induction of cytokines predicted by some of our models would be resolved . The analysis presented in Figure 2B suggests a plausible hypothesis to characterize underlying mechanisms of endotoxin priming . High throughput techniques can be used to identify genes and proteins that are significantly changed by low dose pretreatment . Likely candidates can be assayed during the course of a priming experiment , and the time-course data analyzed as in Figure 2B to identify the critical regulatory factors . Our analyses and simulations reveal that the priming effect is quite sensitive to system dynamics , i . e . , to parameter values and initial conditions . It is well documented that many biological control systems , especially those involving gene expression , are stochastic in nature . Consequently a population of seemingly identical cells may respond heterogeneously to a fixed experimental protocol . In this case , single-cell measurements may reveal cell-to-cell variations in priming and tolerance responses [57]–[59] . Taken together , our integrated and systems analyses reconcile the intriguing paradigm of priming and tolerance in monocytes and macrophages . Given the significance and prevalence of this paradigm in immune cells to diverse stimulants other than LPS , our identified functional motifs will serve as potential guidance for future experimental works related to macrophage polarization as well as dynamic balance of immune homeostasis and pathogenesis of inflammatory diseases .
The following mathematical formalism is used to describe the dynamics of the three-node system , where , and . Notice that lies between 0 and 1 for all t . All variables and parameters are dimensionless . is a generic “sigmoidal” function with steepness ( slope at Wj = 0 ) that increases with σj . Each ωji is a real number in [-1 , 1] with its absolute value denoting the strength of the regulation; ωji>0 for the “activators” and ωji<0 for “inhibitors” of node j . The sum , Wj , is the net activation or inhibition on node j , and ωj0 determines whether node j is “on” or “off” when all input signals are 0 . The parameters γj determine how quickly each variable approaches its goal value , G ( σjWj ) for the present value of Wj . Because the magnitudes of the weights are bounded , |ωji|<1 , it is possible to do a thorough and systematic search of all possible weight matrices , even for networks of moderate complexity , e . g . , K ( = number of non-zero ωji's ) <20 . The formalism is close to that used by Vohradsky [60] , [61] and others [62] , [63] previously . More detailed discussions and applications of the formalism can be found in [64]–[66] . The model contains 18 parameters: 9 ωji's , 3 γj's , 3 σj's and 3 ωj0's . By setting , we fix the time scale of the model to be the response time of the output variable , . We set , so that the response variable is close to in the absence of input . We also chose as a moderate value for the sigmoidicity of the output response . Apart from that , is set to be so that the x2 pathway is responsive to LD stimulation . Our goal is to sample points in a 14-dimensional parameter space that is bounded and continuous . The sampling algorithm needs to search the parameter space thoroughly and generate sample parameter sets that are statistically unbiased and significant . Our strategy is a random walk based on the Metropolis Algorithm [67] through parameter space according to the following rules: We pursue this strategy in two stages . In stage 1 , we set ( see Text S1 ) , so that the random walk has larger tendency to stay in “good” regions of parameter space , but can also jump out of a good region and searches randomly until it falls into another good region ( which may be the same region it left ) . Stage 1 generates a random walk of 109 steps , which is sampled every 100 steps . From this sample of 107 parameter sets only the good ones are saved , giving a sample of ∼ good parameter sets . These data are then analyzed as described below: Stage 2 is a repeat of stage 1 with ρ = 0 . In this case the random walk never leaves a good region . The purpose of stage 3 is to generate a large sample of good parameter sets that may occupy different regions of parameter space . The random walks are sampled every 100 steps , generating 106 good parameter sets from each starting point . Each parameter set must pass an additional test for “biological relevance” ( see Text S1 for details ) before further analysis . While the results reported in the main text are from one run of the search procedure , the whole procedure was repeated several times with random initial starting point in stage 1 . The final results of these repeated runs agree with each other , confirming the convergence of our search procedure . In order to analyze the topological feature of each priming/tolerance mechanism , one needs to map the continuous parameters ωji into a discretized topological matrix τji . In the topological space , variables are only described by ( − , 0 , + ) representing inhibition , no regulation and activation , respectively . A cut off value ( = 0 . 1 ) is used to perform the discretization , following the rules below: Murine bone marrow derived macrophages from C57BL/6 wild type mice were harvested as described previously [53] . Cells were cultured in DMEM medium ( Invitrogen ) supplemented with 100 units/mL penicillin , 100 µg/mL streptomycin , 2 mM l-glutamine , and 10% fetal bovine serum ( Hyclone ) in a humidified incubator with 5% CO2 at 37°C . Cells were treated with LPS ( E . coli 0111:B4 , Sigma ) as indicated in the figure legend . RNAs were harvested using Trizol reagent ( Invitrogen ) as previously described [53] . Quantitative real-time reverse-transcription ( RT ) -PCR were performed as described [68] . The relative levels of IL-6 message were calculated using the ΔΔCt method , using GAPDH as the internal control . The relative levels of mRNA from the untreated samples were adjusted to 1 and served as the basal control value .
|
Inflammation is a fundamental response of animals to pathogen invasion . Among the first responders are macrophage cells , which identify and respond to multiple challenges . Their responses must be carefully regulated to kill invading pathogens without causing too much damage to host cells . Excessive activity of macrophages is associated with serious diseases like sclerosis and cancer . Macrophage responses are governed by a complex signaling network that receives cues , integrates information , implements appropriate responses and communicates with neighboring cells . This network must maintain a short-term memory of pathogen exposure . Endotoxin priming is an example . If macrophages are exposed to a small dose of bacterial toxins , they are primed to respond strongly to a second exposure to a large dose of toxin . Endotoxin tolerance , on the other hand , refers to the fact that macrophages are resistant to endotoxin challenges after a large dose pretreatment . The precise molecular mechanisms of both priming and tolerance are still poorly understood . Through computational systems biology , we have identified basic regulatory motifs for priming and for tolerance . Using information from databases and the literature , we have identified molecules that may contribute to priming and tolerance effects . Our methods are generally applicable to other types of cellular responses .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cytokines",
"statistical",
"mechanics",
"immunity",
"to",
"infections",
"immunology",
"algorithms",
"mathematics",
"inflammation",
"theoretical",
"biology",
"biology",
"nonlinear",
"dynamics",
"immune",
"response",
"immune",
"system",
"physics",
"systems",
"biology",
"computer",
"science",
"computer",
"modeling",
"immunity",
"innate",
"immunity",
"computational",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
Network Topologies and Dynamics Leading to Endotoxin Tolerance and Priming in Innate Immune Cells
|
Sleep is a fundamental biological process conserved across the animal kingdom . The study of how sleep regulatory networks are conserved is needed to better understand sleep across evolution . We present a detailed description of a sleep state in adult zebrafish characterized by reversible periods of immobility , increased arousal threshold , and place preference . Rest deprivation using gentle electrical stimulation is followed by a sleep rebound , indicating homeostatic regulation . In contrast to mammals and similarly to birds , light suppresses sleep in zebrafish , with no evidence for a sleep rebound . We also identify a null mutation in the sole receptor for the wake-promoting neuropeptide hypocretin ( orexin ) in zebrafish . Fish lacking this receptor demonstrate short and fragmented sleep in the dark , in striking contrast to the excessive sleepiness and cataplexy of narcolepsy in mammals . Consistent with this observation , we find that the hypocretin receptor does not colocalize with known major wake-promoting monoaminergic and cholinergic cell groups in the zebrafish . Instead , it colocalizes with large populations of GABAergic neurons , including a subpopulation of Adra2a-positive GABAergic cells in the anterior hypothalamic area , neurons that could assume a sleep modulatory role . Our study validates the use of zebrafish for the study of sleep and indicates molecular diversity in sleep regulatory networks across vertebrates .
The function of sleep is disputed , with hypotheses ranging from energy conservation to synaptic remodeling and memory consolidation , with the possibility of disparate functions across evolution [1 , 2] . One approach to this question is to study sleep and sleep regulatory networks in organisms amendable to molecular , neuronatomical , and genetic studies [3–5] . Using purely behavioral criteria , a sleep-like state has been demonstrated in non-mammalian species [6 , 7] . A sleep-like state has been characterized in Drosophila melanogaster [4] , initially using behavioral criteria , and more recently through electrophysiological studies [8] . Identification and characterization of sleep mutants is ongoing [5] . While unquestionably a superb genetic model organism , the phylogenetic distance between Drosophila and mammals has produced notable and relevant divergences in usage of neuromodulators . Whereas tyramine and octopamine are critically important in Drosophila , they are trace amines of unknown function in mammals and other vertebrates [9] . Further , certain other neurotransmitter systems have not been identified in Drosophila , including hypocretin/orexin , an important sleep modulator . Hypocretins [10] , also called orexins [11] , are neuropeptides of primary interest in the study of sleep . Indeed , they compose the only neurochemical system known to be involved in the generation of a clear human sleep disorder phenotype , narcolepsy [12–17] . In mammals , preprohypocretin is primarily expressed by neurons of the posterior hypothalamus , with widespread projections to the brain and spinal cord [18] . Two closely related receptors are known , with complementary neuronatomical distribution [11 , 19] . Of notable importance for sleep regulation , mammalian hypocretin neurons also project widely and activate monoaminergic cell groups , which are generally known to be wake-active ( adrenergic , dopaminergic , serotoninergic , and histaminergic ) [20–23] . Further , they also project and activate cholinergic cell groups [24–26] important for the regulation of wakefulness and rapid eye movement ( REM ) sleep . Intracerebroventricular ( icv ) injections of hypocretin are potently wake-promoting and increase locomotion in mammals [27–29] , an effect partially blocked by histaminergic [28] and dopaminergic [30] antagonists , with higher doses inducing stereotypies similar to those observed after massive dopaminergic stimulation [30] . These experiments suggest that hypocretin is a major sleep modulator , and that much of its activity is mediated through direct stimulation of aminergic systems . Narcolepsy can be observed in multiple mammalian species . In humans , the disorder is primarily due to hypocretin cell loss [14 , 16] . It is characterized by excessive daytime sleepiness , disturbed nocturnal sleep ( inability to stay asleep at night ) , cataplexy ( transient muscle paralysis triggered by emotions ) , sleep paralysis ( transient muscle paralysis during sleep–wake transitions ) , and hypnagogic hallucinations ( dream-like experiences at sleep onset ) [31 , 32] . Many of the symptoms of narcolepsy are related to an abnormal tendency to enter REM sleep rapidly , and to express abnormal REM sleep features while awake ( e . g . , muscle paralysis ) or dreaming ( hypnagogic hallucinations ) . Contrary to popular belief , narcolepsy is not characterized by increased total sleep time over the 24-h period [33 , 34] . Rather , patients with narcolepsy are unable to consolidate either wakefulness or sleep [33 , 35] . In dogs , narcolepsy has been studied for over 20 years and has been shown to display all the symptoms of human narcolepsy [36] . It can result from mutations in the hypocretin receptor-2 ( HCRTR2 ) gene ( familial cases ) [12] or from hypocretin deficiency ( sporadic narcolepsy ) [37] . In rats and mice that have been engineered to lack hypocretin , hypocretin cells , or HCRTR2 , narcolepsy manifests as sleep and wake fragmentation during the day and the night , sudden episodes of muscle weakness , and rapid transitions from wake to REM sleep [15 , 38 , 39] . In contrast , hypocretin receptor-1 ( HCRTR1 ) knockout mice do not display sudden episodes of muscle weakness , but have sleep abnormalities [39 , 40] . These comparisons highlight the similarity of the phenotype across mammalian species and a primary role for HCRTR2-mediated transmission . The zebrafish is a powerful genetic model that has , as a vertebrate , the advantage of sharing a similar central nervous system organization with mammals [41 , 42] . We , and others , have shown that principal actors of sleep regulation in mammals are largely conserved in zebrafish , including monoaminergic [43–46] , cholinergic [47] , and hypocretinergic ( orexin ) [48–50] cell groups . In addition to conservation of cell groups , responses to various hypnotic drugs such as H1 histaminergic antagonists , melatonin agonists , alpha-2 adrenergic agonists , and GABAergic modulators are also conserved in the zebrafish [51–53] . A sleep-like state was first characterized in 7- to 14-d-old zebrafish larvae by Zhdanova and colleagues , who demonstrated an elevated arousal threshold , a reduced breathing respiratory rate , and a compensatory rest rebound in response to deprivation as well as modulation by hypnotic drugs [51 , 54] . More recent studies have also explored sleep architecture in zebrafish larvae [50] . These findings raise the question of whether conserved neuromodulators function in the same roles to regulate sleep in fish as in mammals , or whether there are presently unappreciated divergences . The study of fish sleep is also interesting because the fish is a cold-blooded vertebrate , and mammalian-like sleep architecture , as defined by REM sleep and non-REM sleep , typically emerged with homeothermy [55] . To address these questions , we performed a detailed characterization of sleep in adult zebrafish and characterized sleep abnormalities in adult zebrafish lacking the sole hypocretin receptor .
We first performed a fine analysis of the characteristics and architecture of the sleep-like state in fully formed adults to develop a working definition of sleep in the zebrafish . We used videotracking of fish illuminated by an infrared source under light and dark conditions ( adult fish sleep recording system [AFSRS]; Figure 1A; see also Materials and Methods ) , and found that zebrafish adults are strongly diurnal , displaying higher activity during the day ( Figure 1B and 1C ) . Brief periods of inactivity , often associated with a drooping caudal fin , were observed , suggesting a sleep-like state ( Video S1 ) . These periods of inactivity alternated with active periods during both the night and the day ( Figure 1B ) . Further , the state was easily reversible by gentle tapping , acoustic stimulation , or weak electrical field stimulation ( e . g . , 2–6 V ) . One important characteristic of sleep is increased arousal threshold . To study this phenomenon , the reaction of fish to a weak electrical field of variable strength was studied ( Video S2 ) . Random voltage stimuli ( 0–2 V ) were applied every 30 min through the day and night ( Figure 1D ) . We noted that fish in an active state were more likely to respond to very low voltage stimuli than those that were inactive ( see legend to Figure 1 for details ) . At higher voltages , all fish responded regardless of activity state ( Figure 1E ) . Inactivity was thus associated with an increased arousal threshold to electrical stimulation , with the greatest differential response observed at 0 . 38 V . Sub-analyses performed during the day versus the night , and in individual animals , yielded similar results ( data not shown ) . We next defined the minimum epoch of immobility distinguishing sleep from simple inactivity . To do so , we used receiver operator curve ( ROC ) analysis [56] of the results of the electrical stimulation experiments . In this analysis , sensitivity ( SE ) and specificity ( SP ) ( and Kappas ) for response to stimulation in inactive versus active states are plotted as the discrimination voltage threshold is varied . A true positive is defined as a demonstration of sleep , as defined by immobility and non-response to stimulation . SE is defined as the percent of non-response to stimulation when inactive/total number of trials when inactive . SP is defined as the percent of response when active/total number of trials when active . These analyses considered the percentage of responses to differing voltages ( 0 . 13 , 0 . 38 , 0 . 5 , 1 and 2 V ) as well as the consecutive number of seconds ( 0–30 s ) of inactivity ( <6 pixels/s ) preceding the stimulus . These SE/SP pairs are plotted on ROC and qualitative ROC planes . Diagnostic lines are drawn on the qualitative ROC plane , and the ideal test point is plotted . This analysis indicated that the best SE/SP ratio ( point closest to ideal test point and closest to the diagnostic line ) was obtained when using 0 . 38 V and 6 s of prior inactivity . As expected , SE increased with increasingly longer periods of inactivity , as the longer the prior period of inactivity was , the likelier true sleep ( without reactivity to electrical stimulation ) was observed ( Figure 1F ) . In contrast , SP decreased with increasingly longer periods of inactivity , as more and more short periods of true sleep were missed and considered “active . ” This analysis provided a working definition of zebrafish sleep: an interval of inactivity ( <6 pixels/s ) lasting at least 6 s ( Figure 1E and 1F ) . All other periods were defined as active ( awake ) . Using this definition , we determined that most sleep episodes occurred at the bottom or the top of the tanks ( Figure 1G ) and were remarkably consolidated during the night ( Figure 1H and 1I ) . No sex differences were found in sleep amounts or distribution ( data not shown ) . We next investigated whether sleep episodes are homeostatically regulated by observing sleep-deprived animals . To do this , we first attempted sleep deprivation by tapping on the aquarium walls or using noise introduced through an underwater speaker . As we noted rapid habituation , electrical stimulation was next attempted . Although this procedure was also imperfect , as we found it extremely difficult to keep the fish awake in the dark for long periods of time ( Video S4 ) , it was retained as the method of choice , as it did not result in rapid habituation . We next designed a computerized system to electrically stimulate a fish each time it displayed sleep behavior ( Figure 2A ) . A yoked control fish was stimulated concurrently at the same voltage ( though not necessarily while resting ) in a separate chamber to control for stress . Increased voltage from 2 to 6 V was applied to both fish if the inactive fish did not react to stimulations ( Figure 2A ) . Animals were sleep deprived during the 6 h of the dark prior to usual light onset ( 9 a . m . ) , and released into either the usual light ( 150 lux ) or an extended period of darkness . This procedure successfully induced sleep deprivation , although partial habituation was observed after 4 h of stimulation , i . e . , toward the end of the procedure ( data not shown ) . Indeed , sleep-deprived animals appeared increasingly immobile and unreactive to stimulation toward the end of the procedure , in contrast to yoked control stimulated fish ( Videos S3 and S4 ) . Further , as sleep episodes normally occur at high frequency at night , random stimulation in the control fish also induced mild partial sleep deprivation ( while controlling for stress ) . Overall , we found that this procedure induced a 30% decrease in sleep in the sleep-deprived group versus undisturbed controls . In contrast , a 10% decrease in sleep was observed in the yoked control group , representing a milder degree of sleep deprivation ( Figure 2B ) and providing a dose-response curve of increasing amounts of sleep deprivation . After release into an extended period of darkness during the subjective day , sleep in undisturbed control animals was lowest . In yoked control animals ( partially deprived ) , minor recovery sleep was observed , while a significant rebound was observed in the sleep-deprived animals , indicating homeostatic regulation of sleep . Differences were statistically significant between sleep-deprived versus yoke control stimulated or undisturbed fish ( Figure 2B ) . Sleep bout length was also increased in both the yoke control and sleep-deprived groups , although not significantly in the yoke control , compared to the undisturbed group . Remarkably , a sleep rebound was not observed when sleep-deprived animals were released into light ( Figure 2B ) . Further , when fish were exposed to 150-lux light during the last 6 h of the biological night , but not electrically stimulated , there was a dramatic suppression of sleep ( 90% decrease ) that was not followed by a rebound when animals were released into darkness ( Figure 2B ) . A similar , nearly complete suppression of sleep was also observed when animals were kept under constant light conditions for 3 d ( Table 1 ) . Again , no significant rebound was observed during the day or the following nights ( data not shown ) . During longer exposure to constant 150-lux light , a progressive return of sleep was noted over a period of 1–2 wk ( Figure S1 ) . As previously reported for activity [57] , sleep was modulated by circadian influences under constant light and dark conditions ( Table 1 ) . Unlike most mammals , however , the direct effect of dark and , more strikingly , light was stronger than circadian influences . Indeed , for most parameters , values varied significantly more with light exposure than with circadian timing ( Table 1 ) . To investigate functional conservation of neurotransmitters regulating sleep in zebrafish , we next anatomically and functionally studied the hypocretin system , the only system known to cause a primary sleep disorder ( narcolepsy ) in mammals . In conjunction with previous work on the hypocretin neuropeptide [48] , we identified a single hypocretin receptor in Tetraodon and in zebrafish ( hcrtr [also known as hcrtr2] ) through bacterial artificial chromosome ( BAC ) library screening and in silico database searches ( see Materials and Methods ) . As recently noted [50] , zebrafish Hcrtr has higher homology to mammalian HCRTR2 , the subtype of primary importance in the mediation of the narcolepsy phenotype [12] . As in mammals [19] , we found widespread hcrtr expression in the telencephalon , hypothalamus , hypophysis , posterior tuberculum , and hindbrain ( Figure 3A and 3B ) and in selected spinal cord neurons ( Figure 3C and 3D ) of larvae at age 2 d postfertilization ( dpf ) . Limited expression was found in thalamic and pallidal areas , reminiscent of overall mammalian HCRTR1 and HCRTR2 mRNA distribution ( cortex and hippocampus , basal forebrain , central midline thalamic areas , hypothalamus , and brainstem ) [19] , although overall neuroanatomical correspondence of structures between these species is only partially established [41] . We next simultaneously mapped the distribution of hcrtr with that of monoaminergic cell groups . In mammals , monoaminergic cell groups modulate wakefulness and are among the most hypocretin-receptor-rich brain regions [19] . These are stimulated by hypocretins and are commonly assumed to mediate much of the downstream effects of hypocretin on sleep regulation [12 , 58] . Interestingly , however , using double in situ hybridization ( ISH ) on 2-dpf larvae , we saw no significant colocalization with adrenergic ( Figure 3G and 3J ) , dopaminergic ( Figure 3E–3S ) , histaminergic ( Figure 3T ) , or serotoninergic ( Figure 3U and 3V ) neurons . Flat mounts and close-ups confirmed these results ( Figure 3H–3J ) . Double fluorescence ISH followed by confocal microscopy ( Figure 3K–3S ) also demonstrated an absence of colocalization in dopaminergic and adrenergic cells , in contrast to a previous report [50] . To determine whether connectivity of the hypocretin and monoaminergic systems emerges later in development , we also performed ISH on adult zebrafish brain sections , using the adult zebrafish atlas established by Wulliman and colleagues [42] . The embryonic distribution of hcrtr was broadly maintained in the adult brain , with prominent localization in the telencephalon , hypothalamus , posterior tuberculum , hypophysis , and brainstem cranial nuclei . In addition to the areas of detected expression in 2-dpf embryos , notable expression was also observed in the periventricular gray zone of the optic tectum ( Figure 4A–4C ) . As in embryos , colocalization of hcrtr with monoaminergic cells was absent except for a few anterior dopaminergic neurons ( Figure 4G ) . Most notably , and unlike previously reported [50] , expression was absent in the large majority of diencephalic dopaminergic neurons ( Figure 4B , 4E , and 4H ) and in the locus coeruleus ( Figure 4C , 4F , and 4I ) . In this last area , however , a few receptor-positive cells were present immediately medially to the locus coeruleus ( Figure 4I ) . Labeling of the locus coeruleus was performed using dbh ( Figure 4I ) , th , and adra2a ( data not shown ) . As hcrtr is not significantly colocalized with major monoaminergic neurons in embryos , we next surveyed zebrafish embryos with other neuronal markers ( acetylcholine , GABA , glycine , and glutamate markers ) that have been proposed as hypocretin targets in various sleep regulatory models [2 , 58] . We found that Adra2a , Gad67 ( also known as Gad1 ) , ChAT , and Glyt2 ( also known as Slc6a5 ) were expressed in regions similar to where hcrtr was expressed ( Figure 5 ) . Double ISH further confirmed that most hcrtr-positive cells were Gad67-positive GABAergic cells ( Figure 5A ) , except in the hypophysis and the ventral posterior tuberculum . A subpopulation of hcrtr-positive GABAergic cells in the anterior hypothalamic area was also positive for Adra2a ( Figure 5B ) . Some overlap was also observed between the cholinergic system and hcrtr-positive cells in the diencephalon and in rhombomere 2 ( Figure 5C ) . In the spinal cord , hcrtr-positive neurons were neither primary sensory neurons nor motoneurons , but were located closer to the primary sensory neuron layer , a region that could be equivalent to lamina-II in mammals; this area is involved in the secondary processing of sensory information such as pain [59] . Most of these neurons were glycinergic ( Figure 5D ) and GABAergic ( Figure 5E ) . In mammals , equivalent neurons receive dense hypocretin projections and are stimulated by the peptide , with a role in the modulation of nociceptive input [59–62] . In a second analysis , double ISH data extended in adults , with a primary focus on the hypothalamic area . As in embryos , we found that many hcrtr-positive cells were Gad67-positive . In the anterior hypothalamus and ventral thalamic nucleus ( Figure 6 , first two columns ) , most hcrtr-positive cells were GABAergic , starting at the diencephalic–telencephalic junction . Further , the majority of the anterior hypothalamic GABAergic cluster was Adra2a-positive . In the posterior diencephalon , only a small region of the central posterior thalamic nucleus was Adra2a- and Gad67-positive ( Figure 6 , last column ) . Studies using cholinergic markers were next performed ( Figure 7 ) . Our primary focus was on the telencephalon and the pons , where cholinergic cells equivalent to sleep regulatory neurons of the nucleus basalis and laterodorsal tegmentum/pedunculopontine nuclei have been reported . Cholinergic staining was abundant in the diencephalon and rhombencephalon , including in many cranial nerve nuclei . Colocalization with hcrtr was only observed in a few areas , most notably in the peripheral gray zone of the optic tectum and periventricular hypothalamus ( Figure 7B and 7C ) . In the telencephalon , we failed to detect any cholinergic neurons ( Figure 7A ) ; this population has been found only in some fishes . Similarly , close to the locus coeruleus , where the equivalent of the laterodorsal tegmentum/pedunculopontine cells are believed to be located , no ChAT expression was noted ( Figure 7D–7G ) . We next screened an ethylnitrosurea-mutagenized TILLING ( for “targeting induced local lesions in genomes” ) library for hcrtr mutations . A premature stop codon mutation ( R168 to stop ) was identified ( hcrtr168 ) that results in the predicted loss of four transmembrane domains as well as the intracellular loop 3 domain required for G-protein coupling . The truncation is also located upstream of two mutations known to produce an inactive protein resulting in canine narcolepsy [12] ( Figure 8 ) . Homozygous hcrtr168 animals developed normally into viable and fertile adults . Extensive observation of larvae and adults did not yield any obvious phenotype , such as the occurrence of sudden REM-sleep-like paralysis episodes ( e . g . , cataplexy or cataplexy-like behaviors ) characteristic of mammalian narcolepsy [12 , 14 , 31 , 63] , either spontaneously or when excited by food administration or mating . Similarly , activity monitoring did not reveal any large differences between mutant , heterozygous , and wild-type larvae of similar background ( data not shown ) . We next studied adult wild-type and mutant fish using our AFSRS ( including comparison of heterozygous and wild-type siblings within the same family ) under typical light/dark conditions . We found that activity of hcrtr168 mutants was slightly increased ( Figure 9A; Table 1 ) and sleep amounts were decreased by 20%–30% during the night ( Figure 9B; Table 1 ) . Most strikingly , fine architecture analysis revealed a 60%–70% increase in the number of sleep–wake transitions , and a 60% decrease in sleep bout length during the night , indicating sleep fragmentation in hcrtr168 ( Figure 9D and 9E ) . Heterozygous animals generally behaved as wild-type siblings , although in some measures ( e . g . , sleep time and sleep transitions ) , an intermediary phenotype was observed ( Table 1 ) . Activity and sleep architecture were normal during the day in all genotypes . Further , wake bout length was essentially unchanged during the day or the night ( Figure 9C; Table 1 ) . These data indicate that the hypocretin receptor is required for proper sleep regulation in adult zebrafish under light/dark conditions . Studies under constant light and dark indicated significantly decreased sleep amounts and significant sleep fragmentation in hcrtr168 compared to wild-type animals at all circadian time points , but these effects were masked by the stimulating effect of light ( Table 1 ) . Similar increases in locomotion ( and decreased sleep ) were observed in all three genotypes when animals were newly moved from their usual aquaria to the recording chambers; thus , disruption of nocturnal sleep in the mutant was unlikely to be due to differential effects of stress or the food deprivation associated with our monitoring method ( Figure S2 ) . Further , food intake satiety monitoring studies ( Figure S3 ) and studies of locomotor activation after feeding were performed , and all genotypes reacted similarly ( Figure S4 ) . Hypocretin-1 icv injections are wake-promoting and increase locomotion in mice , rats , and dogs [27 , 28 , 30] . In contrast , hypocretin-2 is generally inactive because of rapid catabolism [64] . Prior to usual light–dark transition time , adult zebrafish were briefly anesthetized , and hypocretin peptides ( or saline ) were injected icv . Animals were subsequently released in the dark while activity and sleep were measured using the AFSRS . In controls , locomotion was high ( novel environment ) , followed by habituation and reduced activity/increased sleep in the dark ( Figure 10A and 10B ) . In hypocretin-1-injected fish , a reduction in locomotor activity was observed ( Figure 10A and 10B ) . This effect was dose dependent and occurred with either the human or zebrafish hypocretin-1 peptide . Based on sleep scoring of these data , we found that both mammalian and zebrafish hypocretin-1 significantly increased total sleep time ( 23 . 0% ± 11 . 3% and 28 . 7% ± 9 . 0% above baseline sleep , respectively , p < 0 . 05 in both cases after 1 , 400 pmol ) during 9 h of continuous recording ( Figure 10 ) . As expected , zebrafish hypocretin-2 was inactive in wild-type zebrafish ( data not shown ) . In the TILLING hcrtr168 mutant , the sleep-promoting effects of mammalian or zebrafish hypocretin-1 were abolished ( Figure 10C ) . Overall , these experiments indicate antagonism of the sleep-promoting effects of icv hypocretin-1 injection by hcrtr168 , confirming that the effect is mediated through Hcrtr .
Our experiments demonstrate that rest episodes in adult zebrafish represent a genuine sleep-like state , characterized by reversible periods of immobility , place preference ( bottom or surface ) , circadian regulation , and homeostatic rebound . Interestingly , unlike in larvae [50 , 51 , 54] , sleep deprivation was difficult to achieve in adults and was associated with a sleep rebound that was only detectable when the fish were released in the dark . Our study is unique as we studied adults and demonstrated that only 6 s of prior inactivity was sufficient to be associated with decreased arousal threshold and thus to qualify as sleep . Other studies have been primarily performed in larval zebrafish and did not test or report on intervals shorter than 1 min [50 , 51] . In agreement with our finding , a preliminary report in adults reported that long-term sleep deprivation using a moving partition technique or electric shock produced a sleep rebound , associated with increased arousal threshold [54] . Unlike in most mammals , we also found that even moderate levels of light exposure have strong sleep-suppressant effects in zebrafish , and that circadian regulation has a more minor role . The light suppressant effect was not associated with deleterious behavioral effects over a week , but sleep reappeared progressively after 8 d ( Figure S1 ) . These results are in agreement with data from Hurd and colleagues [57] , who found that only a portion of adult fish displayed detectable circadian activity rhythms under constant light or dark conditions at 28 °C , in all cases with significantly lower amplitude than under alternating light/dark conditions . Most strikingly , we also found that light was not only able to suppress sleep ( Figures 2 and S1 ) but that no sleep rebound was observed upon release in the dark . In goldfish and perch , a sleep rebound in the light has been found after sleep deprivation by light , but was mild [7] . Further , the lack of rebound after sleep deprivation by light in our experiment contrasts with the observation of homeostatic regulation after a much lower level of sleep deprivation using electrical stimulation . Overall , whereas it is likely that homeostatic regulation of rest can be demonstrated in some circumstances ( in our case , electrical stimulation when sleeping in the dark ) , we found that wake induced by light in zebrafish was not , on a short-term basis , associated with a mounting sleep debt . How could this unusual effect of light be explained ? Unlike in mammals , but as reported in Drosophila , most zebrafish cells are directly photoreceptive [65 , 66] . Further , melatonin is a strong hypnotic in this species , a property that may be related to the diurnal pattern of activity of zebrafish [51] . The combined effect of light on various cell populations , together with its suppressive effects on melatonin production , may result in multiple redundant wake-promoting inputs into the brain . In favor of this hypothesis , variable effects of light have been observed in other teleosts , where it acts to suppress rest and induce rest rebound in the perch and goldfish [7] , both diurnal fish , whereas it has calming effects in nocturnal fish such as the tench [6] . In this context , the strong effects of light or melatonin may be able to overcome the more minor regulatory effects of other neural networks regulating sleep homeostasis in zebrafish . Recent results in other species , most notably in diurnal birds , indicate that some vertebrates have sleep regulatory characteristics similar to those of zebrafish . Birds are especially interesting as , unlike fish , it is possible to document all the electroencephalographic characteristics of mammalian sleep [6] . Migratory sparrows , for example , are able to survive for long periods of time without sleep under selected ecological conditions and are extremely sensitive to light and dark [67] . Similarly , sleep in pigeons is strongly suppressed by light , without electroencephalogram-defined non-REM sleep rebound in darkness [68–70] . Like zebrafish , diurnal birds such as pigeons are also remarkably sensitive to melatonin , and do not exhibit wake rebound after melatonin-induced sleep [71] . It may thus be that the need for homeostatic regulation of sleep has not strongly evolved in zebrafish , and that it is not as universal in vertebrates as previously believed . Rather , in both diurnal birds and fish , the direct effect of light or melatonin may be able to bypass homeostatic regulation of sleep . Further studies of sleep deprivation by light versus other methods in these species may reveal molecular mechanisms regulating sleep homeostasis . Our studies have also shown significant and informative divergence in the organization of the hypocretin system in zebrafish . We have described a small group of approximately 20 hypocretin cells in the preoptic hypothalamus of embryonic zebrafish and characterized a compact promoter driving expression in these cells [48] . ISH and immunochemistry in adult brains indicates that approximately 40 cells are present in the adult zebrafish preoptic area ( data not shown and Kaslin et al . [49] ) , although an additional more anterior group , probably detected through antibody cross-reactivity , was found using immunochemistry with mammalian antibodies [48 , 49] . The hcrt cluster in zebrafish is distal to the histaminergic cell group expressing histidine decarboxylase , unlike in mammals where these two cells groups are closely adjacent within the posterior hypothalamus . Unlike mammals , the zebrafish has only one hypocretin receptor . This result , surprising when considering the frequency of gene duplications in this species , was confirmed through in silico searches , BAC library screening , genomic Southern blot analysis , and comparisons of syntenies around HCRTR1 and HCRTR2 in mice , humans , and zebrafish . Indeed , only a single hypocretin receptor is identifiable in current releases of other teleosts ( zebrafish , Fugu , Tetraodon , medaka , and stickleback ) . Using ISH , we found that the expression pattern of hcrtr is in agreement with overall mammalian hypocretin receptor expression patterns ( Figures 3 and 4 ) . Indeed , the high density of hypocretin receptor mRNA in the telencephalon , hypothalamus , posterior tuberculum , and hindbrain , but not lateral thalamic and pallidal areas , is reminiscent of overall mammalian HCRTR1 and HCRTR2 distribution and density ( in cortex , hippocampus , basal forebrain , central midline thalamic areas , and hypothalamus ) [19] , although neuroanatomical correspondence of overall structure across these species is only partially established [41] . Similarly , as in mammals , hcrtr is expressed in glycinergic/GABAergic neurons of the spinal cord immediately ventral to sensory neurons . In mammals , these neurons receive dense hypocretin projections and are stimulated by the peptide , with a role in the modulation of nociceptive input [59–62] . Although the overall pattern of expression initially appeared similar to that in mammals , our in-depth analysis indicates differences in expression in relevant sleep regulatory networks ( see below ) . As in mammals , mutation of the hypocretin receptor disrupts the consolidation of sleep/wake behavior . Unlike narcoleptic humans [14 , 31] , canines [12] , and rodents [63] , however , hcrtr168 fish do not display sudden episodes of paralysis ( cataplexy ) , which in mammals are believed to represent dissociated REM sleep . Further , hcrtr168 mutant fish do not have decreased wake bout length , whether in the light or the dark . This contrasts with human narcolepsy , where the primary abnormality is an inability to consolidate long periods of wakefulness during the day . Instead , the phenotype of hcrtr168 is only sleep fragmentation and decreased sleep in darkness ( e . g . , insomnia ) , another disabling symptom of human narcolepsy [31 , 32] . Day and night sleep and wake fragmentation , together with episodes of muscle paralysis , are also a primary feature of rodent and canine narcolepsy [12 , 36 , 63] . One explanation for this discrepancy is likely to be the strong and direct effect of light in inhibiting sleep in zebrafish during the day , as described above . Indeed , the phenotype of hcrtr168 is not apparent under constant light ( Table 1 ) . The direct effect of light in this species may thus have made it unnecessary to evolve a method to consolidate wake , and indeed there was great variability in daytime wake bout length among fish of all genotypes in the absence of light ( Table 1 ) . The lack of wake abnormality in hcrtr168 mutants during the day or the night is also surprising considering recent results indicating increased activity and decreased sleep in zebrafish larvae after heat shock stimulation of hypocretin expression [50] . To address this issue and to test for the specificity of the effects of high doses of hypocretin ( as generated in heat shock overexpressing models ) , we studied the effects of icv injections of mammalian and zebrafish hypocretin-1 at various doses on sleep and locomotion in wild-type zebrafish adults . Mammalian hypocretin-1 is conserved in the functionally important final six C-terminal amino acids of hypocretin-1 and −2 [48 , 49 , 72] needed for binding activity , and thus must be active , as reported for Xenopus hypocretin [73] and in goldfish experiments . Prober et al . [50] found that heat shock activation of hypocretin expression increased locomotion and wake for several days in larvae , although without altering diurnal fluctuations in activity . In our icv injection experiments , we found that zebrafish or mammalian hypocretin-1 , but not zebrafish hypocretin-2 , reduces activity and promotes sleep in the dark ( Figure 10A ) . These effects were abolished in the null hcrtr168 mutant , indicating mediation by Hcrtr . The relatively mild reducing effect on locomotion contrasts with the strong locomotor activation and wake-promoting effects of hypocretin-1 reported in mammals . In rats , for example , 8 , 000 pmol dramatically increases activity [28 , 30] and sustains wake for 5 h , an effect followed by a proportional sleep rebound lasting 10 h [29] . Overall , although it is difficult to reject the possibility that the increased sleep seen in our experiments after icv hypocretin-1 injection is merely a rebound after a primary wake-promoting effect ( an effect that could have been masked by the 10 min of anesthesia and subsequent activation in a novel environment ) , we believe this hypothesis to be unlikely . Indeed , a very minor increase in locomotion has been reported in goldfish after icv injection of mammalian hypocretin-1 [74] , without further follow-up of activity monitoring longer than 1 h after administration . In our experiments , it is difficult to conceive that our injected zebrafish would be activated while in or recovering from anesthesia . Further , even in awake rodents placed in a novel environment , icv hypocretin-1 injection increases locomotion . We thus believe that in zebrafish , unlike in mammals , hypocretin is mildly sedative and certainly not strongly stimulant . Other differences from mammalian physiology include the postulated role of hypocretin in integrating metabolic input in mice [75] , as we did not find differences in locomotor activation after fasting . Environmental modulators of hypocretin release ( e . g . , stress , locomotion , feeding , and fasting ) , as measured using cerebrospinal fluid hypocretin-1 , have also been shown to have differential effects between mammals , possibly reflecting ecological differences in the need for hypocretin to regulate sleep or wake under specific conditions [29] . Together with the mutant data , we therefore favor the hypothesis that hypocretin is a more minor sleep regulatory molecule in zebrafish than in mammals , with mostly sleep-promoting effects in the dark . It is also possible that in mammals , hypocretin has a dual effect on promoting sleep at night and wake during the day , thus explaining insomnia and daytime sleepiness in its pathology . In this case , only sleep-promoting projections of hypocretin would be common to zebrafish and mammals . The differential expression pattern of the sole hypocretin receptor in zebrafish versus those in mammals may explain the mild hcrtr168 insomnia phenotype and the lack of stimulatory effect of hypocretin . We have shown absence of Hcrtr on serotoninergic cells of the raphe nuclei and histaminergic cells . Moreover , unlike in a previous study [50] , we were unable to find hcrtr expression in all diencephalic dopaminergic cells or in the locus coeruleus of 2-dpf embryos ( Figure 3 ) , a result we extended to adult brains ( Figure 4 ) . In mammals , the histaminergic tuberomammillary nucleus , the serotoninergic raphe nuclei , and the adrenergic locus coeruleus are among the densest regions containing hypocretin receptor mRNA [19] . In zebrafish , we previously noted branching of Hcrt neuron projections in the ventral area of the rhombomere 1–2 boundary [48] , but believe this projection to be consistent with the strong hcrtr expression within rhombomere 2 rather than locus coeruleus ( rhombomere 1 ) ( Figure 3 ) . Immunocytochemical experiments have suggested hypocretin projections to these monoaminergic systems in adult zebrafish , but they used mammalian anti-hypocretin that is cross-reactive , as reflected by a mismatch between ISH and immunocytochemical staining patterns of cell bodies [48–50] . Further , in a study using a hypocretin-GFP line , direct contact between hypocretin terminals and monoaminergic cell bodies was not evident [50] , unlike in mammals [20] . Another major regulator of sleep and REM sleep in mammals is the cholinergic system , most notably through projections of the basal forebrain ( regulating wake ) and the pons ( regulating wake and REM sleep ) to the cortex and thalamus , respectively [2 , 24 , 58] . This system is also known to be excited by hypocretin and to be hypocretin-receptor-positive in mammals [24 , 25] . Although data in larvae were suggestive of a possible colocalization in the diencephalon and rhombencephalon ( Figure 5 ) , we found that in adults , areas believed to be equivalent to pontine and basal forebrain cholinergic cell groups in fish—the lateral part of the ventral telencephalon and the peri-locus coeruleus areas , respectively[76 , 77]—no ChAT transcripts were detected . The lack of strong hypocretin input on sleep-related monoaminergic and cholinergic cell groups likely explains the weaker effect of hypocretin on wake consolidation . We have postulated that one of the key functions of hypocretin may be to extend wakefulness in the face of a mounting sleep debt [78] . In mammals , hypocretins strongly innervate the histaminergic cells of the tuberomammillary region [19 , 28] , and the most important mechanism of action of hypocretin on wakefulness may be through the action of histamine on excitatory histamine H1 receptors [28 , 79] . We have recently demonstrated that the wake-promoting action of the H1 receptor is conserved in zebrafish larvae , where strong a hypnotic effect results from treatment with the H1 inverse agonist mepyramine [52] . The absence of hypocretin receptors on histaminergic cells in zebrafish is consistent with an absence of a hypersomnia phenotype in hcrtr mutant animals . Similarly , hypocretin innervation of the locus coeruleus in mammals has been suggested to control cataplexy and muscle atonia during REM sleep [80] . Although speculative at this stage , one pattern of expression , the presence of Hcrtr on a small population of Adra2a-positive , GABAergic cells of the anterior hypothalamic area , may be of relevance to the mediation of short sleep and nighttime sleep fragmentation in hcrtr168 . Indeed , this area may be equivalent to mammalian preoptic/basal forebrain GABAergic , Adra2a-positive cell groups ( including the median and ventrolateral preoptic nuclei and basal forebrain ) known to be primarily active during non-REM sleep [58 , 81] . A loss of hypocretin receptor in this cluster would be predicted to decrease the activity of this sleep-promoting area , in turn decreasing sleep . Interestingly , the equivalent region in mammals , or at least the ventrolateral preoptic nuclei , seems mostly nonresponsive to hypocretin [58 , 82 , 83] . In conclusion , we report on the unique characteristics of sleep regulation and of the hypocretin system in zebrafish . Our data offer intriguing parallels with and surprising divergences from mammalian sleep physiology . Not only did we find that light has a profound effect on sleep , but it abolished the need for short-term homeostatic rebound . Further , in this species , hypocretin is a milder sleep modulator , possibly primarily consolidating sleep rather than wake . This is likely explained by the lack of hypocretin stimulation of monoaminergic and cholinergic systems and a proportionally stronger input on sleep-promoting systems of the hypothalamus and basal forebrain . The need for hypocretin innervation of wake-promoting structures may have evolved later , as the importance of the direct effects of light and melatonin on brain activity decreased , and the need to consolidate wake independently of light effects evolved . It is possible that , as for circadian biology , neural networks regulating behaviors ( e . g . , clock output networks that must vary between nocturnal and diurnal animals ) are less well conserved than core genetic actors to be discovered for the regulation of sleep . Further studies in zebrafish , a poikilotherm vertebrate , together with parallel work in birds and monotremes , may help decipher how sleep regulatory network organization evolved prior to the emergence of homeothermy and the REM/non-REM dichotomy [55 , 84–86] .
Young adult ( 6 mo ) wild-type zebrafish ( Scientific Hatcheries , http://www . scientifichatcheries . com/ ) were used for the sleep characterization . Zebrafish were raised and maintained in Marine Biotech ( http://www . marinebiotech . com/ ) Zmod systems ( 28 . 5 °C , pH 7 . 0 , conductivity = 500 μS ) in a 14 h: 10 h light/dark cycle . The hypocretin receptor knockout hcrtr168 ( originally named hu2098 ) was identified through genomic screening of the hcrtr locus in F1 ethylnitrosurea-mutagenized animals ( TILLING ) as part of the ZF-MODELS Integrated Project in the Sixth Framework Program ( http://www . zf-models . org/ ) . The mutation at codon 168 results in an arginine ( AGA ) to stop ( TGA ) alteration . Heterozygous animals were obtained and out-crossed twice to wild-type ( Scientific Hatcheries ) . Wild-type , heterozygous , and homozygous siblings ( larvae and 6 mo old ) were used for the primary comparison of the three genotypes . As no significant differences were observed across all wild-types , the wild-type categories were pooled in some analyses . The entire AFSRS is kept insulated from light in a sealed black box . Under lights-on conditions , illuminance is fixed at 150 lux , as measured on the water surface ( Tensor VisionMax , 13-W fluorescent light , Tensor Corporation ) . Fish recording chambers consist of four individual cells ( 400 ml each ) . Adult fish are freely swimming ( three dimensional movement ) in vertically and horizontally stacked chambers . System water is circulated at 60 ml/min ( Pondmaster magnetic drive utility pump , Danner Manufacturing , http://www . dannermfg . com/ ) and maintained under controlled conditions ( temperature = 28 . 5 ± 1 °C , conductivity = 500 μS , pH 7 . 0 ) . Filtration ( Penguin 100 , Bio-Wheel power filter , AQUARIA , Marineland , http://www . marineland . com/ ) and heating ( Ebo-Jager 100 Watt , Eheim , http://www . eheim . de/ ) are also provided . Infrared backlighting ( C47192–880 , Advanced Illumination , http://www . advancedillumination . com/ ) is provided during recording . Video images ( 320 pixels × 240 pixels ) are obtained using Sony ( http://www . sony . com/ ) DCR-HC85 or Hamamatsu ( http://www . hamamatsu . com/ ) Photonics C2741–60 cameras equipped with an infrared filter ( B+W 093 , Hamamatsu ) . Images are captured with a frame grabber ( Dazzle Creator80 , Pinnacle Systems , http://www . pinnaclesys . com/ ) , recorded ( VirtualDub for video image , http://www . virtualdub . org/ ) , and processed using a modified algorithm of SleepWatch provided by Irina Zhdanova ( Boston University ) . In our recording conditions , one pixel is equal to 0 . 6 mm ( 100 pixels for one cell ) . Data were further analyzed using Mat Lab ( MathWorks , http://www . mathworks . com/ ) . To quantify fish activity , digital infrared images are recorded ( 30 frames/s ) through backlit chambers , and processed using software extensively modified from SleepWatch . This software first identifies the fish and its barycenter . Successive barycenters are calculated every 1/8th second , with difference in position representing the distance traveled . Total pixels traveled each second is also calculated , representing an activity measurement expressed in pixels/second . Based on video observation , a value below 6 pixels/s was considered as background and reflected an immobile or slightly drifting floating fish . For the electrical stimulation experiments , recording chambers were fitted with stainless steel sidewalls . Stimuli were generated using a NI PCI-6723 ( National Instruments , http://www . ni . com/ ) controlled by LabView software ( National Instruments ) . Water conductivity was fixed at 550 μS . For these experiments , distance traveled and the total surface of the fish image ( pixels ) were calculated frame by frame at a rate of 30 frame/s . Responses were blindly analyzed in video clips both computationally ( calculation of change in velocity , or change in fish image surface , reflecting change in direction of movement ) and through visual scoring of behavioral response . Responses were scored as present or absent . Mammalian hypocretin-1 was obtained through Phoenix Pharmaceutical ( http://www . phoenixpharm . com/ ) . Mature zebrafish hypocretin-1 and −2 were synthesized from predicted sequence by Synthetic Biomolecules ( http://www . syntheticbiomolecules . com/ ) . Adult zebrafish skulls were drilled in the midline at the telencephalon–diencephalon border 5 h prior icv injection . After this surgery , fish recovered in water at standard temperature and conditions . One minute before injection , fish were anesthetized using 0 . 15 mg/ml tricaine ( note that at this concentration fish recover quickly , within a few minutes , when transferred to aquarium water ) . Then 0 . 5 μl of hypocretin peptide solution ( concentrations: 0 . 28 and 2 . 8 nmol/μl ) or saline buffer was injected into the diencephalic and tectal ventricals through this drill hole with a micromanipulator and a glass capillary needle connected to a microinjector ( PLI 90 , Harvard Apparatus , http://www . harvardapparatus . com/ ) . Fish were then released in the recording chambers , and monitoring was initiated 5–10 min after injection . Injection quality was checked by coinjecting hypocretin with a rhodamine dye and by inspection of dissected injected brains the following day . Of note , none of the fish died or appeared abnormal in the following days to weeks after injections . Data are reported as means ± standard error of the mean or percent , where most appropriate . For signal detection analysis , ROC and quantitative ROC analysis were performed as described in Kraemer [56] and above . Comparisions of arousal threshold reactivity ratios between active and asleep animals were performed using Chi2 . To examine the effect of sleep deprivation , ANOVAs with grouping factors were used , followed by t-tests to examine post hoc effects . For icv injections , repeated measure MANOVAs with increasing doses of hypocretin-1 as grouping factors were used , with log 10 transformation of hypocretin dose . A fragment of the hcrtr gene of Tetraodon nigroviridis was identified through a BLAST search of the GenBank Genome Survey Sequences Database ( http://www . ncbi . nlm . nih . gov/dbGSS/; accession AL310684 ) . The corresponding clone ( 52F08 ) was obtained from Genoscope ( http://www . cns . fr/ ) , and the exons of the gene were identified through sequencing . hcrtr-containing BAC clones were first identified using the Genome System BAC library ( Incyte , http://www . incyte . com/ ) and a Tetraodon hcrtr exon 2 probe at low stringency 48 °C ( 20B8 , 20A23 , 31L8 , 59B13 , and 128J8 ) . No evidence for locus heterogeneity was evident after fingerprinting and Southern blotting . Further screening using a hcrtr-specific probe identified the following clones from the CHORI 211 BAC library ( http://bacpac . chori . org/ ) : L80L23 , 16D22 , 103I5 , and 99C14 . Zebrafish exons were identified through sequencing of subclones , and confirmed through reverse transcriptase PCR . The 5′ end of the gene was characterized through 5′ RACE ( Generacer , Invitrogen , http://www . invitrogen . com/ ) . Whole mount ISH was performed as previously described [87] . In some cases , stained embryos were embedded in 1% agarose ( Invitrogen ) and cut with a vibratome ( series 1000 , Sectioning System ) to verify colocalization or to allow better visibility of expression patterns . ISH was also performed on adult brain tissue . In this case , brains were excised and fixed in 8% PFA for 48 h , sectioned by vibratome , and stained as free-floating slices according to the procedure above . The following genes/probes were used: adra2a , chat , dat , dbh , ddc , gad65 , gad67 , glyt2 , hcrtr , hdc , npy , pomca , pyy , tph1 , tph2 , th , vglut1 , vglut2a , and vglut2b .
The NCBI ( http://www . ncbi . nlm . nih . gov/ ) accession number for zebrafish hcrtr is EF150847 . The NCBI ( http://www . ncbi . nlm . nih . gov/ ) or Ensembl ( http://www . ensembl . org/index . html ) accession numbers for the genes/probes used for ISH are adra2a ( NM_207637 ) , chat ( ENSDARG00000015854 ) , dat ( AF318177 ) , dbh ( ENSDARG00000058086 ) , ddc ( ENSDARG00000016494 ) , gad65 ( AF017265 ) , gad67 ( AB183390 ) , glyt2 ( Ab183389 ) , hcrtr ( EF150847 ) , hdc ( EF150846 ) , npy ( NM_001007218 ) , pomca ( ENSDARG00000043135 ) , pyy ( NM_131016 ) , tph1 ( NM_178306 ) , tph2 ( NM_214795 ) , th ( AF075384 ) , vglut1 ( AB183385 ) , vglut2a ( AB1386 ) , and vglut2b ( AB183387 ) .
|
Sleep disorders are common and poorly understood . Further , how and why the brain generates sleep is the object of intense speculations . In this study , we demonstrate that a bony fish used for genetic studies sleeps and that a molecule , hypocretin , involved in causing narcolepsy , is conserved . In humans , narcolepsy is a sleep disorder associated with sleepiness , abnormal dreaming , and paralysis and insomnia . We generated a mutant fish in which the hypocretin system was disrupted . Intriguingly , this fish sleep mutant does not display sleepiness or paralysis but has a 30% reduction of its sleep time at night and a 60% decrease in sleep bout length compared with non-mutant fish . We also studied the relationships between the hypocretin system and other sleep regulatory brain systems in zebrafish and found differences in expression patterns in the brain that may explain the differences in behavior . Our study illustrates how a sleep regulatory system may have evolved across vertebrate phylogeny . Zebrafish , a powerful genetic model that has the advantage of transparency to study neuronal networks in vivo , can be used to study sleep .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"neurological",
"disorders",
"neuroscience",
"genetics",
"and",
"genomics"
] |
2007
|
Characterization of Sleep in Zebrafish and Insomnia in Hypocretin Receptor Mutants
|
In 2000 , we investigated the Rift Valley fever ( RVF ) outbreak on the Arabian Peninsula—the first outside Africa—and the risk of nosocomial transmission . In a cross-sectional design , during the peak of the epidemic at its epicenter , we found four ( 0 . 6% ) of 703 healthcare workers ( HCWs ) IgM seropositive but all with only community-associated exposures . Standard precautions are sufficient for HCWs exposed to known RVF patients , in contrast to other viral hemorrhagic fevers ( VHF ) such as Ebola virus disease ( EVD ) in which the route of transmission differs . Suspected VHF in which the etiology is uncertain should be initially managed with the most cautious infection control measures .
Rift Valley fever ( RVF ) is a zoonotic disease caused by an RNA virus in the genus Phlebovirus , family Bunyaviridae . RVF virus is transmitted to humans primarily by mosquito bites and by direct contact with infected animal body fluids [1] . First described in Kenya in 1910 , the disease has been recognized in many African countries with a severity ranging from localized , well controlled clusters to major epizootics and associated epidemics [2] . In August 2000 , the first confirmed occurrence of RVF outside the African continent was described on the Arabian Peninsula along the Red Sea coast in southwestern Saudi Arabia and Yemen . This outbreak illustrated that the RVF virus can adapt to different ecological conditions and cause infection in humans and domestic ungulates , provided suitable mosquito vectors and animal reservoirs are present . Although most acute RVF virus infections result in a nonspecific febrile illness , the virus is hepatotrophic and associated with hepatitis , and a concomitant nephropathy has been described [3] . In addition , 1% of cases develop hemorrhagic complications and up to 50% of these may result in death . Encephalitis may occur in 1% or more of cases 1 to 4 weeks after the acute illness resolves ( Available via CDC at: http://www . cdc . gov/vhf/rvf/RVF-FactSheet . pdf; available via WHO at: http://www . who . int/mediacentre/factsheets/fs207/en/ ) [4] . During the first 4 weeks after recovery , as many as 15% of cases may result in ocular complications , such as retinitis , and up to 50% may have permanent vision loss [5–7] . Person-to-person transmission has not been described , but laboratory workers are known to be at risk for RVF virus infection possibly , via aerosolization [4] . Human infection readily occurs from contact with infected animal blood and amniotic fluid , in which RVF virus has been reported to reach titers of 1010 virions per ml [8] . Similar titers , 108 among infected humans , who may develop frank hemorrhage , have suggested the possibility that direct person-to-person transmission may occur [9] . However , the true risk to health-care workers ( HCWs ) for acquiring RVF in the hospital setting is unknown . To estimate the magnitude of such a risk , we undertook a descriptive observational cross-sectional study to evaluate nosocomial acquisition of RVF in Jazan , where protective measures were promulgated to hospitals admitting RVF cases .
The study was conducted under the auspices of the Ministry of Health and Field Epidemiology Training Program , Kingdom of Saudi Arabia and with the assistance of CDC as an outbreak response related activity . In addition , we obtained visiting country equivalent institutional review board ( IRB ) approval for a clinical trial of ribavirin for RVF as an adjunct to this study–all part of the overall RVF outbreak response . The risk to HCWs for acquiring RVF in the hospital setting was assessed at four hospitals in the Jazan province–where the outbreak began—during October 22–26 , 2000 , which corresponded to the end of the peak of the outbreak ( three months after it began in August 2000 ) : King Fahad Central Hospital ( KFCH ) , Samtah General Hospital ( SGH ) , Al Ardah Hospital ( AH ) , and Beash Hospital ( BH ) . KFCH was the regional referral hospital , whereas the others were located in the hyperendemic areas . The study was begun approximately three months into the RVF outbreak in Jazan , when on average 50 to 75 new cases were being reported on a weekly basis . From August to October , a total of approximately 400 RVF patients were hospitalized at these four facilities . We were not able to obtain information on how many required intensive care unit admission or had severe manifestations , but these likely represented the minority , given what is known about the natural history of most RVF infections . A cross-sectional cohort from each hospital was selected of approximately 50–150 HCWs who were in close contact with 10 or more RVF patients , their body fluids , or other potentially infectious materials with RVF virus . These cohorts were composed of individuals such as laboratory technicians , phlebotomists , morgue workers , physicians , nurses , orderlies , and cleaning staff working in areas of the hospital with RVF patients and/or potentially contaminated materials ( “high-risk group” ) . Such persons were most likely to have bloodborne pathogen exposure via needlestick , mucosal splashes , or large mucocutaneous exposures to blood or tissue from infected individuals . Another set of cohorts of approximately 50–150 HCWs who did not have much , if any , such exposure was also chosen from the same hospitals; this group included individuals working primarily in the areas of pediatrics , obstetrics and gynecology , psychiatry , pharmacy , social work , and hospital administration ( “low-risk group” ) . Trained interviewers administered a questionnaire to the HCWs in both groups to collect information about their demographics , level and type of hospital exposures , precautionary measures , and possible environmental exposures . All enrolled HCWs were assigned a unique code number designed to assist efforts to accurately identify the specimens and ensure confidentiality . A blood sample ( 5 ml ) was taken from each participant to test for IgM and IgG antibody to RVF virus , using an enzyme-linked immunosorbent assay ( ELISA ) . Antibodies to RVF virus were detected by using both IgM and IgG assays and inactivated RVF virus antigens . Both were done by using methods previously described for Ebola virus [10] . Briefly , the IgM assay was performed in a Mu-capture format using RVF antigen grown in Vero E6 cells and inactivated by gamma irradiation and a hyperimmune anti-RVF mouse ascitic fluid as the detection system for bound antigen . The IgG assay employed a detergent-extracted RVF antigen grown in Vero E6 cells and inactivated by Gamma irradiation; antigen was adsorbed directly onto microtiter plates . Both IgG and IgM assays were performed using mock-infected Vero E6 antigens prepared in the same manner , respectively . Sera were tested at dilutions from 1:100 to 1:6400 in 4-fold dilutions . Samples were considered positive for the IgM assay if 1 ) the sum of the adjusted optical densities from all of the dilutions ( infected antigen less the mock infected antigen ) was greater than 0 . 45 through the entire dilution series and 2 ) the titer was 1:400 . Samples were likewise considered positive in the IgG assay if 1 ) the sum for the adjusted optical densities from all of the dilutions ( infected antigen less the mock infected antigen ) was greater than 0 . 95 through the entire dilution series and 2 ) the titer was 1:400 . The IgM-capture assay employed goat anti-human Mu to capture IgM ( Biosource , Camarilla , CA ) and a horseradish peroxidase conjugated goat anti-mouse IgG from Biosource in the RVF antigen detection system . The IgG assay used a horseradish peroxidase conjugated mouse anti-human Gamma-chain-specific antibody ( Accurate Chemical , Westbury , NY ) to detect bound IgG . These tests were performed at the National Polio Laboratory ( Riyadh , Kingdom of Saudi Arabia ) and confirmed by the Centers for Disease Control and Prevention ( Atlanta , GA , USA ) . Evidence of infection during the epidemic was defined as any individual in the cohort with detectable IgM and IgG antibody to RVF virus . This assessment was performed as a public health emergency declared by the Saudi Ministry of Health and carried out at their urgent request for institutional assistance to assess infection control practices .
A total of 703 HCWs participated in this study . Three hundred and forty-six ( 49% ) were males and the mean age was 33 years ( range: 20–64 years; standard deviation: ± 9 years ) . The most common nationalities included Indians ( 37% ) , Saudi Arabians ( 26% ) , and Filipinos ( 12 . 5% ) . Two hundred sixty-six ( 37 . 8% ) were from KFCH and 240 ( 34 . 1% ) were from SGH , where precautionary measures , such as the use of gloves , gowns , and face masks , were widely implemented . However , of the remaining HCWs , one hundred eleven ( 15 . 8% ) were from AH and 86 ( 12 . 2% ) were from BH , where the use of protective measures was less common . By occupation , 80 ( 11% ) were physicians , 312 ( 44 . 6% ) were nurses , 43 ( 6 . 2% ) were laboratory technicians , 115 ( 16 . 5% ) were cleaners , and 153 ( 21 . 7% ) had other jobs . With respect to community exposure , 74 ( 10 . 7% ) of the 703 participants reported direct contact with animals; among these , 15 ( 20 . 3% ) reported that they slaughtered animals between August and October 2000 . Of the 703 , 42 and 57 ( 6 . 0 and 8 . 1% ) reported a history of exposure to animal abortions or deaths , respectively , around their residence . Mosquitoes were reportedly present at the residence of 347 ( 49% ) participants , but only 242 ( 35% ) participants reported having had mosquito bites . Two hundred sixteen ( 29% ) participants reported contact with 10 or more RVF patients . Twenty ( 2 . 9% ) of the 685 HCWs who completed this item on the questionnaire reported having needlesticks from suspected or confirmed RVF patients within the prior 2 months; up to 57 ( 9% ) reported some unprotected exposure to various body fluids ( e . g . blood , urine , feces , spinal fluid , and saliva ) from suspected or confirmed RVF patients during that same period . Unfortunately , HCW did not recall details of the severity of percutaneous or mucosal exposure to blood or tissue from RVF patients , as parameters such as volume and viral load of occupational exposure has been reported to increase the risk of transmission with other bloodborne viruses such as HIV . Despite these potentially “high-risk” nosocomial exposures , none of these HCWs were found to have evidence of infection with RVF virus . With respect to the self-reported protective measures employed by HCWs , 72 . 1% reported always wearing gloves , 68% face masks , and 60 . 8% gowns when they dealt with suspected or confirmed RVF patients , body fluids , or material ( Table 1 ) . Four ( 0 . 6% ) of 703 HCWs had evidence of recent RVF virus infection as indicated by IgM positivity . All four reported no known or no contact with patients with suspected or confirmed RVF; all reported no needlestick exposure or direct contact with body fluids from RVF patients . All RVF antibody positive HCWs were in the “low-risk group . ” Three worked at AH and one at KFCH . Two were orderlies who worked in the medical ward and intensive care unit , 1 was a security guard , and 1 was a clerk . Three reported having a febrile illness in the past 2 months . All four RVF-antibody-positive HCWs reported having mosquitoes at their place of residence; the number of bites ranged from “sometimes” to “frequently . ” All three HCWs from AH reported close contact with animals; of these , two also reported exposure to dead or aborted animals . Of note , Al Ardah was the area where the first RVF cases and the greatest number of cases in Jazan were reported , and where a 90% antibody prevalence ratio was identified among animals in a survey done in this area [2] . None of the HCWs were IgG positive at the time of the study . Moreover , despite 40% of staff not using contact precautions and the 100% not following airborne/droplet precautions , RVF seroconversion did not occur . Therefore , we infer that standard precautions would suffice in managing RVF patients .
Serological evidence suggests that only four ( 0 . 6% ) of the 703 HCWs were infected by RVF virus . Our data suggest that these infections were probably the result of community exposure rather than nosocomial acquisition . Nosocomial transmission , if it occurs , appears to be very rare in the context of at least rudimentary standard precautions . These data suggest that the risk for hospital-acquired RVF in HCWs is very low and that the use of standard precautions alone afford sufficient protection to HCWs who deal with known RVF patients . Strengths of this study included high-risk and low-risk groups being well-defined and data acquired during the end of the peak of the epidemic , within its epicenter , at a regional referral hospital and 3 hospitals in the surrounding hyper-endemic community in order to minimize ascertainment bias . The study target size was large and extrapolated from the number of cases occurring per week at the time of the study in the Jazan province . However , with the recent Ebola virus disease ( EVD ) epidemic of 2014 –the largest to date , centered in West Africa–caution is advisable in those with suspected viral hemorrhagic fever ( VHF ) . [11] VHF syndromes among the various etiologies can overlap , though the infectiousness and route of transmission may differ considerably . EVD and RVF exemplify this . This EVD outbreak appears to have been caused by Zaire ebola virus , originally identified as the etiology in the 1976 Democratic Republic of the Congo ( DRC ) outbreak that had a case fatality proportion of approximately 90% , and similarly began in rural forest communities [12–14] . In contrast , the 2014 epidemic has had a case fatality proportion of approximately 60–70% . While standard , contact , and droplet precautions are recommended for management of hospitalized patients with known or suspected EVD ( Available at: http://www . cdc . gov/vhf/ebola/hcp/infection-prevention-and-control-recommendations . html ) [15] , our study found that standard precautions alone may suffice for RVF . Given the established high nosocomial transmissibility via body fluids of EVD , while personal protective equipment ( PPE ) for EVD is extensive , including single-use ( disposable ) fluid-resistant gowns that extend to at least mid-calf with single-use “double gloving” and full face shield/facemask ( Available at: http://www . cdc . gov/vhf/ebola/healthcare-us/ppe/guidance . html ) , such PPE appears unnecessary for RVF based upon our study findings . Since the geographical and syndromic distribution of EVD and RVF may coincide , particularly with the frequency of air travel , isolation of individuals suspected to have symptoms of VHF of undetermined etiology seems prudent , adopting the most conservative barrier methods pending the etiologic diagnosis . Co-infection of EVD and RVF is also a possibility . Fortunately , bunyaviruses such as RVF virus and filoviruses such as Ebola are sufficiently different that current serodiagnostic methods should have a high discriminatory index , unlike alphaviruses and flaviviruses . [16] Future studies on rapid diagnostics that shorten the pre-patent in relation to the incubation period and are etiology specific will be invaluable to curb the spread of these deadly VHFs .
|
Rift Valley fever ( RVF ) is a zoonosis that is primarily transmitted to humans via infected mosquito bites . Although most acute RVF infection results in a nonspecific febrile illness , 1% of cases develop hemorrhage and 50% of these may result in death . Person-to-person transmission has not been described , but high viral titers have been observed during hemorrhagic complications , suggesting the potential for direct person-to-person transmission . This article describes an estimated risk of RVF nosocomial transmission during an outbreak setting and contrasts the suggested infection control precautions to that of other viral hemorrhagic fevers ( VHF ) such as Ebola virus disease ( EVD ) .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
The Risk of Nosocomial Transmission of Rift Valley Fever
|
Memory formation is a highly complex and dynamic process . It consists of different phases , which depend on various neuronal and molecular mechanisms . In adult Drosophila it was shown that memory formation after aversive Pavlovian conditioning includes—besides other forms—a labile short-term component that consolidates within hours to a longer-lasting memory . Accordingly , memory formation requires the timely controlled action of different neuronal circuits , neurotransmitters , neuromodulators and molecules that were initially identified by classical forward genetic approaches . Compared to adult Drosophila , memory formation was only sporadically analyzed at its larval stage . Here we deconstruct the larval mnemonic organization after aversive olfactory conditioning . We show that after odor-high salt conditioning larvae form two parallel memory phases; a short lasting component that depends on cyclic adenosine 3’5’-monophosphate ( cAMP ) signaling and synapsin gene function . In addition , we show for the first time for Drosophila larvae an anesthesia resistant component , which relies on radish and bruchpilot gene function , protein kinase C activity , requires presynaptic output of mushroom body Kenyon cells and dopamine function . Given the numerical simplicity of the larval nervous system this work offers a unique prospect for studying memory formation of defined specifications , at full-brain scope with single-cell , and single-synapse resolution .
Experience leaves traces of memory in the nervous system . This assists organisms to predict and adapt to events in their environment . Both invertebrates and vertebrates possess a variety of different learning mechanisms [1 , 2] . Associative learning , for instance , enables animals to draw on past experience to predict the occurrence of food , predators or social partners [3] . Several studies in vertebrates and invertebrates have revealed that associative memories consist of distinct phases , which differ in duration and time of expression . Throughout the animal kingdom , a labile , short-term memory can be distinguished from a robust , long-term memory [4–6] . Long-term memory—in contrast to short-term memory—is resistant to anesthetic disruption and depends on consolidation processes including de novo protein synthesis [4 , 6–9] . Genetic studies in adult Drosophila following olfactory classical conditioning using electric shock as a negative reinforcer have identified distinct temporal memory phases—short-term memory ( STM ) , middle-term memory ( MTM ) , long-term memory ( LTM ) and a so-called anesthesia-resistant memory ( ARM ) [10 , 11] . STM and MTM are both considered to be unconsolidated whereas ARM and LTM are consolidated forms of memory . The main property of STM and MTM is a dependency on the cyclic adenosine 3’5’-monophosphate ( cAMP ) pathway [12] as exemplified by early studies of rutabaga ( rut ) encoded type I Ca2+-dependent adenylyl cyclase ( AC1 ) [13 , 14] and dunce ( dnc ) encoded type 4 cAMP-specific phosphodiesterase ( PDE4 ) [15–17] . Consolidated LTM and ARM are assumed to be represented by separate molecular pathways [18] . In contrast to ARM formation , LTM requires cAMP response element-binding protein ( CREB ) dependent transcription and de-novo protein synthesis [10 , 19 , 20] . Nevertheless , ARM is resistant to anesthetic agents [21] , which cause retrograde amnesia in both invertebrates and vertebrates [6 , 8 , 21 , 22] . Furthermore ARM formation requires the activity of the radish gene [23 , 24] . Taken together , in adult Drosophila classical conditioning following odor-electric shock reinforcement establishes at least four sequential and/or parallel memory phases ( but see also [25] for a further subdivision of ARM ) . However , there is growing evidence that things are unlikely to be as straightforward as originally envisaged . For example , changing parameters of the training regime , such as feeding state , age of flies , timing of the stimuli and the reinforcing stimulus affects distinct aspects of memory formation and in the most extreme case leads to a mechanistically different type of memory being formed [26–29] . Based on the above described , well-established genetic interventions that have functional implications for adult Drosophila we have analyzed memory formation at the larval stage . Although Drosophila larvae are able to form olfactory and visual memories [30–41] , only a few studies have described larval memory formation in more detail . Larval olfactory memory also consists of different phases [32 , 36 , 40 , 42 , 43] . However , some of the studies identified only a short-lasting memory [32 , 42] , while others studies came to the conclusion that the larval memory consists of both , a short-lasting and a long-lasting component [36 , 40 , 43] . Furthermore , genetic dissection of the larval memory linked memory formation to the cAMP pathway [32 , 36 , 40 , 42 , 43] . However , two of these studies have shown in addition , that rsh1 mutants and turnip ( tur ) mutants , which are reduced in protein kinase C ( PKC ) activity , showed an impairment in larval memory [36 , 40] . Recapitulating the appearance of sequential and/or parallel memory phases in larvae is rather difficult , since these molecular processes were suggested to be independent of cAMP signaling . Here we have deconstructed the larval mnemonic organization after odor-high salt conditioning . Therefore we adapted paradigms from adult Drosophila , which allowed us to identify different components of larval memory . We applied ( i ) a cold shock in order to identify an anesthesia resistant form and ( ii ) blocked protein synthesis in order to distinguish protein synthesis independent from the protein synthesis dependent forms . We have shown that depending on the training regime Drosophila larvae are capable of forming distinct memory phases . Following odor-high salt training we identify three different specifications . We describe for the first time an anesthesia resistant memory in larvae ( lARM ) that it is not affected by cold shock treatment and is evident for up to four hours after training . The component ( we use this term here as we were not able to distinguish between the acquisition , consolidation and retrieval of lARM ) relies on radish and bruchpilot gene function , as well as presynaptic output of mushroom body Kenyon cells ( MB KCs ) and dopaminergic signaling . Furthermore , it utilizes the PKC pathway in contrast to traditional cAMP signaling . Second , we describe a short lasting component ( evident for up to 20 minutes after one cycle training ) that depends on traditional cAMP signaling and synapsin gene function . Third , we identify a CREB dependent component that requires a spaced training protocol , which is composed of five cycles of conditioning spaced by rest intervals of 15 minutes .
Third instar Drosophila larvae are able to learn to associate an odor with punishing high salt concentrations [39 , 44] . Thus we utilized a well-established and standardized two odor reciprocal olfactory conditioning paradigm with 1 . 5M sodium chloride ( NaCl ) as negative reinforcement and tested memory persistence by assaying larvae at increasing times after training ( Fig 1A ) . Please note that the standardized paradigm consists of three training trials ( Fig 1B ) . Significant aversive olfactory memory was evident up to four hours after training ( Fig 1C ) . However , the memory exhibited a gradual decay as the time interval increased and was no longer statistically significant after five hours ( Fig 1C ) . The result is supported by nonlinear regression analysis , which describes the retention curve of odor-high salt memory through an exponential decay function ( Fig 1C ) . This suggests that the initially formed odor-high salt memory gradually decays over time . Our data show that larvae can associate odors with high salt punishment and that the learning dependent change in behavior lasts several hours . In adult Drosophila two types of longer-lasting memories were described , called ARM and LTM . Besides being resistant to anesthetic disruption , ARM is apparently independent of protein synthesis [29] . Yet , LTM formation requires de novo protein synthesis [10 , 19] . In order to test if the memory is dependent on de novo protein synthesis , we fed larvae the translation-inhibitor cycloheximide ( CXM ) 20 hours before the experiment [10] . Then odor-high salt memory was tested immediately or 60 minutes after three cycle standard training ( Fig 2A ) . Performance was unaffected by CXM treatment ( Fig 2A and S1A Fig . ) , suggesting that the formed memory is independent of de novo protein synthesis . This conclusion is further supported by two additional findings . First , the deleterious effect of blocking protein synthesis using CXM became apparent by constantly feeding CXM over a longer period of time . CXM treated larvae did neither pupate nor eclose in contrast to both control groups ( S1B Fig . ) . Second , the transcription factor cAMP response element-binding protein ( CREB ) is universally required for LTM , and it has been reported that a dominant-negative dCreb2b repressor transgene driven by a heat-shock promoter ( hs-dCreb2b ) reduces LTM formation in a heat-shock dependent manner [19 , 20] . Expression of dCreb2b via OK107-Gal4 specifically in the larval MB Kenyon cells did not change odor-high salt memory tested immediately or 60 minutes after training when compared to both genetic controls ( Fig 2B and S1C Fig ) . Yet , adult Drosophila are only capable of forming LTM following a spaced training protocol composed of at least five cycles of conditioning separated by inter-trial intervals of 15 minutes [10 , 11 , 25] . Therefore we established a spaced training paradigm for larval odor-high salt conditioning ( S1D Fig . ; five training cycles , 15 minutes inter-trial interval ) . Spaced training induced a learning dependent change of the behavior of two genetic control groups , but not in the behavior of transgenic larvae expressing dCreb2b via OK107-Gal4 specifically in the larval MB Kenyon cells ( S1D Fig . ) . Thus , the obtained results suggest that the established type of odor-high salt memory is paradigm dependent . However , the prominent component established following three cycle standard training is independent of protein synthesis—and therefore by a general criteria of memory formation not LTM . Next , we tested whether odor-high salt memory following three cycle standard training is resistant to anesthesia . We established a cold shock treatment protocol . We trained larvae as described before but put them directly into cold water ( 4°C ) for one minute after training . Larvae were then transferred onto a room temperature agar plate to recover and memory was tested after different retention times . As shown in Fig 2C ( see also S2C Fig . ) applying a cold shock treatment did not disrupt odor-high salt memory tested 10 , 60 , 120 and 180 minutes after training ( 10 minutes is necessary for recovery from the cold ) . Even applying a stronger cold shock of 5 minutes , which completely paralyzed larvae , did not affect odor-high salt memory ( S2B Fig . ) . We also tested whether cold shock treatment applied 0 , 10 , 20 or 40 minutes after training disrupted 60 minutes memory . Again , no significant defect was revealed ( Fig 2D and S2D Fig ) . To test if larval memory following three cycle standard training is in general resistant to cold shock treatment we additionally used 6mM quinine as a negative reinforcer [34 , 45] and 2 . 0M fructose as an appetitive reinforcer ( Fig 2E and 2F ) . For both stimuli the established memory was resistant to cold shock treatment . Please note that in case of fructose reinforcement the obtained memory was partially reduced . Implications for larval appetitive olfactory learning and memory are later discussed . All in all , our results show for the first time that larvae independent of the applied reinforcer are able to form a type of anesthesia resistant memory . It was shown in adult Drosophila that the radish ( rsh ) gene plays a pivotal role for the formation of ARM [11 , 24] . Hereinafter we therefore focused on rsh gene function . We first analyzed the memory performance of rsh mutant larvae following three cycle standard training immediately after training or after 60 minutes ( Fig 3A ) . In both cases rsh1 mutants showed no significant performance ( Fig 3A ) . To ascertain whether this effect is due to the mutation in the radish gene we performed a rescue experiment ( Fig 3B ) . We tested rsh1 mutants that harbor a wild type rsh transgene , hs-rsh , that allows to induce ubiquitous expression of rsh following heat shock [23] . Non-induced larvae showed a lack of anesthesia resistant learning and/or memory , similar to larvae that carry only the rsh1 mutation . Yet , ubiquitous expression of rsh shortly before the experiment rescued the phenotype ( Fig 3B ) . However , at a reduced level as compared to wild type controls ( Fig 3B ) . Yet , task-relevant sensory-motor abilities of rsh1 larvae are defective in responding to the odor benzaldeyhde ( BA ) ( S3B and S3D Fig ) . To clearly show that the impairment for rsh1 mutants is based on a loss of the ability to associate odor with high salt concentrations , we performed additional experiments . We used a one odor reciprocal paradigm ( S3C Fig . ) [46] . Here BA presentation is replaced by paraffin oil that does not provide any olfactory information for the larva . Again rsh1 larvae showed no anesthesia resistant learning and/or memory ( S3C Fig . ) . In summary , we thus conclude that the behavioral phenotype is due to the fact that the mutation in the rsh gene prevents larvae from establishing , consolidating and/or recalling anesthesia resistant memory . Please note that our experiments did not allow to distinguish between the three different processes . Next we analyzed if intrinsic MB KCs are required for anesthesia resistant learning and/or memory following three cycle standard training due to its conserved role in larval and adult olfactory memory formation [32 , 42 , 47 , 48] . Expression of the temperature-sensitive dominant negative form of dynamin shibirets1 ( UAS-shits1 ) [48 , 49] via the OK107-Gal4 in all KCs allows to block synaptic KC output at a restrictive temperature of 35°C due to impaired vesicle recycling ( Fig 4A ) . In contrast to both genetic control groups , OK107-Gal4/UAS-shits1 larvae showed no anesthesia resistant learning and/or memory ( Fig 4A ) . Yet significant difference was only detectable between the UAS-shits1/+ control and OK107-Gal4 /UAS-shits1 ( Fig 4A ) . Control experiments revealed no gross defects in task-relevant sensory-motor abilities ( S4A Fig . ) . In addition UAS-mCD8::GFP expression driven by OK107-Gal4 verified MB specificity in all KCs besides a limited expression in the ventral nerve cord and brain hemispheres ( Fig 4D ) [48] . Repetition of the experiment with a second mushroom body specific driver H24-Gal4 [48] verified the results obtained for OK107-Gal4 ( S4C and S4D Fig ) . Thus , we conclude that KC output is necessary for anesthesia resistant learning and/or memory . In adult Drosophila two presynaptic determinants , Synapsin ( Syn ) and Bruchpilot ( Brp ) , play a pivotal role in controlling the release of KC vesicles . The evolutionary conserved phosphoprotein Syn is responsible for building a reserve pool of vesicles necessary to maintain vesicle release under high action potential frequencies [50–53] . Adult syn97 mutants showed a defect in aversive olfactory memory that is independent of ARM formation [54 , 55] . The active zone protein Brp , which is a homolog to the ELKS/CAST protein family , is an essential component of the presynaptic dense bodies regulating the release probability of synaptic vesicles [56–58] . The presence of Brp in presynaptic terminals of KCs of adults was suggested to be necessary for establishing ARM [59] . To investigate if both proteins are required for anesthesia resistant learning and/or memory following three cycle standard training , we tested a syn deficient mutant syn97 and brp specific RNAi knockdown in all MB KCs via OK107-Gal4 ( Fig 4B and 4C ) . Gene activity of syn was not required for anesthesia resistant learning and/or memory ( Fig 4B ) . The performance of syn97 mutants was statistically indistinguishable from wild type larvae that served as a genetic control ( Fig 4B ) . Lack of the Syn protein in syn97 was verified using a Syn specific antibody ( Fig 4E ) [47] . In contrast Brp function was necessary for anesthesia resistant learning and/or memory ( Fig 4C ) . It was completely absent in OK107-Gal4;UAS-brp-RNAiB3C8 larvae ( Fig 4C ) . Cell specific knockdown of brp in MB KCs was verified by antibody staining ( Fig 4F ) . In addition , brp RNAi knockdown did not reveal gross defects in task-relevant sensory-motor abilities ( S4B Fig . ) . Consequently , we suggest presynaptic activity of the active zone protein Brp in MB KCs is necessary to establish , consolidate and/or retrieve lARM . Please note that our experiments did not allow to distinguish between the three different processes . Molecular studies in several model organisms–including Drosophila—elucidate cAMP as crucial second messenger in memory formation . A proposed model for the molecular mechanism underlying olfactory memory formation is shown in Fig 5A . An association between the odorant and the reinforcement signals elicits an activation of type I Ca2+-dependent AC encoded by the rut gene via calcium/calmodulin and G-protein stimulation [13 , 14 , 60] . This synergistic activation of AC produces an increase in intracellular cAMP concentration [12] . Intolerable cAMP concentration is prevented through the activity of a type 4 cAMP-specific PDE encoded by the dnc gene [12–14] cAMP for its part activates PKA [61] . The activation of PKA leads either to the phosphorylation of a variety of downstream targets ( e . g . Synapsin , Na+ and K+ channels ) for forming a short-lasting memory [47 , 62–64] or the phosphorylation of CREB forming a long-lasting memory [19 , 20 , 65] . To uncover the molecular pathways responsible for anesthesia resistant learning and/or memory following three cycle standard training we tested larvae carrying three classical learning mutations having deficits in cAMP signaling: rutabaga1 , rutabaga2080 and dunce1 [14 , 17 , 24] ( Fig 5B ) . All three mutants showed lARM that was indistinguishable from wild type controls ( Fig 5B ) . These results indicate that the formation , consolidation and retrieval of lARM is independent of the cAMP/PKA signaling pathway . This conclusion is further supported by two additional findings . First , hypomorphic alleles of the DCO gene locus ( DCOB3 and DCOH2 ) , which encodes the major catalytic subunit of the cAMP-dependent PKA ( PKAc ) showed normal lARM similar to genetic controls ( Fig 5C ) . In adults these heterozygous DCOB3/+ and DCOH2/+ mutants show a 50% reduction of PKA activity and suppress age-related memory impairment [27 , 66] . Second , epidermal growth factor receptor ( EGFR ) signaling to a Ras/Neurofibromatosis type I ( NFI ) pathway was suggested to act via a Rut-AC independent AC to activate PKA function [67] . Notably pan neuronal expression of a dominant-negative isoform of EGFR ( EGRFDN ) impairs olfactory memory formation of Drosophila larvae after bidirectional conditioning [68] . Yet , expression of EGFRDN in all KCs using the OK107-Gal4 did not affect lARM ( Fig 5D ) . Please note that our results do not exclude a potential contribution for these genes at later time points after three cycle standard training . Therefore , biochemical pathways independent of cAMP/PKA signaling cascades have to be involved in lARM tested directly after three cycle standard training . PKC signaling ( Fig 5E ) may serve this function as tur mutants that have a reduced PKC activity are impaired in olfactory learning in adult Drosophila [69] . Furthermore , expression of a truncated constitutively active isoform of PKC ( PKCζ ) rescues the memory defects of rsh1 mutants [70] . In fact , transgenic larvae expressing a specific peptide inhibitor of PKC ( PKCi ) in all KCs using OK107-Gal4 showed strongly reduced anesthesia resistant learning and/or memory in contrast to both genetic controls ( Fig 5F ) . Control experiments revealed no gross defects in task-relevant sensory-motor abilities ( S5 Fig . ) . Thus , we conclude that the formation , consolidation and retrieval lARM tested directly after training is independent of the cAMP/PKA pathway and may instead require PKC signaling in KCs . The current model for associative learning in Drosophila states that during training the unconditioned punishing stimulus is mediated by a specific set of dopaminergic neurons onto MB KCs via G-protein receptor signaling [42 , 71–74] . In Drosophila the dopamine D1-like receptor family that includes two different dopamine receptors , called dDA1 and DAMB , was reported to be necessary for larval and adult learning [75 , 76] . Generally , activation of D1-like receptors was shown to be linked with cAMP/PKA-signaling via Gαs signaling ( Fig 5A ) [77] . Yet more recently it was reported that D1-like receptors also activate phospholipase C ( PLC ) via the Gαo signaling , which leads to an activation of PKC ( Fig 5E ) [77] . Thus we were wondering if lARM formation depends on dopaminergic signaling . In line with prior results , we found that mutants for both receptor genes dumb2 ( for dDA1 ) and damb ( for DAMB ) show a defect in anesthesia resistant learning and/or memory following three cycles standard training ( Fig 6A , S6A Fig . ) [72] . In addition , fumin ( fmn ) mutant larvae that have a mutation in the dopamine transporter ( dDAT ) gene [78]—and thus have enhanced DA levels in adults [79]–also show an impairment in anesthesia resistant learning and/or memory ( Fig 6B ) ( for further details see S6B and S6C Fig ) . Finally , acute oral administration of methylphenidate ( “Ritalin” , MPH ) rescued the behavioral phenotype of rsh1 mutant larvae in a dose-dependent manner ( Fig 6C and S6D Fig ) . MPH application in Drosophila , similar to its function in humans , increases DA levels by inhibiting the dopamine transporter ( dDAT ) that mediates dopamine reuptake from the synaptic cleft [80] . In adults it was shown that oral administration of MPH rescues deficits in optomotor responses of rsh1 mutants [81] . Summarizing , three different experiments suggest that dopaminergic signaling is necessary to establish , consolidate and/or retrieve lARM . Please note that–although the effect of DA is likely specific for the establishment of the memory—our experiments did not allow to distinguish between the three different processes . Conditioning Drosophila larvae via three cycle standard training takes about 45 minutes . Yet , two studies on larval aversive olfactory learning suggest that short lasting memory phases exist that are only detectable up to 20 or 50 minutes after training onset [36 , 42] . These results could mean that three cycle standard training–routinely used in our previous experiments–based on its temporal dimension does not allow to identify short lasting memories . Therefore we established a one cycle training paradigm that only takes about 10 minutes for conditioning ( Fig 7 ) . Significant aversive olfactory memory was evident 0 , 10 , 20 and 60 minutes after training ( Fig 7A and S7A Fig . ) . To our surprise , one cycle training increased aversive memory compared to three cycle training ( both groups were tested immediately after training ) ( S7B Fig . ) . Next , we tested whether odor-high salt memory following one cycle training is resistant to anesthesia . As shown in Fig 7B ( see also S7D Fig . ) applying a cold shock treatment did partially disrupt odor-high salt memory tested 10 minutes after training ( 10 minutes is necessary for recovery from the cold ) . In contrast , memory tested 20 or 60 minutes after one cycle training was cold shock resistant ( Fig 7B ) . Based on these results we conclude that—independent of the number of training trials—odor-high salt conditioning leads to the formation of lARM . However , at the same time a second short lasting memory is established that can only be detected for up to 30 minutes after training onset . Therefore the short lasting memory can only be analyzed after one cycle training but not in longer lasting protocols using two or three training cycles ( S7C Fig . ) . This conclusion is further supported by two additional findings . First , genetic interference with rsh and brp gene function , both involved in the formation , consolidation and/or retrieval of lARM , tested immediately after one cycle training only partially impaired the performance of experimental larvae ( Fig 7C and 7F ) . These results are different than the ones obtained following three cycle standard training ( Fig 3 and Fig 4C ) since they imply the presence of a second , lARM independent , memory phase . Second , rut2080 , dnc1 and , syn97 mutants tested immediately following one cycle training performed on a lower level than genetic controls ( Fig 7D and 7E ) . Again , the results are different compared to the ones obtained following three cycle standard training ( Fig 4B and Fig 5B ) and suggest that the formation , consolidation and/or retrieval of a larval short lasting memory ( lSTM ) under these circumstance depends on cAMP signaling .
Memory formation and consolidation usually describes a chronological order , parallel existence or completion of distinct short- , intermediate- and/or long-lasting memory phases . For example , in honeybees , in Aplysia , and also in mammals two longer-lasting memory phases can be distinguished based on their dependence on de novo protein synthesis [82–85] . In adult Drosophila classical odor-electric shock conditioning establishes two co-existing and interacting forms of memory—ARM and LTM—that are encoded by separate molecular pathways [18] . Seen in this light , memory formation in Drosophila larvae established via classical odor-high salt conditioning seems to follow a similar logic . It consist of lSTM and lARM ( Fig 8A ) ( for a spaced training protocol see also S1D Fig . ) . Aversive olfactory lSTM was already described in two larval studies using different negative reinforcers ( electric shock and quinine ) and different training protocols ( differential and absolute conditioning ) [36 , 42] . Our results introduce for the first time lARM that was also evident directly after conditioning but lasts longer than lSTM ( Fig 8A ) . lARM was established following different training protocols that varied in the number of applied training cycles ( S7C Fig . ) and the type of negative or appetitive reinforcer ( Fig 2C–2F ) . Thus , lSTM and lARM likely constitute general aspects of memory formation in Drosophila larvae that are separated on the molecular level . Memory formation depends on the action of distinct molecular pathways that strengthen or weaken synaptic contacts of defined sets of neurons ( reviewed in [1 , 73 , 86–88] ) . The cAMP/PKA pathway is conserved throughout the animal kingdom and plays a key role in regulating synaptic plasticity . Amongst other examples it was shown to be crucial for sensitization and synaptic facilitation in Aplysia [1 , 86] , associative olfactory learning in adult Drosophila and honeybees [85 , 88] , long-term associative memory and long-term potentiation in mammals [89–92] . For Drosophila larvae two studies by Honjo et al . [42] and Khurana et al . [36] suggest that aversive lSTM depends on intact cAMP signaling . In detail , they showed an impaired memory for rut and dnc mutants following absolute odor-bitter quinine conditioning [42] and following differential odor-electric shock conditioning [36] . Thus , both studies support our interpretation of our results . We argue that odor-high salt training established a cAMP dependent lSTM due to the observed phenotypes of rut , dnc and syn mutant larvae ( Fig 7D and 7E ) . The current molecular model is summarized in Fig 8B . Yet , it has to be mentioned that all studies on aversive lSTM in Drosophila larvae did not clearly distinguish between the acquisition , consolidation and retrieval of memory . Thus , future work has to relate the observed genetic functions to these specific processes . In contrast , lARM formation utilizes a different molecular pathway . Based on different experiments , we have ascertained , that lARM formation , consolidation and retrieval is independent of cAMP signaling itself ( Fig 5B ) , PKA function ( Fig 5C ) , upstream and downstream targets of PKA ( Figs 5D and 4B ) and de-novo protein synthesis ( Fig 2A and 2B ) ( but see also for spaced conditioning S1D Fig . ) . Instead we find that lARM formation , consolidation and/or retrieval depends on rsh gene function ( Fig 3 ) , brp gene function ( Fig 4C ) , dopaminergic signaling ( Fig 6 ) and requires presynaptic signaling of MB KCs ( Fig 4A and 4C and S4C–S4E Fig . ) . Interestingly , studies on adult Drosophila show that rsh and brp gene function , as well as dopaminergic signaling and presynaptic MB KC output are also necessary for adult ARM formation [23–25 , 59] . Thus , although a direct comparison of larval and adult ARM is somehow limited due to several variables ( differences in CS , US , training protocols , test intervals , developmental stages , and coexisting memories ) , both forms share some genetic aspects . This is remarkable as adult ARM and lARM use different neuronal substrates . The larval MB is completely reconstructed during metamorphosis and the initial formation of adult ARM requires a set of MB α/β KCs that is born after larval life during puparium formation [25 , 93 , 94] . In addition , we have elicited the necessity of PKC signaling for lARM formation in MB KCs ( Fig 5C ) . The involvement of the PKC pathway for memory formation is also conserved throughout the animal kingdom . For example , it has been shown that PKC signaling is an integral component in memory formation in Aplysia [95–98] , long-term potentiation and contextual fear conditioning in mammals [99–101] and associative learning in honeybees [102] . In Drosophila it was shown that PKC induced phosphorylation cascade is involved in LTM as well as in ARM formation [103] . Although the exact signaling cascade involved in ARM formation in Drosophila still remains unclear , we established a working hypothesis for the underlying genetic pathway forming lARM based on our findings and on prior studies in different model organisms ( Fig 8B ) . Thereby we do not take into account findings from Horiuchi et al . [66] and Scheunemann et al . [104] in adult Drosophila . These studies show that PKA mutants have increased ARM [66] and that dnc sensitive cAMP signaling supports ARM [104] . Thus both studies directly link PKA signaling with ARM formation . It was shown that KCs act on MB output neurons to trigger a conditioned response after training [105 , 106] . Work from different insects suggests that the presynaptic output of an odor activated KCs is strengthened if it receives at the same time a dopaminergic , punishment representing signal . Our results support these models as they show that lARM formation requires accurate dopaminergic signaling ( Fig 6 ) and presynaptic output of MB KCs ( Fig 4A and 4C ) . Yet , for lARM formation dopamine receptor function seems to be linked with PKC pathway activation ( Fig 5 ) . Indeed , in honeybees , adult Drosophila and vertebrates it was shown that dopamine receptors can be coupled to Gαq proteins and activate the PKC pathway via PLC and IP3/DAG signaling [107 , 108] . As potential downstream targets of PKC we suggest radish and bruchpilot . Interference with the function of both genes impairs lARM ( Figs 3 and 4C ) . The radish gene encodes a functionally unknown protein that has many potential phosphorylation sites for PKA and PKC [23] . Thus considerable intersection between the proteins Rsh and PKC signaling pathway can be forecasted . Whether this is also the case for the bruchpilot gene that encodes for a member of the active zone complex remains unknown . The detailed analysis of the molecular interactions has to be a focus of future approaches . Therefore , our working hypothesis can be used to define educated guesses . For instance , it is not clear how the coincidence of the odor stimulus and the punishing stimulus are encoded molecularly . The same is true for ARM formation in adult Drosophila . Based on our working hypothesis we can speculate that PKC may directly serve as a coincidence detector via a US dependent DAG signal and CS dependent Ca2+ activation . Do our findings in general apply to learning and memory in Drosophila larvae ? To this the most comprehensive set of data can be found on sugar reward learning . Drosophila larva are able to form positive associations between an odor and a number of sugars that differ in their nutritional value [31 , 32 , 47 , 109–111] . Using high concentrations of fructose as a reinforcer in a three cycle differential training paradigm ( comparable to the one we used for high salt learning and fructose learning ( Figs 1 and 2F ) ) Michels et al . [47 , 111] found that learning and/or memory in syn97 mutant larvae is reduced to ∼50% of wild type levels . Thus , half of the memory seen directly after conditioning seems to depend on the cAMP-PKA-synapsin pathway . Our results in turn suggest that the residual memory seen in syn97 mutant larvae is likely lARM ( Fig 2F ) . Thus , aversive and appetitive olfactory learning and memory share general molecular aspects . Yet , the precise ratio of the cAMP-dependent and independent components rely on the specificities of the used odor-reinforcer pairings . Two additional findings support this conclusion . First , Kleber et al . [112] recently showed that memory scores in syn97 mutant larvae are only lower than in wild type animals when more salient , higher concentrations of odor or fructose reward are used . Usage of low odor or sugar concentrations does not give rise to a cAMP-PKA-synapsin dependent learning and memory phenotype . Second , Honjo et al . [32] showed that learning and/or memory following absolute one cycle conditioning using sucrose sugar reward is completely impaired in rut1 , rut2080 and dnc1 mutants . Thus , for this particular odor-reinforcer pairing only the cAMP pathway seems to be important . Therefore , a basic understanding of the molecular pathways involved in larval memory formation is emerging . Further studies , however , will be necessary in order to understand how Drosophila larvae make use of the different molecular pathways with respect to a specific CS/US pairing .
Fly strains were reared on standard Drosophila medium at 25°C or 18°C in constant darkness or with a 14/10 hr light/dark cycle . For behavioral analysis we used rut1 , rut2080 , dnc1 , rsh1 [14 , 17 , 24] ( kindly provided by T . Preat ) , DCOB3 , DCOH2 [27] ( kindly provided by M . Saitoe ) , fumin [78] ( kindly provided by M . Heisenberg ) and syn97 [54 , 111] ( kindly provided by B . Gerber ) mutants . All lines were outcrossed over several generations with wild type CantonS that was used as a genetic control . In addition , we used the two dopamine receptor mutants damb and dumb2 and their genetic controls [71] . Note , that in contrast to earlier studies the damb mutant was outcrossed to CantonS over several generations . To rescue the rsh dependent phenotype by artificial ubiquitous rsh expression we used rsh1; hs-rsh larvae [23] ( kindly provided by T . Zars ) . To express Gal4 in all larval KCs we used the driver line OK107 [48 , 113] ( DGRC no . : 106098 ) . UAS-shits was used to acutely block synaptic output from KCs [49] ( BDSC No . : 7068 ) . In addition , we used the four effector lines UAS-EGFRDN [68] ( kindly provided by T . Roeder ) , UAS-RNAiB3C8 [59] ( kindly provided by H . Tanimoto ) , UAS-PKCi [114] ( kindly provided by B . Brembs ) and UAS-Creb2-b [115] ( kindly provided by S . Waddell ) . Experiments were conducted on assay plates ( 85mm diameter , Cat . No . : 82 . 1472 , Sarstedt , Nümbrecht ) filled with a thin layer of 2 . 5% agarose containing either pure agarose ( Sigma Aldrich Cat . No . : A5093 , CAS No . : 9012-36-6 ) or agarose plus reinforcer . We used 1 . 5M and 2 . 0M sodium chloride ( Sigma Aldrich Cat . No . : S7653 , CAS No . : 7647-14-5 ) [44] , 2 . 0M fructose ( Sigma Aldrich Cat . No . : 47740 , CAS No . : 57-48-7 ) [110 , 116] and 6mM quinine ( quinine hemisulfate salt monohydrate , Sigma Aldrich Cat . No . : Q1250 , CAS No . : 207671-44-1 ) [34 , 45] . As olfactory stimuli , we used 10 μl amyl acetate ( AM , Fluka Cat . No . : 46022; CAS No . : 628-63-7; diluted 1:250 in paraffin oil , Fluka Cat . No . : 76235 , CAS No . : 8012-95-1 ) and benzaldehyde ( BA , undiluted; Fluka Cat . No . : 12010 , CAS No . : 100-52-7 ) . Odorants were loaded into custom-made Teflon containers ( 4 . 5-mm diameter ) with perforated lids [109] . Learning ability was tested by exposing a first group of 30 animals to AM while crawling on agarose medium containing in addition sodium chloride as a negative reinforcer . After 5 min , larvae were transferred to a fresh Petri dish in which they were allowed to crawl on pure agarose medium for 5 min while being exposed to BA ( AM+/BA ) . A second group of larvae received the reciprocal training ( AM/BA+ ) . If not stated otherwise , three training cycles are used . Depending on the memory retention larvae were transferred onto another agarose plate prior to training and kept for the indicated time before testing the memory . To increase humidity tap water was added . Memory is tested by transferring larvae onto test plates on which AM and BA were presented on opposite sides . For fructose reinforcement the test plates contains pure agarose whereas for sodium chloride and quinine reinforcement agarose plates containing the respective reinforcer are used . After 5 min , individuals were counted as located on the AM side ( #AM ) , the BA side ( #BA ) , or in a 1 cm neutral zone . By subtracting the number of larvae on the BA side from the number of larvae on the AM side , and dividing by the total number of counted individuals ( #TOTAL ) , we determined a preference index for each training group: To measure specifically the effect of associative learning that is of the odor-reinforcement contingency , we then calculated the associative performance index ( PI ) as the difference in preference between the reciprocally trained larvae: Negative PIs thus represent aversive associative learning , whereas positive PIs indicate appetitive associative learning . Division by 2 ensures scores are bound within ( -1; 1 ) . Heat shocks were applied for six hours . Therefore , food vials with six days old larvae were transferred into an incubator at 35°C . We heat-shocked the w , rhs1;hs-rsh transgenic flies , rsh1 mutants and as controls wild type and w1118 larvae . Afterwards each group received an aversive olfactory training regime at room temperature . For cold-shock experiments larvae were incubated in ice tap water ( 4°C ) for one minute . Larvae were allowed to recover by transferring them onto agarose plates . They started moving within one minute and were kept on the agarose plates at 23°C until testing . To test if aversive olfactory learning is dependent on de novo protein synthesis , wild type larvae were fed 35 mM cycloheximide ( +CXM; Sigma Aldrich Cat . No . : C7698; CAS No . : 66-81-9 ) in 5% sucrose ( w/v ) , 5% sucrose alone ( -CXM ) or tab water ( -CXM , -SUC ) for 20 hours before the experiment [10] . Therefore , 300 ml of solution was added into food vials . Before the experiment larvae were gently transferred to an empty Petri dish and washed with tap water before training and testing . Methylphenidate ( MPH; Sigma Aldrich Cat . No . : M2892; CAS No . : 298-59-9 ) , was orally administered to the larvae , as it was shown that oral consumption of MPH is sufficient to inhibit the Drosophila dopamine transporter [80] . MPH was diluted in tap water with a concentration of 2 mM , stored in the refrigerator and used within 7 days . MPH was applied for one hour to a group of 30 larvae in Petri dishes with an inner diameter of 35 mm ( Greiner ) . Afterwards larvae were gently transferred to an empty Petri dish and washed with tap water before training and testing . To acutely block synaptic output we used UAS-shits1 [49] . Immediately before the experiment , larvae were incubated for 2 min in a water-bath at 37°C . The behavioral experiments were then performed as described before , at restrictive temperature of about 35°C in a custom made chamber placed within a fume hood . Control experiments were performed with incubation at room temperature and at permissive temperature of about 23°C . All statistical analyses and visualizations were done with GraphPad Prism 5 . 0 . Groups that did not violate the assumption of normal distribution ( Shapiro–Wilk test ) and homogeneity of variance ( Bartlett's test ) were analyzed with parametric statistics: unpaired t-test ( comparison between two groups ) or Oneway ANOVA followed by planned pairwise comparisons between the relevant groups with a Tukey honestly significant difference HSD post hoc test ( comparisons between groups larger than two ) . Experiments with data that were significantly different from the assumptions above were analyzed with non-parametric tests , such as Mann–Whitney test ( comparison between two groups ) or Kruskal–Wallis test followed by Dunn's multiple pairwise comparison ( comparisons between groups larger than two ) . To compare single genotypes against chance level , we used One sample t test or Wilcoxon signed rank test . For statistical test concerned with factors equal two or more , two way ANOVA was applied followed by the planned pairwise multiple comparisons ( Bonferroni ) . The significance level of statistical tests was set to 0 . 05 . Figure alignments were done with Adobe Photoshop . Data were presented as box plots , 50% of the values of a given genotype being located within the boxes , whiskers represent the entire set of data . Outliers are indicated as open circles . The median performance index was either indicated as a bold line and the mean as a cross within the box plot or symbol expressed as means ± s . e . m . Unless stated otherwise , all olfactory conditioning experiments are n = 16 . Third instar larvae were put on ice and dissected in phosphate-buffered saline ( PBS ) [71 , 117 , 118] . Brains were fixed in 3 . 6% formaldehyde ( Merck , Darmstadt ) in PBS for 30 min . After eight times rinsing in PBT ( PBS with 3% Triton-X 100 , Sigma-Aldrich , St . Louis , MO ) , brains were blocked with 5% normal goat serum ( Vector Laboratories , Burlingame , CA ) in PBT for 2 hours and then incubated for two days with primary antibodies at 4°C . Before applying the secondary antibodies for two days at 4°C , brains were washed eight times with PBT . After secondary antibody incubation , brains were washed eight times with PBT and mounted in Vectashield ( Vector Laboratories , Burlingame , CA ) between two cover slips and stored at 4°C in darkness . Images were taken with a Zeiss LSM 510M confocal microscope with x25 or x40 glycerol objectives . The resulting image stacks were projected and analyzed with Image-J ( National Institutes of Health , Bethesda , Maryland , http://imagej . nih . gov/ij ) software . Contrast and brightness adjustment as well as rotation and organization of images were performed in Photoshop ( Adobe Systems Inc . , San Jose , CA ) . To analyze the expression pattern of OK107-Gal4 rabbit anti-GFP antibody ( A6455 , Molecular Probes; 1:1000 ) and two different mouse antibodies for staining the cholinergic neuropil ( ChAT4B1; DSHB , Iowa City , IA , 1:150 ) and axonal tracts ( 1d4 anti-Fasciclin 2; DSHB , Iowa City , IA; 1:50 ) were applied [71 , 117] . A specific antibody for the Synapsin protein was used to verify the mutation syn97 ( monoclonal mouse anti-syn , 3C11; DSHB , Iowa City , IA , 1:10 ) [54] . To analyze if the expression level of Bruchpilot is specifically reduced in the MB KCs by driving UAS-RNAiB3C8 via OK107-Gal4 , monoclonal mouse anti-nc82 was used ( nc82 , DSHB , Iowa City , IA , 1:10 ) [58] . As secondary antibodies goat anti-rabbit IgG Alexa Fluor 488 ( A11008 , Molecular Probes , 1:200 ) and goat anti-mouse IgG Alexa Fluor 647 ( A21235 , Molecular Probes , 1:200 ) were used .
|
Learning and memory helps organisms to predict and adapt to events in their environment . Gained experience leaves traces of memory in the nervous system . Yet , memory formation in vertebrates and invertebrates is a highly complex and dynamic process that consists of different phases , which depend on various neuronal and molecular mechanisms . To understand which changes occur in a brain when it learns , we applied a reductionist approach . Instead of studying complex cases , we analyzed learning and memory in Drosophila larvae that have a simple brain that is genetically and behaviorally accessible and consists of only about 10 , 000 neurons . Drosophila larvae are able to learn to associate an odor with punishing high salt concentrations . It is therefore possible to correlate changes in larval behavior with molecular events in identifiable neurons after classical olfactory conditioning . We show that under these circumstances larvae form two parallel memory phases; a short lasting component ( lSTM ) that is molecularly conserved throughout the animal kingdom as it depends on the classical cAMP pathway . In parallel they establish a larval anesthesia resistant memory ( lARM ) that relies on a different molecular signal . lARM has not been described in larvae before .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"learning",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"chemical",
"compounds",
"anesthesiology",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"animals",
"organic",
"compounds",
"hormones",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"cognitive",
"psychology",
"protein",
"synthesis",
"pharmaceutics",
"cognition",
"anesthesia",
"memory",
"amines",
"neurotransmitters",
"catecholamines",
"drosophila",
"dopamine",
"research",
"and",
"analysis",
"methods",
"chemical",
"synthesis",
"proteins",
"life",
"cycles",
"chemistry",
"insects",
"biosynthetic",
"techniques",
"arthropoda",
"biochemistry",
"psychology",
"organic",
"chemistry",
"biogenic",
"amines",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"drug",
"therapy",
"cognitive",
"science",
"larvae",
"organisms"
] |
2016
|
Genetic Dissection of Aversive Associative Olfactory Learning and Memory in Drosophila Larvae
|
Mitochondrial protein import is essential for Trypanosoma brucei across its life cycle and mediated by membrane-embedded heterooligomeric protein complexes , which mainly consist of trypanosomatid-specific subunits . However , trypanosomes contain orthologues of small Tim chaperones that escort hydrophobic proteins across the intermembrane space . Here we have experimentally analyzed three novel trypanosomal small Tim proteins , one of which contains only an incomplete Cx3C motif . RNAi-mediated ablation of TbERV1 shows that their import , as in other organisms , depends on the MIA pathway . Submitochondrial fractionation combined with immunoprecipitation and BN-PAGE reveals two pools of small Tim proteins: a soluble fraction forming 70 kDa complexes , consistent with hexamers and a second fraction that is tightly associated with the single trypanosomal TIM complex . RNAi-mediated ablation of the three proteins leads to a growth arrest and inhibits the formation of the TIM complex . In line with these findings , the changes in the mitochondrial proteome induced by ablation of one small Tim phenocopy the effects observed after ablation of TbTim17 . Thus , the trypanosomal small Tims play an unexpected and essential role in the biogenesis of the single TIM complex , which for one of them is not linked to import of TbTim17 .
The parasitic protozoan Trypanosoma brucei is the causative agent of the devastating human sleeping sickness and of nagana in cattle [1] . However , besides its clinical and economic importance , T . brucei is also an interesting model to investigate variations of basic cell biological processes [2 , 3] . One such process is mitochondrial protein import , which has been studied in great detail in Saccharomyces cerevisiae and in mammalian cells [4 , 5] . Modern phylogeny divides eukaryotes into five to six supergroups that diverged very early in evolution [6] . Fungi and animals belong to the supergroup of the Opisthokonts and therefore are quite closely related . Trypanosomes are a member of the supergroup of the Excavates and thus are phylogenetically very distant to Opisthokonts [7] . Due to this position in the eukaryotic evolutionary tree and its experimental accessibility , T . brucei is excellently suited to investigate which features of mitochondrial protein import are conserved and which ones are not [3 , 4] . Recent studies in T . brucei have characterized the main protein translocase of the mitochondrial outer membrane ( TOM ) , termed archaic translocase of the OM ( ATOM ) [8 , 9] , as well as the translocase of the inner membrane ( TIM ) [10] . Only two subunits of the ATOM complex and one integral membrane subunit of the single trypanosomal TIM complex are orthologous to TOM and TIM complex subunits of any other eukaryote [3 , 4] . This is surprising since protein import is considered to be one of the first , if not the first , mitochondria-specific trait to evolve , which suggests that the machineries mediating the process would be conserved . In the present study we focus on the small Tim family of intermembrane space ( IMS ) localized chaperones ( also known as tiny Tims ) , which is conserved in all mitochondria-containing eukaryotes including trypanosomes [11 , 12] . Members of the small Tim family have a molecular weight of around 10 kDa and contain conserved twin Cx3C motifs that normally are separated by 11–16 residues [13] . The two motifs form intramolecular disulfide bonds that stabilize the helix-loop-helix structure of small Tim proteins [14 , 15] . Their function in yeast and humans is to guide hydrophobic import substrates that emerge from the TOM complex across the IMS to the respective insertases [13] . Mitochondrial carrier proteins ( MCPs ) , which usually contain 6 transmembrane spanning domains , and other inner membrane ( IM ) proteins with internal targeting signals are transferred by small Tims to the TIM22 complex in the IM [16 , 17] . Similarly , β-barrel proteins of the OM are handed over to the sorting and assembly machinery ( SAM ) [18 , 19] . To this end , small Tim proteins form ring-like hetero-hexameric oligomers . In yeast , two such structures have been characterized consisting of either alternating Tim9/Tim10 or Tim8/Tim13 subunits [14 , 20 , 21] . Moreover , a small fraction of Tim9/Tim10 is associated with Tim12 and binds to the TIM22 complex [22] . Similarly , in human mitochondria , two hexameric complexes consisting of Tim9/Tim10a or DDP1 ( deafness dystonia peptide 1 ) /Tim13 as well as a Tim9/Tim10a/Tim10b complex that associates with TIM22 can be differentiated [23 , 24] . Like most mitochondrial proteins the small Tims are imported . Their import , similar to that of other small cysteine-rich IMS proteins , is coupled to oxidation of their cysteine residues [25] . Small Tim proteins are first translocated across the OM through the TOM complex . Subsequently they engage with Mia40 of the mitochondrial IMS import and assembly machinery ( MIA ) , which in cooperation with the sulfhydryl oxidase Erv1 promotes oxidation and folding of the proteins before they assemble into the hexameric complexes or associate with the TIM22 complex [26–28] . Bioinformatic analysis predicts that the T . brucei genome encodes five classical small Tims , as well as one small Tim-like protein that only contains two cysteine residues instead of the classical twin Cx3C motif [3 , 10–12] . Merely a single one of these proteins has been experimentally analyzed to a very limited extent [11] . Here we present the first detailed experimental analysis of trypanosomal small Tim proteins . It focuses on the three most recently discovered proteins , which includes the atypical small Tim . We demonstrate that mitochondrial import of these proteins depends on the trypanosomal TbERV1 orthologue . Moreover , we show that the three small Tims are subunits of the single trypanosomal TIM complex ( approximately 700 kDa ) [10] , but are also present in soluble complexes of approximately 70 kDa . Ablation of any of the three proteins inhibits normal growth , results in the disappearance of the TIM complex and causes mitochondrial protein import defects . While individual ablation of two of the proteins affects import of TbTim17 , a core subunit of the single trypanosomal TIM complex , we provide evidence that one small Tim protein is directly involved in TIM complex assembly .
Using bioinformatics , Gentle et al . detected three trypanosomal small Tim proteins , termed Tim9 , Tim10 and Tim8-13 [11] . In a more recent compositional characterization of the single trypanosomal TIM complex , three additional such proteins were discovered . Based on their molecular weight they were termed TbTim11 ( Tb927 . 5 . 3340 ) , TbTim12 ( Tb927 . 4 . 3430 ) and TbTim13 ( Tb927 . 10 . 11520 ) . The characteristic twin Cx3C motifs that stabilize the hairpin structure of small Tims and target them to the MIA pathway in other eukaryotes are conserved in five of these trypanosomal small Tims [11] ( Fig 1A ) . TbTim12 however contains only two cysteines that are separated by 22 amino acids , indicating a lack of the outer disulfide bond [12] . This is a unique feature not found in any other small Tim so far . Sequence comparison of trypanosomal small Tims with their counterparts in yeast and human indicates that it is not possible to assign clear yeast or human orthologues to the newly identified trypanosomal small Tims ( S1 Table ) . It has been shown that all trypanosomal small Tims are associated with the single trypanosomal TIM complex , irrespectively of whether it is engaged in MCP import or in import of presequence containing proteins [10] . The latter is unexpected as presequence-containing proteins do not require small Tims to cross the IMS . In order to determine the intracellular localization of the three newly discovered small Tim proteins , we established cell lines allowing inducible ectopic expression of C-terminally myc- and HA-tagged versions of the three proteins in various combinations . Cell lines expressing tagged TbTim11 and TbTim13 were subjected to digitonin-based cell fractionation . The results in Fig 1B show that both tagged proteins as well as the endogenous TbTim12 co-fractionated with the mitochondrial marker ATOM40 . Furthermore , alkaline carbonate extraction at pH 11 . 5 of such crude mitochondrial fractions demonstrated that all three proteins are soluble proteins and behave identical to the IMS-localized peripheral membrane protein cytochrome C ( Fig 1C ) . Import of small Tim proteins into the IMS of yeast and human mitochondria depends on the MIA pathway . We therefore expect the same to be the case for the trypanosomal proteins . To test this prediction we expressed the tagged versions of TbTim11 , TbTim12 and TbTim13 individually in the background of an RNAi cell line allowing for ablation of TbERV1 , the only known component of the trypanosomal MIA pathway [29] . We have previously shown that inhibition of protein import by ablation of import factors results in rapid degradation of the corresponding non-imported substrate proteins in the cytosol [9] . In line with this , Fig 1D shows that the levels of all three novel small Tim proteins were drastically reduced upon TbERV1 RNAi induction . The same was true for endogenous Tim9 , but not for cytochrome C , which is imported in a MIA-independent fashion [30] . This indicates that TbTim11 , TbTim13 as well as TbTim12 are substrates of the trypanosomal MIA pathway consistent with a localization in the mitochondrial IMS . The initial identification of the novel small Tims was based on a set of reciprocal co-immunoprecipitations targeting TbTim17 and TbTim13 . In these experiments all six trypanosomal small Tims were specifically interacting with both bait proteins [10] . This stands in contrast to other organisms in which some small Tims exclusively form soluble complexes . Thus , we extended our analysis and performed pulldowns of crude mitochondrial fractions ( termed "crude mito" ) with TbTim11-HA and TbTim12-myc ( Fig 2A ) . The results show that the two tagged proteins , the endogenous Tim9 as well as the integral membrane TIM subunits TbTim17 , TbTim42 and TimRhom I are enriched in the eluate , whereas CoxIV and ATOM40 essentially remain in the unbound fractions . In summary , these results confirm the reciprocal interaction of the trypanosomal small Tims and their association with the single TIM complex . Even though alkaline extraction at pH 11 . 5 had demonstrated that the novel small Tims are soluble proteins ( Fig 1C ) , the interaction with transmembrane components of the TIM complex suggests an association with the mitochondrial inner membrane . In order to analyze this further , we subjected a cell line expressing TbTim11-HA and TbTim13-myc to alkaline extraction at pH 10 . 7 . Under these less stringent conditions , all detectable small Tims were partially found in the insoluble pellet fraction , while the peripheral membrane protein cytochrome C was still exclusively detected in the soluble fraction ( Fig 2B ) . However , if the experiment was repeated in a cell line ablated for TbTim17 , all analyzed small Tims ( TbTim11 , TbTim12 , TbTim13 and Tim9 ) were exclusively detected in the soluble fraction , regardless of the extraction conditions used ( Fig 2B ) . These results confirm that the partial association of small Tims with the mitochondrial IM depends on the presence of the TIM complex . Moreover , they reveal a strong interaction between the small Tims and the integral membrane subunits of the TIM complex that , unlike most typical interactions between peripheral and integral membrane proteins , is in part resistant to highly alkaline conditions . To assess the organization of the small Tim-containing complexes further , we performed 2D-blue native ( BN ) /SDS-PAGE analysis using crude mitochondrial fractions solubilized with 1% digitonin ( Fig 3A ) . The three novel small Tims were found in a heterogenous population of protein complexes of approximately 70 kDa , 150 kDa and ≥700 kDa in size . While most of the high molecular weight complexes co-migrate with the ones formed by the transmembrane subunits TbTim17 and TbTim42 of the TIM complex , the smallest complexes might correspond to the soluble 70 kDa hetero-hexameric complexes formed by small Tims in yeast and humans . To assess if any of these complexes are soluble , we repeated the 2D BN-PAGE using mitochondrial subfractions . To that end , we lysed the OM of crude mitochondria using 0 . 2% digitonin . The released soluble fraction contains proteins of the IMS ( termed "soluble fraction" ) , whereas the remaining pellet fraction consists mainly of IM proteins ( termed "membrane fraction" ) . The results in Fig 3B show that the only small Tim-containing complexes present in the soluble fraction are approximately 70 kDa in size , whereas the higher molecular weight complexes were exclusively found in the membrane fraction . The uniform size of the soluble small Tim complexes of approximately 70 kDa suggests the existence of hexameric assemblies . Next , we separately analyzed the composition of the soluble and membrane-associated complexes formed by the novel small Tims by co-immunoprecipitations based on the submitochondrial fractions described above . Combination cell lines expressing two differently tagged small Tims and the use of peptide antibodies against endogenous Tim9 and TbTim12 allowed us to probe for up to 4 small Tims in a single pulldown experiment . The results show that in pulldowns of the soluble fraction all detectable small Tims co-precipitate with either TbTim13 ( Fig 3C ) or TbTim12 ( Fig 3D ) . The only difference observed in pulldowns of the respective solubilized crude mitochondria was that also TbTim17 was recovered . This demonstrates that there are at least two major populations of trypanosomal small Tim complexes , soluble ones in the IMS which contain all small Tims but no other TIM complex subunits , and another one in which all small Tims are associated with the TIM complex of the inner membrane [10] . To analyze the composition of the soluble small TIM complexes in more detail , we performed SILAC-based co-immunoprecipitation experiments using the T . brucei cell lines allowing inducible , ectopic expression of C-terminally HA-tagged versions of TbTim11 , TbTim12 and TbTim13 , respectively . Each of these cell lines was grown in the presence or absence of tetracycline and isotopically-labeled heavy or light lysine and arginine respectively . Subsequently , identical cell numbers of both populations were mixed , crude mitochondria were prepared and their OM was lysed with 0 . 2% of digitonin . The resulting fractions containing the soluble small Tim-containing complexes were subjected to immunoprecipitations using anti-HA antibodies and analyzed by quantitative MS . The results yield a consistent picture: in all three cases all six small Tims were highly enriched in the IPs and no other significantly enriched proteins were detected ( Fig 4A ) ( S2 Table ) . The molecular weight of the small Tim-containing complexes of close to 70 kDa suggests they are present in hexameric assemblies . However , the uniform recovery of all six small Tims in all pull down experiments excludes that the postulated hexamers , as in yeast and mammals , are formed by specific pairs of small Tims . Two alternative quaternary structures that are consistent with the experimental evidence would be: i ) that all six small Tims build a single defined hexamer consisting of six different subunits , or ii ) that the six different small Tim subunits form promiscuous hexamers without defined subunit compositions . Should a single defined hexamer exist we would expect that ablation of any of the six small Tim subunits would result in the collapse of the hexamer . However , in the case of heterogeneous hexamers ablation of a single Tim subunit would not affect hexamer formation as its absence would be compensated for by the other members of the small Tim protein family . 2D BN-PAGE analysis of the soluble complexes in uninduced and induced TbTim11 , TbTim12 and TbTim13-RNAi cell lines favours the second model , since ablation of the corresponding small Tims does not significantly affect the soluble complexes containing other small Tim proteins that were not targeted by the RNAi ( Fig 4B ) . In order to examine the function of TbTim11 , TbTim12 and TbTim13 , we produced tetracycline-inducible RNAi cell lines . Fig 5A shows that all three proteins , like Tim8-13 analyzed in a previous study [11] , are essential for normal growth of procyclic trypanosomes . Moreover , the TIM complex in these RNAi cell lines rapidly disappeared as shown by BN-PAGE ( Fig 5B ) . However , in the same cells the mitochondrial morphology looked normal and the membrane potential was still intact till the onset of the growth phenotype ( S3 Fig ) . Finally , in agreement with the loss of the TIM complex we observed an accumulation of unprocessed CoxIV precursors in all three small Tim RNAi-cell lines ( Fig 5C ) , which is a hallmark of mitochondrial protein import defects . TbTim17 belongs to the Tim17/22/23 protein family , which in yeast and humans requires small Tim chaperones for its import into mitochondria [24 , 31–33] . In line with this , the steady state levels of TbTim17 are strongly decreased upon ablation of RNAi against TbTim11 and TbTim12 suggesting that the two proteins are involved in import of TbTim17 . In absence of TbTim13 , however , the steady levels of TbTim17 remain essentially constant ( Fig 5C ) . The abundance of TimRhom I , another subunit of the TIM complex , is slightly reduced in the TbTim11-RNAi cell line although not to the same extent than TbTim17 . In the TbTim12 and TbTim13 RNAi cell lines on the other hand the levels of TimRhom I remain essentially unchanged indicating that the protein is stable in the absence of the TIM complex . In order to investigate the function of TbTim13 in more detail we analyzed the global changes in the mitochondrial proteome that were caused by its ablation . A quantitative MS analysis of protein levels in crude mitochondrial fractions of induced versus uninduced TbTim13 RNAi cells demonstrated significantly reduced levels ( ≥1 . 5 fold , p-value ≤ 0 . 05 ) of 443 proteins ( S3 Table ) . Most of these ( 86% ) were mitochondrial proteins as defined in a recent proteomic study [34] . While the TIM complex is not detectable anymore after only 2 days of induction of TbTim13 RNAi ( Fig 5B , right panel ) , the levels of the TbTim17 and all other 10 TIM complex subunits did not decrease even after 2 . 5 days of RNAi induction , which is when the proteomic analysis was performed ( Fig 6A ) [3] . For the core subunit TbTim17 , alkaline carbonate extraction confirmed that the protein is still inserted into the inner membrane in the absence of TbTim13 ( Fig 6B ) . Thus , these results strongly suggest that TbTim13 , in contrast to TbTim11 and TbTim12 , is directly required for the assembly and/or maintenance of the trypanosomal TIM complex but not for import of its subunits . This also explains , why its ablation essentially phenocopies the effects seen after ablation of TbTim17 [10] . We furthermore investigated the fate of two groups of proteins which are typical substrates of small Tim chaperones in other organisms , namely MCPs and β-barrel proteins . Almost half of the detected MCPs [35] were found to be significantly decreased in the TbTim13 RNAi cell line ( Fig 6C ) . However , it is not possible to distinguish whether this phenotype is caused by impairment of the predicted chaperone function of TbTim13 or due to the more direct role it plays in TIM complex assembly and/or maintenance . The biogenesis of β-barrel proteins like ATOM40 or VDAC , on the other hand , is independent of TbTim17 and thus the slightly reduced abundance we observed for all trypanosomal β-barrel proteins could be a direct effect of TbTim13 knockdown ( Fig 6C ) . This is consistent with the idea that TbTim13 may facilitate the transfer of β-barrel protein across the IMS to the SAM complex . However , the observed reductions are too small to be reflected in the amounts of assembled ATOM40 and VDAC that are detected on the BN-PAGE analysis of the TbTim13 RNAi cell line ( S4 Fig ) . The same was observed for the RNAi cell lines targeting TbTim11 and TbTim12 .
The small Tim family of IMS-localized chaperones belongs to the most conserved components of the mitochondrial protein import system . They are found in all mitochondria-containing eukaryotes , even in trypanosomes which contain highly diverged OM and IM protein translocases [3 , 4] . Here we show that the small Tim protein family in T . brucei includes six members , all of which are implicated in mitochondrial protein import [10 , 11] . One of them , TbTim12 , is unusual since it has an incomplete Cx3C small Tim signature motif and thus can only be stabilized by a single intramolecular disulfide bond . Nevertheless , TbTim12 clearly belongs to the small Tim family since its ablation causes the same defects that are observed in knockdown cell lines targeting classical small Tims [11] . Moreover , TbTim12 is conserved throughout kinetoplastids [12] . Thus , it is possible that similar , unusual small Tims might exist in other organisms but have escaped detection precisely because they lack complete small Tims signature domains . Also , TbTim11 and TbTim13 were not identified in previous bioinformatic analyses , even though they contain the expected Cx3C motifs . The probable reason in this case is that their Cx3C motifs are spaced by more than 16 residues [11 , 12] . The trypanosomal small Tims also differ from other members of this protein family regarding the complexes they form . All six trypanosomal small Tims were found to associate with the single trypanosomal TIM complex , irrespective of whether it is engaged in MCP import or in import of presequence-containing proteins [10] . Since presequence-containing proteins generally are less hydrophobic than MCPs or other small Tim substrates , they do not rely on chaperones for the passage of the aqueous IMS . In line with this , it was not possible to show an interaction between small Tims of yeast or human and the respective presequence translocase , the TIM23 complex , even when using highly sensitive SILAC-based quantitative MS of co-immunoprecipitations [36–38] . Thus , the association of all six small Tims with the single trypanosomal TIM complex that is in the process of translocating presequence-containing substrates is very unusual . Besides the TIM complex-associated fraction of small Tims , we demonstrated that all small Tims analyzed in our study ( Tim9 , TbTim11 , TbTim12 and TbTim13 ) are also present in soluble complexes of approximately 70 kDa in mass . This suggests that trypanosomal small Tims form soluble hetero-hexameric assemblies just as small Tims in other organisms [20 , 21 , 23 , 24] . However , the fact that all three SILAC pulldown experiments recover all six members of the small Tim protein family indicates that the postulated hexamers , unlike the ones in yeast and human , are not composed of specific pairs of small Tims . Moreover , in all cases tested , ablation of specific small Tim proteins did not result in the disappearance of the soluble , putative hexameric complexes containing other small Tim subunits . These results are most easily explained by the existence of multiple heterogeneous complexes of highly variable small Tim compositions . Thus , regarding hexamer formation , the small Tims may have at least in part redundant functions . The fact that at least four out of the six trypanosomal small Tims are individually essential for normal growth suggests that their essential function is not linked to the soluble complexes they form but to their tight association with the membrane integrated TIM translocase . Most organisms contain a whole suite of small Tim proteins . S . cerevisiae has five such proteins , whereas humans and trypanosomes have six members of this protein family [11 , 23] . The parasitic apicomplexan Cryptosporidium is an interesting case . It has mitosomes that lack organellar DNA , are not able to perform oxidative phosphorylation and have a highly reduced proteome . A bioinformatic analysis shows that it underwent reductive evolution resulting in a rudimentary mitochondrial protein import system which contains a single small Tim only [39] . Interestingly , this protein was able to form homo-hexameric assemblies . Moreover , even when imported into the IMS of yeast , where it would have the opportunity to interact with endogenous yeast small Tims , it only assembled with itself [39] . Thus , there appear to be three types of small Tim hexamers in nature: i ) the most simple one found in Cryptosporidium consisting of a single small Tim subunit only , ii ) the standard hexamers formed by two different small Tims arranged in alternating order in yeast and mammals and iii ) the small Tim hexamers of trypanosomes which likely have highly variable subunit compositions . This suggests that while the subunits of the soluble IMS chaperone complexes all belong to the conserved small Tim protein family , the quaternary structure formed by them is quite variable in different species . We functionally analyzed TbTim11 , TbTim12 and TbTim13 by RNAi-mediated knockdown . In general , all three cell lines were found to phenocopy the mitochondrial defects observed in TbTim17 RNAi . This is explained by the fact that ablation of any of the three small Tims caused the rapid disappearance of the TIM complex . In the case of TbTim11 and TbTim12 we could demonstrate that the two proteins are required for import of TbTim17 , a core component of the TIM complex . This is in agreement with results from yeast and human mitochondria where an involvement of small Tims in import of members of the Tim17/22/23 has repeatedly been shown [15 , 24 , 31–33] . Ablation of TbTim13 however does not affect import of TbTim17 or of other integral membrane subunits of the TIM complex , indicating that it plays a direct role in the assembly and/or maintenance of the TIM complex . This is a non-canonical function of small Tim proteins that has not been reported before . In summary , while the number and primary structure of trypanosomal small Tims—except for TbTim12 which lacks a complete twin Cx3C motif—are similar to small Tims of other organisms , the quaternary structure of their complexes is very different . Moreover , at least one trypanosomal small Tim has a non-canonical function in TIM complex biogenesis , emphasizing the need to study this important protein family in non-standard model systems that are not closely related to yeast and mammals .
Transgenic procyclic cell lines are based on T . brucei 29–13 [40] and were grown at 27°C in SDM-79 supplemented with 10% ( v/v ) fetal calf serum ( FCS ) . C-terminal epitope tagging was done by fusing the full length open reading frames ( ORFs ) of TbTim11 ( Tb927 . 5 . 3340 ) , TbTim12 ( Tb927 . 4 . 3430 ) and TbTim13 ( Tb927 . 10 . 11520 ) ( numbers appended to TbTim correspond to molecular weight ) to C-terminal triple c-myc- or HA-tags . The fragments encoding the tagged proteins were inserted into modified pLew100 vectors [40] in which the phleomycin resistance gene had been replaced by either the puromycin or the blasticidin resistance gene [41] . RNAi constructs were prepared using stem-loop inserts , the loop being a 439 bp spacer fragment that were integrated into the same pLew100 vectors described above [42] . For TbTim11 , TbTim12 and TbTim13 RNAi cell lines targeting the respective 3’UTRs were established using the primers described in S4 Table . Knockdown of TbERV1 ( Tb927 . 9 . 6060 ) was achieved by RNAi against the ORF ( nt 125–562 ) . TbTim17 was HA-tagged in situ according to published procedures [41] in the background of different RNAi cell lines . Polyclonal rabbit antiserum targeting TbTim12 was produced using the following peptide antigen: TbTim12 , aa 92–109 ( EKARVEMMTQQARKELSR ) . For Western blots ( WB ) the TbTim12 antiserum was used at 1:100 dilution . The specificity was confirmed using WB of whole cell extracts of the uninduced and induced RNAi cell line . Commercially available antibodies were: mouse anti-c-myc ( Invitrogen , Product No . 132500; dilution WB 1:2 , 000 ) ; mouse anti-HA ( Enzo Life Sciences AG , Product No . CO-MMS-101 R-1000 , dilution WB 1:5 , 000 ) ; mouse anti-EF1a ( Merck Millipore , Product No . 05–235 , dilution WB 1:10 , 000 ) . Antibodies previously produced in our lab are: polyclonal rabbit anti-VDAC ( dilution WB 1:1 , 000 ) ; polyclonal rabbit anti-ATOM40 ( dilution WB 1:10 , 000 , IF 1:1 , 000 ) ; polyclonal rabbit anti-CoxIV ( dilution WB 1:1 , 000 ) ; polyclonal rabbit anti-Cyt C ( dilution WB 1:100 ) and polyclonal rabbit anti-Tim9 ( dilution WB 1:20 ) ; polyclonal rat anti-TbTim17 ( dilution WB 1:150 ) [8 , 10 , 43] . Secondary antibodies used: goat anti-rat IRDye 680LT conjugated ( LI-COR Biosciences , P/N 925–68029 , dilution WB 1:10 , 000 ) ; goat anti-mouse IRDye 680LT conjugated ( LI-COR Biosciences , P/N 926–68020 , dilution WB 1:20 , 000 ) ; goat anti-Rabbit IRDye 800CW conjugated ( LI-COR Biosciences , P/N 926–32211 , dilution WB 1:20 , 000 ) ; goat anti-rabbit FITC conjugated ( Sigma , P/N F0382 , dilution IF 1:100 ) . Immunoblots of BN-PAGEs and the tagged small Tims in 2D BN-PAGEs were decorated by HRP-coupled goat anti mouse ( Sigma ) as secondary antibodies , dilution 1:5000 . To generate crude mitochondria enriched fractions by selective lysis of the plasma membrane [44] , 1∙108 cells were incubated for 10 min on ice in 20 mM Tris-HCl pH 7 . 5 , 0 . 6 M sorbitol , 2 mM EDTA containing 0 . 015% ( w/v ) digitonin . After centrifugation ( 6 , 800 g , 4°C ) , the resulting mitochondria enriched pellet was separated from the supernatant and subjected to SDS-PAGE ( 2 . 5 106 cell equivalents of each fraction ) and immunoblotting to demonstrate mitochondrial localization of a protein of interest . Alternatively , the mitochondria enriched pellet was used for further experiments . For visualization of tagged proteins , the respective cell lines were induced for 24 h with tetracycline . Crude mitochondrial fractions produced as described above were further incubated with 0 . 2% digitonin ( 20 mM Tris-HCl pH 7 . 5 , 0 . 6 M sorbitol , 2 mM EDTA ) for 15 min on ice to selectively open the outer mitochondrial membrane . By centrifugation ( 6 , 800 g , 4°C ) the supernatant containing released IMS proteins could be separated from the membrane fraction in the pellet . The individual fractions were subjected to further experiments . To separate soluble proteins from membrane-attached ones , a mitochondria enriched pellet fraction obtained by digitonin extraction was resuspended in 100 mM Na2CO3 at either pH 11 . 5 or pH 10 . 7 , incubated on ice for 10 min and centrifuged ( 100 , 000 g , 4°C , 10 min ) . Equal cell equivalents of all samples were analyzed by SDS-PAGE und immunoblotting . For co-immunoprecipitations of tagged TbTim11 , TbTim12 and TbTim13 , digitonin-extracted crude mitochondrial fractions or submitochondrial fractions of 1∙108 cells each were solubilized for 15 min at 4°C in 20 mM Tris-HCl , pH 7 . 4 , 0 . 1 mM EDTA , 100 mM NaCl , 10% glycerol containing 1% ( w/v ) digitonin and 1X Protease Inhibitor mix ( EDTA-free , Roche ) . Following a clearing spin ( 20 , 000 g , 15 min , 4°C ) , the lysate ( load ) was transferred to affinity purification resin ( 30 μl EZview red anti-c-myc affinity gel from Sigma or 50 μl anti-HA affinity matrix from Roche ) that had been equilibrated in wash buffer ( 20 mM Tris-HCl , pH 7 . 4 , 0 . 1 mM EDTA , 100 mM NaCl , 10% glycerol containing 0 . 2% ( w/v ) digitonin ) . After 2 h of incubation at 4°C , the supernatant ( unbound proteins ) was removed and the resin was washed 3 times with 500 μl wash buffer . To elute the bound proteins , the resin was boiled for 5 min in 2% SDS in 60 mM Tris-HCl pH 6 . 8 ( eluate ) . The resulting samples were analyzed by SDS-PAGE and Western blotting . SILAC co-immunoprecipitation experiments were essentially done as described before [10] . T . brucei cell lines allowing inducible expression of C-terminally HA-tagged TbTim11 , TbTim12 and TbTim13 were used . Expression of the tagged small Tim proteins was induced for 1 d . Cells were grown in SDM-80 medium containing either 12C614Nx ( "light" ) or 13C615Nx ( "heavy" ) arginine and lysine . Digitonin-extracted IMS-containing fractions were processed for immunoprecipitation as described above ( see "Co-immunoprecipitation" ) . Bound proteins were eluted by boiling the resin for 5 min in 60 mM Tris HCl , pH 6 . 8 , containing 0 . 1% SDS and separated by SDS-PAGE on a 4–12% NuPAGE BisTris gradient gel ( Life Technologies ) . Proteins were stained using colloidal Coomassie Brilliant Blue and gel lanes cut into 10 slices . For each bait protein , three biological replicates including a label-switch were performed . Sample preparation including reduction , alkylation , and tryptic in-gel digestion of proteins as well as LC-MS measurements using an Orbitrap Elite mass spectrometer and quantitative MS data analysis with MaxQuant/Andromeda ( version 1 . 5 . 5 . 1 [45];[46] ) were performed as described before [10] . The mean log10 of protein abundance ratios ( TbTim11/12/13-HA versus control ) and the p-value ( one-sided Student's t-test ) across ≥ 2 biological replicates were determined . See S2 Table for information about proteins identified ( overall sequence coverage ≥ 4% ) and quantified in SILAC co-immunoprecipitation experiments . For analysis of the mitochondrial membrane potential , uninduced and induced TbTim11 , TbTim12 and TbTim13 RNAi cell lines were grown for 15 min in the presence of 250 nM MitoTracker Red CMXRos . Cells were harvested , washed with PBS , fixed with 4% paraformaldehyde in PBS and permeabilized with 0 . 2% Triton X-100 in PBS . Between the different incubation steps with primary and secondary antibodies , cells were repeatedly washed with PBS . After postfixation in cold methanol , the slides were mounted using VectaShield containing 4' , 6-diamidino-2-phenylindole ( DAPI ) ( Vector Laboratories , P/N H-1200 ) . Images were acquired with a DFC360 FX monochrome camera ( Leica Microsystrems ) mounted on a DMI6000B microscope ( Leica Microsystems ) . Image analysis was done using LAS X software ( Leica Microsystems ) . Isolation of total RNA from uninduced and induced ( 2 days ) RNAi cells was performed using acid guanidinium thiocyanate-phenol-chloroform extraction [47] . RNA was separated on a 1% agarose gel in 20 mM MOPS buffer , pH7 . 0 containing 0 . 5% formaldehyde and transferred to a Nylon membrane ( GeneScreen Plus , PerkinElmer ) by passive diffusion . Northern probes were prepared from gel-purified PCR products corresponding to the RNAi inserts described above and radioactively labelled using the Prime-a-Gene labelling system ( Promega ) . Mitochondria enriched fractions or submitochondrial fractions were used as starting material and solubilized in 20 mM Tris-HCl pH 7 . 4 , 50 mM NaCl , 10% glycerol , 0 . 1 mM EDTA containing 1% ( w/v ) digitonin . The lysates were cleared by centrifugation prior to separation on 4–13% gradient gels for 1D BN-PAGE of the TIM complex and 6–16 . 5% BN-PAGE for 2D PAGE . For the 2D PAGE the protein complex containing lane was excised , incubated in SDS sample buffer for 15 min at room temperature , boiled for 20 s and again incubated for 20 min at room temperature before the second dimension separation on a SDS-PAGE ( 14% ) . To facilitate transfer of proteins to membranes , BN-PAGE and 2D BN-PAGE gels were incubated in SDS-PAGE running buffer ( 25 mM Tris , 1 mM EDTA , 190 mM glycine , 0 , 05% ( w/v ) SDS ) prior to Western blotting . Mitochondria enriched fractions from induced and uninduced TbTim13 RNAi cells were resuspended in 8 M urea/50 mM NH4HCO3 and processed for liquid chromatography-mass spectrometry ( LC-MS ) analysis as described previously [34] with slight modifications . In brief , proteins were reduced , alkylated , and tryptically digested followed by differential stable isotope dimethyl labeling [48] of desalted peptides using "light" formaldehyde ( CH2O ) and sodium cyanoborohydride ( NaBH3CN ) or the "heavy" , deuterated counterparts ( CD2O/NaBD3CN ) . The experiment was performed in biological triplicates including a label-switch . Light and heavy dimethyl-labeled peptides each derived from 10 μg of protein of mitochondrial fractions from induced and uninduced TbTim13 RNAi cells were mixed , purified using StageTips , and fractionated by high pH reversed-phase ( RP ) chromatography . To this end , acidified peptide mixtures were loaded onto StageTips , washed twice with 0 . 5% ( v/v ) acetic acid and eluted step-wise with 0% , 3% , 6% , 10% , 13% , 15 . 6% , 18 . 7% , and 72% ( v/v each ) acetonitrile in 10 mM NH4OH . LC-MS analyses , performed using an Orbitrap Elite ( Thermo Fisher Scientific , Bremen , Germany ) coupled to an UltiMate 3000 RSLCnano HPLC system ( Thermo Fisher Scientific , Dreieich , Germany ) , and subsequent MS data analysis using the MaxQuant/Andromeda software platform ( version 1 . 5 . 5 . 1 [45];[46] ) with parameters specific for stable dimethyl labeling were carried out as described by [34] . The mean log2 of normalized protein abundance ratios ( induced/uninduced ) and the p-value ( two-sided Student's t-test ) across ≥ 2 biological replicates were determined . See S3 Table for information about proteins identified ( overall sequence coverage ≥ 4% ) and quantified in this analysis .
|
Trypanosoma brucei and its relatives are prominent pathogens causing human and animal diseases , which mainly affect developing countries . The single mitochondrion of trypanosomes is essential across its entire life cycle . Most organellar proteins are imported by hetero-oligomeric protein complexes in the two mitochondrial membranes . Interestingly , the composition of the two import machineries is remarkably different from their corresponding counterparts in other organisms . In contrast , chaperones termed small Tims that escort hydrophobic proteins across the aqueous intermembrane space are conserved in almost all eukaryotes including trypanosomes . Here we show that a fraction of them interact tightly with the inner membrane translocase . Another fraction is present as soluble 70 kDa complexes , which likely consists of hexamers of small Tims without a defined subunit composition . In other eukaryotes these hexamers are usually composed of two alternating small Tims . Moreover , while some small Tims are involved in import of a core subunit of the inner membrane protein translocase , we found one small Tim that directly mediates the assembly of the translocase complex . Knowing which components of the trypanosomal protein import systems are conserved and which ones are not is essential to evaluate whether mitochondrial protein import might be a suitable drug target .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"rna",
"interference",
"polyacrylamide",
"gel",
"electrophoresis",
"blue",
"native",
"polyacrylamide",
"gel",
"electrophoresis",
"membrane",
"proteins",
"fungi",
"immunoprecipitation",
"chaperone",
"proteins",
"mitochondria",
"epigenetics",
"bioenergetics",
"co-immunoprecipitation",
"cellular",
"structures",
"and",
"organelles",
"gel",
"electrophoresis",
"research",
"and",
"analysis",
"methods",
"electrophoretic",
"techniques",
"genetic",
"interference",
"proteins",
"gene",
"expression",
"cell",
"membranes",
"precipitation",
"techniques",
"yeast",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"integral",
"membrane",
"proteins",
"genetics",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"organisms"
] |
2017
|
A trypanosomal orthologue of an intermembrane space chaperone has a non-canonical function in biogenesis of the single mitochondrial inner membrane protein translocase
|
Microbial pathogenesis studies traditionally encompass dissection of virulence properties such as the bacterium's ability to elaborate toxins , adhere to and invade host cells , cause tissue damage , or otherwise disrupt normal host immune and cellular functions . In contrast , bacterial metabolism during infection has only been recently appreciated to contribute to persistence as much as their virulence properties . In this study , we used comparative proteomics to investigate the expression of uropathogenic Escherichia coli ( UPEC ) cytoplasmic proteins during growth in the urinary tract environment and systematic disruption of central metabolic pathways to better understand bacterial metabolism during infection . Using two-dimensional fluorescence difference in gel electrophoresis ( 2D-DIGE ) and tandem mass spectrometry , it was found that UPEC differentially expresses 84 cytoplasmic proteins between growth in LB medium and growth in human urine ( P<0 . 005 ) . Proteins induced during growth in urine included those involved in the import of short peptides and enzymes required for the transport and catabolism of sialic acid , gluconate , and the pentose sugars xylose and arabinose . Proteins required for the biosynthesis of arginine and serine along with the enzyme agmatinase that is used to produce the polyamine putrescine were also up-regulated in urine . To complement these data , we constructed mutants in these genes and created mutants defective in each central metabolic pathway and tested the relative fitness of these UPEC mutants in vivo in an infection model . Import of peptides , gluconeogenesis , and the tricarboxylic acid cycle are required for E . coli fitness during urinary tract infection while glycolysis , both the non-oxidative and oxidative branches of the pentose phosphate pathway , and the Entner-Doudoroff pathway were dispensable in vivo . These findings suggest that peptides and amino acids are the primary carbon source for E . coli during infection of the urinary tract . Because anaplerosis , or using central pathways to replenish metabolic intermediates , is required for UPEC fitness in vivo , we propose that central metabolic pathways of bacteria could be considered critical components of virulence for pathogenic microbes .
Traditional studies of bacterial pathogenesis have focused on pathogen-specific virulence properties including toxins , adhesins , secretion , and iron acquisition systems , and mechanisms to avoid the innate and adaptive immune response . Examining bacterial metabolism during the course of an infection is also critical to further our understanding of pathogenesis and identifying potential targets for new antimicrobial agents . Infectious diseases represent a serious threat to global health because many bacteria that cause disease in humans such as Staphylococcus aureus , Mycobacterium tuberculosis , and E . coli are steadily developing resistance to many of the available treatments [1]–[3] . Since the introduction of antibiotics in the last century , the emergence of bacteria that resist these compounds has rapidly outpaced the discovery and development of new antimicrobial agents [4] . The need to understand bacterial physiology during infection of the host is critical for the development of new antimicrobials or antibiotics that will reduce their burden upon human health . Among common infections , urinary tract infections ( UTI ) are the most frequently diagnosed urologic disease . The majority of UTIs are caused by E . coli and these uropathogenic E . coli ( UPEC ) infections place a significant financial burden on the healthcare system by generating annual costs in excess of two billion dollars [5] , [6] . Because UTIs are a significant healthcare burden and E . coli is one of the best studied model organisms for studying metabolism , these traits can be exploited to understand and identify metabolic pathways that are required for the growth of the bacterium during infection of the host . Despite being arguably the most studied organism , E . coli metabolism during colonization of the intestine has only recently been explored [7] , [8] . Commensal E . coli acquires nutrients from intestinal mucus , a complex mixture of glycoconjugates , and subsequently expresses genes involved in the catabolism of N-acetylglucosamine , sialic acid , glucosamine , gluconate , arabinose and fucose [8] , [9] . E . coli mutants in the Entner-Doudoroff and glycolytic central metabolic pathways have diminished colonization levels reflecting the importance of sugar acid catabolism [8] . These findings suggest that commensal E . coli uses multiple limiting sugars for growth in the intestine [8] . Together , this developing body of evidence supports the assertion that E . coli grows in the intestine using simple sugars released by the breakdown of complex polysaccharides by anaerobes [9] , [10] . Much less is known about the metabolism of enteric pathogens during colonization of the gastrointestinal tract . Enterohemorrhagic E . coli ( EHEC ) O157∶H7 requires similar carbon metabolic pathways as do commensal strains , however , mutations in pathways that utilize galactose , hexuronates , mannose , and ribose resulted in colonization defects only for EHEC [9] . It was also found that multiple mutations in a single EHEC strain had an additive effect on colonization levels suggesting that this pathogen depends on the simultaneous metabolism of up to six sugars to support the colonization of the intestine [9] . When faced with limiting sugars due to consumption by other colonizing bacteria , EHEC may switch from glycolytic to gluconeogenic substrates to sustain growth in the intestine [11] . Synthesis and degradation of glycogen , an endogenous glucose polymer , plays an important role for EHEC and pathogenic Salmonella during colonization of the mouse intestine presumably by functioning as an internal carbon source during nutrient limitation [12]–[14] . Although it is not known which external carbon sources are used by S . enterica serovar Typhimurium during colonization it has been demonstrated that full virulence requires the conversion of succinate to fumarate in the tricarboxylic acid ( TCA ) cycle [15] , [16] . These studies have contributed much to the understanding of the in vivo metabolic requirements of EHEC colonization; however , these studies were done in an animal model that is not suitable for studying pathogenesis because these animals do not exhibit signs of EHEC infection [9] , [11] , [13] . In contrast to the nutritionally diverse intestine , the urinary tract is a high-osmolarity , moderately oxygenated , iron-limited environment that contains mostly amino acids and small peptides [17] , [18] . The available studies on UPEC metabolism during UTI has revealed that the ability to catabolize the amino acid D-serine in urine , which not only supports UPEC growth , appears important as a signaling mechanism to trigger virulence gene expression [19] , [20] . Metabolism of nucleobases has been demonstrated to play a role for UPEC colonization of the urinary tract; signature-tagged mutagenesis screening identified a mutant in the dihydroorotate dehydrogenase gene pyrD that was outcompeted by wild-type UPEC in vivo [21] and in a separate transposon screen a gene involved in guanine biosynthesis , guaA , was identified and found to be attenuated during experimental UTI [22] . To better understand bacterial metabolism during infection , we used a combination of comparative proteomics and systematic disruption of central metabolism to identify pathways that are required for UPEC fitness in vivo . By examining the expression of UPEC cytoplasmic proteins during growth in human urine , we confirmed that E . coli is scavenging amino acids and peptides and found that disruption of peptide import in UPEC significantly compromised fitness during infection . Consistent with the notion that peptides are a key in vivo carbon source for UPEC , only mutations ablating gluconeogenesis and the TCA cycle demonstrated reduced fitness in vivo during experimental UTI . These findings represent the first study of pathogenic E . coli central metabolism in an infection model and further our understanding of the role of metabolism in bacterial pathogenesis .
Culturing UPEC in human urine partially mimics the urinary tract environment and has proven to be a useful tool to identify bacterial genes and proteins involved in UTI [18] , [22]–[24] . Because it is well established that urine is iron-limited and our previous studies clearly demonstrated that the majority of differentially expressed genes and proteins are involved in iron acquisition [18] , [23] , we determined the protein expression profile of E . coli CFT073 during growth in human urine and compared that with bacterial cells cultured in iron-limited LB medium to unmask proteins involved in processes other than iron metabolism . Using this strategy and 2D-DIGE it was possible to visualize 700 cytoplasmic protein spots , 84 of which were differentially expressed ( P<0 . 05 ) between urine and iron-limited LB medium ( Fig . 1 ) . Of these , 56 were more highly expressed in human urine ( green ) than in iron-limited LB medium , while 28 demonstrated greater expression in iron-limited LB medium ( red ) than in urine ( Fig . 1 ) . Proteins induced in human urine with >2-fold differences from expression levels in iron-limited LB medium were identified by tandem mass spectroscopy ( Table 1 ) . The results indicate that E . coli growing in urine are expressing proteins involved in the catabolism of pentose sugars; XylA ( xylose isomerase ) , AraF ( high-affinity arabinose-binding protein ) , and the non-oxidative pentose phosphate pathway enzyme TalA ( transaldolase ) were induced 5 . 25- , 2 . 02- , and 5 . 66-fold , respectively ( P<0 . 001 ) ( Table 1 ) . Other proteins that were induced are the involved in metabolism of the sugar acids gluconate ( UxuA , mannonate dehydratase ) , gluconolactone ( YbhE , 6-phosphogluconolactonase ) , sialic acid ( NanA , N-acetylneuraminate lyase ) , and fructose ( FruB , fructose-specific IIA/FPr PTS system component ) . Multiple isoforms of the periplasmic dipeptide and oligopeptide substrate-binding proteins DppA and OppA were also induced ( >2-fold , P<0 . 009 ) in urine confirming the notion that amino acids and small peptides are being acquired from this milieu ( Table 1 ) . Proteins involved in amino acid metabolism were also identified and include SerA ( D-3-phosphoglycerate dehydrogenase ) that is involved in serine biosynthesis and two enzymes in the arginine biosynthesis pathway , ArgG ( argininosuccinate dehydrogenase ) and SpeB ( agmatinase ) ( Table 1 ) . As expected , none of the proteins identified were involved in iron uptake or metabolism , although DppA has been reported to bind heme albeit with less affinity than dipeptide substrates [25] . Notably , there was an increase in abundance for two central metabolism enzymes , TalA , as mentioned above , and TpiA that was increased 4 . 58-fold ( P<0 . 0001 ) in urine ( Table 1 ) . TalA , a non-oxidative pentose phosphate pathway enzyme , converts sedoheptulose-7-phosphate and glyceraldehyde-3-phosphate to erythrose-4-phosphate and fructose-6-phosphate . Due to the transfer of the glycolytic intermediate glyceraldehyde-3-phosphate by TalA , this enzyme is an important link between the pentose phosphate pathway and glycolysis [26] . TpiA is a glycolytic enzyme that catalyzes the reversible isomerization of glyceraldehyde-3-phosphate and dihydroxyacetone phosphate [27] . The induction of TalA and TpiA suggested that the coupling of the pentose phosphate pathway and glycolysis or gluconeogenesis via the transfer and isomerization of glyceraldehyde-3-phosphate may be an important route of carbon flux through these central pathways during the bacterium's growth in human urine . To determine whether some proteins identified by 2D-DIGE are required for UPEC fitness during UTI , CFT073 mutants were constructed in the genes: talA , xylA , tpiA , serA , speB , uxuA , nanA , argG , araF , dppA , and oppA . For these studies , an experimental competition between each mutant strain and wild-type parental CFT073 was performed . Wild-type UPEC and the mutant strain were prepared in a 1∶1 ratio and transurethrally inoculated into the bladders of mice . The number of mutant ( kanamycin-resistant ) and wild-type ( kanamycin-sensitive ) bacteria recovered from the bladder and kidneys was determined by plating the tissue homogenates for CFU on both LB agar and LB agar containing kanamycin . Mutants containing defects in genes that affect fitness in vivo are out-competed by the wild-type strain when inoculated into the same animal . This was determined by comparing the ratio of colony forming units ( CFU ) of bacteria recovered from the infection to the ratio of bacteria contained within the inoculum to obtain a competitive index ( CI ) . A CI>1 indicates the wild-type out-competes the mutant strain and a CI<1 indicates the wild-type is out-competed by the mutant . In these series of experimental infections , only mutants defective in peptide transport ( ΔdppA and ΔoppA ) were dramatically out-competed by wild-type UPEC in vivo , CI>50 , P<0 . 005 for the bladder ( Table 2 ) . One additional mutant , ΔtpiA , that functions in both glycolysis and gluconeogenesis , was out-competed by wild-type in the kidneys at 48 hpi , CI = 2 . 54 , P = 0 . 0206 ( Table 2 ) . Despite the lack of attenuation in vivo for the many of the mutants , these results reveal a number of important findings . The agmatinase mutant ΔspeB out-competed wild-type in the bladder at 48 hpi , CI = 0 . 14 , P = 0 . 0122 ( Table 2 ) . Agmatinase is part of arginine metabolism and catalyzes the formation of the polyamine putrescine and urea from agmatine and H2O . This suggests that accumulation of agmatine or reduced production of urea and putrescine by the mutant may provide a modest advantage over wild-type UPEC during infection of the bladder . CFT073 ΔargG was unable to grow in MOPS defined medium unless supplemented with 10 mM arginine ( Fig . 2A ) , validating the expected auxotrophic phenotype . Similarly , the ΔserA serine auxotroph required supplementation with either 10 mM serine or glycine in MOPS , D-serine was unable to rescue the in vitro growth defect ( Fig . 2B ) . Lack of arginine or serine biosynthesis had little effect upon the ability of UPEC to grow logarithmically in human urine , although the ΔargG mutant consistently entered stationary phase at a lower cell density , with an O . D . 600 of 0 . 45±0 . 04 compared to 0 . 59±0 . 03 for wild-type ( P = 0 . 051 ) ( Fig . 2C ) . When tested for in vivo fitness , neither the ΔargG nor ΔserA strain were significantly out-competed by wild-type UPEC at 48 hpi ( Fig . 2D , 2E , and Table 2 ) . Additionally , there was no preference for serine over arginine or vice versa for UPEC colonization at 48 hpi . When the auxotrophic strains were co-inoculated into the same mice both mutants were recovered at similar levels ( Fig . 2F ) . These data clearly demonstrate that there are sufficient concentrations of arginine , serine and/or glycine in the urinary tract to support growth of these auxotrophic strains . As mentioned , deletion of the genes encoding periplasmic peptide substrate-binding proteins , dppA and oppA , had the greatest impact on UPEC fitness in vivo of the CFT073 mutants in genes whose products were induced during growth in human urine ( Table 2 ) . The dipeptide transport mutant , ΔdppA , failed to maintain colonization in the bladder at 48 hpi , 11/11 bladders had undetectable levels ( <200 CFU/g ) for this mutant , while wild-type levels from the same bladders reached a median of 104 CFU/g ( P = 0 . 0020 ) ( Fig . 3A ) . Because these mice had low levels of recoverable UPEC from the kidneys it was not possible to determine the contribution of dipeptide transport for kidney colonization . Import of oligopeptides via the OppA substrate-binding protein is also required for UPEC fitness in vivo . CFT073 ΔoppA was out-competed nearly 500∶1 wild-type∶mutant in the bladder ( Table 2 ) with a 3-log reduction in the median CFU/g from bladder tissue at 48 hpi ( P = 0 . 0047 ) ( Fig . 3B ) . In these co-challenge infections , wild-type UPEC colonized 10/16 ( 62% ) of kidneys , while ΔoppA was detectable in 4/16 ( 25% ) of kidneys at 48 hpi . The ratio of wild-type∶mutant recovered from the kidneys at this time point was 156∶1 ( Table 2 ) where wild-type UPEC had 3-logs greater CFU/g than ΔoppA ( P = 0 . 0420 ) ( Fig . 3B ) . Together , the in vivo fitness defect for CFT073 harboring a deletion of either dppA or oppA suggests that peptides may be an important carbon source for UPEC during urinary tract infection . Previously , we have shown that the low copy pGEN plasmid is maintained in CFT073 in the absence of antibiotic pressure for up to 48 h [28] . Using this ampicillin resistant plasmid system , we cloned the entire dppA gene including 200 bp upstream from the predicted start site of translation and introduced the resulting construct , pGEN-dppA , into the CFT073 ΔdppA strain . To determine if it was possible to complement the ΔdppA defect in vivo , co-challenge infections were performed as described and modified to enumerate bacteria in tissue homogenates by plating on agar containing ampicillin ( wild-type CFT073 harboring pGEN ) or ampicillin and kanamycin ( CFT073 ΔdppA containing pGEN or pGEN-dppA ) . The ΔdppA mutant containing empty vector ( pGEN- ) demonstrated the expected fitness defect in bladder colonization when co-inoculated with wild-type CFT073 ( pGEN- ) ( P = 0 . 0002 ) while ΔdppA containing a wild-type copy of dppA ( pGEN-dppA ) restored colonization to wild-type levels in the bladder at 48 hpi ( Fig . 3C ) . Although both mutant ( pGEN- ) and wild-type ( pGEN- ) demonstrated poor colonization in the kidneys of these animals , complementation of ΔdppA ( pGEN-dppA ) resulted in a 2-log increase in median kidney CFU/g at 48 hpi ( Fig . 3D ) . The requirement for peptide transport for UPEC fitness during infection implicates peptides as an important carbon source in vivo . This predicts that certain central metabolism pathways that operate during catabolism of amino acids or peptides may be more important for in vivo growth of UPEC than pathways that function primarily to catabolize sugars . To test the role of central metabolic pathways during an actual infection mutants were constructed in UPEC strain CFT073 to produce defects in glycolysis ( pgi , phosphoglucose isomerase and tpiA , triosephosphate isomerase ) [29] , the Entner-Doudoroff pathway ( edd , 6-phosphogluconate dehydratase ) [10] , the oxidative branch ( gnd , 6-phosphogluconate dehydrogenase ) and the non-oxidative branch ( talA , transaldolase ) of the pentose phosphate pathway [26] , gluconeogenesis ( pckA , phosphoenolpyruvate carboxykinase ) [30] , and the TCA cycle ( sdhB , succinate dehydrogenase ) [31] . The in vitro growth of these central metabolism mutants were examined and compared to wild-type UPEC during culture in human urine , LB medium , and MOPS defined medium containing 0 . 02% glucose . All of the central metabolism mutants produced similar logarithmic growth as wild-type when cultured in human urine ( Fig . 4A ) and LB medium ( data not shown ) under defined inoculation conditions . As expected , only mutants with defects in glycolysis demonstrated diminished growth in MOPS medium containing glucose as the sole carbon source ( Fig . 4B ) . The Δpgi strain produced an extended lag phase of 5 . 5±1 . 1 h compared with wild-type ( P = 0 . 001 ) and ΔtpiA failed to reach exponential phase after 18 h ( Fig . 4B ) . These data and the indistinguishable growth of the glycolysis mutants from wild-type in urine supported the proteomics data and indicated that UPEC growing in urine utilizes carbon sources other than glucose . To determine the role for central metabolism during E . coli infection of the urinary tract , the ascending model of murine UTI was used as described above to measure the impact that a lesion in central metabolism has upon the relative fitness of the strain in vivo . Mutants with defects in glycolysis had levels of colonization in the bladder at 48 hpi similar to wild-type ( P>0 . 400 ) ( Fig . 5A and 5B ) . In the kidneys , Δpgi CFU/g were comparable to wild-type ( Fig . 5A ) , while ΔtpiA demonstrated a 10-fold reduction in the median CFU/g ( P = 0 . 0206 ) ( Fig . 5B ) . The pentose phosphate pathway mutants , Δgnd ( Fig . 5C ) and ΔtalA ( Table 2 ) , were not significantly out-competed by wild-type in vivo . The mutant with a defect in the Entner-Doudoroff pathway ( Δedd ) also was not impaired in the ability to infect both the bladder and kidneys as indicated by its similar colonization to wild-type at 48 hpi ( Fig . 5D ) . UPEC in vivo fitness was significantly reduced in the TCA cycle mutant ΔsdhB , this mutation resulted in a 50-fold reduction in median CFU/g in the bladder ( P = 0 . 0134 ) and a 1 . 5-log decrease in kidney CFU at 48 hpi ( P = 0 . 0400 ) ( Fig . 5E ) . This defect in the TCA cycle impacted fitness to a greater extent in the bladder , where 11/15 ( 73% ) of mice had undetectable levels of mutant bacteria , than in the kidneys where 6/15 ( 40% ) mice had undetectable counts ( Fig . 5E ) . The gluconeogenesis mutant , ΔpckA had a 2-log reduction in median CFU/g in both the bladder ( P = 0 . 0005 ) and kidneys ( P = 0 . 0322 ) and half of the mice ( 7/14 ) displayed undetectable levels of ΔpckA at 48 hpi ( Fig . 5F ) . To verify that this mutation is non-polar as expected and the defect in colonization is not due to a secondary mutation , in vivo complementation experiments were conducted . The ΔpckA mutant with the pGEN empty vector demonstrated a 2-log reduction in CFU/g at 48 hpi ( P = 0 . 0039 ) in the bladder when co-inoculated into mice with wild-type UPEC containing pGEN ( Fig . 6 ) . When CFT073 ΔpckA ( pGEN-pckA ) were co-inoculated with CFT073 ( pGEN- ) there was no significant difference in bladder CFU/g at 48 hpi between the strains ( Fig . 6 ) . Thus , by re-introducing the pckA gene into the mutant it was possible to complement the ΔpckA defect in bladder colonization at 48 hpi . The in vitro growth and in vivo fitness for the UPEC central metabolism mutants is summarized in Table 3 . As expected , only mutations in glycolysis had a negative effect on growth in defined medium with glucose . Only gluconeogenesis or TCA cycle mutants demonstrated reduced persistence at 48 hpi in both the bladder and kidneys ( Table 3 ) . Non-oxidative and oxidative pentose phosphate pathway and Entner-Doudoroff pathway mutants did not demonstrate any colonization defect and of the glycolytic mutants only the triosephosphate isomerase deletion had a measurable defect in the kidneys but not in the bladder ( Table 3 ) . Together , the fitness defect for the peptide transport mutants and these data indicate UPEC could be using amino acids as the primary carbon source during infection . Surprisingly , there was no correlation between the ability of the central metabolism mutants to grow in human urine ex vivo and grow in the urinary tract in vivo .
Bacterial pathogenesis traditionally involves studying virulence traits involved in the production of toxins and effectors , iron acquisition , adherence , invasion , and immune system avoidance . Although many paradigms exist that describe mechanisms of pathogenesis , the contribution of microbial metabolism to bacterial virulence during an infection is less understood . Much work has been done studying E . coli as model organism for characterizing individual central metabolism pathways and enzymes [10] , [27] , [32]–[38] . We have shown here that central metabolism studies in E . coli can be extended to investigate the contribution of central pathways to bacterial pathogenesis using a virulent uropathogenic E . coli strain and a well-established animal model of UTI . It is known that commensal E . coli require the Entner-Doudoroff pathway and glycolysis for colonization in vivo; while the TCA cycle , pentose phosphate pathway , and gluconeogenesis are dispensable in the intestine [8] . In contrast , we have shown that during E . coli infection of the urinary tract , the pathways required for commensal colonization are dispensable while the TCA cycle and gluconeogenesis are necessary for UPEC fitness in vivo . Adaptation to distinct host environments has been previously shown to involve shared traits between commensal and pathogenic strains [39] , [40] . Because commensal E . coli are an important natural component of the intestine one concern faced when developing antimicrobials that target pathogenic strains is how to avoid eradicating commensal bacteria . Thus , these findings highlight important differences between commensal and pathogenic E . coli that could be exploited for the development of antimicrobials that target these pathways in this pathogen during infections that may not affect commensal strains . Interestingly , in addition to UPEC , gluconeogenesis is required for virulence in microbes that represent an array of pathogenic lifestyles , from intracellular bacteria and parasites [41] , [42] , plant-pathogenic [43] , and intestinal pathogens [16]; suggesting that anaplerosis may be a common mechanism of microbial pathogenesis . This study comprehensively examines the role of pathogenic E . coli central metabolism in a disease model and provides insight not only into UPEC metabolism in vivo but also information regarding the nutrients available to support the growth of E . coli within the urinary tract . The proteomics experiments did reveal that UPEC growing in human urine induces expression of multiple isoforms of both dipeptide- and oligopeptide-binding proteins , both of which were found to be required for UPEC to effectively colonize the urinary tract . This indicates that these bacteria actively import short peptides in urine and this function may indicate that peptides are an important carbon source in vivo . Consistent with this , only bacteria with defects in peptide transport , gluconeogenesis , or the TCA cycle demonstrated a significant reduction in fitness in vivo in both the bladder and kidneys . These findings suggest a model that describes the biochemistry of E . coli during UTI . For optimal growth during infection , short peptides are taken up by UPEC and degraded into amino acids that are catabolized and used in a series of anaplerotic reactions that replenish TCA cycle intermediates and generate gluconeogenesis substrates ( Fig . 7 ) . Certain glycolytic steps are irreversible and the reverse gluconeogenic reaction is performed by an enzyme specific for gluconeogenesis . Carbon flux through glycolysis and gluconeogenesis must be carefully controlled by the cell to avoid a futile cycle of carbon metabolism [44] . Allosteric regulation of enzymes that catalyze irreversible reactions in these pathways and catabolite repression are mechanisms used to avoid the futile cycle [45] , [46] . A gluconeogenic-specific enzyme subject to allosteric regulation is phophoenolpyruvate carboxykinase that converts oxaloacetate to phosphoenolpyruvate [47] . Deletion of the gene pckA that encodes this enzyme resulted in a significant reduction in UPEC fitness in vivo . Because bacteria prevent glycolysis and gluconeogenesis from occurring simultaneously and deletion of pckA reduced fitness in vivo , we reason that carbon flux through gluconeogenesis during UPEC infection may be an important indication of amino acid catabolism in vivo . It is not surprising that , in addition to gluconeogenesis , the TCA cycle is also required for UPEC fitness in vivo . These two pathways are connected and collectively described as “filling in” or anaplerotic reactions . The TCA cycle is necessary to provide substrates for gluconeogenesis when cells use amino acids as a carbon source . Gluconeogenic amino acids can be degraded to oxaloacetate or to pyruvate that can be converted to acetyl-CoA and enter the TCA cycle [47] . Oxaloacetate , a TCA cycle intermediate , is converted to phophoenolpyruvate during gluconeogenesis by PckA as described above . A mutation in the TCA cycle enzyme succinate dehydrogenase , sdhB , results in a UPEC strain that has reduced fitness in vivo . This finding suggests that UPEC are growing aerobically in the urinary tract because succinate dehydrogenase is replaced by fumarate reductase during anaerobic growth and therefore , future work could confirm if the reductive TCA cycle is not operating during UPEC infection . The requirement for peptide import and the TCA cycle for UPEC fitness during infection is consistent with the hypothesis that acetyl-CoA production from the degradation of amino acids could be occurring in vivo as has been shown by another group [48] . Interestingly , with the exception of peptide-transport proteins , up-regulation of protein expression in urine ex vivo did not correlate with functional importance in vivo . This could be due to the fact that many central metabolism genes are constitutively expressed and that human urine only partially mimics the complex lifestyle of UPEC during UTI [49] . The absence of host cells and the immune response during growth in urine ex vivo could in part account for this discrepancy . It also remains possible that mutants that lack growth defects in urine but demonstrate reduced fitness in vivo could represent genes or metabolic pathways that are required for intracellular phases of growth during cystitis [50] . Despite these disadvantages , up-regulation of both DppA and OppA expression was seen in urine and loss of either dppA or oppA was found to negatively impact UPEC colonization in vivo . Induction of dppA has been reported in a hypervirulent UPEC strain that has a lacks a functional D-serine deaminase gene ( dsdA ) [51] . Deletion of dppA in this mutant strain resulted in a loss of the hypervirulent phenotype in vivo and significantly reduced its ability to colonize the urinary tract in competition with wild-type [51] . Surprisingly , in contrast to our findings , this group found that mutation of dppA alone had no effect on UPEC fitness in vivo [51] . Due to lack of complementation , it is unclear from that work why loss of dppA dramatically attenuated a hypervirulent strain but had no effect on wild-type . Despite this inconsistency in that work , the importance of peptide transport for UPEC fitness in vivo is supported by the findings that loss of either dppA or oppA significantly reduced colonization of the urinary tract and that the reduced bacterial colonization in the ΔdppA strain can be restored to wild-type levels by complementing the mutant with a wild-type dppA gene . In summary , defects in the both branches of the pentose phosphate pathway , the Entner-Doudoroff pathway , and glycolysis had limited or no impact on UPEC fitness in vivo . On the other hand , the TCA cycle- and gluconeogenesis-defective strains demonstrate significant fitness reductions during UTI . The utilization of short peptides and amino acids as a carbon source during bacterial infection of the urinary tract is supported by the observation that UPEC mutants defective in peptide import have reduced fitness in vivo while auxotrophic strains do not . Together , these findings provide compelling evidence to support the notion that catabolism of amino acids to form TCA cycle intermediates and gluconeogenic substrates is important for the ability of UPEC to infect the urinary tract efficiently . This shows that anaplerotic and central metabolism pathways are required for UPEC fitness in vivo and suggest microbial metabolism should be considered important for bacterial pathogenesis .
Strains were derived from E . coli strain CFT073 , a prototypic UPEC strain isolated from the blood and urine of a patient with acute pyelonephritis [52]; its genome has been sequenced and fully annotated [53] . Isolated colonies were used to inoculate overnight Luria-Bertani ( LB ) cultures . Bacteria from overnight cultures were collected by centrifugation , washed with sterile PBS , and 106 CFU were used to inoculate pre-warmed LB or human urine . To mimic iron-limitation in urine , LB containing 10 mM deferoxamine mesylate ( Sigma ) was used as a growth medium for comparative proteomics . For human urine cultures , mid-stream urine was collected into sterile sample containers from 8–10 male and female donors , pooled , and sterilized by vacuum filtration through a 0 . 22 µm pore filter . MOPS defined medium containing 0 . 2% glucose [54] with and without 10 mM L-arginine , L-serine , glycine , aspartatic acid , or D-serine ( Sigma ) was also used to test growth of mutant strains . Growth curves were established in triplicate using a Bioscreen bioanalyzer in 0 . 4 ml volumes; OD600 was recorded every 15 min . All cultures were incubated at 37°C; LB overnight and MOPS cultures were incubated with aeration; urine cultures were incubated statically . For preparation of proteins , UPEC isolate CFT073 was grown statically to exponential phase ( OD600 = 0 . 25 ) in pre-warmed LB or human urine at 37°C in 5×100 ml cultures for each growth medium . Bacteria were harvested from 500 ml of culture by centrifugation ( 10 , 000× g , 30 min , 4°C ) and lysed in a French pressure cell at 20 , 000 psi . Harvested cells were washed and resuspended in 10 ml of 10 mM HEPES , pH 7 . 0 containing 100 U of Benzonase ( Sigma ) . Following two passes through the chilled pressure cell , lysates were centrifuged ( 7500× g , 10 min , 4°C ) to remove unbroken cells and supernatants were ultracentrifuged ( 120 , 000× g , 1 h , 4°C ) to remove membranes and insoluble material . Soluble proteins were quantified using the 2D Quant Kit ( GE Healthcare ) following the manufacturer's protocol and either used immediately in DIGE-labeling procedures or stored at −80°C . For fluorescence difference in gel electrophoresis ( 2D-DIGE ) [55] , bacterial proteins were minimally labeled with cyanine-derived fluors ( CyDyes ) containing an NHS ester-reactive group as recommended by the manufacturer ( GE Healthcare ) . To determine quantitative differences within the UPEC soluble proteome during growth in human urine , cytoplasmic proteins prepared from human urine cultures were labeled with Cy3 , from LB broth with Cy5 , and a pooled internal standard representing equal amounts of both urine and LB preparations with Cy2 as described previously [23] . Briefly , 50 µg of protein was incubated with 400 pmol CyDye for 30 min and the reaction was stopped by added 10 mM lysine . Following labeling , samples labeled with each CyDye were pooled ( 150 µg total protein ) , mixed with an equal volume of 2× DIGE sample buffer; 7 M urea , 2 M thiourea , 10 mM tributylphosphine ( TBP ) ( Sigma ) , 2× biolytes 3–10 ( Bio-Rad ) , 2% ASB-14 and incubated on ice for 10 min . For rehydration , samples were brought to 0 . 35 ml with 1× DIGE rehydration buffer ( 7 M urea , 2 M thiourea , 5 mM TBP , 1× biolytes 3–10 , 1% ASB-14 ) and used to passively rehydrate pH 4–7 IPG strips ( Bio-Rad ) overnight at room temperature . Rehydrated IPG strips were equilibrated and subjected to isoelectric focusing for 50 , 000 V·h and second dimension SDS-PAGE on 10% gels within low fluorescence glass plates ( Jule Biotechnologies , Inc . ) and were run at a constant current of 55 mA at 4°C for 4 hr . Following SDS-PAGE , image acquisition and pixel intensity was obtained using a Typhoon scanner ( GE Healthcare ) and differential in-gel analysis and biological analysis of variance were performed using the DeCyder 6 . 5 software suite ( GE Healthcare ) . Using this software , the normalized spot volume ratios from Cy3 or Cy5 labeled spots were quantified relative to the Cy2-labeled internal standard from the same gel . The Cy2-labeled standard was then used to standardize and compare normalized volume ratios between the Cy3 and Cy5 labeled proteins between gels representing three independent experiments to generate statistical confidence for abundance changes using student's t-test and ANOVA . To identify the proteins , 500 µg of cytoplasmic proteins were focused as described above and spots of interest were excised from a colloidal Coomassie-stained 2D SDS-PAGE gel and subjected to enzymatic digestion with trypsin . Mass spectra were acquired on an Applied Biosystems 4700 Proteomics Analyzer ( TOF/TOF ) . MS spectra were acquired from 800–3500 Da and the eight most intense peaks in each MS spectrum were selected for MS/MS analysis . Peptide identifications were obtained using GPS Explorer ( v3 . 0 , Applied Biosystems ) , which utilizes the MASCOT search engine . Each MS/MS spectrum was searched against NCBInr . Tryptic digestion and tandem mass spectrometry were performed at the University of Michigan Proteome Consortium . Deletion mutants were generated using the lambda red recombinase system [56] . Primers homologous to sequences within the 5′ and 3′ ends of the target genes were designed and used to replace target genes with a nonpolar kanamycin resistance cassette derived from the template plasmid pKD4 [56] . Kanamycin ( 25 µg/ml ) was used for selection of all mutant strains . Gene deletions begin with the start codon and end with the stop codon for each gene . To determine whether the kanamycin resistance cassette recombined within the target gene site , primers that flank the target gene sequence were designed and used for PCR . After amplification , each PCR product was compared to wild-type PCR product and in cases where size-differences are negligible; PCR products were digested with the restriction enzyme EagI ( New England Biolabs ) . Both the PCR products and restriction digests were visualized on a 0 . 8% agarose gel stained with ethidium bromide . For in vivo complementation , the dppA and pckA genes were amplified from CFT073 genomic DNA using Easy-A high-fidelity polymerase ( Stratagene ) and independently cloned into pGEN-MCS [28] , [57] using appropriate restriction enzymes . The sequences of pGEN-dppA and pGEN-pckA were verified by DNA sequence analysis prior to electroporation into CFT073 ΔdppA or ΔpckA mutant strains . Six-to eight-week-old female CBA/J mice ( 20 to 22 g; Jackson Laboratories ) were anesthetized with ketamine/xylazine and inoculated transurethrally over a 30 sec period with a 50 µl bacterial suspension per mouse using a sterile polyethylene catheter ( I . D . 0 . 28 mm×O . D . 0 . 61 mm ) connected to an infusion pump ( Harvard Apparatus ) . To measure relative fitness , overnight LB cultures for CFT073 and the mutant strain were collected by centrifugation and resuspended in sterile PBS , mixed 1∶1 and adjusted to deliver 2×108 CFU per mouse . Dilutions of each inoculum were spiral plated onto LB with and without kanamycin using an Autoplate 4000 ( Spiral Biotech ) to determine the input CFU/mL . After 48 hpi , mice were sacrificed by overdose with isoflurane and the bladder and kidneys were aseptically removed , weighed , and homogenized in sterile culture tubes containing 3 ml of PBS using an OMNI mechanical homogenizer ( OMNI International ) . Appropriate dilutions of the homogenized tissue were then spiral plated onto duplicate LB plates with and without kanamycin to determine the output CFU/g of tissue . Plate counts obtained on kanamycin were subtracted from those on plates lacking antibiotic to determine the number of wild-type bacteria . Competitive indices were calculated by dividing the ratio of wild-type to mutant at 48 hpi by the ratio of wild-type to mutant input CFU/mL . Groups of 5 mice per co-challenge were used to determine defects in fitness , when a defect was apparent the co-challenge was repeated two more times with groups of 5 mice . Statistically significant differences in colonization ( P-value<0 . 05 ) were determined using a two-tailed Wilcoxon matched pairs test . All animal protocols were approved by the University Committee on Use and Care of Animals at the University of Michigan Medical School .
|
Bacteria that cause infections often have genes known as virulence factors that are required for bacteria to cause disease . Studying virulence factors such as toxins , adhesins , and secretion and iron-acquisition systems is a fundamental part of understanding infectious disease mechanisms . In contrast , little is known about the contribution of bacterial metabolism to infectious disease . This study shows that E . coli , which cause most urinary tract infections , utilize peptides as a preferred carbon source in vivo and requires some , but not all , of the central metabolic pathways to infect the urinary tract . Specifically , pathways that can be used to replenish metabolites , known as anaplerotic reactions , are important for uropathogenic E . coli infections . These findings help explain how metabolism can contribute to the ability of bacteria to cause a common infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases/urological",
"infections",
"biochemistry",
"infectious",
"diseases/bacterial",
"infections",
"microbiology/microbial",
"physiology",
"and",
"metabolism",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis"
] |
2009
|
Fitness of Escherichia coli during Urinary Tract Infection Requires Gluconeogenesis and the TCA Cycle
|
Fatty liver is a major health problem worldwide . People with hereditary deficiency of hormone-sensitive lipase ( HSL ) are reported to develop fatty liver . In this study , systemic and tissue-specific HSL-deficient mice were used as models to explore the underlying mechanism of this association . We found that systemic HSL deficient mice developed fatty liver in an age-dependent fashion between 3 and 8 months of age . To further explore the mechanism of fatty liver in HSL deficiency , liver-specific HSL knockout mice were created . Surprisingly , liver HSL deficiency did not influence liver fat content , suggesting that fatty liver in HSL deficiency is not liver autonomous . Given the importance of adipose tissue in systemic triglyceride metabolism , we created adipose-specific HSL knockout mice and found that adipose HSL deficiency , to a similar extent as systemic HSL deficiency , causes age-dependent fatty liver in mice . Mechanistic study revealed that deficiency of HSL in adipose tissue caused inflammatory macrophage infiltrates , progressive lipodystrophy , abnormal adipokine secretion and systemic insulin resistance . These changes in adipose tissue were associated with a constellation of changes in liver: low levels of fatty acid oxidation , of very low density lipoprotein secretion and of triglyceride hydrolase activity , each favoring the development of hepatic steatosis . In conclusion , HSL-deficient mice revealed a complex interorgan interaction between adipose tissue and liver: the role of HSL in the liver is minimal but adipose tissue deficiency of HSL can cause age-dependent hepatic steatosis . Adipose tissue is a potential target for treating the hepatic steatosis of HSL deficiency .
Disorders of lipid accumulation such as obesity and fatty liver ( hepatic steatosis ) are among the greatest risk factors for health in developed countries [1–4] . Hepatic steatosis is linked to the development of liver fibrosis , cirrhosis and cancer [5 , 6] and is rapidly increasing in prevalence [7 , 8] . Increasing interest centers on the biology of triglyceride ( TG ) -containing cytoplasmic lipid droplets , TG synthesis and TG degradation ( lipolysis ) . In adipose tissue , hormone-sensitive lipase ( HSL ) , a cytoplasmic lipase encoded by the LIPE gene , is important for lipolysis . After the initial cleavage of a TG to a diacylglycerol ( DG ) plus a fatty acid ( FA ) , performed by adipose triglyceride lipase ( ATGL ) [9] , HSL-mediated hydrolysis of DG to a monoacylglycerol ( MG ) plus a FA [10] . MG is cleaved in turn by a MG hydrolase to release glycerol plus a FA . In humans and mice , systemic ATGL deficiency causes hepatic steatosis [11 , 12] . By using liver-specific ATGL deficient mouse models , we and others further showed that ATGL deficiency in liver causes marked hepatic steatosis in mice , suggesting that the underlying mechanism of ATGL-related hepatic steatosis is liver autonomous [13 , 14] . As in ATGL deficiency , the small number of individuals with genetic deficiency of HSL reported so far also show liver steatosis in middle age [15] . The underlying mechanism of HSL-related hepatic steatosis is still elusive . HSL-deficient mice have been described and , like HSL-deficient humans , show protection from obesity , a low capacity to increase lipolysis following beta-adrenergic stimulation and higher levels of diglycerides in adipose tissue [10 , 16 , 17] . Paradoxically , some HSL-deficient mouse strains have been reported to develop hepatic steatosis [18 , 19] but other reports mention low liver fat content in HSL-deficient mice [10 , 20 , 21] . To explore the potential mechanism of HSL deficiency-related hepatic steatosis , we studied the effect of HSL deficiency on liver fat content in different mouse models . The result show that hepatic steatosis occurs with aging in HSL-deficient mice . Using three models of HSL deficiency ( systemic , hepatic and adipose ) we show that , surprisingly , unlike ATGL , liver fat levels are unrelated to liver HSL but that adipose HSL deficiency alone is sufficient to produce a similar level of hepatic steatosis as in systemic HSL deficiency .
Comparison of reports of liver fat content of HSL-deficient mice revealed that studies reporting low liver TG content [10 , 20 , 21] were performed in mice before 4 months of age , whereas all those reporting hepatic steatosis [18 , 19] were in older mice . To test the hypothesis that systemic HSL knockout ( HSLSKO ) mice develop hepatic steatosis with aging , two groups of mice were studied , aged 3 and 8 months . Three-month-old HSLSKO mice had similar body weight ( Fig 1A ) , liver weight ( Fig 1B ) and liver fat content ( Fig 1C ) to controls . However , at 8 months of age , although HSLSKO mice were lean ( Fig 1A ) , their liver mass ( Fig 1B ) and TG content ( Fig 1C ) were greater than those of controls . Therefore , available data show the age-dependent development of hepatic steatosis in HSLSKO mice . To investigate the mechanism of hepatic steatosis in HSLSKO mice , we hypothesized that HSL , which is essential for normal lipolysis in adipose tissue , might also be important for degradation of acylglycerols in liver and thus directly influence liver fat content . To test this , liver-specific HSL knockout ( HSLLKO ) mice were created . Deficiency of HSL in liver was demonstrated by the absence of detectable HSL protein S1A Fig , and very low HSL mRNA S1B Fig in liver . Surprisingly , in contrast to HSLSKO mice , HSLLKO mice were similar to normal controls in body weight ( Fig 1D ) , liver weight ( Fig 1E ) and liver TG content ( Fig 1F ) . These results proved that hepatic HSL does not contribute to fatty liver in HSL deficiency , suggesting that the mechanism of hepatic steatosis in HSL deficiency depends upon organs other than liver . Because adipose tissue is a major regulator of TG storage and of FA release , we hypothesized that HSL deficiency in adipose tissue might cause systemic metabolic changes leading to hepatic steatosis . To test this , mice with adipose HSL deficiency ( HSLAKO ) were created as described [22] . Compared to normal controls , at 3 months of age , HSLAKO mice had similar body weight ( Fig 1G ) , liver weight ( Fig 1H ) and liver TG content ( Fig 1I ) . However , at 8 months of age , HSLAKO mice showed lower body weight ( Fig 1G ) , but higher liver mass ( Fig 1H ) and higher liver TG content ( Fig 1I ) than controls . The severity of the steatosis of HSLAKO mice was similar to that observed in HSLSKO mice ( Fig 1C and 1I ) . Liver histology of 8-month-old mice confirmed these findings , showing hepatic steatosis in HSLSKO and HSLAKO mice , but not in HSLLKO mice ( Fig 2 ) . Therefore , HSL deficiency in adipose tissue alone is sufficient to cause the age-dependent hepatic steatosis observed in systemic HSL deficiency . Compared to matched controls , 3-month-old HSLAKO mice had similar body weight to controls ( Fig 1G ) , but 8-month-old HSLAKO mice had lower body weight ( Fig 1G ) . Further measurements showed that HSLAKO mice have similar fat mass as controls at 3 months ( Fig 3A ) , but lower mass at 8 months ( Fig 3B ) . Consistent with this , markers of lipogenesis and of TG synthesis in adipose tissue , including Fas , Acc1 , Cd36 , Fabp4 , Ppar-γ and Dgat2 , were similar between HSLAKO mice and the corresponding controls at 3 months ( Fig 3C ) , but significantly lower in 8-month-old HSLAKO adipose tissue than in the corresponding controls ( Fig 3D ) . To further explore lipogenesis and TG synthesis in 8-month-old HSLAKO adipose tissue , we studied the expression of key proteins of lipogenesis ( FAS ) and of TG synthesis ( DGAT2 ) . As seen in the corresponding mRNAs , FAS and DGAT2 protein levels were lower than those of controls ( Fig 3E ) . Histologically , adipose tissue of HSLAKO mice showed macrophage infiltration ( Fig 3F ) and increased levels of macrophage and inflammatory markers ( Fig 3E and 3G ) . HSLAKO white adipose tissue ( WAT ) had heterogeneity of cell size ( Fig 3H ) , with a bimodal distribution in which small adipocytes ( ≤50 μm ) and large adipocytes ( ≥150 μm ) are each significantly more prevalent than in controls ( Fig 3H ) . Together , these results show that HSL deficiency in adipose tissue causes age-related lipodystrophy , with decreased fat mass and inflammation in adipose tissue . Plasma metabolites related to energy metabolism were measured in 14-hour overnight fasted mice . The result showed that plasma glucose was lower in 3-month-old HSLAKO mice than in corresponding controls ( Fig 4A ) . No difference in plasma glucose was observed in 8-month-old mice ( Fig 4A ) . Plasma FA level was not significantly different from controls values in 3-month-old HSLAKO mice ( Fig 4B ) , but was significantly lower in 8-month-old HSLAKO mice ( Fig 4B ) . Compared to normal controls , HSLAKO mice showed lower levels of plasma TG ( Fig 4C ) and of plasma adiponectin ( Fig 4D ) both at 3 and at 8 months of age . HSLAKO mice failed to show the age-related increase in leptin levels seen in normal controls ( Fig 4E ) . Interestingly , we found that 3-month-old HSLAKO mice had lower levels of insulin than controls , but 8-month-old HSLAKO mice had higher levels ( Fig 4F ) , suggesting an age-related development of insulin resistance . To further test insulin sensitivity , we performed insulin tolerance tests in HSLAKO and control mice . Compared to normal controls , HSLAKO mice showed improved insulin sensitivity at 3 months ( Fig 5A–5B ) but were insulin resistant at 8 months of age ( Fig 5C–5D ) , demonstrating that systemic insulin resistance develops with age in HSLAKO mice . Iinsulin sensitivity is driven in large part by insulin-stimulated muscle glucose uptake , and this is suppressed by muscle fat content . Therefore , muscle fat content was measured in the HSLSKO , HSLLKO , and HSLAKO mice . As with liver TG content , HSLSKO and HSLAKO mice showed similar levels of skeletal muscle fat content to their corresponding controls at 3m , but higher levels at 8m ( Fig 5E–5F ) . This difference was not seen in 8 month old HSLLKO mice ( Fig 5G ) . Glucose tolerance was similar in HSLAKO mice and normal controls ( Fig 5H–5K ) . In summary , primary adipose deficiency of HSL in mice results in atrophy and inflammation of adipose tissue and systemic insulin resistance . Each of these has been reported to promote hepatic steatosis [23–25] . We next studied the pathways of TG and FA disposal in liver , including FA oxidation , TG export in VLDL and TG hydrolysis to see whether one or more might be affected by adipose HSL deficiency . Plasma 3-hydroxybutyrate ( 3-HB ) levels provide one indication of liver FA oxidation and ketone body production . After a 5-hour fast , 3-HB levels were similar in HSLAKO and control mice , but after a 14-hour fast , 3-HB levels were significantly lower in HSLAKO mice than controls ( Fig 6A ) , consistent with lower hepatic FA oxidation and ketogenesis . Also , low expression of genes related to FA oxidation , including Cpt1a , Pparα , Lcad and Vlcad , was observed in HSLAKO liver ( Fig 6B ) . Finally , hepatic FA oxidation was measured directly in liver slices using 1-14C palmitic acid as substrate . HSLAKO livers had a lower rate of FA oxidation than control livers ( Fig 6C ) . Together , these results suggest that HSLAKO mice had lower hepatic FA oxidation than controls . This could contribute to hepatic steatosis . Reduction of hepatic release of TG , measured as VLDL secretion , could also contribute to hepatic steatosis [26] . To test whether adipose HSL deficiency affects hepatic VLDL production , we measured the increase of plasma TG level 4 hours after intraperitoneal injection of poloxamer 407 ( P407 ) , an inhibitor of lipoprotein lipase . Compared to normal controls , HSLAKO mice showed lower post-P407 plasma TG levels at 4h after injection ( Fig 6D ) , indicating lower production of VLDL . These results demonstrate lower production of VLDL in HSLAKO mice than in controls . This could potentially contribute to hepatic steatosis in HSLAKO mice . Defective hepatic lipolysis is another potential contributor to steatosis [13] . We therefore measured the mRNA and protein levels of ATGL , the major hepatic TG hydrolase [13] . In liver , HSLAKO mice had lower levels of ATGL mRNA and protein compared to normal controls ( Fig 7A–7B ) , suggesting a lower lipolytic capacity . When hepatic TG hydrolase activity was measured directly with radiolabeled TG as substrate , the activity of HSLAKO liver was 46% that of control mice ( Fig 7C ) . Together , these results suggest a lower capacity for hepatic TG degradation in HSLAKO than in normal control liver . In summary , lipid metabolism in HSLAKO liver is characterized by low levels of hepatic FA oxidation , of VLDL secretion and of TG hydrolase activity , each of which could contribute to hepatic steatosis . Plasma levels of ALT ( Fig 7D ) and liver expression of mRNAs of the pro-inflammatory cytokines TNFα and IL-6 and of the M1 macrophage markers iNOS , Cxcl9 and Cxcl10 were similar in HSLAKO and normal mice ( Fig 7E ) . The levels of three fibrosis-related transcripts , MMP9 , TGFβ1 and α-SMA , were also similar in HSLAKO and normal control mice ( Fig 7F ) . Together , these results demonstrate that at 8 months of age , HSLAKO mice develop isolated hepatic steatosis .
In this study , we showed that age-dependent hepatic steatosis and insulin resistance develop in HSL-deficient mice and that this occurs by an adipose tissue-dependent mechanism ( Fig 8 ) . Mice with systemic or adipose HSL deficiency show marked macrophage infiltration in adipose tissue and progressive development of lipodystrophy . In striking contrast , mice with hepatic HSL deficiency had normal liver weight and fat content , excluding a cell-autonomous effect of HSL deficiency on hepatocyte TG content . Together , these results show that adipose HSL deficiency has a major effect on liver TG content . Adipose tissue influences systemic energy balance . Conditions like exogenous obesity [27 , 28] and lipoatrophy [23 , 29] reduce the capacity of adipose tissue to take up and to store additional triglycerides , and they are associated with systemic insulin resistance and hepatic steatosis . A small number of mouse models with adipose-specific genetic changes have been reported to develop hepatic steatosis [29–31] . This group includes mice with adipose-specific deficiency of the insulin receptor [29] , or of Raptor/mTORC1 [31] , both of which develop lipodystrophy , and mice with adipose-specific overexpression of RBP-4 which show adipose tissue inflammation with macrophage infiltration [30] . Congenital and acquired lipodystrophies cause insulin resistance and hepatic steatosis [23 , 24] . The insulin receptor , Raptor/mTORC1 and RBP-4 each has direct links to insulin signaling [32–37] . In HSLSKO mice , “crown-like structures” ( dead adipocytes surrounded by macrophages ) occur [38] , as they do in humans with metabolic syndrome [39] . HSLSKO and HSLAKO mice therefore have several features in common with the other models of secondary , adipose-driven hepatic steatosis , including adipose macrophage infiltration and inflammation , age-related progressive lipodystrophy and systemic insulin resistance . Young HSLAKO mice exhibit improved insulin tolerance despite unchanged liver TG content and body weight . Insulin sensitivity is affected by at least four tissues: skeletal muscle , liver , pancreas , and fat tissue [40] . In young HSLAKO mice , fat tissue is relatively normal ( not lipodystrophic yet ) compared to the old HSLAKO mice , their skeletal muscle and liver have similar amounts of fat as controls and we detected no evidence of muscle , liver or pancreatic dysfunction . HSLAKO mice show low plasma FFA , likely due to their primary deficiency of adipocyte lipolysis . When other tissues function normally , low plasma FFA levels contribute to improved insulin sensitivity , as previously shown in the ATGL adipose tissue knockout mice [41] . Therefore , the main contributor to improved insulin tolerance in young HSLAKO mice is lower plasma FFA . Of note , we found that HSLAKO livers showed lower rates of fat disposal by oxidation , VLDL production and lipolysis . In general , there are two potential causes for this . ( 1 ) Less availability of fatty acid substrate for these processes is one potential cause . For example , liver specific deficiency of ATGL , the main lipase responsible for hepatic triglyceride degradation , causes TGs accumulation in liver . In these mice , due to lack of FA availability , FA oxidation and VLDL package were low [13] . ( 2 ) Impaired subcellular organelle function could be another reason . High TG levels in hepatocytes has been associated with ER stress and/or mitochondrial dysfunction [42] , with subsequent lower rates of fat oxidation and VLDL production . In HSLAKO mice , lower plasma FA due to defective lipolysis with reduced availability of fatty acid substrates is one direct cause of lower mitochondrial FA oxidation . Also , when equal among of FAs were given to HSLAKO liver , lower fatty acid oxidation rate were shown in HSLAKO mice than controls ( Fig 6D ) suggesting an impaired mitochondrial function . Therefore , both substrate availability and dysfunctional subcellular organelle contribute to lower hepatic fatty acid disposal in HSLAKO mice . Although in adipose tissue , two major lipases ATGL and HSL catalyze sequential steps in lipolysis , their deficiency causes fatty liver in completely different fashions . The secondary hepatic steatosis that occurs in systemic and adipose HSL deficiency contrasts with primary hepatic steatosis in liver-specific ATGL deficient mice as we showed previously [13] . Defining the mechanisms of hepatic steatosis is necessary to individualize treatment . For the hepatic steatosis of HSL-deficient mice , adipose tissue , not liver , should be considered as a main target for prevention and treatment . The HSL-deficient patients described to date and the 8-month-old HSLAKO mice both develop partial lipodystrophy , with inflammation and low levels of adipogenic and lipogenic markers in adipose tissue . These similarities suggest that the development of hepatic steatosis in HSL-deficient patients may also be mechanistically similar to that of HSL-deficient mice . If so , adipose tissue may be a therapeutic target for preventing and treating hepatic steatosis in patients with HSL deficiency and possibly other forms of secondary hepatic steatosis .
All experiments were approved by Animal Facility Committee of CHU Sainte-Justine Hospital ( protocol 620 ) according to the guidelines of the Canadian Council on Animal Care ( http://www . ccac . ca/en_/ ) . Mice from the previously-described strain of systemic HSL knockout ( HSLSKO ) mice [17] were bred to a C57BL/6J background for at least 8 generations . Liver-specific HSL knockout mice ( HSLLKO ) were created by breeding a gene-targeted HSL allele with Lox sites flanking exon 1 of Lipe as we previously described [22] , with a transgene expressing Cre recombinase from the albumin promoter . Mice with adipose HSL deficiency ( HSLAKO mice ) were created as we previously described [22] . Controls were sibling littermates with wild type HSL alleles that expressed the albumin-Cre transgenic mice in the controls for HSLLKO mice , and the Fabp4-Cre transgene in the controls for HSLAKO mice . After weaning , mice received Global Rodent Diet ( Teklad #2019 ) . All the mice were transferred to a C57BL/6J background for at least eight generations . Male mice were used for all experiments . PCR and Western blotting were performed as described [13] . Primers for PCR are listed in S1 Table . Antibodies for Western blotting were: ATGL ( #2138 , Cell Signaling Technology , Danvers , MA ) ; HSL [43]; and TGH ( a gift from Richard Lehner , University of Alberta , Edmonton ) [44] . Commercial kits were used to assay plasma fatty acids ( FA ) ( Wako HR Series NEFA-HR , Wako Pure Chemical Industries , Chou-ku , Osaka ) , TG ( 12016648 122 , Roche Diagnostics , Indianapolis , IN ) , glucose ( ALL-IN-ONE blood glucose monitoring system , ACCU-CHEK Compact Plus , Roche Diagnostics , Indianapolis , IN ) and 3-hydroxybutyrate ( 3-HB ) ( Precision Xtra blood glucose & ketone monitoring system , Abbott Diabetes Care , Mississauga , Ontario ) . Hormones were measured with commercially-available kits: insulin ( 80-INSMSU-E01 , Alpco Diagnostics , Salem , NH ) , leptin ( MOB00 , R&D Systems , Minneapolis , MN ) and adiponectin ( MRP300 , R&D Systems , Minneapolis , MN ) . Lipids were extracted from liver and skeletal muscle by the Folch method [45] . Lipid classes were resolved by thin-layer chromatography and TG content was measured as described [13] . Tissue fragments were fixed in buffered formalin , then paraffin-embedded for hematoxylin-eosin or Masson trichrome staining . Image J was used for adipocyte diameter measurements . The distribution of cell size was expressed as percentage of total counted adipocytes . A minimum of 6 high power fields ( X200 ) were counted per mouse . Four mice of each genotype were studied . ITT and GTT were performed as described [13] . Hepatic FA oxidation was tested as described , using 1-14C-palmitic acid ( Pekin Elmer ) as substrate [13] . Hepatic VLDL secretion was measured as described [13] . This was assayed in vitro using Triolein [42] ( Perkin Elmer ) as substrate , as described [13] . Values are presented as means ± SEM . Groups were compared using the unpaired two-tailed Student’s t-test .
|
Fatty liver is a major complication of obesity and of type 2 diabetes mellitus . It carries a high risk of cirrhosis and liver cancer . In fatty liver , triglycerides accumulate to high levels in the cytoplasm of hepatocytes . Triglycerides are degraded by lipolysis , which has been most studied in fat cells where its three steps are catalyzed by different enzymes . The second step , hydrolysis of diglyceride to a monoglyceride , can be mediated by hormone-sensitive lipase ( HSL ) . Patients with genetic deficiency of HSL have fatty liver . In this study , we found that systemic HSL deficient mice developed fatty liver with aging . To study the mechanism of steatosis , we made liver-specific HSL-deficient mice . Surprisingly , these mice had normal liver fat content . We then studied mice with HSL deficiency in adipose tissue . Adipose HSL-deficient mice developed hepatic steatosis to a similar extent as mice with systemic HSL deficiency , showing that adipose HSL deficiency is sufficient to cause fatty liver . Furthermore , like reported HSL-deficient humans , mice with adipose HSL deficiency had systemic insulin resistance , reduced fat mass and inflammation in fat tissue . Each of these is known to promote hepatic steatosis . Livers of adipose HSL-deficient mice showed low levels of hepatic fatty acid ( FA ) oxidation , of very low density lipoprotein ( VLDL ) secretion and of triglycerides ( TG ) hydrolase activity , each of which could contribute to fat accumulation in liver . Tissue-selective genetic alterations may help in identifying and understanding the tissues responsible for complex metabolic phenotypes like fatty liver . Our data suggest that at least in mice , strategies for treatment of fatty liver related to HSL deficiency should concentrate on adipose tissue .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"steatosis",
"anatomical",
"pathology",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"cytopathology",
"liver",
"diseases",
"animal",
"models",
"model",
"organisms",
"signs",
"and",
"symptoms",
"gastroenterology",
"and",
"hepatology",
"experimental",
"organism",
"systems",
"research",
"and",
"analysis",
"methods",
"lipids",
"inflammation",
"fats",
"biological",
"tissue",
"mouse",
"models",
"chemistry",
"oxidation",
"immune",
"response",
"biochemistry",
"diagnostic",
"medicine",
"anatomy",
"adipose",
"tissue",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"fatty",
"acids",
"fatty",
"liver"
] |
2017
|
Adipose tissue deficiency of hormone-sensitive lipase causes fatty liver in mice
|
Although much of the information regarding genes' expressions is encoded in the genome , deciphering such information has been very challenging . We reexamined Beer and Tavazoie's ( BT ) approach to predict mRNA expression patterns of 2 , 587 genes in Saccharomyces cerevisiae from the information in their respective promoter sequences . Instead of fitting complex Bayesian network models , we trained naïve Bayes classifiers using only the sequence-motif matching scores provided by BT . Our simple models correctly predict expression patterns for 79% of the genes , based on the same criterion and the same cross-validation ( CV ) procedure as BT , which compares favorably to the 73% accuracy of BT . The fact that our approach did not use position and orientation information of the predicted binding sites but achieved a higher prediction accuracy , motivated us to investigate a few biological predictions made by BT . We found that some of their predictions , especially those related to motif orientations and positions , are at best circumstantial . For example , the combinatorial rules suggested by BT for the PAC and RRPE motifs are not unique to the cluster of genes from which the predictive model was inferred , and there are simpler rules that are statistically more significant than BT's ones . We also show that CV procedure used by BT to estimate their method's prediction accuracy is inappropriate and may have overestimated the prediction accuracy by about 10% .
Developing computational strategies for predicting transcription factor binding sites ( TFBSs ) and transcription regulatory networks has been a central problem in computational biology for more than a decade . Reviews on this problem and various proposed methods can be found in [1–3] . A popular strategy is to search from upstream sequences of a set of co-regulated genes for over-represented ( i . e . , enriched ) sequence features ( motifs ) [4–7] . With the help of gene expression microarray technology , the expression level of thousands of genes can be measured at the same time [8–10] , which makes the discovery of sets of co-regulated genes and their respective regulatory signals at the genome-wide level a reality for many species . Bussemaker et al . [11] pioneered the use of regression models to relate a gene's expression with numbers of occurrences of certain k-mer “words” in the upstream sequence of this gene . Motivated by their work , researchers have developed various methods to extract features that are predictive of gene expression levels . Keles et al . [12 , 13] tackled the problem using logic regression , which treats motif occurrences as binary covariates and selects important predictors adaptively . Conlon et al . [14] proposed a stepwise regression procedure called Motif Regressor , which uses motif matching scores at promoter regions instead of k-mer occurrences as covariates . Zhong et al . [15] extended these methods by introducing a more flexible regression model with an unspecified nonlinear link function . Das et al . [16] implemented a smoothing-spline regression in the place of the linear regression used by Motif Regressor . Further along this general direction , Segal et al . [17] showed that DNA sequence and gene expression information can be combined to construct transcriptional modules . Lee et al . [18] used the ChIP-chip technology and genome-wide location analysis to infer transcriptional regulatory networks in S . cerevisiae . Beer and Tavazoie ( BT ) [19] proposed a novel formulation of the sequence–expression problem . They asked the very intriguing , but seemingly impossible , question: how much can we predict gene expressions from gene upstream sequences ? To address the question , they first clustered a large portion of genes in S . cerevisiae into 49 tight co-expression groups , found enriched sequence patterns ( motifs ) among the promoter sequences of genes in each group using de novo motif prediction tools [6 , 20] , and then trained a set of Bayesian network models to predict the group membership of each gene using the matching scores of its promoter sequence to the set of sequence motifs as well as the orientation and position of the predicted binding sites . They conducted a 5-fold cross-validation ( CV ) procedure to estimate their model's prediction power and found its prediction accuracy to be as high as 73% . A great benefit of the Bayesian network , as shown by BT , is its ability to learn “combinatorial codes” for gene regulation . Hvidsten et al . [21] have applied a similar approach to infer “IF–THEN” rules for transcription regulation . While Bussemaker et al . [11] and Conlon et al . [14] aimed at using gene expression information to help discover transcription factor binding motifs ( TFBMs ) and binding sites , BT focused directly on the prediction problem . However , a few key questions remain . First , BT's assessment of their method's prediction power is over-optimistic , as their CV procedure did not include the motif-finding step ( more details later ) . But , how much can we really predict ? Second , is the Bayesian network an appropriate model for the task or just too complex a black box , prone to overfitting for the stated tasks ? Third , do those inferred combinatorial rules have real predictive power , or are they only observational oddities after the model fitting ? How should we think about and quantify uncertainties inherent in such inferred models ? Given the limited amount of data and the vast number of potential predictors ( e . g . , 666 sequence motifs , orientations , and positions of candidate motif sites , etc . ) , it is not clear if a complex-structured model can be fitted with any confidence . Our plan to address the above concerns is as follows . We first use the same data and the same ( but wrong ) CV procedure as in [19] to develop our predictive models , naïve Bayes classifiers with feature preselections , so as to study the problem of model fitting . Then , we study contributions of various sequence features , such as orientations and positions of the predicted binding sites , to the prediction accuracy . Lastly , we implement a correct CV procedure and show the difference of prediction accuracies resulting from correct versus incorrect CV procedures . Based on the same gene clustering information , putative TF binding motifs , and gene upstream sequences as in [19] , our naïve Bayes classifiers outperformed BT's Bayesian network without using any information regarding the position and orientation of the predicted TFBSs . Our classifiers typically select more motif features , but have far fewer model parameters than the Bayesian network models in [19] . We also found that adding the information regarding TFBS orientation and position cannot further improve the naïve Bayes classifier's predictive power in a global way , which casts doubts on several biological predictions made in [19] regarding combinatorial rules of gene regulation . We further studied a few cases in detail and found that the supports for the inferred combinatorial rules are at best circumstantial . Finally , we speculate that the incorrect CV procedure used in [19] has likely overestimated the accuracy rate of their method by 10% .
The data used in this study were obtained from the supplemental Web site of [19] , which contains matching scores ( i . e . , the likelihood of a promoter sequence to contain good sequence matches to a candidate TFBM ) , and orientations and positions of the predicted matches of 666 putative TFBMs for 2 , 587 genes in S . cerevisiae . In [19] , these 2 , 587 genes were clustered into 49 different co-expression groups according to their expression profiles in 255 conditions , such as environmental stress [22] and cell cycle [8] . We trained a set of naïve Bayes classifiers to predict the cluster label ( membership ) for each gene using only its motif matching scores . Since genes in the same cluster have very similar expression profiles , a gene's cluster membership can serve as a surrogate of its expression behavior under different conditions . We built one naïve Bayes model for each cluster , resulting in a total of 49 classifiers . For each cluster , we first ranked all the 666 sequence motifs according to a Chi-square test procedure , which reflects these motifs' capability of differentiating genes in this cluster from all other genes . Then , we selected the top m most significant motifs as explanatory variables to train a naïve Bayes classifier ( for this cluster ) , where m can range from 1 to 666 . We used the same 5-fold CV procedure as that in BT to test the predictive power of our models . As shown in Figure 1 , using the same criteria for classification accuracy as in [19] ( i . e . , for any pair of clusters , if the correlation between their mean expression is greater than 0 . 65 , then misclassifying genes in one cluster into the other is not counted as errors ) , naïve Bayes classifiers correctly predicted expression patterns for 75% of the genes when the number of preselected motifs m is 5 . When m is increased to 20 , naïve Bayes classifiers achieved a 79% prediction accuracy ( see Table S1 ) . In addition , the naïve Bayes models contain almost all the motif features selected by BT in [19] and include many more ( see Figures S1 and S2 ) . It can also be seen that , although the training accuracy always increases as m increases , the prediction accuracy starts to plateau and then decrease as m exceeds 20 , which is indicative of overfitting as more variables are included . Following BT , we also calculated the mean correlation of each gene to its predicted expression pattern . For a gene , its predicted expression pattern is the mean expression pattern of the cluster that it is predicted to belong to . With our 20-motif naïve Bayes model , we obtained a mean correlation of 0 . 56 without using any position and orientation information , which is also higher than BT's result of 0 . 51 . Having fitted the classification models , we now study how the 666 motifs are present in the model of each cluster . Our first observation is that most clusters have their distinct sets of motif features . But a few motifs are selected by multiple clusters , which may indicate that either the transcription factors corresponding to these motifs are somewhat multi-taskers , or the clusters that share these common motifs are closely related . For example , Motifs PAC and RRPE are selected in the models for clusters 4 , 10 , 17 , 26 , and 29 . This suggests that many genes in these five clusters may be targeted by the TFs that bind to PAC and RRPE . Clusters 47 and 48 share 17 out of 20 motifs in their models ( p < 1 × 10−21 ) . Coupled with the fact that the correlation of the mean expression patterns of these two clusters is more than 0 . 8 , it strongly suggests that genes in these two clusters are co-regulated . Motif PAC is associated with polymerase A and C subunits [20 , 23] . Motif RRPE specifically exists in genes involved in rRNA processing [20] . BT extracted from their model a combinatorial prediction rule for cluster 4 [19]: PAC should have a score higher than 0 . 6 and be within 140 bp of ATG; RRPE should have a score higher than 0 . 65 and be within 240 bp of ATG . Table 1 shows numbers of genes in a few different clusters that satisfy these constraints . The statistics suggest that PAC and RRPE are both significantly enriched in cluster 4 , but not uniquely . Clusters 10 , 17 , 26 , and 29 also have significant portions of genes that satisfy the constraints of both motifs . Our naïve Bayes method successfully picked PAC and RRPE for all these five clusters , whereas BT did not select RRPE for cluster 10 , or PAC for cluster 29 . It suggests that , due to its complex nature , the Bayesian network model in [19] can easily miss important features . Furthermore , our method using no information about TFBS orientation and position correctly predicted 94% of the genes in cluster 4 and 87% of the genes in clusters 10 , 17 , 26 , and 29 , which is comparable to the 92% and 87% accuracy of [19] for the same clusters . RAP1 is a main regulator of ribosomal proteins in S . cerevisiae , and many ribosomal protein coding genes are reported to have RAP1 binding site ( s ) in their upstream sequences [24] . BT [19] found that cluster 1 is enriched with RAP1 binding sites , and their Bayesian network inferred a rule for genes in this cluster: their RAP1 score on upstream sequences has to be greater than 0 . 6 , and their RAP1 sites have to be oriented toward a certain direction . We examined this rule carefully and observed the following . First , we found that 82 genes in cluster 1 ( a total of 124 genes ) and 165 genes in other clusters ( a total of 2 , 463 genes ) have putative RAP1 binding sites ( i . e . , with RAP1 matching score >0 . 6 ) , which gives rise to a p-value of 1 × 10−59 ( based on Fisher's exact test ) for the enrichment of RAP1 sites in cluster 1 . Seventy-three genes in cluster 1 and only 85 genes in other clusters satisfy both the orientation and the site score requirements , which yields an even more significant contrast p-value , 1 × 10−64 . It seems that the RAP1 orientation can indeed help enhance the prediction specificity , although only slightly . However , our naïve Bayes model selected motif M198 as its main predictor for genes in cluster 1 . This motif has a very similar weight matrix to that of RAP1 but includes an extra position ( Figure 2 ) . By setting 0 . 6 as the score threshold of M198 , we found that 100 genes in cluster 1 and 126 genes in other clusters contain the M198 site , which gives us a p-value of 4 × 10−94 for the M198 enrichment in cluster 1 . Thus , if judged by statistical significance of the prediction specificity , the naïve Bayes model with one simple predictor easily outperformed the more complex combinatorial rule inferred by BT's Bayesian network . In order to evaluate the effectiveness of RAP1 ( with orientation constraint , denoted as RAP1d for short ) and M198 as covariates in our classifier , we compared two procedures . In both procedures , one single best motif was selected for each cluster . The only difference was that , for cluster 1 , M198 was used in Procedure One and RAP1d was used in Procedure Two . As a result , Procedure One predicted 20 more genes correctly than Procedure Two , and the improvement is mainly in cluster 1 . For cluster 1 alone , Procedure One has a 30% false positive and 18% false negative rates , while Procedure Two has a 38% false positive and a 34% false negative rate . These results further suggest that M198 is a better motif for cluster 1 than the oriented RAP1 . In the next subsection , we provide a more thorough investigation on the biological relevancy of motif site orientation and its effect on the classification accuracy . The result in the previous subsection does not mean that the motif site orientation is not biologically important . In fact , we found that 91 of the 100 predicted M198 sites for genes in cluster 1 are oriented toward one direction . In comparison , only 56 of the 126 predicted M198 sites for genes in other clusters are oriented the same way . Clearly , including both the M198-score and its site orientation constraints can improve the prediction specificity for cluster 1 , as observed by BT for RAP1 . However , in a similar procedure comparison as in the previous subsection , adding the orientation constraint of M198 does not improve the global prediction . This orientation constraint may help reduce the false positive rate for cluster 1 , but it at the same time increases false positive rates in other clusters . Thus , a fundamental question is: is it appropriate to justify the “authenticity” of a prediction model based on its prediction performance ? Our analysis suggests that a combinatorial regulation rule , and perhaps many other causal relationships , may not be reliably inferred using an automatic “learning machine” under a global classification accuracy criterion . To assess globally whether the TFBS orientation and position information can further help predict gene expression , we added the covariates representing TFBS orientations and positions to the feature list of our model . We performed the same feature preselection and naïve Bayes procedures as described above on the augmented dataset . The classification accuracies for the training sets were very close to the result from using motif score alone . However , the classification accuracies for the test sets were slightly worse than before . This result implies that , although it may be biologically true that orientations and positions of authentic TFBSs have an effect on the binding of the corresponding TFs in some cases , such information for predicted TFBSs do not help in predicting co-expression of genes globally when motif matching scores are given . Even in BT's Bayesian network models , position and orientation constraints were selected only 5 . 1% and 0 . 6% of the time , respectively . In both of the strong cases detailed in [19] , we were able to find a simpler rule ( matching scores only ) that is as sensitive and specific as or better than the combinatorial rules reported by BT . We would like to caution the reader again , however , that our results cast doubts on some of these delicate model interpretations of BT but do not imply that the position and orientation of TFBSs are biologically unimportant . So far we have followed BT's approach as closely as possible: using the same set of motif features generated by [19] and employing exactly the same CV procedure as theirs . The only difference between our and their approach is that we used the naïve Bayes model , whereas they used the more complex Bayesian network . However , we cannot help notice that the 615 de novo motifs ( excluding the 51 known motifs ) generated by [19] were found by using the Gibbs motif sampler AlignACE [20] to search the upstream sequences of all genes in both the training and the test datasets for each cluster . These motifs were further optimized so as to be more specific to the respective clusters they were discovered from by a simulated annealing procedure [19] , still using all genes in both the training and test datasets . These steps inevitably generate motifs ( features ) that are already biased in favor of the existing clustering in the test set . In a valid CV procedure , only the information for the training set genes , including both their upstream sequences and their cluster labels , are allowed to be used in both feature extraction and model training . To correctly measure how much of gene expression information can be predicted by DNA sequence features , we implemented a valid 5-fold CV procedure , still using the gene clustering result of BT . First , genes in each cluster were divided into five sets of approximately equal sizes at random . Each time , we left out 20% of genes ( one subset of genes for each cluster ) , and used the remaining 80% of genes ( i . e . , the training set ) and their upstream sequences for de novo motif finding via AlignACE [20] . These motifs were then optimized by a simulated annealing algorithm . The total number of motifs we found ranged from 600 to 700 for each training set , which is consistent with the number of 666 motifs in [19] . We then preselected the top 20 motifs ( see Figure S3 ) for each cluster and trained naïve Bayes classifiers based on the training set and the preselected motifs . Finally , the classifiers so trained were used to predict the cluster memberships of the left-out 20% genes . The classification accuracy of this correct CV procedure is 61% according to the criterion in [19] , which is still significantly higher than random guessing . When we further added the 51 known motifs to the motif sets , the classification accuracy increased to 64% . Note that we cannot directly use the motif finding and model-fitting procedure of [19] because their complete algorithm is not publicly available . Furthermore , their-model fitting procedure needs bootstrapping replications and can be overly time consuming , unstable , and nonreproducible . Thus , there is a possibility that the low accuracy of our correct CV procedure is caused by the lower capability of our motif finding strategy compared to that of [19] . To calibrate with BT's approach , we also applied the exact same incorrect CV procedure as in [19] using our own motif finding , optimization , and model-fitting strategies described above . When using all the genes in all clusters , our de novo motif discovery strategy found altogether 650 motifs , and the whole procedure yielded a classification accuracy of 75% , which is slightly higher than the result of [19] ( 73% ) . Based on these results , we conclude that the incorrect CV procedure of [19] has likely overestimated the true prediction accuracy of their expression prediction method by 10%–15% .
The naïve Bayes model we adopted is essentially the simplest version of the Bayesian network . The assumption of conditional independence of the covariates is far from realistic in most applications , as well as in this study . However , it outperformed the more complicated Bayesian network , as well as SVM , CART , logistic regression , and Bayesian logistic regression [25] ( unpublished data ) for this study . As described by Domingos and Pazzani [26] , optimality in terms of zero-one loss ( classification error ) is not necessarily directly connected to the quality of the fit of a probability distribution . Rather , as long as both actual and estimated distributions agree on a most-probable class , the classifier will have a reasonable performance . Although it is not rare to see successful examples of the naïve Bayes method , the feature selection step is always challenging . In our method , features are considered independently . Each feature is dichotomized to 0 or 1 according to a threshold that maximizes a Chi-square test statistic . In this way , features that are highly associated with a target cluster will be selected as covariates in the naïve Bayes model of this cluster . Our method selects not only the features that are enriched in the target cluster , but also those that are “depleted” in the target cluster but enriched in other clusters . The latter type of features can be explained as a logic operator “NOT” . Dichotomization of motif scores in our procedure is a gross simplification . Although the binding of a TF to DNA may not be a simple 0–1 trigger , it is easier to model it in this way , and it is also interesting to see whether this simple model can help predict gene expression . We expect to lose some information through discretization , but it is not clear how much the lost information can help the classification problem . It is a worthwhile future project to explore possibilities of using the continuous data , both motif scores , and gene expression values , directly and more efficiently . Our study has shown that it is perhaps not very sensible to justify a model's “authenticity” by its global prediction performance , and one may easily inject subjective interpretations into the inference results , especially when the prediction uncertainty is not explicitly quantified . This in fact is a challenge for many machine learning approaches , and researchers have begun to pay attention to the problem of estimating prediction uncertainties . In this regard , it is perhaps beneficial to act more like a real Bayesian when using Bayesian tools . That is , these tools not only provide point estimates , but also posterior distributions , which summarize all the information in the data and quantify uncertainties of the estimates . The keen difference between the correct and incorrect CV procedures reminds us how easy it is to be overconfident . Similar mistakes have also been uncovered in some computational biology studies in which knowledge from literature is used to help construct gene clusters or biological networks and these results are then evaluated and validated by GO analysis , which is by itself a product partially based on the literature . Although it has been accepted as common knowledge in biology that TFBSs' orientation and position have a functional role in affecting gene regulation activities , and anecdotal examples abound [27 , 28] , it is still nonconclusive how the orientation and position information of putative TFBSs can help one discern true TFBSs from sporadic sequence matches that exert no regulatory functions . In particular , the TFBS orientation and position information did not help us improve the classification accuracy globally , and was not even obviously useful in the two strongest cases detailed in [19] . Since the Bayesian network in [19] is more prone to overfitting , the danger of overinterpreting the fitted models can be a serious threat . In a recent study of nucleosome positioning in yeast , Yuan et al . [29] observed that true regulatory elements are highly enriched in nucleosome depleted regions . Thus , certain sequence information at a scale of nucleosome binding regions ( larger than TF binding sites ) may be more useful than orientation and position information in differentiating true TFBSs from false ones .
For motif j , its score for gene i is denoted as sij , which is computed in [19] as either zero , when motif j has no predicted occurrence in the promoter of gene i , or the highest matching score among all predicted occurrences of the motif in the promoter of gene i . In this way , a score matrix S = ( sij ) 2587×666 can be built directly from the supplement data of [19] . The continuous scores sij are discretized into 0 or 1 by a thresholding procedure described below . In a word , a threshold for the scores corresponding to a motif is chosen so as to maximize the specificity of TFBSs for the cluster of interest . Let N be the number of all the genes in consideration ( i . e . , 2 , 587 ) and let yi be the class label of gene i ( i ∈ {1 , ··· , }N ) . Among these N genes , Nk , 1 of them are in class k ( defined as positive set ) and Nk , 0 are not in class k ( defined as negative set ) . Thus Nk , 1 = #{i:yi = k} , Nk , 0 = #{i:yi ≠ k} and Nk , 1 + Nk , 0 = N . For motif j ( j ∈ {1 , ··· , 666} ) and a threshold c , define The best threshold for motif j in model k is defined as: where More intuitively , the above procedure finds the most significant Chi-square test result for the 2 × 2 contingency table of the N's . This procedure makes the distribution of TFBSs in positive set and negative set most different . The thresholds calculated above discretize the score matrix S into a 0–1 matrix and it is denoted as X . Note that the discretized covariate matrix X will be different for fitting models in different classes . The feature preselection step is simply an extension of the threshold finding procedure . For model k , the best threshold is calculated for motif j along with its highest χ2 statistic . Features ( motifs ) are sorted by their χ2 statistics , and the top m ones are included the models . This selection is done for each model separately . The naïve Bayes method has been widely used in statistical learning . It is based on the very simple assumption that all feature variables ( covariates ) are independent given the class label of the sample . We use cluster 1 and its preselected m motifs as an example to describe our naïve Bayes model fitting procedure . Denote the class label variable as Y and the preselected top m covariates as X1 , ··· , X m . Using the Bayes theorem , we have Thus , the odds ratio can be computed as We further assume Bernoulli models for each Xj given Y and class label variable Y itself , i . e . , The prior distributions for py , p0j , and p1j are set to be uniform . The training set consists of a class label vector y = ( y1 , ··· , yN ) and the discretized TFBS score matrix X = ( xij ) , i = 1 , ··· , N; j = 1 , ··· , m . Given the training set , the posterior distribution of py , p0j , and p1j can be easily calculated as For a new observation with the covariates vector Xnew = ( X1 , new , . . . , Xm , new ) , we have Thus , we have the predictive odds ratio for this new observation as For the 49 classes , 49 models are fitted and the genes in the test set are assigned to the class with the respective model that fits the data best . Specifically , for k = 1 , ··· , 49 , the odds ratio can be calculated and a gene will be assigned to a class k* with the highest odds ratio . To reduce the complexity , for each motif on each gene we only consider the orientation and position of the site with the highest matching score . The site orientation is coded into two separate binary variables , xl and xr , where xl = 1 indicates that the predicted site is left-oriented ( away from ATG ) , xr = 1 for right-oriented , and xl = 0 or xr = 0 otherwise . Note that when a gene does not contain TFBS for a specific motif , the corresponding xl and xr are both 0 . The TFBS position in [19] is a continuous variable representing the distance of the TFBS to ATG . We set it to a very large number if a motif has no occurrence in the promoter region of a gene . In our naïve Bayes procedure , the new variable d is a dichotomized version of the original position variable based on an optimized distance threshold , so that d = 1 means that the distance from the predicted site to ATG is smaller than the chosen threshold .
|
Through binding to certain sequence-specific sites upstream of the target genes , a special class of proteins called transcription factors ( TFs ) control transcription activities , i . e . , expression amounts , of the downstream genes . The DNA sequence patterns bound by TFs are called motifs . It has been shown in an article by Beer and Tavazoie ( BT ) published in Cell in 2004 that a gene's expression pattern can be well-predicted based only on its upstream sequence information in the form of matching scores of a set of sequence motifs and the location and orientation of corresponding predicted binding sites . Here we report a new naïve Bayes method for such a prediction task . Compared to BT's work , our model is simpler , more robust , and achieves a higher prediction accuracy using only the motif matching score . In our method , the location and orientation information do not further help the prediction in a global way . Our result also casts doubt on several biological hypotheses generated by BT based on their model . Finally , we show that the cross-validation procedure used by BT to estimate their method's prediction accuracy is inappropriate and may have overestimated the accuracy by about 10% .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"and",
"genomics",
"saccharomyces",
"cerevisiae",
"computational",
"biology"
] |
2007
|
Predicting Gene Expression from Sequence: A Reexamination
|
Transportin-SR ( TRN-SR ) is a member of the importin-β super-family that functions as the nuclear import receptor for serine-arginine rich ( SR ) proteins , which play diverse roles in RNA metabolism . Here we report the identification and cloning of mos14 ( modifier of snc1-1 , 14 ) , a mutation that suppresses the immune responses conditioned by the auto-activated Resistance ( R ) protein snc1 ( suppressor of npr1-1 , constitutive 1 ) . MOS14 encodes a nuclear protein with high similarity to previously characterized TRN-SR proteins in animals . Yeast two-hybrid assays showed that MOS14 interacts with AtRAN1 via its N-terminus and SR proteins via its C-terminus . In mos14-1 , localization of several SR proteins to the nucleus was impaired , confirming that MOS14 functions as a TRN-SR . The mos14-1 mutation results in altered splicing patterns of SNC1 and another R gene RPS4 and compromised resistance mediated by snc1 and RPS4 , suggesting that nuclear import of SR proteins by MOS14 is required for proper splicing of these two R genes and is important for their functions in plant immunity .
In eukaryotes , the nuclear envelope forms a barrier between the cytoplasm and the nucleus . Trafficking of macromolecules across the nuclear envelope occurs through the nuclear pore complex ( NPC ) [1] . Previous studies on MOS3 [2] , MOS6 [3] , MOS7 [4] and MOS11 [5] have revealed the importance of nucleocytoplasmic trafficking in plant immunity . Mutations in MOS3 , MOS6 , MOS7 and MOS11 suppress the constitutive defense responses of snc1 ( suppressor of npr1-1 , constitutive 1 ) , a gain-of-function mutant carrying a mutation in a Toll/interleukin-1 receptor-Nucleotide Binding-Leucine Rich Repeat ( TIR-NB-LRR ) R protein [6] . MOS3 encodes the nucleoporin Nup96 [2] , whereas MOS11 encodes a putative RNA binding protein [5] . Both MOS3 and MOS11 are required for mRNA export . MOS6 encodes a putative importin-α [3] , whereas MOS7 encodes another nucleoporin , Nup88 , which is required for nuclear accumulation of snc1 and two general defense regulators , Enhanced Disease Susceptibility 1 ( EDS1 ) and Nonexpresser of PR genes 1 ( NPR1 ) [4] . Nuclear import receptors play essential roles in transferring proteins from the cytoplasm to the nucleus . The largest group of nuclear import receptors belong to the importin-β super-family . Members of the importin-β super-family have rather low overall sequence similarity but they all have a conserved N-terminal RAN-binding domain [7] , [8] . The import receptors recognize the nuclear localization sequence ( NLS ) of target proteins to facilitate their transport through the NPC . Upon RAN-GTP binding to importin-β , the importin-β complex is dissociated and the cargo is released into the nucleus . The importin-β super-family can be divided into several sub-families according to the direction and the cargo type they transport [9] . Among them , the transportin-SR ( TRN-SR ) subfamily functions as nuclear import receptors for serine-arginine rich ( SR ) proteins . TRN-SR was originally identified as an interactor of SR domains of ASF/SF2 [10] and papillomavirus E2 [11] . In humans , the C-terminus of TRN-SR interacts with SR proteins and the interaction can be disrupted upon RAN-binding to its N-terminus [10] . SR proteins are a highly conserved family of nuclear proteins that play important roles in splicing [12]–[14] . They contain RNA recognition motifs ( RRM ) at the N-terminus and an arginine-serine rich ( RS ) domain at the C-terminus . The NLS is located in the RS domain . SR proteins not only function as splicing factors for constitutive splicing [15] , [16] , they also regulate alternative splicing through splice site selection in a concentration-dependent manner [17] , [18] . Several plant R genes including the tobacco N gene [19] , the barley Mla6 [20] , Arabidopsis SNC1 [21] and RPS4 [22]–[24] are alternatively spliced . For example , six transcript variants ( TV ) have been identified for RPS4 [23] , [24] . Compromised RPS4-mediated resistance resulting from a lack of the TVs suggests that alternative splicing of RPS4 is required for its function [23] . However , it is unclear how alternative splicing of these R genes is controlled and why it is necessary for immunity . In this study , we report that Arabidopsis MOS14 encodes a TRN-SR that is required for proper splicing of SNC1 and RPS4 , suggesting that SR proteins may play important roles in the control of the splicing of these two R genes .
Arabidopsis snc1 constitutively activates defense responses and displays enhanced resistance to pathogens . snc1 mutant plants exhibit dwarf morphology with curly leaves . Suppressor screens of snc1 have previously been carried out using fast neutron and T-DNA insertional mutagenesis [2] , [25] . To identify additional suppressor mutants of snc1 , we treated snc1 npr1 seeds with ethane methyl sulfonate ( EMS ) and screened the M2 plants for mutants that suppressed snc1 dwarfism . From this population , we identified mos14-1 snc1 npr1 ( Figure 1A ) . In snc1 npr1 , defense marker gene PR1 and PR2 are constitutively expressed . As shown in Figure 1B and 1C , constitutive activation of PR1 and PR2 is suppressed in mos14-1 snc1 npr1 . Analysis of SA levels also showed that the elevated SA levels in snc1 npr1 are suppressed by mos14-1 ( Figure 1D and 1E ) . To test whether enhanced pathogen resistance in snc1 npr1 is affected by mos14-1 , mos14-1 snc1 npr1 seedlings were challenged with the virulent oomycete pathogen Hyaloperonospora arabidopsidis ( H . a . ) Noco2 . As shown in Figure 1F , resistance to H . a . Noco2 is lost in mos14-1 snc1 npr1 . To map the mos14-1 mutation , we crossed mos14-1 snc1 npr1 ( in the Columbiaecotype background ) with Landsberg erecta ( Ler ) -snc1 [2] . In the F2 mapping population , about a quarter of the progeny showed morphology similar to the triple mutant . Crude mapping using 24 F2 plants revealed that mos14-1 is linked to the lower arm of chromosome 5 ( Figure 2A ) . Further analysis indicated that mos14-1 is flanked by marker MMN10 and MUB3 . Fine mapping using about 1200 F2 plants narrowed mos14-1 to a 60 kb region between marker K19B1 and MRG21 . To identify the mos14-1 mutation , PCR fragments covering this 60 kb region was amplified directly from mos14-1 snc1 npr1 and sequenced . A single G to A mutation was found in At5g62600 ( Figure 2B ) , which is located at the junction of the 13th intron and 13th exon of the gene . RT-PCR analysis using primers flanking the mutation showed that splicing of At5g62600 was affected by the mutation ( Figure 2C ) . The RT-PCR fragments were cloned into the pGEM-T vector . Subsequent sequence analysis of cDNA clones from mos14-1 revealed that they fell into six different classes . All of them represent transcript variants that were incorrectly spliced . An alignment of wild type cDNA and the cDNA variants from mos14-1 are shown in Figure S1 . To confirm that the mutation in At5g62600 is responsible for the suppression of snc1 npr1 mutant phenotypes , a genomic clone containing At5g62600 was constructed and transformed into mos14-1 snc1 npr1 . Transgenic plants from five independent lines carrying the wild type At5g62600 displayed snc1-like morphology ( Figure 2D ) . Further analysis of a representative transgenic line showed that the expression of PR1 and PR2 was similar to snc1 npr1 ( Figure 2E and 2F ) . In addition , resistance to H . a . Noco2 was also restored in the transgenic line ( Figure 2G ) , confirming that At5g62600 complemented mos14-1 and MOS14 is At5g62600 . To obtain the mos14-1 single mutant , we backcrossed mos14-1 snc1 npr1 with wild type plants . The mos14-1 single mutant was obtained by genotyping the F2 plants . The mos14-1 single mutant flowers late and has reduced fertility . Besides , it exhibits small stature ( Figure S2 ) . When the genomic clone of At5g62600 was introduced into the mos14-1 single mutants , it reverted the size and fertility of the mutant to wild type-like and also suppressed the late flowering phenotype , showing that the developmental phenotypes observed in mos14-1 are caused by the mos14-1 mutation . MOS14 is a single copy gene in Arabidopsis . It encodes a protein with 25% identity and 45% similarity to the TRN-SR in Drosophila , suggesting that MOS14 may be a transporter for SR proteins . MOS14 and its animal homologs are highly conserved at their N-terminus ( Figure S3 ) , which contain the importin-β N-terminal domains . To determine the subcellular localization of MOS14 , transgenic plants expressing MOS14 under its native promoter with a C-terminal GFP tag were generated in both wild type and mos14-1 backgrounds . Expression of MOS14-GFP in mos14-1 suppresses the developmental phenotypes of mos14-1 ( Figure S4 ) , suggesting that the fusion protein is functional . Confocal fluorescence microscopy analysis of transgenic plants expressing the MOS14-GFP fusion protein showed that the GFP signal is found exclusively in the nucleus ( Figure 3 ) , indicating that MOS14 is a nuclear protein . In the nuclei of root cells , GFP fluorescence was excluded from a large part of the nucleus , probably the nucleolus . We did not observe similar exclusion of MOS14-GFP from parts of the nuclei in epidermal cells , probably because these nuclei are much smaller than those in root cells . In animals , TRN-SR binds SR proteins via its C-terminus and transport SR proteins through the nuclear envelope . Binding of RAN-GTP to the N-terminus of TRN-SR in nucleus results in the release of SR proteins . To test whether MOS14 is able to interact with SR proteins , the N-terminus ( 1–281 ) and C-terminus ( 282–958 ) of MOS14 were expressed in the bait vector and four selected Arabidopsis SR proteins ( AtRS2Z33 , AtRSZ21 , AtRS31 and AtSR34 ) were expressed in the prey vector for yeast two-hybrid assays . As shown in Figure 4A , the C-terminus , but not the N-terminus of MOS14 interacts with the SR proteins . We also tested the interactions between MOS14 and AtRAN1 . As shown in Figure 4B , the N-terminus , but not the C-terminus of MOS14 interacts with AtRAN1 . To test whether the mos14-1 mutation affects the nuclear import of Arabidopsis SR proteins , we made constructs expressing four SR genes AtRS2Z33 , AtRSZ21 , AtRS31 and AtSR34 with a C-terminal GFP tag . These constructs were transformed into protoplasts of wild type and mos14-1 plants to check for the localization of the SR-GFP proteins . A construct expressing the SARD1-GFP fusion protein was included as the control [26] . As shown in Figure 5A , in both wild type and mos14-1 protoplasts , SARD1 was localized in the nucleus . Consistent with previous studies [27] , the SR-GFP proteins were clearly localized in the nucleus of wild type protoplasts . However , in mos14-1 protoplasts , the SR-GFP proteins were mainly localized in the cytoplasm ( Figure 5A and Table 1 ) , suggesting that MOS14 is required for the nuclear localization of SR proteins . Unlike GFP expressed under 35S promoter which is distributed throughout the whole cell , the SR-GFP proteins were localized to discrete foci in the cytoplasm of mos14-1 protoplasts . The pattern of these foci resembles that of P-bodies , which are distinct foci in the cytoplasm of eukaryotic cells containing many enzymes involved in mRNA turnover . Because of the diverse roles of SR proteins in RNA metabolism , it would not be surprising if they also function in P-bodies . The effect of mos14-1 on the localization of AtRSZ21 and AtSR34 was further confirmed in transgenic plants expressing the AtRSZ21-GFP and AtSR34-GFP fusion proteins . As shown in Figure 5B and 5C , AtRSZ21-GFP and AtSR34-GFP were localized in discrete foci in the cytoplasm of guard cells in mos14-1 background . The GFP fusion proteins were also observed in the cytoplasm of leaf pavement cells in mos14-1 . Taken together , these experiments indicate that MOS14 is a transporter for SR proteins . Multiple SNC1 transcripts with intron 2 and intron 3 removed or retained have previously been detected [21] . Because none of the transgenic plants expressing the snc1 cDNA exhibit dwarf morphology like snc1 mutant plants ( Figure S5 ) , alternative splicing is probably required for the function of SNC1 . Since mos14-1 affects the nuclear localization of SR proteins and SR proteins participate in pre-mRNA splice site recognition and spliceosome assembly , we tested whether splicing of SNC1 was affected in mos14-1 . Primers flanking the introns of SNC1 were designed to evaluate its splicing pattern of SNC1 ( Figure 6A ) . Consistent with the previous report , we detected transcripts with either intron 2 or 3 retained ( Figure S6 ) . As shown in Figure 6B , we detected another transcript that contains both intron 2 and 3 ( TV1 ) in addition to the regular transcripts with both intron 2 and 3 removed ( TV4 ) in mos14-1 snc1 npr1 . In wild type plants , the amount of TV2 and TV3 is small compared to that of TV4 . Both TV2 and TV3 increased dramatically in the mos14-1 snc1 npr1 mutant plants ( Figure 6B ) . Similar alteration of SNC1 transcript patterns was also observed in the mos14-1 single mutant ( Figure S8B ) . Since PCR reaction using the RNA samples showed no amplification , the DNA fragments from RT-PCR represent SNC1 transcripts rather than genomic DNA contamination . Further analysis of SNC1 transcript variants in mos14-1 and mos14-1 snc1 npr1 lines carrying the wild type MOS14 transgene showed that the splicing patterns of SNC1 in the transgenic lines are similar to those in the wild type plants ( Figure S8A and S8B ) . These data indicate that mos14-1 affects the splicing of the SNC1 transcript . The R gene RPS4 was also reported to be alternatively spliced [23] . We designed primers to detect the transcript variants for RPS4 by RT-PCR . As shown in Figure 6C , the levels of TV1 are similar in wild type and mos14-1 . However , TV2+TV3 increased considerably and TV4 was significantly reduced in mos14-1 , indicating that mos14-1 also affects the splicing pattern of RPS4 transcripts . The altered RPS4 transcript patterns in mos14-1 snc1 npr1 and mos14-1 can be complemented by the MOS14 transgene ( Figure S8C and S8D ) . To determine whether MOS14 has a general role in RNA splicing , we analyzed splicing of two housekeeping genes Actin1 and β-tubulin4 by RT-PCR using primers that flank introns . ROC1 was used as the control because it contains no intron . We found that splicing of Actin1 and β-tubulin4 was not affected in mos14-1 ( Figure S7 ) . We also analyzed the splicing patterns of U1-70K , AtSR30 and AtSR34 , three genes reported to be alternatively spliced [28] , [29] . As shown in Figure S7 , the splicing of AtSR30 and AtSR34 , but not U1-70K was clearly affected by mos14-1 . Alteration of the transcription patterns of AtSR30 and AtSR34 in mos14-1 further supports the role of MOS14 in alternative splicing . Since the splicing of Actin1 , β-tubulin4 and U1-70K is not affected by mos14-1 , there may be a certain level of specificity in MOS14-mediated pre-mRNA processing . To test whether the splicing defect in mos14-1 leads to a decrease in snc1 and RPS4 transcripts , real-time RT-PCR was carried out using primers to amplify an unspliced region at the 3′ end of the two genes . As shown in Figure 6D and 6E , expression levels of both snc1 and RPS4 decreased significantly in the presence of the mos14-1 mutation . Since mos14-1 altered the splicing pattern of RPS4 and reduced its expression , we tested whether RPS4-mediated immunity is affected by mos14-1 . As shown in Figure 7A , growth of Pseudomonas syringae pv . tomato ( P . s . t . ) DC3000 avrRps4 in mos14-1 is about ten-fold higher than that in wild type , suggesting RPS4-mediated immunity is compromised in mos14-1 . We also tested whether MOS14 is required for basal resistance by challenging the mos14-1 plants with the virulent pathogen P . s . t . DC3000 . As shown in Figure 7B , bacterial growth is about ten-fold higher in mos14-1 compared to wild type , indicating that MOS14 is also required for basal resistance .
Previous studies on snc1 suppressor mutants revealed that multiple components are involved in the regulation of plant immunity . In particular , pathways involved in mRNA export , protein import and protein export were found to contribute to immune responses . Here we report the identification of MOS14 as a novel component of nucleocytoplasmic trafficking required for plant immunity . Loss of MOS14 function suppresses the constitutive defense responses of snc1 , compromises resistance mediated by RPS4 and impairs basal resistance against P . s . t . DC3000 . These findings show that MOS14 plays a critical role in plant immunity . MOS14 encodes a nuclear protein with high sequence similarity to TRN-SR proteins in animals . TRN-SR proteins have been shown to function as nuclear import receptors for both phosphorylated SR proteins as well as the splicing repressor protein RSF1 which antagonizes SR proteins in the nucleus [11] , [30] . Since their discovery , TRN-SR proteins have not been extensively studied [10] . MOS14 is a single-copy gene , while the Arabidopsis genome has 18 genes belonging to six subfamilies of SR proteins , of which three are plant-specific [31] . There is no close homolog of RSF1 in Arabidopsis . Like the TRN-SR proteins in animals , the N-terminus of MOS14 interacts with AtRAN1 and the C-terminus interacts with SR proteins . In addition , localization of several SR proteins to the nucleus was impaired by mos14-1 . These data indicate that the mechanism of nuclear import of SR proteins is conserved between plants and animals . Very limited studies have been performed on the genetic characterization of TRN-SR proteins . In C . elegans , RNAi of the MOS14 homolog Transporter of SR-1 ( TSR-1 ) leads to embryonic lethality , suggesting TRN-SR proteins can be essential for viability [32] . Intriguingly , the mos14-1 mutation is not lethal , although it does cause multiple development phenotypes such as reduced stature and fertility . In addition to its functions in development , our genetic analysis of MOS14 revealed that it plays important roles in both R gene-mediated resistance as well as basal defense , suggesting that nuclear import of SR proteins is important for plant immunity . The reasons why mos14-1 leads to these pleiotropic defects and not lethality awaits further investigation . SR proteins play important roles in general RNA splicing , alternative splicing , as well as other processes of RNA metabolism . Consistent with the function of MOS14 in the nuclear import of SR proteins , the mos14-1 mutation affects the splicing of SNC1 and RPS4 . Several R genes including SNC1 , RPS4 and tobacco N gene are alternatively spliced , and alternative splicing of RPS4 and N gene are required for their function [23] , [33] . In mos14-1 , alternative splicing of both SNC1 and RPS4 are altered . This effect probably contributes to the suppression of snc1 mutant phenotypes by mos14-1 and compromised RPS4 function in the mos14-1 single mutant . In addition to the altered ratio of transcript variants , the expression levels of snc1 and RPS4 were also reduced . The reduced expression of snc1 and RPS4 is probably caused by splicing defects resulting from the reduced nuclear localization of SR proteins . In mos14-1 snc1 npr1 , the SNC1 TV-4 transcript level is only modestly reduced , suggesting that reduced accumulation of TV-4 may not be the only factor that contributes to the complete suppression of snc1 mutant phenotype . In addition to reduced accumulation of TV-4 , levels of SNC1 TV-1 , TV-2 and TV-3 are considerably increased in mos14-1 snc1 npr1 . These transcripts are predicted to produce truncated snc1 proteins because of introduction of early stop codons . It is possible that these truncated proteins may interfere with the function of the full-length snc1 . Because snc1 and RPS4 are not the only genes whose splicing are affected by mos14-1 , altered splicing of one or more unknown positive regulators of plant defense could also contribute to the suppression of snc1 mutant phenotypes . In addition to the compromised resistance responses mediated by snc1 and RPS4 , basal resistance against P . s . t . DC3000 is also compromised in mos14-1 . It remains to be determined how mos14-1 affects basal resistance . One possibility is that MOS14 is required for the splicing of one or more R genes that contribute to basal resistance against P . s . t . DC3000 . Alternatively , mos14-1 may cause splicing defects in defense regulators required for basal resistance . In summary , we have identified MOS14 as a nuclear transporter of SR proteins . Our data suggest that regulation of R gene splicing by SR proteins is critical for plant immunity . Future studies on individual SR proteins will help us better understand how SR proteins regulate the splicing of R genes .
All plants were grown at 23°C under 16 hr light/8 hr dark in plant growth rooms or chambers , if not specifically mentioned . To identify mutations that suppress the mutant phenotypes of snc1 , snc1 npr1 seeds were mutagenized with EMS . About 30 , 000 M2 plants representing about 1 , 500 M1 families were screened for suppression of the dwarf morphology of snc1 npr1-1 . Mutants lacking the dwarf phenotype were further analyzed for suppression of the constitutive defense responses in snc1 npr1 . About 0 . 1 g tissue was collected and RNA was extracted by Takara RNAiso reagent . The RNA was treated with Promega RQ1 RNase-Free DNase to remove contaminating genomic DNA . Reverse transcription was subsequently carried out using oligo-dT and the M-MLV RTase cDNA synthesis kit from Takara . About 200 ng of total RNA was included in each RT reaction . For semi-quantitative and real-time PCR , one fiftieth of the cDNA was used in each reaction . A total of 40 cycles were performed for semi-quantitative RCR except 28 cycles for ROC1 . Real-time PCR was carried out using Takara SYBR® Premix Ex Taq™ II . The primers for real-time PCR analysis of PR1 , PR2 , SNC1 [34] and ROC1 ( also called cyclophilin ) [35] were described previously . ROC1 is a housekeeping gene without introns . The sequences of primers used for SNC1 and RPS4 transcript variants analysis are shown in Table S1 . Primers to amplify U1-70K [28] , AtSR30 and AtSR34 [29] were described previously . For infections with H . a . Noco2 , three-week-old soil-grown plants were sprayed with H . a . Noco2 at 5×104 spores/ml . The inoculated seedlings were subsequently kept in a growth chamber with high humidity ( >80% ) at 18°C under 12 hr light/12 hr dark cycle for seven days before growth of H . a . Noco2 was quantified , as previously described [36] . For infections with P . s . t . DC3000 or P . s . t . DC3000 avrRps4 , five-week-old soil-grown plants were infiltrated with bacterial suspensions ( OD600 = 0 . 001 ) in 10 mM MgCl2 . Samples were taken at day 0 and day 3 . To analyze the SA levels in the mutant plants , SA was extracted using a previously described procedure [37] and measured by high-performance liquid chromatography . For the transgenic complementation test , three PCR fragments , F12R37 ( 3 . 9K ) , F14R38 ( 3 . 8K ) and F23R19 ( 2 . 9K ) covering the 11 kb region where MOS14 is located were amplified from wild type genomic DNA . The primers used for amplification of F12R37 , F14R38 and F23R19 are F12 , R37 , F14 , R38 , F23 and R19 respectively , and their sequences are provided in Table S1 . These fragments were sequentially sub-cloned into pBluescript SK+ . The complete 11 kb fragment was subsequently cloned into a modified pGreen0229 vector containing the NOS terminator to obtain the construct pMOS14:MOS14 . The final construct containing MOS14 was transformed into mos14-1 snc1 npr1 through Agrobacterium-mediated transformation . For the subcellular localization study of MOS14 , PCR fragments F12R37 ( 3 . 9K ) , F14R38 ( 3 . 8K ) and F23R20 ( 2 . 9K ) were sequentially sub-cloned into pBluescript SK+ . The primers used for amplification of F23R20 are F23 and R20 and their sequences are listed in the Table S1 . The 11 kb fragment described above was cloned into a modified pCambia1305 vector expressing C-terminal tagged GFP to obtain pMOS14:MOS14-GFP . For transient expression of SR proteins in protoplasts , full-length cDNAs of AtRS2Z33 , AtRSZ21 , AtRS31 and AtSR34 without the stop codons were amplified by PCR and cloned into the modified pUC19 vector pUC19-35S-cmGFP4 that expresses GFP under the 35S promoter . To obtain transgenic plants expressing snc1 cDNA , full-length snc1 cDNA was amplified from total cDNA of snc1 and cloned into a modified pGreen0229 vector . The cDNA clone was sequenced to make sure the sequence is correct and no intron was retained . To obtain transgenic plants expressing AtSR34-GFP and AtRSZ21-GFP , full-length cDNAs of AtSR34 and AtRSZ21 without the stop codons were amplified by PCR and cloned into a modified pCambia1300 vector expressing C-terminal tagged GFP under 35S promoter . The constructs were transformed into Col-0 and mos14-1 through Agrobacterium-mediated transformation . To make constructs for the yeast two hybrid assays , an SfiI restriction site was first introduced to the multiple cloning site of pGBKT7 and pGADT7 to obtain pGBKT7a and pGADT7a , respectively . cDNA expressing the N-terminal or C-terminal region of MOS14 and AtRAN1 were amplified by PCR and cloned into pGBKT7a . Full-length cDNAs of AtRS2Z33 , AtRSZ21 , AtRS31 and AtSR34 were amplified by PCR and cloned into pGADT7a . cDNA expressing the N-terminal or C-terminal region of MOS14 were also cloned into pGADT7a . The plasmids expressing the MOS14 fragments were co-transformed with the vectors expressing AtRAN1 or one of the SR proteins into the yeast strain PJ694α for yeast two-hybrid analysis . For confocal fluorescence microscopy analysis of MOS14-GFP , the roots or leaves of six-day-old seedling grown on MS plates were first stained with propidium iodide ( PI ) for 1 min and then washed in ddH2O for at least three times . The concentration of PI used for staining the roots was 10 µg/ml , whereas the concentration of PI used for the leaves is 10 mg/ml . The stained sample was observed using a Zeiss Meta 510 confocal microscope . Excitation wavelengths for GFP and PI were 488 nm and 543 nm , respectively . For root samples , the emission filter used for PI was LP560 nm . For leaf samples , the emission filter used for PI was BP560 nm-615 nm . For both root and leaf samples , the emission filter for GFP was BP505 nm-530 nm . Plasmids used for protoplast transfections were purified with Invitrogen PureLink™ HiPure Plasmid Filter Purification Kit . Transformation of wild type or mos14-1 protoplasts was performed as previously described [38] . After transformation , protoplasts were kept in the dark for about 16 hours . The transformed protoplasts were examined using a Zeiss Axiovert 200 fluorescence microscope . The pictures of representative protoplasts were taken using confocal fluorescence microcopy . For autofluorescence , the emission filter used was 650 nm-740 nm . Confocal fluorescence microscopy analysis of transgenic plants expressing AtSR34-GFP and AtRSZ21-GFP was performed on three-week-old seedlings using a procedure described in the analysis of MOS14-GFP localization . Sequence data from this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers: At5g62600 ( MOS14 ) , At2g14610 ( PR1 ) , At3g57260 ( PR2 ) , At4g38470 ( ROC1 ) , At2g37620 ( Actin1 ) , At5g44340 ( β-tubulin4 ) , AAD38537 ( hTRN-SR1 ) , CAB42634 ( hTRN-SR2 ) , NP608708 ( dTRN-SR ) , AF025464 ( TSR1 ) and CAA99366 ( MTR10a ) .
|
Plant immune receptors encoded by Resistance ( R ) genes play essential roles in defense against pathogens . Multiple R genes are alternatively spliced . How plants regulate the splicing of these R genes is unclear . In this study , we identified MOS14 as an important regulator of two R genes , SNC1 and RPS4 . Further analysis showed that MOS14 functions as the nuclear import receptor for serine-arginine rich ( SR ) proteins , which play diverse roles in RNA metabolism . Loss of the function of MOS14 results in altered splicing patterns of SNC1 and RPS4 and compromised resistance mediated by snc1 and RPS4 , suggesting that nuclear import of SR proteins by MOS14 is required for proper splicing of these two R genes and is important for their functions in plant immunity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"plant",
"science",
"plant",
"biology",
"plant",
"pathogens",
"plant",
"pathology",
"plant",
"genetics",
"biology",
"plant",
"physiology"
] |
2011
|
Transportin-SR Is Required for Proper Splicing of Resistance Genes and Plant Immunity
|
Nodular Oesophagostomum genus nematodes are a major public health concern in some African regions because they can be lethal to humans . Their relatively high prevalence in people has been described in Uganda recently . While non-human primates also harbor Oesophagostomum spp . , the epidemiology of this oesophagostomosis and the role of these animals as reservoirs of the infection in Eastern Africa are not yet well documented . The present study aimed to investigate Oesophagostomum infection in terms of parasite species diversity , prevalence and load in three non-human primates ( Pan troglodytes , Papio anubis , Colobus guereza ) and humans living in close proximity in a forested area of Sebitoli , Kibale National Park ( KNP ) , Uganda . The molecular phylogenetic analyses provided the first evidence that humans living in the Sebitoli area harbored O . stephanostomum , a common species in free-ranging chimpanzees . Chimpanzees were also infected by O . bifurcum , a common species described in human populations throughout Africa . The recently described Oesophagostomum sp . found in colobine monkeys and humans and which was absent from baboons in the neighboring site of Kanyawara in KNP ( 10 km from Sebitoli ) , was only found in baboons . Microscopic analyses revealed that the infection prevalence and parasite load in chimpanzees were significantly lower in Kanyawara than in Sebitoli , an area more impacted by human activities at its borders . Three different Oesophagostomum species circulate in humans and non-human primates in the Sebitoli area and our results confirm the presence of a new genotype of Oesophagostomum recently described in Uganda . The high spatiotemporal overlap between humans and chimpanzees in the studied area coupled with the high infection prevalence among chimpanzees represent factors that could increase the risk of transmission for O . stephanostomum between the two primate species . Finally , the importance of local-scale research for zoonosis risk management is important because environmental disturbance and species contact can differ , leading to different parasitological profiles between sites that are close together within the same forest patches .
Emerging zoonotic diseases are a serious threat to public health and animal conservation . This is especially true for apes , whose close phylogenetic relationship with humans increases the risk of zoonotic transmission between them . Although humans have always shared habitats with non-human primates , the dynamics of their relationships are rapidly changing nowadays . Indeed , non-human primate populations suffer from forest loss and fragmentation [1–3] and an increasing number of them live in anthropogenically disturbed habitats such as farmlands , human settlements , fragments of forest , and isolated protected areas [4–6] . As a consequence , people and non-human primates live in increasing spatial proximity to each other [7] . So far , several cases of pathogen transmission have been reported to have occurred between non-human primates and humans; these include the transmission of viruses ( e . g . [8–10] ) , bacteria ( e . g . [11–13] ) as well as blood-borne parasites ( e . g . [14–16] ) , and intestinal parasites ( e . g . [17–21] ) . Nematodes of the genus Oesophagostomum are intestinal parasites , which frequently infect primates ( including monkeys , apes and humans ) , domestic and wild pigs and ruminants [22 , 23] . Uninodular oesophagostomosis ( i . e . Dapaong tumor ) and multinodular oesophagostomosis [24] , which are caused by O . bifurcum in humans [25] , have been reported in endemic foci in West Africa ( Togo and Ghana ) with an estimated 250 , 000 infected people and a further one million at risk of contracting these parasitic diseases [26] . Third stage larval development in the colon wall induces the aforementioned inflammatory masses that cause severe abdominal pain , diarrhea and weight loss , and occasional death from peritonitis and intestinal occlusion [24] . While a variety of drugs kill Oesophagostomum nematodes , these drugs appear to be less effective on the tissue-dwelling stage because of the difficulty they have passing through the nodule wall [25] ) . Only nine cases published in six studies before the 1980s [23 , 27] and six cases published in one study in 2014 [18] reported the presence of human oesophagostomosis in Uganda . The few reported cases might be related to a low rate of infection . Nevertheless , oesophagostomosis infections may also be underdiagnosed , notably because obtaining a definitive diagnosis by ultrasound examination [26] is rarely undertaken in Ugandan hospitals and dispensaries [18] . Because transmission occurs through ingestion of the infective third-stage larvae [25] present in water , in food or on the ground , oesophagostomosis is a potential zoonotic risk when humans and non-human primates share the same habitats . In Ghana , identification of genetic differences among Oesophagostomum nematodes found in different primate hosts suggested that infection with these nematodes was rarely zoonotic [28 , 29] . Still , the risk of zoonotic infection from the presence of infected chimpanzees in the vicinity of humans was mentioned in a study conducted in Kibale forest in western Uganda [17] , while a recent study undertaken in the same area described a novel Oesophagostomum clade that infects humans and five sympatric species of non-human primates [18] . Because of the severity of the clinical consequences of oesophagostomosis , it should not remain a neglected area of public health . Surprisingly , while captive chimpanzees suffer from oesophagostomosis , no evident clinical signs have been observed in wild chimpanzees so far [30] , except for one individual from Gombe ( Tanzania ) who developed weight loss as well as diarrhea prior to death , without other predisposing factors [31] . It has been suggested that ingestion of rough leaves via swallowing might decrease chimpanzee infestation with the parasite . Hairy leaves have a mechanical effect in preventing third stage larvae from penetrating the colon wall and by abrading the intestinal wall thereby leading to the expulsion of the immature and adult worms present in the nodules and in the gut lumen [32] . Wild chimpanzees that harbor these parasites usually do not suffer lethal infections , whereas humans can die from oesophagostomosis . More specifically , the Sebitoli area , located in the extreme north of Kibale National Park ( KNP ) in Uganda , is an area with high anthropogenic pressure . Indeed , the human demographic density is high at the forest borders ( villages with croplands , tea and eucalyptus plantations , and tea factories ) and a tarmac road crosses the forest and the Sebitoli chimpanzee home range [33] . Additionally , the Sebitoli forest was commercially logged in the 1970s [34 , 35] and is today mostly degraded and regenerating with 70% of the land cover affected [36] . Studies conducted in Sebitoli but also in KNP and in close-by forest fragments have shown that non-human primates ( i . e . , chimpanzees , Pan troglodytes; baboons , Papio anubis; black and white colobus , Colobus guereza; redtail monkeys , Cercopithecus ascanius; and vervet monkeys–Chlorocebus pygerythrus ) frequently feed on croplands at the forest edge [37 , 38] . At these sites , farmers rank baboons as the worst pests among the non-human primates because they regularly forage in large groups on several different crops at varying stages of maturity [37 , 39] . Since then , baboons have come to represent a particularly important risk for people living close to the forest area , equally to chimpanzees , which are less frequent crop raiders but our closest relatives . Also , in the Kibale region , almost 9% of the population has reported direct contact with non-human primates [40] via touching carcasses ( 60 . 8% of cases ) or butchering these animals ( 16% of cases ) ; this is important because these activities represent a high-risk for zoonotic transmission of pathogens . In addition to poachers , other humans ( e . g . researchers , field assistants and rangers ) also regularly enter the forest and are in close proximity with the non-human primates living there . These observations underline the necessity to determine which species pose a risk of transmission in an environment where the risk of zoonosis appears to be particularly important . This study aimed to investigate Oesophagostomum infection in terms of parasite species diversity , rate of infection and parasite loads , in four primate species ( humans , chimpanzees–an ape species and closest relative to humans , baboons–a terrestrial monkey species , black and white colobus–an arboreal monkey species ) to better understand the zoonotic risk associated with increased spatial proximity locally between humans and wildlife in an area subject to a high rate of environmental disturbance .
KNP is located in southwestern Uganda ( 0°13′ to 0°41′N and 0°19′ to 30°32′E ) , covers 795 km2 [41] and declines in elevation from 1590 m in the north to 1110 m in the south [42] , while temperatures in the park range from 23 . 3 to 24 . 2°C ( annual mean daily maximum; [43] ) . The park , home of 13 non-human primates species [44] , is a mosaic of mature forest ( 58% ) , colonizing forest formally used for agriculture ( 19% ) , grassland ( 15% ) , woodland ( 6% ) and lakes and wetlands ( 2% ) [45] ( Fig 1 ) . At Sebitoli , a long-term research project the “Sebitoli Chimpanzee Project” was initiated in 2008 . The chimpanzee habituation level in this area did not allow researchers to collect identified feces from these animals at the time of the study ( see below ) and other non-human primates are not under habituation . Human pressure around Sebitoli is high . In fact , the human population density within 5 km of the boundary is ~260 inhabitants/km2 to the west and 335 inhabitants/km2 to the east of the park [46] , and 82% of the Sebitoli chimpanzee home range borders are in contact with anthropogenic features [36] . Additionally , an asphalted road with high traffic intensity linking Kampala to the Democratic Republic of Congo crosses the forest [33] . Human fecal samples ( N = 326 ) were collected during five different periods ( July–August 2010 , July–August 2011 , March–April 2012 , February–March–April 2013 , December 2013–January 2014 ) from people in six villages less than 500 m from the border of the park ( Fig 1 ) . Some people were sampled several times at a minimum of 1-month intervals . Participants received instructions on how to collect and store the fecal samples and researchers retrieved them within 1 day of collection . Fecal samples from baboons ( Papio anubis ) ( N = 97 ) and black and white colobus ( Colobus guereza ) ( N = 96 ) were collected during the first three study periods and fecal samples from chimpanzees ( Pan troglodytes schweinfurthii ) ( N = 228 ) were collected during the five study periods . These samples ( < 6 hours old ) were collected non-invasively in the forest and immediately transferred to plastic bags . It is likely that samples from the same animals were collected several times because we did not know the identity of the individual from whom the stool was collected . Also , the sample sizes were larger than the number of animals within the groups or community studied . Fecal samples were collected from a relatively large area of the forest ( 25 km2 ) . Fecal samples from humans and non-human primates were inspected before processing them to check for the presence of macroscopic parasites and to note the consistency ( liquid , soft or pasty , solid or normal , and dry or hard , according to a method published previously [47] ) . Two grams of a fresh fecal sample from a human or a non-human primate was preserved in 18 mL of 10% formalin saline solution . These samples were analyzed at the Department of Parasitology ( Ecole Nationale Vétérinaire d’Alfort , France ) and a direct microscopic examination of two 50-μL smears was performed to access the presence of hookworm-like eggs ( at 100–400x magnification after homogenization ) . Because collecting fecal samples from different animal species and from humans living in separated villages was time consuming , as well as the fecal sample storage for both microscopy and molecular analyses , we were unable to conduct systematically other microscopic examination methods such as fecal flotation or sedimentation on fresh samples [48] . Each egg was identified according to its size , color , shape and morula aspect . Eggs were not classified at the genus level because the Oesophagostomum genus cannot be distinguished with certainty from hookworm nematodes ( i . e . Ancylostoma sp . , Necator sp . ) by microscopy alone . To establish an arithmetic parasite load ( eggs per gram of feces; epg ) , the total number of eggs , larvae and adults from a 100-μL aliquot was counted and multiplied by 100 . Then , a corrected parasite load ( CPL ) was obtained according to the stool consistency ( i . e . x2 when the feces were soft , x3 when the feces were liquid [49] ) . Fecal samples ( N = 15 baboon samples; N = 22 black and white colobus samples; N = 39 chimpanzee samples; N = 39 human samples ) were randomly selected and then stored differently according to the study periods: ( 1 ) at least 4 g was diluted in 18 mL of 95% ethanol ( 2010 ) ; ( 2 ) at least 4 g was diluted in 18 mL of 95% ethanol over a 24 h period , after which the supernatant was removed and the sedimented feces dried on a silica gel beads ( 2011 , 2012 ) ; and ( 3 ) 10 mL of coproculture products were stored in 50 mL of 95% ethanol ( 2012 , 2013 ) . One gram of fresh feces was mixed with charcoal and vermiculite , and cultured for 10 to 17 days in a Petri dish at room temperature ( approximately between 18°C and 26°C ) . During the culture , regular inspection was done to keep the culture moist and to softly stir the mixture to minimize fungal growth . At the end of the culture , all the mixture was transferred on two layers of gauze and then larvae products were collected via the Baermann procedure , described in [48] . Molecular analyses were performed at the Eco-anthropology and Ethnobiology Laboratory ( National Museum of Natural History , France ) . DNA was extracted from coproculture products , feces in ethanol or dried feces . DNA from 10 mL of a larval culture was extracted with a QIAamp DNA Mini Kit Tissue ( Qiagen , Chatsworth , CA , USA ) according to manufacturer’s protocols but with the following modifications . In step 1 , 1 mL of phosphate-buffered saline ( PBS ) was added to the sample then mixed thoroughly by vortexing , centrifuged and the supernatant removed ( 3 times ) , followed by an overnight incubation at -20°C with the ASL buffer included in the kit . DNA was extracted from 100 mg of dried sample or from 2 mL of ethanol sample with a QIAamp DNA Stool Kit ( Qiagen ) . We made a modification to step 1 where 1 mL of PBS was added to the sample , the sample allowed to sit for 10 min at room temperature , and then incubated overnight at 70°C with ASL buffer . An external PCR targeting the ribosomal internal transcribed spacer 2 gene ( ITS2 ) using NC1 and NC2 primers [50] , followed by an internal semi-nested PCR using OesophITS2-21 [18] and NC2 primers were performed . PCR reactions were cycled in a BioRad CFX ( Bio-Rad Laboratories , Hercules , CA , USA ) with a mix of sterile water , Taq polymerase , buffer , primers , dNTPs , fluorochrome and an intercalating agent ( Ssofast Evagreen ) . The following temperature profile was used for the external PCR: 94°C for 2 min; 45 cycles of 94°C for 10 sec , 60°C for 30 sec , 72°C for 1 min , and a final extension at 72°C for 10 min . The semi-nested PCR followed a slightly different temperature profile: 95°C for 2 min; 45 cycles of 95°C for 10 sec , 55°C for 30 sec , 72°C for 1 min; and a final extension at 72°C for 5 min . The PCR products were sequenced at the Pasteur Institute ( Lille , France ) by the Genoscreen Laboratory using primer NC2 and OesophITS2-21 . DNA sequences were hand-edited and cleaned with 4Peaks software . Sequence alignments were performed on SeaView software by inputting our sequences with the sequences obtained by Ghai et al . [18] ( accession numbers: KF250585 –KF250660 ) and by Krief et al . [17] ( KT592234 , KT592235 ) . In addition , we included three Oesophagostomum stephanostomum reference sequences ( AF136576 , AB821022 , AB821031 ) , one O . bifurcum sequence ( AF136575 ) and five outgroups ( HQ844232 , Y11736 , Y11735 , Y10790 , AJ006149 ) . Phylogenetic trees were established using the maximum likelihood method in MEGA [51] and the Hasegawa-Kishino-Yano substitution model with five discrete gamma categories [52] . To assess the phylogenetic robustness of the tree , 1000 bootstrap replicates were performed . All the relevant sequences have been deposited in GenBank under the accession numbers: KR149646 –KR149658 . Special attention was paid to avoiding contamination during all the process stages , that is , during the field collections ( utilization of gloves and tongue depressors ) , during sample storage ( use of sterile instruments on different days of collection for each primate species ) , and during DNA extraction and amplification ( separation of the samples by species on the plates and repetitions ) . The percentage values of the fecal samples that were positive for hookworm-like eggs were considered a proxy for the infection prevalence and the mean corrected parasite load ( including infected and non-infected samples ) as a proxy for the infection intensity . All statistical tests were performed via R software [53] and were two-tailed with the criterion of statistical significance set at P < 0 . 05 . When samples sizes were small or data were not normally distributed , nonparametric procedures were used . The Uganda National Council for Science and Technology , the Uganda Wildlife Authority and the National Museum of Natural History in France ( Memorandum of Understanding SJ 445–12 ) reviewed and approved the animal care and human research protocols . The free-ranging chimpanzees and monkeys were studied without invasive methods and without interacting with the researchers . Additionally , we obtained the approval of each village chairperson to conduct our research , and human volunteers gave their written informed consent . All volunteers were free to withdraw from the study at any time . The purpose , methods and preliminary findings of the research were explained to all volunteers . Each fresh human sample collected herein was analyzed microscopically within 12h of collection via a fecal flotation to ascertain the parasite species present [49] and to immediately inform the volunteer whether or not he or she had an infection . Following the recommendations of the local dispensary in the area , a single dose anthelmintic treatment ( albendazole ) was given to any person infected with nematodes ( hookworm-like species , Ascaris lumbricoides or Trichuris trichiura ) . Persons who received anthelmintic drug treatment could be resampled but at 8-monthly intervals as a minimum .
The proportions of samples containing hookworm-like eggs varied significantly among the host species ( Chi-square = 395 . 2; df = 3; P<<0 . 001 ) . Chimpanzees had the highest percentage of positive samples ( 77 . 2%; 176/228 ) , followed by baboons ( 71 . 1%; 69/97 ) . Humans ( 6 . 4%; 21/326 ) and black and white colobus ( 2 . 1%; 2/96 ) were the two species with low proportions of samples positive in hookworm-like eggs . The mean hookworm-like egg load was also significantly higher in chimpanzee feces ( CPLmoy = 535 epg ) and in baboon feces ( CPLmoy = 343 epg ) compared with human feces ( Mann-Whitney tests: W = 64443; P<<0 . 001 and W = 26443; P<<0 . 001 , respectively ) and in black and white colobus feces , whose mean corrected parasite loads were < 100 epg ( Mann-Whitney tests: W = 19512; P<<0 . 001 and W = 8013; P<<0 . 001 , respectively ) ( Fig 2 ) . During the dry season , the parasite load was higher in baboons than in chimpanzees ( Mann-Whitney test: W = 2269; P<0 . 01 ) but in the wet season , it was three times higher ( statistically significant ) in chimpanzees than in baboons ( Mann-Whitney test: W = 3071; P<0 . 001 ) . In comparison with the dry season , during the wet season , the mean hookworm-like egg load was significantly lower in baboon feces ( Mann-Whitney test: W = 551; P<<0 . 001 ) while it was higher in chimpanzee feces ( Mann-Whitney test: W = 7014; P<<0 . 001 ) ( Fig 2 ) . PCR products from 61 out of the 115 fecal samples tested ( 92 . 3% ( 36/39 ) of the chimpanzee samples , 93 . 3% ( 14/15 ) of the baboon samples , 36 . 4% ( 8/22 ) of the black and white colobus samples and 28 . 2% ( 11/39 ) of the human samples ) produced interpretable DNA sequences . Approximately 50% of the samples stored in ethanol or in ethanol followed by silica gel and more than 75% of the samples that generated coproculture products gave interpretable DNA sequences ( Table 1 ) . Fifty-seven DNA sequences ( 34 from chimpanzees , 14 from baboons , 5 from black and white colobus and 4 from humans ) matched the Oesophagostomum ITS2 sequences already published . Sequences corresponding to O . stephanostomum were found in 82 . 1% ( 32/39 ) of the chimpanzee samples , 18 . 2% ( 4/22 ) of the black and white colobus samples and 10 . 3% ( 4/39 ) of the human feces samples . Sequences from one man ( 60 years of age ) and three women ( 16 , 32 and 41 years of age ) clustered with the published O . stephanostomum sequence . Sequences matching with O . bifurcum were obtained from 80% ( 12/15 ) of the baboon samples , from 5 . 1% ( 2/39 ) of the chimpanzee samples and from 4 . 5% ( 1/22 ) of the black and white colobus sample ( Fig 3 ) . Only 13 . 3% ( 2/15 ) of the baboon fecal samples were positive for the new sequence type of Oesophagostomum sp . , recently described in Ghai et al . [18] ( Fig 3 ) . Microscopic examinations showed that despite the high infection prevalence in olive baboons and the low infection prevalence in humans being common between sites , Sebitoli black and white colobus monkeys had a low prevalence of infection , as has been described previously in Kanyawara by Gillespie et al . [54] but not by Ghai et al . [18] . The infection prevalence in chimpanzees and the arithmetic mean corrected for parasite load were significantly higher in Sebitoli than in Kanyawara ( Mann-Whitney test: W = 6020 , P<0 . 001 ) ( Table 2 ) . At the molecular level , the Oesophagostomum clades obtained from chimpanzees in Sebitoli and Kanyawara were similar in terms of the predominance of O . stephanostomum and O . bifurcum ( less frequent ) ( Table 3 ) . All of the Kanyawara baboon samples were positive for the O . bifurcum clade . In addition to O . bifurcum , the newly described Oesophagostomum clade was identified in 14 . 3% of the Sebitoli baboon fecal samples ( Table 3 ) . The rate of Oesophagostomum infestation of human feces and colobus feces was twice as low in Sebitoli as in Kanyawara . Specifically , 60% of the Oesophagostomum species in Kanyawara black and white colobus monkeys comprised the newly described Oesophagostomum sp . clade [18] , whereas 80% were O . stephanostomum in Sebitoli colobine monkeys but Oesophagostomum sp . was not detected . Humans living in the Sebitoli area harbored only O . stephanostomum , while villagers from the Kanyawara area were only infected by Oesophagostomum sp . ( Table 3 ) .
In the present study , microscopic and molecular approaches were used to reveal the prevalence and parasitological load of Oesophagostomum sp . in three non-human primate species and humans living in close proximity in a forested area . Our results provide the first evidence that some humans living in the Sebitoli area are infected by O . stephanostomum , a common species in free-ranging chimpanzees . Moreover , the chimpanzees also harboured O . bifurcum , a species commonly described in humans . Finally , the existence of the new clade Oesophagostomum sp . described in black and white colobus monkeys and humans in Kanyawara ( a neighboring site to Sebitoli ) by Ghai et al . [18] , was confirmed in the Sebitoli region as two baboon fecal samples were infected with it . Microscopy revealed that the infection prevalence and the parasite load were significantly higher in the Sebitoli chimpanzees than in the Kanyawara ones . Sebitoli chimpanzees had a high infection prevalence compared with colobine monkeys and humans . While the Oesophagostomum species isolated from Sebitoli and Kanyawara chimpanzees are similar , both the prevalence of infection and the corrected arithmetic mean parasite load were higher in the Sebitoli apes . Without additional data , it appears to be difficult to attribute these differences to specific causes . Indeed , individuals may differ in terms of infection rates and parasitic loads according to demography ( 1 . 5 individuals/km2 at Kanyawara vs . 3 . 2 individuals/km2 at Sebitoli , leading to increased transmission among individuals; [58 , 59] ) and behaviour ( e . g . higher association strength between members of the same community with increased grooming sessions; [60] ) . These observations might also result from a difference in the chimpanzees’ physiology , immunity , and environment ( e . g . difference in proximity to humans and their livestock ) . Because of their proximity to humans , Sebitoli chimpanzees are more affected by stress , which can be evidence through increased signs of anxiety when chimpanzees leave the forest to crop raid at the borders , and when they cross a tarmac road with high traffic cutting their home range [33 , 38] ) . Stress could decrease their immunity level [61 , 62] , and thereby make them more susceptible to parasitic infections . Additionally , about 30% of the chimpanzees have limb deformities caused by poaching , 10% of the individuals suffer from facial dysplasia [63] and one of them has a cleft lip [64] . Such mutilations and congenital diseases—suspected to be caused by prenatal exposure to teratogen chemicals—may be associated with other health disorders in the affected individuals and decrease individual’s immunity to pathogens such as parasites [65–67] . Other individuals—without abnormal phenotypes—may also experience effects of such exposure . A similar prevalence of Oesophagostomum spp . and other hookworms was observed in villagers living in Sebitoli and Kanyawara ( 28 . 2% of the Sebitoli villagers and 25% of the Kanyawara villagers ) , but two different Oesophagostomum clades were distinguished: O . stephanostomum in Sebitoli and Oesophagostomum sp . in Kanyawara . O . stephanostomum is common in non-human primates , particularly great apes [17 , 18] . In the present study , humans infected with O . stephanostomum came from Sebitoli village ( one man and two women ) and from Kyansimbi village ( one woman ) . They were used to seeing chimpanzees and baboons in their gardens , at locations less than 500 m from the forest edge . Interestingly , one of them was used to sleeping in a small hut to prevent animal crop raiding during the night . However , Sebitoli chimpanzees often feed in maize gardens in large parties and they can stay a long time in croplands [38] , likely contaminating the fields when defecating . In addition , maize begins to ripen and be consumed by both humans and chimpanzees at the end of the wet season ( Cibot et al . , submitted ) when the rate of Oesophagostomum infection in chimpanzees was the highest . This period poses the highest risk of pathogen transmission through the fecal-oral route . Taken together , these results and observations showing a high degree of spatiotemporal overlap between humans and chimpanzees , which are phylogenetically close species , represent factors that could enhance the risk of transmission for O . stephanostomum between chimpanzees and people . In this study , black and white colobus had a low prevalence of infection , which could be related to their arboreality ( making them less vulnerable to nematode parasites with a life cycle including soil [18 , 68] ) and to their diet ( important ingestion of secondary compounds in leaves with potential anthelminthic properties [30] ) , and they are rarely in contact with humans ( Cibot et al . , submitted ) . While oesophagostomosis lesions in baboons are identical to the ones described in chimpanzees and humans ( nodules in the intestinal wall; [31 , 69] ) , O . bifurcum , which appears to be the most common species in Sebitoli baboons , was not found in any of the villagers . Nevertheless , because Sebitoli baboons are reported to be the most frequent crop raiders ( Cibot et al . , submitted ) and two baboon fecal samples were infected with the new Oesophagostomum sp . clade described in humans from Kanyawara , we still cannot exclude the potential disease transmission between the two species . Surprisingly , in baboons , the mean corrected parasitic load of hookworm-like eggs was higher during the dry season compared than it was in the wet season ( the opposite of what was observed in humans and chimpanzees in this study ) . Indeed , lower ambient temperatures and higher humidity rates likely favor survival of eggs and larvae in feces , increasing the risk of infection during the wet season [70] . A long-term survey with an increased sample set should be initiated to confirm this result and investigate why an opposite seasonality pattern exists between chimpanzees and baboons . Our findings raised the need for better public health awareness of oesophagostomosis in the Kibale region . Further studies should be conducted to better understand the epidemiology of Oesophagostomum infections in Uganda , and the role played by domestic animals ( cows , sheep , goats , pigs ) , and other wild animals ( antelopes , buffalos , wild pigs ) in its transmission . Indeed , a recent study undertaken in Tanzania revealed that crop fields regularly used by both chimpanzees and domesticated animals represented potential hotspots for Cryptosporidium transmission [71] . However , when comparing studies , we should interpret carefully data because different methods may have been used . Indeed , in the present study , we detected N . americanus using the OesophITS2-21 primer , which was supposed to be specific and only allow amplification of the Oesophagostomum genus [18] . This result could be caused in part by a difference in the storage procedure between the two studies , with coproculture storage likely favoring the amplification of Necator sp . Similarly , we need to be cautious with the prevalence we obtained after PCR and sequencing since we amplified materials issued from different methods and we sampled unidentified individuals for wild primates compared to other studies . Indeed , the relatively low prevalence in Oesophagostomum spp . in humans and in colobine monkeys could result from an underestimation of all the nematode species or of certain species of Oesophagostomum , due to the culture of fecal samples , which could allow differential development of larva-stage nematodes and which is less sensitive than other methods ( e . g . agar plate culture , which requires sterilization systems that are not available in field settings ) . Then , we likely underestimate the public health problem . Finally , we should also remain prudent when we employ the term “species” in our study , since DNA amplification was only based on a short region of a single gene . While we demonstrated a relatively high genetic diversity within the Oesophagostomum genus , the sequencing of additional genes or the morphological identification of the third stage larvae and adults worms should be established to confirm the species level differentiation . As Bortolamiol et al . [36] , comparing three sites within the same park including Sebitoli and Kanyawara , revealed that small-scale analysis was needed to obtain a better understanding of chimpanzee diet , repartition , density and land use , the parasitological profiles were different between sites and showed that research is required at a more local scale for zoonosis management . For example , the higher prevalence of Oesophagostomum spp . in colobine monkeys observed in Kanyawara compared to our present study may be explained by a difference in the Oesophagostomum species harbored in the black and white colobus monkeys , which could have consequences for zoonosis management . Moreover , working at a small scale is also essential for the human-wildlife zoonotic management as local knowledge and traditional beliefs can differ significantly between people living in relatively close locations and even within the same villages . In fact , near the KNP area , different ethnicities live within the same villages [37] and an increasing number of migrants , notably Congolese people fleeing conflicts in the Democratic Republic of Congo , join western Uganda [72] . Migrants may have a different perception of the risks associated with living in close proximity to wild animals , or the risks associated with hunting animals and eating bush meat [40 , 73] . In any case , health-risk education programs should be better integrated into conservation programs and measures against crop-raiding and other practices such as throwing food from passing vehicles on the Sebitoli road to feed baboons should be implemented . It should also be important to better follow Ugandan patients in hospitals and dispensaries with clinical examination and ultrasonography to better evaluate impacts on human health . Today , in Northern Togo and Ghana , the Oesophagostomum infections seem to have been eradicated after a large scale and intense mass treatment on humans [23] . Finally , our findings reinforce the fact that zoonotic parasites , in the context of increased proximity between non-human primates and humans , should be considered a priority concern for researchers , wildlife managers and health care systems .
|
Nodular worms frequently infect primates , pigs and ruminants . These intestinal nematodes induce inflammatory masses in the colon wall that cause severe abdominal pain , diarrhea , weight loss , and potential death . Through microscopic and molecular analyses , we studied the presence of nodular worms in three non-human primates ( chimpanzees , baboons , black and white colobus ) and humans inhabiting the Sebitoli area , at the extreme north of Kibale National Park in Uganda . Three different Oesophagostomum species were identified in the primates studied and we confirmed the existence of a recently described clade in baboons . Because the Sebitoli chimpanzees displayed a high prevalence of infection and because a high spatiotemporal overlap between humans and apes occurred in our study area , the risk of transmission of O . stephanostomum between the two species cannot be neglected . Thus , our results add to our understanding of nodular worm infection in location where non human primates and humans are co-existing , and underline the necessity to conduct further research at a local scale in a public health concern .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Nodular Worm Infections in Wild Non-human Primates and Humans Living in the Sebitoli Area (Kibale National Park, Uganda): Do High Spatial Proximity Favor Zoonotic Transmission?
|
Traits such as clutch size vary markedly across species and environmental gradients but have usually been investigated from either a comparative or a geographic perspective , respectively . We analyzed the global variation in clutch size across 5 , 290 bird species , excluding brood parasites and pelagic species . We integrated intrinsic ( morphological , behavioural ) , extrinsic ( environmental ) , and phylogenetic effects in a combined model that predicts up to 68% of the interspecific variation in clutch size . We then applied the same species-level model to predict mean clutch size across 2 , 521 assemblages worldwide and found that it explains the observed eco-geographic pattern very well . Clutches are consistently largest in cavity nesters and in species occupying seasonal environments , highlighting the importance of offspring and adult mortality that is jointly expressed in intrinsic and extrinsic correlates . The findings offer a conceptual bridge between macroecology and comparative biology and provide a global and integrative understanding of the eco-geographic and cross-species variation in a core life-history trait .
There is enormous variation in life-history among species and across regions , which ecologists have long sought to explain [1–3] . One trait of particular interest is the number of eggs laid per nest ( clutch size ) by birds , which is central to avian reproductive effort and probably the best-recorded animal life-history trait . The causes of its substantial variation have fascinated behavioural , ecological , and evolutionary biologists for more than 60 years [2 , 3] , but remain highly debated [4–6] . Life-history theory aims to discover the factors that determine intra- and inter-specific variation in life-history traits . This discipline has recently benefited from observational and experimental studies that have quantified important trade-offs , reaction norms , and phenotypic plasticity of the variation within populations and species [7 , 8] . However , this work is limited in its ability to explain the tremendous interspecific and geographic variation in life-histories—from warblers to raptors , and from the tropics to the poles . Inspired by David Lack's original observations [3] , comparative biologists have documented that clutch size tends to be conserved within clades and often co-varies with intrinsic ( biological ) attributes such as body size , nestling development , and nest type . At the same time , others have pointed to extrinsic ( environmental ) influences on clutch size with respect to latitude [1 , 2] , gradients of resource availability , and seasonality [4 , 9 , 10] , and between biogeographic regions . Lack [3 , 11] hypothesised that clutch size may be determined by food abundance during the breeding period , per se , and that northern species have large clutches because daylight periods during the breeding season are longer than those in the tropics . In contrast , seasonality of food abundance is suggested to be linked to clutch size by two alternative mechanisms . Classical life-history theory predicts that high seasonality in the temperate regions , causing high adult mortality , will lead to the evolution of high investment in current reproduction and large clutch sizes because the likelihood to survive until the next breeding season is low [5] . Alternatively , Ashmole [9] argued that high adult mortality in the temperate regions reduces population density , increases per-individual resource availability in the breeding season , and allows temperate birds to nourish large clutches [4 , 12] . Critically , like these suggested extrinsic drivers , clades and their intrinsic traits that may affect clutch size are also not randomly distributed along environmental gradients or realms . Consequently , separate viewpoints focusing on just intrinsic or extrinsic drivers have limited the unification and generalization of our understanding of life-history variation . Here we present an analysis that integrates these perspectives and we assess the variation in clutch size across species and assemblages worldwide . We compiled information on clutch size and other intrinsic ( body mass , migratory behavior , development mode , nest type , diet ) and extrinsic attributes ( latitude , temperature , precipitation , net primary productivity , seasonality , and realm ) for a total of 5 , 290 species of landbirds . This allows us to develop and test a first global model of clutch size that integrates existing viewpoints of life-history variation .
While clutch sizes vary over a large range , more than half of all birds lay 2 or 3 eggs ( mode: 2 , median: 2 . 8; Figure 1 ) . The right-skewed frequency distribution indicates that from a global perspective , the large clutch sizes of northern temperate bird species—few in numbers , but most frequently studied in life-history research [5 , 13 , 14]—are in fact unusual . This highlights the importance for a perspective that extends to the tropics [15] . We find that a key intrinsic factor that distinguishes clades in their typical clutch size is the mode of development [16–19] . Precocial species , with their more mobile offspring , have much larger clutches ( x̄ = 4 . 49 , N = 864 species ) than altricial species ( x̄ = 2 . 85 , N = 4 , 426; t = 21 . 73 , p < 0 . 001 ) . This may be the result of the shorter and less intensive parental care required by precocial young , thus reducing the fitness costs of additional offspring and allowing parents to raise larger clutches [18 , 19] . Development mode is phylogenetically highly conserved ( in our dataset there is no altricial versus precocial variation below the family level ) , and we note that its consequences and associated selection pressures likely influence other intrinsic effects on clutch size . Because of the prevalent influence of development mode and its high collinearity with other potential predictors of clutch size ( Table S1 ) , we assessed all intrinsic effects in combination ( for single-predictor results , see Table S2 ) . A previously noted negative effect of body mass on clutch size [20] is only weakly borne out across the global avifauna for altricial species ( Table 1 and Figure 2 ) . From a global perspective , altricial migrants have larger clutches than nonmigrants , especially when extrinsic effects are not accounted for ( see below ) , which is different from studies that do not include tropical species [17 , 21 , 22] . Another strong intrinsic determinant of clutch size is nest type [23–25] . Cavity nesters , which are naturally exposed to lower rates of nest predation , tend to have larger clutch sizes than open nesters , and species with half-open nests are in between ( Figure 2 ) . Finally , clutch size also varies by diet [3] , with granivores and omnivores laying larger clutches than frugivores and nectarivores . A second suite of constraints on life histories arises from extrinsic factors characterizing the environment of species . One popular “catch-all” surrogate is latitude , which captures much of the global environmental variation because energy availability becomes more seasonal and is usually reduced at higher latitudes . An increase in clutch size toward the poles has long been noted [1 , 2] and is confirmed by our data ( Table S1 ) . The extensive geographical coverage of our data allows us to disentangle the various environmental trends underlying latitude . We use environmental information integrated across the global breeding distribution of each species to quantify the average extrinsic conditions characterizing its broad-scale niche . Specifically , we evaluate the seasonal difference between summer and winter temperatures ( TempMax – TempMin , averaged over 3-mo periods ) , which emerges as the strongest extrinsic predictor: clutch sizes are smallest in species inhabiting relatively aseasonal environments and increase linearly with temperature seasonality ( Figure 2 ) . When this seasonality is controlled for , energy availability in the breeding season ( NPPMax ) has a very weak positive effect on clutch size . This supports the idea that seasonality in resource conditions has a much stronger effect on clutch size than the absolute level of resources in the breeding season [4 , 9 , 10 , 26] . Even though fine-scale variation in productivity may limit NPPMax as estimate of per-individual energy availability during the breeding season , the consistently weak trend over a wide range of environments offers little support for Lack's original hypothesis [3 , 11] . Finally , after accounting for these two environmental variables , only a limited biogeographic signal ( variable Realm ) remains: birds of Australasia , the Afrotropics , and especially Oceania tend to have smaller clutch sizes than birds of other regions . While the strong deviation in Oceania may arise from a potential over-sampling of species with large clutches ( data were available for only 17% of species , compared to 57% elsewhere ) , the life-history strategies of island taxa may be partially shaped by higher population densities and elevated intraspecific competition [27 , 28] . Intrinsic and extrinsic life-history determinants do not act in isolation , and our analytical approach allows us to assess their respective contribution in combination . In the joint model , all five intrinsic and three extrinsic predictors so far discussed emerged as significant ( Table 1 and Figure 2 ) . All extrinsic and most of the intrinsic variables continue to have very strong effects . This is not true for migratory tendency and diet , which are closely tied to climatic conditions ( Table S1 ) . After accounting for temperature seasonality in the combined model , these variables retain relatively little residual effect . This suggests that the larger clutches of altricial migrants arise at least in part from their occupying high-latitude , seasonal environments . By themselves , the pure intrinsic and extrinsic models account for a substantial amount of cross-species variation in clutch size ( 32% and 27% , respectively ) . The combined model explains 44% and predicts the absolute variation well ( Figure 3A; xobserved = −0 . 28 + 1 . 14 ( s . e . = 0 . 018 ) xpredicted; F1 , 5288 = 3 , 914 ) . Different lineages may evolve fundamentally different morphological , physiological , and ecological niches and as a consequence exhibit conservatism in both life-history traits and their intrinsic and extrinsic correlates [29] . Over 90% of the variation in key life-history traits in birds occurs at the level of families and higher [30] . We therefore tested the ability of our combined model to predict variability of clutch size in a phylogenetic nested model that takes into account the order and family membership of the species . The results confirm the strong phylogenetic constraints on many intrinsic predictors of clutch size . After accounting for clade membership , the relative strength of the intrinsic compared to the extrinsic model in the cross-species analysis is reversed . Under phylogenetic control , the extrinsic portion of the model offers stronger predictions ( r2 = 0 . 26 ) than the more phylogenetically conserved intrinsic part ( r2 = 0 . 21 ) , and the latter does not improve overall model fit as much ( delta Akaike information criterion ( AIC ) of 305 and 983 to full model , respectively ) . Extrinsic predictors appear orthogonal to phylogeny and the outstanding importance of temperature seasonality is confirmed . As expected , biogeographic realm membership , which is tightly linked to clade-specific biogeographic history , loses importance in the phylogenetic model . Accounting for the phylogenetic variation at the order and family level explains substantial additional variation , increasing r2 to 0 . 68 ( Figure 3B ) . The geographic context of our data allows us to test the ability of a comparative analysis to predict a global eco-geographic pattern [31] . Specifically , we evaluate how well predictions for each species from our combined cross-species and phylogenetic model fit the observed average ( geometric mean ) clutch size in 2 , 521 bird assemblages of 220 × 220-km size . This provides a test of whether the proposed integration of intrinsic and extrinsic factors is able to recreate observed geographic gradients . Observed average clutch sizes across assemblages show a remarkably strong geographic gradient from an average of 4 . 5 eggs at the high northern latitudes to just over two eggs in the tropics ( Figure 4A ) . We find that our combined cross-species model successfully predicts this geographic pattern ( Figure 3C ) , with a slope almost indistinguishable from 1 ( = 0 . 092 + 0 . 99 ( s . e . = 0 . 004 ) ; r2 = 0 . 97 , F= 75 , 680 ) . The phylogenetic nested model provides an even better fit ( = −0 . 09 + 1 . 05 ( s . e . = 0 . 003 ) ; r2 = 0 . 98 , F = 129 , 800 ) . This exceptional match is confirmed by a visual inspection of the geographic patterns ( Figure 4B ) . It illustrates the statistical strength that assemblage attributes , averaged across species , can achieve even when underlying detectable trends across species are weaker . Geographic trends in the attributes of assemblages ( e . g . , mean assemblage clutch size ) are affected by species' different geographic range sizes , as wide-ranging species occur in a disproportionate number of assemblages and thereby dominate geographic patterns [32] . Geographic trait patterns based on assemblage averages therefore carry a signal of both trait and range size variation across space . Consequently , models of eco-geographic patterns confound correlates of trait variation with correlates of species distributions and range size ( and their respective patterns of spatial autocorrelation ) . For an understanding of potential extrinsic determinants of trait variation , we therefore advocate the use of a comparative approach for biological inference . Additionally testing whether a model can predict geographic patterns allows validation and bridges to the eco-geographic perspective . Our findings on 56% of the world's landbirds empirically support recent theoretical work that highlighted the importance of food seasonality via adult mortality on clutch size [26 , 33] . Highly seasonal environments can cause increased adult mortality [34] , e . g . , because birds have to survive low temperatures and resource conditions in situ or because they have to migrate , which carries risks and costs . Additional effects on population density and , indirectly , per-individual resource availability in the breeding season , then combine to make seasonality of resources the predominant driver of clutch size variation across geographic gradients [26 , 33] . The significance of mortality , in this case mostly of offspring , for the evolution of clutch size is also expressed in the most important intrinsic determinant: nest type . Closed-nesters are subject to much smaller rates of nest predation or loss [23 , 24] , and nest safety may influence clutch size through clutch size–dependent nest predation [35] or the effect of chick survival on adult density [4 , 9] . Nest type is phylogenetically conserved ( its importance decreases strongly when phylogeny is addressed , Table 1 ) , and it is clearly an intrinsic attribute . But its importance may itself be modulated by extrinsic constraints connected to mortality , such as nest predation pressure . This illustrates yet further the strong link between environmental conditions and the evolution and geographic distribution of biological traits such as nest type , which in turn affect life-history traits . Intricate disruptions of such trait associations may arise from climate change and its differential consequences for extrinsic vs . intrinsic determinants . This study confirms many of the previously asserted correlates of clutch size , but moreover demonstrates how life-history traits are jointly determined by the interplay of intrinsic biological traits , the phylogenetic affinities , and the environment of a species . Understanding these interactions is vital for gauging broad-scale life-history consequences of future climate change and their potential impacts on biodiversity . Our findings call for a combination of traditional cross-species comparative analyses with spatial and macroecological approaches to gain a more integrative , conceptual understanding of life-history variation . Using this approach offers a compelling integration of the intrinsic and extrinsic determinants of trait variation that help understand long-noted eco-geographic patterns and critical linkages in a world of change .
We obtained the minimum and maximum clutch size data for 5 , 290 landbird species from a range of literature sources ( see Tables S4–S6 and [36] for detailed overview ) . In this compilation , we did not include brood parasites , as their clutch size is difficult to define ( female birds usually spread a large number of eggs over many host nests ) and is obviously exposed to very different selection pressures . We also excluded predominately pelagic and marine species , because the environmental data in the analysis ( see below ) prevent a straightforward comparison with predominantly terrestrial species . We calculated the species-typical clutch size as the geometric mean of the typical minimum and maximum clutch size ( for an evaluation of intraspecific variation , see below ) . For the same species , we compiled data on species-typical values of potential intrinsic determinants ( development mode , body mass , migratory behavior , nest type , diet ) from the literature ( see Tables S4–S6 for details ) . We classified species into precocial ( newly born young are relatively mobile , covered in feathers , and independent ) and altricial ( newly born young are relatively immobile , naked , and usually require care and feeding by the parents ) . Mass information ( body mass in grams ) was compiled from a variety of sources and averaged across up to four sources , and , if they differed , across sexes . Diet data came from the dataset described in [36] . Species dietary preferences were first recorded across nine major diet categories , and species were subsequently assigned to one out of five primary diets ( vertebrates , invertebrates , fruits or nectar , other plant material or seeds , and omnivore ) . We were able to compile data on nest type data for 2 , 816 species in the analysis and for all 1 , 293 genera . Based on these data , we scored nest type according to levels of nest cover as follows: 1 , open ( e . g . , no nest , cup , scrape , saucer , platform ) ; 2 , half-open nest ( e . g . , pendant , sphere , dome , pouch , crevice ) ; 3 , closed ( cavity , burrow ) . This nest cover score only showed minor variation within genera , and across the 2 , 816 species with data the average genus score was an adequate surrogate for the species-level score ( Nest type ( genus ) = −0 . 00 + 1 . 00 × nest type ( species ) . r2 = 0 . 90 , F1 , 2814 = 26 , 740 ) . We therefore used genus-typical nest type scores for all species lacking data . We acknowledge that this may inflate the type I error for this variable in our nonphylogenetic analysis . We assessed the migratory tendency of species ( Migrant ) and classified them into non-migrants ( non-migrants , only altitudinal and local migrants ) and migrants ( inter- and intra-continental ) following [37] and [38] . We used the extent of occurrence maps of breeding ranges derived from a variety of literature sources ( for details see Figure S1 , Table S4 , and [39] ) ( for justification of grain size see [40] ) to characterize the broad-scale environmental attributes of species . We calculated the centroid of the range in shapefile format to derive the average absolute latitude of a species' range location ( Abs . Latitude ) . We extracted range map occurrences across a 55 × 55 km2 equal area grid ( cylindrical equal area projection ) , which we then linked to an environmental dataset extracted across the same grid at 0 . 01° spatial resolution . Temperature ( °C ) and precipitation ( mm ) data came from University of East Anglia's Climatic Research Unit gridded climatology 1961–1990 dataset [41] at native 10-min resolution . We determined average annual temperature ( TempAvg ) , and total annual precipitation ( PrecTotal ) . For seasonality of temperature , we used average temperature of the coldest and warmest three months across all years ( TempMin , TempMax ) and calculated the annual temperature range ( TempMax – TempMin ) . In order to achieve a representative estimate of both total and seasonal net primary productivity across the second half of the 20th century , we used the output for above-ground NPP ( g Carbon m−2 ) from a recently developed global productivity model [42] based on the Lund-Potsdam-Jena dynamic global vegetation model , including land-use [43] . We averaged model output across 1961–1990 and in the analysis used total annual NPP ( NPPTotal ) , NPP of the most productive three months ( NPPMax ) , and a ratio characterizing seasonality in NPP ( 1 – ( NPPMin/NPPMax ) ) . We use NPP as a broad-scale , general proxy for food abundance [44] , with NPPTotal reflecting average food abundance during the year , NPPMax being food abundance in the breeding season , and ( 1 – ( NPPMin/NPPMax ) ) mirroring seasonal variation in food abundance . Both tropical and temperate birds have been shown to most likely breed in the months with highest NPP [26] . For biogeographic realm , we determined the realm that contains the majority of a species breeding range . We followed the regionalization originally given by [45] as spatially implemented by the World Wildlife Fund ( WWF ) [46] , excluding the Antarctic region . Our taxonomy and phylogenetic placement of families and orders follows [37] , with several updates ( see [39] and Tables S4–S6 for details ) . Following [37] , we excluded from the analysis brood parasites and species that forage predominately in pelagic and marine waters during the breeding season , resulting in a global list of 9 , 391 bird species . Data were not available for all predictor variables for some species , which therefore had to be excluded . The restricted final dataset with full information consists of 5 , 290 species , i . e . , 56% of all qualifying species worldwide ( for a list of species used in the analysis see Table S4 ) . Global representation of qualifying species across realms ranged from high in Nearctic ( 95% ) and Palearctic ( 80% ) , to medium in Indomalaya ( 57% ) and Aftrotropics ( 71% ) , to low in Australasia ( 49% ) and Neotropics ( 45% ) , and poor in Oceania ( 16% ) . We log10-transformed mean clutch size . Visual inspection indicated that this sufficiently stabilized model residuals . We transformed Mass and PrecTotal as log10 ( x ) , TempAvg as log10 ( x + 100 ) , and TempMax – TempMin as log10 ( x + 1 ) . Predictor variables were mostly weakly correlated , although 8 out of 65 variable combinations reached Spearman rank correlations ≥ 0 . 75 ( Table S1 ) . We first ran single-predictor cross-species linear models to test the influence of each predictor variables on log-transformed clutch size ( Table S2 ) , and we then used AIC values to guide variable selection for multi-predictor models . Given the different selection pressures on clutch size between precocial and altricial birds , we analyzed single predictor effects separately for these two groups and examined interactions between the factorial variable precocial and all other continuous predictors . We built the multi-predictor model starting with the variable that had the lowest AIC value in the single predictor models and sequentially added the next-strongest predictor . We first developed multi-predictor models of the selected intrinsic and extrinsic variables separately by sequentially adding the best-performing variables , and then combined both sets of predictors in a joint model ( the “Both” model , Table 1 ) . We only included variables in the final model if the improvement in AIC they provided to the combined model was over ten . In a second set of analyses , we built nested phylogenetic models to address the strong phylogenetic signal in clutch size variation . Specifically , we fitted a linear mixed effects model using the function lme in the nlme library version 3 . 1–83 run in R 2 . 5 . 1 . We used the taxonomic ranks Order and Family in [37] to assign nested clade membership and fitted Order alone and Family nested in Order as random effects in addition to the fixed effects selected in the across-species multi-predictor model ( see above ) . We evaluated model fits using AIC and r2 , which was calculated as the fit between observed and predicted clutch size across effect types . To investigate the spatial patterns in clutch size across assemblages , we compiled 2 , 521 lists of bird species found in 220 × 220-km grid cells across the globe . Grid cells with less than 50% land and under 30 bird species with clutch size data were excluded . For each grid cell assemblage , we calculated the geometric mean clutch size of its members , as observed and as predicted by the cross-species and nested phylogenetic multi-predictor models . Our analysis is based on species-typical intrinsic and extrinsic attributes that are integrated or averaged across all the individual and population level observations available and across the entire geographic range . This collapses the intraspecific variation to one single value , but has the advantage that values are buffered against between-site variation and potential population-level idiosyncrasies . We considered whether the substantial intraspecific variation in clutch size , which our analysis misses , may potentially bias observed patterns . Intraspecific clutch size variation may occur within individuals ( within and between years ) and between individuals ( within and among populations ) . While a full global analysis at the various levels of within-species variation of clutch size would be desirable , the data do not allow it . However , we believe that natural selection on clutch size occurs at all listed levels of organization , and effects across species should be consistent and comparable to effects within species . Potential biases may arise if intrinsic or extrinsic within-species gradients were drastically different to those at the across-species level . This issue should increase in importance toward wide-ranging species as ( i ) their larger geographic spread would likely result in increased trait variance and ( ii ) average intrinsic and extrinsic attributes would be representative for a successively smaller subset of populations . Lacking standardized data on the intraspecific variance in clutch size across a representative set of species , we use as proxy the range of clutch sizes recorded for a species in the literature standardized by the mean ( range to mean ratio ) . We find that thus-measured clutch size variance shows a weak , but significant increase with geographic range size: ( log10-transformed ratio + 1 ) = −0 . 06 + 0 . 03 log10 ( geographic range size ) ; r2 = 0 . 05 , F1 , 5288 = 263 . 6 , p < 0 . 001 ) . To examine the sensitivity of the results of our study to this variation , we selected the roughly half of all species for which the range-to-mean ratio was under 0 . 28 , i . e . , the species with either no recorded intraspecific variation ( 2 , 102 species ) , those with a mean clutch size of 4 that varied at most by one egg ( 276 species ) , and those with a mean clutch size of 7 that varied at most by two eggs ( 33 species ) . We then repeated our combined model ( Table S3 ) . All core results for this subset of species are broadly similar to those observed for all species . We conclude that the trends in our study should be robust to potential biases from intraspecific variation .
|
Why do some bird species lay only one egg in their nest , and others ten ? The clutch size of birds is one of the best-studied life-history traits of animals . Nevertheless , research has so far focused either on a comparative approach , relating clutch size to other biological traits of the species , such as body weight; or on a macroecological approach , testing how environmental factors , such as seasonality , influence clutch size . We used the most comprehensive dataset on clutch size ever compiled , including 5 , 290 species , and combined it with data on the biology and the environment of these species . This approach enabled us to merge comparative and macroecological methods and to test biological and environmental factors together in one analysis . With this approach , we are able to explain a major proportion of the global variation in clutch size and also to predict with high confidence the average clutch size of a bird assemblage on earth . For example , cavity nesters , such as woodpeckers , have larger clutches than open-nesting species; and species in seasonal environments , especially at northern latitudes , have larger clutches than tropical birds . The findings offer a bridge between macroecology and comparative biology , and provide a global and integrative understanding of a core life-history trait .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"evolutionary",
"biology",
"ecology"
] |
2008
|
The Worldwide Variation in Avian Clutch Size across Species and Space
|
Genome-wide association studies ( GWAS ) have identified 38 larger genetic regions affecting classical blood lipid levels without adjusting for important environmental influences . We modeled diet and physical activity in a GWAS in order to identify novel loci affecting total cholesterol , LDL cholesterol , HDL cholesterol , and triglyceride levels . The Swedish ( SE ) EUROSPAN cohort ( NSE = 656 ) was screened for candidate genes and the non-Swedish ( NS ) EUROSPAN cohorts ( NNS = 3 , 282 ) were used for replication . In total , 3 SNPs were associated in the Swedish sample and were replicated in the non-Swedish cohorts . While SNP rs1532624 was a replication of the previously published association between CETP and HDL cholesterol , the other two were novel findings . For the latter SNPs , the p-value for association was substantially improved by inclusion of environmental covariates: SNP rs5400 ( pSE , unadjusted = 3 . 6×10−5 , pSE , adjusted = 2 . 2×10−6 , pNS , unadjusted = 0 . 047 ) in the SLC2A2 ( Glucose transporter type 2 ) and rs2000999 ( pSE , unadjusted = 1 . 1×10−3 , pSE , adjusted = 3 . 8×10−4 , pNS , unadjusted = 0 . 035 ) in the HP gene ( Haptoglobin-related protein precursor ) . Both showed evidence of association with total cholesterol . These results demonstrate that inclusion of important environmental factors in the analysis model can reveal new genetic susceptibility loci .
Genome-wide association studies ( GWAS ) have identified more than 38 larger genetic regions which influence blood levels of total cholesterol ( TC ) , low-density lipoprotein cholesterol ( LDL-C ) , high-density lipoprotein cholesterol ( HDL-C ) and triglycerides ( TG ) [1]–[3] . These studies modeled basic anthropometric confounders , such as sex and age , while leaving out important environmental influences , such as diet and activity . This strategy is statistically suboptimal since the unexplained variation in the phenotype can increase the measurement error and as a result require larger sample sizes to detect a significant effect . Manolio [4] argued strongly for modeling of environmental covariates in GWAS and recommended lipid levels as a paradigmatic phenotype for studying the genetic and environmental architecture of quantitative traits . In order to explore the usefulness of including both environmental and genetic factors in the analysis model , we used lipid measurements from the EUROSPAN study , comprising 3 , 938 individuals for whom genome-wide SNP data ( NSNP = 311 , 388 ) were available [5] . We measured daily intake of food and physical activity at work and at leisure and modeled the influence of those environmental covariates on serum lipid levels in a GWAS . First , data from the Northern Sweden Population Health Study ( NSPHS ) were used as a discovery cohort to screen for SNPs that displayed the lowest p-values when the model was adjusted for environmental covariates . We then used the other , non-Swedish EUROSPAN cohorts for replication of our strongest associations in a candidate gene association study ( CGAS ) . We chose a population living in northern Sweden for the selection of candidate loci because it shows strong natural heterogeneity in certain lifestyle factors ( e . g . diet , activity ) , but homogeneity in other environmental aspects such as climate [6] . Whereas one group is living a modern , sedentary lifestyle found also in the southern part of Sweden and other western European countries , a subgroup of Swedes follows a traditional , semi-nomadic way of life based on reindeer herding . Reindeer herders typically show higher intake of game meat ( reindeer , moose ) , which has a high protein and low fat content , and lower intake of non-game meat , fish , and dairy products among other , lesser differences . They also exert more physical activity at work to tend their reindeer herds , but less activity at leisure [7] .
We performed a GWAS with a lifestyle-adjusted model which included not only sex and age , but also daily intake of game meat , non-game meat , fish , milk products , physical activity at work and at leisure as covariates . We focused on the 0 . 05% of all SNPs with the lowest p-values in the diet- and activity-adjusted model ( corresponding to about 150 SNPs per lipid ) . For total cholesterol , 88 of these were located in a gene and 14 in genes that have been associated with energy metabolism ( http://www . ncbi . nlm . nih . gov/omim/ ) . For LDL-C , 65 SNPs were located in a gene , of which 8 were functionally relevant . Several of the SNPs for LDL-C were identical with those affecting total cholesterol , as expected from the high correlation ( r = 0 . 91 ) between both phenotypes . For HDL-C , SNP rs2292883 , located in the MLPH gene ( Melanophilin ) , showed a genome-wide significant p-value ( p = 1 . 06×10−07 ) . 69 SNPs for HDL-C were located in a gene and 14 of those genes were reported as having a metabolic effect . Finally , for triglycerides , 63 SNPs were located in a gene , but only 4 SNPs in genes with a functional annotation of interest ( Table 1 and Table S1A , S1B , S1C , S1D ) . In order to evaluate the effect of including diet and activity covariates in the association analysis , we overlaid the p-values in the Manhattan plots from the NSPHS for the unadjusted and adjusted GWAS models ( Figure 1 , Figure 2 , Figure 3 , Figure 4 ) . More refined GWAS results separating the effect of adjusting for either diet or physical activity are presented in Figure S1A , S1B , S1C , S1D; and Figure S2A , S2B , S2C , S2D . As expected , the p-values for a number of SNPs were sensitive to the inclusion of both diet and activity covariates in the model . We matched the 0 . 05% SNPs with the lowest p-values ( top SNP list ) between the unadjusted and the adjusted model . For TC , 83 ( 53% ) SNPs were found in both top SNP lists . Those lists contained 102 ( 64% ) identical SNPs for LDL-C and 103 ( 65% ) for HDL-C . The analyses resulted in the same 74 ( 47% ) top SNPs for TG levels ( Table S1A , S1B , S1C , S1D ) . Finally , we compared the p-value changes of the resulting 39 candidate SNPs that are located in genes with a metabolic effect between the diet and activity-adjusted ( full ) model and the unadjusted ( restricted ) model resulting in an up to 27-fold p-value decrease ( Table 1 ) . A food- and activity-adjusted candidate gene association study of the final 39 candidate SNPs in the Scottish ( SC ) sample ( N = 714 ) was applied using similar lifestyle covariates ( Table 2; Table S1E , S1F , S1G , S1H; Table S2 ) . We replicated the effect of rs2000999 ( pSC , unadj = 6 . 16×10−03 , pSC , adj = 4 . 33×10−03 ) in the HP gene ( Haptoglobin-related protein Precursor ) on TC level and the effect of rs1532624 ( pSC , unadj = 2 . 40×10−09 , pSC , adj = 1 . 96×10−09 ) in CETP ( Cholesteryl ester transfer protein ) on HDL-C . In the Swedish cohort ( SE ) , the unadjusted genetic effect of rs2000999 in the HP gene is equivalent to a moderately large difference in average TC level of 20 . 21 mg/dl between the homozyguous genotypes ( MeanSE , unadj ( TC|A/A ) −MeanSE , unadj ( TC|G/G ) = 243 . 16−222 . 95 , Effect SizeSE , unadj = 0 . 41 , Effect SizeSE , adj = 0 . 44 ) ( Effect Size ( ES ) = ( MA/A−MB/B ) /SDpooled ) . Equivalent effects were observed in the Scottish replication sample ( MSC , unadj ( TC|A/A ) −MSC , unadj ( TC|G/G ) = 235 . 36 mg/dl−222 . 54 mg/dl = 12 . 82 mg/dl , ESSC , unadj = 0 . 29 , ESSC , adj = 0 . 52 ) . SNP rs1532624 in the CETP gene is associated with a large , unadjusted difference in HDL-C level of 9 . 99 mg/dl ( MSE , unadj ( HDL-C|A/A ) −MSE , unadj ( HDL-C|C/C ) = 68 . 14 mg/dl−58 . 15 mg/dl , ESSE , unadj = 0 . 73 , ESSE , adj = 0 . 48 ) in the discovery cohort and similar effects regarding direction and size in the replication cohort ( MSC , unadj ( HDL-C|A/A ) −MSC , unadj ( HDL-C|C/C ) = 69 . 79 mg/dl−60 . 75 mg/dl = 9 . 04 mg/dl; ESSC , unadj = 0 . 59 , ESSC , adj = 0 . 57 ) . We also performed an unadjusted candidate gene analysis of the 39 candidate SNPs in all non-Swedish ( NS ) EUROSPAN cohorts ( Scotland , Croatia , The Netherlands , and Italy , NNS = 3 , 282 ) and aggregated the results in a meta-analysis ( Table 2; Table S1I , S1J , S1K , S1L ) . We confirmed the effects of rs5400 ( pNS = 4 . 68×10−02 ) in SLC2A2 on TC . We again found that rs2000999 ( pNS , unadj = 3 . 54×10−2 ) in HP influences TC levels and rs1532624 ( pNS , unadj = 2 . 87×10−20 ) in CETP ( Cholesteryl ester transfer protein ) affects HDL-C levels . The unadjusted genetic effect of rs5400 is equivalent to a moderately large difference in mean TC level of 27 . 11 mg/dl between homozyguous genotypes ( MSE , unadj ( TC|A/A ) −MSE , unadj ( TC|G/G ) = 249 . 30 mg/dl−222 . 19 mg/dl , ESSE , unadj = 0 . 57 , ESSE , adj = 0 . 66 ) in the Swedish Cohort and a small total effect in all non-Swedish samples ( MNS , unadj ( TC|A/A ) −MNS , unadj ( TC|G/G ) = 236 . 69 mg/dl−223 . 34 mg/dl = 13 . 35 mg/dl , ESNS , unadj = 0 . 30 ) . No other associations , including LDL cholesterol or triglycerides levels , were replicated ( all p>0 . 05 ) . The genome-wide significant SNP rs2292883 in the Melanophilin ( MLPH ) gene found in the Swedish cohort was not confirmed .
Environmental covariates may either act as moderators , mediators or even suppressors , thereby affecting the discovery of genetic susceptibility loci [8] , [9] . Therefore , we conducted a GWAS , modeling genetic and important environmental effects , such as food intake and physical activity , on serum levels of classical lipids . To our knowledge , this is the first GWAS on blood lipid levels modeling environmental factors , in particular major food categories and physical activity , in international cohorts . Our analysis replicated one known locus in the CETP gene [1] and identified two other gene loci in the SLC2A2 and HP gene , respectively , involved in energy metabolism but not previously reported to be associated with cholesterol levels . SLC2A2 encodes the facilitated glucose transporter member 2 ( GLUT-2 , Solute carrier family 2 ) and is predominantly expressed in the liver . Mice deficient in GLUT-2 are hyperglycemic and have elevated plasma levels of glucagon and free fatty acids [10] . Mutations in GLUT-2 cause the Fanconi-Bickel syndrome ( FBS ) characterized by hypercholesterolemia and hyperlipidemia [11] , [12] . Cerf [13] argued that a high-fat diet causes a decreased expression of the GLUT-2 glucose receptor on β-cell islets . As a result , glucose stimulation of insulin exocytosis is impaired causing hyperglycemia , a clinical hallmark of type 2 diabetes . In addition , Kilpelainen et al . [14] found that physical activity moderates the genetic effect of SLC2A2 on type 2 diabetes . These studies suggest that these lifestyle factors could have masked genetic effects in previous , unadjusted GWAS . This is emphasized by the strong increase in statistical significance of the SLC2A2 polymorphisms after adjusting for diet and physical activity , indicating that the examined lifestyle factors modified the effect of this gene . Our supplemental results show that physical activity markedly moderated the genetic effect on total cholesterol . The HP gene encodes the Haptoglobin-related Protein Precursor ( Hp ) , which binds hemoglobin ( Hb ) to form a stable Hp-Hb complex and , thereby , prevents Hb-induced oxidative tissue damage . Asleh et al . [15] identified severe impairment in the ability of Hp to prevent oxidation caused by glycosylated Hb . Diabetes is also associated with an increase in the non-enzymatic glycosylation of serum proteins , so these authors suggested that there is a specific interaction between diabetes , cardiovascular disease and the Hp genotype . It results from the increased need of rapidly clearing glycosylated Hb-Hp complexes from the subendothelial space before they oxidatively modify low-density lipoprotein to form the atherogenic oxidized low-density lipoprotein . The p-value for association between the HP SNP rs2000999 and total serum cholesterol concentration decreased in the model adjusted for diet and physical activity , suggesting that the genetic effect is moderated by diet and physical activity . Our supporting material points out the moderating role of physical activity in particular . We also observed a highly significant association between rs1532624 in CETP and HDL-C levels . The CETP protein catalyzes the transfer of insoluble cholesteryl esters among lipoprotein particles . Variation in CETP is known to affect the susceptibility to atherosclerosis and other cardiovascular diseases [16] . Adjustment for diet and physical activity in our model caused an increase of the p-value of this SNP . Our supporting results indicate that the genetic effect is mediated by diet or by physical activity in a similar way . This study also has some limitations . First , we are aware that our candidate gene association approach covers only a very small fraction of all genomic loci , which is one of the potential reasons why some classical lipid-influencing genes , such as APOE , are not represented in our candidate SNP list . Therefore , our approach is not comprehensive and may have failed to identify other relevant lifestyle-sensitive genetic variants . Nonetheless , we decided to apply this approach to make the best out of the available lifestyle data . Second , our study provides only limited information on the role of individual lifestyle factors for a genetic variant . However , in this study we aimed at amplifying genetic effects by adjusting for a maximum amount of environmental variance in a single model and , therefore , we neglected some of these aspects here . Third , we did not model genetic covariates in known lipid-relevant genes which may also moderate the effect of other genetic predictors . This is due to the focus of this paper on gene-environment relationships . In summary , we have demonstrated that modeling environmental factors , in particular major food categories and physical activity , can improve statistical power and lead to the discovery of novel susceptibility loci . Such models also provide an understanding of the complex interplay of genetic and environmental factors affecting human quantitative traits . Inclusion of environmental covariates represents a much needed next step in the quest to model the complete environmental and genetic architecture of complex traits .
All EUROSPAN studies were approved by the appropriate research ethics committees according to the Declaration of Helsinki [17] . The Northern Swedish Population Health Study ( NSPHS ) was approved by the local ethics committee at the University of Uppsala ( Regionala Etikprövningsnämnden , Uppsala ) . The Scottish ORCADES study was approved by the NHS Orkney Research Ethics Committee and the North of Scotland REC . The Croatian VIS study was approved by the ethics committee of the medical faculty in Zagreb and the Multi-Centre Research Ethics Committee for Scotland . The Dutch ERF study was approved by the Erasmus institutional medical ethics committee in Rotterdam , The Netherlands . The Italian MICROS study was approved by the ethical committee of the Autonomous Province of Bolzano , Italy . The examined subjects stem from five different population-representative , pedigree-based cohorts from the EUROSPAN consortium ( http://www . eurospan . org ) . All studies include a comprehensive collection of data on family structure , lifestyle , blood samples for clinical chemistry , RNA and DNA analyses , medical history , and current health status . All participants gave their written informed consent [18] . A brief description of each population is given below: The Northern Swedish Population Health Study ( NSPHS ) represents a cross-sectional study conducted in the community of Karesuando in the subartic region of the County of Norrbotten , Sweden , in 2006 [5] . This parish has about 1500 eligible inhabitants of whom 740 participated in the study . The final sample consisted of 309 men and 347 women who were aged between 14 and 91 years . The inclusion of diet and activity covariates in the analytical model and according missing values reduced the effective sample size by less than 5% . The Orkney Complex Disease Study ( ORCADES ) is a longitudinal study in the isolated Scottish archipelago of Orkney [19] . Participants from a subgroup of ten islands ( N = 719 ) were used for the presented analysis . The sample comprised 334 men and 385 women aged between 18 and 100 years . The inclusion of diet and activity covariates in the analytical model and according missing values reduced the effective sample size by less than 5% . The VIS study is a cross-sectional study in the villages of Vis and Komiza on the Dalmatian island of Vis , Croatia , and was conducted between 2003 and 2004 [20]–[22] . 795 participants who had both genotype and phenotypic data available were analysed . This cohort included 328 men and 467 women with an age between 18 and 93 years . The Microisolates in South Tyrol Study ( MICROS ) is a cross-sectional study carried out in the villages of Stelvio , Vallelunga , and Martello , Venosta valley , South Tyrol , Italy , from 2001 to 2003 [23] . The 1 , 097 participants ( 475 males , 622 females , age between 18 and 88 years ) presented in this study are those for whom both relevant genotype and phenotype data were available . The Erasmus Rucphen Family Study ( ERF ) is a longitudinal study on a population living in the Rucphen region , the Netherlands , in the 19th century [24] . Fasting total cholesterol , HDL cholesterol and triglyceride levels were available . LDL cholesterol was estimated using the Friedewald formula [25] . The 918 individuals included in this study consisted of the first series of participants with 354 men and 564 women aged between 18 and 92 years . DNA samples were genotyped according to the manufacturer's instructions on Illumina Infinium HumanHap300v2 or HumanCNV370v1 SNP bead microarrays . Both arrays have 311 , 388 SNP markers in common that are distributed across the human genome . Analysis of the raw data was done in the BeadStudio software with the recommended parameters for the Infinium assay and using the genotype cluster files provided by Illumina . Individuals with a call rate below 95% and SNPs with a call rate below 98% , deviating from Hard-Weinberg equilibrium ( pHWE<1×10−6 ) or with a minor allele frequency of less than 1% were excluded from the analysis . Total cholesterol ( TC ) , low-density lipoprotein cholesterol ( LDL-C ) , high-density lipoprotein cholesterol ( HDL-C ) , and triglycerides ( TG ) were quantified by enzymatic photometric assays using an ADVIA1650 clinical chemistry analyzer ( Siemens Healthcare Diagnostics GmbH , Eschborn , Germany ) at the Institute for Clinical Chemistry and Laboratory Medicine , Regensburg University Medical Center , Germany . In the NSPHS cohort , we collected data with a food frequency questionnaire based on the Northern Sweden 84-item Food Frequency Questionnaire ( NoS-84-FFQ ) [26] . We included in the questionnaire several items on foods specific for the lifestyle in this geographic region , in particular on game consumption ( reindeer , moose ) . The answer options consisted of an 11-point format: 0 = “Never” , 1 = “less than 1 time per month” , 2 = “1 to 3 times per month” , 3 = “1 time per week” , 4 = “2 to 4 times per week” , 5 = “5 to 6 times per week” , 6 = “1 time per day” , 7 = “2 to 3 times per day” , 8 = “4 to 5 times per day” , 9 = “6 to 8 times per day” , 10 = “9 to 10 times per day” . The questionnaire was applied in electronic format by a trained study nurse as an interviewer . For each food item we calculated daily intake in gram per day as a standardized unit of measurement and aggregated the items to food categories , such game meat , non-game meat , fish , and dairy products . We evaluated the construct validity ( known-groups validity ) of the added items on game consumption in the NoS-84-FFQ questionnaire . We compared reindeer herders ( N = 94 ) versus non-reindeer herders ( N = 505 ) . We observed highly significant , large effect sizes in men ( ES = 1 . 25 , p = 9 . 7×10−04 ) and women ( ES = 1 . 15 , p = 2 . 9×10−05 ) in the expected direction corresponding with an approximately three times higher consumption of absolute overall game intake in reindeer herders compared to others . A similar approach was used for the measurement and analysis of dietary data collected with a food frequency questionnaire in the Scottish cohort ( Table S2 ) . In the NSPHS cohort , we used two self-report scales to measure overall physical activity at work and at leisure . The Work Activity Scale ( WAS , 6 items ) addresses typical occupational physical activities: sitting , standing , walking , lifting , and general indicators of physical activity , i . e . sweating and tiredness after work . The Leisure Activity Scale ( LAS , 4 items ) asks for various typical freetime activities such walking , cycling , other sporting activities , and sweating as a general indicator of physical activity . Participants reported the frequency of each activity on a 5-point rating scale ( 1 = “never” , 2 = “seldom” , 3 = “sometimes” , 4 = “often” , and 5 = “always” ) . Both scales showed satisfying internal consistency with Cronbach's α ( WAS ) = 0 . 73 and Cronbach's α ( LAS ) = 0 . 70 . A similar approach was used for the measurement and analysis of data on physical activity collected with a self-report questionnaire in the Scottish cohort ( Table S2 ) .
|
In this article we report a genome-wide association study on cholesterol levels in the human blood . We used a Swedish cohort to select genetic polymorphisms that showed the strongest association with cholesterol levels adjusted for diet and physical activity . We replicated several genetic loci in other European cohorts . This approach extends present genome-wide association studies on lipid levels , which did not take these lifestyle factors into account , to improve statistical results and discover novel genes . In our analysis , we could identify two genetic loci in the SLC2A2 ( Glucose transporter type 2 ) and the HP ( Haptoglobin-related protein precursor ) gene whose effects on total cholesterol have not been reported yet . The results show that inclusion of important environmental factors in the analysis model can reveal new insights into genetic determinants of clinical parameters relevant for metabolic and cardiovascular disease .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genetics",
"and",
"genomics/gene",
"discovery",
"nutrition",
"genetics",
"and",
"genomics/complex",
"traits",
"public",
"health",
"and",
"epidemiology/epidemiology",
"public",
"health",
"and",
"epidemiology/social",
"and",
"behavioral",
"determinants",
"of",
"health",
"genetics",
"and",
"genomics/medical",
"genetics"
] |
2010
|
Modeling of Environmental Effects in Genome-Wide Association Studies Identifies SLC2A2 and HP as Novel Loci Influencing Serum Cholesterol Levels
|
Apoptosis is essential to prevent oncogenic transformation by triggering self-destruction of harmful cells , including those unable to differentiate . However , the mechanisms linking impaired cell differentiation and apoptosis during development and disease are not well understood . Here we report that the Drosophila transcription factor Cut coordinately controls differentiation and repression of apoptosis via direct regulation of the pro-apoptotic gene reaper . We also demonstrate that this regulatory circuit acts in diverse cell lineages to remove uncommitted precursor cells in status nascendi and thereby interferes with their potential to develop into cancer cells . Consistent with the role of Cut homologues in controlling cell death in vertebrates , we find repression of apoptosis regulators by Cux1 in human cancer cells . Finally , we present evidence that suggests that other lineage-restricted specification factors employ a similar mechanism to put the brakes on the oncogenic process .
It has been a long-standing paradigm that impaired cell fate commitment is a key initiator of cancer development [1] , [2] , since cancer cells display reduced differentiation properties compared to normal cells , while tumor formation can be suppressed by inducing the terminal cell fate in cancer cells [3] . The molecular basis of the interplay between cell differentiation and cancer has only recently been established . Bossuyt and colleagues ( 2009 ) demonstrated that loss of the proneural transcription factor Atonal not only leads to a loss of differentiated eye tissue but also promotes tumor formation and progression in this tissue context [4] . Thus , their work provided evidence that the maintenance of a differentiated state , which is critically controlled by a cell-type specification factor , is one crucial aspect to prevent the oncogenic process , whereas loss of this master regulator , together with other mutations creating a sensitized background , leads to the initiation of tumorigenesis . In order to evade tumor development , organisms have evolved potent mechanisms to protect themselves from the effects of mutations in their soma [5] . Programmed cell death , or apoptosis , plays a crucial role in removing abnormal cells , which could develop into tumors . This is supported by the observation that most types of cancers are associated with genetic alterations that deactivate this rescue pathway , most commonly via up-regulation of anti-apoptotic genes [6] . Since loss of terminal differentiation and the inability to activate apoptosis are crucial steps in cancer development , the existence of regulatory mechanisms preventing the accumulation of cells harboring mutations in both pathways seems essential for the survival of multi-cellular organisms . Consistently , mutations in differentiation genes very often result in the activation of the programmed cell death machinery [7] , [8] . However , the mechanisms linking loss of differentiation and induction of apoptosis , which is crucial for the prevention of tumor formation , are still missing . Here we have used the Drosophila posterior spiracle ( PS ) as a model to analyze the interplay of differentiation and apoptosis at the mechanistic level . By studying the morphogenesis of this organ , we have identified a hard-wired program through which the cell-type specifying transcription factor Cut ( Ct ) controls in a subset of PS cells , the filzkörper cells , initiation of differentiation and simultaneous repression of apoptosis via the direct transcriptional regulation of the pro-apoptotic gene rpr . Using two well-established Drosophila in vivo eye cancer models , we demonstrate that this regulatory circuit instructed by the transcription factor Ct is a very potent mechanism to prevent and/or reduce tumor growth , as it allows the lineage-specific removal of abnormal cells at the time of their genesis . Moreover , our data show that a related regulatory wiring is used in vertebrates and that other cell-type specification factors might employ a similar mechanism for tumor suppression , thus suggesting that the coupling of differentiation and apoptosis by individual transcription factors is a widely used and evolutionary conserved cancer prevention module , which is hard-wired into the developmental program .
The PS connects the Drosophila respiratory system to the environment and consists of an internal tube , the spiracular chamber with a refractile filter , the filzkörper , which is specified by the transcription factor Ct , and an external protrusion in which the spiracular chamber is located , the stigmatophore , which is under the control of the transcription factor Spalt ( Sal ) ( Figure 1A; Figure S1A–S1D ) [9] . In 1st instar ct mutant larvae filzkörper cells are not detectable ( Figure 2A , 2D; Figure S5A , S5B ) , which a priori suggests that Ct is primarily required for the specification of the filzkörper cell fate . However , due to the fact that Ct has also been shown to regulate programmed cell death [7] , we assumed that filzkörper cells in ct mutant embryos could be completely missing due to the induction of apoptosis . To test this hypothesis , we analyzed the expression of all Drosophila pro-apoptotic genes , which revealed the specific repression of reaper ( rpr ) ( Figure 1B , 1C; Figure S1E , S1F; Figure S2A , S2B ) but not of head involution defective ( hid ) , grim and sickle ( skl ) ( Figure S1I–S1N ) transcription by Ct in embryonic filzkörper precursor cells . Strikingly , we only observed rpr de-repression in Ct-positive filzkörper , but never in Ct-neighboring , Sal-positive stigmatophore precursor cells ( Figure 1B , 1C; Figure S1E , S1F ) , evidencing the cell-autonomous regulation of rpr by Ct . Rpr binds to Inhibitor of Apoptosis Protein ( IAP ) , thereby releasing inhibition of caspases and promoting apoptosis [10] . Consistently we could demonstrate enhanced cell death in ct deficient filzkörper precursor cells of stage 11 embryos using the genetically-encoded caspase reporter Apoliner [11] as well as TUNEL and Acridine Orange ( AO ) stainings ( Figure 1D , 1E , 1F , 1G; Figure S2G , S2H ) . Thus , rpr de-repression is followed by apoptosis induction in ct mutant embryos . To study the interplay between cell-type specification and cell death at the mechanistic level , we identified conserved Ct-dependent regulatory regions in the rpr intergenic regions using computational methods . Due to the principal requirement of the Hox transcription factor Abdominal-B ( Abd-B ) for PS development [9] , we searched for clusters of binding sites for Abd-B and Ct and found a highly conserved 571 bp DNA element close to the rpr coding region , which we termed rpr-HRE-571 ( Figure S4 ) . Sequence-specific interaction of recombinant Ct protein with part of the enhancer module , the S2 sub-fragment , was detected by electrophoretic mobility shift assays ( Figure 1K; Figure S1G ) . Immunostainings revealed rpr-HRE-571-GFP activity solely in stigmatophore ( Figure 1I , 1L; Figure S3A , S3E , S3I , S3M ) but not in Ct-positive filzkörper precursor cells ( Figure 1L; Figure S3A , S3E , S3I , S3M ) , that do not express rpr ( Figure 1B , 1H; Figure S1E ) . To validate the in vivo interaction of Ct with the identified enhancer module , we interfered with Ct-enhancer interaction in two ways: we mutated Ct binding sites within the rpr-HRE-571-S2 fragment ( Figure 1J ) , a truncated module with identical activity as the rpr-HRE-571 enhancer ( Figure 1M; Figure S3B , S3F , S3J , S3N ) , and eliminated all three sites in a small deletion version of the rpr-HRE-571 enhancer , termed rpr-HRE-571-S1 ( Figure 1J ) . In both cases , GFP expression was ectopically activated in Ct-positive filzkörper precursor cells ( Figure 1N , 1O; Figure S3C , S3D , S3G , S3H , S3K , S3L , S3O , S3P ) . These experiments demonstrated that Ct directly represses rpr transcription and thus apoptosis in the filzkörper precursor cells of the PS in a cell-autonomous manner by interacting with a small enhancer module located in the rpr intergenic region . Since our result showed that filzkörper cells are very efficiently eliminated by apoptosis in the absence of Ct function , we next asked whether Ct primarily acts as a repressor of programmed cell death or whether this factor is also required for the differentiation of filzkörper cells . To this end , we analyzed ct deficient cells , which were kept alive by expressing the caspase inhibitor p35 [12] in ct mutant embryos using the PS-specific driver ems-GAL4 [13] . In order to follow the cells normally under the control of Ct , these cells were GFP-labeled using the same driver , which is active only in a subset of Ct-expressing cells ( Figure 2G , 2K ) . Our experiments revealed that Ct- and GFP-positive filzkörper cells found in the wild-type situation ( Figure 2A , 2F , 2F′ , 2B , 2G , 2G′ , 2K; Figure S5A , S5D , S5G ) are eliminated in ct mutant embryos ( Figure 2D , 2I , 2I′ , 2L; Figure S5B , S5E , S5H ) , whereas they remained viable when apoptosis is blocked ( ct−; ems::p35 ) ( Figure 2E , 2J , 2J ) . However , these ct deficient , undead cells had developmental defects , as they did not properly invaginate and did not acquire their terminal cell fate as indicated by reduced expression of the apical cell polarity marker Crumbs ( Crb ) and the cell adhesion molecule DE-Cadherin ( Figure 2F′ , 2J′; Figure S5D , S5F , S5G , S5I ) . Consistently , these cells never adopted a filzkörper cell fate ( Figure 2E; Figure S5C ) . These defects were a consequence of blocking cell death in ct deficient , undifferentiated cells and were not due to a general response to the apoptosis inhibitor p35 , as the filzkörper of ems::p35 control embryos ( Figure 2B , 2G , 2G′ ) was indistinguishable form those of wild-type embryos ( Figure 2A , 2F , 2F′ ) . Local activation of apoptosis was sufficient to induce cell death in filzkörper cells , as expression of a rpr transgene resulted in their elimination ( Figure 2C , 2H , 2H′ ) . Taken together , our results revealed that Ct carries out two functions during PS morphogenesis: it allows the survival of uncommitted precursor cells by the transcriptional repression of the pro-apoptotic gene rpr and subsequently it drives these cells into a filzkörper-specific cell fate . Ct is expressed in many different cell and tissue types [14] , thus we tested the Ct switch function in diverse developmental contexts . Ct activity was eliminated in the Drosophila eye using RNAi ( Figure 3C , 3G ) , resulting in an overall reduction of the eye size ( Figure 4A , 4B ) , and a loss of interommatidial bristles ( Figure 3A , 3E ) , which normally express Ct ( Figure 3C ) . Consistently , expression of the bristle shaft progenitor marker DE-Cadherin [15] was lost in GMR::ctRNAi pupal retinas ( Figure 3B , 3F ) . Expression of the apoptotic executor activated Caspase-3 was significantly increased in eye discs of ey::ctRNAi 3rd instar larvae ( Figure 3D , 3H ) , and a significant induction of rpr RNA levels was observed using quantitative Real Time-PCR ( qRT-PCR ) ( Figure 3K ) . Co-expression of either the apoptosis inhibitor p35 , which rescues the eye size ( Figure 4C ) , or of a rprRNAi construct along with the ctRNAi transgene resulted in a survival of ct deficient cells , as evidenced by the expression of the bristle shaft progenitor maker DE-Cadherin ( Figure 3J ) . However , reminiscent to the phenotypes in the PS ( Figure 2E , 2J′ ) , these cells were unable to adopt their terminal fate due to the absence of the cell-specification factor Ct , and consequently fully differentiated interommatidial bristles were absent ( Figure 3I ) . Similar results were obtained in other cell types specified by Ct ( Figure S6 ) , suggesting that the Ct-dependent switch between cell-type specification and programmed cell death is of general relevance . By analyzing the Ct-rpr interaction in two well-established in vivo Drosophila cancer models , we asked whether the combined transcriptional regulation of differentiation and apoptosis repression by Ct could represent a cancer prevention mechanism . In the oncogenic “eyeful” model [16] , eye tumors occurred in 72 . 5% of control flies , with 4 . 9% of them showing macroscopically visible secondary tumor growths derived from the developing retina ( Figure 4H , 4N ) due to the eye-specific over-expression of the Notch ligand Delta ( Dl ) and the two epigenetic regulators longitudinals lacking ( lola ) and pipsqueak ( psq ) [16] . In contrast , pre-oncogenic ey::Dl flies over-expressing Dl exclusively in eye tissue [16] never displayed any eye tumors or invasive tumors but only mildly overgrown eyes ( Figure 4D , 4N ) . Eye-specific inhibition of Ct activity alone only caused a small increase in primary and secondary tumor incidences in both sensitized backgrounds ( Figure 4E , 4I , 4N ) , however , these numbers were dramatically increased when Ct function and the ability to activate apoptosis were simultaneously inhibited ( Figure 4F , 4G , 4K , 4L , 4M , 4N ) . Consistently , increased numbers of apoptotic cells were found in tumorous tissue with reduced Ct levels ( ey::Dl;2xctRNAi ) ( Figure 4P , 4Q ) , demonstrating that the coupled regulation of differentiation and apoptosis by a single transcription factor is an important mechanism to suppress cancer . However , despite increased apoptosis activation in ey::Dl;2xctRNAi eye imaginal discs ( Figure 4P ) , which should result in a reduction of tumor growth , tumor formation in these animals was increased ( Figure 4N , 4Q ) . Using the proliferation marker Phosphorylated histone H3 ( PH3 ) , we could demonstrate that the tumor growth induced by differentiation loss is due to excessive cell proliferation ( Figure 4P ) , which is in line with previous results [4] . What is the molecular basis for this phenotype ? RT-PCR analysis of candidate genes involved in cell cycle and growth control using ey::Dl and ey::Dl;ctRNAi eye imaginal discs revealed a strong induction of phosphoinositide 3-kinase ( PI3K ) upon Ct depletion ( Figure 4O ) . It has been shown before that PI3K overexpression in the ey::Dl pre-oncogenic background leads to tumor formation [17] and that PI3K is a limiting factor for RasV12 DlgRNAi induced tumor growth [18] . Thus we tested its contribution to tumor formation in Ct-induced oncogenic eyes by reducing its level in ey::Dl;2xctRNAi animals . Interestingly , we not only found a rescue of the tumorous eye growth , but also a dramatic increase in the occurrence of smaller eyes in ey::Dl;2xctRNAi;PI3KRNAi animals ( Figure 4Q ) , which is similar to the apoptosis-induced “small eye” phenotype observed upon Ct depletion in the wild-type background ( Figure 4B ) . Taken together , these results show that the Ct-dependent tumor growth is in part mediated by the up-regulation of the PI3K signaling pathway and that this pro-tumorigenic effect counteracts the anti-tumorigenic apoptosis effect of Ct . We found cell clusters expressing the eye differentiation marker ELAV at abnormal , ectopic positions in undifferentiated tissue of 3rd instar eye-antennal discs ( Figure S7 ) , and it had been shown before that changes in the adhesive properties of cells are critical in inducing migratory behavior [19] , [20] . Consistently , transcriptome profiling experiments revealed a reduction in the expression of cell adhesion genes in eye-imaginal discs of Ct depleted animals exhibiting primary and secondary tumor formation ( ey::Dl;2xctRNAi ) in comparison to control animals ( ey::Dl ) ( Figure 5A ) . To test the significance of this finding , we interfered with the function of α-PS4 integrin , one of the genes identified as Ct responsive ( Figure 5A ) , by reducing its expression and the expression of its heterodimeric interaction partner β-PS integrin ( mys ) [21] in the ey::Dl pre-oncogenic background . We observed an increase in primary and secondary tumor formation in both situations , while reducing the activity of a related but Ct-independent integrin , the α-PS2 integrin ( if ) , did not have any effect ( Figure 5B ) . Since decreasing the activity of another Ct responsive cell adhesion gene , namely Tissue inhibitor of metalloproteases ( Timp ) , also induced an increase in secondary tumors ( Figure 5B ) , we asked if restoration of cell adhesion would be able to rescue this phenotype in the Ct loss-of-function setting . To this end , we expressed one of the major adhesion genes regulated by Ct , DE-Cad ( Figure 5C ) , in eye cells of eyeful+ctRNAi;p35 animals , which display high rates of invasive tumors ( Figure 4N , Figure 5D ) , and observed a reduction of secondary tumor growth rate by more than 50% ( Figure 5D ) . These results demonstrate that regulation of cell adhesiveness is one of the essential Ct-dependent mechanisms to suppress tumor spread . In vertebrates , invasive tumor growth requires the detachment of abnormal cells from tumor tissue and their circulation in the bloodstream [22] . To test if secondary tumor formation mediated by loss of Ct function is dependent on a similar mechanism , we analyzed the hemolymph , the insect “blood” , in our fly lines . Strikingly , we detected a significant increase in GFP-labeled eye-imaginal disc cells in the hemolymph of animals forming invasive tumors ( eyeful+GFP;ctRNAi;p35 ) in comparison to control animals ( ey::GFP ) ( Figure 5E , 5F ) , suggesting that tumor cells in flies indeed circulate through the bloodstream and invade ectopic locations . In sum , these results demonstrate that transcriptional coupling of differentiation and apoptosis is a cell-intrinsic mechanism to ensure normal development and to prevent tumor initiation , progression and invasion , which is at least in part achieved by fine-tuning the adhesive properties of cells required for tissue integrity . We next explored whether the effective regulation of programmed cell death by Ct has been conserved during evolution . The vertebrate homologue of Cut , Cux1 , has a well-documented function in cell differentiation during normal development as well as in tumor initiation and progression in specific cancer types [23] . In addition , several studies show that Cux1 represses apoptosis during normal vertebrate development [24] , [25] , [26] , and just recently it has been demonstrated that Cux1 knock-down leads to activated apoptosis and to reduced growth of xenograft tumors in vivo [25] , [27] . To further investigate the mechanistic basis of Cux1 function in mediating apoptosis repression in vertebrates , we suppressed Cux1 in Panc1 pancreatic cancer cells ( Figure 6A ) and determined the transcriptional response of human apoptosis genes . Strikingly , mRNA levels of the pro-apoptotic gene puma were consistently elevated , whereas the anti-apoptotic gene Bcl-2 was down-regulated upon Cux1 depletion ( Figure 6A ) . Since BH3-only proteins , like Puma , and Bcl-2 are important for the release of the vertebrate functional equivalent of Rpr , Smac/DIABLO , from mitochondria [28] and since Cux1 binding sites are present close to the puma coding region ( Table S1 ) , these results suggest that the regulatory wiring of differentiation and apoptosis , at the level of Cut , is functionally conserved in vertebrates . Does this regulatory layout represent a general mechanism employed by other differentiation factors ? This would require a whole suite of cell-type specifying transcription factors to repress cell death genes by interacting with distinct enhancer modules located in their regulatory regions . In addition , these modules should follow a similar functional logic to the rpr-HRE-571 enhancer , in that cell-type specific gene activation is counteracted by strong repressing inputs from linked cis-elements ( Figure 1L–1O ) . In line with this , we found that a different conserved enhancer module on the Drosophila rpr regulatory region ( rpr-HRE-707 ) drove expression in CNS midline cells of stage 14 embryos ( Figure 6F ) , which never express rpr at this and subsequent developmental stages ( Figure 6L ) [29] . However , extending the enhancer to include additional cis-elements ( rpr-HRE-707+156 ) ( Figure 6D ) resulted in loss of enhancer activity ( Figure 6H ) . Using the JASPAR database [30] , we found consensus binding sequences for POU-domain containing transcription factors on the extended enhancer module , and one of these factors , Ventral veins lacking ( Vvl ) , is known to function in midline glial cells and to repress apoptosis [31] , [32] . Our analysis revealed a partial overlap of Vvl and reporter gene expression in rpr-HRE-707 embryos ( Figure 6J ) , and consistently ectopic rpr transcripts were detected in several midline cells of vvl mutants ( Figure 6L , 6M ) . Due to the existence of GFP-positive cells not expressing Vvl ( Figure 6J ) , we assume that not only Vvl but also other POU transcription factors interact with the rpr-HRE-707+156 enhancer to repress rpr transcription in midline cells . Revisiting the rpr-HRE-571 enhancer module revealed that extension of the enhancer also led to a complete loss of reporter activity ( compare Figure 1L , Figure 6E , 6G ) . Thus , complete repression of rpr transcription in the PS requires two inhibitory inputs: one active in filzkörper cells , which we had identified to be mediated by Ct , and one so-far unknown repressor functional in stigmatophore cells . Importantly , the functional analogy of Vvl and Ct also extended to the tumor suppression activity , since , like in the case of Ct ( Figure 4N ) , primary and secondary tumor frequencies were increased when the ability to activate apoptosis and Vvl function was impaired at the same time ( Figure 6K ) . Furthermore , we identified two unrelated cell-type specifying transcription factors in addition to Ct and Vvl , which showed similar behavior with regards to tumor suppression ( Figure S8 ) . Together with the fact that the regulatory sequences flanking the Drosophila rpr coding region show significantly less sequence divergence than expected and a high occurrence of conserved transcription factor binding motifs ( Figure 6B , 6C ) , these findings lead us to propose that coupling of differentiation and cell death repression via a single transcription factor represents a general cancer prevention mechanism ( Figure 7 ) , which could be employed by a large number of developmental regulators in diverse organisms .
Programmed cell death is an integral aspect of animal development [33] . Genetic studies in C . elegans , Drosophila and mouse have shown that apoptosis is used to sculpt tissues and to remove excessive and unwanted cells , thus defining the morphology required for diverse physiological functions [34] . In this context , apoptosis is usually regulated by cell signaling pathways [33] , [35] , [36] . In addition to its role in tissue morphogenesis , apoptosis is also required to eliminate potentially deleterious cells , which in most cases involves complex multi-step control mechanisms [33] , [37] . One such situation generating harmful cells is the inability to differentiate or adopt the appropriate cell fate , which very often results in uncontrolled cell proliferation and cancer development , and thus requires the immediate killing of these cells . However , even though it is established that apoptosis is a protective mechanism against tumorigenesis in cases of aberrant cell differentiation [1] , [34] , [38] , the interplay of the two processes at the mechanistic level has remained unclear . In our study , we show that the simultaneous and antagonistic regulation of differentiation and apoptosis is a hard-wired developmental program and carried out by individual transcription factors , such as Cut . Our results demonstrate that impairment of differentiation in the cell lineage specified by Cut instantaneously triggers locally restricted apoptosis by releasing transcriptional repression of the pro-apoptotic gene rpr in these cells . Due to its immediate effect , the coupling of differentiation and apoptosis on the transcriptional level represents one of the fastest and most direct mechanisms to eliminate abnormal cells in status nascendi and thereby immediately interferes with their potential to develop into harmful cells . Interestingly , apoptosis induction as a consequence of aberrant cell-type specification is not only mediated by the cell death promoting gene rpr but also by hid [8] . However , despite the same trigger , which is the inability to properly differentiate , the transcriptional basis for inducing the expression of one of these two apoptosis genes seems to be quite different: in Drosophila early developmental mutants only the expression of the pro-apoptotic gene hid is up-regulated [8] , whereas our study shows that exclusively the transcription of rpr is induced when a factor specifying a distinct cell type is lost . Although it is currently unknown how hid expression is regulated at the transcriptional level , this raises the possibility that the apoptosis gene hid acts a safeguard when broad positional information at the onset of embryogenesis is absent , whereas rpr might take over this function later in development when individual and specific cell types are defined by transcription factors restricting cell fate choices . Given the well-known role of the vertebrate homologue of Cut , Cux1 , in tumor initiation and progression in specific cancer types [23] , we addressed whether the switch function of the cell specification factor Cut is also relevant in a pathological context . We found that simultaneous inhibition of Cut function and apoptosis within a sensitized background increases tumor formation and metastasis to secondary sites in the animal . In contrast , down-regulation of Cut and inhibition of apoptosis in a normal developmental context , such as in the Drosophila PS or the developing eye , only results in the survival of the Cut deprived cells , but not in tumor development . These results demonstrate that cells , which are unable to undergo the cell lineage-specific differentiation program , have to be eliminated , since they have the potential to develop into cancerous cells when other genetic or micro-environmental changes accumulate [19] , [27] , [39] . But why do differentiation-deprived cells form tumors in a cancer-prone tissue environment despite the ability to activate the apoptotic rescue pathway ? This is due to the fact that the transcription factor Cut , as part of its selector gene function , coordinately regulates multiple cellular processes , including differentiation , apoptosis , cell adhesion , but also proliferation , which are all required for proper cell fate specification and the maintenance of a differentiated state ( thereby preventing tumor formation ) . If , however , Cut activity is abolished , all its downstream functions are affected , leading not only to the activation of apoptosis , but also to reduced differentiation and adhesion properties and the activation of cell proliferation , which is , in the case of Cut , mediated ( at least in part ) by the PI3K signaling pathway . Thus , loss of Cut function stimulates tumor growth in a sensitized background , since the pro-tumorigenic effects of deregulated proliferation and cell adhesiveness out-compete the anti-tumorigenic apoptosis effects at work . However , when the anti-tumorigenic effect is eliminated in the differentiation-compromised cancer tissue , tumorigenesis is strongly enhanced , which resembles a prevalent situation in aggressive human cancers characterized by the loss of differentiation , the resistance to apoptosis activation and the mis-regulation of adhesion properties [1] , [40] , [41] . Several lines of evidence suggest that the dual role of Cut in differentiation and apoptosis for cancer prevention is conserved in evolution . First of all , the two vertebrate homologues of Cut , Cux1 and Cux2 , code for homeobox-containing transcription factors , which are crucially involved in cell-type specific terminal differentiation [14] , [23] , [42] . Both , Cux1 and Cux2 , have similar binding specificities to Drosophila Cut [43] , they also operate as transcriptional repressors and activators of genes in multi-lineage differentiation pathways [26] and , like Drosophila Cut , they act as downstream effectors of the Notch signaling pathway [44] , [45] . In addition to their well-established role in development and differentiation , there are also several examples linking the vertebrate Cut homologue Cux1 to apoptosis and cancer . First , inhibition or partial disruption of Cux1 function in mice leads to increased apoptosis rates in vivo [24] , [26] . Second , Cux1 regulates normal hematopoiesis , in part by modulating the levels of survival and apoptosis factors [26] . Third , Cux1 plays a prominent role in cancer progression [23] . And fourth , induced down-regulation of Cux1 in subcutaneous xenograft tumors leads to activation of apoptosis and to reduced tumor growth [25] . Our results now show that the Cut-Rpr regulatory wiring of apoptosis and differentiation is conserved in vertebrates . In mammalian cells , the Rpr functional homologue , Smac/DIABLO , which is normally compartmentalized within mitochondria , has to be released to execute its pro-apoptotic function by binding to and inactivating Inhibitors of Apoptosis ( IAPs ) [46] . This process requires the permeabilization of the outer mitochondrial membrane ( MOMP ) , which is achieved by the interaction of pro-apoptotic proteins like Puma with anti-apoptotic proteins like Bcl-2 , which normally inhibit MOMP [47] . We now show that down-regulation of Cux1 in pancreatic cancer cell lines leads specifically to the transcriptional induction of the pro-apoptotic gene puma and the down-regulation of the anti-apoptotic gene Bcl-2 . Thus , two crucial regulators for Smac/DIABLO release are controlled by Cux1 on the transcriptional level , showing that the basic design principle of the Cut-Rpr regulatory wiring is conserved but has been adapted to the system requirements in evolution . In future , it will be intriguing to study this mechanism in diverse cellular backgrounds , including stem cells , which neither die nor differentiate .
To identify Abd-B binding sites we used the method of Wasserman and Sandelin ( 2004 ) [48] with the Abd-B Position Frequency Matrix [49] ( http://jaspar . genereg . net/ ) and 90% cut-off . Abd-B binding site clusters were identified if at least three Abd-B sites were present in a 400 bp window . Conserved enhancers were identified using PhastCon score [50] . Within conserved regions , Ct binding sites were identified using published sequence data [49] . Vvl and Cux1 binding sites were identified using the Vvl and Cux1 Position Frequency Matrices available at the JASPAR database [49] ( http://jaspar . genereg . net/ ) . Drosophila melanogaster strain Oregon R was used as wild type . Amorphic allele ctdb7/FM7 [51] , ems-Gal4 and ems-Gal4 , UAS-GFP/TM6B [13] were obtained from J . Castelli-Gair Hombria , UAS-Dcr2; ey-Gal4 from B . Dickson , eq-Gal4/TM6B [52] from H . Pi , UAS-Apoliner5 [11] from J . P . Vincent , UAS-ctEHK2/CyO [53] , UAS-ctRNAi; UAS-ctRNAi ( Grueber and Jan , unpublished ) from Y . N . Jan and ey-Gal4 , UAS-Dl/CyO and eyeful flies ( ey-Gal4 , GS88A8 , UAS-Dl/CyO ) from M . Domiguez [16] . UAS-Dcr2; C96-Gal4 ( BL-25757 ) , UAS-CD8::GFP ( BL-5130 ) from Bloomington stock center . GMR-Gal4 , UAS-p35 , UAS-Abd-B , arm-Gal4 , UAS-rpr , UAS-lacZ were described elsewhere [54] , [55] , [56] , [57] . Other UAS-RNAi lines were obtained either from BDSC , VDRC or TRiP: DE-Cad ( v8024 ) , rpr ( v12045 ) , vvl ( JF02126 ) , gro ( v6316 ) , H ( v24466 ) , αPS2 ( if ) ( BL27544 ) , αPS4 ( v109783 ) , βPS ( mys ) ( HMS00043 ) , PI3K ( v107390 ) and Timp ( v109427 ) . Five independent transgenic lines were analyzed for each reporter construct . Panc-1 human pancreatic cancer cells ( Department of Surgery , Medical Faculty , University of Heidelberg ) and 293 T cells were maintained in DMEM supplemented with 10% fetal bovine serum , 2 mM L-glutamine , non-essential amino acids , 100 U/ml penicillin , 100 U/ml streptomycin , and 0 . 25 µg/ml amphotericin B . shRNA directed against human Cux1 was generated using the following complementary oligonucleotides ( forward and reverse ) : Cux1_KDa: 5′CCGGAAGAAGAACACTCCAGAGGATCTCGAGATCCTCTGGAGTGTTCTTCTTTTTTTG3′ and 5′AATTCAAAAAAAGAAGAACACTCCAGAGGATCTCGAGATCCTCTGGAGTGTTCTTCTT3′; Cux1_KDb: 5′CCGGAAGAATCTTCTCGTTTGAAACCTCGAGGTTTCAAACGAGAAGATTCTTTTTTTG3′ and 5′AATTCAAAAAAAGAATCTTCTCGTTTGAAACCTCGAGGTTTCAAACGAGAAGATTCTT3′; shRNA control ( C ) , 5′CCGGAATTGCCAGCTGGTTCCATCACTCGAGTGATGGAACCAGCTGGCAATTTTTTTG3′ and 5′CCGGAATTGCCAGCTGGTTCCATCACTCGAGTGATGGAACCAGCTGGCAATTTTTTTG3′ . pLKO lentiviral vectors containing shRNA were transfected into 293 T cells together with psPAX2 ( packaging vector ) and pMD2 . G ( VSV-G envelope protein expression vector ) using the calcium-phosphate transfection kit ( Sigma ) . Panc-1 cells were infected using lentivirus-containing 293 T cell supernatant and Cux1 protein levels were assessed by Western blotting using anti-Cux1 ( Sigma-Aldrich ) and anti-β-Actin ( GeneTex ) antibodies . All enhancer fragments were PCR amplified from genomic DNA and cloned in pHPelican-GFP [58] or pHPelican-GFP_DEST [59] . For binding site mutations , a two-step overlap PCR was performed . For mutating Ct binding sites within the rpr-HRE-571-S2 enhancer fragment , mutation introduced into the Cut consensus sequences were identical to the ones introduced into the EMSA probes ( see below ) . Primer sequences are available upon request . Real Time PCR was performed following standard protocols using SYBR green . Expression was normalized to GAPDH for mammalian cells and to endogenous actin5C mRNA for imaginal disc analysis . Relative expression levels are based on three biological replicates . Drosophila embryos were fixed as described [56] . Eye discs or wing discs were dissected in PBS and fixed in 4% paraformaldehyde/PBS for 10 min for immunostaining . In situ hybridization and immunochemistry were performed as described [56] . Fluorescent mRNA/protein double labeling and fluorescent duplex in situ hybridizations were done as described previously [60] . Primary Antibodies used were: mouse anti-Ct 2B10 ( 1∶200 , DSHB ) , mouse anti-Crb cq4 ( 1∶200 , DSHB ) , rat anti-DE-Cad DCAD2 ( 1∶100 , DSHB ) , mouse anti-ELAV ( 1∶200 , DSHB ) , rat anti-Sal ( 1∶800 kind gift from R . Barrio ) , rabbit anti-PH3 ( 1∶200 , Santa Cruz ) , rabbit anti-Cleaved Caspase-3 ( 1∶50 Cell Signalling ) , mouse anti-GFP ( 1∶1000 , Roche ) , rabbit anti-GFP ( 1∶2000 , Sigma ) , anti-DIG POD ( 1∶200 , Roche ) , Streptavidin HRP ( 1∶200 , PerkinElmer ) . Scanning electron microscopy , Acridine Orange staining and cuticle preparation were carried out as described in Lohmann et al . ( 2002 ) [55] and Zhai et al . ( 2010 ) [61] . TUNEL assay was performed with the In Situ Cell Death Detection Kit ( TMR ) from Roche according to the manufacturer's instruction . Cut CR3HD ( 4849–5412 of ctRA from Y . N . Jan ) was cloned into pMAL2-c2x vector ( NEB ) and expressed as Maltose Binding Protein ( MBP ) fusion proteins . EMSAs were carried out as described in Stöbe et al . ( 2009 ) [56] . The following oligonucleotides ( S2 subfragment ) were used for analyzing the Cut binding sequence in EMSA ( only forward strand is shown ) : Wild-type . 5′GCACTTTTGCCTGCAGTTCAACTCGGTTCAGTTCGGTTGTGTCATAAAAAATC3′ Mutated . 5′GCACTTTTGCCTGCAGTGGAACTCGGTGGAGTGGGGTTGTGTCATAAAAAATC3′ Cut consensus sequences are underlined , exchanged nucleotides in the mutated versus the wild-type sequence are shown in bold . ey::CD8-GFP or eyeful+CD8-GFP;ctRNAi;p35 3rd instar larvae were dissected by rupturing the larval cuticle at the posterior end with a pair of fine forceps , the hemolymph was collected in ice-cold Schneider's medium ( Invitrogen GIBCO ) containing 1× Complete protease inhibitor mixture ( Roche ) . Hemolymph cells were analyzed via FACSAria to quantify GFP-labeled cells circulating within the hemolymph . In addition , GFP mRNA levels within the hemolymph were measured by qRT-PCR . Eye-antennal discs of ey::Dl or ey::Dl::2xctRNAi 3rd instar larvae were dissected in cold PBS , total RNA was extracted using standard procedures . Microarray analysis was conducted at the Genomics Core Facility , EMBL , Heidelberg , Germany . Microarray data analysis was performed using the R package as described previously [54] .
|
Apoptosis is a highly conserved cellular function to remove excessive or unstable cells in diverse developmental processes and disease-responses . An important example is the elimination of cells unable to differentiate , which have the potential to generate tumors . Despite the significance of this process , the mechanisms coupling loss of differentiation and apoptosis have remained elusive . Using cell-type specification in Drosophila as a model , we now identify a conserved regulatory logic that underlies cell-type specific removal of uncommitted cells by apoptosis . We find that the transcription factor Cut activates differentiation , while it simultaneously represses cell death via the direct regulation of a pro-apoptotic gene . We show that this regulatory interaction occurs in many diverse cell types and is essential for normal development . Using in vivo Drosophila cancer models , we demonstrate that apoptosis activation in differentiation-compromised cells is an immediate-early cancer prevention mechanism . Importantly , we show that this type of regulatory wiring is also found in vertebrates and that other cell-type specification factors might employ a similar mechanism for tumor suppression . Thus , our findings suggest that the coupling of differentiation and apoptosis by individual transcription factors is a widely used and evolutionarily conserved cancer prevention module , which is hard-wired into the developmental program .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"molecular",
"cell",
"biology",
"genetics",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
Antagonistic Regulation of Apoptosis and Differentiation by the Cut Transcription Factor Represents a Tumor-Suppressing Mechanism in Drosophila
|
The Piwi pathway is deeply conserved amongst animals because one of its essential functions is to repress transposons . However , many Piwi-interacting RNAs ( piRNAs ) do not base-pair to transposons and remain mysterious in their targeting function . The sheer number of piRNA cluster ( piC ) loci in animal genomes and infrequent piRNA sequence conservation also present challenges in determining which piC loci are most important for development . To address this question , we determined the piRNA expression patterns of piC loci across a wide phylogenetic spectrum of animals , and reveal that most genic and intergenic piC loci evolve rapidly in their capacity to generate piRNAs , regardless of known transposon silencing function . Surprisingly , we also uncovered a distinct set of piC loci with piRNA expression conserved deeply in Eutherian mammals . We name these loci Eutherian-Conserved piRNA cluster ( ECpiC ) loci . Supporting the hypothesis that conservation of piRNA expression across ~100 million years of Eutherian evolution implies function , we determined that one ECpiC locus generates abundant piRNAs antisense to the STOX1 transcript , a gene clinically associated with preeclampsia . Furthermore , we confirmed reduced piRNAs in existing mouse mutations at ECpiC-Asb1 and -Cbl , which also display spermatogenic defects . The Asb1 mutant testes with strongly reduced Asb1 piRNAs also exhibit up-regulated gene expression profiles . These data indicate ECpiC loci may be specially adapted to support Eutherian reproduction .
In animal RNA interference ( RNAi ) pathways , the Argonaute ( AGO ) and Piwi proteins are deeply conserved from humans to basal animals like cnidarians and poriferans [1 , 2] . Likewise , several AGO-bound microRNAs ( miRNAs ) are also deeply conserved across bilaterians [3] , but Piwi-interacting RNAs ( piRNAs ) differ by evolving so rapidly that there are few individual piRNAs conserved between even closely related species [4 , 5] . A rationale for the fast rate of piRNA sequence evolution is to keep pace with the rapidly evolving transposable elements ( TE ) that some piRNAs are targeting [6 , 7] . However , most mammalian piRNA cluster ( piC ) loci are depleted of TE sequences [4 , 5] , and in spite of the TE-repressing function of the Piwi pathway , animal genomes vary widely in size due mainly to TE content , from ~10% of the euchromatic genome in flies to >40% in mammals [8] . Thus , we have an incomplete understanding of what additional roles the Piwi pathway plays beyond TE repression . The piRNAs arise as clusters from two main types of loci . Intergenic piRNA cluster loci are large 10–100 kb long regions , and some of these loci function to silence TEs with TE-directed piRNAs . However , the function of most intergenic piRNA cluster loci in mammals remains mysterious because they are depleted of TE sequences and the transcripts are non-coding [9] . In contrast , genic piRNA cluster loci are derived from protein-coding transcripts that possess extensive 3'Untranslated Regions ( 3'UTRs ) which efficiently enter the piRNA biogenesis pathway [10 , 11] . Hundreds of genic piRNA cluster loci are also depleted in TE sequences , and germ cells specifically select genic transcripts via an unknown signature for piRNA biogenesis [11] . Nevertheless , the abundant piRNAs from the Drosophila genic piC traffic jam ( tj ) may regulate gene targets important for follicle cell development [12] . Although piRNAs are essential for animal fertility , we do not know which of the hundreds of piC loci are most important for animal reproduction , particularly in mammals . Therefore , we hypothesize that determining which piC loci are most conserved in piRNA expression throughout animal evolution would yield functional insight to this question above . However , the field has only touched upon the evolutionary patterns of piC loci with a handful of previous studies comparing piC loci between a limited set of animal species . In Drosophilid flies , the intergenic piC locus flamenco ( flam ) is syntenically conserved between D . melanogaster ( D . mel ) and D . erecta ( D . ere ) for ~10 million years ( MY ) of evolution [13] . In rodents and humans , which diverged in evolution ~90 MY ago , some piC loci are also syntenically conserved [4 , 5] , and Assis and Kondrashov applied a classification process to these human and rodent piRNA datasets to conclude that piC loci were rapidly expanding at a high rate , with mainly new gains of clusters between these species and no piC loci losses [14] . Assis and Kondrashov further proposed a positive selection process for piC expansion to keep pace with the rapid evolution of TE sequences [14] , however the capacity of piCs to expand via copy number evolution has recently been re-inspected in humans by Gould et al , whom suggest instead that negative selection may be acting upon human piCs to limit their copy number expansion [15] . Earlier studies only examined human , mouse and rat piRNA datasets that are now considered shallow by today’s deep-sequencing standards [4 , 5] , and thus were unable to determine whether piRNA expression from piC loci could be deeply conserved like miRNA expression . To gain a more complete picture of the evolution of piC loci expression , this requires the experimental support of new , deeper piRNA datasets from a broader spectrum of animals . Fortunately , the diversity of species with sequenced piRNA libraries have greatly expanded recently with these various studies [11 , 16–26] , some of which mainly focused on the characterizing the novel piRNA features specific to the species . Thus , we sought to discover: ( 1 ) which piC loci , if any , may exhibit deep conservation of piRNA expression across animal lineages , ( 2 ) how frequently do piC loci expression patterns vary between species , and ( 3 ) which gene regulation step ( s ) prominently influences these piC loci expression patterns . Here , we formally define conserved piC loci expression as the detection of mature piRNAs from syntenic loci between species , which is a process that depends on both: ( 1 ) the transcription of the loci , and ( 2 ) selection of the transcribed precursor RNA by Piwi-pathway proteins for processing into mature piRNAs . In this study , we describe a comprehensive comparative genomics approach focusing first on the genic piRNA biogenesis mechanism that we had previously shown was deeply conserved between flies and mice [11] . Although gene homologs for mouse and fly genic piC loci can be predicted bioinformatically ( S1A Fig ) , it is not clear if these are actually orthologous piC loci because gene families have greatly expanded in mouse during the >600 MY of evolution between these species . Thus , we decided to examine piRNA expression patterns within closely-related species amongst the Drosophilids and tetrapod lineages to determine how many piC loci expression patterns for piRNA biogenesis are deeply conserved . We confirm that genic piC loci , which are depleted in TE sequences , are rapidly evolving and expanding into repertoires that are unique to animal species . Surprisingly , we also discover a cohort of piC loci with piRNA expression patterns conserved across ~100 MY of Eutherian mammal evolution , which stand out from the generally rapid evolution of most animal piC loci expression patterns .
We developed a piC loci discovery approach on new piRNA datasets that we had generated as well as analyzing publicly-deposited datasets covering four Drosophilids , the chicken and eleven mammals within Primate , Glire , and Laurasiatherian clades ( Fig 1A and 1B ) . We sequenced Drosophilids piRNAs from ovarium samples because piRNAs are more plentiful and diverse in fly ovaries compared to fly testes [13 , 27] . The tetrapod piRNAs were sampled from adult testes , which contained piRNAs from both early and post-pachytene stages of meiotic germ cells [19] . We generated a variety of piRNA libraries from Glires , including libraries enriched in pre-pachytene piRNAs from 10 days-post-partum ( 10dpp ) testes and PIWIL2 immunoprecipitates ( IPs ) , whereas the remaining tetrapod libraries were previously published in these studies: [11 , 19–26] . This piRNA compendium was chosen for comprehensiveness and quality of piRNA coverage , since all libraries displayed the expected main read length distribution of 23–32 nucleotides and sufficient read depth ( 4–143 million reads , S1 Table ) . This phylogenetic spectrum of piRNA dataset allowed us to compare and determine the conservation of piC loci expression patterns spanning both short ( <50 MY ) and long ( >150 MY ) divergence times . To search for conserved expression of piRNAs across potential piC loci between species , we focused first on genic piC loci because the protein-coding genes are clear proxies for assigning orthology between potential piC loci . We deployed a small RNA profiling and bioinformatics procedure that identified genic piC loci by tracking piRNAs from protein-coding transcript 3'UTRs ( S1B Fig ) . This procedure was highly accurate in tracking piRNAs at specific genic piC loci , because the reads exhibited the correct length ranges of 24-28nt for Drosophilid piRNAs and 24-31nt for mammalian piRNAs , as well as a bias for uridine at the 5' nucleotide of the small RNAs ( S1C–S1F Fig ) . We examined conservation of piRNA expression from piC loci in four model Drosophilids where TE-directed piRNAs were previously characterized ( Fig 1B ) [13 , 29] . D . mel protein-coding transcript alignments served as excellent proxies for transcripts in the incomplete draft assemblies of the D . ere , D . yakuba ( D . yak ) and D . virilis ( D . vir ) genomes . To prevent intergenic piC loci from overshadowing genic piC loci , we processed the ovarium samples to enrich for follicle cells and this improved genic piC locus detection ( Fig 1C and 1D , [28] ) . We discovered >270 genic piC loci expressed in both D . mel and D . ere , 82 genic piC loci in D . vir , but surprisingly only 27 loci in D . yak ( S2A Fig and S2 Table ) . The most abundant genic piC locus in D . mel , piC-tj , was only conserved in piRNA expression in D . ere , whereas the genic piC locus at diminutive ( dm ) , the Drosophila c-Myc oncogene homolog , was more deeply conserved to D . vir in piRNA expression ( Fig 1E ) . Although we discovered plenty of piRNAs for the D . yak ortholog of the flam intergenic piC locus ( Fig 1G ) , indicating the D . yak library sufficiently contained follicle-cell specific piRNAs , genic piC loci were still depleted in D . yak . We examined gene expression profiles in D . yak follicle cells for known Piwi pathway genes and new genes associated with PIWI in D . mel OSS follicle cells as well as Northern blot comparisons ( S2C and S2D Fig ) , but these investigations were unable to explain the depletion of genic piC loci in D . yak . Therefore , we only further considered the genic piC loci expression patterns in D . mel , D . ere , and D . vir ( Fig 1F ) , and observed that most Drosophila genic piC loci were unique to a single species , such as the piC-Adar that is specific to D . vir ( Fig 1E–1Eiii ) . In fact , the mRNA gene expression profiles in the ovarium samples shared much more similarity between these three Drosophilids than the genic piC loci piRNA expression patterns ( Fig 1F ) . Furthermore , the top piRNA-producing genic piC loci from D . mel were also expressed as mRNAs in the other Drosophilid ovariums despite frequently losing the capacity to generate piRNAs ( S2E Fig ) . Finally , we determined the gain and loss rates of genic piC loci in Drosophilids , with the highest rapid gain rates displayed in D . mel and D . ere ( S2F Fig ) . These data suggest that Drosophilid genic piC loci evolved rapidly at the level of sequence elements within the genic transcripts rather than in the control of gene expression , thus leading to this diversity in genic piC loci expression patterns across species . Given the compactness of Drosophilid genomes , a major proportion of intergenic piRNAs in D . mel derive from the flam and 42AB piC loci , the two master control loci implicated in repressing TEs to maintain fertility in females [13 , 30 , 31] . Therefore , we turned our attention to examining these two major intergenic piC loci , and found both to be remarkably young evolutionary inventions ( <12MY ) , having only recently arisen in the melanogaster subgroup ( Fig 1G , S3 Fig ) . By tracking piRNA expression or an intervening TE-rich region , we detected signatures of piC-42AB orthologs in D . simulans ( D . sim ) and D . sechellia ( D . sec ) , and flam orthologs in D . ere and D . yak . However , piC-42AB was absent from D . ere and D . yak genomes , and no flam orthologs were conserved beyond these species despite conserved gene synteny around flam and piC-42AB loci across ~40 MY of Drosophilid evolution ( Fig 1H ) . Previous evolutionary studies suggested a selective advantage for Drosophilid piRNAs to silence TEs , but only up to a limit , when the host organism also tolerates the residing TEs [32–34] . Our data echoes this fluidity for the TE silencing function of major intergenic piC loci . Although they appear essential for TE repression and fertility , large intergenic piC loci as elements can arise and evolve as rapidly as the smaller genic piC loci . Next , we applied our piC locus discovery approach to adult testes small RNA datasets from Primates ( human , macaque , marmoset ) , Glires ( mouse , rat , rabbit ) and Laurasiatherians ( dog , horse , pig ) , ( Fig 2 , S2 Table ) . We observed conserved piRNA expression for piC loci within Rodents ( Fig 2B ) , and within Primates ( Fig 2C ) . However , most genic piC loci , as defined by production of mature piRNAs , were uniquely detected in a single species , including human-specific genic piC loci that in two independent human testes studies appear to generate potentially overlapping 3'UTR-piRNAs ( Fig 2D , S1F Fig ) . Although endogenous overlapping small RNAs are prevalent in invertebrates [35] , this configuration of two potentially overlapping human piRNA cluster loci may hint to the existence of dsRNAs in the mammalian germline beyond endo-siRNA transcripts in mouse oocytes [36–38] . We examined three species-specific piC loci ( piC-Arhgap20 , piC-Piwil1 , and piC-Riok1 ) that were only expressed in mouse , rat , and rabbit , respectively ( Fig 2E–2G ) , and we confirmed these species-specific piRNA expression patterns by Northern blotting ( S4A Fig ) . We also considered whether different stages of testes development or partitioning of piRNAs into the different mammalian Piwi proteins could be the root of these species-specific detections of genic piC loci ( see S4B–S4H Fig and Supplementary Text Discussion ) . Despite comprehensive profiling of genic piC loci across PIWIL2 IP’s and different stages of Glire testes , the genic piC loci repertoires remained highly diverse between species . With regards to intergenic piC loci , their definitions amongst species were greatly influenced by the completeness of the species’ genome assembly , such that mouse and human exhibited the greatest number of intergenic piC loci ( S3 Table ) . Nevertheless , at least 70% or more of genic piC loci could be reproducibly called by our pipeline in two independent human and chicken piRNA libraries ( S4I Fig ) . These analyses mitigate the concern of limits in our piC discovery approach and confirm that the diversity in piC loci expression patterns between species is not an artifact . Our approach was most effective in D . mel , human and mouse , where we detected the expression of >290 , >1700 , >2900 genic piC loci , respectively ( Figs 1F , 2H and S4H ) . Considering the extensive piRNA sequencing coverage in human and mouse libraries , the comparison of piC loci expression patterns from all samples between these two species still show many species-specific genic piC loci expression patterns , which is also reflected by the Primate-specific and Glire-specific genic piC loci patterns ( S4G Fig , middle , right ) . In addition , we observe within each of the three different clades ( Primates , Glires , and Laurasiatherians ) a tendency for adult testes gene expression profiles to show greater overlap when contrasted to the diversity of genic piC loci ( Fig 2H ) . These patterns persisted even when we raised the stringency of cutoffs by 2-fold for piRNAs needed to call a shared piC locus expression and for gene expression profiles to be called overlapping ( S2B and S4H Figs ) . Amongst the top piRNA-generating genic piC loci in mouse and human , the gene orthologs that lost the capacity to make piRNAs in the most distant relative within the clade ( rabbit and marmoset , respectively ) , still tended to be expressed in the adult testes as mRNAs ( S4J and S4K Fig ) . Therefore , we propose that clade-specific and species-specific piC locus expression is a common feature in both flies and mammals; and that the rapid evolution of piC loci expression patterns is occurring at the transcript sequence level to alter entry into piRNA biogenesis pathways rather than at the transcript expression level . We also measured the rate of genic piC loci gain and loss throughout the phylogeny of tetrapods examined in this study , and consistently observed bursts of genic piC loci gains and very few losses since the number of ancestral genic piC loci were low ( S4L Fig ) . This result is consistent with the gain and loss rates for genic piC loci in Drosophilids and further validates the earlier proposal that piC loci expansion is a common phenomenon [14] . Despite rapid evolution of piC loci expression patterns between species , we wondered if analyzing genic piC loci sequences within fly and mouse genomes could reveal sequence motifs and structured RNA elements that might differentiate genic piC loci from standard mRNAs that do not generate piRNAs . To test this question , we selected the top piRNA-producing transcripts from the fly ( D . mel ) and the mouse , respectively , and also selected a list of negative control transcripts that had similar annotated 3'UTR lengths to genic piC loci yet did not make piRNAs ( S5A Fig and methods ) . We then searched each list with an RNA feature analysis against the Open Reading Frames ( ORF ) and 3'UTR sequences . First , we counted the average number of significant structured RNA elements predicted genome-wide by the RNAZ , EvoFold and REAPR algorithms [39–41] . Compared to the negative control sets , the top set ( highest piRNA-producing set ) of fly genic piC transcripts appeared to be enriched in predicted structured RNA elements in the ORF sequences compared to the negative control set , whereas the 3’UTRs of top set of mouse genic piC transcripts were more enriched in structured RNA elements ( S5B and S5E Fig ) . Two different sequence motif discovery programs , MEME and GLAM2 [42] , both predicted a Poly-U-rich motif with greater statistically-significant enrichment amongst the 3'UTRs of both fly and mouse genic piC transcripts compared to the negative control sets , whereas no clear motif was enriched amongst the ORF’s . These initial surveys hint at intrinsic features that may distinguish a genic piC transcript from a regular protein-coding mRNA , and these features may frequently arise or disappear during animal evolution to yield the diversity of clade- and species-specific sets of genic piC loci . To examine whether genic piC loci were subjected to different evolutionary forces compared to non-piRNA producing control transcripts , we compared sequence conservation and evolution rates for fly and mouse genic piC loci with normalized values of the phastCons [43] and phyloP [44] scores for ORF and 3'UTR sequences . High average phastCons scores reflect stronger conservation [43] , whereas positive phyloP scores suggest selective constraints on the sequences’ evolution [44] . Our analyses suggest that genic piC ORFs with the greatest number of piRNAs in both flies and mice were more conserved and under greater selective constraint than negative control mRNAs ( S5F and S5G Fig ) . The faster evolving 3'UTR sequences were expected to be less conserved , reflecting 2–5 fold lower normalized phastCons scores in 3'UTRs compared to ORFs , and these 3’UTRs displayed phyloP scores that were positive but not particularly high . In addition , the 3'UTR phastCons and phyloP scores were not statistically distinct between genic piC loci and negative control transcripts . Since the bulk of genic piRNAs derive from the 3'UTR , this result is consistent with the overall poor conservation of individual piRNA sequences between animal species . Although most piC loci are rapidly evolving , we were struck by the conserved expression of 8% and 5% of genic piC loci in human and mouse , respectively ( Fig 2H ) . To look for deeper conservation of piC loci expression beyond the ~80 MY of divergence between humans and mice , we discovered and compared piC loci expression patterns across nine mammals and the chicken , thus spanning ~300MY of tetrapod evolution [45] . This search revealed 21 intergenic and 56 genic piC loci conserved in piRNA expression across Primates , Glires , and Laurasiatherians ( Fig 3 ) , for which these three clades had diverged from a common Eutherian ( placental ) mammalian ancestor ~100 MY ago [46] . We name these piC loci as Eutherian-Conserved piRNA cluster ( ECpiC ) loci , which tended to yield many more total piRNAs per loci compared to the other Less-Conserved piRNA cluster ( LCpiC ) loci ( Fig 3B ) . Surprisingly , ECpiC loci were not consistently expressed in the opossum , a marsupial; and the platypus , a monotreme; which had diverged even further from Eutherian mammals at an additional ~60 and ~100 MY ago , respectively [45] . This is striking given that the synteny of genes around the ECpiC loci was still preserved , persisting from Eutherian mammals out to the chicken ( Fig 3A ) . The mRNAs for the opossum , platypus , and chicken genes orthologous to genic ECpiC loci were also robustly expressed in the adult testes despite lacking the ability to generate piRNAs ( Fig 4A ) . This suggests that the absence of conserved piRNA expression from the ECpiC loci in opossum , platypus , and chicken testes are not due to gene expression profile differences compared to Eutherian mammal testes . Furthermore , piRNA coverage in the opossum , platypus and chicken adult testes small RNA libraries was excellent and unobscured by miRNAs ( Fig 4B ) [19 , 20] . Finally , we readily identified genic and intergenic piC loci in opossum , platypus , and chicken despite their draft genome states in the UCSC Genome Browser [47] ( Fig 4C and 4D ) . Together , these results confirm ECpiC loci are only conserved in piRNA expression in Eutherian mammals , and suggest that the evolutionary conservation of ECpiC loci is due to preserving sequence elements directing piRNA biogenesis rather than piRNA cluster precursor transcript expression . To further evaluate whether transcriptional regulation determines the choice of transcripts to become piRNA precursors , we considered a previous study showing the A-MYB transcription factor is conserved from mouse to chicken in binding to promoters of Piwi pathway genes to promote a feed-forward loop of piRNA biogenesis in the mouse and chicken testes ( i . e . A-MYB peak at Piwil1 [19] , Fig 4G ) . Although we observed many clear A-MYB peaks at the putative promoters of both chicken and mouse piC loci ( Fig 4C–4F ) , there were notable examples of piRNA absence despite transcriptional activation , such as conserved A-MYB binding and H3K4me3 peaks in a chicken locus orthologous to mouse intergenic ECpiC#1 , and conserved A-MYB and H3K4me3 peaks at mouse Piwil1 . However , few piRNAs were detected from the intergenic piC syntenic chicken locus or from mouse and chicken Piwil1 ( Fig 4E–4G ) . This contrasts with ample Piwil1 genic piRNAs in the rat ( Fig 2F ) . We conclude that transcriptional activation in tetrapod testes , including by A-MYB , is insufficient to determine piRNA biogenesis . This evidence further supports our hypothesis that the evolution of piRNA biogenesis signatures is occurring within the sequence of the precursor RNA transcript rather than in the control of transcription initiation . Although opossum and platypus piRNA clusters were configured similarly to mammalian piRNA clusters for piRNA biogenesis from non-overlapping transcripts ( Fig 4D and 4F ) , chicken piC loci were surprisingly distinct in their configuration of piRNA biogenesis compared to mammalian piC loci . Mammalian piRNAs typically map only to single strands such as single-stranded mRNA transcripts in genic piC loci or to two transcripts that emanate as bi-directional non-overlapping strands from a common promoter region for some intergenic piC loci [4 , 5 , 19 , 48] ( Fig 4F ) . Many chicken piRNAs instead mapped to both plus and minus strands for multiple genic and intergenic piC loci ( Fig 4C and 4D ) . When we calculated a single-stranded configuration index for the major piC loci from three mammals , the chicken , the frog and the fly D . mel ( Fig 4H ) , this index accurately distinguished the single-stranded flam and 20A piC loci from the 42AB and 102E piC loci in the fly , and clearly confirmed that piC loci in all three mammals are predominantly single-stranded in their configurations . Interestingly , chicken piC loci displayed a wide range of piC configurations , with many more double-stranded piC loci with indexes similar to the fly 42AB and 102E double-stranded piC loci . Although the frog is evolutionarily a more distant tetrapod than chicken ( ~370 vs ~320 MY , respectively ) in relation to humans [45] , the frog piC loci determined from oocytes were surprisingly more single-stranded and similar in configuration to mammalian piC loci [49] ( Fig 4H ) . These analyses and the shorter piRNA length distributions of chicken piRNAs ( Fig 4B ) strongly suggest that chicken piRNA biogenesis pathways may be more similar to flies compared to other tetrapods . We hypothesized that extensive conservation of piRNA expression patterns across ~100 MY of evolution implies conserved developmental functions for ECpiC loci . A likely conserved function for intergenic ECpiC loci would be to generate particularly essential TE-directed piRNAs , but TE sequences are neither more enriched nor more conserved in ECpiC loci versus LCpiC loci ( S3 Table ) , and mouse genetic studies knocking out parts of intergenic piC loci are only now beginning to emerge [50] . However , a compelling non-TE repression role may be associated with intergenic ECpiC #18 that is located between the CCAR1 and DDX50 genes and generates many piRNAs antisense to the STOX1 transcript ( Fig 3C , S6 Fig ) . ECpiC#18 is an intergenic piC locus because its transcript is antisense to STOX1 , has lower coding potential than genic piC loci , and H3K4me3 ChIP-seq patterns suggests transcription begins in the intergenic region between STOX1 and DDX50 . The STOX1 gene is implicated as a genetic factor linked to the placental-based disease of preeclampsia in humans [51–53] . A previous Gene Ontology analysis of mouse genic piC loci displayed an enrichment of gene function processes such as nucleic acid metabolism , transcription and regulation-related processes [11] . Such functions are also known for the genes that are ECpiC loci , such as the oncogenic transcription factors ABL2 and ELK4 , the miRNA effector AGO2 , and TBL2 and KCTD7 genes that are in the deleted locus of Williams-Beuren syndrome patients [54] , which display developmental defects . While future genetic studies are required to investigate broader sets of ECpiCs , we looked for existing mutations in ECpiC loci that displayed developmental phenotypes and wondered if the mutations disrupted piRNA biogenesis . We noticed two prominent genic ECpiC loci corresponding to genes Cbl and Asb1 ( Fig 5A ) . Knockout ( KO ) mice in each of these genes were created more than a decade ago because the genes were highly expressed in the blood [55 , 56] . However , these mutants exhibited no major abnormalities , with the exception of being hypofertile . This defect was in both cases due to depleted spermatogenesis [55 , 57] ( Fig 5B ) . Could the mutations in Cbl and Asb1 be causing a loss of genic piRNAs in the testes that might explain the spermatogenic defects ? Northern blotting revealed that Cbl piRNAs were reduced ~2-fold in Cbl Heterozygotes ( HET ) and KO mutants when compared to the Wild-type ( WT ) ( Fig 5C ) , which coincides with difficulties breeding both HET and KO Cbl mutants [57] . Next , we resurrected from sperm the original Asb1 mutant mouse line and sequenced adult testes small RNA-seq and mRNA-seq libraries . Indeed , we found that the LacZ-PGK-Neo insert disrupting the Asb1 transcript greatly reduced Asb1 piRNAs ~6-fold in the KO and ~2-fold in the HET testes ( Fig 5D and 5E ) . Other genic and intergenic piC loci were unaffected in the Asb1 KO testes compared to HET testes ( Fig 5F ) , indicating the specificity of the mutation only affecting the piC-Asb1 . Expression profiling indicated a significantly greater number of up-regulated genes than down-regulated genes when comparing Asb1 KO testes to HET testes ( Fig 5G ) , whereas far fewer genes were either up- or down- regulated in KO versus HET kidney , a somatic tissue expressing Asb1 . In addition , no mammalian TE transcripts were detected to be up-regulated in the Asb1 mutant testes . Some but not all predicted targets complementary to Asb1 piRNAs were up-regulated in Asb1 KO versus HET testes ( Fig 5H ) , so future studies will be focused at distinguishing which up-regulated transcripts in the Asb1 mutant are direct or indirect targets of the piRNAs .
Our study provides a new understanding of the evolutionary patterns for animal piRNAs and piC loci . These integral components of the ancient Piwi pathway evolve much more quickly than RNAi protein factors and other small regulatory RNAs like miRNAs , which can be conserved as far back as ~500 MY of evolution [1 , 2] . By matching orthologous piC loci via synteny and protein orthology , and then broadly profiling piRNAs and mRNAs across animal gonads , we confirm rapid evolution in the piRNA expression patterns for both genic and intergenic piC loci even between close relatives within insect and mammalian clades that have only diverged by ~10 MY of evolution . This rapid gain of species-specific genic piC loci in both insects and tetrapods is consistent with a previous study’s proposal that piC loci are expanding rapidly via positive selection processes [14] . However , Eutherian mammals have distinctly conserved the expression of piRNAs from a notable set of ECpiC loci through ~100 MY of evolution , whereas very few deeply conserved piRNA expression patterns were observed in flies . What might explain the lack of deeply-conserved piRNA expression patterns in flies ? Perhaps piC loci evolution may be accelerated in Drosophilids due to certain aspects of the Piwi pathway that are currently unique to Drosophila , such as the capacity of de novo TE insertions to promote new piRNAs from flanking genomic sequence [58 , 59] , Drosophila-specific piRNA biogenesis factors like RHINO , CUTOFF , and DEADLOCK [58 , 60–62] , and epigenetic induction and suppression of piRNA biogenesis [63–68] . In addition , whereas only a minority of mammalian piRNAs are complementary to TEs [4 , 5] , the majority of Drosophila piRNAs are complementary to TEs [13 , 30 , 31] , perhaps reflecting a more dynamic “arms race” between TEs and Drosophilid piRNAs [32–34] . Nonetheless , the rapid evolution for the majority of piC loci in both mammals and flies may also suggest that many piC loci are evolving by non-adaptive evolutionary forces that result in diverse piRNA repertoires , which germ cells may simply tolerate along with their highly diverse transcriptomes [69] . In addition , this diversity of piC loci between species may be attributed to higher frequencies of mutations in non-coding portions of genes and intergenic regions that allow transcripts to enter or leave the piRNA biogenesis pathway with greater frequency , since these turnover events may be under relaxed evolutionary constraints [14] . Nevertheless , some specific piC loci might also behave differently from the bulk of piC loci with regards to evolutionary constraints , as has been proposed for some human piC loci [15 , 70] . Thus , the conservation of piRNA expression from ECpiC loci is striking , and we speculate these loci may have been selected to yield high levels of specific piRNAs in order to promote gene expression profiles favoring spermatogenic and embryonic fitness . This list of ECpiC loci will help us prioritize which loci to generate future piC loci mutants , and leads us to wonder if their functions will be tied to unique aspects of Eutherian reproduction , such as placental development and paternal genome regulation during spermatogenesis . Our analysis also suggests that most piC loci in vertebrates and flies are evolving under possibly non-adaptive evolutionary forces , whereby genetic drift may frequently create species- and clade-specific sequence motifs or RNA structural elements that allow diverse repertoires of transcripts to frequently enter ( and exit ) the piRNA biogenesis pathway . Although animal gonads are notable for their generally promiscuous transcriptional activity as well as rapid evolutionary turnover of genes [71 , 72] , we still observed that gene expression profiles between species are more similar than genic piC loci piRNA expression profiles ( Figs 1 and 2 ) . We propose a model for most piC loci being neutral for gonadogenesis fitness , and the plethora of individual piC loci may result in functional redundancy and allow fluidity in the emergence and evolutionary turnover of piC loci ( Fig 6 ) . However , some piC loci have been subjected to adaptive evolutionary forces in order to help germ cells suppress TE mobilization , such as prominent intergenic piRNA cluster loci serving as master TE control loci in Drosophilids [30] . Mutations disrupting Asb1 and Cbl in the mouse appear to also disrupt the piRNAs from these individual piC loci , hinting that the loss of piRNAs may be causing the spermatogenic defect . Indeed , more genes are up-regulated than down-regulated in the Asb1 mutant testes , with several up-regulated genes having complementarity to Asb1 piRNAs ( Fig 5 ) . Although we cannot exclude the idea that loss of Asb1 protein function is causing the spermatogenic defect , we note that five other close Asb1 homologs ( Asb3 , -4 , -5 , -8 , and -9 ) are still highly expressed in the Asb1 mutant and might provide protein function compensation [55] , whereas only Asb1 generates an abundant cluster of piRNAs . Future efforts should be directed at creating new mouse mutants that replace the endogenous 3'UTRs of Asb1 and Cbl with an equivalently long 3'UTR of a different gene that is also expressed highly in the germline but does not generate piRNAs . This approach may be better than simply deleting the endogenous 3'UTRs because Asb1 and Cbl mRNA stability and translation would likely be strongly diminished if they completely lacked a 3'UTR . Our discovery of ECpiC loci suggests new regulatory functions beyond TE repression to accommodate specific aspects of Eutherian reproduction . There is a precedent for an epigenetic process specific to Eutherian reproduction , such as random X-chromosome inactivation via the Xist non-coding RNA , which differs from the paternally imprinted inactive X chromosome in marsupials and stochastic dosage compensation in monotremes [73] . ECpiC loci such as piC-Asb1 , piC-Cbl , and ECpiC#18 may represent new gene targets to examine for cases of hypo-fertility , which may be more likely to persist in a population than completely sterile mutations . Albeit reduced in fecundity , reproduction may be still viable in young piC locus mutant animals , but germ cell longevity may also be limited . When new ECpiC loci mutants are made that will only disrupt the piRNAs while leaving the protein coding gene intact , we will be better able to discern which target genes are mis-regulated during spermatogenic decline or defects on placental development . Since the piRNAs from ECpiC#18 are mainly antisense to the STOX1 mRNA , a gene clinically implicated in placental development [51–53] , we speculate that mis-regulation of ECpiC#18 might likely impact STOX1 expression and contribute to sperm phenotypes that fall within a hypothesis for paternal factors in the etiology of preeclampsia [74] . Future characterization and genetic studies in mouse will be useful to determine if ECpiC#18 directly regulates the STOX1 locus .
This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . Mice and rats were euthanized with CO2 and followed with cervical dislocation or decapitation , prior to testes dissection . Rabbit tissues were purchased , while other vertebrate datasets were downloaded from the NCBI Sequencing Read Archive . This work was conducted under the approval of the Brandeis University Institutional Animal Care and Use Committee ( IACUC ) under the protocol #13013 to NCL . To deplete the transposon-directed piRNAs which overwhelm genic piRNAs in typical total Drosophila ovaries small RNA libraries [13] , we dissected >500 ovaries in cold PBS and that were treated to a trypsinization and gravity sedimentation protocol that ruptures most nurse cells and oocytes but leaves the layer of follicle cells intact [28] . The D . mel strain was Oregon-R , while D . ere , D . yak , and D . vir strains were the standard wild-type strains used in the genome projects and obtained from the UCSD Drosophila Species Stock Center . Total RNA from rodent testes was extracted using TRI-reagent ( MRC ) using manufacturer instructions . Small RNA in size range from 18 to 35 bases were gel purified , for Drosophila samples depletion from 2S rRNA was performed and libraries were constructed as described in [28] . To construct mRNAseq libraries , total RNA samples were subjected to two rounds of poly-A enrichment using biotinylated ( dT ) x18 oligonucleotides and the polyATtract kit ( Promega ) . RNA-seq libraries were generated using the ScriptSeq V2 construction protocol ( Epicenter , performed according to manufacturer’s instructions ) . Sequencing was performed on an Illumina HiSeq 2000 , and 50 bases long reads were processed and split according to their index primer barcodes . All read data that we generated for this study has been deposited in the Sequencing Reads Archive ( SRA ) of the Gene Expression Omnibus ( GEO ) under the accession number of GSE62556 . All the other small RNA and mRNA datasets were downloaded from GEO and SRA , with all accession numbers recorded in S1 Table . The initial deep sequencing library preparation steps with standard processing tools is described in the Supplementary Materials Additional Experimental Procedures section . Genic piRNA clusters were determined by a process ( S1c Fig ) that begins with a custom script ( ngs_genecentric_wig . c ) that compares the counts in a WIG file to a pre-defined gene structure BED file ( obtained and modified from UCSC genome browser ) . The script calculates the 5’UTR , exon and 3’UTR read counts for each gene in the BED file . When preforming the 5’UTR and 3’UTR counting , the script automatically extends 5’UTR or 3’UTR regions by a window of 2000bp ( 500bp for Drosophilids ) if it detects at least 1 RPM reads within the window . After downloading original Refseq BED tables from the UCSC genome browser website , we modified the RefSeq BED files to conform to our custom script . For human , fly , and mouse , we kept the original species-specific RefSeq annotations . But for the rat and rabbit , we used the mouse RefSeq annotation first , and then supplemented it further with human RefSeq annotations where mouse annotations were missing . For the marmoset , monkey , dog , horse , pig , platypus , opossum , and chicken; we first used the human annotation , then the mouse annotation , and finally add the species specific annotations . We curated these tables by only retaining genes with ≥10 RPMs of piRNA counts in the 3’UTR of the same strand orientation of the mRNA . In mammals , between 1–5% of genic piRNA clusters also had at least 50 RPMs of piRNAs mapping to the ORF that were kept in the analysis . We further groomed these tables by removing false positives that were genes with exceedingly high 3’UTR counts which were ambiguously attributed to a nearby transposable element , or were adjacent to another gene that was actually generating the piRNAs , was a duplicated gene name entry , or was a non-coding RNA ( i . e . many mouse RefSeq non-coding RNAs with the “gm####” identifier ) . We finally removed additional false positives that were genes with incorrect homolog alignments caused by their simple and repetitive protein domains , such as histones , olfactory receptors and zinc finger protein genes . Mouse , human and flies intergenic clusters lists were previously determined in [4 , 5 , 13 , 29 , 30] . Intergenic clusters for other species were detected with a custom script ( ngs_coverage-nelson . c ) that uses a 5 Kb sliding window , initiates a cluster when the window read count is > 1 RPM , and terminates the cluster when the read count is < 1 RPM . Intergenic piRNA cluster loci defined by contiguous intervals were then kept if they totalled ≥10RPMs . We computed the Single-stranded configuration index of piRNA clusters by first taking each cluster locus and splitting it into windows of 5 Kb each , and then for each of these windows we calculated the formula [ ( A-B ) ^2]/ ( A2+B2 ) , where “A” is the plus strand read counts in RPM and “B” is the minus strand read counts in RPM within a window . The single-stranded configuration index of a cluster is the average single-stranded configuration index of all the 5Kb windows . The gene expression profiles for samples from human , mouse and D . mel were determined by mapping mRNA-seq reads to RefSeq transcripts with Bowtie ( up to 2 mismatches ) . Read counts were normalized to mapping library size ( RPM ) and then gene length ( RPKM ) , and finally each gene was normalized to a housekeeping gene ( RpL32 in mammals and in flies , RpL32 is also known as Rp49 in flies ) . Since RefSeq transcript libraries are deficient in other animals , for the other Drosophilids , mapped reads were then transformed to D . mel gene models through orthologs listed in the OrthoDB database and then RPM counts were determined as described in [75] . mRNA-seq reads for non-human Primates , Laurasiatherians , and other tetrapods were mapped to their respective genomes , and then the counting algorithm from genic piC discovery was applied to yield exonic counts that were as accurate as mapping to RefSeq transcripts , except that no additional curation was applied . Differential gene expression analysis between Asb1 KO and Het mutant mouse tissues was conducted with the edgeR package [76] on two biological replicates from testes and one sample of kidney . For predicting targets of piC-Asb1 , Asb1 piRNAs were queried against the mouse RefSeq transcripts using the “blastn-short” command in BLAST [77] and then sorting for BLAST results with an e-value≤1 , which in this query frequently demanded at least 13bp of complementarity between the piRNA and the predicted target . Custom Perl scripts and SQL queries counted the BLAST matches and gene expression changes . For the species comparisons Venn diagrams , the genic piC names and the gene expression profiles were rooted to the best annotated model organism gene names from D . mel for Drosophilids; and human and mouse for mammals and chicken . Genes and genic piC loci could thus be compared between species with SQL queries . Overlaps analyses in genic piC locus conservation required threshold piRNA expressions to be ≥10 RPMs ( Figs 1F and 2G ) and ≥20 RPMs ( S2B and S4G Figs ) . Overlap in gene expression profiles were defined as similar expression levels of RPKM values normalized against the RpL32/Rp49 housekeeping gene to be within 3 fold of each two species comparison ( Log10 delta value ≤ 0 . 5 ) and within 2 fold for three species comparisons ( Log10 delta value standard deviation ≤ 0 . 3 ) for analyses in Figs 1F and 2G . For gene expression analyses in S2B and S4G Figs , the Log10 delta value ≤0 . 3 and Log10 delta value standard deviation ≤ 0 . 177 was used . To test if the distribution of piC locus overlaps were significantly different from the distribution of similar gene expression profiles , we calculated both the Chi-square test and the Ratios proportion Z-score test with Bonferroni correction . The human Study #2 dataset is a second independent sample of small RNAs generated from human testes , and it shares much similarity with the human Study #1 dataset , such as similar piRNA cluster loci patterns in Figs 2B , 2C and S4I . All human piC loci comparisons in Fig 3 were done with the human Study#1 datasets after we had confirmed highly similar profiles and counts for all ECpiC loci in the human Study#2 dataset . We ranked genic piC loci by piRNA abundance ( RPMs ) and selected top cohorts of 278 and 332 genic piC loci for fly and mouse , respectively . These sets were divided into a top and bottom half , and compared to lists of negative control genes that do not generate piRNAs . The negative control list contained a total of 3 times as many genes as genic piC loci to diminish selection bias , and >90% of these negative control genes were also expressed in both fly follicle cell enriched samples and mouse adult testes . The 3’UTR boundaries of the genic piC loci were defined by the previous piRNA tracking algorithm while the negative controls were selected based on the longest Refseq-annotated 3’UTR . In order to ensure that genes in negative control set have approximately similar 3’UTR length compared to genic piC loci set , we padded the 3’UTR in fly by 250 to 500 bases and in mouse by 500 to 1100 bases , unless the extension overlaps with a gene in the same strand . The genomic coordinates for the 3’UTRs and CDS ( minus introns ) were determined for each negative control gene and genic piC; and used to track and count predicted RNA structural elements and phastCons and phyloP scores . RNA structural elements from fly genome ( Release 5/Dm3 ) were determined with RNAz , Evofold , and REAPR applied to the 12 Drosophilids alignments with the deviation parameter , dev = 20 and confidence scores >0 . 6 [39–41] , while only REAPR was applied to the mouse genome ( Mm10 ) using an alignment of 8 Glire genomes and measured at deviation levels of dev = 10 and confidence scores >0 . 8 . The per-base values of phastCons [43] and phyloP [44] scores were downloaded from UCSC Genome browser with the exception of fly phyloP scores computed by David Garfield at the EMBL , and were based off the 12 Drosophilids genomes and 60 vertebrate genomes alignments . Total per-base values were summed and then averaged by the base length for each gene using BEDtools [78] and custom Perl scripts to count the features for each gene . Sequence motif analysis was performed with MEME [42] using an “OOPS ( only once per sequence ) ” model; and with GLAM2 on CDS and 3’UTR sequences of the longest isoform from the masked genome . To determine the gain and loss rates for genic piC loci in Drosophilids and tetrapods , we constructed phylogenetic trees according to the Gregorian clock of millions of years of divergence and the tabulated expression or absence of expression for each piC loci with gene orthologs present between all the species . We followed the similar procedure of measuring gain and loss rates of miRNA genes as detailed in [20] , using the COUNTS program [79] and the Wagner parsimony approach . Additional experimental procedures are in the Supplementary Materials document describing the verification of gene expression using RT-qPCR , small RNA northern blotting , and the PIWI IP from OSS cells and PIWIL2 IP from Glire testes .
|
Animal genomes from flies to humans contain many hundreds of non-coding elements called Piwi-interacting RNAs ( piRNAs ) cluster loci ( piC loci ) . Some of these elements generate piRNAs that direct the silencing of transposable elements , which are pervasive genetic parasites . However , we lack an understanding of the targeting function for the remaining bulk of piRNAs because their loci are not complementarity to transposable elements . In addition , the field does not know if all piC loci are quickly evolving , or if some piC loci might be deeply conserved in piRNA expression , an indication of its potentially functional importance . Our study confirms the highly rapid evolution in piRNA expression capacity for the majority of piC loci in flies and mammals , with many clade- and species-specific piC loci expression patterns . In spite of this , we also discover a cohort of piC loci that are deeply conserved in piRNA expression from the human to the dog , a significantly broad phylogenetic spectrum of eutherian mammals . However , this conservation in piRNA expression ends at non-eutherian mammals like marsupials and monotremes . Existing mutations in two of these Eutherian-Conserved piC ( ECpiC ) loci impair mouse reproduction and abrogate piRNA production . Therefore , we suggest these ECpiC loci are conserved for piRNA expression due to their important function in eutherian reproduction and stand out as prime candidates for future genetic studies .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Conserved piRNA Expression from a Distinct Set of piRNA Cluster Loci in Eutherian Mammals
|
Staphylococcus aureus causes acute and chronic infections resulting in significant morbidity . Urease , an enzyme that generates NH3 and CO2 from urea , is key to pH homeostasis in bacterial pathogens under acidic stress and nitrogen limitation . However , the function of urease in S . aureus niche colonization and nitrogen metabolism has not been extensively studied . We discovered that urease is essential for pH homeostasis and viability in urea-rich environments under weak acid stress . The regulation of urease transcription by CcpA , Agr , and CodY was identified in this study , implying a complex network that controls urease expression in response to changes in metabolic flux . In addition , it was determined that the endogenous urea derived from arginine is not a significant contributor to the intracellular nitrogen pool in non-acidic conditions . Furthermore , we found that during a murine chronic renal infection , urease facilitates S . aureus persistence by promoting bacterial fitness in the low-pH , urea-rich kidney . Overall , our study establishes that urease in S . aureus is not only a primary component of the acid response network but also an important factor required for persistent murine renal infections .
Bacterial pathogens often encounter acidic environments within host tissues and employ several direct and indirect defense measures [1] . Direct measures include the utilization of proton pumps and generation of alkaline compounds such as ammonia to neutralize pH . Indirect methods such as damage repair , biofilm formation , and metabolic alterations , are utilized to rescue cell viability . Staphylococcus aureus is a leading cause of opportunistic infections in community and health care settings [2 , 3] . S . aureus resides in multiple acidic niches during colonization and infection of the human host that include the surface of the skin and within abscesses [4–6] . It is important to note that S . aureus is sensitive to acetic acid stress when growing in the presence of excess glucose [7] . Weak acids such as acetic acid are unique in potentiating stationary phase cell death , in that unlike strong acids that fully dissociate in water , the undissociated weak acids can easily enter into the cytoplasm and reduce the intracellular pH by releasing protons . Therefore , S . aureus must overcome different kinds of acid stress to maintain viability . However , mechanisms of acid resistance in S . aureus are not well described . It has been shown that sodA , which encodes a superoxide dismutase , is induced upon acid stress and facilitates acid tolerance by alleviating the cell damage caused by reactive oxygen species [8] . In addition , a σB–dependent acid-adaptive response has been described that facilitates S . aureus survival in media with a pH of 2 after pre-exposure to a sub-lethal pH of 4 [9] . Based on global transcriptional studies , increased urease activity is thought to be a major contributor to acid resistance in S . aureus [10–13] . In humans , urea is produced in the liver via the urea cycle as a means to remove excess nitrogen . Urea enters the bloodstream , becomes concentrated in the kidneys , and is excreted during urination . The concentration of urea in the blood is normally 2 . 5–7 . 1 mM , and it is found in other body fluids such as gastric acid , sweat , and saliva . The level of urea in the saliva is 3–10 mM in healthy individuals but can reach 15 mM in patients with renal diseases [14] . Notably , the re-absorbance of urea from the collecting ducts makes the interstitium of the kidney inner medulla a urea-rich environment . Urease ( EC: 3 . 5 . 1 . 5 ) is a nickel-dependent metalloenzyme that catalyzes the hydrolysis of urea into ammonia ( NH3 ) and carbon dioxide ( CO2 ) [15–17] . For some bacterial species , urease is an integral part of the bacterial acid response network , as the hydrolysis product ammonia is readily protonated into ammonium ( NH4+ ) , during which process protons are consumed , resulting in an increase in pH [1] . Urease is crucial for niche adaptation of many bacterial pathogens . For example , urease is essential for the survival of Helicobacter pylori in the stomach lining , where the pH can be as low as 2 . 5 [18] . With a high affinity for urea , urease from H . pylori is required not only for the establishment of infections but also for the maintenance of a chronic infection [19] . Also , Streptococcus salivarius produces urease to utilize salivary urea as a nitrogen source for growth while resisting acid stress [20] . Over 90% of S . aureus strains are urease-producing [21] , which is encoded by the urease gene cluster ureABCEFGD . The α , β , and γ subunits that comprise the apoenzyme are encoded by ureC , ureB , and ureA , whereas ureEFGD genes encode accessory proteins . Previous studies have shown that urease genes are highly transcribed during biofilm growth conditions [13 , 22] . However , the function and utilization of urease in S . aureus has not been comprehensively studied . In this work , we explored the in vitro and in vivo functions of urease in S . aureus . We found that S . aureus primarily utilizes urease to facilitate pH homeostasis under weak acidic stress , but it does not utilize urea as a nitrogen source under neutral pH . Lastly , our data demonstrate that urease is essential for S . aureus to persist in mouse kidneys , where a significant pH gradient exists and urea is an abundant nitrogen source .
Previous work has documented that aerobic growth of S . aureus in tryptic soy broth ( TSB ) containing excess glucose ( 35–45 mM ) impairs stationary phase survival of S . aureus [7] . Under these growth conditions , the acetate derived from glucose catabolism is not consumed as a secondary carbon source , and the pH in the medium remains low , which potentiates cell death . Based on those observations , we hypothesized that the presence of urea in TSB containing 45 mM glucose would rescue cell death due to ammonia generation via urease activity . To test this hypothesis , we performed a growth assay in which JE2 wildtype ( WT ) and JE2 ureB::ΦΝΣ ( ureB ) were aerobically cultured in TSB containing 45 mM glucose with or without 10 mM urea . Over the period of 120 h , colony forming units ( CFU/ml ) and extracellular pH were monitored every 24 h . In addition , culture supernatant was analyzed to measure glucose , acetate , urea , and ammonia concentrations . In the absence of urea supplementation , both WT and the ureB mutant showed a drastic decrease in cell viability ( ~9 log10 difference ) while maintaining an acidic extracellular pH ( ~4 . 8 ) ( Fig 1A and 1B ) . However , we observed a urease-dependent increase in viability ( ~8 log10 difference ) and medium pH ( ~4 pH unit difference ) in the presence of exogenous urea ( Fig 1A and 1B ) . Both pH and viability phenotypes of the ureB mutant were complementable by integrating ureABCEFGD into the chromosomal SaPI1 attC site ( S1 Fig ) . Glucose in the medium was depleted by 24 h for both WT and the ureB mutant either with or without urea supplementation ( Fig 1C ) . However , only WT grown in TSB supplemented with 10 mM urea consumed acetate ( Fig 1D ) . Lastly , we observed a urease-dependent consumption of urea coincident with the generation of ~20 mM NH3 ( Fig 1E and 1F ) . Thus , these results suggested that the S . aureus urease functions as part of an acid response network to facilitate pH homeostasis in the presence of urea . Weak acids and subsequent intracellular acidification has been previously shown to generate endogenous reactive oxygen species and potentiate cell death [7] . Thus , to evaluate the physiological status of WT and the ureB mutant in the above growth assay , flow cytometry was performed to assess cellular respiration and reactive oxygen species ( ROS ) levels . The results confirmed that urease-mediated pH homeostasis rescued cellular respiration and protects cells from endogenous ROS under weak acid stress in the presence of urea at 72 h of growth when viability differences are evident ( Fig 1A and S2 Fig ) . It is important to note that in TSB the concentration of arginine , which can be catabolized to generate NH3 via arginase/urease , nitric oxide synthase ( NOS ) , or two separate arginine deiminase ( ADI ) systems ( Fig 2A ) , is not sufficient to rescue the survival of JE2 in this assay . Therefore , we repeated the assay with excess arginine ( 5 mM ) in TSB containing 45 mM glucose . As a result , excess arginine was unable to rescue viability of JE2 to the same extent as urea , although we did observe an arginine deiminase-dependent increase in viability ( ~2 log10 ) ( S3A and S3B Fig ) . Collectively , these data suggest that urea derived from arginine via RocF and subsequent urease activity is not functional in this assay . However , Staphylococcus epidermidis catabolized either arginine ( S3C and S3D Fig ) [23] or urea ( S3E and S3F Fig ) to rescue growth under weak acid stress . To further investigate the transcriptional regulation of the ure operon , a lacZ reporter plasmid pNF315 was generated in which the promoter of ure was fused to the promoterless lacZ gene and transduced into JE2 . The previous experiments ( Fig 1 ) suggested that ure transcription or urease function is induced under weak acid stress [10] . Indeed , as the pH deceased due to the accumulation of acetate , ure transcription was induced 3 . 3-fold at 6 h comparing to 2 h , in TSB containing 45 mM glucose with or without 10 mM urea ( Fig 3A–3C ) . When the media were buffered to a pH of 7 . 25 with 100 mM 3- ( N-morpholino ) propanesulfonic acid ( MOPS ) , the transcription of ure was significantly inhibited regardless of urea supplementation ( Fig 3A ) . These results indicate that the transcription of urease genes is induced by weak acid stress . Multiple global transcriptional studies have suggested that the accessory gene regulator ( Agr ) quorum sensing system , as well as global regulators CcpA and CodY , function to regulate the transcription of the urease operon [24–26] . To assess these relationships in our model , JE2 WT , ΔccpA::tetL ( ΔccpA ) , Δagr::tetM ( Δagr ) , and ΔcodY::ermB ( ΔcodY ) each containing pNF315 were grown aerobically in TSB containing 45 mM glucose and 10 mM urea . β-galactosidase activity assays were performed with cell lysate collected during early- ( 2 h ) , mid- ( 6 h ) , and post- ( 10 h ) exponential phases of growth . The transcription of the ure operon was significantly decreased in ΔccpA/pNF315 ( 6 h ) and Δagr/pNF315 ( 6 and 10 h ) , and significantly increased in ΔcodY/pNF315 at 6 and 10 h ( Fig 3D ) , indicating that the transcription of ure genes is activated by CcpA and Agr and negatively regulated by CodY . However , it is unclear if ure transcriptional regulation by CcpA , Agr , or CodY is via direct or indirect regulation . To corroborate the effects of these regulators on urease activity , JE2 WT , Δure , ΔccpA , Δagr and ΔcodY were grown in TSB containing 45 mM glucose and 10 mM urea for 120 h ( Fig 3E–3H ) . As expected based on the transcriptional analysis , ΔcodY essentially phenocopied WT with regards to pH , acetate production and viability ( Fig 3E–3H ) . However , since Δagr displayed reduced ure transcription , it was predicted that the viability would be significantly reduced in the 120-h growth assay . Indeed , the extracellular pH of Δagr was significantly different from WT by 6 h of growth ( Fig 3E ) and remained at 5 . 1 from 24–120 h similar to Δure . Further , in comparison to JE2 WT , the viability of Δagr decreased ~6 log10 to ~102 CFU/ml at 120 h similar to Δure . Lastly , based on Fig 3D and the β-galactosidase activity assay documenting a decrease in ure transcription , we would expect the ΔccpA mutant to have decreased viability similar to Δagr and Δure . However , we found that the pH was significantly higher than WT/ ΔcodY from 6–12 h of growth ( Fig 3E ) . In addition , it produced less extracellular acetate than WT ( Fig 3F ) , presumably due to consumption of acetyl-CoA via the tricarboxylic acid ( TCA ) cycle as CcpA represses TCA cycle activity [27–29] . Further , it survived as well as WT , and the pH remained alkaline over the entire 120 h ( Fig 3G and 3H ) . Based on our previous work documenting the function of CcpA in repressing amino acid catabolism [30] , we hypothesized that the survival of the ΔccpA mutant in the above growth assay was urease-independent due to derepression of amino acid catabolism and subsequent generation of ammonia . Previous investigations have documented that in the presence of a preferred carbon source such as glucose , CcpA represses amino acid catabolic genes including gudB ( encoding glutamate dehydrogenase ) , rocF ( encoding arginase ) , putA ( encoding proline dehydrogenase ) , and arcA1/arcA2 ( encoding arginine deiminases ) [23 , 24 , 30–32] , all of which produce ammonia as a byproduct . To determine whether the survival of ΔccpA was dependent upon urease or catabolism of a particular amino acid , a growth assay was performed in which JE2 WT , ΔccpA , ΔccpA/gudB::ΦΝΣ , ΔccpA/Δure , ΔccpA/putA::ΦΝΣ , and ΔccpA/arcA1::kan/arcA2::ΦΝΣ were cultured in TSB containing 45 mM glucose . In the absence of urea , JE2 WT was unable to survive as expected , presumably due to a dramatic decrease in pH observed over the 120 h experimental timeframe ( Fig 4A and 4B ) . However , ΔccpA/gudB::ΦΝΣ , ΔccpA/Δure , ΔccpA/putA::ΦΝΣ , ΔccpA/arcA1::kan/arcA2::ΦΝΣ all phenocopied ΔccpA , suggesting that it was not the catabolism of one specific amino acid that was responsible for cell survival in this assay ( Fig 4A and 4B ) . These data led to the hypothesis that the derepression of overall amino acid catabolism provides the ΔccpA mutant a growth advantage . Further , as shown in Fig 3F and previously it is known that ΔccpA mutants produce less extracellular acetate due to derepression of the TCA cycle [27–29] . To address this hypothesis , JE2 WT and ΔccpA were grown in TSB containing 45 mM glucose . As expected , the ΔccpA mutant generated significantly more ammonia and the extracellular acetate was eventually consumed by ΔccpA compared to JE2 WT ( Fig 4C ) . Amino acid analysis of the same supernatant demonstrated that the ΔccpA mutant consumed histidine , aspartate , proline , glutamate , alanine , and arginine at a faster rate than WT ( Fig 4D–4H ) . Taken together , these data suggest that the absence of CcpA in S . aureus facilitates survival in the presence of excess glucose due to decreased acetate generation and increased ammonia generation by amino acid catabolism of multiple amino acids . In previous experiments using TSB containing 45mM glucose , catabolism of arginine in the media is repressed by CcpA ( Fig 4H ) . Therefore , we were unable to determine the potential function , if any , of endogenously derived urea . Further , it is unclear if urease is active at neutral pH and generates NH3 for use in nitrogen metabolism . To determine if urease is functional in media where arginine is rapidly catabolized thus generating urea from arginine via arginase ( RocF ) , we grew JE2 in buffered chemically defined medium ( CDM ) lacking glucose [32 , 33] . CDM is a defined medium that lacks glucose but contains 18 amino acids except glutamine and asparagine [34] , buffered at pH 7 . 5 . We reasoned that if urease catalyzes the reaction generating NH3 from urea , the ammonia would be actively utilized by glutamine synthetase to synthesize glutamine from glutamate ( Fig 2A ) . Both glutamine and glutamate are major amino donors for cellular reactions [35] . Therefore , JE2 WT and the ureABCEFGD deletion mutant ( Δure ) were grown aerobically in CDM containing 0 . 1 g/L 15N-arginine ( guanidino-labeled only ) and cells were harvested at 7 h , at which time both supernatant and intracellular components were assessed by nuclear magnetic resonance ( NMR ) to detect 15N-glutamine [36] . Since the ammonia generated from amino acid catabolism is also utilized for glutamine synthetase , the rapidly catabolized serine was labeled with 15N as a control [31] . As a result , 17 times more 15N-glutamine was detected when 15N-serine was added to CDM than when 15N-arginine was added ( Fig 2B ) . In addition , no difference was noted when comparing the 15N-glutamine detected from WT or Δure when grown in CDM containing 15N-arginine . Therefore the small amount of detected 15N-glutamine was derived from ADI or NOS-generated 15NH3 ( Fig 2B ) . However , in CDM containing 15N-labeled arginine , significant 15N-labeled urea was detected extracellularly in both WT and Δure ( Fig 2C ) , indicating that the nitrogen from arginine catabolism does not enter the intracellular nitrogen pool , but is rather excreted as urea under neutral pH . Our data demonstrated that urease facilitates pH homeostasis and cell survival in vitro under weak acid stress in the presence of urea . However , it is unclear if urease functions to facilitate staphylococcal colonization or virulence . One niche where urease may be important is the host skin , where S . aureus resides within hair follicles and sweat glands [37] . Moreover , it is known that human sweat contains 22 . 2 mM urea [38] and the pH of human skin is ~4 to ~6 [39] . However , animal models of S . aureus skin colonization are difficult to replicate since mice and other rodents do not sweat . Therefore , we reasoned that another host niche where urease might be required was in the colonization of the kidney , which has a low tissue pH and a relatively high concentration of urea . In addition , it is well known that S . aureus causes chronic kidney infections in mice and thus the kidney provides a nidus for subsequent staphylococcal metastasis ( S4 Fig ) [40] . To test this hypothesis , we used a mouse bacteremia model in which C57BL/6 mice were retro-orbitally injected with JE2 WT and Δure . On days 8 , 12 and 19 post-infection , bacterial burden in the kidney was determined ( Fig 5 ) . Although no difference between WT and Δure was noted on day 8 ( Fig 5A ) , kidneys infected with Δure had significantly lower bacterial burden on days 12 and 19 , with more kidneys below the limit of detection infection compared to WT ( Fig 5B and 5C ) , indicating that urease contributes to the persistence of S . aureus during a mouse chronic kidney infection . To determine if the host immune response differed between mice infected with either WT or Δure , leukocyte populations were assessed from infected kidneys on day 8 , an interval where bacterial burdens were equivalent , to prevent bias from animals that had cleared the infection . However , no significant differences were noted between these two groups ( S5 Fig ) suggesting the absence of ure did not skew the immune response to facilitate enhanced clearance .
Acid stress , along with other environmental risk factors such as extreme temperatures , osmotic pressure , and nutrient depletion , is challenging for bacterial survival [41] . Accordingly , a variety of strategies are utilized to resist low pH which also contribute to bacterial virulence [1] . In Escherichia coli , four main acid resistance systems ( ARs ) have been described: the oxidative system AR1 , the glutamate-dependent AR2 , the arginine-dependent AR3 , and the lysine-dependent AR4 [42] . The amino acid-dependent ARs are composed of a decarboxylase which consumes protons , and an inner membrane antiporter which imports the decarboxylase substrate while exporting the product . Bacillus cereus activates not only the general stress response genes via σB but also proton transporters and amino acid decarboxylases , as well as the ADI system that produces ammonia [43] . H . pylori is known for the ability to proliferate in extremely low pH environments such as the host gastric acid [44 , 45] . In addition to amino acid catabolism , enhanced urease activity sustains favorable intracellular pH by generating ammonia [18 , 45–48] . Mycobacterium tuberculosis resists acid stress through nitrogen assimilation from asparagine hydrolysis [49] , as well as urea hydrolysis [50] . However , there is little known about acid resistance in S . aureus although it proliferates in multiple mildly acidic niches of the human host . In the aerobic growth assay of S . aureus cultured in TSB containing 45 mM glucose , the acetate produced via glucose catabolism is excreted into the growth medium ( Fig 1D ) . When the extracellular pH of the growth medium reaches the pKa of acetic acid ( 4 . 8 ) , acetate becomes protonated and is able to traverse the cell membrane and release the proton in the near neutral pH of the cytoplasm . The drop in intracellular pH potentiates cell death in S . aureus by intracellular acidification and ROS generation [7] . We documented that in the presence of urea , urease functions to facilitate pH homeostasis in a weak acid environment through the generation of ammonia that inhibits the acetate-dependent intracellular acidification ( Fig 1E and 1F ) [7] . In addition , it was confirmed that the urease activity and subsequent NH3 generation rescued cellular respiration and prevented endogenous ROS generation ( S2 Fig ) . The ammonia generated facilitated acetate consumption through acetyl-CoA synthetase and the TCA cycle for subsequent cell growth . Collectively , these data suggest that urease is a significant component of the acid response network of S . aureus in the presence of urea . Many gram-positive species , including S . epidermidis , utilize arginine catabolism as a rapid mechanism to generate ammonia during acidic pH stress [23 , 51] . The pathway most utilized is the ADI pathway yielding ammonia , ornithine , and ATP . However , in contrast to S . epidermidis [23] ( S3C and S3D Fig ) , excess arginine was unable to remarkably rescue S . aureus during weak acid stress suggesting that ammonia generating pathways via arginine are not significantly active in our aerobic assay containing glucose ( S3A and S3B Fig ) . These results agree with a previous report documenting that the ADI pathway genes are not significantly induced under weak acid stress in S . aureus [10] . In this work we also confirmed that in S . epidermidis 1457 , additional arginine and urea provided a growth advantage under weak acid stress ( S3C and S3E Fig ) , suggesting that both ADI and urease are active in S . epidermidis . This result is consistent with another study documenting the differential transcriptional response following sapienic acid stress in S . epidermidis and S . aureus [52] . Under these growth conditions , S . aureus upregulates urease whereas S . epidermidis upregulates ADI , the NreABC nitrogen regulation system , in addition to the nitrate and nitrite reduction pathways . The lack of NH3 generation via arginine catabolism in S . aureus is not unexpected as catabolism of arginine via RocF is under the control of carbon catabolite repression and CcpA [30 , 31] . Our results suggest that when S . aureus is growing in the presence of glucose or another preferred carbon source , urease must utilize exogenous urea to facilitate pH homeostasis . Previous results from our laboratory demonstrated that when S . aureus grows in a defined medium lacking glucose , arginine is catabolized via RocF generating ornithine and urea [31] . Thus , we wanted to determine if urease was active under growth conditions where urea was generated via arginine catabolism and the medium was not acidic . These NMR experiments suggested that under neutral growth conditions , little ammonia from urea is detected ( via detection of 15N-labeled glutamine ) . In fact , the majority of urea was detected in the culture medium as it is excreted , and is potentially used as a nitrogen storage molecule . Indeed , we found that a ccpA mutant grown in TSB lacking urea was able to survive weak acid stress via derepression of global amino acid catabolism . This observation suggests that when S . aureus is growing in acidic environments where peptide and amino acids are the major carbon source , arginine catabolism and urease activity is not required to facilitate pH homeostasis , which is due to rapid catabolism of amino acids and subsequent NH3 release . Previous transcriptional analyses have suggested that ureABCEFGD is upregulated upon acid stress [10 , 11 , 53] . In the current study , we confirmed via β-galactosidase assays that the transcription of the ure genes was inhibited when the medium was buffered to a pH of 7 . 25 with MOPS ( Fig 2A ) . Moreover , we found that the transcription of the urease genes was activated by CcpA and Agr , while inhibited by CodY ( Fig 2B ) . The regulation of urease transcription by CcpA is supported by the putative catabolite-responsive element ( cre ) site identified 139 base pairs upstream of the ureA start codon [24] . Also , CcpA activation of ure transcription agrees with a previous finding that S . aureus urease , as a part of the CcpA regulon , has higher transcription as well as enzymatic activity in WT as compared to a ΔccpA mutant [24] . The significant decrease in ure gene transcription in the Δagr mutant that we observed is consistent with the transcriptional array data documenting that urease genes are upregulated by Agr [25] . The involvement of the urease genes in the Agr regulon strengthens the link between urease activity and virulence in S . aureus . Importantly , phagosomal acidification induces Agr activity , which is essential for S . aureus survival inside macrophages [54] . Under this circumstance , Agr may upregulate urease to counter acidic pH in coordination with enhanced virulence . CodY is also a global regulator that controls the expression of a variety of genes in gram-positive bacteria [55] . In particular , CodY senses the level of branched-chain amino acids and intracellular GTP and controls the transcription of many metabolic genes that are involved in amino acid synthesis , TCA cycle , and carbon overflow metabolism [56] . Our results agreed with previous reports that CodY represses urease gene transcription in Bacillus subtilis [57 , 58] and S . salivarius [26] . Although urease genes are not the direct targets of CodY in S . aureus UAMS-1 [59] , it is possible that CodY negatively regulates urease gene transcription through repressing Agr , since the agrA gene encoding the Agr response regulator is upregulated in the UAMS-1 ΔcodY mutant [59] . More in-depth future studies are required regarding the regulation of urease , as other transcriptional regulators such as Sae [60 , 61] , ClpP [62–64] , and MgrA [65] , are suggested to contribute to the regulatory network that fine-tunes urease activity . Lastly , it is interesting that CcpA activates ure transcription but represses rocF ( arginase ) transcription . These data suggest that the generation of urea by arginase is not linked to urease activity . Thus , ure transcription is activated when S . aureus is growing with a preferred carbon source such as glucose , which generates weak acids such as lactate or acetate . However , this also suggests that the urea utilized by urease must be exogeneous and not generated by arginase activity , which is only active when S . aureus is growing on non-preferred carbon sources such as peptides and amino acids . To interrogate the function of urease in vivo , we hypothesized that the kidney is a favorable niche for S . aureus colonization and resisting the host immune response for the following reasons . First , renal blood flow is about 20% of the cardiac output [66] . Thus , S . aureus has ample opportunities to invade kidney tissue during blood filtration . Second , as urea becomes concentrated when transported through renal tubules during the production of urine , the collecting ducts in the inner medulla display the highest permeability to urea [67]; hence , not only the renal tubules but also the medullary interstitium is rich in urea , providing sufficient substrates for urease . Third , kidney medullary interstitium has a low pH ( ~5 . 5 ) , comparing to the neutral cortical interstitium pH ( ~7 . 4 ) [68 , 69] . In our mouse S . aureus bacteremia model , the temporal distribution of the organ bacterial burden followed what has been previously documented ( S4 Fig ) [40 , 70 , 71] . Among all the examined organs , the kidney was the only niche that developed a chronic infection over time . On days 12 and 19 , mice inoculated with the Δure mutant had a significant decrease in CFU count in the kidneys ( Fig 5 ) , indicating a selective pressure in S . aureus to maintain urease function , similar to what has been reported in H . pylori [19] . The increased persistence of JE2 WT over Δure during chronic kidney infections demonstrated that urease functionality enhances the fitness of S . aureus within low pH and high urea environments such as the kidney . In order to determine whether differences in leukocyte recruitment are responsible for the changes in bacterial persistence between JE2 WT and Δure , individual kidneys were analyzed by flow cytometry ( S5 Fig ) . Overall , no significant changes in the leukocyte populations were observed , indicating that urease primarily enhances bacterial persistence rather than directly altering leukocyte infiltration . The reason why we chose to evaluate the leukocyte populations on day 8 was that the kidney infections started to be cleared on approximately day 12 ( Fig 5 ) . Thus the drastic differences in the bacterial burden between JE2 WT and Δure could skew the immune responses on day 12 . Further studies are required to determine the mechanism of how urease facilitates survival during infection . It is also possible that urease allows for persistence in the phagolysosomes upon phagocytosis by macrophages [72] . As macrophages are found in renal medulla [73] , S . aureus needs to employ strategies to survive within or escape from the phagolysosomes during colonization in the kidney . Moreover , the phagolysosomes are acidic in pH , which may induce urease activity for acid resistance and survival of S . aureus [74] . Indeed , anti-inflammatory macrophages , which are prevalent during late stages of S . aureus infection , produce urea via arginase-1 [75] . For the above reasons , it would be appropriate to expand our future studies to examine the function of urease in phagolysosome survival or the escape of S . aureus , especially regarding kidney macrophages . In summary , we identified that urease in S . aureus functions to facilitate pH homeostasis and survival under weak acid stress in the presence of urea; in non-acidic conditions , the endogenous urea derived from arginine is secreted extracellularly but not catabolized to fuel nitrogen metabolism . We found that urease is induced by weak acid stress and is within the regulation network that consists of CcpA , Agr , and CodY , interconnecting S . aureus stress response , metabolism , and virulence . We illustrated that urease provides a fitness advantage for S . aureus to persist during chronic kidney colonization of mice . These data all point to the conclusion that urease is not only a critical component of the acid stress response system of S . aureus , it is also an important factor in S . aureus pathogenesis .
Animal experiments were performed in ABSL2 facilities in accordance with a protocol ( #11-076-08-FC ) approved by the Institutional Animal Care and Use Committee ( IACUC ) . All animals at the University of Nebraska Medical Center are maintained in compliance with the Animal Welfare Act and the Department of Health and Human Service “Guide for the Care and Use of Laboratory Animals . ” Animals were anesthetized with ketamine and xylazine . Post injection of S . aureus retro-orbitally , all anesthesized mice were continuously monitored until they regained sternal recumbency and were capable of holding their heads up . The animals were monitored once/day on a daily basis following infection to ensure animal welfare . At all monitoring intervals , post-infection general appearance and body weights were recorded . Animals were euthanized by exposure to CO2 in a chamber ( chamber was not pre-charged ) . Animals were in the CO2 filled chamber for 5 minutes after all evidence of respiration and cardiac function was absent . CO2 was chosen as a method of euthanasia because it has a rapid anesthetic effect and quickly results in loss of consciousness and respiratory arrest . The E . coli , S . aureus , and S . epidermidis strains , plasmids , as well as primers used in this study are listed in S1 Table . E . coli cultures were grown in Luria-Bertani broth ( LB; Difco; Becton , NJ ) . S . aureus and S . epidermidis were grown in tryptic soy broth ( TSB; Difco; Becton , NJ ) containing 14 or 45 mM glucose . CDM was prepared essentially as previously described [34] , and no glucose was added . Overnight cultures grown in TSB were washed with phosphate-buffered saline ( PBS ) twice before inoculation to an optical density at 600 nm ( OD600 ) of 0 . 05 . Cultures were grown aerobically at 37°C in flasks with a 10:1 flask-to-volume ratio shaking at 250 rpm . When necessary , antibiotics were added to cultures as follows: ampicillin ( 50 μg/ml ) ; erythromycin ( 10 μg/ml ) ; tetracycline ( 10 μg/ml ) ; and chloramphenicol ( 10 μg/ml ) . Bacterial growth yield was assessed by measuring the OD600 . Culture pH was measured with a pH meter ( Mettler Toledo , Columbus , OH ) . Bacterial viability was measured as CFU/ml by serial dilutions on TSB agar plates . PCR amplifications were performed using Q5 High-Fidelity DNA polymerase ( New England Biolabs , Beverly , MA ) , Midas Mix ( Monserate Biotechnology Group , San Diego , CA ) , and oligonucleotides ( S1 Table ) synthesized by Sigma-Aldrich ( St . Louis , MO ) . Restriction endonucleases and ligase from New England Biolabs ( Beverly , MA ) were used for DNA digestion and ligation . Purification of DNA fragments prior to subsequent cloning steps was achieved by recovery from agarose gels using a DNA Clean and Concentrator-5 Kit ( Zymo Research , Orange , CA ) . Recombinant plasmids were purified using a Zyppy Plasmid Miniprep Kit ( Zymo Research , Orange , CA ) . All plasmid inserts were sequenced at Eurofins Genomics ( Louisville , KY ) to ensure the absence of mutations . The reporter plasmid pNF315 was constructed by amplifying the intergenic region upstream of ureA with primers 2833 and 2835 so that the native ribosomal binding site ( RBS ) was replaced with a plasmid-encoded RBS . The DNA fragment was digested and ligated into the BamHI and XhoI sites of the vector plasmid pJB185 , which contains a promoterless lacZ [76] . pNF315 was electroporated into S . aureus RN4220 and subsequently transduced into JE2 strains using bacteriophage Φ11-mediated transduction [77] . To create the markerless JE2 Δure mutant , the allelic exchange plasmid pNF320 was generated by inserting the DNA sequences 1 kb upstream and 1 kb downstream of the ureABCEFGD operon into the temperature-sensitive E . coli-S . aureus shuttle vector plasmid pJB38 [78] , using a NEBuilder HiFi DNA Assembly Cloning Kit ( New England Biolabs , Beverly , MA ) , with primers 2980 and 2983 , as well as primers 2986 and 2987 . pNF320 was electroporated into S . aureus RN4220 and subsequently transduced into JE2 WT using bacteriophage Φ11-mediated transduction . Once the plasmid pNF320 was introduced into JE2 , the allelic replacement to introduce the deletion mutation into the S . aureus chromosome was performed as previously described [79] . The deletion of the urease operon was confirmed phenotypically by plating on a Christensen's urea agar plate and by PCR using primers 2984 and 2985 . Chromosomally complementation of Δure was performed as previously described [80] . Briefly , plasmid pNF363 was constructed containing ureABCEFGD genes with their native promoter by amplifying an approximately 5 . 5 kb region from the JE2 genome using primers 2991 and 3306 . The resulting DNA fragment was inserted into BamHI and PstI sites of the shuttle vector pJC1111 yielding pNF363 . pJC1111 and pNF363 were subsequently transformed into RN9011 for chromosomal integration . Φ11 mediated transduction was performed to move the integrated pJC1111 and pNF363 into both JE2 WT and JE2 ureB::ΦΝΣ . For all metabolite assays , 1 ml bacterial culture was collected and pelleted for 2 . 5 min at 15 , 000 rpm . The supernatant was collected and stored at -80°C until use . Glucose , acetate , urea , and ammonia concentrations were determined using commercial kits ( R-Biopharm AG , Darmstadt , Germany ) according to the manufacturer's instructions . As previously described [31 , 36] , five independent 50 ml cultures of S . aureus JE2 WT and the Δure mutant were grown to stationary phase ( OD600 = 1 . 9 ) in CDM containing 15N2-labeled arginine ( Isotec , Sigma-Aldrich , Miamisburg , OH ) or 15N-labeled serine ( Isotec , Sigma-Aldrich , Miamisburg , OH ) . For each culture , a total OD600 of 40 was collected and pelleted by centrifugation at 4000 rpm for 5 min at 4°C . 2 ml culture supernatant was collected as the media sample . Pellets were washed with 10 ml of cold sterile water twice and resuspended in 1 ml cold sterile water . The cells were lysed using a bead ruptor ( OMNI International , Kennesaw , GA ) and centrifuged at 15 , 000 rpm for 15 min at 4°C . The pellet was re-extracted with 1 ml cold sterile water . The combined cell lysate supernatant from both extractions , as well as the culture supernatant , were snap frozen in liquid nitrogen and lyophilized using a FreeZone freeze dryer . The data collection and analysis of NMR was conducted as previously described [36] . A Bruker AVANCE IIIHD 700 MHz spectrometer equipped with a 5 mm quadruple resonance QCI-P cryoprobe ( 1H , 13C , 15N , and 31P ) , an automatic tune and match system ( ATM ) , and a SampleJet automated sample changer system with Bruker ICON-NMR software were utilized . The 2D 1H−15N HSQC spectra collected for S . aureus cell lysates and culture media were assigned using a database of 2D 1H−15N HSQC reference spectra for known metabolites [36] . A chemical shift tolerance of 0 . 08 ppm for 1H and 0 . 25 ppm for 15N were used to match metabolites to our reference database . Beta-galactosidase assays were performed essentially as previously described [81] . Briefly , overnight cultures of JE2/pNF315 grown in TSB were inoculated in TSB containing 45 mM glucose , TSB containing 45 mM glucose buffered with 100 mM MOPS , TSB containing 45 mM glucose with 10 mM urea , and TSB containing 45 mM glucose with 10 mM urea buffered with 100 mM MOPS . At 2 h and 6 h , 2 ml and 0 . 5 ml of cells were collected and centrifuged ( Fig 3A ) . Additionally , overnight cultures of JE2/pNF315 , ΔccpA/pNF315 , Δagr/pNF315 , ΔcodY/pNF315 grown in TSB were inoculated to TSB containing 45 mM glucose and 10 mM urea . At 2 h , 6 h , and 10 h , 2 ml , 0 . 5 ml , and 0 . 5 ml of cells were collected and centrifuged respectively ( Fig 3D ) . The cell pellets were resuspended in 1 . 2 ml Z-buffer ( 60 mM Na2HPO4 , 40 mM NaH2PO4 , 10 mM KCl , 1 mM MgSO4 , 50 mM β-mercaptoethanol , pH 7 . 0 ) and lysed with a bead ruptor ( OMNI International , Kennesaw , GA ) . 700 μl supernatant of the cell lysate was collected , and 140 μl of 4 mg/ml ortho-nitrophenyl-β-galactoside ( ONPG ) was added . The samples were incubated at 37°C until the color turned slightly yellow ( under OD420 1 . 0 ) . 200 μl of 1 M Na2CO3 was added to stop the reaction . Protein concentrations were determined by Bradford assays using the Protein Assay Dye Solution ( Bio-Rad , Hercules , California ) . Absorbances at 420 nm and 550 nm were measured with an Infinite 200 plate reader ( Tecan , Männedorf , Switzerland ) . Overnight cultures of JE2 WT and ΔccpA were inoculated to an OD600 of 0 . 05 in TSB containing 45 mM glucose . At 0 h , 3 h , 6 h , 9 h , and 12 h , 0 . 5 ml culture was collected and pelleted for 3 min at 15 , 000 rpm . Supernatant was collected and filtered through the Pierce Protein Concentrators ( 3 , 000 molecular weight cutoff; Thermo Scientific , Rockford , IL ) according to the manufacturer’s instructions . Amino acid analysis was performed with a Hitachi L-8800 amino acid analyzer by the Protein Structure Core Facility , University of Nebraska Medical Center . The flow cytometry analyses following the growth assays were performed as previously described [7] , using a BD LSRII flow cytometer ( Becton and Dickinson , San Jose , California ) . Cells collected at 24 h and 72 h from the growth assay where JE2 WT and ureB::ΦΝΣ were cultured in TSB containing 45 mM glucose and 10 mM urea . Cell samples were washed with PBS to a final concentration of 107 cells/ml and stained with 5-cyano-2 , 3-ditolyl tetrazolium chloride ( CTC , 5 mM ) and 3- ( p-hydroxyphenyl ) fluorescein ( HPF , 15 μM ) . The fluorescence-activated cell sorting ( FACS ) was performed at a flow rate of ∼1 , 000 cells per second with 10 , 000 events per sample . Samples were excited at 488 nm , with HPF emission being detected at 530±30 nm , and CTC emission being detected at 695±40 nm . The FlowJo software was used to analyze the raw data . For the flow cytometry analyses following the animal experiments , kidneys were collected in 1 . 0 mL of FACS buffer , which was composed of PBS and 2% heat-inactivated fetal bovine serum ( FBS ) . Kidneys were homogenized with the blunt end of a 3 . 0 mL syringe and filtered through a 70 μm filter ( BD Falcon , BD Biosciences ) . Filtrate was washed with PBS and collected by centrifugation ( 300 x g , 5 min ) , whereupon the filtrate was digested with Collagenase A and DNase while mixing at 37°C . The reaction was stopped after 15 minutes with heat-inactivated FBS on ice , filtered , and washed with FACS buffer , whereupon red blood cells ( RBC ) were lysed using the RBC Lysis Buffer ( BioLegend , San Diego , CA ) . Single cell suspensions were washed and resuspended in FACS buffer and incubated with TruStain fcX ( BioLegend , San Diego , CA ) to minimize non-specific antibody binding . Samples were divided in two to analyze innate immune cell ( MDSCs , neutrophils , monocytes , and macrophages ) populations and lymphocyte ( CD3 , CD4 , CD8 , and γδ T cells ) populations separately . Both samples were stained with Live/Dead Fixable Blue Dead Cell Stain ( Invitrogen , Eugene , OR ) . Innate immune cells were stained with CD45-APC , Ly6G-PE , Ly6C-PerCP-Cy5 . 5 , and F4/80-PE-Cy7 , CD11b-FITC ( BioLegend , San Diego , CA ) . Lymphocytes were stained with CD45-PE-Cy7 , CD3-APC , CD4-PacBlue , CD8-FITC , and γδTCR-PE . An aliquot of pooled cells was stained with isotype-matched control antibodies to assess the degree of non-specific staining per treatment group [82] . For individual samples , 10 , 000–100 , 000 events were analyzed using BD FACSDiva software with cell populations expressed as percentage of total viable CD45+ leukocytes . Seven-week-old male and female C57BL/6 mice ( Charles River Laboratories , Wilmington , MA ) were used in all animal experiments . Overnight cultures of S . aureus JE2 WT and Δure in TSB were washed with PBS twice and suspended in PBS to yield an OD600 of 10 . The cultures were further diluted 1:50 with PBS , prior to the retro-orbital injection of 50 μl ( 106 CFU ) final bacterial suspension . The inocula were verified by serial dilution plating and colony enumeration on TSB agar plates . Mice were anesthetized by intraperitoneal injection of ketamine/xylazine ( 60 mg/kg and 3 mg/kg , respectively ) . Mice were euthanized for the quantification of bacterial burden ( expressed as Log [ ( CFU/g of tissue ) +1] ) , by serial dilution plating and colony enumeration of homogenized organs . For all studies , statistical analysis was performed using GraphPad Prism 5 . 0 software ( La Jolla , CA ) . P-values < 0 . 05 were considered significant . For comparisons of two groups , Mann-Whitney test was used . One-way analysis of variance ( ANOVA ) was performed to compare three or more groups . Two-way repeated measures ANOVA was performed to compare differences between groups with two independent variables .
|
Urease has been reported to be crucial to bacteria in environmental adaptation , virulence , and defense against host immunity . Although the function of urease in S . aureus is not clear , recent evidence suggests that urease is important for acid resistance in various niches . Our study deciphered a function of S . aureus urease both in laboratory conditions and during host colonization . Furthermore , we uncovered the major components of the regulatory system that fine-tunes the expression of urease . Collectively , this study established the dual function of urease which serves as a significant part of the S . aureus acid response while also serving as an enzyme required for persistent kidney infections and potential subsequent staphylococcal metastasis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"urea",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"ureases",
"enzymes",
"pathogens",
"enzymology",
"microbiology",
"carbohydrates",
"staphylococcus",
"aureus",
"organic",
"compounds",
"glucose",
"basic",
"amino",
"acids",
"amino",
"acids",
"kidneys",
"bacteria",
"bacterial",
"pathogens",
"proteins",
"staphylococcus",
"medical",
"microbiology",
"microbial",
"pathogens",
"chemistry",
"biochemistry",
"ammonia",
"arginine",
"organic",
"chemistry",
"anatomy",
"monosaccharides",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"renal",
"system",
"catabolism",
"metabolism",
"organisms"
] |
2019
|
Urease is an essential component of the acid response network of Staphylococcus aureus and is required for a persistent murine kidney infection
|
The packaging and organization of genomic DNA into chromatin represents an additional regulatory layer of gene expression , with specific nucleosome positions that restrict the accessibility of regulatory DNA elements . The mechanisms that position nucleosomes in vivo are thought to depend on the biophysical properties of the histones , sequence patterns , like phased di-nucleotide repeats and the architecture of the histone octamer that folds DNA in 1 . 65 tight turns . Comparative studies of human and P . falciparum histones reveal that the latter have a strongly reduced ability to recognize internal sequence dependent nucleosome positioning signals . In contrast , the nucleosomes are positioned by AT-repeat sequences flanking nucleosomes in vivo and in vitro . Further , the strong sequence variations in the plasmodium histones , compared to other mammalian histones , do not present adaptations to its AT-rich genome . Human and parasite histones bind with higher affinity to GC-rich DNA and with lower affinity to AT-rich DNA . However , the plasmodium nucleosomes are overall less stable , with increased temperature induced mobility , decreased salt stability of the histones H2A and H2B and considerable reduced binding affinity to GC-rich DNA , as compared with the human nucleosomes . In addition , we show that plasmodium histone octamers form the shortest known nucleosome repeat length ( 155bp ) in vitro and in vivo . Our data suggest that the biochemical properties of the parasite histones are distinct from the typical characteristics of other eukaryotic histones and these properties reflect the increased accessibility of the P . falciparum genome .
The human malaria parasite , Plasmodium falciparum , yearly responsible for an estimated 600 , 000 deaths ( WHO Report 2014 ) , has the AT-richest genome sequenced to date . The AT-content averages 80 . 6% genome wide , but reaches up to 90% in introns and intergenic regions [1] . P . falciparum shows a complex life cycle in two hosts , exhibiting dramatic changes in the gene expression program . At least 60% of the genome is transcriptionally active during erythrocytic development with gene expression being activated in form of a cascade and tightly regulated during developmental stage transition [2 , 3] . Like other eukaryotic genomes , the plasmodium forms nucleosomes , possesses chromatin modifiers , chromatin remodeling activities and potentially active DNA methyltransferases [4] . The nucleosome core is composed of 147bp of DNA wrapped 1 . 65 turns around the histone octamer , consisting of two copies of each H2A , H2B , H3 , and H4 [5] . In organizing the nucleosome , H3 homo-dimerizes using the C-terminal ends and heterodimerizes with H4 to form a H3–H4 tetramer [6] . Histone dimers composed either of H2A and H2B , or H3 and H4 organize each 30bp of DNA . Two H3-H4 dimers bind to the central 60bp of nucleosomal DNA and each H2A-H2B dimer organizes 30bp towards the end of the particle [5] . DNA binding and distortion is brought about by the interaction of the histones with the minor groove of the DNA at 14 independent DNA binding regions , termed super helix locations ( SHL ) . Almost 400 direct and indirect histone-DNA interactions render the nucleosome one of the most stable protein-DNA complexes under physiological conditions [5] . The observation that specific DNA sequences favor the formation of nucleosomes in vitro and in vivo , correlates well with the important role of positioned nucleosomes in organizing the chromatin landscape to regulate gene expression [7–9] . Also for plasmodium it was shown that proper promoter functioning and regulation of var gene expression requires the presence of positioned nucleosomes [10–13] . High-throughput sequencing analyses of nucleosome positions and chromatin structure analyses in different life cycle stages correlate changes in chromatin structure with the regulation of gene expression [14–16] . Interestingly , the chromatin structure of P . falciparum is distinct from other eukaryotes in that the genome is surprisingly accessible containing poorly positioned nucleosomes [14–17] . These effects are suggested to be related to the extremely high AT-content ( 81% ) , generating an inherently inflexible DNA molecule , reducing the potential to form positioned nucleosomes [18 , 19] . Due to the central role of the histones in the cell , these proteins have been highly conserved throughout eukaryotic evolution [20] . P . falciparum possesses the most divergent histones in sequence , with identities of only 64% , 67 , 7% , 92 , 2% and 92 , 6% between human and plasmodium H2A , H2B , H4 and H3 . This difference may well reflect adaptations for gene regulation and potentially present an adaption to the AT-richness of the genome . In our study , we show that recombinant plasmodium histones , like human histones , bind poorly to AT-rich DNA . Histone sequence variations result in reduced histone octamer stability and the nucleosomal arrays form the shortest known nucleosome repeat length , measuring only about 155bp . Most interestingly , the octamers lost the capability to recognize intrinsic DNA encoded nucleosome positioning sequences , challenging the current view that DNA structure and di-nucleotide repeats determine translational and rotational nucleosome positioning . Albeit , we observed that the few positioned nucleosomes in Plasmodium falciparum are flanked by long AT-repeats in the DNA linker sequences and show that these sequences serve to guide nucleosome positioning . The biochemical properties of the histones mirror the organization of the plasmodium chromatin structure in vivo and we show that in contrast to other eukaryotes , long nucleosome flanking , AT-rich sequence elements are required for their positioning .
In order to study the functional effects of the distinct P . falciparum histone sequences , we first analyzed the location of the amino acid differences with respect to nucleosome structure ( Fig 1A and 1B ) . The amino acid differences in H3 and H4 do not affect the sites of direct histone-DNA interaction . H2A and H2B exhibit the majority of differences in their N-terminal tails and the H2A C-terminal tail . In addition clusters of amino acid differences occur in the regions of the superhelix locations ( SHL ) , the regions forming the L1L2 loops , and the α1α2 DNA-binding motifs [21] . The amino acid differences at the SHL3 . 5 and 4 . 5 regions of H2B and H2A affect direct histone-DNA interactions; thereby they could alter nucleosome stability ( Fig 1A ) . In contrast , the variability in the flexible histone tails are suggested not to contribute to complex stability [21] . We used the available nucleosome structure , in the absence of the flexible histone tails ( PDB entry 3AFA [22] ) to model the strength of the histone-DNA interactions in silico . Compared to the full-length sequence , the first 15 and last 13 residues of H2A and the first 27 of H2B , 42 of H3 , and 23 residues of H4 were missing . A homology model for the plasmodium nucleosome was built based on the human nucleosome structure ( PDB entry 3AFA ) . The fact that 84% of all amino acid residues in the human nucleosome are identical ensures a high quality 3D-model of the plasmodium nucleosome . Molecular dynamics simulations were performed and DNA-protein interactions for the full complex as well as for individual residues were scored . The histone- and species-specific differences in binding energy were determined by subtracting for each snapshot the score of the plasmodial DNA-histone interaction from the median score calculated for the human DNA-histone interaction ( Fig 1C ) . Only H2A#1 showed a slightly stronger DNA binding in human histones , which we do not consider significant for the following reasons: The reported accuracy of FoldX is 0 . 46 kcal/mol [23] , which is the standard deviation of the difference between ΔΔGs calculated by FoldX and the experimental values . Human H2A sequences differ from plasmodial H2A sequences by 24 residues in the modelled core region , while 14 interacted with DNA in our analysis ( see below and S1 Fig ) . The median difference in H2A#1 energies is 2 . 03 kcal/mol . Thus , the mean contribution of each mutation is 2 . 03 kcal/mol / 14 = 0 . 15 kcal/mol , which is below the reported accuracy . The mean contributions of each mutation are even smaller for the other histones and thus considered as a neutral effect . Therefore , we suggest similar DNA binding for the core regions of all corresponding human and plasmodial histones . For individual residues , π-π stacking , cation-π stacking , contacts , hydrophobic interactions , and hydrogen-bond networks were assessed based on the outcome of YASARA [24] . π-π stacking did not contribute noticeable to DNA-protein interactions and the comparison of all other residue-specific scores did not indicate striking differences ( S1 Fig ) . In summary , the in silico modelling suggests a highly similar strength of the interactions between human or plasmodial histone cores and DNA , both in total and on a per residue basis . Histones were expressed in bacteria , purified octamers were reconstituted and used for nucleosome assembly ( Fig 2 ) . The positions of nucleosomes on DNA can change as a result of thermally induced nucleosome sliding [25 , 26] . Temperature induced nucleosome sliding exhibits the tendency of moving the nucleosomes to thermodynamically more stable positions , which depend on the DNA sequence [26 , 27] . We use this method to compare the human and plasmodium nucleosome stability , when assembled on a DNA fragment containing the 601 nucleosome positioning sequence [28] . Human histone octamers were reconstituted on the Cy5 labeled DNA fragment , whereas the plasmodium octamers were separately reconstituted on the Cy3 labelled DNA fragment ( Fig 2B ) . Plasmodium nucleosomes like human nucleosomes recognize and are specifically positioned on the artificial 601 nucleosome positioning sequence . Both , parasite and human nucleosomes , form a defined nucleoprotein particle covering a single position on the 208bp long DNA fragment ( S2A Fig ) . The Cy3 and Cy5 labelled substrates were mixed to allow an internally controlled experiment , addressing nucleosome mobility with increasing temperatures ( Fig 2B ) . Changes in nucleosome positions were analyzed on native polyacrylamide gels , scanning the individual fluorescence channels , followed by ethidium bromide staining to visualize the mixture of human and plasmodium nucleosomes . In contrast to the molecular dynamics simulation , the parasite octamers exhibited increased nucleosome sliding activity at elevated temperatures , when compared to the human nucleosomes . The result is suggesting weaker histone-DNA interactions within the plasmodium nucleosomes . As the 601 DNA presents an artificial , GC-rich sequence ( 56 , 9% GC ) , and the histones may be evolved to bind AT-rich DNA , we repeated the experiment with the AT-rich KahrP sequence amplified from the P . falciparum genome ( 14 , 6%; S2B Fig ) . Nucleosomes are formed on several locations on the DNA fragment resulting in a more fuzzy EMSA pattern and the pattern is distinct from human nucleosomes ( addressed below ) . However , with this nucleosomal template thermal mobilization could be observed as well . Again , the plasmodium histones showed an increased thermal mobility , starting at lower temperatures ( S2B Fig ) , suggesting a reduced stability of histone–DNA interactions . Nucleosomes were shifted to slower migrating bands in the mobility shift assay , suggesting a relocation of the histone octamers to more central positions [29] . Still , to rule out that changes in band-shift position are not due to histone loss , we isolated nucleosomal bands from the gel and analyzed them for the equimolar histone ratio . Fuzzy nucleosomes and additional bands of the 601 nucleosomes contain the complement of all four histones ( S2C Fig and Fig 5D and 5E ) , suggesting different nucleosome positions , rather than histone loss [25] . In order to test the stability of histone-DNA interactions we incubated the reconstituted nucleosomes with increasing concentrations of chloroquine and monitored the release of free DNA . Chloroquine is a DNA intercalating drug and its binding results in changing the geometry of DNA and stiffening the molecule . DNA bound proteins compete with chloroquine for binding , depending on their interaction strength . Proteins with relatively lower DNA binding affinity are displaced at correspondingly lower chloroquine concentrations [30] . Human and plasmodium nucleosomes were incubated with increasing concentrations of chloroquine , revealing the appearance of free DNA and the disruption of the plasmodium nucleosomes at 1mM chloroquine , whereas at least three times higher concentrations are required to initiate the disruption of the human nucleosomes ( Fig 2C ) . To analyze whether all histones or only a subset of the 4 histones reveal reduced histone-DNA interaction stability , we incubated nucleosomal arrays with increasing salt concentrations . The histones H2A and H2B can be dissociated from the particle starting at concentrations of 0 . 7M NaCl and above 1 . 2M NaCl also H3 and H4 start to dissociate from DNA [31 , 32] . Nucleosomal arrays were reconstituted on biotinylated DNA , bound to magnetic beads , incubated with increasing salt concentrations and the supernatants were collected ( Fig 2D ) . Salt eluted proteins were analyzed by SDS-PAGE and visualized by silver staining . Correlating with the high number of amino acid changes in H2A and H2B , these two proteins exhibit significantly reduced binding affinities towards the DNA and the H3-H4 tetramer . At salt concentrations of about 0 . 7M NaCl , both , plasmodium H2A and H2B , were quantitatively eluted from the nucleoprotein particle , whereas their human counterparts still remained associated . In contrast , the plasmodium H3 and H4 proteins are stably bound to DNA , like the human H3/H4 proteins , even at high NaCl concentrations . Taken together the results show an overall decreased nucleosome stability , as compared to the human nucleosomes , due to the weaker interactions of H2A and H2B , resulting in the increased mobility of the nucleosomes on DNA . The plasmodium genome has an average of 80 . 6% AT content and poly dA:dT sequences form straight and rigid helical structures with the potential to exclude nucleosomes [33–35] . It is shown that nucleosomes tend to form over GC-rich DNA with specific dinucleotide phasing [36–38] . To test whether plasmodium histones adapted to preferentially bind to AT-rich DNA , we fragmented plasmodium and human genomic DNA by sonification to create DNA fragments with a mean length of 200–300bp for nucleosome reconstitution experiments . The individual DNA samples were chemically labelled with fluorescent dyes , Cy5 for the human DNA and Cy3 for the plasmodium genomic DNA . The human and plasmodium genomic DNAs were mixed at equimolar ratios and reconstituted into nucleosomes by salt dialysis , using either plasmodium or human histone octamers ( Fig 3A , lanes 1–5 ) . The competitive assembly reaction revealed that the human GC-rich DNA was reconstituted more efficiently , by both , the human and plasmodium histones . The assembly efficiency was similar for human and plasmodium histones , showing no preference of the parasite histones for the AT-rich genomic DNA . In order to directly compare AT-rich plasmodium DNA with selected GC-rich DNA , we used the DNA sequence of the human ribosomal gene with a mean GC content of 60% ( lanes 6 to 8 ) . Again , both , the plasmodium and human histones bound the GC-rich DNA with higher affinity , fully assembling the GC rich DNA and only partially binding to plasmodium genomic DNA . Finally , we tested whether the high GC-content is required throughout the nucleosomal DNA for efficient nucleosome assembly . Three different PCR fragments , containing 147 bp , AT-rich ( 85 , 6% ) DNA , flanked by DNA linkers with GC-content and increasing length , were created ( S3A Fig ) . The parasite and human histone octamers again behave similar in that all three DNA molecules are reconstituted into nucleosomes with similar affinity . The result suggests that GC-rich DNA sequences at the border of the nucleosomes , where the H2A and H2B proteins contact DNA , are not sufficient to increase the GC-dependent binding affinity . Next , we had a closer look at the available high throughput sequencing data and re-analyzed the datasets studying the accessibility of chromatin ( Data sets used: SRX013309 [14] , SRX013302 [14] , SRX885811-SRX885819 [16] ) . Ponts and colleagues performed FAIRE assays [39] to reveal the accessible genomic regions , relative to their MNase resistant nucleosomal fraction . A genome browser snapshot of the FAIRE data , nucleosome occupancy and the GC content is given in S4A Fig . The accessible chromatin regions ( FAIRE: median GC-content 15%; 19377 nucleosome free regions , blue line ) and the sequences occupied by nucleosomes were quantified ( nucleosomal: median GC-content 29% , 22770 nucleosome positions identified; red line ) . As control we used the same number of randomly chosen genomic regions ( median GC-content 19%; grey line; S4B Fig ) . The data shows that intergenic regions are highly accessible , giving rise to large FAIRE domains , whereas the nucleosomes occupy the genic regions . The data suggests that nucleosomes are preferentially formed , or remain stable on GC-rich sequences that are mainly located in exonic regions . In contrast , nucleosomes are preferentially depleted from the AT-rich intronic and intergenic regions ( S4C Fig ) . The FAIRE and MNase assays are complementary , suggesting more accessible chromatin at intergenic regions . However , these results are based on highly over-digested chromatin and it is known that MNase exhibits a strong preference for AT-rich sequences [40] . A recent genome wide study shows that nucleosomes in the intergenic regions are not depleted , but potentially disappear by over-digestion of chromatin with MNase [16] . Analyzing the new , low digested dataset , still revealed higher nucleosome occupancy at GC-rich sequences ( S4D and S4E Fig ) . The enrichment of nucleosomes may reflect and correlate with the higher binding affinity of the nucleosomes towards GC-rich sequences ( Data sets used: SRX885814 [16] ) , being an inherent feature of the parasite histones . Apparently , the Plasmodium falciparum histones did not evolve to efficiently package the AT-rich genome . Using the chloroquine intercalation assay we now directly tested the stability of plasmodium and human nucleosomes on AT- and GC-rich DNA . Human and plasmodium nucleosomes were reconstituted on three different DNAs of varying AT-content , length and distinct fluorescent labels to monitor their chloroquine stability in a single reaction to allow direct comparison ( Fig 3B; S3B Fig ) . The DNA molecules ( a mixture of AT- and GC-rich DNA in Fig 3B; a 210 bp long AT-rich DNA in S3B Fig ) were individually reconstituted into nucleosomes and then mixed , in order to perform an internally controlled competition assay to monitor the appearance of free DNA by chloroquine mediated nucleosome disruption . Interestingly human and plasmodium nucleosomes are similarly stable on the AT-rich DNA , giving rise to small amounts of disrupted nucleosomes with increasing chloroquine concentrations . However , on the GC-rich DNA different results were obtained , as revealed by the appearance of the free GC-rich DNA at lower chloroquine concentrations , as compared to the AT-rich DNA ( Fig 3B ) . Plasmodium histones are less stably bound to GC-rich DNA than human nucleosomes , suggesting that changes in the plasmodium histone sequence result in nucleosome de-stabilization . Cellular nucleosomes are mobilized by ATP dependent chromatin remodeling enzymes that use the energy of ATP hydrolysis to move the histone octamer on the underlying DNA [41 , 42] . Next , we tested whether the increased mobility and reduced stability of plasmodium nucleosomes would also affect their dynamics in an enzyme catalyzed remodeling reaction . Nucleosome remodeling reactions were performed with recombinant remodeling enzymes from the Chd and ISWI families; i . e . Chd3 and Snf2H ( Fig 4A ) . Again , Cy3 and Cy5 labeled 601 DNA , either reconstituted with the plasmodium or the human nucleosomes were mixed in one test-tube to allow the direct comparison of the remodeling kinetics . Chd3 ( Fig 4A , lanes 1 to 9 ) exhibited very similar remodeling kinetics on both types of substrate . The nucleosomes positioned at the border of the DNA fragment were partially moved to the center of the DNA fragment with no apparent difference of positioning and remodeling efficiency . To our surprise , the human ISWI type remodeler Snf2H did move the parasite nucleosomes less efficient than the human nucleosomes ( Fig 4A , lanes 10 to 18 ) . The less stable plasmodium nucleosome is a worse substrate for Snf2H , suggesting that Snf2H either binds the nucleosome with lower affinity , or nucleoprotein structures are lacking that are required for Snf2H-dependent remodeling . We first tested the binding affinity of Snf2H towards the Cy5-plasmodium and Cy3-human nucleosomes in competitive bandshift analysis ( S5A Fig ) , where we did not observe differences . Next , we had a closer look at the histone H4 tail . ISWI type remodeling enzymes strictly depend on the intact H4 tail and more detailed on the amino acids R17-H18-R19 of the H4 tail [43 , 44] . A closer inspection of the plasmodium histone H4 revealed the presence of the RHR motif . However , at position 21 plasmodium H4 exhibits a V to I change in sequence that could influence remodeling efficiency ( S5B Fig ) . In order to test if this is the case , we prepared hybrid octamers ( H4hyb ) containing the human H4 in combination with the plasmodium histones H2A , H2B and H3 ( Fig 1A , lane 8 ) . Hybrid nucleosomes were compared in the competitive remodeling assays with the human nucleosomes , showing that the amino acid exchange at position 21 affects the recognition of the histone H4 tail ( S5C Fig ) . Remodeling is similar efficient when comparing the hybrid nucleosomes with the human counterpart , suggesting that the reduced stability of the parasite nucleosomes does not automatically increase the ATP dependent nucleosome remodeling rate . Nucleosomes form arrays with defined linker lengths on DNA . The length of the DNA linker is cell type specific in higher eukaryotes and also reveals organism-specific inter-nucleosomal distances [45] . Partial MNase digestions of native P . falciparum chromatin exhibits a nucleosomal ladder , revealing the regular array of nucleosomes on DNA [11 , 46] . We tested whether the parasite histones would form similar nucleosomal arrays as the human histones , when reconstituted on circular , supercoiled DNA by the salt dialysis method ( Fig 4B ) . Plasmodium histones form nucleosomal ladders , but interestingly , the plasmodium nucleosomal ladder exhibited repeatedly a more smeary appearance , but still revealing the repetitive pattern that corresponds to the array-form . The smeary appearance could be explained by a reduced nucleosomal stability , corresponding to increased MNase sensitivity of the nucleosome core , and/or an increased heterogeneity of DNA linker lengths . However , in contrast to the spacing of the human nucleosomal array , the plasmodium nucleosomes obey significantly closer inter-nucleosomal spacing , with a mean distance of about 155bp . To rule out effects of reduced nucleosomal stability , associated with the generation of DNA fragments of sub-nucleosomal size , we performed extended MNase digestions and detailed DNA fragment length analysis ( S6A and S6B Fig ) . Prolonged MNase digestions gave rise to stable digestion intermediates of about 150bp in size ( S6B Fig ) and bioinformatic analysis , of the mono-nucleosomal DNA fragment size distribution in the Kensche data set , did also reveal a bona fide protection of about 150 bp of DNA by the histone octamers ( S6C Fig ) . Even though our assays suggest a reduced stability of the plasmodium nucleosome , the histone octamer does stably protect the 147bp nucleosome core sequence from MNase digestion , as shown for the human nucleosome core . Next , we used the experimental data of Kensche and colleagues to mine for di-nucleosomal DNA fragments that are abundant due to their low MNase digestion regimen . In order to have internal controls , we applied the bioinformatic analysis to all 8 erythrocytic stages and extracted the plasmodium and human DNA fragments from the data . The fragment sizes were plotted and di-nucleosomal fragment lengths were calculated ( Fig 4C ) . Indeed , we find the same short inter-nucleosome distances in vivo as in vitro , and longer nucleosome repeat lengths ( NRL ) for the human nucleosomes . Our in vitro analysis and the in vivo data show that the NRL is even shorter than in yeast , the organism with the shortest repeat length known so far . Taken together our data suggests that the global NRL distribution in the genome is driven by the biochemical properties of the histones and not by additional cellular activities . Interestingly , experimental and modeling studies have shown that short NRLs ( up to 177bp ) inhibit the folding into higher order structures of chromatin , being in well agreement with the accessible chromatin of P . falciparum [47] . DNA sequence directs nucleosome positions in vivo , as the histone octamer does preferentially bind with a ~10bp periodicity to anti-phased A/T and G/C dinucleotides , corresponding to the helical turn of DNA wrapped around the histone core [9 , 36 , 48] . As documented above , we used the 601 DNA sequence , representing an artificial high affinity binding site for nucleosomes , to prepare nucleosomes with defined positions ( Fig 2B ) . However , we noticed that on natural DNA sequences no discrete positioning is obtained and that the reconstitution pattern deviates from the human nucleosomes . To unravel the underlying reason , we used a set of native mouse ( rDNA promoter -190/+90 ) , D . melanogater ( HSP70 ) and genomic P . falciparum sequences ( Pf3D7v3: 1307500–1308399; S7A Fig ) to visualize and study nucleosome positioning . DNA fragments were reconstituted into nucleosomes , using either recombinant human or parasite histone octamers . As expected , the human histone octamers form discrete nucleosome positioning patterns on all DNA templates used , being visible as specific bands in the native polyacrylamide gels ( Fig 5A–5C ) . Multiple bands arise from different , specific mono-nucleosomal positions on the given DNA molecules . As previously shown , the nucleosomal patterns do depend on the sequence and structure of the DNA molecule [49] . In contrast , plasmodium nucleosomes exhibit a smeary appearance , with only a few prominent bands in the electromobility shift assay . The results suggest that the majority of the octamers do not assemble on discrete , preferential sites , but are randomly distributed along the DNA . Discrete bands are most often the lowest bands , corresponding to a nucleosome covering the thermodynamically stable end position of the DNA [49] . The loss of nucleosome positioning is apparently driven by the changes in the amino acid sequences of the plasmodium histones , but independent of histone H4 , as shown by using a plasmodium hybrid octamer , carrying the human histone H4 ( Fig 5A and 5B ) . The fuzziness of the assembly reaction is most probably not an effect of weaker histone DNA interactions and DNA breathing in the nucleosome , as MNase digestions show that the DNA entry/exit sites of the nucleosomes are stably protected from nuclease cleavage ( S6A and S6B Fig ) . These experiments do also reveal the lack of detectable sub-nucleosomal , or non-canonical histone-DNA complexes . In addition , plasmodium histone octamers form intact nucleosomes on DNA , as we have shown by assembly on the 601 DNA , and these nucleosomes are properly structured and stoichiometric , as revealed by the enzymatic nucleosome remodeling assays ( Fig 2B; S2C and S4A Figs ) . In order to show that this is also the case for the experiments shown here and to exclude nucleosomal dis-integration to be responsible for the lack of positioning , we analyzed the histone content of the reconstituted nucleosomes ( Fig 5D and 5E ) . The biotinylated HSP70 DNA was reconstituted into nucleosomes ( Fig 5D ) and the bound histones were eluted from DNA , after binding it to magnetic beads . Even though there is no clear nucleosomal pattern , all core histones are present in stoichiometric amounts ( Fig 5E ) , suggesting the formation of bona fide nucleosomes . This result was also confirmed by the salt elution experiment performed with the nucleosomal arrays ( Fig 2D ) . Our observation is supported by in vivo data , when analyzing the plasmodium nucleosome positions from the different genome wide studies . The high throughput sequencing dataset of Bunnik and colleagues contains contaminations of human nucleosomal DNA ( 3 . 2 mio reads ) in addition to the plasmodium DNA sequences , serving us as internal controls for the bioinformatic analysis [15 , 50] . We selected nucleosomal DNA fragments , ranging in size from 146 to 148bp , and analyzed their nucleotide frequency along the DNA path ( 61 , 263 plasmodium and 121 , 819 human sequences ) . As shown in previous studies , the human nucleosome positions exhibit an enrichment in G/C every 10bp and shifted by 5bp an enrichment in A/T di-nucleotides [9 , 51] ( S7B Fig ) , revealing sequence dependent nucleosome positioning signals . In contrast , this periodicity is lacking in the DNA sequences of the plasmodium nucleosomes , suggesting that the plasmodium octamers do not obey the same positioning rules as the human histones . Like in our in vitro positioning experiments , the in vivo analysis shows that the parasite histones do not recognize the sequence dependent nucleosome positioning signals . As this dataset was derived from heavily digested chromatin , we also re-analyzed the dataset of Kensche and colleagues with our bioinformatics pipeline [16] . These authors show that nucleosome positioning can be observed at regulatory regions , but is rarely detected in the intergenic regions [16] , being consistent with our experimental data . For a detailed analysis , we isolated the contaminating human nucleosomal DNA to visualize the oscillation of GC and AT sequences every 10bp ( S7C Fig ) . The sequence oscillation of human nucleosome occupancy can be observed , but due to the relatively low number of read counts the pattern is similar , but not identical to the Bunnik study . In contrast to the Bunnik data , the 146-148bp long DNA fragments of P . falciparum did reveal a weak oscillation of the AT-dinucleotides , with an unusual enrichment at the dyad axis and additional AT-peaks shifted by 5bp . However , no oscillation of the CG-dinucleotides could be observed . The pattern is not completely lost as suggested in the dataset of the LeRoch manuscript , but very different and relaxed in comparison to the human dinucleotide pattern . Differences in the dinucleotide repeat pattern between human and plasmodium nucleosomes can be also attributed to the high AT-content of the plasmodium genome , albeit clearly shifted peak distributions indicate intrinsic differences in motif recognition . In addition , the absence of positioned nucleosomes in the genome do confirm relaxed recognition of nucleosome positioning signals , like the side by side comparison of human and plasmodium nucleosome positioning patterns . The data suggests that plasmodium histones either recognize different sequence dependent signals , or have a reduced affinity to DNA motifs being located in the realm of the nucleosome . In order to reveal an alternative nucleosome positioning pattern in detail , future experiments would have to reconstitute plasmodium nucleosomes on whole genomic DNA libraries in vitro and perform sequence analysis of the MNase protected DNA fragments . Our biochemical studies do provide an explanation for the lack of positioning patterns in vivo , arguing for the presence of alternative and additional sequence-dependent positioning signals in the regulatory regions of P . falciparum . As the parasite nucleosome lost its capability to adopt sequence-specific nucleosome positioning , we analyzed which signals could determine the positioning of nucleosomes at regulatory regions . We tested whether the associated linker DNA sequences are involved in the translational positioning of nucleosomes . Using the genome wide nucleosome occupancy data , we defined a fuzziness parameter of nucleosome positioning . The idea is that well positioned nucleosomes create a small occupancy footprint on genomic DNA , with narrow peaks of nucleosome read annotations . In order to calculate the degree of fuzziness we used the established DANPOS2 toolkit [52] . Next we aligned the P . falciparum genes according to the start of the protein coding region ( ATG ) , the end of the coding region ( ECR ) and the exon/intron boundaries , plotting nucleosome occupancy and the degree of nucleosome fuzziness above . We have performed the analysis for the datasets available from the LeRoch and Bártfai labs , as described above , giving similar results ( Fig 6A and 6B and S8A and S8B Fig respectively ) . Even though the nucleosome core exhibits relaxed recognition of sequence dependent nucleosome positioning signals , we observed specifically positioned nucleosomes at these regulatory regions . An inspection of the AA/TT and AT/TA dinucleotide frequency revealed their non-random distribution around the strongly positioned nucleosomes . AT dinucleotide frequencies are enriched , whereas AA dinucleotide frequencies are reduced in the linker regions of these nucleosomes ( Fig 6A; S8A Fig ) . Next we sorted the genomic peaks of nucleosome occupancy according to the fuzziness parameter and plotted the respective AA and AT dinucleotide abundance ( Fig 6B , S8B Fig ) . As shown for the site-specific analysis , the best positioned nucleosomes are characterized by the largest depletion of AA/TT di-nucleotides and the strongest enrichment of AT/TA di-nucleotides , flanking the nucleosome core . The bioinformatics analysis clearly reveals that flanking sequences of well positioned parasite nucleosomes exhibit increased stretches of AT-rich elements , suggesting the importance of linker DNA in translational nucleosome positioning . In contrast , yeast nucleosomes do not reveal such linker DNA dependent positioning signals ( Fig 6B ) . In order to test this behavior experimentally , we generated three DNA fragments with a 150bp long HSP70 sequence , flanked either by DNA linkers consisting of 15 bp of AT di-nucleotides , AA di-nucleotides or GC-rich sequences ( Fig 6C ) . The three different DNA fragments were used for nucleosome assembly and analyzed on native polyacrylamide gels . Whereas the AA- and GC-rich linkers favor the assembly of nucleosomes at more than two distinct sites , the AT linker containing DNA fragment reveals a distinct positioning pattern with a reduced number of positions . Still the 15bp of flanking DNA did not convincingly determine a unique translational position , showing that the 15bp long AT repeat is not a sufficiently strong signal . However , in vivo the best positioned nucleosomes are flanked by AT-repeats of up to 50 bp , suggesting that longer sequence elements are required . Next we used the genomic data to retrieve highly positioned nucleosomes in plasmodium . We selected sequences close to the MAL1P1 . 31 ( MAL ) and HSP86 ( HSP ) promoter , harboring a positioned nucleosome . The genomic regions were cloned and amplified by PCR , containing AT-rich linker regions from 32 to 42bp . DNA fragments were used for nucleosome assembly ( Fig 6D ) . Indeed , these DNA fragments revealed only one main nucleosomal position in the EMSA assay , showing that long AT-repeats are required to guide nucleosome positioning in vitro and potentially in vivo ( lanes 1–3 and 9–11 ) . To prove that the AT-sequences are responsible for positioning within these genomic sequence contexts , we replaced the AT-repeats by GC-rich DNA elements . The different DNA templates were reconstituted in parallel and revealed that in the absence of the AT-repeats discrete nucleosome positioning is lost ( lanes 4–7 ) . Our results suggest that plasmodium histone octamers have a reduced affinity for AT di-nucleotide repeats containing DNA elements and are therefore forced to assemble next to these sequences . In summary , we show that the biochemical properties of the histone octamers dictate the positioning of the parasite nucleosome cores and the accessibility of the chromatin structure .
The sequences of plasmodial histones are highly divergent from those of other eukaryotes . We questioned whether this difference represents an adaptation to the extraordinarily high AT-content of P . falciparum and whether these amino acid replacements do alter the physicochemical properties of the nucleosome . The results presented in this study are un-expected , showing that the observed mutations do not result in better binding of AT-rich DNA . In contrast , we even observe a reduced stability of the nucleosomes on GC-rich DNA , accompanied by a reduction in the thermal stability of the octamer on DNA . In agreement with the reduced thermal stability , we also observe reduced salt stability of H2A and H2B , the histones with the majority of sequence alterations . The plasmodium nucleosomes lost their strong ability to recognize the phased AT and GC di-nucleotide patterns , weakening the intrinsic capability of sequence dependent nucleosome positioning . The in vitro reconstitution experiments are backed up by the comparative genomic analyses of the human and plasmodium nucleosome positions in vivo , showing either no ( LeRoch data ) , or a strongly reduced and altered recognition of di-nucleotide patterns ( Bártfai data ) . Still , positioned nucleosomes can be detected at regulatory regions like the transcription start sites , transcription termination sites and intron/exon boundaries . We can show that nucleosome positioning at these sites is achieved by the presence of long AT-repeats in the linker regions of these nucleosomes , being unfavorable sites and therefore placing the nucleosomes next to such sequences . These AT-rich sequence boundaries are preferentially located at regulatory elements in plasmodium , but not in yeast and human genomic DNA , presenting an alternative mechanism of nucleosome positioning . Eukaryotic nucleosome cores are spaced by 20 ( budding yeast ) to 75bp ( echinoderm sperm ) of linker DNA connecting neighboring nucleosomes [45] . The plasmodial nucleosomal arrays reconstituted in this study reveal the shortest linker lengths in the eukaryotic kingdom known today . With a repeat length of about 155bp , nucleosome spacing is significantly shorter than in yeast . Published in vivo data is conflicting , with a study suggesting repeat lengths of 180bp [53] and others that correlate well with our results , describing extremely short nucleosome repeat lengths of 155bp ( +/-5bp ) [11 , 46] . In our opinion , the combined biochemical and genomic data analysis does convincingly reveal the short NRL that is present throughout the erythrocytic stages of the plasmodium life cycle . We suggest that sequence alterations in the histones allow the generation of compact nucleosomal arrays determining the unique chromatin architecture of P . falciparum . Like yeast , P . falciparum is lacking histone H1 and our experiments show that the short NRLs are a result of the biochemical properties of the plasmodial histone-octamer rather than depending on the absence of H1 binding . Like the yeast genome , the plasmodial genome is relatively accessible , as judged by MNase and FAIRE assays [14] implying the lack of condensed heterochromatin structures . Several reports and modeling studies show that short nucleosome spacing interferes with the formation of organized higher order structures of chromatin , resulting in an accessible genome architecture [54 , 55] . Our data suggest that plasmodial histones evolved to enable short nucleosome repeat lengths potentially inhibiting the formation of compact higher order structures of chromatin . Besides the atypical spacing , plasmodial histones exhibit reduced thermal and salt stabilities revealing weakened histone DNA interactions . The abridged levels of higher order structures of chromatin is accompanied with reduced nucleosome stability , potentially simplifying the access of sequence specific binding proteins to DNA . Kensche and colleagues showed that during the transition between the cell cycle stages chromatin structure and nucleosome positioning changes occur around the transcription factor binding sites [16] . Screens of histone mutations in yeast , relieving the dependence of transcription on the SWI/SNF remodeling complex ( SIN ) identified specific changes in the histones H3 and H4 [56 , 57] . The amino acid changes are preferentially located at SHL locations , where the histones do directly interact with DNA and do alter nucleosome stability and chromatin compaction [58 , 59] . The P . falciparum H3 and H4 sequences do not exhibit these classical SIN mutations . But , the H2A and H2B histone sequences exhibit clusters of amino acid changes at and close to SHL locations , with two sites being equivalent to SIN mutations ( G to T in H2B position 72 and T to S in H2A position 76 ) . The role of H2A and H2B SHL regions in the stabilization of the nucleoprotein structure was not yet analyzed , but they are suggested to play an important role [60] . Interestingly , the P . falciparum histones exhibit clustered amino acid changes at the SHL3 . 5 and 4 . 5 regions of H2B and H2A that could alter nucleosome stability and contribute to the open chromatin structure in P . falciparum . According to our in silico assessment of protein-DNA interactions , the interactions of human and plasmodial histones with DNA are indistinguishable , which is in contrast to our experimental results . How can one reconcile these seemingly conflicting findings ? Due to their flexibility , N- and C-termini of the histones were not considered in homology modeling; consequently , our binding analysis was blind for their putative interactions with DNA and effect on nucleosome stabilization . On the other hand , a comparison of human and plasmodial histone sequences reveals drastic differences in these histone termini ( Fig 1 and S1 Fig ) . We suggest a crucial contribution of these flexible histone termini to DNA binding and nucleosome stabilization , which is a hypothesis to be tested in future experiments . The mechanism driving nucleosome positioning is an essential field of study in chromatin research , as the locations of the nucleosomes on DNA determine the accessibility for sequence specific DNA binding proteins . The first description of nucleosome positioning in vivo and revealing the important role of multiple histone-DNA interactions and DNA structure in mediating positioning [61 , 62] initiated a search for DNA dependent positioning signals . Many structure based differences in DNA sequence patterns , like AA [62] , GG [63 , 64] , TA and GC oscillations , several tri- to poly-nucleotide patterns ( for a review see ref . [65] ) , motifs being essential for anisotropic DNA bending ( for a review see ref . [66] ) were proposed to determine the intrinsic nucleosome positioning behavior . In principle , sequences that are already pre-curved , requiring low energy levels to wind around a histone octamer should preferentially bind and position nucleosomes . Models were devised to predict nucleosome positioning in vitro and in vivo from DNA sequence [9] , albeit the predictive power is currently being questioned [67] . With the study presented here , we further question the sequence dependent view of nucleosome positioning . The overall structure of the nucleosome of P . falciparum nucleosomes is identical to other eukaryotic structures , but has a strongly reduced ability to recognize sequence encoded nucleosome positioning signals in vivo and in vitro . Phased di-nucleotide repeat patterns , clearly detectable for the human nucleosomes , are mostly diminished in plasmodium . Biochemical reconstitution of nucleosomes on DNA templates of different origin and AT content only exhibit a few discrete nucleosome positions . The lack of positioning argues that additional constraints must exist to determine nucleosome positioning . As the exchange of human histone H4 for the plasmodial H4 in the octamer did not restore positioning , the signals must be read by the other histones . The best candidates are the histones H2A and H2B that exhibit the most differences to the other canonical histones ( Fig 1 ) . We hypothesize that the histone tails , bearing most of the amino acid exchanges , may interact with DNA and influence nucleosome positioning , but details have to be addressed in future experiments . Lack of intrinsic nucleosome positioning capabilities are exchanged by linker DNA dependent nucleosome positioning mechanisms in P . falciparum , like the AT repeats enriched in the associated linker DNA of well positioned plasmodial , but not in yeast and human nucleosomes . A closer look also reveals that such external positioning sequences do flank regulatory regions and thereby ensure specific positioning of regulatory nucleosomes .
Plasmids encoding the canonical human and plasmodium histone sequences were optimized for bacterial expression and ordered as synthetic genes . Recombinant expression , purification of histones from inclusion bodies and octamer refolding was done as described previously [68] . Recombinant SNF2H , Chd1 , Chd3 and Chd4 were expressed in Sf21 cells ( Invitrogen ) and prepared according to standard procedures [49] . DNA fragments were synthesized by PCR , using fluorescently labelled oligonucleotides binding to the described murine rRNA gene promoter ( -190 to +90 , relative to the transcription start site ) , to the D . melanogaster HSP70 promoter and to the DNA 601 sequence . The AT rich P . falciparum sequence Pf3D7v3:1307900–8200 was subcloned and DNA fragments were prepared by PCR or restriction enzyme digestion and purification . MAL and HSP sequences were generated by oligonucleotide-annealing , -ligation and cloning into pUC19 . A plasmid containing the sequence encompassing the Knob-associated histidine-rich protein ( KahrP ) gene promoter , was kindly provided by Till Voss . Nucleosomes were assembled according to Rhodes and Laskey using the salt gradient dialysis technique [69] . A typical assembly reaction ( 50 μl ) contained 4 . 0 μg DNA , varying amounts of recombinant histone octamer , 200 ng BSA/ml , in high salt buffer ( 10 mM Tris , pH 7 . 6 , 2 M NaCl , 1 mM EDTA , 0 . 05% NP-40 , 2 mM ß-mercaptoethanol ) . The salt was continuously reduced for 16–20 h and nucleosomes were assayed in 80 mM salt buffers . The quality of the assembly reaction was assayed by electromobility shift assays on native polyacrylamide gels or by partial MNase digestion and analysis of the nucleosomal ladder on agarose gels . Mobility shift assays utilizing thermally induced movement of nucleosomes were carried out as described [70] . Nucleosomal DNA , either labelled with Cy3 or Cy5 ( 150ng of each template ) were incubated in a total volume of 20 μl Ex80/BSA-buffer ( 10 mM Tris , pH 7 . 6 , 80 mM NaCl , 1 . 5 mM MgCl2 , 1 mM EDTA , 0 . 05% NP-40; 200 mg/l BSA ) for 60 min at 48° to 66°C . Nucleosome positions were analyzed on a native 6% polyacrylamide gels ( 0 . 4x TBE ) and visualized fluorescence scanning . To monitor nucleosomal stability with increasing chloroquine ( Sigma ) concentrations , differentially fluorescent AT-rich and GC-rich DNA fragments were fully reconstituted into nucleosomes . Nucleosomal species were mixed to allow competitive and comparative assays . Increasing concentrations of chloroquine were added to the nucleosomes in EX80/BSA-buffer and incubated for 10 min at 37°C . Nucleosome positions were analyzed as described above . Biotonylated DNA was prepared by PCR , using one biotinylated primer and the plasmid pUC19 as a template . Plasmodium and human nucleosomal arrays were reconstituted on the 2375bp long DNA fragment by the salt dialysis method . Nucleosomal arrays ( 4μg ) were coupled to magnetic streptavidin coated Dynal Beads and unbound DNA was washed with Ex80-buffer . Chromatin was incubated with LO-buffer , stepwise increasing the NaCl concentration from 400 to 1400 mM , then bound to the magnetic beads and the supernatant was collected . Salt eluted histones were analyzed by SDS-PAGE ( 17% ) and visualized by silver staining . Nucleosome mobility was assayed as described [49] . Briefly , reactions contained 20 nM Cy5 and Cy3 labelled DNA reconstituted into nucleosomes , 1 mM ATP , 100 ng/μl BSA , in Ex80 buffer ( 20 mM Tris pH 7 . 6 , 80 mM KCl , 1 . 5 mM MgCl2 , 0 . 5 mM EGTA ) and recombinant remodeling enzymes . Nucleosomes were incubated with the enzymes for 60 min at 30°C . The reactions were stopped by the addition of 1μg of plasmid DNA and incubated for 5 min on ice . The nucleosome positions were analyzed by electrophoresis on 6% native polyacrylamide gels in 0 . 4x TBE and fluorescence scanning . Nucleosome remodeler interactions were probed by the incubation of 20nM of nucleosomes with increasing concentrations of remodeling enzymes , for 60 min at 30°C , then loaded on 6% native polyacrylamide gels in 0 . 4x TBE and fluorescence scanning . Sequence data from formaldehyde-assisted isolation of regulatory elements ( FAIRE; SRX013302 ) and MNase mediated purification of mononuclesomes ( SRX013309 , SRX316306 , SRX316307 , SRX885811-SRX885819 ) [14–16] was downloaded and mapped to the P . falciparum 3D7D genome version 3 , or the UCSC human genome version 37 ( hg19 ) using the local alignment option of bowtie2 with default settings [71] . The reads were filtered for mapping quality ( MAPQ > 20 ) and concordant alignment in the case of using paired-end sequencing data . The bowtie output was further processed using SAMtools [72] and BEDtools [73] for customized analysis purposes . Human as well as plasmodium derived nucleosomal fragments with a size of 146-148bp were selected ( SRX316307 , SRX885814 ) [15 , 16] . Nucleotide frequency was calculated using annotatePeaks . pl script from the HOMER software package [74] with the following parameters: -hist 1 –di–size -150 , 150 . The findPeaks script from the HOMER software package was applied to FAIRE ( SRX013302 ) [15] and MNase single-end sequencing data ( SRX013309 ) with the following parameters: -style histone–size 147 –region–norm 1e7 –gsize 23292104 –minDist 20 –inputSize 147 –C 0 –F 2 –minTagThreshold 50 . To determine the nucleosome peaks the MNase sequencing data was used as input and the FAIRE sequencing data as background ( vice versa for the FAIRE peaks ) . Peaks were centered at the position with the highest read coverage using getPeakTags script ( -start 80 –end 80 –center–fragLength 70 ) . Peaks overlapping more than 100bp were merged and GC content of peaks was calculated using BEDtools . annotatePeaks . pl script was applied to annotate the peaks to genomic features ( exon , intron , intergenic ) on the plasmodium genome . Nucleosome positions of MNase paired-end sequencing data ( SRX316306 , SRX885814 ) [15 , 16] were determined using danpos . py script from DANPOS2 toolkit [52] with the default parameters in dpos mode . The reported fuzziness score was used as measurement of nucleosome positioning and subsequent filtering . Peak centers were annotated to genomic features ( TSS , TTS and Exon/Intron boundary ) using annotatePeaks . pl from HOMER package . Total peak count was normalized respectively by the occurrence of the feature . annotatePeaks . pl script was applied to calculate the nucleotide distribution of features and peaks as well . The fragment size distribution of human as well as plasmodium derived di-nucleosomal fragments was estimated by a gaussian kernel density using fragments larger than 250 bp ( SRX885811-SRX885819 ) [16] . The average nucleosome-nucleosome distance was inferred from the maximum of the kernel density estimate . The standard protocol of YASARA [75] ( version 16 . 4 . 6 ) was used to create homology models of all histones and the complete nucleosome consisting of an octamer that had 146bp of DNA wrapped around it . For each model of a nucleosomal complex , the input of YASARA was a multiple FASTA file with two DNA- and eight protein- sequences . The DNA-sequences were two copies of the palindromic DNA fragment ( 146bp long ) from human X-chromosome alpha satellite DNA as found in PDB entry 3AFA . The protein-sequences were from the histones of Homo sapiens or P . falciparum , respectively . The GenBank accession numbers for the human histones were AAA63191 . 1 ( H2A ) , AAN59961 . 1 ( H2B ) , NP_066403 . 2 ( H3 ) , NP_003539 . 1 ( H4 ) and for the plasmodial histones AAA29612 . 1 ( H2A ) , XP_001347738 . 1 ( H2B ) , AAO23910 . 1 ( H3 ) , AAP45785 . 1 ( H4 ) . Due to their flexibility , the N- and C-termini of histones could not be resolved in X-ray structures; thus , their 3D-orientation is unclear . This is why the histone sequences were trimmed according to the resolved 3D structure reported in PDB entry 3AFA . In order to determine the homology models for plasmodium , three rounds of PSI-BLAST restricted to PDB entries [22] were conducted and YASARA selected PDB entries 3AFA , 5AV6 , 3TU4 , 3X1T and 2NQB as templates . These datasets represent the structures of nucleosomal core particles from different eukaryotic species . For the human template ( 3AFA ) , all of the 740 target residues could be aligned to template residues; for these the sequence identity was 84% . After building models for each template , YASARA combined the best scoring fragments of all models to deduce a hybrid homology model . The resulting hybrid model scored best and the internal quality assessment of YASARA determined an overall Z-score of0 . 056 , which indicates model quality 0 . 056 standard deviations better than an average high-resolution X-ray structure . Note that model quality is most reliable for globular proteins and can be misleading for other protein types . Dihedrals and packing in 1D and 3D were rated as optimal by YASARA . The full models of a nucleosome in PDB-format can be found in the two Supplementary Files hu_complex . pdb ( Homo sapiens ) and pl_complex . pdb ( P . falciparum ) . YASARA [24] was used to run three MD simulations for each nucleosome complex . In all runs , the complex was embedded into a water box; in order to vary experimental conditions , simulations were initiated with the different start-temperatures of 298 . 14 K , 298 . 15 K , and 298 . 16 K , respectively . The data sets consisted of three MD trajectories each comprising 200 snapshots that represented varying poses of a 50 ns interval and FoldX [23] was used to calculate a score assessing the interaction energy . 200 snapshots with a time period of 250 ps were saved for subsequent processing; snapshots were stored in pdb format and contained the complex plus all water molecules within a maximal distance of 3 Å to a protein or DNA molecule . These snapshots were used to deduce mean values of scores assessing the following interactions , which were determined in a residue-specific manner: π-π stacking , cation-π stacking , contacts , hydrophobic interactions , and hydrogen-bond networks . For the first four interactions , scores were taken from the YASARA output; see YASARA documentation for details of computation . To score hydrogen-bond networks , distances were analyzed between residues , DNA , and water molecules in a snapshot-specific manner . Thus , a graph was computed that consisted of nodes that represent putatively interacting atoms on the surface of the considered molecules and of edges modelling hydrogen bonds . An edge was inserted , if the distance between a donor and an acceptor atom was not larger than 2 . 5 Å . Based on this network , a score was computed for each path interconnecting a pair of atoms from DNA and a protein according to: Spath ( atomik , atomjl ) =1/ ( edges ( atomik , atomjl ) # path_ident_len ) ( 1 ) Here , edges ( atomik , atomjl ) is the number of edges interconnecting an atom k of residue I with atom l of nucleotide j and the normalization factor #path_ident_len is the number of paths with the same length observed in the full data set . Thus , the score for a hydrogen-mediated interaction decreases with the number of involved water molecules and results in a higher score for a more direct one . The maximal number of co-operative water molecules was limited to one and for each residue resj , all spath-values were summed up . For each of the averaged scores with noticeable amplitude , the log2-value was plotted for corresponding residues of the histones from H . sapiens and P . falciparum together with the sequences by means of a circos graph [76] . To analyze the interaction energy between individual histone cores and the DNA , FoldX was used ( version 4 , [23] ) . First , the side-chain orientation of all snapshots was optimized with the RepairPDB command to prepare the structures for the FoldX force-field . Then , mean interaction energies between Histone and DNA as well as their standard deviations were then deduced with the AnalyzeComplex command . The high-throughput sequencing data are available in the NCBI’s Sequence Read Archive ( SRA ) ( accession: SRP055417 , SRP026365 , SRP001451 , SRP001452 ) .
|
Nucleosomes are not positioned randomly on DNA but on preferential sites with respect to the underlying DNA sequence . Histones belong to the most conserved eukaryotic proteins , as sequence dependent nucleosome positioning is an essential regulatory feature of nucleosomes , determining the accessibility of regulatory factors to DNA . We determined the biochemical properties of plasmodium histones and show that they are distinct from human forms , explaining the accessible chromatin structure of P . falciparum . Amino acid exchanges in the histones do not present an adaption to the AT-rich genome , but rather reduce the binding affinity to GC-rich DNA sequences , resulting in rather unstable nucleosomes with labile H2A and H2B , requiring extra-nucleosomal positioning signals to keep them on place . Plasmodium chromatin exhibits the shortest nucleosome spacing known to date potentially inhibiting the formation of higher order structures and maintaining chromatin accessible .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"parasite",
"groups",
"plasmodium",
"dna-binding",
"proteins",
"plasmodium",
"falciparum",
"parasitic",
"protozoans",
"sequence",
"assembly",
"tools",
"parasitology",
"apicomplexa",
"protozoans",
"genome",
"analysis",
"sequence",
"motif",
"analysis",
"epigenetics",
"chromatin",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"genomics",
"malarial",
"parasites",
"chromosome",
"biology",
"proteins",
"bioinformatics",
"gene",
"expression",
"histones",
"nucleosomes",
"biochemistry",
"dna",
"sequence",
"analysis",
"cell",
"biology",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"organisms"
] |
2016
|
Plasmodium falciparum Nucleosomes Exhibit Reduced Stability and Lost Sequence Dependent Nucleosome Positioning
|
During many chronic infections virus-specific CD8 T cells succumb to exhaustion as they lose their ability to respond to antigenic activation . Combinations of IL-12 , IL-18 , and IL-21 have been shown to induce the antigen-independent production of interferon ( IFN ) -γ by effector and memory CD8 T cells . In this study we investigated whether exhausted CD8 T cells are sensitive to activation by these cytokines . We show that effector and memory , but not exhausted , CD8 T cells produce IFN-γ and upregulate CD25 following exposure to certain combinations of IL-12 , IL-18 , and IL-21 . The unresponsiveness of exhausted CD8 T cells is associated with downregulation of the IL-18-receptor-α ( IL-18Rα ) . Although IL-18Rα expression is connected with the ability of memory CD8 T cells to self-renew and efflux rhodamine 123 , the IL-18Rαlo exhausted cells remained capable of secreting this dye . To further evaluate the consequences of IL-18Rα downregulation , we tracked the fate of IL-18Rα-deficient CD8 T cells in chronically infected mixed bone marrow chimeras and discovered that IL-18Rα affects the initial but not later phases of the response . The antigen-independent responsiveness of exhausted CD8 T cells was also investigated following co-infection with Listeria monocytogenes , which induces the expression of IL-12 and IL-18 . Although IL-18Rαhi memory cells upregulated CD25 and produced IFN-γ , the IL-18Rαlo exhausted cells failed to respond . Collectively , these findings indicate that as exhausted T cells adjust to the chronically infected environment , they lose their susceptibility to antigen-independent activation by cytokines , which compromises their ability to detect bacterial co-infections .
Memory CD8 T cells typically develop following a short period of antigenic activation , which occurs during acute infections with intracellular pathogens . A hallmark of these memory T cells is their ability to rapidly respond following re-exposure to their original inducing antigen [reviewed in 1 , 2] . This recall response includes the production of cytokines such as interferon ( IFN ) -γ , the elaboration of cytotoxic effector activities , proliferation , and changes in the expression of cytokine receptors and adhesion molecules . In addition to their exquisite ability to respond to antigenic activation , memory CD8 T cells have also been shown to possess an innate-like ability to respond to certain sets of cytokines in the absence of antigen exposure [3]–[10] . Most notably , a combination of the proinflammatory cytokines IL-12 and IL-18 causes the pronounced production of IFN-γ by memory T cells [5] , [6] , [8]–[10] . Other cytokine combinations including IL-18 and IL-21 , as well as IL-18 and type I IFN have been shown to have similar activating effects [3] , [4] , [11] . This sensitivity to cytokine stimulation endows memory T cells with the capacity to respond in an antigen- and T cell receptor ( TCR ) -independent manner to certain infections that induce inflammation , such as Listeria monocytogenes ( LM ) and Burkholderia pseudomallei [5] , [6] , [10] . Thus , pre-existing memory CD8 T cells can potentially contribute to the control of a broad set of infections due to their ability to detect changes in the inflammatory cytokine milieu . During chronic viral infections the development of prototypic memory CD8 T cells is disrupted . Although initial CD8 T cell responses are usually elaborated , the responding virus-specific CD8 T cells undergo a differentiation process that results in their exhaustion [reviewed in 1 , 12 , 13] . The most severely exhausted CD8 T cells develop under conditions of high viral loads and ineffective CD4 T cell help . Although severely exhausted CD8 T cells retain expression of IFN-γ mRNA , they fail to produce IFN-γ protein after exposure to their cognate antigen [14] , [15] . In addition , exhausted cells exhibit altered maintenance requirements , as they lose the self-renewal properties associated with normal memory T cells and may become deleted over time [16]–[18] . It is possible that exhaustion has evolved to allow antigen-specific T cells to become tuned to an environment of persisting antigen . Thus their loss of responsiveness to antigenic activation may serve as a safety mechanism that limits pronounced and sustained effector activities , which could be immunopathogenic . It is less clear , however , whether exhaustion alters the ability of T cells to mount antigen-independent responses to inflammatory cytokines . Although T cell exhaustion has been described during several chronic viral infections including HIV and hepatitis C virus infections , it is most well characterized in mice persistently infected with lymphocytic choriomeningitis virus ( LCMV ) . The LCMV system is particularly informative as different durations of infection can be established depending upon the isolate of virus and strains of mice used [17] , [19] . In the current study we have used LCMV infection of mice to investigate whether the development of T cell exhaustion alters the ability of virus-specific CD8 T cells to perceive and respond to antigen-independent activation with combinations of IL-12 , IL-18 , and IL-21 . The findings show that unlike effector and memory CD8 T cells , exhausted cells are not activated by these cytokines , and this correlates with differential expression of the IL-18-receptor-α ( IL-18Rα ) . The decrease of IL-18Rα expression on exhausted CD8 T cells is consequential as it renders these cells more prone to deletion during the initial phase of persistent LCMV infection . In addition , lower IL-18Rα expression is associated with the failure of exhausted CD8 T cells to respond to bacterial co-infection by upregulating CD25 and producing IFN-γ .
For these studies we harnessed the LCMV system which provides an informative experimental platform for analyzing effector , memory , and exhausted CD8 T cell responses . We induced acute infections in C57BL/6 ( B6 ) mice using the LCMV-Arm isolate which elicits a pronounced effector CD8 T cell response and subsequently establishes a long-lived pool of highly functional memory CD8 T cells following the complete resolution of the infection . By contrast , LCMV-cl13 infection of B6 mice results in a disseminated infection which is slowly brought under control over a period of several months , but continues to smolder in certain organs such as the kidney . This protracted infection induces an initial effector-like CD8 T cell response which is followed by the development of exhaustion . LCMV-cl13 infection of CD4-/- mice is never brought under control and , although an initial effector-like response is mounted , severe CD8 T cell exhaustion develops as the chronic infection persists [17] . Previous studies have shown that various combinations of IL-12 , IL-18 , and IL-21 activate primed CD8 T cells to produce IFN-γ in the absence of antigenic stimulation [3]–[6] , [8]–[11] . We confirmed and extended these findings using LCMV-specific effector CD8 T cells derived from B6 mice infected with LCMV-Arm ( an acute infection ) and observed qualitative and quantitative differences in the ability of the cytokine mixtures to stimulate IFN-γ production ( Fig . 1 ) . IL-12 , IL-18 , or IL-21 alone caused , at best , minimal IFN-γ production by LCMV GP33-specific effector CD8 T cells ( Fig . 1A ) . IL-12+IL-18 with or without IL-21 stimulated IFN-γ production by 68±8% ( SD ) of the cells; however , IL-18+IL-21 had a more modest effect , activating 33±17% of these CD8 T cells , and the levels of IFN-γ production were also lower based upon the mean fluorescence intensity ( MFI ) of the IFN-γ+ population ( Fig . 1B ) . IL-12+IL-21 was less effective at stimulating IFN-γ production ( Fig . 1 ) . Since severely exhausted CD8 T cells fail to produce IFN-γ in response to stimulation with their cognate antigen [15]–[20] , we next investigated whether effector , memory , and exhausted virus-specific CD8 T cells were susceptible to activation with the various cytokine combinations . To address this , acute ( LCMV-Arm ) , protracted ( LCMV-cl13 infection of B6 mice ) , and chronic ( LCMV-cl13 infection of CD4-/- mice ) infections were established . During the effector phase of the CD8 T cell response , at day 9 following infection , LCMV GP33-specific CD8 T cells from all of the cohorts analyzed produced IFN-γ in response to a brief ( 5 . 5 hr ) exposure to IL-12+IL-18 and IL-12+IL-18+IL-21 ( Fig . 2 ) . The other cytokine combinations and the single cytokines alone had minimal to modest stimulatory effects ( Fig . 2D ) . By day 35 post-infection , the memory CD8 T cells that developed in acutely infected hosts retained the ability to respond to the cytokine combinations . By contrast , the development of exhaustion ( days 35-36 post-infection of cl13 infected mice ) was associated with the inability to produce IFN-γ in response to these cytokines ( Fig . 2 ) . We also checked whether exposure to IL-12 , IL-18 , and IL-21 either alone or in combination activated effector , memory , or exhausted CD8 T cells to upregulate expression of the IL-2 receptor-α chain , CD25 . Changes in CD25 expression by LCMV-GP33 epitope-specific CD8 T cells were assessed following stimulation with the various cytokine combinations . The levels of CD25 increased on memory CD8 T cells from B6 LCMV-Arm infected mice following stimulation ( Fig . 3 ) . The magnitude of upregulation paralleled the trends observed for IFN-γ production with the most marked expression induced by IL-12+IL-18 and IL-12+IL-18+IL-21 , while IL-18+IL-21 had less potent activating abilities ( Fig . 2D and Fig . 3B ) . By contrast , LCMV-specific CD8 T cells from protracted and chronically infected mice failed to increase CD25 expression in response to stimulation with the cytokine panels ( Fig . 3 ) . Collectively these findings show the divergence between memory and exhausted CD8 T cells and demonstrate that persistent viral infections corrupt the ability of virus-specific CD8 T cells to respond to antigen-independent stimuli . To investigate why exhausted cells lose responsiveness to cytokine stimulation , the expression of the cognate cytokine receptors on virus-specific CD8 T cells was evaluated . Whereas the expression of IL-12Rβ2 was similar on LCMV GP33-specific CD8 T cells from mice undergoing acute , protracted , and chronic LCMV infections ( Fig . 4A , D ) , the levels of the IL-21R tended to be higher in the protracted and chronically infected cohorts ( Fig . 4B , E ) . The IL-18Rα was clearly expressed at high levels during the effector phase of the response ( day 9 ) and was maintained on memory CD8 T cells ( day 35 , B6 Arm ) ; however , the development of exhaustion in the protracted and chronically infected groups was associated with the downregulation of IL-18Rα ( days 35-36 , B6 cl13 and CD4-/- cl13 ) ( Fig . 4C , F ) . Efflux of the fluorescent dye rhodamine 123 ( Rh123 ) is associated with populations of IL-18Rhi memory CD8 T cells which have self-renewing capabilities [21] . Since severely exhausted CD8 T cells are not always maintained over time [17]–[19] and express only low levels of the IL-18Rα , we next evaluated whether these cells were capable of effluxing Rh123 . Consistent with previous findings [21] , we found that IL-18Rαhi CD8 T cells from acutely infected mice ( Fig . 5A ) , which encompass the virus-specific memory T cell population ( Fig . 5B ) , efficiently effluxed Rh123 during a one-hour period . This efflux of Rh123 was blocked by either Cyclosporine A ( CsA; Fig . 5 ) or vinblastine ( Fig . 5B ) , which inhibit ABCB1 transporters required for the removal of Rh123 [21] , [22] . By contrast , virus-specific CD8 T cells from protracted ( B6 cl13 ) and chronically ( CD4-/- cl13 ) infected mice , which have downregulated the IL-18Rα , effluxed Rh123 at least as efficiently as their memory counterparts from acutely infected mice ( Fig . 5 ) . Thus , even though exhausted CD8 T cells have altered maintenance requirements and proliferative properties , in chronic LCMV infection the levels of IL-18Rα on virus-specific CD8 T cells do not correlate with their ability to efflux Rh123 . To further examine the significance of decreased IL-18Rα expression on virus-specific CD8 T cell responses , we analyzed bone marrow chimeras generated by reconstituting lethally irradiated mice with a mixture of CD45 . 1 IL-18Rα+/+ cells and either CD45 . 2 IL-18Rα-/- ( experimental ) or CD45 . 2 IL-18Rα+/+ ( control ) cells . By 8 weeks following reconstitution , the mean proportion of CD8 T cells that were IL-18Rα-/- ( CD45 . 2 ) was 59% with a range of 53-64% in the experimental chimeras , and in the control chimeras the fraction of CD8 T cells that were CD45 . 2 ( IL-18Rα+/+ ) was 42% with a range of 35-46% . Both the experimental and control chimeras responded to infection with LCMV-cl13 . By 7 days following infection the mean proportion of CD8 T cells that were CD45 . 2 ( IL-18Rα-/- ) was 53% with a range of 38-63% in the experimental chimeras , and in the control chimeras the fraction of CD8 T cells that were CD45 . 2 ( IL-18Rα+/+ ) was 39% with a range of 31-46% . Virus-specific CD45 . 2+ CD8 T cell responses also became detectable in both cohorts ( Fig . 6A ) and the fraction of CD45 . 2 tetramer binding cells at this initial time point essentially reflect the pre-infection degree of chimerism . Between days 7-16 following infection , the virus-specific IL-18Rα-/- ( CD45 . 2 ) cells appeared to preferentially contract in the experimental chimeras , as the fraction of these cells decreased during this period . However , the proportion of virus-specific CD8 T cells that were CD45 . 2 ( IL-18Rα+/+ ) in the control chimeras remained stable ( Fig . 6A ) . Although the IL-18Rα-/- CD8 T cells appeared to be outcompeted during the second week of infection , this disproportionate loss stabilized between days 16-26 ( Fig . 6A , B ) . This stabilization was concurrent with the downregulation of IL-18Rα on the DbGP33+CD8 T cells in the control and experimental chimeras ( Fig . 6C ) . Given the marked downregulation of IL-18Rα on exhausted CD8 T cells , we evaluated whether this would compromise their ability to respond in an antigen-independent manner to a bacterial co-infection . LM has been previously shown to stimulate IFN-γ production by memory CD8 T cells that are not specific for LM antigens but respond due to their sensitivity to IL-12 and IL-18 induced by the bacterial infection [6] . Thus to provide a stringent in vivo readout of whether the innate-like responses of exhausted CD8 T cells were corrupted , cohorts of acute , protracted , or chronically infected mice were challenged with wild-type LM , which does not encode any known LCMV epitopes . We used a challenge inoculum of 106 cfu as this dose has been previously shown to induce a pronounced IFN-γ + response by memory CD8 T cells [6] , and evaluated the impact of this co-infection on the “bystander” LCMV-specific CD8 T cells 20hr later ( Fig . 7A ) . In all cases , the CD8 T cells that became IFN-γ+ following LM challenge were IL-18Rαhi , indicating that the ability to sense this proinflammatory cytokine was critical for the response ( Fig . 7B , C ) . In the acutely infected group LM infection caused between 66-91% , of LCMV GP33-specific memory CD8 T cells to produce IFN-γ ( Fig . 7E ) . This response was severely curtailed in hosts undergoing protracted or chronic LCMV infections as only 11-35% and 0 . 9-9% , respectively , of the GP33-specific CD8 T cells became IFN-γ positive following LM co-infection , and similar trends were observed for the GP276 viral epitope-specific responses ( Fig . 7E ) . We further checked whether LM infection also resulted in increased CD25 expression on the LCMV-specific CD8 T cells . Again , although the memory CD8 T cells did upregulate CD25 , the responsiveness of the exhausted cells was significantly diminished , which was consistent with the observations of IFN-γ production ( Fig . 7D , F ) . Thus , whereas memory CD8 T cells can vigorously respond to sets of inflammatory cytokines and certain bacterial infections , the development of exhaustion is associated with downregulation of IL-18Rα and the inability to respond to both the underlying viral infection and the inflammatory cytokine milieu . Despite this clear cut defect in the innate-like properties of virus-specific CD8 T cells in the protracted and chronically infected cohorts , by 20 hr following LM infection bacterial titers were somewhat lower in the spleens ( Fig . 7G ) and livers ( Fig . 7H ) of these mice . This paradoxical finding indicates that although exhausted CD8 T cells lose their ability to respond to changes in the levels of inflammatory cytokines induced by the bacterial challenge , other immunological alterations in the chronically infected host help confer resistance to secondary infections .
Our investigation of the sensitivity of virus-specific CD8 T cells to stimulation with combinations of IL-12 , IL-18 , and IL-21 highlights the divergence between memory and exhausted T cells [1] , [12] , [13] . The effector populations that develop during the early stages of acute , protracted , and chronic LCMV infection are more similar , and in all cases a substantial fraction express the IL-18Rα and produce IFN-γ as well as increase expression of CD25 , in response to antigen-independent stimulation by the cytokines tested . Distinct differences manifest as constituents of the effector pool transition into the memory compartment following resolution of the acute infection , and as the exhausted state progressively develops if the infection persists . The memory cells are clearly distinguished by the maintenance of IL-18Rα expression and retain the ability to respond to activation by specific cytokines in the absence of antigen . By contrast , as exhausted CD8 T cells emerge in persistently infected mice , the IL-18Rα is downregulated . This mirrors the reduction in IL-18Rα observed on HIV-specific CD8 T cells , suggesting that decreased levels of IL-18Rα is a common feature of exhaustion [23] . As shown in this report , the loss of IL-18Rα on exhausted CD8 T cells correlates with their failure to prominently react to both in vitro activation with specific cytokine combinations and also to in vivo exposure to bystander LM infection . Although bona fide memory CD8 T cells are highly sensitive to both TCR triggers and cytokine mediated signals that induce IFN-γ , the well-described inability of severely exhausted T cells to produce marked amounts of IFN-γ in response to TCR-dependent antigenic activation [12] , [13] , [15] , [17]–[19] is not rescued by cytokine exposure . Virus-specific memory CD8 T cells not only respond to cytokine activation and bystander LM infection by producing IFN-γ but also upregulate expression of the IL-2Rα chain , CD25 . The synergistic ability of IL-18 in combination with either IL-12 or IL-21 to induce CD25 may serve to pre-trigger memory CD8 T cells encountering an inflammatory environment and other helper mediators , and this possibly augments their proliferation and reacquisition of effector traits if antigenic signals are also received . This response is abolished in exhausted CD8 T cells and likely further impedes their responsiveness to either ongoing antigenic activation or co-infection associated danger signals . Thus , as virus-specific T cells succumb to exhaustion , they adjust to the sustained presence of antigen as well as the unique cytokine milieu that occurs in the persistently infected host , and appear to undergo a global desensitization in their ability to respond to effector function-inducing signals . The upregulation of IL-18Rα transcript levels occurs during the contraction phase of the response , and several studies have shown that memory T cells express high levels of IL-18Rα [14] , [23] , [24] . Nevertheless , memory T cell formation and maintenance does not require IL-18 signals as the kinetics , magnitudes , and longevity of LM-specific CD8 T cell responses in IL-18- and IL-18Rα- deficient mice resemble those observed in immunocompetent animals [25] . During chronic LCMV infection , exhausted T cells form and also remain present even though expression of IL-18Rα becomes downregulated . Our analyses of mixed bone marrow chimeras show that although both IL-18Rα+/+ and IL-18Rα-/- CD8 T cells do initially participate in the response to LCMV cl13 infection , the virus-specific IL-18Rα-/- CD8 T cells are outcompeted by their IL-18Rα+/+ counterparts following the peak of the effector phase of the response . The proportion of IL-18Rα-/- T cells subsequently stabilizes , and this coincides with the decrease in receptor expression on the index IL-18Rα+/+ population . The apparent preferred maintenance of IL-18Rα+/+ CD8 T cells during the contraction phase of the response is consistent with the documented pro-survival functions of IL-18 and its roles in limiting activation-induced cell death [26] . IL-18 has also been shown to increase the proliferation of memory CD8 T cells as they mount a recall response upon reencountering presented antigen [27] . In chronically infected mice viral antigen is present throughout the response , and therefore it is plausible that in this setting IL-18 exerts a dual survival and proliferative role . These effects are temporally limited due to the progressive downregulation of IL-18Rα that occurs as exhaustion develops . The prototypic memory CD8 T cells and exhausted T cells that are present following the contraction phase of the response have distinct maintenance requirements [28] , [29] . Memory CD8 T cells are maintained at remarkably stable levels in the absence of antigen , principally due to the common-γ chain cytokine family members IL-7 and IL-15; however , the cognate receptors for these homeostatic cytokines are downregulated on exhausted CD8 T cells [28]–[31] . Instead , the continued presence of viral antigen appears to preserve the exhausted CD8 T cells and allows them to bypass the normal necessity for IL-7 and IL-15 . Interestingly , in humans expression of the IL-18Rα is linked with the ability of memory phenotype cells to reconstitute and “self-renew” following chemotherapy , and this property is also associated with the capacity of these cells to efflux Rh123 [21] . We confirmed this finding in acutely infected mice as the ability to efflux Rh123 was primarily detected in the IL-18Rαhi subset of cells; however , the IL-18Rαlo exhausted T cells were also similarly capable of effluxing Rh123 . This differs from the observations of Turtle et al . [21] , but likely reflects the unique properties of exhausted T cells , which are maintained despite downregulation of the IL-18Rα and retain the ability to efflux Rh123 . The establishment of virus-specific memory CD8 T cells which can detect and respond to increased levels of inflammatory and regulatory cytokines allows them to provide a broader level of immunological protection that is not governed by their precise antigen-specificity [5]–[7] , [10] . This is well described following LM infection of mice as memory CD8 T cells specific for non-LM encoded epitopes confer some level of protection against bacterial challenge [6] , [10] . We document clear differences in the ability of acute and chronic infections to elicit virus-specific CD8 T cell populations capable of detecting bystander LM infections . Our findings show that LCMV-specific memory CD8 T cells from acutely infected mice mount a marked bystander response to LM infection resulting in the production of IFN-γ and upregulation of CD25 . By contrast , the ability of exhausted virus-specific CD8 T cells to mount a non-antigen specific , innate-like , response to a bacterial co-infection is severely curtailed . The most pronounced exhaustion of CD8 T cells develops during chronic infection in the absence of sufficient CD4 T cell help . During HIV infection there is a profound loss of CD4 T cells in the gastrointestinal tract [32]–[34] . This is associated with translocation of microbial flora and products from the intestinal lumen and the detection of lipopolysaccharides and bacterial DNA in the plasma , which results in chronic immune activation and accelerated disease progression [35] , [36] . Therefore , it will be interesting to dissect inter-relationships between viral loads , CD8 T cell exhaustion , and the extent of microbial translocation following HIV infection , as well as during pathogenic and non-pathogenic SIV infections . There is increasing interest in understanding how underlying persistent viral infections have a generalized impact on immune responses [37] . Commonly , individuals are exposed to a myriad of infections and the collective induction or suppression of immune system functions caused by the “virome” likely mold the formation of subsequent responses . Although the innate-like response of exhausted LCMV-specific CD8 T cells to LM co-infection was significantly reduced , the bacterial burden in the LCMV chronically infected cohort was either no different or lower than that detected in the acutely infected group . This paradoxical finding suggests that other immunological attributes of the chronically infected host contribute to overall resistance or susceptibility to secondary or opportunistic infections . For example , macrophage activation caused as a result of acute LCMV infection has been proposed to account for increased resistance to LM infection [38] . Similarly macrophage activation in mice persistently infected with either murine γ-herpesvirus 68 or mouse cytomegalovirus confers protection against LM and Yersinia pestis challenge [39] . Conversely , other immunological alterations , such as ablation of type I IFN production by plasmacytoid dendritic cells , which occurs during chronic LCMV infection , can promote establishment of certain opportunistic infections [40] . The development of exhaustion associated with chronic LCMV infection clearly results in a profound disruption of usually highly responsive and effective anti-viral T cells . These alterations impact both their ability to respond to antigenic stimuli as well as inflammatory cytokines . Such changes in the properties of T cells occur in the context of global shifts in immune system functions which emerge as the hosts adjusts to the ensuing chronic infection , including disturbances in splenic architecture , and alterations in the composition and activation status of classic innate immune effectors . The collective changes in the immunological environment that occur likely act in concert to limit bacterial growth following LM challenge , compensating for the dysregulation in the innate-like properties of the virus-specific CD8 T cell population . Overall , the signature loss of IL-18Rα expression by exhausted virus-specific CD8 T cells represents one mechanism of re-calibrating the cellular immune response to ongoing chronic infection . This phenotypic shift likely represents one of many evolutionary adaptations that help prevent severe immunopathology . Nevertheless , the combined alterations to the host response caused by virus-persistence may help or hinder resistance to new or current infections .
All procedures with experimental mice were approved by the University of Alabama at Birmingham Institutional Animal Care and Use Committee in accordance with NIH guidelines . C57BL/6J ( B6 ) , C57BL/6-Cd4tm1Mak/J ( CD4-/- ) , B6 . SJL-PtprcaPepcb/BoyJ ( CD45 . 1 ) , B6 . 129S7-Rag1tm1Mom/J ( Rag-1-/- ) , and B6 . 129P2-Il18r1tm1Aki/J ( IL-18Rα-/- ) mice were originally purchased from Jackson Laboratory ( Bar Harbor , ME ) . All mice were bred and/or maintained in fully accredited facilities at the University of Alabama at Birmingham . For acute infections mice were infected by i . p . injection with 2×105 PFU LCMV-Armstrong ( Arm ) . Protracted and chronic infections were established by i . v . inoculation with 2–4×106 PFU LCMV-clone 13 ( cl13 ) into B6 and CD4-/- mice , respectively [17] . In certain experiments 0 . 96–3 . 3×106 colony-forming units ( CFU ) of the 10403S strain of Listeria monocytogenes ( LM ) ( kindly provided by Dr . D . Portnoy , University of California , Berkeley , CA ) was administered by i . v . injection into mice that had been infected with LCMV 59-81 days previously . To determine LM titers 20 hr following co-infection , suspensions of splenocytes were diluted with an equal volume of 0 . 5% Triton X-100 ( Fisher Scientific , Fair Lawn , NJ ) . Livers were collected into PBS , weighed , diluted with an equal volume of 0 . 5% Triton X-100 and then homogenized [essentially as in 41] . Numbers of CFU were determined by plating serial dilutions on BHI agar plates . Splenocytes and blood samples were processed as previously described [42] . For LM co-infection studies splenic samples were collected , prepared , and cultured for 3hr at 37°C in medium without antibiotics but containing 10 µg/ml Brefeldin A ( Sigma-Aldrich , St . Louis , MO ) prior to staining and flow cytometric analyses [6] . Splenocytes were cultured for 5 . 5 hr at 37°C in the presence or absence of recombinant murine IL-12 ( Biosource/Invitrogen , Camarillo , CA or Peprotech , Rocky Hill , NJ ) , IL-18 ( Biosource/Invitrogen , Camarillo , CA ) , IL-21 ( R&D Systems , Minneapolis , MN or Peprotech , Rocky Hill , NJ ) , or various combinations of the three cytokines . All cytokines were used at a final concentration of 20 ng/ml . Brefeldin A ( Golgi Plug , BD Biosciences , San Jose , CA ) was added for the last 1 . 5 hr of culture to facilitate the intracellular accumulation of IFN-γ . Surface and intracellular staining was performed essentially as previously described [17] . All samples were pre-treated with anti-CD16/CD32 mAb ( clone 2 . 4G2 ) ( UAB Immunoreagent Core ) prior to staining . Surface staining was performed using various combinations of anti-CD44 ( clone IM7 , BD Biosciences ) , anti-CD25 ( clone 3C7 or PC61 , Biolegend , San Diego , CA ) , anti-IL-12Rβ2 ( clone 305719 , R&D Systems ) , anti-IL-18Rα ( clone 112614 , R&D Systems ) , and anti-IL-21R ( clone 4A9 , Biolegend ) mAbs together with anti-CD8α clone 53-6 . 7 ( eBioscience ) , and PE or allophycocyanin conjugated MHC class I tetramers . MHC tetramers were produced in house or obtained from the National Institute of Allergy and Infectious Diseases tetramer core facility , Atlanta , GA . For intracellular staining the anti-IFN-γ antibody XMG1 . 2 was used ( eBioscience ) . All samples were acquired on an LSRII flow cytometer ( BD Biosciences ) , and data were analyzed using FlowJo software ( Tree Star , Ashland , OR ) . Splenocytes were loaded with Rhodamine 123 ( Rh123 ) ( Invitrogen or Sigma-Aldrich ) in RPMI-1640 supplemented with 10% FCS , 50 µM β-mercaptoethanol , 100 U/ml penicillin , and 100 µg/ml streptomycin for 30 minutes on ice . Samples were then washed and cultured for 1hr at 37°C in the absence or presence of either 0 . 1 µg/ml Cyclosporine A ( CsA ) ( Sigma-Aldrich ) or 5 µg/ml Vinblastine ( Sigma-Aldrich ) . After incubation cells were washed and stained for CD8α and IL-18Rα together with PE-conjugated DbGP33 MHC tetramers , as described above . Samples were resuspended in 0 . 1% BSA and 0 . 01% NaN3 in PBS prior to flow cytometric analyses , and Rh123 fluorescence was detected using a 530/30 bandpass filter . Chimeras were generated essentially as previously described [42] . Bone marrow from CD45 . 1 ( IL-18Rα+/+ ) , CD45 . 2 ( IL-18Rα-/- ) , and CD45 . 2 ( IL-18Rα+/+ ) B6 mice was T cell depleted using anti-CD5 ( Ly-1 ) microbeads ( Miltenyi Biotec , Auburn , CA ) . Recipient Rag-1-/- mice were administered a split dose of radiation to give a total exposure of ∼1000 rads . These recipient mice were then reconstituted by i . v . injection with an approximate 50∶50 ratio ( 1 . 4×106 total cells ) of either CD45 . 1 ( IL-18Rα+/+ ) : CD45 . 2 ( IL-18Rα+/+ ) bone marrow , to generate a control cohort , or CD45 . 1 ( IL-18Rα+/+ ) : CD45 . 2 ( IL-18Rα-/- ) bone marrow , to generate an experimental cohort . The mice were infected with 4×106 PFU cl13 at 9 or 11 weeks after reconstitution . One-way ANOVA was used to determine statistical significance . P values were calculated using Prism software ( Graph Pad , San Diego , CA ) . Statistical significance is defined as *p<0 . 05 , **p<0 . 01 , ***p<0 . 001 .
|
Ongoing viral infections can corrupt the immune defense system . One way in which this occurs is by the induction of exhaustion in the host's anti-viral CD8 T cells , a key component of the immune response . Investigating the causes and consequences of T cell exhaustion will likely provide insights into potential strategies for improving immunity to ongoing infections . In addition , there is an increasing interest in determining how underlying chronic infections broadly impact immune responses to new infections . In this study we demonstrate that exhausted anti-viral T cells downregulate the receptor for the immune system factor IL-18 . This downregulation is consequential as it prevents the function of exhausted cells from being rescued by certain combinations of immune system factors , cytokines , which potently activate conventional , highly protective responses . Moreover , the loss of this receptor disfavors the accumulation of anti-viral CD8 T cells during the initial phase of the response . We also show that whereas conventional anti-viral CD8 T cells can sense a bacterial co-infection that induces IL-12 and IL-18 production , the exhausted cells fail to respond . Collectively , these studies reveal how components of the immune defense system become recalibrated during an ongoing infection , altering their ability to respond to certain new infections .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immune",
"cells",
"immunity",
"to",
"infections",
"immunology",
"microbiology",
"adaptive",
"immunity",
"immune",
"defense",
"animal",
"models",
"of",
"infection",
"t",
"cells",
"microbial",
"pathogens",
"biology",
"immune",
"response",
"viral",
"persistence",
"and",
"latency",
"immunity",
"virology",
"co-infections"
] |
2011
|
Exhausted CD8 T Cells Downregulate the IL-18 Receptor and Become Unresponsive to Inflammatory Cytokines and Bacterial Co-infections
|
Muscle coordination studies repeatedly show low-dimensionality of muscle activations for a wide variety of motor tasks . The basis vectors of this low-dimensional subspace , termed muscle synergies , are hypothesized to reflect neurally-established functional muscle groupings that simplify body control . However , the muscle synergy hypothesis has been notoriously difficult to prove or falsify . We use cadaveric experiments and computational models to perform a crucial thought experiment and develop an alternative explanation of how muscle synergies could be observed without the nervous system having controlled muscles in groups . We first show that the biomechanics of the limb constrains musculotendon length changes to a low-dimensional subspace across all possible movement directions . We then show that a modest assumption—that each muscle is independently instructed to resist length change—leads to the result that electromyographic ( EMG ) synergies will arise without the need to conclude that they are a product of neural coupling among muscles . Finally , we show that there are dimensionality-reducing constraints in the isometric production of force in a variety of directions , but that these constraints are more easily controlled for , suggesting new experimental directions . These counter-examples to current thinking clearly show how experimenters could adequately control for the constraints described here when designing experiments to test for muscle synergies—but , to the best of our knowledge , this has not yet been done .
The muscle synergy hypothesis has received considerable attention in the neuroscience community ( see [1] for a review ) . It posits that the central nervous system ( CNS ) activates muscles using the flexible combination of a small number of patterns . This hypothesis is commonly motivated as a potential mechanism by which the nervous system can simplify the control of a large number of muscles [2] , [3] , [4] . Counter-examples to the muscle synergy hypothesis have been observed for the control of hand musculature [5] , [6] . We therefore set out to answer the question: is the human hand a unique system for not employing synergies , or are the muscle synergies detected in other neuromuscular systems actually of non-neural origin ? Answering this question is crucial to making progress in the field of motor neuroscience . The muscle synergy hypothesis has been notoriously difficult to prove or falsify [1] . Two distinct strategies have been employed to generate muscle activity to test this hypothesis: behavior in humans or animals , and direct stimulation of the motor system . The behavioral approach simply observes the electromyographic ( EMG ) activity in a large number of muscles during natural motor behavior , and uses computational techniques to identify consistent structure in the EMG signals across different tasks [3] , [7] , [8] . The stimulation approach artificially excites a variety of locations in the nervous system and shows that a relatively small number of muscle activation patterns emerge [9] . The behavioral approach has the advantage that it can be applied to a human or completely intact animal during natural behavior , but has the disadvantage that the task constraints could favor particular muscle activation patterns , independent of neural control [1] . The stimulation approach has the advantage that it is unaffected by the task constraints , but it is unclear whether the complete repertoire of muscle activation patterns can be elicited by these techniques [9] . Thus , existence of muscle synergies of neural origin has not been conclusively proven . Muscle coordination studies using the behavioral approach are more relevant to natural human behavior [8] and disease states [7] , and repeatedly show that muscle activations are constrained to a low-dimensional subspace across a variety of tasks . This potential evidence for the muscle synergy hypothesis comes from a number of behavioral studies , including cat postural control [3] , [10] , human postural control [11] , [12] , [13] , human arm control [8] , [14] , human leg control [15] , primate grasping [16] , and natural lower limb behaviors of the frog [2] . The basis vectors of these low-dimensional subspaces are often called muscle synergies , and are taken to represent the underlying neural strategies to simplify multi-muscle control . An important class of behavioral experiments examining the muscle synergy hypothesis examines EMG responses to external perturbations ( e . g . [3] , [10] , [12] , [13] , [14] ) . In this work , we show that such low dimensionality induced by external perturbation can be a product of unavoidable constraints related to movement . Another important class of behavioral experiments examines EMG during voluntary activation of muscles ( e . g . [7] , [8] , [15] ) . In this work , we show that low EMG dimensionality during voluntary muscle force production could be related to task selection . Thus , we are fundamentally questioning the utility of the behavioral approach and the validity of its interpretation , because the synergies detected by these methods may not uniquely reflect neural strategies to simplify the control of multiple muscles .
The behavioral approach to muscle synergies involves examining a limb controlled by multiple muscles ( Figure 1A ) . The limb is moved , either voluntarily or externally , in a large number of directions in its workspace ( Figure 1A ) . The set of EMG vectors , each of which describes the activity in multiple muscles for a particular direction of movement , is observed to be low-dimensional ( Figure 1A ) . This observation is used to support the muscle synergy hypothesis , which posits that muscles can only be activated in groups ( Figure 1A ) . We begin by illustrating constraints that could appear as low-dimensional EMG without the muscle synergy hypothesis being true . We first do this graphically in simple system so it is clear that they apply generally and are not specific to any particular system or an artifact of a particular computational model ( Figure 1 ) . We then proceed to the detailed analysis of realistically complex neuromuscular systems . The th element of moment arm matrix , denoted , is the moment arm of the th muscle about the th joint . The angle of joint is denoted , and the length of muscle is denoted . Any movement of this leg around a particular posture ( Figure 1B ) will induce changes in muscle length given by the equation . These unavoidable muscle length changes can be visualized as lying in a subspace ( plane ) spanned by two basis vectors , which are in fact the moment arms grouped by joint ( Figure 1B ) . To account for the causal interaction between musculotendon length changes and EMG during behavioral experiments with external perturbations , we perform a simple thought experiment . What pattern of EMG would we expect to see if there were no neural muscle synergies controlling muscles in groups , but each muscle independently resisted lengthening during the perturbation ( Figure 1B ) ? This scenario would lead to increased EMG if the muscle were stretched , but no EMG if the movement induced the muscle to passively shorten , and would have the effect of stabilizing the reference posture . Examining the predicted EMG , we see that it would still be low-dimensional ( Figure 1B ) , with 2 principal components accounting for 96% of the data variance . However , this low-dimensionality is not related to any neural controller designed to control muscles in groups ( none was active ) . Rather , the low dimensionality arises naturally from biomechanical constraints and independent response of each muscle . We refer to these as feedback-related muscle synergies because they are mediated by afferent information . Notice that 2 synergies do not completely account for all of the simulated EMG variance , despite the fact that the external perturbations are 2-dimensional . This arises from the nonlinear relation between musculotendon length change and the resulting EMG . Even if a task is internally driven and there is no external perturbation , the set of muscle activations will have a low dimensional structure even when the limb endpoint is driven in an exhaustive set of directions ( Figure 1C shows the case of omnidirectional static force production ) . We refer to these as feedforward-related muscle synergies because the low dimensional structure of the muscle activations arises directly from the structure of the set of feasible motor commands . Because of muscle redundancy , a range of different muscle coordination patterns equivalently produce a same endpoint force vector ( Figure 1C ) . The muscle coordination patterns that produce any single endpoint force vector are themselves a low-dimensional subset of muscle force space ( in this case , a line that we accurately computed for this schematic model ) . But , perhaps counter-intuitively , even when all options for endpoint force in all directions are combined , the set of options available to the CNS is still low-dimensional ( approximately spanned by only 2 principal components in this case ) . This is because the experiment , however exhaustive , still constrains the magnitude and direction of test forces . Therefore , the experimental design automatically constrains the observed combinations of muscle activity to a low-dimensional subspace , which could be misinterpreted as neurally-generated muscle synergies . We found experimental evidence of feedback-related and feedforward-related muscle synergies in a cadaveric human hand ( Figure 2 ) and evidence for them in a realistic model of the human leg , as described below . To demonstrate feedback-related and feedforward-related muscle synergies , we actuated the seven tendons of cadaveric index fingers with computer-controlled motors ( Figure 2 ) . As in prior work , we resected four fresh frozen cadaver arms at the mid-forearm level and dissected them to reveal the proximal end of the insertion tendons of all seven muscles controlling the index finger [17]: flexor digitorum profundus ( FDP ) , flexor digitorum superficialis ( FDS ) , extensor indicis ( EI ) , extensor digitorum communis ( EDC ) , first lumbrical ( LUM ) , first dorsal interosseous ( FDI ) , and first palmar interosseous ( FPI ) . We fixed the specimen rigidly to a tabletop using an external fixator ( Agee-WristJack , Hand Biomechanics Lab , Inc . , Sacramento , CA ) , and we tied and glued the proximal tendons to Nylon cords attached to rotational motors . A real-time controller and custom-written software controlled the motors . Load cells measured the tension in each cord , which was fed back to the motor so that a desired amount of tension could be maintained on each tendon . A motion capture system ( Vicon Motion Systems , Oxford , UK ) recorded the angles of all index finger joints . To demonstrate feedback-related muscle synergies the experimenter moved the finger in its workspace while changes in tendon length were recorded ( Figure 2B ) . The motors actively maintained 5 N of tension on each tendon to prevent slackness . We generated movements of the fingertip at random until we filled the planar workspace of the finger around a starting posture ( Figure 2C ) . We examined two postures , one with the index finger more extended and one with the index finger more flexed . To demonstrate feedforward-related muscle synergies , we rigidly secured the index fingertip to a 6 DOF load cell ( JR3 , Woodland , CA ) ( Figure 2D ) ) . A pre-programmed sequence of tension was then delivered to the tendons . An “active” tendon had 10 N applied to it , whereas an “inactive” tendon had 0 N applied to it . We delivered all possible activity combinations ( for seven tendons there are 128 combinations: all possible combinations of 1 muscle active , 2 muscles active , 3 muscles active , … , all muscles active ) in sequence to the specimen , holding each combination for 3 seconds and recording the average fingertip wrench ( all forces and torques ) exerted during this 3 second period . We then performed linear regression on the fingertip wrench using the applied tendon tension as the independent variable . This regression provided an action matrix that predicted the fingertip wrench vector given the muscle activation vector ( Figure 2E ) , and allowed us to quantify the goodness of that prediction . This action matrix included estimates for the maximum muscle force for the index finger muscles [17] . Analysis of the action matrix using computational geometry ( see below ) revealed all possible muscle coordination pattern options for endpoint force in all directions . We developed a model of the human leg to demonstrate that our results also apply to other parts of the motor system . To analyze feedback-related muscle synergies , we constructed a leg model using the 44 muscles and moment arms contained in a previously-validated lower extremity model [18] . We obtained the OpenSim implementation of this model from the Neuromuscular Models Library ( simtk . org ) . For simplicity and without loss of generality , we only considered sagittal plane movement . We extracted the moment arms for the hip , knee , and ankle as a function of posture , and generated a moment arm matrix . In addition , we extracted the estimated maximum isometric force for each muscle , and generated a diagonal maximum force matrix . The Jacobian matrix was derived for a three-link planar manipulator [19] , based on estimated anthropomorphic lengths for a 170 cm tall male [20] . We also adapted a well-accepted 14-muscle version [21] to analyze feedforward-related muscle synergies . We found that the 44-muscle model was of too high a dimensionality ( 44-D ) for current computational geometry algorithms ( see below ) , as complexity grows exponentially in the number of muscles . Fourteen muscles are on the same order as number of muscle recordings used to test for synergies , thus this model demonstrates the principles at work without loss of generality . The 14-D model contained 14 muscles/muscle groups ( muscle/muscle group abbreviation in parentheses ) : medial and lateral gastrocnemius ( gastroc ) , soleus ( soleus ) , tibialis posterior ( tibpost ) , peroneous brevis ( perbrev ) , tibialis anterior ( tibant ) , semimembranoseus/semitendenosis/biceps femoris long head ( hamstring ) , biceps femoris short head ( bfsh ) , rectus femoris ( rectfem ) , gluteus medialis/glueteus minimus ( glmed/min ) , adductor longus ( addlong ) , iliacus ( iliacus ) , tensor facia lata ( tensfl ) , gluteus maximus ( glmax ) . We obtained the maximal torque output for these muscle groups by averaging moment arms from the 44-muscle model , weighted by the estimated maximum isometric force . In so doing , the muscle groups have the same torque generating capabilities as the 44 individual muscles .
We find that the dimension of the simulated EMG associated with feedback-related muscle synergies reflects the low dimensionality of the movement , creating the appearance of muscle synergies even in the absence of a specific neural controller ( Figure 3 ) . Movements of the fingertip had more than 80% of their variance explained by 2 principal components , and thus were largely confined to a plane ( Figure 3A ) . These movements induced low-dimensional ( synergistic ) patterns of simulated EMG ( Figure 3B ) that mirrored the dimension of the movement ( Figure 3C ) . We tested if such feedback-related muscle synergies would exhibit features thought to support the neural-origin of muscle synergies . We chose to examine generalization across posture [10] , meaning that the low-dimensional basis of muscle synergy vectors extracted from muscle activity in one posture could be used to reconstruct the muscle activity in a different posture . For our 44-muscle computational model of the human leg , 5 principal components were sufficient to explain more than 80% of the variance in simulated EMG across movements in 16 directions in a reference posture ( Figure 4A ) . These 5 principal components were then used to reconstruct the simulated EMG from a different posture ( test posture ) , in a test of synergy generalization ( Figure 4B ) . A map of test postures where the reconstructed EMG accounted for more than 80% of the simulated EMG in the test posture indicates that synergy generalization is expected over a wide range of postures , without indicating that the muscle synergy hypothesis is true . An example of one test posture in which synergy generalization would be expected is shown ( Figure 4C ) . We demonstrate feedforward-related muscle synergies first on the cadaver index finger . We find , regardless of force magnitude ( Figure 5A ) , that the set of muscle coordination patterns associated with fingertip forces in all directions is low-dimensional ( Figure 5B ) . Thus , a low-dimensional muscle activation space should not be surprising in experiments with hand musculature , and does not by itself suggest a specific simplifying neural controller . We also demonstrate feedforward-related muscle synergies on our simplified 14-muscle computer model of the human leg ( Figure 6A ) . Again , regardless of force magnitude ( Figure 6B ) , the set of muscle coordination patterns associated with foot forces in all directions is low-dimensional ( Figure 6C ) . However , it is clear from the rather large dimensionality of feedforward-related muscle synergies for the leg ( 7 synergies of 14 muscles are required to account for 80% of force variance ) that the dimensionality of feedforward-related muscle synergies is not limited by the dimension of the task . Thus , while a low-dimensional muscle activation space should also not be surprising in experiments with leg musculature , and does not by itself suggest a specific neural controller , very low-dimensional EMG data during isometric force production would not be predicted by feedforward-related muscle synergies of the human leg . However , such predictions must be done using a biomechanical model on a experimental-specific basis for the limb being examined .
We performed this study to test whether non-neural constraints could produce the dimensionality reduction hypothesized to reflect neurally-established functional muscle groupings to relieve higher brain centers from controlling numerous muscles individually [1] , [5] , [6] . By providing clear counterexamples for limbs exerting endpoint static forces—or moving—in multiple directions , we demonstrate that two previously unrecognized non-neural constraints among muscles can also enforce such low-dimensionality , and thus give the appearance of muscle synergies . By confirming this alternative explanation to prior data and their interpretation , our work brings to light two fundamental non-neural constraints that need to be understood before muscle synergies of neural origin can be confidently disambiguated , found , and studied . We believe that properly controlling for these non-neural constraints is possible , and will enable studies that are capable of testing whether neural constraints are indeed present to reduce the dimensionality of the controller ( i . e . , the muscle synergy hypothesis ) . Thus , for example , it may be premature to attribute the emergence of such low dimensionality to neural sources [7] before other alternatives have been ruled out . To the best of our knowledge , the threshold for proving muscle synergies of neural origin exist has not been met because no study has adequately controlled for these non-neural interactions . While other authors have previously used biomechanical models to explain muscle synergies [24] , [25] , to our knowledge no previous study has asked the more fundamental question of whether an experimenter could be led to conclude that the muscle synergy hypothesis was true when , in fact , constraints among EMG activity among muscles were coming from different sources . Before discussing our results , we point out a potential source of confusion . If the nervous system is clearly generating the observed muscle activation patterns , how could one say that muscle synergies can be of non-neural origin ? The key here is to distinguish between choices of motor commands the nervous system makes vs . constraints on the feasible motor commands the nervous can use [26] . If the nervous system could have used a large variety of different muscle coordination patterns for a given task , and only certain patterns are ever observed , then clearly some specific neural strategy selected the observed patterns . However , if only a given variety of muscle coordination patterns are feasible for a given task ( due to , say , musculoskeletal geometry or the experimenter's selection of tasks ) , then clearly detecting a small variety of muscle coordination patterns does not suggest any specific neural strategy . Therefore , the goal of this study is to simply show by counterexample that non-neural interactions can also give rise to the dimensionality reduction thought to support the existence do muscle synergies of neural origin . Along the feedback-related front , our study revealed a previously-unappreciated similarity between low-dimensional muscle synergies and a dimensionality reduction arising from coupling among muscle moment arms . In the muscle synergy hypothesis , this low-dimensional subspace is interpreted as a reflection of the CNS controlling the musculature with a small number of activation patterns . We show that such dimensionality reduction can also arise from mechanical coupling because there are far fewer joints than muscles . Our simulated EMG data came from a thought experiment that assumed that each muscle independently resists ( and produced EMG ) when a small externally imposed perturbation causes its lengthening [3] , [10] , [12] , [13] , [14] . This transformation between musculotendon length changes and EMG preserved the low-dimensionality of length changes and reflected them in the EMG signals . Of course , this is a simple transformation that may not apply to all experimental preparations . However , this transformation would produce low-dimensional EMG without the muscle synergy hypothesis being true , including generalization of EMG synergies across postures . It is likely that a wide variety of “well-behaved” transformations ( e . g . monotonic ) between length change and EMG will produce the same results found here . We found similar results for linear , exponential and sigmoidal relationships between muscle stretch and EMG . It is , therefore , the burden of the experimenter using the behavioral approach with external perturbations to demonstrate whether or not the actual transformation is causing EMG to reflect the low-dimensionality of induced muscultendon length changes . Feedback-related muscle synergies are strikingly similar to EMG synergies used in support of the muscle synergy hypothesis both in their low-dimensionality and in their generalization across posture [10] , [12] , [13] . Along the feedforward-related front , we have also demonstrated reasons to doubt available evidence for synergies of neural origin . Feedforward-related muscle synergies are inevitable in paradigms studying the generation of voluntary muscle force in redundant muscle systems . We show that , even in these presumably exhaustive explorations of force production , i . e . when endpoint forces of the same magnitude are generated in multiple directions , the mechanically defined set of muscle activation options available to the CNS is low-dimensional in the absence of any neural interactions . Feedforward-related muscle synergies in muscle activation space emerge , not from specific neural interactions , but from unconstrained variation in a task-irrelevant subspace . Thus , if experimentally-observed muscle synergies are feedforward-related , we would predict significant variation in muscle activation pattern among repeated trials of the same task . Such task-irrelevant inter-trial variability has been observed in muscle activation patterns [11] , supporting the strong possibility that muscle synergies are feedforward-related , and do not reflect a specific dimensionality-reducing strategy employed by the CNS . Whereas we explored feedforward-related muscle synergies in the isometric context , coordination patterns would be similarly constrained during the production of voluntary feedforward movement because the equations of motion of the limb , combined with the desired trajectory in state space , define a manifold of feasible solutions for the control of movement [26] , [27] . We demonstrated feedforward-related muscle synergies using both cadaveric data and a simplified 14-muscle sagittal-plane computational model of the human leg . While it would have been ideal to analyze feedforward-related muscle synergies in the full 44-muscle leg model , vertex enumeration algorithms in computational geometry simply do not accommodate such large dimensionality . The simplified 14-muscle model suffices to provide the necessary counterexample to studies of muscle synergies for several reasons . It had many more muscles than endpoint degrees-of-freedom ( 14 muscles , 2 translational and 1 rotational DOFs at the endpoint ) . It had been previously validated and employed [21] . And 14 is on the order of the number of muscles that have been recorded from in studies of muscle synergies in the human leg ( [12] , [13] - 16 muscles ) . Thus we found that all muscle coordination patterns that produced endpoint force in all directions are embedded in a low-dimensional subspace of muscle activation space using three different approaches: a three-muscle schematic model ( Figure 1 ) , a 7-muscle model constructed directly from cadaveric data ( Figure 6 ) , and the 14-muscle leg model ( Figure 7 ) . Therefore , we conclude that feedforward-related muscle synergies are a general feature of neuromuscular systems producing forces in multiple directions , as demonstrated by three lines of evidence from computational and experimental data . We note that in the posture examined , analysis of this simplified leg model showed the least steep curve in variance explained vs . number of PC's . Nevertheless , these results support our conclusions because the aim of our work is demonstrate that there is real and unavoidable contribution of non-neural constraints on the low dimensionality of the neural command . In all cases this effect is real as the curves clearly depart from the unity line indicating no non-neural effects . So long as our examples show an effect of non-neural constraints , our point is made . An important , but often underestimated , issue in the literature is the subjective question of what constitutes a “significantly steep” slope that shows important neural or non-neural effects , and when does it begins to affect the interpretation of the data ( for a detailed overview of estimation of degrees of freedom in motor systems and critical evaluation of PCA , please see [28] ) . Whether or not there is a significant non-neural effect is a question for researchers to decide depending on the goals of the study—and is beyond the scope of this work . Moreover , a motivation for our experimental work was to sidestep the tangential debate about whether or not any given leg or arm model is appropriately complex ( see [29] ) . Our experiments to study force-related muscle synergies using actual cadaveric index fingers in multiple postures at their full natural complexity—where we know the ground truth of tendon actuation—reveal that as few as two synergies can account for more than 80% percent of force variance in seven muscles . This ratio of non-neural synergies that can explain c . 80% of the variance to number of muscles examined ( 2/7 = 0 . 29 , see Figure 5B ) is comparable to that attributed to synergies of neural origin in some studies , such as 5/19 = 0 . 26 in human reaching [8] , 4/15 = 0 . 26 in cat postural control [3] and 6/16 = 0 . 38 in human postural control [12] . Therefore , our results provide direct counterexamples that strongly suggest that analyzing muscle synergies in the context of experiment-specific biomechanical models ( Figure 7 ) is necessary to determine if feedforward-related muscle synergies could interfere with the interpretation of EMG data . Can muscle synergies of neural origin ever be found ? We use a flowchart to summarize how our findings suggest a new way forward to finding muscle synergies of neural origin ( Figure 7 ) . The key is to disambiguate synergies of neural origin from potential confounds . Our work reveals that feedback-related muscle synergies can be controlled for using a model that predicts musculotendon length changes on the basis of estimates of muscle activity , like EMG . We do not believe that such a model is currently feasible , largely because existing detailed mechanistic models of EMG apply to isometric contractions only [30] and often fail to realistically replicate fundamental features of EMG [31] . In contrast , current EMG techniques are a more reasonable estimate of activation level ( and muscle force ) than of musculotendon length change . The well-known limitations of EMG for these applications are discussed elsewhere [23] , [26] , [32] , [33] , but EMG nevertheless remains a reasonable and practical tool . Thus , our work reveals that force-induced muscle synergies could be controlled for in two ways . The first approach would be an experimental paradigm which could proceed with an incomplete sampling of the feasible force output of the limb . For example , the experiment could employ endpoint outputs in every direction as is currently used , but it would be coupled with an experiment-specific computational model to estimate the features of muscle activation that are not explained by the constraints of the task . Such a computational model could predict the possible coordination patterns , and determine if muscle synergies were observed simply because the possible coordination patterns were low-dimensional . This approach has not yet been taken to study muscle synergies , but we believe it to be the most productive way forward . A prior studies has approximated such approach [6] , [34] , and their conclusions did not support the presence of synergies of neural origin . The second approach would be an experimental paradigm that requires the limb to generate every possible endpoint output ( in magnitude and direction ) . If dimensionality reduction in muscle activation space were observed , it could then be unambiguously interpreted as muscle synergies of neural origin . While the feasible force set has been estimated in human subjects [35] , such experiments may prove prohibitively long or fatiguing while maintaining intramuscular electrodes in position to reliably record from all muscles of the limb or finger . Disambiguating muscle synergies of neural origin from those of non-neural origin is essential not only for basic research in motor neuroscience , but also for clinical populations . Muscle synergy structure has been reported to be similar between the affected and unaffected arms of stroke survivors [7] . Since the biomechanical structure of these limbs may be very similar , non-neural feedback-related and feedforward-related muscle synergies could have the primary contributors to this finding . Thus , controlling for these non-neural interactions may be essential to designing the most effective rehabilitation strategies .
|
How the brain and spinal cord control the body is a fundamental question of critical scientific and clinical importance . The preferred experimental approach to answer this question has been to infer the neural control strategy by analyzing recordings of muscle activity and limb mechanics collected while animals and people use their limbs . This has led to a popular , but not yet proven , hypothesis that the brain and spinal cord simplify the control of the numerous muscles by grouping them into few functional units called neural synergies . Our detailed experiments and simulations challenge the utility of this approach and the validity of its interpretation . We point out that mechanical constraints can also explain those experimental recordings . In particular , the anatomy of the limb combined with the type of tasks studied and analysis used , suffice to give the appearance of neural synergies . To be clear , we do not disprove the neural synergy hypothesis . Rather , in the tradition of scientific debate , by showing an alternative explanation to the available data we challenge the community and ourselves to design novel experiments and analyses to conclusively test that hypothesis by ruling out the confounds we point out .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"physics",
"medicine",
"diagnostic",
"medicine",
"clinical",
"neurophysiology",
"electromyography",
"computational",
"neuroscience",
"physiology",
"biology",
"biophysics",
"neuroscience",
"biomechanics"
] |
2012
|
Challenges and New Approaches to Proving the Existence of Muscle Synergies of Neural Origin
|
Mycobacterium leprae ( M . leprae ) lives and replicates within macrophages in a foamy , lipid-laden phagosome . The lipids provide essential nutrition for the mycobacteria , and M . leprae infection modulates expression of important host proteins related to lipid metabolism . Thus , M . leprae infection increases the expression of adipophilin/adipose differentiation-related protein ( ADRP ) and decreases hormone-sensitive lipase ( HSL ) , facilitating the accumulation and maintenance of lipid-rich environments suitable for the intracellular survival of M . leprae . HSL levels are not detectable in skin smear specimens taken from leprosy patients , but re-appear shortly after multidrug therapy ( MDT ) . This study examined the effect of MDT components on host lipid metabolism in vitro , and the outcome of rifampicin , dapsone and clofazimine treatment on ADRP and HSL expression in THP-1 cells . Clofazimine attenuated the mRNA and protein levels of ADRP in M . leprae-infected cells , while those of HSL were increased . Rifampicin and dapsone did not show any significant effects on ADRP and HSL expression levels . A transient increase of interferon ( IFN ) -β and IFN-γ mRNA was also observed in cells infected with M . leprae and treated with clofazimine . Lipid droplets accumulated by M . leprae-infection were significantly decreased 48 h after clofazimine treatment . Such effects were not evident in cells without M . leprae infection . In clinical samples , ADRP expression was decreased and HSL expression was increased after treatment . These results suggest that clofazimine modulates lipid metabolism in M . leprae-infected macrophages by modulating the expression of ADRP and HSL . It also induces IFN production in M . leprae-infected cells . The resultant decrease in lipid accumulation , increase in lipolysis , and activation of innate immunity may be some of the key actions of clofazimine .
Leprosy is a chronic infectious disease caused by Mycobacterium leprae ( M . leprae ) , which is a typical intracellular pathogen that parasitizes tissue macrophages ( histiocytes ) and Schwann cells of the peripheral nerves of the dermis . Although its prevalence has declined over the last several decades due to the introduction of multi-drug therapy ( MDT ) by the World Health Organization ( WHO ) , leprosy remains a major public health problem in many developing countries: In 2010 , 228 , 474 new cases were registered worldwide [1] . Based on their clinical , histological and immunological manifestations , leprosy patients are classified into five groups that comprise one continuous spectrum: Tuberculoid ( TT ) , Borderline Tuberculoid ( BT ) , Borderline ( BB ) , Borderline Lepromatous ( BL ) and Lepromatous ( LL ) [2] . LL is characterized by widespread skin lesions containing numerous bacilli that live in the foamy or enlarged lipid-filled phagosome within macrophages . Schwann cells in LL nerves also have the foamy , lipid-laden appearance that favors mycobacterial survival and persistence . In Schwann cells , M . leprae infection-induced biogenesis of lipid droplets correlates with increased prostaglandin E2 ( PGE2 ) and interleukin-10 ( IL-10 ) secretion , which is essential for leprosy pathogenesis [3] , [4] . Although lipid-laden macrophages are also observed in other mycobacterial infections , including tuberculosis [5] , [6] , the amount of lipid and the number of infected macrophages are most prominent in cases of LL [7] , [8] . The PAT protein family is named after three of its members: perilipin , adipophilin/adipose differentiation-related protein ( ADRP ) , and tail-interacting protein of 47 kDa ( TIP47 ) . PAT family members are responsible for the transportation of lipids and the formation of lipid droplets in a variety of tissues and cultured cell lines , including adipocytes [9]–[12] . ADRP selectively increases the uptake of long chain fatty acids and has an essential role in fatty acid transport [13] , [14] . Hormone-sensitive lipase ( HSL ) , as the first enzyme identified in the induction of lipo-catabolic action initiated by hormones , is the predominant lipase effector of catecholamine-stimulated lipolysis in adipocytes [15] . Therefore , ADRP and HSL have opposing functions , i . e . , lipid accumulation vs . its degradation . ADRP and HSL also play important roles in lipid accumulation in M . leprae-infected macrophages [8] , [16] . M . leprae infection increased the expression of ADRP mRNA and protein , facilitating the accumulation and maintenance of a lipid-rich environment suitable for intracellular survival [8] . Conversely , HSL expression was suppressed in macrophages infected with M . leprae [16] . These results suggest that both ADRP and HSL influence the lipid-rich environment that favors M . leprae parasitization and survival in infected host cells . In our previous study , HSL expression was not detectable in slit-skin smear specimens from non-treated LL and BL patients , but it re-appeared shortly after MDT treatment [16] . However , how treatment modulates HSL expression is not clear . In the present study , we determine the effect of MDT components on host lipid metabolism by investigating the influence of rifampicin , dapsone and clofazimine on the expression of ADRP and HSL in THP-1 cells .
Human specimens were used according to the guidelines approved by the Ethical Committee of the National Institute of Infectious Diseases ( Tokyo , Japan ) . All samples were anonymized before use . Clofazimine ( Sigma-Aldrich Co . , St . Louis , MO ) , rifampicin ( Wako Pure Chemical Industries Ltd . , Osaka , Japan ) and dapsone ( Wako Pure Chemical Industries Ltd . ) were dissolved in dimethyl sulfoxide ( DMSO ) and stored at 4°C . The final concentration used in the culture medium was 8 . 0 µg/ml rifampicin , 5 . 0 µg/ml dapsone or 2 . 0 µg/ml clofazimine . Hypertensive nude rats ( SHR/NCrj-rnu ) , infected with the Thai53 strain of M . leprae [17] , [18] were kindly provided by Dr . Y . Yogi of the Leprosy Research Center , National Institute of Infectious Diseases . Japan . The protocol was approved by the Experimental Animal Committee , of the National Institute of Infectious Diseases , Tokyo , Japan ( Permit Number: 206055 ) . Animal studies were carried out in strict accordance with the recommendations from Japan's Animal Protection Law . M . leprae was isolated as previously described [19] , [20] . The human premonocytic cell line THP-1 was obtained from the American Type Culture Collection ( ATCC; Manassas , VA ) . The cells were cultured in six-well plates in RPMI medium supplemented with 10% charcoal-treated fetal bovine serum ( FBS ) , 2% non-essential amino acids , 100 IU/ml penicillin and 100 µg/ml streptomycin at 37°C in 5% CO2 [7] , [8] . Typically , 3×107 bacilli were added to 3×106 THP-1 cells ( multiplicity of infection: MOI = 10 ) . Total RNA from cultured cells was prepared using RNeasy Mini Kits ( Qiagen Inc . , Valencia , CA ) as described previously [7] , [8] . Total RNA preparation from slit-skin smear samples was performed as described [8] , [16] . Briefly , stainless steel blades ( Feather Safety Razor Co . , Osaka , Japan ) used to obtain slit-skin smear specimens were rinsed in 1 ml of sterile 70% ethanol . The tube was then centrifuged at 20 , 000×g for 1 min at 4°C . After removing the supernatant , RNA was purified with the same protocol that was used for cultured cells . The RNA was eluted in 20 µl of elution buffer and treated with 0 . 1 U/µl DNase I ( TaKaRa Bio , Kyoto , Japan ) at 37°C for 60 min to degrade any contaminating genomic DNA . All RNA samples had an OD260/280 of 1 . 8–2 . 0 and an OD260/230 >1 . 8 . RNA sample quality was also confirmed using denaturing agarose gel electrophoresis and the Agilent 2100 Bioanalyzer ( Agilent Technologies , Santa Clara , CA ) ( Fig . S1 ) . Total RNA from each sample was reverse-transcribed to cDNA using a High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems , Foster City , CA ) with random primers [8] , [16] . The following primers were used to amplify specific cDNAs: ADRP: 5′-TGTGGAGAAGACCAAGTCTGTG-3′ ( forward ) and 5′-GCTTCTGAACCAGATCAAATCC-3′ ( reverse ) ; HSL: 5′-CTCCTCATGGCTCAACTCCTTCC-3′ ( forward ) and 5′-AGGGGTTCTTGACTATGGGTG-3′ ( reverse ) ; interferon ( IFN ) -β: 5′-TGCTCTCCTGTTGTGCTTCTCCAC-3′ ( forward ) and 5′-CAATAGTCTCATTCCAGCCAGTGC-3′ ( reverse ) ; IFN-γ: 5′-GCAGAGCCAAATTGTCTCCTTTTAC-3′ ( forward ) and 5′-ATGCTCTTCGACCTCGAAACAGC-3′ ( reverse ) and actin: 5′-AGCCATGTACGTAGCCATCC-3′ ( forward ) and 5′-TGTGGTGGTGAAGCTGTAGC-3′ ( reverse ) . Touchdown PCR was performed using a PCR Thermal Cycler DICE ( TaKaRa Bio , Tokyo , Japan ) [7] , [8] . Briefly , the PCR mixture was first denatured for 5 min at 94°C , followed by 20 cycles of three-temperature PCR consisting of a 30-sec denaturation at 94°C , a 30-sec annealing that started at 65°C and decreased 0 . 5°C every cycle to 55°C , and a 45-sec extension at 72°C . An additional 10 cycles were performed for ADRP and β-actin , and 14 cycles for HSL with a fixed annealing temperature of 55°C . The products were analyzed by 2% agarose gel electrophoresis . Cellular protein was extracted and analyzed as previously described [16] , [21] . Briefly , cells were lysed in a lysis buffer containing 50 mM HEPES , 150 mM NaCl , 5 mM EDTA , 0 . 1% NP40 , 20% glycerol , and protease inhibitor cocktail ( Complete Mini , Roche , Indianapolis , IN ) for 1 h . After centrifugation , the supernatant was transferred and 10 µg of protein was used for analysis . Cellular proteins were mixed with 4× LDS sample buffer and 10× reducing agent ( Invitrogen , Life Technologies , Carlsbad , CA ) and incubated for 10 min at 70°C prior to electrophoresis . Proteins were separated on NuPage 4–12% Bis Tris Gels and transferred using an iBlot Gel Transfer Device ( Invitrogen ) . The membrane was washed with PBST ( phosphate buffered saline ( PBS ) with 0 . 1% Tween 20 ) , blocked in blocking buffer ( PBST containing 5% skim milk ) overnight , and then incubated with either rabbit anti-ADRP antibody ( Santa Cruz Biotechnology Inc . , Santa Cruz , CA; 1∶2 , 000 dilution ) , rabbit anti-HSL antibody ( Cell Signaling Technology , Danvers , MA; 1∶1 , 000 dilution ) or goat anti-β-actin antibody ( Santa Cruz; dilution 1∶2 , 000 ) . After washing with PBST , the membrane was incubated for 1 h with biotinylated donkey anti-rabbit antibody for ADRP and HSL ( GE Healthcare , Fairfield , CT; 1∶2 , 000 dilution ) or biotinylated donkey anti-goat antibody for β-actin ( Millipore , Billerica , MA; dilution 1∶10 , 000 ) followed by streptavidin-HRP ( GE Healthcare; 1∶10 , 000 dilution ) for 1 h . The signal was developed using ECL Plus Reagent ( GE Healthcare ) . THP-1 cells were grown on glass coverslips in 24-well plates for 24 h , before the culture medium was exchanged with RPMI 1640 containing M . leprae and clofazimine . Control and drug-treated THP-1 cells were fixed in 10% formalin for 10 min . They were then washed with Dulbecco's PBS ( DPBS ) and balanced with 60% isopropanol for 1 min before staining with oil-red-O ( Muto Pure Chemicals , Tokyo , Japan ) for 10 min . The cells were counterstained with hematoxylin for 5 min followed by ethanol dehydration and coverslip sealing . Archived formalin-fixed , paraffin-embedded tissue sections were subjected to immunohistochemical staining as described [7] . Briefly , deparaffinized sections were heated in 1 mM NaOH at 120°C for 5 min for antigen retrieval . They were then washed with PBST and blocked in blocking buffer ( DAKO , Carpinteria , CA ) for 10 min , and then incubated with either anti-ADRP antibody ( Santa Cruz Biotechnology Inc . ; 1∶200 dilution ) or anti-HSL antibody ( Cell Signaling Technology; 1∶100 dilution ) , for 1 h at room temperature . After washing the slides with PBST , peroxidase-labeled streptavidin-biotin method was employed using the LSAB2 kit ( DAKO ) and 3 , 3-diaminobenzidine tetrahydrochloride ( DAB ) for the staining of ADRP . Tyramide signal amplification ( TSA ) -HRP method was utilized to amplify HSL staining signals using the TSA Biotin System ( PerkinElmer , Inc . , Waltham , MA ) according to the manufacturer's protocol . Sections were then stained using carbol fuchsin to visualize acid-fast mycobacteria and counterstained with hematoxylin . All experiments were repeated at least three times . Since the replicates produced essentially the same outcomes , representative results from these independent experiments are shown in the figures .
The effect of MDT drugs on lipid metabolism in M . leprae-infected macrophages was examined by infecting human premonocytic THP-1 cells with M . leprae ( MOI = 10 ) in the presence of 8 . 0 µg/ml rifampicin , 5 . 0 µg/ml dapsone or 2 . 0 µg/ml clofazimine for 24 h . Total RNA was isolated and RT-PCR analysis was performed to evaluate possible changes in ADRP and HSL mRNA levels . In our previous studies , M . leprae infection has been shown to increase ADRP and decrease HSL expression , which will in turn increase the lipid accumulation that is thought to contribute to maintaining a phagosome environment which permits M . leprae to parasitize tissue macrophages [8] , [16] . However , when M . leprae-infected THP-1 cells were treated with clofazimine , ADRP expression levels decreased and HSL expression increased ( Fig . 1 ) . Rifampicin and dapsone did not show significant effects on the mRNA expression of ADRP , while they decreased HSL expression by augmenting the effect of M . leprae infection . To further evaluate the effect of clofazimine on ADRP and HSL expression , THP-1 cells were treated with clofazimine in the presence or absence of M . leprae infection for 6 , 12 , 24 and 48 h . Total RNA and cellular protein were extracted and used for RT-PCR analysis and Western blot analysis , respectively . Linearity of the RT-PCR amplifications of ADRP , HSL and β-actin was confirmed by serial dilution of RNA samples and densitometric analysis of the bands ( Fig . S2 ) . RT-PCR showed that clofazimine alone had no effect on ADRP and HSL mRNA levels in control THP-1 cells ( Fig . 2 , left panel ) . Consistent with previous reports , ADRP mRNA expression was increased and HSL mRNA expression was decreased when THP-1 cells were infected with M . leprae ( Fig . 2 , middle panel ) [8] , [16] . However , simultaneous clofazimine treatment and M . leprae infection of THP-1 cells led to decreased ADRP and increased HSL mRNA levels ( Fig . 2 , right panel ) . The decrease of ADRP and increase of HSL mRNA expression were further confirmed by quantitative real-time PCR ( Fig . S3 ) , which also supports the linearity of our RT-PCR data . Thus , it was shown that clofazimine modulated expression of ADRP and HSL only in M . leprae-infected cells . Similar results were observed for ADRP and HSL protein expression levels in each experiment . In the above studies , THP-1 cells were simultaneously treated with clofazimine and infected with M . leprae . Therefore , there was a possibility that clofazimine might have modulated the cellular environments of THP-1 cells before engulfing M . leprae . To eliminate this possibility and to imitate clinical situations , THP-1 cells were first infected with M . leprae for 24 h , to allow cells to engulf enough bacilli , before they were treated with clofazimine . M . leprae infection enhanced ADRP expression and suppressed HSL expression for up to 72 h ( Fig . 3 , left panel ) , which is consistent with the results shown in Fig . 2 , middle panel . However , adding clofazimine 24 h after M . leprae infection produced lower levels of ADRP expression , but increased HSL expression ( Fig . 3 , right panel ) . Interestingly , ADRP expression fell even lower than the original level , and HSL rose higher than original levels , following clofazimine treatment . These results suggest that the lipid catabolic activity once suppressed by M . leprae infection was reactivated by clofazimine treatment , which in turn would promote lipolysis in infected macrophages and decrease cellular lipids . Also , these results are consistent with clinical situations in which HSL mRNA levels were recovered following successful treatment with MDT in LL and BL patients [16] . The decrease in ADRP expression and increase in HSL expression produced by clofazimine treatment were also observed when M . leprae-infected cells were further treated with peptidoglycan ( PGN ) , a ligand for Toll-like receptor ( TLR ) -2 , to activate innate immunity [8] , [16] . We therefore hypothesized that clofazimine treatment might activate the innate immune response of THP-1 cells , which also confers bactericidal activities . To assess activation of innate immunity , production of interferon IFN-β and IFN-γ mRNA was evaluated in control and M . leprae-infected THP-1 cells treated with clofazimine . A transient increase of IFN-β and induction of IFN-γ were observed only in THP-1 cells infected with M . leprae and treated with clofazimine ( Figs . 4A and 4B ) . Transient induction of IFNs as a result of macrophage activation is consistent with previous reports [22]–[24] . Innate immune activation of infected cells will further contribute to the elimination of intracellular bacilli , which is also consistent with the observation that the active form of vitamin D suppresses CORO1A expression in THP-1 cells [21] . To test whether the decrease in ADRP expression and increase in HSL expression after clofazimine treatment would result in less accumulation of cellular lipids after M . leprae infection , THP-1 cells were infected with M . leprae ( MOI = 10 ) in the presence or absence of 2 . 0 µg/ml clofazimine for 48 h . Oil-red-O staining clearly demonstrated the accumulation of cellular lipid droplets following M . leprae infection ( Fig . 5B vs . Fig . 5A ) . In M . leprae-infected cells treated with clofazimine , the amount of lipid droplets in the cell had significantly decreased by 48 h ( Fig . 5C vs . 5B ) . The decrease in cellular lipid droplets is in agreement with the results shown in this study in which clofazimine decreased ADRP and increased HSL expression in M . leprae-infected cells . To confirm the expression pattern of ADRP and HSL in clinical courses of leprosy , ADRP and HSL mRNA levels were evaluated in slit-skin smear specimens by RT-PCR analysis . ADRP mRNA was detected in all LL and most BL cases tested ( Fig . 6A , right panel ) . HSL mRNA was detected in four BL cases; however , ADRP mRNA expression in these cases was absent or weaker than in other BL samples ( Fig . 6A , cases 2 , 4 , 6 and 8 ) . In one case , from which serial samples were obtained , the expression of ADRP mRNA decreased and HSL mRNA levels increased after treatment ( Fig . 6B ) . To further confirm changes in ADRP and HSL expression following treatment , immunohistochemical and acid-fast staining were performed using formalin-fixed paraffin-embedded skin tissue sections . Consistent with a previous report , ADRP localized to phagosome membranes that contains solid-shaped M . leprae ( Fig . 7A ) [8] . HSL staining was not evident before treatment ( Fig . 7C ) . Three months after treatment , staining of the bacilli showed a dotted pattern with no solid-staining , indicating degeneration of M . leprae ( Fig . 7 B and 7D ) . At this point , ADRP staining was faint ( Fig . 7B ) , but strong HSL staining was observed along the phagosomal membrane ( Fig . 7D ) . These staining patterns correlate with changes in mRNA levels of ADRP and HSL in the skin smears ( Fig . 6B ) .
In previous studies , we showed that M . leprae infection increases ADRP expression and decreases HSL expression in host macrophages [8] , [16] . The results of the present study demonstrate that clofazimine , one of the three major drugs used to treat leprosy , counteracts the effect of M . leprae to reduce ADRP and increase HSL expression of both mRNA and protein levels . These results are consistent with our observations in clinical samples obtained from leprosy patients , in which HSL levels were not detectable in skin smear specimens before treatment , but re-appeared shortly after MDT [8] , [16] . The other two MDT drugs , dapsone and rifampicin , revealed no effects on the expression of either ADRP or HSL . Mycobacteria survive by evading the host immune system and accessing host metabolic pathways to obtain nutrients for growth . M . leprae has undergone reductive evolution and pseudogenes now occupy half of its genome [25]–[27] , thus M . leprae is thought to be the mycobacterium most dependent on host metabolic pathways , including host-derived lipids . As we previously reported , PGN can activate TLR2 to increase the expression of HSL [16] and suppress ADRP and perilipin expression [7] , [8] , [21] . These effects mediated by the TLR-initiated signaling pathway will induce lipid degradation , which makes it difficult for M . leprae to survive within host cells . M . leprae infection not only suppresses HSL expression , but also invalidates all effects of PGN on ADRP and perilipin , thus ensuring a phagosome environment that is favorable for mycobacterial survival [16] . In the present study clofazimine increased HSL expression and decreased ADRP expression only in M . leprae-infected cells . The amounts of lipids accumulated in the cells decreased when clofazimine was added to the cell culture medium . The decrease of the lipid-rich environment against the survival of M . leprae may be one of the key actions of clofazimine . Clofazimine was the first clinically developed riminophenazine for the treatment of tuberculosis [28] . Its use has been extended to many Gram-positive bacterial infections as well as mycobacterial diseases [28]–[30] . The drug is now widely used for the treatment of leprosy , but its mechanism remains unclear [31]–[33] . The drug is extremely lipophilic and is also active in membrane destabilization and possible promotion of antigen processing . Stimulated phospholipase A2 activity and subsequent accumulation of arachidonic acid and lysophospholipids were confirmed in clofazimine-induced membrane destabilization [29] , [34] . Increased major histocompatibility complex ( MHC ) class II expression in peripheral blood monocytes [35] , up-regulated lysosomal enzyme activity of cultured macrophages [36] and decreased suppressor T-cell activity in mycobacteria-infected mice [37] reveal the potential role of clofazimine in facilitating immune recognition . Although the underlying molecular mechanisms are not clear , clofazimine suppressed ADRP and induced HSL , IFN-β and IFN-γ expression only in cells infected with M . leprae , the same effects products by PGN [8] , [16] , [21] . Therefore , it is possible that clofazimine revives at least some of the activities of PGN , which is normally shielded by redundant mycolic acid at the M . leprae cell wall . Given the extreme lipophilicity of clofazimine and its activity against many Gram-positive bacteria , clofazimine may interact with the mycolic acid in the M . leprae cell wall that facilitates the exposure of PGN , which in turn activates TLR2-mediated signaling cascades , subsequently decreasing ADRP and increasing HSL [8] , [16] , [21] . Furthermore , since most lepra reactions , a cell-mediated , delayed-type hypersensitivity immune response , occur during or after MDT [38] , [39] , the prospect that clofazimine rescues shielded PGN activities , promoting lysosomal fusion and antigen processing , would be a plausible explanation for the trigger of lepra reactions . The results from present and previous studies may explain the underlying mechanisms , at least in part , of successful parasitization of M . leprae and the effects of MDT treatment observed in patients . In conclusion , we have shown that clofazimine devastates the lipid-rich environment in M . leprae-infected host macrophages by modulating the expression of ADRP and HSL and activates the innate immune response of infected cells , both of which would be important in fighting mycobacterial infection .
|
Leprosy , caused by Mycobacterium leprae ( M . leprae ) , is an ancient infectious disease that remains the leading infectious cause of disability . After infection , M . leprae lives inside host macrophages that contain a large amount of lipids , which is thought to be an essential microenvironment for M . leprae to survive in host cells . M . leprae infection increases lipid accumulation in macrophages and decreases the metabolic breakdown of lipids ( catabolism ) . In addition , the treatment of leprosy with multidrug therapy ( MDT ) reverses the effect of infection on the modulation of lipid metabolism . We therefore aimed to use cultured human macrophage cells to determine which of the three MDT drugs ( clofazimine , dapsone , or rifampicin ) is responsible for this effect . We found that only clofazimine affects lipid accumulation and catabolism in M . leprae-infected cells in vitro . The amounts of lipids accumulated in the cells decreased when clofazimine was added to the cell culture medium . Clofazimine also activated immune responses in M . leprae-infected cells . These results suggest that the effectiveness of clofazimine against leprosy is due to the modulation of lipid metabolism and activation of immune reactions in M . leprae-infected host cells .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"biochemistry",
"infectious",
"diseases",
"bacterial",
"diseases",
"mycobacterium",
"drugs",
"and",
"devices",
"biology",
"metabolism",
"lipid",
"metabolism"
] |
2012
|
Clofazimine Modulates the Expression of Lipid Metabolism Proteins in Mycobacterium leprae-Infected Macrophages
|
Trade-offs between energy consumption and neuronal performance must shape the design and evolution of nervous systems , but we lack empirical data showing how neuronal energy costs vary according to performance . Using intracellular recordings from the intact retinas of four flies , Drosophila melanogaster , D . virilis , Calliphora vicina , and Sarcophaga carnaria , we measured the rates at which homologous R1–6 photoreceptors of these species transmit information from the same stimuli and estimated the energy they consumed . In all species , both information rate and energy consumption increase with light intensity . Energy consumption rises from a baseline , the energy required to maintain the dark resting potential . This substantial fixed cost , ∼20% of a photoreceptor's maximum consumption , causes the unit cost of information ( ATP molecules hydrolysed per bit ) to fall as information rate increases . The highest information rates , achieved at bright daylight levels , differed according to species , from ∼200 bits s−1 in D . melanogaster to ∼1 , 000 bits s−1 in S . carnaria . Comparing species , the fixed cost , the total cost of signalling , and the unit cost ( cost per bit ) all increase with a photoreceptor's highest information rate to make information more expensive in higher performance cells . This law of diminishing returns promotes the evolution of economical structures by severely penalising overcapacity . Similar relationships could influence the function and design of many neurons because they are subject to similar biophysical constraints on information throughput .
The balance between cost and benefit plays an important role in directing the evolution of biological systems [1 , 2] . Costs and benefits are many and various; for example , the elongated tail of the male long-tailed widow bird is very effective at attracting females , but it also makes the male more conspicuous to predators and greatly increases the energetic cost of flight [3 , 4] . Many of the costs that are incurred in the manufacture , maintenance , operation , and carriage of systems can be reduced to a common currency , the expenditure of metabolic energy , while the benefits can be measured in terms of a system's performance . A system's energy cost and its performance interact , within the context of the organism and its habitat , to determine fitness . Relationships between cost and performance have undoubtedly shaped the evolution of nervous systems [5–7] . The enlargement of structures for particularly important and demanding behavioural tasks , such as the auditory system of a bat [8] , and the reduction of redundant structures , such as the thalamo-cortical visual system of the subterranean mole rat Spalax [9] , suggest that larger structures perform better and cost more . Economical wiring patterns and layouts [10–12] and mechanisms that improve the energy efficiency of neurons [13 , 14] , circuits [15] , and codes [16–19] have evolved in nervous systems , and these adaptations suggest that there is pressure on nervous systems to maximise performance and minimise expenditure on materials and metabolic energy [20] . Much of the metabolic energy consumed by a nervous system is used to generate and transmit signals , and most of this goes to the Na+/K+ pump , to restore the ionic concentration gradients that drive rapid electrical signalling and neurotransmitter uptake [21] . This energy usage is directly related to performance— more power is required to transmit signals at higher rates [22–24] . Furthermore , the quantities of energy used by neurons are sufficiently large to limit the coding , processing , and transmission of information . Thus the limited availability of energy not only constrains the size and total number of neurons in the brain [7 , 25] , it limits representational capacity by placing a remarkably low ceiling on mean firing rates [21 , 26] . Although the balance between energy costs and performance could well play a formative role in the evolution of nervous systems , to our knowledge no single study has set out to establish these relationships by measuring both costs and performance across a set of comparable neurons . Fly photoreceptors offer several advantages for such a systematic comparative study of the trade-offs between neuronal energy costs and neuronal performance . The biochemical and electrical signalling mechanisms , the phototransduction cascade [27] , and the photoreceptor membrane [28 , 29] , are exceptionally well described [30–32] . High quality intracellular recordings from identified photoreceptors in intact retina allow one to measure both cost and performance in the same cell . Performance can be measured directly , as the rate at which the photoreceptor transmits information , from recordings of voltage signals [33] . The metabolic cost of this information can be obtained by measuring membrane voltage and conductance , and then applying these measurements to a membrane model to calculate the ionic currents used to generate responses and the rate at which Na+/K+ pumps must consume ATP to maintain the ionic concentration gradients that drive electrical signalling . This empirical method yields the unit cost of information , measured in ATP molecules hydrolysed per bit of information coded [14 , 34] . We present a systematic comparative study of fly photoreceptors , which sets out to discover how neuronal energy costs change with neuronal performance . We compare homologous photoreceptors taken from four species of Diptera , the blowfly Calliphora , the fleshfly Sarcophaga , and two Drosophilids . The blowfly and the fleshfly have larger eyes with better spatial and temporal resolving power , presumably because these large flies fly faster and further and are more manoeuvrable than the Drosophilids . Photoreceptor performance is measured directly , as information throughput in bits s−1 , and energy costs are estimated as the rate at which the Na+/K+ pump must hydrolyse ATP molecules in order to sustain signalling . We confirm that blowfly R1–6 photoreceptors achieve higher bit rates than D . melanogaster [31 , 33] at greater cost [14 , 34] , and we extend this comparison to the full operating range of background light levels . Furthermore , by applying identical methods to four species , we describe how costs scale against performance . We find that it is costly to improve performance , because membrane conductance increases supralinearly with maximum bit rate , and this makes information more expensive in higher capacity cells . Our measurements confirm theoretical findings [16 , 18 , 35] that the fixed cost of maintaining a cell at rest , ready to signal , is a major determinant of metabolic efficiency and also establish a basic microeconomic relationship; namely that the fixed cost of maintaining a cell ready to signal increases with its maximum information rate . In this sense fly photoreceptors resemble cars; a high performance Porsche Carrera GT consumes three times as much fuel km−1 as a lower performance Honda Civic [36] , even when driven at the same low speeds ( urban cycle ) . Because this new example of a neuronal law of diminishing returns appears to be enforced by the basic biophysics of electrical signalling , we suggest that it operates in many neurons and could , therefore , play a significant role in determining the function , design , and evolution of nervous systems .
We compared information rates with energy costs in R1–6 photoreceptors from four species C . vicina , S . carnaria , D . virilis , and D . melanogaster . C . vicina and D . melanogaster R1–6 were chosen because they are known to transmit at very different rates . In daylight the large Calliphora cells transmit approximately 1 , 000 bits s−1 [30 , 33 , 34] , whereas the smaller D . melanogaster cells transmit at just over 200 bits s−1 [31 , 32 , 37] . We developed a new preparation , the intact retina of D . virilis , to provide R1–6 cells of intermediate size and information rate . We also recorded from the large R1–6 photoreceptors of another vigorous fly with a large eye , S . carnaria , in order to confirm that the high costs measured in Calliphora R1–6 are associated with high information rates . We measured information rates from intracellular recordings of voltage responses to optical signals ( Figure 1 ) [33] . The photoreceptor was first adapted to a background light whose effective intensity had been calibrated as an effective photon rate by counting that same photoreceptor's discrete responses to single photons ( see Materials and Methods; Figure 1A ) . This calibration takes account of differences in acceptance angle and sensitivity and enables us to compare the performance of photoreceptors receiving the same number of photons . Once stably adapted , the photoreceptor was presented with multiple repeats of the same brief sequence of pseudorandom modulation of the light around the background intensity ( Figure 1B and 1C ) . The mean contrast of this modulation ( standard deviation/mean ) was 0 . 32 , a value close to that of natural scenes ( see Materials and Methods ) . Photoreceptors encode the fluctuations in stimulus contrast as a graded ( analogue ) modulation of membrane potential that is contaminated by noise . We extracted the photoreceptor's voltage signal ( Figure 1B and 1C ) by averaging the responses ( averaging eliminates noise ) and then extracted the noise by subtracting our estimate of the signal from the response to each stimulus repeat . Our estimate of the signal was transformed into the signal power spectrum S ( f ) . Each of the extracted noise traces was transformed , and the resulting ensemble of spectra was averaged to generate the noise power spectrum N ( f ) . Both signal and noise were distributed normally , allowing the rate at which the photoreceptor transmits information I in bits s−1 to be determined by applying Shannon's formula [38] to the power spectra of the signal S ( f ) and noise N ( f ) : The logarithmic term in this equation is the distribution of information across frequencies , as plotted in Figure 1D . At the lowest photon rates ( 102−103 effective photons s−1 ) the information rates of all four photoreceptors were almost identical , suggesting that under these conditions the information rates in all four species were limited by photon noise , rather than response bandwidth ( Figure 2 ) . At higher photon rates ( >103 effective photons s−1 ) the information rates of the photoreceptors diverged ( Figure 2 ) . At the highest effective photon rates ∼107 s−1 , which are within 0 . 7 log units of the highest daylight intensities [39] , Calliphora and Sarcophaga photoreceptors attained throughputs close to 1 , 000 bits s−1 ( 955 ± 70 , n = 3 for Calliphora and 1 , 130 ± 67 , n = 11 for Sarcophaga ) , compared with ∼510 bits s−1 in D . virilis photoreceptors ( 512 ± 26 , n = 21 ) and ∼200 bits s−1 in D . melanogaster photoreceptors ( 197 ± 31 , n = 26 ) . Note that , as explained in the Materials and Methods , our set of values from 26 D . melanogaster photoreceptors includes data from 21 cells that were published in an earlier study [32] . The rates measured in Calliphora and D . melanogaster R1–6 photoreceptors are similar to those measured previously with comparable methods [31 , 34 , 37] . The information rate of D . melanogaster R1–6 photoreceptors saturated at our highest intensities , but the information rates in the other species did not ( Figure 2 ) . Nonetheless , because our highest photon rate is close to that experienced in full daylight [39] , the photoreceptors are operating close to their natural intensity limit . The R1–6 photoreceptors of the larger more active flies , Calliphora and Sarcophaga , code information at higher rates because they maintain a higher signal-to-noise ratio ( SNR ) over a broader bandwidth of response . The contributions of SNR and bandwidth to performance are illustrated by comparing a plot of information versus frequency for the highest information rate photoreceptor , Sarcophaga R1–6 , with a plot for the lowest information rate photoreceptor , D . melanogaster R1–6 ( Figure 1D ) . At any given frequency the Sarcophaga R1–6 carries more information than the D . melanogaster R1–6 because its SNR , S ( f ) /N ( f ) in Equation 1 , is larger . The Sarcophaga R1–6 codes almost half of its information at frequencies in the range 100–300 Hz but , because of its poorer bandwidth , D . melanogaster R1–6 codes very little information at frequencies above 100 Hz ( Figure 1D ) . We used an established electrical model of the photoreceptor membrane to estimate the rate at which photoreceptors consume metabolic energy ( see Materials and Methods ) . The model [14 , 34] incorporates the two major conductances , light-gated and potassium , as well as the electrogenic Na+/K+ pump , and calculates the flux of ions through these components from measurements of total conductance and membrane potential ( Figure 3 ) . The flux of ions through the Na+/K+ pump gives the rate at which ATP is hydrolysed in order to maintain ionic concentration gradients , and ATP hydrolysis rate in molecules s−1 is our measure of metabolic energy cost . To obtain the data used to estimate metabolic costs , we measured a photoreceptor's membrane potential and input resistance ( see Materials and Methods ) , first in the dark and then at each of the background light intensities at which we measured the information rate ( Figure 3A and 3B ) . All photoreceptors were depolarised by light , and the steady-state depolarisation produced by a sustained background light increased with background intensity ( Figure 3A ) . At lower backgrounds , below 103 effective photons s−1 , Calliphora R1–6 photoreceptors were most depolarised , and Sarcophaga R1–6 photoreceptors were least depolarised , whereas at higher backgrounds , above 105 effective photons s−1 , D . melanogaster R1–6 photoreceptors were most depolarised , and D . virilis R1–6 photoreceptors were least depolarised ( Figure 3A ) . Photoreceptor input resistance dropped with increasing light intensity ( Figure 3B ) because of increased activation of light-gated channels and voltage-gated potassium channels [28 , 32] . In the dark , and at any particular photon rate , D . melanogaster R1–6 photoreceptors had the highest input resistance , D . virilis R1–6 were intermediate , and the Calliphora and Sarcophaga R1–6 photoreceptors had the lowest input resistances ( Figure 3B ) . Putting these measurements of membrane potential and resistance into our electrical model ( Figure 3C ) , we obtained the rate of ATP consumption ( Figure 3D ) . In the dark and at any particular photon rate , Calliphora and Sarcophaga R1–6 photoreceptors had the highest rate of ATP consumption , D . virilis R1–6 were intermediate , and D . melanogaster R1–6 photoreceptors had the lowest rate of ATP consumption ( Figure 3D ) . Thus small photoreceptors that transmitted at lower bit rates ( Figure 2 ) had lower energy costs ( Figure 3D ) . We can separate photoreceptor energy cost into two components , dark and signalling . The dark cost is the rate at which ATP is hydrolysed to maintain the cell's resting potential in the dark . The signalling cost is the increase in ATP hydrolysis rate induced by light , i . e . , The signalling cost is , in microeconomic terms [40] , a variable cost that increases with the level of output , whereas the dark cost is a fixed cost that , like the rent on a factory , is paid at a fixed rate , irrespective of output . In nervous systems the fixed cost of maintaining an inactive neuron's resting potential is an important determinant of the metabolic efficiency of distributed neural codes [16] , of spike trains in single cells [18] , and of stochastic signalling mechanisms such as ion channels and synapses [35] . The R1–6 photoreceptors of all four species have significant fixed costs—all photoreceptors consume substantial quantities of ATP in the dark ( Figure 3D ) . This fixed cost differs greatly between species , according to photoreceptor input resistance and membrane potential . Comparing the photoreceptors of different species , although there are appreciable differences ( up to 10 mV ) in the dark resting potential ( Figure 3A ) , the dark input resistances vary by more than an order of magnitude ( Figure 3B ) and are , therefore , primarily responsible for the large differences in ATP consumption rates in the dark ( Figure 3D ) . The dark consumption is approximately 2 × 109 ATP molecules s−1 in Calliphora and Sarcophaga R1–6 photoreceptors , whereas in D . melanogaster R1–6 the dark consumption is approximately twenty times less , 1 × 108 ATP molecules s−1 . Illumination increases the rate of ATP consumption in all of the R1–6 photoreceptors and , as in the dark , those with the lowest input resistances ( Calliphora and Sarcophaga ) consumed the most ATP ( Figure 3D ) . The similarity between the log–log plots of the total cost versus the effective photon rate suggests that the signalling costs of the photoreceptors from the different species have similar dependencies on light level and scale with the dark cost ( Figure 3D ) . This suggestion led us to compare the dark costs and the signalling costs at different light levels , in the R1–6 photoreceptors of the four species . In each of the four species , the photoreceptor signalling cost rises with intensity and approaches an asymptote at bright daylight levels ( Figure 4A ) . Even though there was an approximately 25-fold difference in both the total and the signalling ATP consumption at the brightest light intensities across the four species , the ratio between the signalling cost at the brightest light levels ( the maximum signalling cost ) and the dark cost was similar in each photoreceptor and was , on average , 4 . 7 ( Figure 4B ) . This scaling suggests that the energy consumption in the dark is directly related to the highest rates of consumption in bright light . When the signalling cost of a photoreceptor type is normalized with respect to dark consumption and plotted against the log of photon rate ( Figure 4C ) , the curves for the different photoreceptor types are similar but not identical , as expected of a set of homologous photoreceptors that use similar mechanisms to generate and regulate responses , but fine tune these mechanisms to their particular requirements . By dividing the total rate of energy consumption ( Figure 3D ) by the corresponding rate of information transmission ( Figure 2 ) at each light level , we derived the metabolic cost of information , as ATP molecules bit−1 . This measure allows us to assess how economically each type of photoreceptor transmits information ( Figure 5A ) . We discovered that all four photoreceptors showed the same behaviour: increasing the light intensity not only increases information rate ( Figure 2 ) , it also decreases the total cost per bit ( Figure 5A ) . The proportional decrease is smallest in D . melanogaster R1–6 , approximately 3:1 , and largest in Sarcophaga R1–6 and Calliphora R1–6 , approximately 10:1 . A substantial part of this decrease in the total cost per bit can be attributed to the dark cost . At low light levels the dark cost is a substantial fraction of the total , and dividing this fixed cost by the low bit rate produces a high cost per bit , which then decreases as bit rate increases . At the highest light intensities the cost per bit starts to level out ( Figure 5A ) , suggesting that under daylight conditions R1–6 photoreceptors are operating close to their minimum cost per bit . To see if other factors contribute to the fall in bit cost with increasing light level we calculated the signalling cost per bit , by dividing the rate at which ATP molecules are consumed for signalling by the bit rate . In the two high bit rate photoreceptors , Sarcophaga R1–6 and Calliphora R1–6 , the signalling cost per bit is highest at low light intensities and then declines over the intensity range 102–107 effective photons s−1 to approximately 30% of its original value ( Figure 5B ) . This increase in efficiency with photon rate is not observed in the two lower bit rate photoreceptors . In D . melanogaster R1–6 the signalling cost per bit first rises slightly with increasing intensity , peaks between 103 and 104 photons s−1 and then falls back to the previous level , while in D . virilis the signalling cost per bit doubles over the range 103 to 106 photons s−1 and then dips slightly ( Figure 5B ) . Note , however , that over most of the intensity range the signalling cost per bit is lower in the low bit rate cells ( Figure 5B ) . The signalling cost per bit could be falling with increasing light level in the Calliphora and Sarcophaga R1–6 , because these photoreceptors expand their bandwidth to higher frequencies ( e . g . , Figure 1D ) to achieve higher bit rates . Contributions to a fall in cost per bit could also be made by the improvement in photoreceptor SNR and by light adaptation of the phototransduction cascade , which , by reducing the light-gated conductance activated per photon [41] , reduces the energy cost per photon . By plotting bit cost versus bit rate ( Figure 6 ) we are able to compare the efficiency of cells operating at the same information rate . The total cost per bit varies consistently between the different species . At a given information rate D . melanogaster R1–6 photoreceptors encode most economically , it is approximately three times more expensive to operate D . virilis R1–6 photoreceptors at the same information rate , and approximately ten times more expensive for Calliphora and Sarcophaga R1–6 photoreceptors . These proportional differences in cost are , to a first approximation , maintained over the range of bit rates , and this suggests that the higher total cost per bit in the two larger flies , Calliphora and Sarcophaga ( Figure 6A ) , is primarily associated with their higher dark cost . Because signalling cost tends to increase with dark cost ( Figure 4 ) , the signalling costs per bit are also substantially higher in Calliphora and Sarcophaga R1–6 than in Drosophilid photoreceptors operating at the same bit rate ( Figure 6B ) . The information rates measured with our brightest stimuli are indicative of a photoreceptor's maximum performance . Comparing metabolic costs with these highest rates , we see that both the total cost and the dark cost increase supralinearly with performance . Plots of the logarithms of costs against the logarithms of highest rates ( Figure 7 ) suggest that the total cost increases as ( performance ) 1 . 7 , and the dark cost increases close to ( performance ) 1 . 5 , but with only four species these exponents are preliminary estimates . Nonetheless , there is no doubt that both the unit cost of information ( the total cost per bit ) and the dark cost are directly related to a photoreceptor's ability to transmit information ( Figures 5 and 6 ) ; the higher a photoreceptor's maximum bit rate , the higher the dark cost , the higher the signalling cost , and the higher the total cost per bit . In conclusion , our recordings from intact fly photoreceptors demonstrate that the cost per bit varies with bit rate in two ways . In any single photoreceptor ( e . g . , a Calliphora R1–6 ) , bit rate increases with light level while the total cost per bit falls ( Figure 5 ) . This fall is due to two factors , offsetting the dark cost ( the substantial fixed cost of maintaining the photoreceptor's resting potential in darkness ) and in the high bit rate photoreceptors , a substantial reduction in signalling cost per bit ( Figure 5 ) . However , when we compare homologous R1–6 photoreceptors in different species , we see that the energy cost per bit increases with a photoreceptor's performance ( Figure 6 ) , where performance is assessed from the highest information rate measured with our brightest stimulus . Again , the relatively high fixed cost of the dark resting potential ( Figures 3 , 4 , and 7 ) is implicated in this relationship between maximum bit rate and bit cost .
It is important that our measures of information rate and metabolic cost are biologically relevant and reliable . We argue that information rate , in bits s−1 , is both a convenient and an appropriate measure of photoreceptor performance , despite the fact that Shannon's treatment of information rate ( Equation 1 ) treats all parts of the signal equivalently , irrespective of the features they represent . Most of the features extracted by fly visual systems are unknown to us , nonetheless there are three reasons why we are confident that the “feature-free” measure , information rate in bits per second , is appropriate . First , unlike the R1–6 photoreceptors of the male housefly lovespot , which are specialised to detect rapidly moving high contrast targets [44] , the R1–6 photoreceptors in our species do not appear to be adapted to detect particular features . Second , R1–6 photoreceptors support many aspects of vision because they feed a variety of parallel circuits in the optic lobes [45] . Third , even when visual systems are devoted to processing a small number of biologically relevant objects , photoreceptors still have to code a wide range of signals . The range is wide because a photoreceptor signals the presence of an object in its field of view as a change in photon rate . This change varies greatly in relative amplitude ( contrast ) and time course , depending on the object's position , illumination , orientation , distance , movement , and the background against which the object is viewed . Thus , because viewing conditions cause a single object to generate a range of photoreceptor signals , the general measure of performance , bit rate , is appropriate . The argument for this general measure is strengthened still further by the fact that synaptic transfer from photoreceptors to interneurons is optimised to maximize bit rate [46–48] . Finally , bit rate takes account of the basic biophysics of coding by combining two more fundamental determinants of signal quality: the accuracy of response and the ability to follow rapidly changing signals . Accuracy contributes to information rate through the SNR , and response speed contributes by determining the bandwidth over which signals can be transmitted ( Equation 1 ) . Given that information rate , in bits per second , is an appropriate measure of photoreceptor performance , are the highest bit rates reached in bright light ( Figure 2 ) adequately representing the different abilities of the four photoreceptors to code information ? Our experiments were designed to apply the two determinants of information rate , bandwidth and SNR , equally to all photoreceptors . White noise tests the full bandwidth by injecting equal power at frequencies that extend well beyond each cell's cutoff . By using photon counts to compare photoreceptors from different species , we ensured that our comparisons are not biased by optical differences ( e . g . , in facet lens diameter , focal length , and rhabdomere width ) that influence the number of photons individual photoreceptors receive from the same stimulus [49 , 50] and hence the photon noise limit to SNR . Although our stimuli enable us to compare photoreceptors on equal terms , there are two reasons why our measured information rates fall short of full capacity . With the exception of D . melanogaster , we were unable to saturate photoreceptors' information rates ( Figure 2 ) , even though our highest intensities are within a factor of five of the photon rates experienced in full daylight [39] . In addition , the information capacity is , by definition , determined using a stimulus that is tailored to distribute power optimally across the photoreceptor bandwidth . Our bit rates in Calliphora ( Figure 2 ) are 50% below the capacities measured at the same photon rates [33] but , although our white-noise stimuli underestimate information capacity , they overestimate the rates generated under natural conditions because natural stimuli have less power at high frequencies . However , the overestimate appears to be small ( 10%–20% ) [51] compared with the 5-fold differences in the highest rates measured in the photoreceptors of the four species ( Figure 2 ) . We conclude that , although the highest information rates we measured underestimate full capacities , our data reflect the highest rates expected under natural conditions . We can , therefore , conclude that our measured bit rates adequately describe differences in photoreceptor performance between species . Turning to our comparison of metabolic cost , the measurements of input resistance and membrane potential used to calculate energy consumption were consistent from cell to cell and agreed with previous studies of Calliphora and D . melanogaster [28 , 30 , 32 , 52] . As expected , the larger photoreceptors with lower input resistance consume more energy . Our conductance-based method underestimates total energy consumption , because it neglects both the intermediate processes in the phototransduction cascade that consume energy [27] and essential maintenance processes , such as macromolecular synthesis . However , the intermediate processes of phototransduction are likely to add less than 10% to the total energy cost because ion flux is the final stage in signal amplification [34] and , in active neural tissue , macromolecular synthesis contributes less than 10% to the total energy consumption [21] . Most importantly , our estimate of energy consumption for a fully light-adapted Calliphora photoreceptor , 7 × 109 ATP molecules s−1 , agrees remarkably well with the most recent value obtained from measurements of retinal oxygen consumption , 6 . 5 × 109 ATP molecules s−1 [53] . A number of studies suggest that energy cost and performance are related to photoreceptor structure and biophysics via two fundamental measures of signal quality , SNR and bandwidth . These two measures determine the measure of performance adopted in this study , information rate ( Equation 1 ) , and the photoreceptors that achieve higher rates do so because they have a better SNR in bright light and a wider bandwidth ( Figure 1D ) . Photoreceptor SNR rises with the rate at which photons are being transduced , and in insect photoreceptors the SNR often tends to plateau at the highest light levels , as photomechanical mechanisms attenuate the incoming photon flux to prevent saturation . The proposal that the maximum attainable SNR has a structural basis , the number of photoreceptive microvilli in a photoreceptor [54] , is strongly supported by more recent evidence . A microvillus contains all of the signalling molecules of the phototransduction cascade , and a single photon hit appears to produce an all or nothing response , a quantum bump , from one entire microvillus [27 , 41] . The corollary that the maximum rate of photon conversion , and hence the maximum SNR , is limited by the number of microvilli , is supported by measurements of SNR under saturated conditions [39] and by the observation that photoreceptors with more microvilli achieve higher SNRs [31 , 32 , 55] . The second determinant of information rate , bandwidth , is regulated by two sets of factors , the molecular dynamics of phototransduction and the electrical properties of the photoreceptor membrane [56–58] . Photoreceptors regulate their bandwidth by controlling the dynamics of the phototransduction cascade and by tuning the frequency response of the membrane with voltage-gated potassium channels [28 , 31 , 32 , 59] . In a given photoreceptor , the bandwidth is adjusted according to light level to improve information throughput , and , comparing different photoreceptors , the maximum bandwidth varies systematically according to retinal position , colour type , and visual ecology [30 , 55 , 60] . Thus photoreceptor bandwidth is carefully regulated to adapt response dynamics to operating conditions . These two factors , the number of microvilli and the membrane bandwidth , link information rate to energy consumption . Increasing the number of microvilli to improve the SNR will increase the photoreceptor's membrane area and hence its total conductance and capacitance , leading to larger ionic currents . Increasing the membrane's potassium conductance to widen its bandwidth ( by reducing its time constant ) also increases the flow of ions across the membrane . Indeed , the high metabolic cost of increasing membrane bandwidth has been invoked to explain why slowly flying insects , exemplified by Tipulid flies , have slow photoreceptors with a low potassium conductance , long time constant , and narrow bandwidth [59 , 60] . The low potassium conductance of slow cells is achieved by inactivation [58] , and the contribution of potassium channel inactivation to energy efficiency has been demonstrated directly by genetically manipulating and modelling photoreceptors in D . melanogaster . When the rapidly inactivating Shaker K+ -channel of R1–6 photoreceptors is deleted by mutation , there is an increase in tonic conductance , and the cost of information , in ATP molecules bit−1 , increases [14] . These findings strongly suggest that the biophysics of SNR and bandwidth link information rate to energy consumption to produce the trade-off between cost and performance observed here . However , this suggestion must be confirmed by relating measurements of SNR , microvillus number , membrane conductance , and membrane bandwidth to measurements of cost and capacity . Such a detailed analysis of photoreceptor structure , biophysics , and performance will reveal whether large increases in bandwidth explain the fall in signalling cost per bit seen in high bit rate photoreceptors ( Figures 5B and 6B ) . This detailed comparison will also decide whether fly photoreceptors divide their energy investment between SNR and bandwidth optimally , to maximize energy efficiency . To the best of our knowledge , this is the first study to measure the fixed cost of maintaining a neuron ready to signal and then relate this fixed cost to performance and metabolic efficiency . The fixed cost of maintaining a photoreceptor's dark-resting potential is high ( approximately 20% of the cost in full-daylight ) and , following earlier calculations [39] , we find that in Calliphora this dark current amounts to approximately 2% of a blowfly's total resting metabolic rate . Although photoreceptor fixed costs vary between species by more than an order of magnitude , they are a remarkably constant proportion , between a fifth and a quarter , of the energy consumed in full daylight ( Figure 4 ) . This proportion suggests that the metabolic scope of insect photoreceptors ( the ratio between the maximum sustainable metabolic rate and the resting metabolic rate ) lies between four and five . Fixed costs increase with capacity ( Figure 7 ) , as also observed in comparative studies of energy throughput and metabolic rate in mice [61] . Species of mice that are adapted to live in areas where food is more plentiful convert food to energy at higher rates than species that live in areas where food is scarce . The high-energy users also have higher basal metabolic rates , presumably to support the extra fixed cost of the larger organs , such as gut and heart , needed to handle higher rates of energy throughput [61] . The reasons why R1–6 photoreceptors have a high fixed cost are unclear , but the proximate cause is a dark resting potential that is approximately 20 mV less negative than the potassium reversal potential [28] . The inward currents that produce this depolarisation have not been identified , but the limited evidence suggests two possibilities , both of which are related to maintaining a high sensitivity . The first source is spontaneous activation of the phototransduction machinery [62 , 63] . Because this spontaneous activity increases with the number of microvilli , its cost will increase with capacity . The second source of inward current could be voltage-sensitive conductances and feedback synapses associated with signal amplification and band pass filtering at the photoreceptor's high sensitivity output synapses [64 , 65] , which are tonically active in the dark [66] . We note in passing that strictly diurnal insects could economise on energy consumption by down-regulating the phototransduction cascade and reducing synaptic activity at night . The fact that a photoreceptor's fixed costs could be related to both its input and its output emphasises that energy efficiency is a systems' property that depends upon relationships within and between components [15 , 35 , 67] . The energy-information trade-offs that we have described in photoreceptors have implications for the design and evolution of insect retinas . The cost of increasing the maximum rate at which a photoreceptor can handle information is substantial and involves large increases in both the cost per bit and the fixed cost of maintaining the photoreceptor in the dark . This leads to a law of diminishing returns whereby a small increase in information capacity requires a larger proportional increase in energy cost . This law increases evolutionary pressure to reduce photoreceptor performance to the minimum required for satisfactory visually-guided behaviour by penalizing excess capacity . The result , allocation of resources according to need , could help to explain why , in males of Calliphora vicina , the R1–6 photoreceptors that look ahead at approaching objects through superior optics have higher information rates than those looking sideways and backwards through inferior optics [30] . The fixed costs of phototransduction could be particularly important for nocturnal insects . Their photoreceptors often have a large area of photosensitive membrane to improve photon capture [49 , 68 , 69] , and this could create problems due to high fixed costs . Furthermore , nocturnal photoreceptors operate at extremely low light levels where fixed costs make each bit of information extremely costly ( Figures 4 and 5 ) . Because the membrane area of the photoreceptive microvilli cannot be sacrificed without losing photons , the only way to reduce fixed costs is to reduce membrane conductance . In extreme circumstances this could result in the photoreceptor membrane having such a long time constant that this , rather than the number of microvilli , limits information capacity at higher light levels . The photoreceptors of nocturnal Tipulids have high resistances and long time constants [58 , 60] and may well , therefore , be implementing this strategy . The relationships between energy and information observed here in fly photoreceptors will apply to signalling systems that share similar biophysical relationships between SNR , bandwidth , and energy cost . Although neurons use synapses as discrete signalling units , rather than microvilli , they too are subject to the stochastic activation of conductance , and are constrained by membrane time constant [70] . Consequently , improvements in neuronal reliability , speed of response , and information rate will probably involve increased energy consumption; namely the additional signalling cost of operating extra channels and synapses and the additional fixed cost of this extra signalling machinery . Recent experiments on spiking neurons support the suggestion that additional signalling and fixed costs make information more expensive in neurons that transmit at higher rates [71] . Comparing the different classes of ganglion cell in guinea pig retina , brisk cells transmit information at higher rates than sluggish cells , because brisk cells fire spikes more frequently with greater temporal precision . Information will be more expensive in a brisk cell because , as expected of a cell that fires at a higher rate , a spike in a brisk cell carries less information than a spike in a sluggish cell [71] . In addition , because brisk cells are larger than sluggish cells , a brisk cell spike will use more energy . By analogy with fly photoreceptors , we further suggest that fixed costs will be higher in brisk cells , because their superior temporal precision requires more channels and synapses leading to a higher baseline conductance . This extra conductance will also increase the signalling cost of generating spikes . For these reasons information will cost more in the higher rate brisk cells than in the lower rate sluggish cells . This cost differential could help to explain why the retinal output is divided among different classes of ganglion cell with over 60% of the information being transmitted by low cost sluggish cells [71] . Thus , this classic example of parallel coding could be improving energy efficiency by directing the signals that require less temporal precision into lower cost channels [72 , 73] . This design principle could well extend beyond the retina to higher visual centres and to the coding of other sensory modalities , such as hearing . The relationships between fixed costs , signalling costs , and bit rate could have a significant impact on coding and neural circuit design . In fly R1–6 photoreceptors , the fixed and total costs increase as power functions of maximum bit rate , with exponents of approximately 1 . 5 and 1 . 7 , respectively ( Figure 7 ) . Thus the relationship between cost and performance follows the law of diminishing returns . Similar examples of this law have been observed in theoretical studies of spiking neurons , synaptic arrays , and neural circuits [15 , 16 , 34] . In general [18] , this law makes it advantageous to implement energy efficient neural codes [17] that distribute information among spikes or neurons so as to avoid high rates . The relationship between costs and capacity measured in fly photoreceptors demonstrate that this law applies not only to signalling costs , but also to fixed costs . Again , theoretical studies have shown that fixed costs are important determinants of energy efficiency , which help set the optimum numbers of synapses and channels [35] and the optimum sparseness of energy-efficient neural codes [16] . Thus the empirical data presented in this study ( Figure 7 ) demonstrate a relationship between representational capacity ( i . e . , highest bit rate ) and fixed cost , which will influence the energy efficiency of circuits and codes . Information rate is not the only measure of neuronal performance by which to judge efficiency . The measures that are most appropriate for a neuron will be defined by the role the neuron plays , processing signals in circuits , and determining behaviour . Relevant measures of performance could include the sharpness of frequency tuning in auditory systems [74] , latency in reflex arcs [75] , and storage capacity in cortical networks [76] . Just as the basic biophysical constraints of bandwidth and noise link photoreceptor information rate to energy consumption , so might improvements in these other performance measures involve additional costs . As examples , frequency tuning could be linked to ion flux by the conductances used to regulate the membrane time constant and to actively suppress or amplify particular frequency bands , while rapid responses and high temporal precision are associated with shorter time constants , larger diameter cells , and larger synapses [77] . On this basis , we suggest that comparative studies of neuronal cost and performance , similar to the experiments presented here , will confirm that trade-offs between energy cost and performance are widespread . The balance between energy cost and performance appears to have played a significant role in determining the evolution of nervous systems [5 , 6] . Numerous examples exist of the relative reduction or expansion of the whole brain or particular brain regions during evolution [7 , 78] . For example , in the extinct bovid genus Myotragus , brain size was reduced by 50% relative to similar bovids of comparable body mass following isolation on a Mediterranean island . It is argued that this reduction was a response to two factors; reduced predation pressure and increased competition for a limited food supply [79] . In birds the degree of specialization for food hoarding correlates with the volume of the hippocampus , expressed relative to both body mass and telencephelon volume [80] . Thus both energetic costs and behavioural requirements are likely to be important selective pressures influencing relative brain size [7] . Improvements in behavioural performance can come about in at least three ways; by acquiring more information from the environment , by improving the nervous system's ability to process and represent information , and by finer or more appropriately coordinated control of motor outputs and muscles . The energy-information trade-offs discovered in fly photoreceptors R1–6 demonstrate that even small improvements in the ability of single cells to acquire and transmit information , and hence to process information more accurately and rapidly , come at a high energetic cost . Costs rise more rapidly than performance and this intensifies selection on neural structures and promotes evolutionary adaptation by increasing the sensitivity of trade-offs between costs and benefits .
We used four species of fly for this study; C . vicina , S . carnaria , D . virilis , and D . melanogaster . Populations of three of these species C . vicina , D . virilis , and D . melanogaster were maintained in the Department of Zoology , University of Cambridge , United Kingdom . Individuals of S . carnaria were obtained from wild populations near Cambridge between May and September , 2004 . The two larger fly species , C . vicina and S . carnaria , were mounted with their dorsal surface uppermost on a wax platform . Additional wax was used to fix the head and thorax but not the abdomen , which was left free to allow breathing . Both Drosophila species were mounted in a custom-built holder , and their head and thorax fixed using wax . In all species a small window ( no more than a few facets in diameter ) was cut manually into the top of the right compound eye and sealed immediately with silicon grease to prevent dehydration . The grease is soft enough to allow intracellular microelectrodes to be inserted through the seal , without damage . A second window was cut into the left compound eye to allow access for the indifferent electrode , a 50-μm-diameter silver wire . In vivo intracellular microelectrode recordings were obtained from R1–6 photoreceptors of C . vicina , S . carnaria , D . virilis , and D . melanogaster . All recordings were made using borosilicate glass electrodes filled with 3 M KCl . The electrode resistance varied considerably depending on the species from which the recording was being made; electrodes with resistances of 100–130 Ω were used for C . vicina and S . carnaria R1–6 photoreceptors , whereas 200 Ω or greater resistance electrodes were used for R1–6 photoreceptor recordings from D . virilis and D . melanogaster . The pipettes were pulled from 10-cm borosilicate glass capillaries ( 1 . 0 mm outer diameter , 0 . 58 mm inner diameter; GC100F-10 , Harvard Apparatus , http://www . harvardapparatus . co . uk ) using a Sutter P97 puller ( Sutter Instruments , http://www . sutter . com ) and inserted into the eye , through the silicon grease seal using a Zeiss Jena grease-plate micromanipulator . All recordings were made using an Axoclamp 2A amplifier ( Molecular Devices , http://www . moleculardevices . com ) . Throughout recordings the temperature of the flies was maintained between 22 °C and 24 °C . Photoreceptors were considered for analysis only if their membrane potentials were hyperpolarised by more than −55 mV in the case of photoreceptors from the drosophilid species and −60 mV in the case of photoreceptors from C . vicina and S . carnaria . Additional criteria such as the amplitude of the saturating impulse response in dark-adapted conditions and the photoreceptor input resistance were also used to determine recording quality . The photoreceptor responses to light were recorded in bridge mode . To determine the input resistance , current was injected , and the voltage response measured in switched current clamp mode . Stimulus generation and data acquisition were carried out using a digital computer and a purpose-built interface . Both stimuli and responses were usually digitised at 2 kHz and , to prevent aliasing , responses were low pass filtered by a four-pole Butterworth with a cutoff at half the Nyquist frequency , i . e . , 500 Hz . Photoreceptors were stimulated by a point source , the tip of a light guide that was positioned on the optical axis and subtended six degrees at the cornea . In the setup used for Calliphora R1–6 , white light was provided by a 450-W high-pressure xenon arc lamp ( PRA model 301s ) , which was stabilised with optical feedback to suppress unwanted fluctuations in the light intensity delivered to the waveguide to below 0 . 5% ( root mean square ) . To provide white-noise stimulation , the arc was modulated by feeding a voltage command waveform from the computer to the optical feedback unit . In the setup used for the other photoreceptors , the light source was a high intensity LED ( 505 nm , LEDtronics , http://www . ledtronics . com ) whose output was controlled directly by a voltage to current converter , driven directly by the computer . All voltage commands were corrected for the nonlinear characteristics of the LED . Light was attenuated by calibrated neutral density filters to provide a series of background light levels . The effective intensity of the light source was determined for each photoreceptor by counting its responses to single photons , quantum bumps [81] . The photoreceptor was dark-adapted for at least 20 min , until it was sufficiently sensitive to produce clearly resolvable bumps ( Figure 1A ) . The light level was then adjusted to produce from three to ten quantum bumps per second , by inserting neutral density filters , and the bump rate determined by counting at least 100 bumps in a measured time interval . The background light level was increased in steps by removing neutral density filters , and the effective photon rate at each background was extrapolated by multiplying the bump rate by the reduction in filter attenuation . Information rates were measured from a photoreceptor's voltage response to Gaussian white noise [33] using well-established procedures [30–32 , 34] . The light source was modulated randomly for 0 . 512 s with a contrast c ( t ) = I ( t ) /Io , where I ( t ) and c ( t ) specify the intensity and contrast with time t , and Io is the mean light level . The root mean square contrast was 0 . 32 , which is close to the mean value of 0 . 4 measured for natural scenes [46] . The voltage waveform used to modulate the light source was generated digitally , by inverse Fourier transformation of a spectrum with constant amplitude and random phase , up to a cutoff frequency of 500 Hz . To iron out small inconsistencies in signal power spectra , three pseudorandom Gaussian time traces , I ( t ) were used , each repeated 50 times . This was reduced to two pseudorandom traces in D . melanogaster , where controls showed that this reduction had a negligible effect on measured information rates . The ensemble average of the photoreceptor voltage response to each sequence was derived to give the voltage signal S ( t ) ( Figure 1B and 1C ) , and this estimate of signal was subtracted from each of the responses to derive 50 noise traces . The noise traces were transformed to power spectra and ensemble averaged to give the noise power spectrum , N ( f ) , which was corrected for recording noise by subtracting the noise spectrum recorded with the electrode outside the cell . The two or three signal traces were transformed , and the spectra averaged to give the signal power spectrum , S ( f ) . A four-term Blackmann-Harris window was applied to the signal traces and the noise traces prior to transformation to the frequency domain , and the SNRs , S ( f ) /N ( f ) , were corrected for statistical bias [82] . The amplitude distributions of signal and noise were approximately Gaussian and could , therefore , be used to calculate the information rate according to Equation 1 . To improve the reliability of our conclusions , the measurements of information rates made in the five D . melanogaster photoreceptors R1–6 recorded for this study were supplemented with published data from 21 cells [32] , giving 26 cells in all . The information rates obtained from the five new cells were very similar to those obtained earlier , despite having been obtained on a different setup . The membrane resistance was measured from recordings of the photoreceptor membrane's voltage response to current that was injected via the recording electrode , using a discontinuous switched clamp . In the experiments performed on Calliphora the resistance was estimated from the response to injected white-noise current , because this method also measures the dynamic impedance of the membrane [83] . The current was modulated using digitally generated waveforms , as described above for the white-noise optical stimulus . The pseudorandom current sequences had a zero mean , and their root mean square amplitude was adjusted to recording conditions , to generate a peak-to-peak membrane response of 2–4 mV . The average voltage response to current v ( t ) was calculated by ensemble averaging 200 repeats of the pseudorandom white-noise stimulus and transformed to the response power spectrum V ( f ) . Dividing V ( f ) by the power spectrum of applied current i ( f ) , yielded the impedance Z ( f ) . The membrane resistance , RM , was estimated from the zero frequency asymptote of Z ( f ) . In the other three species , the photoreceptor membrane resistance was measured from the change in membrane potential produced by a low amplitude current pulse . The pulse's duration was adjusted so that it fully charged the membrane capacitance and , to ensure that the activation of voltage-sensitive conductance had a negligible effect on these measurements , the current was reduced to a level ( ∼50 μA ) , where positive and negative pulses produced symmetrical responses . The responses to several hundred current pulses were averaged to generate a reliable estimate of the voltage change . By using very small currents , this second method returns a value of membrane resistance that is closer to the steady state because it reduces artefacts due to rectification . This may explain why the resistances measured in Sarcophaga R1–6 using current pulses are slightly higher than those measured in Calliphora , using white-noise current ( Figure 3 ) , even though Sarcophaga achieves high information rates ( Figure 2 ) . In addition the smaller currents have less of a deleterious effect on recording stability . ATP consumption was estimated by applying measurements of membrane resistance and potential to a standard membrane model ( Figure 3 ) of the insect photoreceptor [34] . Additional data on the membrane resistance and potential of D . melanogaster photoreceptors were obtained from the literature as follows . The single set of values of membrane resistance versus background intensity reported by Juusola and Hardie [31] were added to the measurements from five cells obtained for the present study , to give sets of values for six cells . We took 21 sets of measurements of membrane potential versus background light intensity from an earlier study by Niven et al . [32] , which , with the five new cells recorded for this study , gave 26 sets of values in all . The model incorporates the three dominant membrane mechanisms , a light-gated conductance gL with reversal potential EL = −5 mV [84] , a potassium conductance gK with reversal potential EK = −85 mV [28] , and a standard Na+/K+ pump that generates a pump current , ip by exporting three Na+ ions and importing two K+ ions per ATP molecule hydrolysed [85 , 86] . When the photoreceptor is in the steady state and has a membrane potential EM , the light-gated conductance and the potassium conductance produce transmembrane currents iL = ( Em − EL ) gL and iK = ( Em − EK ) gK . In order to maintain ionic homeostasis the pump current must be ip = iK/2 . The pump current can be derived from the membrane model by equating currents across the model membrane iK + iL + ip = 0 setting gK + gL = 1/RM where RM is the measured membrane resistance , and inserting the measured membrane potential , EM . The rate of ATP hydrolysis required to generate this steady-state pump current is our estimate of the rate at which the photoreceptor consumes energy . Following a stable electrode penetration , the photoreceptor was dark adapted for at least 20 minutes and then calibrated by counting quantum bumps ( Figure 1A ) , as described above . A neutral density filter was removed from the light beam to set the first background light level , the background light was switched on , and the cell was adapted for at least 2 min , until its membrane potential reached a stable steady state . Once the photoreceptor was stably light adapted , the membrane potential EM , the impedance Z ( f ) , and the information rate were successively measured , using the procedures described above . The membrane potential was then checked for drift . The light was then extinguished , the stability of the resting potential was checked for 30 s–1 min , and following withdrawal of another neutral density filter , the next highest background was switched on . This sequence of stable light adaptation and measurement was repeated until the maximum effective intensity was reached . The light was then extinguished , and the cell left in the dark to check that the resting potential returned to a value that was within 2 mV of that measured at the start of the experiment . Data from cells that failed this final test were rejected . Finally , the electrode noise was measured with the electrode just outside the cell , in a position where the noise amplitude was at a minimum .
|
Many animals show striking reductions or enlargements of sense organs or brain regions according to their lifestyle and habitat . For example , cave dwelling or subterranean animals often have reduced eyes and brain regions involved in visual processing . These differences suggest that although there are benefits to possessing a particular sense organ or brain region , there are also significant costs that shape the evolution of the nervous system , but little is known about this trade-off , particularly at the level of single neurons . We measured the trade-off between performance and energetic costs by recording electrical signals from single photoreceptors in different fly species . We discovered that photoreceptors in the blowfly transmit five times more information than the smaller photoreceptors of the diminutive fruit fly Drosophila . The blowfly pays a high price for better performance; its photoreceptor uses ten times more energy to code the same quantity of information . We conclude that , for basic biophysical reasons , neuronal energy consumption increases much more steeply than performance , and this intensifies the evolutionary pressure to reduce performance to the minimum required for adequate function . Thus the biophysical properties of sensory neurons help to explain why the sense organs and brains of different species vary in size and performance .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"physiology",
"computational",
"biology",
"evolutionary",
"biology",
"drosophila",
"biophysics",
"neuroscience"
] |
2007
|
Fly Photoreceptors Demonstrate Energy-Information Trade-Offs in Neural Coding
|
We present a data-driven mathematical model of a key initiating step in platelet activation , a central process in the prevention of bleeding following Injury . In vascular disease , this process is activated inappropriately and causes thrombosis , heart attacks and stroke . The collagen receptor GPVI is the primary trigger for platelet activation at sites of injury . Understanding the complex molecular mechanisms initiated by this receptor is important for development of more effective antithrombotic medicines . In this work we developed a series of nonlinear ordinary differential equation models that are direct representations of biological hypotheses surrounding the initial steps in GPVI-stimulated signal transduction . At each stage model simulations were compared to our own quantitative , high-temporal experimental data that guides further experimental design , data collection and model refinement . Much is known about the linear forward reactions within platelet signalling pathways but knowledge of the roles of putative reverse reactions are poorly understood . An initial model , that includes a simple constitutively active phosphatase , was unable to explain experimental data . Model revisions , incorporating a complex pathway of interactions ( and specifically the phosphatase TULA-2 ) , provided a good description of the experimental data both based on observations of phosphorylation in samples from one donor and in those of a wider population . Our model was used to investigate the levels of proteins involved in regulating the pathway and the effect of low GPVI levels that have been associated with disease . Results indicate a clear separation in healthy and GPVI deficient states in respect of the signalling cascade dynamics associated with Syk tyrosine phosphorylation and activation . Our approach reveals the central importance of this negative feedback pathway that results in the temporal regulation of a specific class of protein tyrosine phosphatases in controlling the rate , and therefore extent , of GPVI-stimulated platelet activation .
Platelets , small anuclear cells , are quiescent in undamaged blood vessels . They respond to injury by activating , triggering blood to clot . Whilst platelet activation is essential to prevent excessive bleeding at sites of injury , their inappropriate activation , for example as a consequence of vascular disease , can lead to the formation of clots within the circulation , or thrombosis which triggers heart attacks and strokes [1] . Damage to blood vessels results in the exposure of extracellular matrix proteins , particularly collagens , which form structural components within the vessel wall . Collagens provide an initiating signal for platelet activation . Indirect interactions between platelets and collagen , in the arterial circulation , are mediated by von Willebrand factor ( vWF ) which binds to collagen and to GPIb on the platelet's surface . This causes platelets to slow down and roll along the site of vessel damage allowing direct binding to platelet collagen receptors , including integrin α2β1 , which largely supports adhesion , and glycoprotein ( GP ) VI which stimulates cell signalling and activation [2] . Platelet activation is marked by a dramatic change in platelet shape , the secretion of various prothrombotic factors and conformational change in the integrin αIIbβ3 . These secreted factors initiate a second wave of signalling . Integrin αIIbβ3 binding to fibrinogen , which in itself stimulates signalling , supports platelet aggregate formation allowing the assembly of a platelet thrombus ( or haemostatic plug ) to stem the loss of blood [2 , 3] . GPVI is a member of the immunoglobulin family of receptors that shares aspects of its mechanisms of activation with immunoreceptors , including B and T cell antigen receptors [4] . These receptors possess or associate with transmembrane proteins that have cytoplasmic domains that contain immunoreceptor tyrosine-based activation motifs ( ITAMs ) . The binding of a ligand leads to associated Src Family Kinases ( SFKs ) phosphorylating conserved tyrosine residues on the ITAM . In the case of GPVI , Syk ( spleen tyrosine kinase ) , or the related protein ZAP70 in T cells , is recruited to the receptor complex through direct binding to the phosphorylated ITAM . Recruitment of Syk in turn provokes its phosphorylation [5] , pivotal to downstream signalling events: in platelets this leads to activation , shape change and aggregation ( Fig 1 ) [6] . The development of new drugs to suppress platelet function and thereby prevent thrombosis has been shown to be an effective strategy [7–9]; however , current anti-platelet therapies are ineffective in many patients and are associated with side effects . A more detailed understanding of the signalling pathways that lead to platelet activation is needed to develop more sophisticated , effective and safer anti-platelet therapies [10] . GPVI is a candidate anti-platelet drug target [11 , 12] and the central role of Syk in immune cell signalling also makes it an attractive therapeutic option , with implications for the treatment of allergy , auto-immune diseases and haematological malignancies [13] . In other cell types mathematical and computational models have aided in the understanding of signalling pathways and assisted further experimental work [14–16] . The formation of blood clots and thrombi has long been of mathematical interest but , probably due to the ready availability of data , mathematical modelling has focussed on the coagulation cascade [17–23] , i . e . the network of interactions between blood proteins that results in the formation of fibrin polymers that stabilise a clot . These models show promise in translating an individuals' coagulant profile into a measure of thrombotic risk [24 , 25] . Even though platelets are an ideal cell type to model ( they lack a nucleus thereby removing the complexities of gene transcription and translation ) it is only recently that the platelet has become of mathematical interest . The main focus of recent developments have been the interactions between a platelet and the coagulation cascade [26 , 27] and the role of platelets in a thrombus shaped by blood flow [28–31] , there being several excellent reviews [32–34] . Where platelets have been studied in more detail investigations focus on the functional response of platelets to agonists [35] and the mechanics of signalling pathways [36–39] . Dandekar and co-workers [36 , 37] focused on platelet signalling mechanisms and the effects of prostacyclin , a molecule secreted from endothelial cells that contributes to platelets remaining in a quiescent state in the undamaged circulation . Their studies were based on platelet-specific data and simulated and tested different pharmacological conditions and the contributions of various platelet receptors . A detailed mathematical model [38] that captures current knowledge of platelet signal transduction through the ADP receptor is able accurately to predict calcium flux and ADP dose-response . This work was used as a basis for a model of PAR-1 mediated platelet activation [39] that includes the critical step of linking inside-out to outside-in integrin signalling . These latter two studies focus on the events downstream of G-protein coupled receptors through to intracellular calcium mobilisation , a critical step in platelet activation that is common to all activation pathways . While GPVI activation is critical in health and disease no mathematical study has focussed on the platelet's response to collagen . The pathway of reactions downstream of the GPVI receptor is considerably more complex than those stimulated by G-protein coupled receptors . We therefore focus our investigations on the early reactions , proximal to the GPVI receptor , that lead to Syk activity that is thought to be a fundamental switch that controls GPVI mediated signal transduction . Studies to dissect the component parts and associated processes within the GPVI signalling pathway have provided a picture of the series of steps that occur [40] but these studies often lack detailed kinetic analysis and molecular detail . For example , they may measure the level of tyrosine phosphorylation of a specific protein , rather than assessing the phosphorylation of specific amino acid residues , despite these individual phosphorylated residues being linked to specific functions . Furthermore , previous attempts to model signalling pathways in this and other cell types have often been based on qualitative data , which may make it difficult to detect differences between experimental data and model predictions . Here , these limitations in available data were resolved by the collection of high density quantitative experimental observations that capture the temporal increase and decrease in tyrosine phosphorylation of key proteins at specific amino acid residues in the early steps of activation of the GPVI signalling pathway . Existing knowledge of platelet signalling pathways tends to focus on the sequential forward reactions that lead to platelet activation . While tight negative regulation of the pathways appears crucial for appropriate and rapid platelet activation , little effort has focussed on understanding what the key regulators are that modulate these reactions and how they integrate to precisely control platelet activation . Stimulation of GPVI results in the initiation of a signalling pathway that is highly dependent on the activities of a range of protein kinases . A protein tyrosine kinase transfers a phosphate group to a protein . Such phosphorylation may alter the function of a given protein , for example through inducing a conformational change , creating docking sites , causing intracellular relocation , and modulation of enzymatic activity . Protein tyrosine phosphatases reverse these protein modifications since they are able to remove phosphate groups returning proteins to their original state . Historically protein tyrosine kinases have been the focus of more research than protein tyrosine phosphatases . Recent studies indicate that human platelets possess at least 18 different protein tyrosine phosphatases , numbering more than 52 , 000 copies in total per platelet [41] . While traditionally phosphatases have been thought to be promiscuous in activity , constitutively active and able to remove phosphate moieties from proteins , it is now evident that this view is too simplistic and that phosphatases like kinases are actively recruited and regulated within specific pathways [10 , 42 , 43] . The identities and roles of phosphatases within the GPVI signalling pathway are not well understood , although it is clear that some play a fundamental role in the control of platelet function following stimulation with collagen [44 , 45 , 46] . Biological knowledge of components and interactions of signalling pathways is often expressed by cartoons or network diagrams . These descriptive static models provide little information about how the many components evolve over time , making it difficult to predict system behaviour . Here we reinterpret network diagrams , representing current biological knowledge and hypotheses surrounding the early regulation of the GPVI signalling pathway , into a series of dynamic mathematical models that include various possible modes of regulation by phosphatases . This allows direct comparison between knowledge and hypotheses and high density quantitative , temporal experimental data generated specifically to inform the development of the models . Through a process of parameter estimation , model validation and comparison with experimental data we gain understanding of the temporal dynamics that control the initiation of the GPVI signalling pathway allowing us to explore how perturbations of different kinds ( e . g . ligand availability , negative regulation processes , and expression of key regulatory proteins ) effect the ability of GPVI to signal downstream .
In all models we treated GPVI , its associated Fc receptor γ-chain and SFKs as a single entity ( see Fig 2 ) . The ligand ( CRP and indeed collagen ) is a complex one being of indeterminant length , weight and number of binding sites for the GPVI receptor . While ligand mediated clustering is thought to be an important attribute the difficulty with obtaining experimental data covering these very early timepoints leads us to neglect these complications and assume an abundant ligand source . A ligand binds to this receptor bundle , which is subsequently phosphorylated , allowing for recruitment and activation of the cytosolic protein-tyrosine kinase Syk . Binding to the receptor leads to auto-phosphorylation of Syk , on tyrosine 525 ( Y525 ) , that allows the receptor to signal downstream [48–50] . The early reactions of ITAM phosphorylation and Syk binding are unlikely to be rapidly reversible . The binding affinity of Syk to the ITAM is high , occurring through two tyrosine residues [5] that protect the ITAM from dephosphorylation . Experimental data shows quantitative differences between protein copy numbers and equilibrium levels of Syk phosphorylation , pointing to regulatory activity . Syk phosphorylation is an obvious candidate for regulation , most likely through a phosphatase . Phosphatases have traditionally been viewed as promiscuous housekeeping molecules that act quickly on exposed phosphoproteins [43] . Our initial model ( A ) incorporated the continuous presence of a simple abundant generic phosphatase that is able to dephosphorylate Syk on Y525 , thereby restricting the ability of Syk ( and therefore of the GPVI receptor ) to signal downstream . Full details of the model and the parameter values obtained are given in S1 Text . Model solutions describing Syk phosphorylation monotonically rise and settle to the steady state seen in experimental data ( see Fig 4 , top row ) . The steady state is solely determined by Syk's rate of phosphorylation and its regulation , such that Bound , inactive Syk =γ1p1+γ1sI , Active Syk =p1γ1+p1sI , ( 1 ) Where sI denotes Syk copy numbers , p1 the rate that Syk is phosphorylated and γ1 the rate at which it is dephosphorylated . Local sensitivity analysis ( see Methods ) reveals that the model's steady state is insensitive to variation in the rate of ITAM phosphorylation and the rate that Syk binds to the phosphorylated ITAM . While Model A is able to describe the equilibrium to which Syk activity settles , it is unable to capture its early transient peak in phosphorylation . This early peak in Syk phosphorylation is reminiscent of negative regulatory behaviour affecting Syk activity and points to the constitutively activated phosphatase being insufficient to explain this behaviour . Therefore the model was adapted , replacing the 'housekeeping' phosphatase with a more complex pathway that involves a specific phosphatase , TULA-2 . While it is known that TULA-2 can interact with Syk , in a ubiquitin-dependent manner [51] , the purpose of the interaction had not previously been established . This interaction becomes the template for model adaption . The mathematical model was extended to incorporate a negative feedback , which leads from Syk phosphorylation to its own regulation , with the intention of explaining the early peak in Syk activity seen in experimental data . This newly introduced regulatory pathway centres on the proteins c-Cbl and TULA-2 . c-Cbl is a member of the Cbl ( Casitas B-lineage lymphoma ) family of adaptor proteins and is known to be found in platelets [50 , 52] . It has been shown to play a role in regulating signals by similar receptors in other cell types [53 , 54] and it has been implicated in GPVI regulation in platelets [52] . c-Cbl is a ubiquitin ligase that is able to associate with phosphorylated tyrosine kinases , such as Syk , whereby it can promote their ubiquitination [55] . Syk is known to be ubiquitinated rapidly upon activation by CRP and collagen [50] . c-Cbl has no phosphatase domain but has been suggested as a necessary intermediate scaffold between Syk and its phosphatase [50] . The TULA ( T-cell Ubiquitin Ligand ) family of proteins are histidine tyrosine phosphatases that have been shown to be important negative regulators in immune cells [56] . TULA proteins are able to bind ubiquitinylated proteins and they play a role in regulating T-cells , decreasing Syk related protein ( ZAP-70 ) phosphorylation [57 , 58] . TULA-2 exists in platelets [41 , 51] where it has been shown to associate with , and subsequently dephosphorylate , Syk [56] . Model B incorporates an additional Syk phosphorylation site ( Y323 ) that is a known binding site for c-Cbl [59] . In the model , binding of c-Cbl to Y323 leads to ubiquitination of Syk , which allows TULA-2 to associate with Syk and dephosphorylate its activatory phosphorylation site ( Y525 ) , returning the receptor complex to an inactive state . All reactions are assumed to be reversible . A schematic of these newly introduced reactions is given in Fig 2 , a network diagram in Fig 3 and the corresponding equations and parameters obtained from the parameter fitting process are given in S1 Text . Experimental data specifically describing the newly introduced phosphorylation site was collected and Model B was fitted to both sets of experimental data simultaneously ( Fig 5a and 5b ) . Model profiles describe accurately the steady state representing Syk phosphorylation on Y525 and the newly introduced regulatory site ( Y323 ) . They failed , however , to capture the early transient peak seen in experimental observations of both phosphorylation sites . If Model B is fitted to data restricted to one phosphorylation site ( Y525 ) then while model solutions capture the full dynamics seen in Syk Y525 phosphorylation predictions for phosphorylation on Y323 are inaccurate , settling to a level that is three fold higher than that seen experimentally ( Fig 5c and 5d ) . The model steady state representing Syk phosphorylation on Y525 is given by Active Syk =sI−γ1p1G11r , ( 2 ) where G11r denotes the phosphatase TULA-2 bound to Syk , and hence able to dephosphorylate it . Local sensitivity analysis ( see Methods ) revealed that the time to reach equilibrium was still predominantly influenced by the rate that the ITAM is phosphorylated; the level of steady state ( for the activatory and regulatory phosphorylation site ) is most strongly influenced by the strength of the parameters that comprise the feedback loop . The level of Syk Y525 at steady state is also sensitive to the rate of Syk phosphorylation on Y525 and its reversal . This is reflected in the change in parameter values when fitted to both sets of data; the rate of all reactions within the feedback pathway and the rate that Y525 are increased to allow the model to fit the data . Model B accurately describes the steady states seen in the experimental observations but , though showing promise , fails to describe fully their kinetics . This led us to investigate biologically plausible refinements that would enable the model to recapture the early transitory behaviour we had seen when the model was fitted to data solely describing phosphorylation in Syk Y525 . In formulating Model C two biologically plausible modifications to our model were explored . The first modification ( H1 ) allows the TULA-2 , when bound to the receptor complex , to dephosphorylate other Syk molecules . The second modification ( H2 ) assumes that the increase in Syk activity ( phosphorylation on Y525 ) increases the rate at which Y323 is phosphorylated—increases in Syk activity following Y525 phosphorylation have previously been reported [47] . The simultaneous inclusion of both modifications is denoted by Model C , H3 . These reactions are depicted in Figs 2 and 3 and the corresponding equations are given in S1 Text . Model C , H1 and H2 profiles ( Fig 6a and 6b broken and dotted lines respectively ) fail to describe transitory behaviour in both sets of experimental data . Model C , H3 ( Fig 6a and 6b , solid line ) incorporates H1 and H2 simultaneously and is able to describe accurately both sets of experimental observations . These optimal fits were dependent on the models being simultaneously fitted to both sets of experimental data . Like Model B , if fitted to one set of observations ( i . e . Y525 ) , model predictions for Y323 were inaccurate ( Fig 7c and 7d ) . The steady state that model profiles representing Syk phosphorylation on Y525 is determined by Active Syk =sI−γ1p1G11r− ( TULA-2 bound ) ( GY525−G11r ) , ( 3 ) where GY525 denotes the portion of Syk molecules that are bound to the receptor complex and phosphorylated on Y525 . Local sensitivity analysis ( see Methods ) revealed ( see Fig G in S1 Text ) that the steady state is influenced by parameters that comprise the whole regulatory pathway reflected in the above by the proportion of TULA-2 that is bound to the receptor complex . In agreement with Thomas et al [51] the above expression ( 3 ) and sensitivity analysis ( see Fig J in S1 Text ) confirm that a decrease in the levels of TULA-2 lead to hyperphosphorylation of Syk . The time to reach the above steady state rate is shown to be predominantly influenced by the rate of ITAM phosphorylation while the time to reach the steady state of the regulatory phosphorylation site ( Y323 ) is influenced by the rate of ITAM phosphorylation and by protein copy numbers of GPVI and Syk . Of the parameters that were held fixed during the parameter fitting process ( platelet volume , extracellular volume and rates of ligand binding and dissociation ) only the platelet volume influences Syk activity . In summary , the ability of the model to describe the full range of dynamics seen in the experimental observations requires that Syk activity increases the rate of phosphorylation on Y323 and that TULA-2 is able to dephosphorylate not only the receptor to which it is bound but also any nearby receptor complex . To understand if the extension of each model is justified through a better fit to data we use the Akaike information criterion ( AIC ) and a modified form that takes into account small experimental sample sizes ( AICc ) —a description of the metric and how we apply it is given in Methods and the results are shown in Table 3 . The model with the lowest AICc provides the best balance between its replication of data and the added complexity introduced to achieve it . Model C ( H3 ) has the lowest AICc; it provides the most faithful replication of both sets of data ( a total of forty-four experimental observations ) and is not overly compromised by its additional complexity when compared to Model B . If data are restricted to that describing one phosphorylation site ( Y525 , twenty-two time-points ) then Model A has the lowest AICc; Model B and C are too complicated ( too many parameters ) to be inferred from such a set of observations . Overall , these results confirm Model C , H3 as our preferred model and this was used for all further simulations .
A major success over the last 20 years has been the identification of the key components of the major signal transduction pathways in cells [2–7] . What is less well understood is how these components come together in space and time in a regulated manor to bring about appropriate cellular responses . Whilst the activation of protein kinases has been studied in some detail , much less work has been done on the regulation of phosphatases in part due to the lack of specific inhibitors . A search of the literature shows that there are 20 times the number of papers on protein kinases in platelets as there are on protein phosphatases . However as noted earlier most cells , including platelets , contain many different types of protein-tyrosine phosphatases ( there are roughly the same number of different protein-tyrosine phosphatases as there are protein tyrosine kinases in platelets [40] ) , that show distinct subcellular compartmentalisation and regulation . Some appear to be simple abundant cytosolic phosphatases without obvious regulatory domains which may be constitutively active , some are transmembrane receptor type phosphatases such as CD148 , whilst others are highly regulated containing domains that allow them to be temporally recruited to specific signalling complexes either via SH2 domain phosphotyrosine interactions ( eg . , SHP1 and SHP2 ) or through ubiquitin-dependent binding ( TULA-1 and TULA-2 ) [40–45] . Thus the challenge now is to determine which of these phosphatases are required and sufficient to explain the regulation of the key determining steps in individual signalling cascades . The objective was to develop a mathematical model of GPVI proximal signalling that allows the comparison of current biological knowledge and hypotheses to high-density temporal experimental data explicitly collected to inform the model . The experimental observations describe in particular phosphorylation of Syk , a key protein that is recruited to the GPVI receptor and which initiates downstream signalling . It was anticipated that this approach would lead to increased knowledge surrounding the interaction of the components of this key step in platelet activation and , importantly , how this is controlled . The model was developed incrementally . Starting with a simple model ( A ) that captured current biological knowledge of the components and interactions that occur just downstream of the GPVI receptor more complicated interactions were incorporated until a good description of experimental data was achieved . While the components that comprise the forward steps of platelet signalling have been well studied , regulatory processes have not . The experimental data pointed to the pathway being regulated and , in consequence , a simple housekeeping phosphatase that limits the ability of Syk ( and therefore the receptor ) to signal downstream was incorporated into Model A . The model was unable to explain the full dynamics seen in experimental observations and led us to conclude that the inclusion of a simple phosphatase , despite their abundance in platelets [40] , is insufficient to explain data . A potential regulatory pathway was selected , supported by current literature , which focuses on the Syk protein and its ability to participate in its own regulation through proteins c-Cbl and TULA-2 [49–55] . Our initial attempts to explain the experimental data while promising did not capture the full range of dynamics displayed in observations of the central protein Syk on a phosphorylation site that we equate to activity and one that initiates the ability of Syk to regulate itself [46–47] . With further model refinements ( Model C ) we found that a model with c-Cbl/TULA-2 incorporated could provide not only a good quantitative fit to experimental data but could capture its full dynamics . As demonstrated in sensitivity analysis the inclusion of the negative feedback loop protects Syk activity from variation in the number of Syk molecules . Interestingly , if the experimental data is limited to that describing a single Syk phosphorylation site ( Y525 ) then the inclusion of the more complex regulatory mechanism ( centring on c-Cbl and TULA-2 ) cannot be justified . It is only when data describing a second phosphorylation site was collected that Model B and Model C could be successfully differentiated . This emphasises the need for high density data to allow the development of such complex models that incorporate more subtle processes ( and many parameters ) . The model predicts that , on average , very few molecules of c-Cbl and TULA-2 need to be bound to the receptor complex at any point in time . This in turn suggests a significant stochastic component to the behaviour . This , and the prediction that regulation of activated Syk molecules relies on the ability of the bound phosphatase , TULA-2 , being able to dephosphorylate many activated Syk molecules , point to receptors being clustered at the plasma membrane . This fits well with the localisation of GPVI to specific membrane domains and the known ability of the multivalent ligand , collagen , to drive localised clustering of GPVI at the platelet surface [65] but also highlights the importance of further spatial mathematical investigations . To be of wider use it is essential that the model can accurately predict experimental signalling outputs under varying conditions . The model was tailored to an environment where the ligand is abundant but is able to predict Syk activity in response to decreasing levels of ligand availability . A decrease in the availability of the ligand in the model results in a reduction of the early transient peak and a delay to reach the steady state which shows good agreement with independently generated experimental data . Platelet signalling is likely to demonstrate a wide degree of variation across a healthy population but little is known about the extent and at what point this leads to potential pathologies . This normal variation is seen in the levels of the individual proteins involved in our model , both in our data and the recent quantitative human platelet proteome [40] and also in the signalling responses of individual donors studied experimentally ( Fig 7 ) . In the first instance the model was fitted to data based on triplicate samples from one healthy donor . By comparison to data describing Syk phosphorylation , on both the activatory ( Y525 ) and regulatory ( Y323 ) sites , in nine additional donors we found the model was able to replicate phosphorylation profiles observed in a wider healthy population . The model was also utilised to investigate signalling responses in a population that has reduced levels of GPVI receptors [1 , 2] with normal population variability of expression levels of GPVI and other signalling components involved ( Fig 10 ) . This suggested that the time to peak of Syk activation in normal individuals seems to be tightly controlled whilst the actual peak maxima of Syk Y525 phosphorylation is more variable , as was also seen experimentally in our normal panel of nine healthy donors who all had very similar times to peak but quite variable absolute levels of Syk Y525 phosphorylation ( Fig 7 ) . The simulated population with reduced GPVI levels however had a longer time to reach this peak and the timing seems to be less tightly controlled within the simulated population . This may suggest that temporal regulation of this signalling pathway may be more important to a functional response than the absolute magnitude of the response . In summary , this study aims to bridge the gap between biological knowledge and hypotheses and temporal experimental data in aiding understanding of how the components of a key initiating step in platelet activation interact to control precisely a platelet's response to its environment . Platelet activation has traditionally been thought of as a series of forward steps resulting in the platelet switching on . While the fine details of even this initiating step will continue to be elucidated , the principal of negative pathways playing a central role is clear and highlights the complexity of these signalling pathways . Indeed there is evidence of a number of inhibitory receptors and signalling mechnisms including those stimulated by nitric oxide and prostacyclin that serve to prevent un-required activation [69] . Some mechanisms such as those stimulated by the adhesion receptor PECAM-1 are known to modulate signalling downstream of the processes explored in this study , although we cannot rule the potential impact of these mechanisms on the initiation of GPVI signalling . Given the outcomes of this study , this will be the focus for future work . While our approach here has focused on elucidating a key regulatory step in a platelet signalling pathway we believe that this approach of integrating biological knowledge with time-rich kinetic data sets and mathematical models is well suited to the analysis of the regulation of any signalling cascade for which suitable reagents are available .
Here we briefly describe recently established methods utilised for the measurement and quantification of protein copy numbers and their post translational modifications . Please see [70] for further details . The quantification method is based on the relationship between a known amount of a specific protein to a quantitative signal emitted by fluorescently conjugated antibodies . Individual phosphorylated proteins were first isolated by immunopreciptation ( IP ) using suitable site-specific antibodies . Identical serial dilutions of the IP were loaded onto two SDS-PAGE gels . One gel was also loaded with a serial dilution of a known concentration of a corresponding non-phosphorylated recombinant protein . The resulting immunoblot was probed with an antibody recognising the total protein ( i . e . non phosphorylated ) . The second gel loaded with the IP serial dilutions was blotted then probed with the same phospho-specific antibody used for the IP . Both gels also contained serial dilutions of known amount of IgG necessary to calibrate with subsequent experiments . The different immunoblots were then treated with appropriate fluorescent conjugated secondary antibodies and scanned with a fluorescence imaging system . The resulting values for the recombinant protein dilution were then used to construct a standard curve and determine the concentration of molecules immunoprecipitated . Combined with the results from the second immunoblot allows the direct relationship between the amount of protein actually present on the immunoblot and fluorescent levels obtained when probing with a phosphospecific antibody to be determined . Further experimental samples were loaded on SDS-PAGE gels together with identical IgG serial dilutions for normalisation with the reference datasets and quantification of the amount of phosphorylated proteins . The amount of total protein was established by comparing the normalised fluorescence intensity of the samples and the recombinant protein standard curve when probed with the antibody recognising the total protein . Blood samples were obtained from healthy volunteers that had given consent , using procedures approved by University of Reading Research Ethics Committee . Washed platelets were prepared by differential centrifugation , as described previously [70] , and resuspended in Tyrodes buffer containing 0 . 4U/ml Apyrase , 1mM EGTA and 100 μM Indomethacin; to suppress secondary signalling and secretion . Samples were stimulated with CRP-XL ( provided by Prof Richard Farndale , University of Cambridge , UK ) at a final concentration of 10 μg/ml , then lysed , denatured and loaded onto 10% SDS-PAGE gels . CRP-XL , unlike collagen , is a GPVI selective agonist . Reference datasets were constructed using platelets treated with pervanadate [71] ( 10 μg/ml ) CRP for 1 min . Immunoprecipitation was carried out using PureProteome Protein A magnetic beads ( Millipore ) . Immunoprecipitates and experimental samples were loaded alongside the corresponding recombinant protein ( Syk , GST-tagged , Abnova; c-CBL proprietary tag , Abcam ) and a series of dilutions of IgG ( Rabbit and Murine Isotype controls , US Biological ) . Westernblot transfer to Immobilon-FL membrane ( Millipore ) was performed using a semi-dry blotter ( Bio-Rad ) . The membranes were blocked with 5% ( w/v ) BSA and probed as instructed by the manufacturer using the Millipore SNAP i . d . protein detection system and the appropriate antibodies ( Anti-Syk N-19 , Santa-Cruz Biotechnology Inc; anti-c-Cbl , BD Biosciences; anti-Syk phospho Y525+Y526 , anti-c-Syk Y323 ) . Immunoblots were treated with either a fluorescent Cy5 dye-labeled goat anti-Rabbit or a Alexa-Fluor 647 dye-labeled donkey anti-mouse ( Life Technologies ) . Digital scan values of the fluorescence emission were obtained using a Typhoon Trio scanner ( GE Healthcare ) . Quantification was performed using the ImageQuant TL software and data analysed using R [72] . We estimated all unknown parameter values for our models by utilising a constrained local optimization routine ( MATLAB's fmincon ) that varies all unknown parameters to minimize the differences between the model and experimental data via a cost function Sum Squared due to Error ( SSE ) = ( yi ( θ ) −Datai ) 2 ( 4 ) where yi ( θ ) is the model’s prediction for the relevant model variable ( which depends on the parameters θ ) and Datai represents the experimental observations at time points i . Given two sets of parameters , the one with the smaller cost function is the one that provides the better description of experimental observables . The values that parameters can take are restricted within limits that are guided by biological knowledge and literature ( S1 Text includes a summary of the limits used and how they were set ) . The optimization algorithm takes steps that successively decrease the value of the cost function , beginning from an initial guess of the parameter values . To avoid parameter sets being selected that fit local , rather than global , minima we utilize a multi-start approach where repeated optimization runs are carried out with multiple initial guesses . We sampled initial guesses from the complete allowable parameter range utilising Latin Hypercube Sampling ( LHS ) . The range that parameter values can take is discretised into N bins . Where parameter limits extend several orders of magnitude , a logscale is used , which leads to a better representation of parameter space . Each interval in the parameter range is sampled once ( without replacement ) , so that the entire range for each parameter is explored . The resulting N samples are used as initial guesses . Parameter estimation is performed utilising samples N = 1000 . The five parameter sets with the lowest SSE N = 1000 are reported in each case to provide confidence that the best fit is converging , for the final model ( C , H3 ) samples of N = 10000 are reported , and needed to converge . We performed a local sensitivity analysis , where each parameter was varied by fifty and ninety percent above and below their initially estimated value , and computed the normalised local sensitivity of either the final steady state or the time to reach this steady state according to Sensitivity Score = ( Oa−Oi ) Oa ( 5 ) where Oi and Oa represent the model output ( e . g . the steady state and time to reach the peak in Syk activity ) in respect of the initial parameter set ( obtained from parameter fitting ) and the adapted parameter respectively . To aid model comparisons we used Akaike’s Information Criterion ( AIC ) [73] . This is an information-theoretic criterion for model comparison , which incorporates not only the cost function value from the parameter fitting process ( SSE ) but also a penalty based on the number of parameters ( K ) in the model . We used a modified criterion [68] that takes into account the experimental sample size ( n ) by increasing the relative penalty for model complexity with small datasets defined by AICc=AIC+2K ( K+1 ) ( n−K−1 ) ( 6 ) where AIC=n ( ln ( SSE/n ) ) +2K . ( 7 ) AICc converges to AIC as n becomes large with respect to K . The value of AICc has no meaning in isolation , its relevance only becoming apparent when it is used to compare ( and rank ) models fitted to the same experimental data . In general a better fit to experimental data is achieved when the complexity of a model is increased , simply because the number of free parameters increases . The AICc ( and indeed the AIC ) aims to protect against this by a penalty based on the number of parameters in the model with the lowest AICc being the better model .
|
Platelets are blood cells that , upon injury , trigger the blood to clot . Following blood vessel damage platelets encounter the extracellular matrix protein collagen to which they respond . They become activated , aggregating to form a major component of blood clots . The platelet collagen receptor GPVI stimulates platelet activation through a complex signalling pathway , and while many of the molecules involved in the activation of this pathway have been identified , their specific roles in determining the rate and extent of the exceptionally rapid platelet response have not been determined . Furthermore , while signalling proteins responsible for forward reactions are known , reverse or negative feedback elements are not well understood . Platelets also trigger thrombosis in diseased arteries , causing heart attacks and strokes , and therefore platelets , and particularly the GPVI signalling pathway , are therapeutic targets . To begin to understand the components in the GPVI signalling pathways that may represent tractable therapeutic targets we have developed a mathematical model of the key initiating events that occur upon stimulation of GPVI . In so doing , we have established the importance of a specific phosphatase-controlled negative feedback in determining the rate of initiation of platelet activation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[] |
2015
|
Regulation of Early Steps of GPVI Signal Transduction by Phosphatases: A Systems Biology Approach
|
In healthy adult mice , the β cell population is not maintained by stem cells but instead by the replication of differentiated β cells . It is not known , however , whether all β cells contribute equally to growth and maintenance , as it may be that some cells replicate while others do not . Understanding precisely which cells are responsible for β cell replication will inform attempts to expand β cells in vitro , a potential source for cell replacement therapy to treat diabetes . Two experiments were performed to address this issue . First , the level of fluorescence generated by a pulse of histone 2B–green fluorescent protein ( H2BGFP ) expression was followed over time to determine how this marker is diluted with cell division; a uniform loss of label across the entire β cell population was observed . Second , clonal analysis of dividing β cells was completed; all clones were of comparable size . These results support the conclusion that the β cell pool is homogeneous with respect to replicative capacity and suggest that all β cells are candidates for in vitro expansion . Given similar observations in the hepatocyte population , we speculate that for tissues lacking an adult stem cell , they are replenished equally by replication of all differentiated cells .
Stem cells are defined by an ability to self-renew and differentiate into a variety of cell types . Some adult organs , including the intestine , skin , blood , and parts of the brain , are maintained by stem cells [1–5] . In cases where the differentiated cells are postmitotic , such as erythrocytes and olfactory neurons , tissue turnover depends entirely on stem cell differentiation . To explain the mechanism of β cell maintenance and regenerative repair , it has been hypothesized that renewal occurs via an adult stem cell residing in the pancreatic ducts [6] , acini [7] , islets [8 , 9] , spleen [10] , or bone marrow [11] . In contrast , Dor et al . found that pre-existing β cells , rather than stem cells , are the major source of new β cells in healthy and pancreatectomized mice [12] . Furthermore , the forced cell cycle arrest of β cells severely restricts postnatal β cell mass [13] , indicating that non–β cells ( such as putative adult stem cells ) cannot maintain β cell mass . Together , these results demonstrate that β cell mass is predominately , if not exclusively , sustained through the replication of β cells . It remains unclear whether all β cells contribute equally to growth and maintenance . Two possible models might explain the expansion of β cells . The β cell population may be heterogeneous , comprised of both highly replicative cells and very slowly dividing , possibly postmitotic , cells . This would be consistent with the hypothesis that a subpopulation of insulin-expressing cells may maintain the entire pool , perhaps as unipotent adult stem cells [14] or by reversible dedifferentiation to a replicative state [15] . Alternately , the β cell population may be homogeneous , with all β cells contributing equally to growth . Two approaches were used to address this issue ( Figure 1 ) . First , a broad survey of the replicative potential of the entire β cell pool was performed by monitoring the dilution or disappearance of a fluorescent marker accompanying cell division . β cells were pulse labeled with a tetracycline-inducible histone 2B–green fluorescent protein ( H2BGFP ) [16] and , following a chase period , the level of fluorescence detectable within β cells was measured . Second , the clonal descendents of individual β cells were examined using a reporter system developed for mosaic analysis with double markers ( MADM ) [17] . Both assays are designed to assess whether β cells are a heterogeneous population . If β cells are heterogeneous , highly replicative β cells will lose the H2BGFP label quickly as they replicate and generate large clones , while slowly dividing β cells will retain the H2BGFP label and generate small clones . Alternately , if β cells are a homogenous population , all β cells would be expected to lose the H2BGFP label at similar rates , and all clones should be of comparable size . We observed uniform loss of the H2BGFP label with time , and detected only similarly sized clones in the chase population . The tetracycline-inducible H2BGFP and MADM systems are complementary approaches , both supporting the conclusion that all β cells contribute equally to β cell growth and maintenance .
Tumbar et al . engineered transgenic mice expressing H2BGFP from a tetracycline-responsive promoter ( tetracycline-inducible promoter [tetO]–H2BGFP ) to mark cells and assess their rates of division [16] . To verify that H2BGFP is diluted with cell division and distributed equally between daughter cells , we characterized the tetO-H2BGFP system in vitro . Rosa26 and CAGGs ( constitutive promoters containing the CMV enhancer and the chicken β-actin promoter ) are commonly used in mouse embryonic fibroblasts ( mEFs ) and mouse embryonic stem ( mES ) cells . We used these promoters to drive expression of the reverse tetracycline transactivator ( rtTA ) in the presence of doxycycline . Rosa26-rtTA; tetO-H2BGFP mEFs and CAGGs-rtTA; tetO-H2BGFP mES cells express H2BGFP within 12 h of doxycycline application ( Figure 2A and 2C ) . Doxycycline was removed from the media , and the progressive dilution of H2BGFP protein resulting from cell division was measured by fluorescence-activated cell sorter ( FACS; Figure 2B and data not shown ) . A uniform loss of label was observed and the median GFP intensity of GFP-positive cells decreased with time . Given that mEFs divide every 24 h ( unpublished data ) , and that H2BGFP fluorescence can no longer be detected after 5 d , H2BGFP fluorescence is no longer detectable above background in vitro by FACS after a population has undergone approximately five cell divisions ( Figure 2B ) . The standard deviation of fluorescent intensity within the labeled pulse population is too large to precisely measure the number of cell divisions within the chase population . To verify that H2BGFP is segregated equally between daughter cells , CAGGs-rtTA: tetO-H2BGFP mES cells were cultured on the stage of a confocal microscope and imaged every 12 min . After the first division , total GFP fluorescence in the two daughter cells , measured as integrated pixel inten-sity , added up to the fluorescence of the original cell , and H2BGFP was split equally between daughter cells in the first and second divisions ( Figure 2D ) . In both dividing and nondividing cells , minimal bleaching was observed , despite imaging every 12 min over 18 h ( Figure 2D and unpublished data ) . Detection of H2BGFP fluorescence is dependent on the laser settings used . In this case , H2BGFP fluorescence was no longer detectable after three rounds of cell replication , though lower laser intensities were used than typically employed for fixed tissue sections . Because β cells divide slowly , and some may be postmitotic , it was important to determine whether all cells , regardless of replicative activity , can be labeled by the inducible H2BGFP system . We found that H2BGFP labeling occurs independent of cell division in cultured Rosa26-rtTA; tetO-H2BGFP mEFs ( Figure S1 ) . Cells treated with mitomycin C are irreversibly arrested in S phase , but still express H2BGFP , which incorporates into the nucleus within 12 h of the administration of doxycycline . Furthermore , cells reversibly arrested with either aphidicolin ( G0/G1 block ) or nocadozole ( G2/M block ) express H2BGFP upon treatment with doxycycline . At 3 h after release from nocadozole , cells develop labeled mitotic spindles , indicating that the H2BGFP has been integrated into chromatin and not just added to a nuclear histone pool ( unpublished data ) . The Rosa26 locus is active in most cells of the mouse [18] , suggesting that in the presence of doxycycline , Rosa26-rtTA should drive tetO-H2BGFP expression and label most mouse cells . Rosa26-rtTA; tetO-H2BGFP labeled diverse cell types , including but not limited to pancreas , intestine , fat , bone marrow , muscle , skin , and retina ( Figures 3 , 4 , and unpublished data ) , but not cortical neurons , olfactory bulb , or spinal cord , possibly due to the inability of doxycycline to cross the blood–brain barrier ( unpublished data ) . Under repressive conditions ( without doxycycline ) , no expression was observed in these organs ( n = 4; Figures 3 and 4 ) . The stability of H2BGFP can be most easily assessed in postmitotic cells , where any loss of fluorescence with time can only be explained by degradation of the H2BGFP protein . Mammalian photoreceptor cells are postmitotic and are not replaced over the lifespan of the animal; they are identified by their position within the outer nuclear layer of the retina and by expression of the calcium-binding protein recoverin [19] . Photoreceptor cells can be labeled by Rosa26-rtTA; tetO-H2BGFP ( Figure 4A ) . Following a chase of 6 mo ( Figure 3C ) , H2BGFP was detected in whole eyes and sectioned retinas at the same imaging settings used to collect pulse data . Staining with recoverin verified that label retention is restricted to the photoreceptor layer of the retina ( Figure 4A ) . Thus , the H2BGFP label is stable and retained in postmitotic cells . It is formally possible that the H2BGFP protein has a shorter half-life in pancreatic β cells than in postmitotic photoreceptor cells . As a further validation that the tetO-H2BGFP system can identify heterogeneity in cell populations , we confirmed that we could detect nonuniform loss of H2BGFP in tissues where slow-cycling cells are known to exist . Stem cells are often proliferatively quiescent compared with neighboring transit-amplifying cells . Because of their slow-dividing nature , stem cells tend to remain labeled in experiments that use DNA synthesis labels such as tritiated thymidine or bromodeoxyuridine . Label-retaining cells ( LRCs ) in Rosa26-rtTA; tetO-H2BGFP mice are clearly visible in the hair follicle bulge cells after a 2 mo chase , as previously shown by Tumbar et al . [16] ( Figure 4B ) . Furthermore , GFP-positive intestinal crypt cells can be identified up to 1 mo following the pulse , though the vast majority of the labeled cells in chase intestines seem to be slowly dividing mesenchymal cells and intestinal neurons ( Figure 4B ) . Finally , Linlowkit+sca+ sorted hematopoietic stem cells within the bone marrow also retain GFP fluorescence to a much greater extent than whole bone marrow ( unpublished data ) . Because pulse–chase experiments using tetO-H2BGFP can identify tissue heterogeneity in the form of slowly dividing stem cell populations , it should also be able to identify a subpopulation of slowly dividing β cells , should it exist . To determine whether all β cells divide at the same rate , we used the tetO-H2BGFP strategy in combination with promoters that express either tetracycline transactivator ( tTA ) or rtTA within the pancreas . Pdx1 expression in the postnatal pancreas is enriched in β cells , where it regulates insulin expression [20] . In tetO-H2BGFP animals , expression of tTA from the native Pdx1 locus ( Pdx1-tTA ) should label β cells in the absence of doxycycline , while the transgenic rat insulin promoter ( RIP ) –rtTA should label β cells in the presence of doxycycline . However , RIP-rtTA , but not Pdx1-tTA , labeled β cells even in unpulsed animals ( unpublished data and Figure 3B ) , indicating that the RIP-rtTA transgene system is leaky . Therefore , Pdx1-tTA is the only β cell–specific driver suitable for these experiments . It should be noted that Pdx1-tTA animals are haploinsufficient; however , mice with one inactivated Pdx1 allele can be maintained in the heterozygous state and have normal pancreatic development and β cell maintenance , though they show modestly impaired glucose tolerance [20] . Pulsing Pdx1-tTA; tetO-H2BGFP animals for 6 wk after birth labeled 80% of β cells ( Figure 3B ) . In addition to labeling β cells , some labeled nuclei occurred outside the β cell pool ( Figure S2 ) . Somatostatin and pancreatic polypeptide–expressing cells were frequently labeled in Pdx1-tTA; tetO-H2BGFP animals , consistent with the fact that Pdx1 was originally cloned from somatostatin-producing islet cell lines [21 , 22] , and that Pdx1 expression can be detected in these cells [23] . Rare glucagon cells were labeled , while exocrine cells ( amylase-positive ) were consistently weakly labeled , and no H2BGFP expression was observed in the ducts ( CK19-positive; Figure S2 ) . These observations are also consistent with the findings of Oster et al . , who observed rare Pdx1-immunoreactive nuclei in all pancreatic cell types [23] . The entire β cell pool was assayed for LRCs . A group of 38 Pdx1-tTA; tetO-H2BGFP animals ( 19 female , 19 male ) were pulsed from birth until 6 wk of age by ceasing doxycycline administration at postpartum day 0 ( P0; Figure 3C ) . Six of these animals ( three male , three female ) were euthanized following a 6-wk postnatal pulse . All animals collected in the pulse group showed 80% of β cells labeled with H2BGFP; labeling was consistent throughout the pancreas and between animals ( n = 6 ) . The remaining mice were again administered doxycycline water to repress transcription of H2BGFP , and were euthanized after chase periods of 1 wk ( n = 4 ) , 2 wk ( n = 4 ) , 1 mo ( n = 6 ) , 2 mo ( n = 6 ) , 3 mo ( n = 6 ) , and 6 mo ( n = 6 ) . Sections of the pancreati were stained with insulin to identify β cells . We observed a uniform loss of label in β cells with time ( Figure 5A ) . To confirm this finding , the experiment was repeated with Rosa26-rtTA; tetO-H2BGFP animals . A group of 40 animals ( 18 female , 22 male ) were pulsed from conception until 6 wk of age by administration of doxycycline water . Eight animals ( four male , four female ) were euthanized at 6 wk of age . Pulse expression in these mice was more variable between animals than for Pdx1-tTA; tetO-H2BGFP mice , so all experiments were performed on sibling cohorts . The remaining mice were removed from doxycycline and were euthanized after chase periods of 1 wk ( n = 4 ) , 2 wk ( n = 4 ) , 1 mo ( n = 6 ) , 2 mo ( n = 6 ) , 3 mo ( n = 6 ) , and 6 mo ( n = 6 ) . Sections of the pancreati were stained with insulin to identify β cells . Again , we observed uniform loss of label in β cells with time ( Figure 5A ) . To measure the relative intensity of GFP-positive cells , we used FACS analysis of dissociated islet cells . For these experiments , littermates were pulsed for at least 6 wk , and chases were structured so that all animals could be euthanized and analyzed by FACS on the same day . Importantly , comparisons of animals pulsed for 6 or 14 wk showed no significant increase in the median intensity of the GFP-positive population ( unpublished data ) . By analyzing all timepoints in parallel , we were able to compare the relative GFP intensity between the pulse and chase populations . This experiment was repeated four times using Rosa26-rtTA; tetO-H2BGFP mice and twice using Pdx1-tTA; tetO-H2BGFP mice; the results were consistent each time . Exposure-matched photographs of whole islets taken prior to dissociation ( Figure 5B ) and FACS plots of dissociated islets ( Figure 5C ) indicate that the median intensity of the GFP fluorescence within the β cell pool decreases with time . No outlying population of LRCs in the β cell pool can be identified . Hepatocytes , like β cells , are an endodermal cell type , and are thought to be maintained by self-replication [24] . It is unknown , however , whether all hepatocytes contribute equally to growth and maintenance . We examined livers from Rosa26-rtTA; tetO-H2BGFP pulse–chase animals and observed a uniform loss of label with time ( Figure S3 ) . Similar to the results with β cells , no outlying population of LRCs in the hepatocyte pool was identified . To directly compare the replication capacity of individual β cells , a lineage-based clonal analysis in the pancreas was performed . RIP-CreER transgenic mice [12] drove expression of tamoxifen-dependent Cre recombinase specifically in β cells , while the MADM reporter system [17] was used to label individual β cells . The MADM system is a unique tool that allows low-frequency labeling of cells , a prerequisite for clonal analysis ( Figure 6A ) . It contains two alleles at the Rosa26 locus , Rosa26GR and Rosa26RG , each containing reciprocal parts of chimeric marker genes ( GFP and RFP ) interrupted by a loxP site . Neither allele generates an active fluorescent protein until Cre-mediated interchromosomal recombination restores functional expression of GFP and RFP . Cells are labeled differently depending on when recombination occurs during the cell cycle . Recombination at G0 or G1 results in double-colored cells ( expressing both GFP and RFP ) . Alternately , recombination at G2 results in two outcomes at equal frequency: either one red and one green daughter cell ( single-colored cells ) , or one colorless and one double-colored daughter cell . As expected , no labeling was observed in RIP-CreER; Rosa26GR/Rosa26RG mice in the absence of tamoxifen . We injected tamoxifen over 3 d into 23 RIP-CreER; Rosa26GR/Rosa26RG mice between 4 and 8 wk of age . Two animals were euthanized within 4 d of tamoxifen injection ( the pulse group ) , and we found that 0 . 1%–0 . 5% of β cells were labeled . All clones observed were insulin positive , and 90% were single-cell clones ( the remainder were composed of two cells ) . In pulse and chase animals , all labeled cells expressed both GFP and RFP ( RFP expression required antibody staining for detection; Figure S4A ) . The existence of only double-colored clones indicates that recombination occurred during G1 or G0 , which is expected for a slow-dividing cell population that spends little time in G2 . Conversely , in experiments performed using Pdx1-Cre; Rosa26GR/Rosa26RG mice , where Cre recombinase was expressed in the rapidly dividing embryonic pancreas , single-colored clones were detected occasionally ( Figure S4B ) . These observations are consistent with those of Zong et al . [17] , who found that single-colored cells are a minority , though their proportion to double-colored cells increases when Cre recombinase is expressed in a rapidly dividing population . Given that all β cell clones are double-colored , all clones are shown only in green to allow for ease of presentation , and insulin staining is shown in red . For the purpose of this experiment , clones are defined as clusters of labeled cells within a single islet; all cells within a clone are assumed to be derived from a single β cell . Tamoxifen-treated mice were euthanized 1 mo ( n = 12 ) or 2 mo ( n = 9 ) following the pulse . A total of 175 clones from 1-mo chase animals and 122 clones from 2-mo chase animals were sampled using single random sections , and no large clones were detected . The absence of large clones qualitatively suggests an absence of fast-dividing β cells . To estimate the expansion of β cells , we analyzed serial sections . After 1 mo of chase , the average size of a clone was 5 . 1 ± 5 . 4 cells ( n = 45 clones ) , whereas after 2 mo of chase , the average size of a clone was 8 . 2 ± 6 . 9 cells ( n = 40 clones ) . In addition , the experiment was repeated using CAGGs-CreER; Rosa26GR/Rosa26RG mice , and again , no large β cell clones were detected ( unpublished data ) . Labeled cells found within a single islet are of clonal origin: the probability that three or more labeled cells found within the same islet are not clonal is 1 . 8 × 10−2 ( based on Poisson distribution analysis assuming labeling frequency of 0 . 1% and islet size of 500–1 , 000 cells ) . This clonal analysis supports the model that the growth and maintenance of β cell mass in the adult pancreas is achieved by the replication of individual β cells that have similar replicative capacities .
Our experiments were designed to address the mechanism of growth and maintenance of mature β cells . To determine whether all β cells divide at the same rate in the adult mouse , two experiments were undertaken . The tetracycline-inducible H2BGFP and MADM systems are complementary approaches: whereas tetO-H2BGFP labels most β cells and provides a broad view of the population dynamic , MADM labels single cells and provides an accurate clonal analysis of the progeny of individual cells within the β cell pool . Both the uniform loss of the H2BGFP label with time in the β cell population and the comparable β cell clone sizes generated through MADM analysis indicate homogeneity exists within the β cell pool . Stated otherwise , all β cells appear to contribute equally to growth and maintenance . The β cell mass is dynamic and can respond to environmental cues such insulin and glucose [25] . The β cell number increases dramatically in the first year of rodent life [12 , 26] , up to 10-fold in cases of insulin resistance [27] , and up to 1 . 5-fold during pregnancy [28 , 29] . Recent experiments suggest that when not hindered by persistent autoimmune attack or the toxicity of high blood glucose levels [30] , β cells have the capacity to regenerate . While the mechanism regulating β cell expansion remains unclear , our findings indicate that all β cells are capable of replication and are therefore viable targets for in vitro or in vivo expansion . Seaberg et al . recently reported that single-cell clones derived from adult islets generated colonies of 2 , 000–10 , 000 cells that expressed markers of neural , glial , pancreatic endocrine , exocrine , and duct identities [31] . These clones were generated from ~0 . 02% of islet cells , though their identity and relationship to in vivo growth is yet to be determined . We cannot rule out the possibility that a rare type of β cell was missed in our examination of individual clones using the MADM marking experiments . However , because the rate of clonal expansion is sufficient to account for the growth of the β cell population during the chase period , a rare highly proliferative β cell did not contribute significantly to the expansion of β cell mass . Published rates for β cell replication in adult mice ( 12 wk old ) are highly variable , from 2% [32] to 15% per day [33] . Assuming 5% of β cells replicate per day , and that all β cells are equivalent , β cells should divide approximately every 20 d . This would dilute the H2BGFP label beyond detection ( by completing up to five rounds of replication ) within 100 d . In addition , clone size at 2 mo should be approximately eight cells . These straightforward calculations predict results that are entirely consistent with our findings . These estimates , of course , assume no β cell death over the duration of our experiments . β cells have a finite lifespan , but the absolute β cell death rate is unknown . Based on β cell mass measurements and an estimate of β cell proliferation of 2% per day throughout adulthood , Finegood et al . calculated the β cell lifespan to be 52 d [32] . Recent findings demonstrate that β cell proliferation rates decline to less than 0 . 1% in 1-y-old mice [33] , casting doubt on the often quoted rates for β cell turnover in mice . Furthermore , TUNEL analysis of wild-type β cells consistently fail to identify apoptotic cells [13 , 27 , 33 , 34] . Regardless of the true rate of β cell turnover , our findings of a uniform loss of label and a consistent clone size indicate that all β cells have equivalent replicative capacity . Pancreatic β cells are not the only differentiated cell type capable of growth and maintenance without the support of an adult stem cell population . Hepatocytes are highly replicative and not thought to be supported by a facultative stem cell under normal conditions [35] . Pulse–chase analysis with the tetracycline-inducible H2BGFP label shows that all hepatocytes lose their label at the same rate . Therefore , like the β cell population , the hepatocyte population seems to be homogeneous . We do not know of an example of a mature differentiated cell type that has two populations ( one replicative and the other not ) . We speculate that when tissues are without an adult stem cell , they are replenished by equal replication of all differentiated cells . The demonstration that all β cells are equivalent , contributing equally to the growth and maintenance of the β cell population , has clinical implications if we assume that rodents and man use the same mechanism for pancreatic homeostasis . The destruction of β cells that causes type I diabetes has been counteracted by the transplantation of β cells . The clinical impact of this approach is currently limited , in part , by the scarcity of available pancreatic tissue [36] . A better understanding of adult β cell replication may help attempts to expand pancreatic β cells in vitro as a source of transplant material to treat diabetes .
Pdx1-tTA , Rosa26-rtTA , tetO-H2BGFP , and MADM mice were generously provided by Ray MacDonald ( University of Texas Southwestern Medical Center , Dallas , Texas , United States ) , Rudolf Jaenisch ( Massachusetts Institute of Technology , Cambridge , Massachusetts , United States ) , Elaine Fuchs ( Howard Hughes Medical Institute and Rockefeller University , New York , New York , United States ) , and Liqun Luo ( Howard Hughes Medical Institute and Stanford University , Stanford , California , United States ) , respectively . Mice were maintained at a barrier facility in the Department of Molecular and Cellular Biology at Harvard University under animal protocol 93–15 . Pdx1-tTA , Rosa26-rtTA , and tetO-H2BGFP mice were backcrossed to >95% C57BL/6 inbred background; RIP-CreER and MADM mice were maintained on a mixed background . Greater variation of H2BGFP expression was observed in Rosa26-rtTA; tetO-H2BGFP animals than with the other drivers; experiments were conducted on littermates to reduce variability . Notably , the highest-expressing animals were often runted , indicating that either Rosa26-rtTA activity or H2BGFP expression in some cell types is harmful . Breeding to a C57BL/6 inbred background reduced variability to some extent . Genotyping was performed by adding a tail biopsy to 100 μl DirectPCR ( ViaGen , http://www . viagen . com ) with 30 μg proteinase K ( Roche , http://www . roche . com ) , incubating overnight at 55 °C and denaturing proteinase K for 20 min at 95 °C . PCR primers specific to tTA ( forward 5′-ctggtcgagctggacggcgacgta aac-3′ , reverse 5′-atgtgatcgcgcttctcgttgggg-3′ ) , Rosa26-rtTA ( A 5′-aaagtcgctctgagttgtTAt-3′ , B 5′-gcgaagagtttgtcctcaacc-3′ , C 5′-ggagcgggagaaatggatatg-3′ ) , GFP ( forward 5′-ctggtcgagctggacggcgacgtaaac-3′ , reverse 5′-atgtgatcgcgcttctcgttgggg-3′ ) , Cre ( forward 5′-tgccacgaccaagtgacagc-3′ , reverse 5′-ccaggttacggatatagttcatg-3′ ) , MADM wild-type ( forward 5′-ctctgctgcctcctggcttct-3′ , reverse 5′-cgaggcggatcacaagcaata-3′ ) , and MADM knockin alleles ( forward 5′-ctctgctgcctcctggcttct-3′ , reverse 5′-tcaatgggcgggggtcgtt-3′ ) amplified 600 bp , 300 bp , 600 bp , 600 bp , 330 bp , and 250 bp fragments , respectively . PCR conditions: 95 °C for 5 min , then 35 cycles of 95 °C for 30 sec , 55 °C for 30 sec , 72 °C for 60 sec , and finally 72 °C for 5 min . Doxycycline ( Sigma , http://www . sigmaaldrich . com ) was added to drinking water at 1 mg/ml and sweetened with sucrose ( 1% ) . Water bottles were changed weekly with freshly prepared solution . Tamoxifen ( Sigma ) was dissolved in corn oil at 20 mg/ml and mice were injected intraperitoneally ( 6 or 8 mg/d for 3 consecutive days ) . Tissue was dissected from mice , fixed in 4% paraformaldehyde/PBS solution for two hours at 4 °C , washed in PBS , incubated in 30% sucrose/PBS solution overnight , embedded in OCT ( Tissue-Tek; Electron Microscopy Sciences , http://www . emsdiasum . com ) and stored at −80 °C . Frozen samples were sectioned at 10 μm for staining or up to 50 μm for serial analysis . The following primary antibodies and dilutions were used: guinea pig anti-Pdx1 antibody ( kindly provided by C . Wright , Vanderbilt University , Nashville , Tennessee , United States ) , 1:1000; guinea pig anti-swine insulin ( DakoCytomation , http://www . dako . com ) , 1:200; guinea pig anti-glucagon antibody ( Linco , http://www . linco . com ) , 1:200; rabbit anti-human pancreatic polypeptide ( DakoCytomation ) 1:200; rabbit anti-human somatostatin ( DakoCytomation ) , 1:200; rabbit anti-amylase ( Sigma ) , 1:200; rabbit anti-CK19 ( Developmental Studies Hybridoma Bank , http://dshb . biology . uiowa . edu ) , 1:1 , 000; rabbit anti-GFP ( Molecular Probes , http://probes . invitrogen . com ) , 1:200; rabbit anti-DsRed ( Clontech , http://www . clontech . com ) , 1:100; and rabbit anti-recoverin ( Chemicon , http://www . chemicon . com ) , 1:2 , 000 . Secondary antibodies donkey rhodamine RedX anti–guinea pig ( Jackson ImmunoResearch , http://www . jacksonimmuno . com ) , donkey rhodamine RedX anti-rabbit ( Jackson ImmunoResearch ) , and donkey rhodamine RedX anti-goat ( Jackson ImmunoResearch ) were used at 1:200 dilution . To visualize nuclei , slides were stained with 0 . 5 μg/ml DAPI and then mounted with VectaShield Mounting Medium ( Vector Laboratories , http://www . vectorlabs . com ) . Triple-labeled GFP/rhodamine/DAPI images were acquired using a Zeiss LSM510 Meta confocal microscope ( http://www . zeiss . com ) . The pancreas was perfused through the bile duct with 5 ml digestion solution ( low-glucose DMEM [Gibco , http://www . invitrogen . com] with 10 mM HEPES [Gibco] , 0 . 25 mg/ml liberase RI [Roche] , and 0 . 1 mg/ml ovalbumin trypsin inhibitor [Roche] ) , dissected and incubated at 37 °C for 20 min . Cold washing solution ( low-glucose DMEM with 10 mM HEPES , 10% FBS [Hyclone , www . hyclone . com] , and 0 . 1 mg/ml ovalbumin trypsin inhibitor ) was added , and islets were centrifuged , washed twice , and filtered through a 500 μm diameter wire mesh . Islets were centrifuged , washed twice in washing solution , resuspended in Histopaque 1077 ( Sigma ) , and vortexed . The islet suspension was carefully overlaid with washing solution ( without serum ) and centrifuged for 20 min at 10 °C , separating islets from exocrine tissue . The islet layer was collected at the interface , pelleted , washed twice , and further purified by two rounds of gravity sedimentation . Finally , pure islets were handpicked under a dissecting scope . Islets were dissociated by incubation with 0 . 25% trypsin-EDTA ( Gibco ) at 37 °C for 5 min , washed , fixed for 15 min in 1% paraformaldehyde/PBS solution , resuspended in 5% donkey serum/PBS , and FACS sorted on a BD Aria ( BD Biosciences , http://www . bdbiosciences . com ) . Rosa26-rtTA; tetO-H2BGFP mEFs were obtained by collecting timed plugs and dissecting embryonic day ( E ) 12 . 5 embryos . Embryos were eviscerated , trypsinized , plated on gelatinized plates , and cultured in standard mEF media ( DMEM with 10% FBS and 1x penicillin/streptomycin [Gibco] ) . Cells were grown with 10 μg/ml doxycycline to induce H2BGFP transcription . CAGGs-rtTA; tetO-H2BGFP mES cells were obtained by electroporating a CAGGs-rtTA-IRES-puromycin plasmid into tetO-H2BGFP mES cells derived from blastocysts , selecting with 4 μg/ml puromycin in 10 μg/ml doxycycline and picking and expanding green colonies . mES cells were grown in standard mES conditions ( knockout DMEM [Gibco] with 15% defined FBS [Hyclone] , 200 mM L-glutamine [Gibco] , 10 mM nonessential amino acids [Gibco] , 1× penicillin/streptomycin , 0 . 001% β-mercaptoethanol [Sigma] , and 1 , 000 U/ml LIF [Chemicon] ) . Dividing cells were imaged by culturing on the stage of a Zeiss LSM confocal microscope , and 15-image z-stacks were scanned every 12 min at 0 . 5% laser intensity . Images were quantified using MetaMorph software ( Molecular Devices , http://www . moleculardevices . com ) , and values were recorded as integrated pixel intensity . Cell-cycle experiments were performed as follows: 10 μg/ml mitomycin C ( Sigma ) was applied to cells for 3 h to irreversibly inhibit the cell cycle in S phase , 0 . 25 ng/ml aphidicolin ( Sigma ) or 100 ng/ml nocadozole ( Sigma ) was applied to cells for 6 h to reversibly arrest the cell cycle at G0/G1 or G2/M , respectively .
|
The β cells of the pancreas are responsible for insulin production and their destruction results in type I diabetes . β cell maintenance , growth , and regenerative repair is thought to occur predominately , if not exclusively , through the replication of existing β cells , not via an adult stem cell . It was previously unknown , however , whether all β cells divide at the same rate , or if multiple subpopulations of β cells exist , some highly replicative and others very slowly dividing , possibly postmitotic . We performed two types of experiments to determine whether all β cells are alike: label-retaining analysis and clonal analysis . Our results indicate that all β cells contribute equally to islet growth and maintenance .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"cell",
"biology",
"physiology",
"in",
"vitro",
"diabetes",
"and",
"endocrinology",
"molecular",
"biology"
] |
2007
|
All β Cells Contribute Equally to Islet Growth and Maintenance
|
Evolutionary expansion of signaling pathway families often underlies the evolution of regulatory complexity . Expansion requires the acquisition of a novel homologous pathway and the diversification of pathway specificity . Acquisition can occur either vertically , by duplication , or through horizontal transfer , while divergence of specificity is thought to occur through a promiscuous protein intermediate . The way by which these mechanisms shape the evolution of rapidly diverging signaling families is unclear . Here , we examine this question using the highly diversified Rap-Phr cell–cell signaling system , which has undergone massive expansion in the genus Bacillus . To this end , genomic sequence analysis of >300 Bacilli genomes was combined with experimental analysis of the interaction of Rap receptors with Phr autoinducers and downstream targets . Rap-Phr expansion is shown to have occurred independently in multiple Bacillus lineages , with >80 different putative rap-phr alleles evolving in the Bacillius subtilis group alone . The specificity of many rap-phr alleles and the rapid gain and loss of Rap targets are experimentally demonstrated . Strikingly , both horizontal and vertical processes were shown to participate in this expansion , each with a distinct role . Horizontal gene transfer governs the acquisition of already diverged rap-phr alleles , while intralocus duplication and divergence of the phr gene create the promiscuous intermediate required for the divergence of Rap-Phr specificity . Our results suggest a novel role for transient gene duplication and divergence during evolutionary shifts in specificity .
The evolution of signaling complexity often occurs by diversification and repeated utilization of signal transduction pathways [1–6] . This generally requires two processes: the acquisition of homologous copies of the pathway's components and the co-diversification of interacting components to ensure specificity of interaction within a pathway while avoiding cross-talk between pathways ( Fig 1A ) [7 , 8] . Bacteria have a multitude of signal transduction pathways , which have undergone evolutionary expansion and divergence of specificity , such as two-component systems [6 , 8–10] , antisigma-sigma factors [11] , and toxin-antitoxin systems [12 , 13] . The large number of available bacterial genomes allows for high-resolution analysis of evolutionary expansion , rendering bacterial signal transduction a favorable model system for studying diversification . While eukaryotes can only acquire paralogous genes through duplications , bacteria can acquire them either by gene duplication ( Fig 1C , right ) or by horizontal transfer ( Fig 1C , left ) [14] . Previous works on the prevalence of these two processes in the acquisition of bacterial two-component signal transduction pathways have indicated that it is dominated by gene duplication , but it is also affected by horizontal transfer [15–17] . However , the coarse-grained resolution of these studies prevents the distinction between vertical acquisition and horizontal transfer between closely related strains [18 , 19] . The second requirement for paralogous expansion is the divergence of interaction specificity between pathways ( Fig 1B ) . This is generally thought to evolve using a promiscuous form of one of the interacting components , which can interact with both variants of its partner ( Fig 1B , top ) [2 , 4 , 12 , 20–22] . The promiscuous form can be the ancestral state , subsequently evolving into two states of different specificity [4 , 20] , or it can be an evolutionary intermediate between the ancestral specific state to a novel state [22 , 23] . The ability to distinguish between these two diversification scenarios typically depends on our capacity to infer and analyze the ancestral state from phylogenetic data [2 , 20 , 21] . A recent work used deep mutational scanning to show the abundance of promiscuous bacterial intermediates in the evolution of a bacterial toxin-antitoxin family [12] . This approach , however , cannot distinguish whether the promiscuous form is ancestral or intermediate or determine the evolutionary relevance of the identified diversifying trajectories . The modes by which rapidly diversifying signaling families expand are therefore still unclear . The Rap-Phr cell–cell signaling system of Bacilli can serve as a model system to study bacterial modes of expansion and diversification [24–26] . The cytoplasmic Rap receptor can bind , and sometimes dephosphorylate , its target , leading to inhibition of target activity [27 , 28] . The cognate phr gene codes for a pre-polypeptide , which undergoes multiple cleavage events during its secretion , resulting in the release of a mature penta- or hexa-peptide Phr autoinducer [25 , 29–31] . The mature Phr peptide is transported into the cytoplasm through the oligopeptide permease system , where it can interact with Rap receptors [26] , subsequently leading to major conformational changes in the Rap protein and preventing Rap from repressing its target [27 , 28 , 32 , 33] . Rap-Phr systems have mostly been studied in the B . subtilis 168 lab strain . This strain encodes for eight paralogous rap-phr loci , each coding for a different Phr autoinducer . In addition , it encodes for three orphan rap genes that lack a cognate phr locus [33–35] . Despite the genomic expansion of paralogous Rap-Phr systems , they all have the same overall structural organization and most have a redundant function in repressing either Spo0F or ComA , two key response regulators of the Bacillus stress response network [31 , 36–40] . We recently demonstrated how social selection can explain the acquisition of additional Rap-Phr systems , despite their redundant regulation of the same target [41] . Some rap-phr loci are encoded by mobile genetic elements [31 , 39 , 42–47] , and while many mobile-element-associated Rap systems maintain their repressive effect on Spo0F or ComA , some also play a direct role in controlling the mobility of their associated mobile genetic elements [44 , 47] . To study the expansion and diversification of the Rap-Phr family , we combined computational mining of available Bacillus genomes and experimental characterization of the target and autoinducer specificity of multiple Rap-Phr systems . We found that at the organismal level , acquisition of a novel Rap-Phr paralogous system occurred by horizontal gene transfer ( Fig 1C , left ) . At the locus level , diversification of Rap-Phr specificity was facilitated by phr gene duplication or intragenic duplication of the Phr autoinducer coding sequence , followed by diversification of the autoinducer sequence . We show that the diverged duplicated phr form can serve as a promiscuous intermediate between two states of specificity ( Fig 1B , bottom ) . Therefore , the extreme diversity of the Rap-Phr system results from a combination of horizontal and vertical processes operating at two different levels of genetic organization .
To understand the extent to which Rap-Phr systems are prevalent in the Bacillus genus , we downloaded 413 whole-genome sequences of strains from this genus from the National Center for Biotechnology Information ( NCBI ) ( S1 Data ) . The species association of these genomes is heavily biased towards genomes from the B . subtilis ( 127 strains ) and B . cereus ( 216 strains ) groups of species ( S1 Fig ) . The conserved Rap structure , which includes a three-helix N-terminal and tetra-tricopeptide repeat C-terminal domain [24 , 32 , 33] , allowed us to search for Rap homologs in all the genomes using the basic local alignment search tool for translated DNA ( BLAST tblastn program ) . Following identification of Rap homologs , we searched for candidate phr genes , relying on the known organization of annotated phr genes ( S2 Fig ) —short open reading frames with a secretion signal sequence , located immediately downstream of the Rap gene in the same direction ( see Methods ) [24] . Our final database contained ~2 , 700 functional rap genes ( S1 and S2 Data ) . rap genes were identified in all strains of the B . subtilis and B . cereus groups as well as in two evolutionarily distinct Bacillus species—B . halodurans and B . clausii ( S1 Fig ) . Notably , all strains harboring at least one rap gene encoded for multiple rap paralogs ( Fig 2A , S5 Data ) . The number of rap genes differed between groups , averaging 11 ± 2 ( mean ± standard deviation [st . dev . ] ) in the B . subtilis group and 6 ± 3 in the B . cereus group . In B . subtilis 168 , there are three orphan rap genes not accompanied by an adjoining phr gene , a scenario typical for B . subtilis strains , which have an average 2 . 7 ± 1 orphan raps per strain ( mean ± st . dev . ) . In contrast , almost all ( >95% ) of the rap genes from the B . cereus group had an adjoining putative phr gene . We next performed a phylogenetic analysis of the ensemble of rap genes ( see Methods for details of the phylogenetic analysis and S3 and S4 Data ) . Although the overall divergence of Rap homologs was large , the family was clearly divided into two groups , corresponding to the division between the B . cereus and B . subtilis groups of species ( Fig 2B ) . This suggests that the divergence of rap genes in each of these groups occurred after their evolutionary separation , with no horizontal transfer between groups . The phylogeny of the B . clausii and B . halodurans Rap proteins suggests that they acquired their rap genes by one or two horizontal gene transfer events , respectively , from B . subtilis group isolates , followed by intraspecific diversification and accumulation , as seen in the two major groups ( Fig 2B ) . No evidence of divergence of the Rap protein through recombination of different Rap homologs was found ( Methods ) . These observations suggest that diversification and paralogous expansion occurred independently multiple times during the evolution of the Rap-Phr system . To gain further insight into the population genetics underlying the diversification of the Rap-Phr family , we focused on the B . subtilis group , in which multiple Rap-Phr systems have been previously characterized . The known Phr autoinducer sequences and the patterns of phr sequence conservation along the Rap phylogenetic tree were used to identify putative penta or hexa-peptide autoinducers and to cluster the Rap proteins ( Fig 2C ) . All together , we defined 102 clusters with 81 unique Phr autoinducer peptides . To the best of our knowledge , this extreme autoinducer diversity is much greater than that observed in any other family of quorum-sensing systems . In order to identify the mode of acquisition of a novel Rap-Phr system into a genome , we analyzed the level of horizontal gene transfer of Rap-Phr systems . We used two independent measures to estimate this trait—guanine-cytosine ( GC ) -content analysis and abundance analysis . First , because mobile element-related genes in B . subtilis typically have a significantly lower GC-content as compared to the rest of the genome [48] , we characterized Rap-Phr as mobile if their GC-content was significantly lower than the average GC-content of their respective strain ( Methods , S3 Fig ) . We found that 75% of Rap-Phr clusters were mobile ( Fig 2C , bottom right ) . In parallel , mobile ( or accessory ) genes can be identified by their intermittent appearance within strains of a given species . Thus , we constructed an association matrix , in whcih each Rap cluster was marked as either present or absent in each of the genomes of B . subtilis group isolates ( S4 Fig ) . With few exceptions , raps identified as core genes by their GC-content appeared in the great majority of isolates from a given species , whereas mobile Rap systems appeared in only a few strains ( S5 Fig , S5 Data ) , demonstrating a good correlation between the two measures of mobility . These results suggest that the acquisition of a novel Rap-Phr into a genome is dominated by horizontal gene transfer . To determine whether duplication , as the alternative mode of acquisition , occurred as well , we searched for cases in which two Rap-Phr systems from the same cluster were coded in the same genome ( S4 Fig ) . Only five such cases were identified , three of which occurred in clusters that were categorized as mobile by both criteria above and may result from a recent duplicated introduction of a single mobile element . Therefore , there is little evidence for direct duplication events in Rap-Phr that belong to the “core” genome . We noted that different species in the B . subtilis group have different core Rap paralogs ( S4 Fig ) . The observed diversity pattern fits a slow ongoing process of fixation of Rap-Phr systems , in which some Rap variants ( e . g . , RapA and RapC ) are fixed in multiple related species , while others are fixed only in a single species within the group . Interestingly , all orphan Rap systems belonged to the core group , by both modes of assessment , with RapD present in all but one species ( B . licheniformis ) . Although mobile Rap-Phr systems dominated the population diversity , most Rap paralogs in any given strain ( 60% ± 10% mean ± st . dev . ) belonged to the core group ( Fig 2C , top right ) . Despite the abundance of available genomic data , we do not observe a saturation of Rap-Phr diversity with strain number ( S6 Fig ) . The large diversity is also evident from the fact that several rap-phr clusters were found only on separately sequenced plasmids [49–52] . Current data indicate that despite the large sequence and autoinducer diversity , many of the Rap receptors target either Spo0F , ComA , or both . The evolutionary rate of target specificity shift is unclear . To experimentally determine the specificity of multiple Rap systems , we used two isolates ( marked in S4 Fig ) , B . amyloliquefaciens FZB42 and B . licheniformis ATCC 14580 , as templates for the cloning of ten novel rap genes , most of them core genes of their respective species ( Fig 3 and S4 Fig , S1 Table ) . These newly cloned Rap genes belong to previously unexplored branches of the Rap phylogeny . Correspondingly , these Rap homologs almost double the phylogenetic diversity [53] of characterized Rap proteins ( from ~6% to ~10% of the total phylogenetic diversity of Rap proteins in the B . subtilis group ) . In addition , seven B . subtilis-associated rap genes ( rapC , F , I , P , J , B , and D ) were cloned under the control of the inducible hyper-spank promoter [54] . All genes were introduced into the genome of strain PY79 . To prevent interference by the endogenous Phr product , the rapC , F genes were introduced into a strain with a deletion of these two systems ( S7 Fig , S5 Data ) . Finally , an available phrA deletion was used to assay the effect of RapA . The different Rap proteins were assayed for their effects on the Spo0A and ComA pathways by measuring their impact on sporulation efficiency and expression of the ComA-regulated srfA promoter using a yellow fluorescent protein ( YFP ) reporter [43] , respectively ( Fig 3A , S5 Data , Methods ) . Because ComA activity is indirectly affected by the Spo0A pathway [56] , we introduced a spo0A deletion into each of the YFP-reporting strains . Five out of the ten novel Rap proteins affected both sporulation efficiency and srfA expression , while RapBL4 affected only sporulation . Four of the novel rap overexpression constructs did not strongly affect either pathway . The eight B . subtilis-associated Rap proteins had the expected , previously characterized effect , with RapA , I , B , and J affecting Spo0A [26 , 33] , RapF , C , and D affecting ComA [34 , 37] , and RapP affecting both pathways [43] . These data allow us to better estimate the rate at which target choices change along the evolution of the Rap lineage . Based on our results and those reported by others , we assembled a phylogenetic tree of 25 Rap variants whose targets have been at least partially characterized ( Fig 3B ) . We used the GLOOME program [55] to estimate the rate of gains or losses of regulation of the Spo0A and ComA pathways . We found the most parsimonious switching model to include 12 gain and loss events ( Fig 3B ) , starting with an ancestral strain that regulated spo0A activity . This ancestral strain acquired the ability to control ComA in multiple independent events ( see Discussion ) . The interactions between Rap proteins with the aforementioned targets were recently analyzed at the structural level , allowing for the identification of specific Rap residues that directly interact with each target [27 , 28] . Upon analysis of the conservation of these residues in the characterized Rap proteins according to their functional targets ( S8 Fig ) , we found that the amino acid residues , where RapF interacts with its target ComA , were not conserved in many other ComA-interacting Rap proteins . In contrast , the RapH amino acid residues at the interface with Spo0F were highly conserved in characterized Rap proteins , irrespective of whether they regulate the Spo0A pathway or not . The high level of conservation of Spo0F-interacting residues and low level of conservation of ComA-interacting residues were also demonstrated upon analysis of all B . subtilis group-associated Rap proteins using the ConSurf program ( S9 Fig ) [57] . These results further support the ancestral origin of the interaction between Rap proteins and Spo0F and the independent gain of ComA interaction by multiple sub-lineages of Rap proteins , as suggested by the parsimony analysis ( Fig 3B ) . Anecdotal experiments in strain 168 have indicated that divergent Rap-Phr pairs are orthogonal—a receptor from one Rap-Phr strain will predominantly respond only to its cognate autoinducer [33 , 35 , 38 , 41] . This notion has not been studied systematically . It is also unclear whether divergent Rap-Phr systems encoded on different chromosomes would maintain orthogonality . We took advantage of the large collection of inducible Rap systems to thoroughly analyze these points . Fourteen custom-made putative autoinducer peptides were assayed for their ability to restore gene expression in the presence of different inducible Raps ( Fig 3C ) . We used a peptide concentration of 10 μM , a level that exceeds both the measured affinity of Phr peptides to cognate Raps and the physiological levels of Phr ( Methods ) [25 , 30 , 43] . The interactions between peptides and Rap proteins were monitored using either the PsrfA-YFP or PspoIIG-YFP reporter constructs , depending on whether the Rap targets the ComA or Spo0A pathways , respectively ( Fig 3C , S5 Data ) . Rap proteins that affect both pathways were assayed only once . We found that the repressive effect of all Rap proteins on gene expression was alleviated by addition of saturating amount of their respective cognate Phr peptide . One exception to this rule was RapBL4 , which did not interact with its putative Phr pentapeptide ( with amino-acid sequence GRAIF ) . We also found that the orphan RapB , J proteins were not affected by any Phr but that a 10-fold higher concentration of PhrC did activate them , in accordance with previous works that suggested this weak interaction ( Fig 3C ) [33 , 35] . We observed strict maintenance of orthogonality of Rap-Phr systems residing on the same genome . In two cases , cross-talk between two systems encoded by different strains was detected , with RapBL5 responding to both its cognate PhrBL5 autoinducer and to the related PhrBA1 autoinducer , and RapBA2 responding to its cognate PhrBA2 and more weakly to PhrBL6 . Notably , the RapBL5 and RapBA1 proteins were only weakly divergent ( Fig 3B ) , but RapBA2 and RapBL6 were unrelated . Our data therefore support the notion of strong orthogonality of Rap-Phr systems encoded in the same genome and some orthogonality between divergent Rap-Phr systems encoded by different genomes . Thus far , our results suggest that the Phr peptides coevolve with their cognate Rap receptors to maintain the specificity of interaction , while Rap receptor affinity to its main two targets can change . To understand the evolution of the Phr peptides , we studied the diversity of the phr sequences in further detail . All Rap-Phr systems analyzed to date have a single phr gene , coding for a single Phr autoinducer penta- or hexa-peptide . In contrast , we found multiple cases in our database where peptide autoinducer coding regions were duplicated ( Fig 4 , S10 Fig ) . In some cases , all putative autoinducer repeats were identical ( Fig 4A and S10E Fig ) , while in others , a single phr gene coded for multiple similar , but nonidentical , putative peptide autoinducers ( Fig 4B and S10A–S10D Fig ) . For example , the Phr prepeptides of a group of closely homologous Rap proteins ( related to RapH ) contain multiple varying repeats of the motif [S/I][D/I/N/Y]RNT[T/I] ( S10B Fig ) . We also identified two subclusters of the B . subtilis Rap-Phr systems , in which the entire phr gene had undergone a duplication event ( S10C and S10D Fig ) . The putative peptide autoinducers of the two phr genes had also diverged . We also observed sequence duplications ( either intragenic or full-gene ) events in rap-phr loci of other Bacilli ( S11A–S11C Fig ) . A similar analysis of the related NprR-NprX quorum-sensing family [58 , 59] showed duplications in some nprX genes as well ( S11D Fig , Methods ) . These results indicate that autoinducer duplications are abundant and that putative autoinducer sequences diverge after duplication . Autoinducer duplication may facilitate the coevolution of Rap-Phr pairs by allowing a duplicated and diverged phr gene to serve as a transient promiscuous intermediate ( Fig 1B ) . To experimentally examine this possibility , we analyzed a subset of Rap-Phr systems of the RapK/RapG/RapBL3 cluster ( Fig 4B ) [37] . The phr gene of three closely related systems in this cluster encodes for a pre-peptide with two putative autoinducer peptides . One system , designated RapK2-RR , encodes twice for the putative penta- or hexapeptide ERPVG ( T ) . The second system ( RapK2-KR ) encodes for the putative autoinducers ERPVG ( T ) and EKPVG ( T ) , while the third ( RapK2-KK ) encodes twice for the putative autoinducer EKPVG ( T ) . RapK2 variants ( S1 Table ) were cloned under the control of the hyper-spank promoter and monitored for the effect of their overexpression on a PspoIIG-YFP reporter , as described above ( Fig 4C , S5 Data ) . spoIIG promoter activity was repressed to background levels in all three overexpression strains . spoIIG promoter activity was restored in strains overexpressing either RapK2-RR or RapK2-KK upon addition of their cognate hexapeptides ( ERPVGT and EKPVGT , respectively ) , but not when their non-cognate hexapeptide was added ( Fig 4C ) . RapK2-KR , whose cognate Phr encodes both type of peptides , responded only to the addition of ERPVGT . None of the strains responded to the addition of the relevant pentapeptides . The specificity shift between the different RapK2 variants may therefore provide an example for the role of duplication in such an event .
The diversity and functional orthogonality of diverging signal transduction paralogs in bacteria are well characterized [9 , 11 , 12] . However , the way by which new paralogs are acquired and diversified and , specifically , the roles of horizontal and vertical events in this process are not clear . In this work , we showed that both horizontal and vertical processes are crucial for the expansion of the Rap-Phr quorum-sensing system but operate at different levels of organization . Acquisition of a novel Rap-Phr system was shown to be facilitated by horizontal gene transfer , while diversification was facilitated by phr duplications within the diversifying locus ( Fig 5 ) . A key novelty of our finding is exposure of the role of duplications in the divergence of Rap-Phr system specificity . The prevalence of intragenic and whole-gene phr duplications and the experimental analysis of the RapK2-PhrK2 variants ( Fig 4 and S10 and S11 Figs ) suggest that transient Phr duplication and divergence play a role during evolutionary shifts in Rap-Phr specificity ( Fig 4D ) . The ancestral copy of Phr interacts with the ancestral Rap form , while the duplicated and diverged autoinducer copy has the potential to interact with a coevolved receptor . Importantly , this mechanism differs in two aspects from the common views of pathway diversification . First , divergence of a signaling pathway is typically linked with a promiscuous protein that can interact with the two forms of its partner [2 , 12] . Here , the promiscuous form is the duplicated phr and not a single Phr peptide , which interacts with both Rap variants . Second , duplications are typically only considered important if both diverged duplicates survive over evolutionary timescales [3] . In contrast , Phr duplication and divergence are evolutionarily crucial for a specificity shift , but to complete the shift it has to be transient—with either divergence or duplication itself being lost . In the specific case we examined , the duplication persisted , while duplicate diversity was transient ( Fig 4D ) . We found that Rap acquisition and divergence occurred independently in multiple evolutionary lineages , indicating that it is an intrinsic feature of the function of this system . In the B . subtilis group alone , we identified dozens of putative Phr autoinducer peptides , rendering it the largest known quorum-sensing family . While the high orthogonality of a significant number of pairs was experimentally verified ( Fig 3B ) , further experimental work will be required to explore the level of orthogonality between all clusters . In fact , some cases of cross-interactions were detected ( Fig 3C ) . Notably , strong cross-interactions between Rap-Phr pairs encoded in the same genome were not observed , while weak interactions , as seen between PhrC and the orphans RapB and J ( Fig 3C ) [33 , 35] , or RapF [41] , were noted . In general , the functional importance of nonspecific interactions is unclear . More specifically , the interaction between orphan Raps and non-cognate Phrs may be physiologically irrelevant , given the low affinity of these interactions ( ~100 μM ) compared with the physiological concentration of Phrs ( ~100 nM ) [25] . Whether orphan Raps interact with other signaling molecules remains to be determined . Notably , formation of orphan Raps is rare , and all major B . subtilis orphan Raps are anciently fixed in their genome ( Fig 2C and S4 Fig ) . Rap proteins also diverge with regards to the targets they regulate . Our data indicate that the distinction between targets is not the result of ancient diversification but rather is the result of an ongoing process of multiple events of gain and loss of target regulation ( Fig 3A and 3B ) . The GLOOME parsimony analysis indicated that the ancestral Rap receptor regulated the Spo0A pathway and that the regulation of ComA by Rap proteins has been gained multiple times within the B . subtilis group . This is in agreement with the low sequence conservation between different ComA-regulating Rap variants ( S7 and S8 Figs ) and the absence of ComA homologs in the majority of B . cereus strains . In addition , the mechanism of ComA regulation differs across Raps . RapF blocks comA binding to DNA through direct competition for the DNA binding domain [27] . In contrast , Rap60 does not block ComA DNA binding but prevents the DNA-bound ComA from activating transcription [46] . Phr diversification is accompanied by coevolution of the Rap receptor . A key future challenge is to identify the amino-acid positions of Rap that are crucial for its coevolution and the underlying structural principles of peptide-receptor specificity . Our results suggest that even key conserved features of this interaction can be lost during diversification . Specifically , arginine at the second position of the Phr autoinducer is highly conserved due to a salt-bridge with a conserved negatively charged residue on the Rap protein ( Fig 2 ) [32 , 33] . While substitution to lysine ( as in PhrG or PhrK2-KK ) does not dramatically interfere with this interaction , this residue is substituted with the uncharged leucine in PhrBL3 . Correspondingly , the conserved Rap aspartate residue is substituted by the non-charged glutamine residue in RapBL3 , suggesting that the electrostatic interaction has been replaced by another type . The mobile nature of the majority of Rap-Phr systems ( Fig 2C and S3 and S4 Figs ) and their functional adaptive role in the transfer of mobile elements [47 , 52] and in social interactions [41] indicate that they act as “kind-discrimination” systems [62] . Such systems mediate discriminative interactions between bacteria or between their genetic parasites . Kind-discrimination can operate through various mechanisms such as cell–cell signaling [60] , intracellular toxin-antitoxin [63] , aggregation [64 , 65] , surface exclusion [66] , bacteriocin-immunity [67] , or contact-mediated toxins [68 , 69] . Most kind-discrimination systems are two-gene systems with a high divergence of specificity between interacting pairs . Like the Rap-Phr system , some of these systems tend to accumulate in large numbers within bacteria [70] . Notably , the social nature of kind-discrimination implies that alleles may strongly interact even if they are not encoded in the same bacterium . This presumably increases selection pressure for specificity ( Fig 5 ) . Intralocus duplication events can yield a transient promiscuous intermediate in other kind-discrimination systems . A recent analysis showed that toxin-antitoxin systems can shift specificity through promiscuous toxin intermediates , which can mediate interactions with two different antitoxins [12] . However , this does not rule out the possibility of duplication as an alternative mechanism . Further analysis of natural variation will be required to further assess these phenomena . One difference between Rap-Phr and other kind-discrimination systems is the short length of both the phr gene and the mature autoinducer peptide . This may promote duplication and neofunctionalization in peptide-based quorum-sensing systems in comparison to other systems where both interacting partners are larger globular proteins . Interestingly , fungi mating pheromones show a striking similarity to the Phr diversity , with both genic and intragenic duplications of the pheromone peptide coding sequences as well as some cases where there is sequence variability between duplicates [71] . Duplications may arise by sexual selection [72] , and it has been suggested , but not proven , that they may facilitate diversification . Taken together , the data from this work and previous works [41 , 47] show how duplication and rapid horizontal transfer can work together to rapidly expand a signaling pathway family operating at multiple levels of selection . Further work will be needed to determine the generality of this phenomenon .
|
Many molecular pathways are found multiple times in a given organism , where they are often reutilized for different functions . Such expansion of a family of pathways requires two main evolutionary processes—acquisition of additional copies of the pathway's genes and divergence of interaction specificity to prevent cross-talk between pathways while preserving interactions within each copy of the pathway . In bacteria , acquisition can occur horizontally , by transfer between different lineages , or vertically , by duplication within the lineage . Interaction specificity is thought to diverge through a promiscuous intermediate component that prevents loss of interaction during the process . In this work , we study the mechanisms underlying the extreme expansion of the Rap-Phr cell–cell signaling family in the Bacillus genus . Specificity of Rap-Phr interaction is critical for guiding preferential action towards kin . We find that horizontal transfer and not duplication guides the acquisition of an already divergent Rap-Phr variant . Surprisingly , duplication still has a key role during expansion , as duplication and subsequent divergence of the signaling molecule gene provide the promiscuous intermediate state needed for divergence of specificity . We therefore identify two complementary roles for horizontal and vertical processes in the evolution of social bacterial pathways .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion"
] |
[
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"bacillus",
"microbiology",
"prokaryotic",
"models",
"phylogenetics",
"data",
"management",
"phylogenetic",
"analysis",
"experimental",
"organism",
"systems",
"genome",
"analysis",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"bacillus",
"cereus",
"genomics",
"sequence",
"alignment",
"bioinformatics",
"medical",
"microbiology",
"microbial",
"pathogens",
"biological",
"databases",
"evolutionary",
"systematics",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"sequence",
"databases",
"database",
"and",
"informatics",
"methods",
"bacillus",
"subtilis",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"genomic",
"databases",
"organisms"
] |
2016
|
Transient Duplication-Dependent Divergence and Horizontal Transfer Underlie the Evolutionary Dynamics of Bacterial Cell–Cell Signaling
|
Small RNAs are functional molecules that modulate mRNA transcripts and have been implicated in the aetiology of several common diseases . However , little is known about the extent of their variability within the human population . Here , we characterise the extent , causes , and effects of naturally occurring variation in expression and sequence of small RNAs from adipose tissue in relation to genotype , gene expression , and metabolic traits in the MuTHER reference cohort . We profiled the expression of 15 to 30 base pair RNA molecules in subcutaneous adipose tissue from 131 individuals using high-throughput sequencing , and quantified levels of 591 microRNAs and small nucleolar RNAs . We identified three genetic variants and three RNA editing events . Highly expressed small RNAs are more conserved within mammals than average , as are those with highly variable expression . We identified 14 genetic loci significantly associated with nearby small RNA expression levels , seven of which also regulate an mRNA transcript level in the same region . In addition , these loci are enriched for variants significant in genome-wide association studies for body mass index . Contrary to expectation , we found no evidence for negative correlation between expression level of a microRNA and its target mRNAs . Trunk fat mass , body mass index , and fasting insulin were associated with more than twenty small RNA expression levels each , while fasting glucose had no significant associations . This study highlights the similar genetic complexity and shared genetic control of small RNA and mRNA transcripts , and gives a quantitative picture of small RNA expression variation in the human population .
A world of noncoding RNA molecules has been uncovered in the last decades , expanding our understanding of functional elements in the genome [1] . After it was found that the small ( ∼15–30 nt ) noncoding RNAs can directly modulate protein levels [2] , [3] , and via that , almost any cellular process [4] , they have been subject to vigorous study , leading to the recognition that several different types of small RNAs can act as posttranscriptional regulators [5] . MicroRNA genes ( miRNAs ) were the first animal small RNA genes to be discovered [6] , and over 1 , 500 examples have been found in humans to date [7] . The primary miRNA transcript has a stem loop structure that is recognised and cleaved via RNA processing enzymes to produce a double stranded duplex [8] . The mature miRNA strand is loaded into a complex containing Argonaute family proteins and guided to targeting , while the other strand is assumed to be degraded . miRNAs target mRNA transcripts via base pair complementarity , typically in the 3′ untranslated region [8] , [9] , but also coding sequence [10] . This targeting can induce transcript cleavage , degradation , destabilisation , or repression of translation , thus modulating protein levels . Small nucleolar RNAs ( snoRNAs ) are typically longer genes ( 60–300 nt ) that facilitate RNA editing within ribosomal or spliceosomal RNAs [11] . However , their full sequences can also be processed into snoRNA derived RNAs that exert a similar mode of action as miRNAs [12] , [13] , [14] . The recent ability to quantify levels of small RNA expression invites questions about the extent and causes of their variability in the human population . Importantly , the quantity and quality of transcripts are the only way genetic variation can influence phenotype . Thus , the genetic contribution to small RNA expression trait variability has to be assessed for accurate understanding of transmission of heritable information . Such questions have already been successfully addressed for mRNA expression levels , where variability between tissues [15] , populations [16] , and diseased and healthy individuals [17] , as well as the contribution of genotype [16] , [18] , [19] , [20] have been thoroughly characterised . Previous studies have found genetic contribution to miRNA levels in both human fibroblasts [21] as well as adipose tissue [22] using miRNA microarrays . However , other types of small RNAs have not been assayed , and a full account of small RNA sequence and transcriptome variability in a reference cohort is missing . Small RNA expression can be viewed as a primary genetic trait to be mapped in isolation , but also as a quantitative trait with downstream influences on gene expression and other phenotypes . Recent studies have been successful in combining information about genotype and intermediate phenotypes ( such as mRNA levels [17] , [23] , [24] or inferred cellular activations [25] ) to understand how the genetic signal is mediated . In this light , it is especially interesting to analyse small RNA transcript levels as intermediate traits potentially causative for downstream effects , as both miRNAs and snoRNAs have already been implicated in many human disease phenotypes ranging from obesity and autism to cancer [26] , [27] , [28] , [29] , [30] , [31] , [32] . The MuTHER ( Multi-Tissue Heritability Resource ) cohort was established with the aim of analysing the genetics of gene expression in multiple human tissues in over 800 individuals [19] , [20] , [33] . This cohort is a subset of the UK Twins [34] , and has extensive information on genotype and gene expression , as well as a plethora of clinical phenotypes . We set out to characterise small RNA variability in 131 abdominal fat samples from MuTHER resource using high throughput sequencing technology . We quantified the content of the small RNA transcriptome , the extent of sequence and transcript level variation , the relative levels of miRNA expression from both arms of the molecule , as well as coexpression of miRNAs from the same cluster . Since high density genotype data , mRNA levels from the same RNA sample as well as obesity-related phenotypes were available for these individuals , we associated these measurements with the small RNA levels to find out about the extent of genetic control , mRNA and miRNA expression correlates , and relation of small RNAs and global metabolic traits .
We sequenced subcutaneous adipose tissue small RNAs of 131 females from the UK TWINS cohort [34] included in the MuTHER study [19] on the Illumina GAII platform ( Materials and Methods , data available at the EGA , submission ID EGAS00001000212 ) . After filtering , quality control , and mapping , we obtained 331 million total reads , with a median of 2 . 3 million reads per sample aligning to the genome ( Materials and Methods , Figure S1 ) . The majority of the reads ( 93% ) mapped to annotated mature miRNA sequences ( mirBase v17 [7] ) , with the rest divided between tRNAs ( 2% ) , snoRNAs ( 0 . 6% ) , lincRNAs ( 0 . 3% ) , and other noncoding RNA features annotated in Ensembl v63 [35] ( Table S1 ) . This distribution is expected , as we size-selected for 15–30 base pair fragments , which excludes other functional RNA species except for degradation products . In addition , we found reads mapped to loci previously unannotated for noncoding RNA transcription . We identified 12 novel miRNA gene candidates using MapMi ( [36] , Materials and Methods , Dataset S1 ) , and 701 short ( <100 bp ) regions with at least 1000 total mapped reads across all samples ( 2% of all mapped reads , Table S2 ) . These regions were significantly enriched in DNAse hypersensitivity sites ( 237/701 , one-tailed binomial p<10−10 , Materials and Methods ) , which often harbour enhancer elements that are known to give rise to short transcripts [37] . The rest overlapped exons ( 149/701 , one-tailed binomial p<10−10 ) and introns ( 190/701 , not significant ) , with 233 regions arising from intergenic sequence . We quantified expression levels of 418 known miRNA gene products , 239 tRNAs , 173 snoRNAs , 111 lincRNAs and 107 other RNAs that had at least 1000 total sequencing reads ( Figure 1A , Table S1 ) . For further analyses , we focused on miRNAs and snoRNA derived sequences as the only known functional molecules in our selected size range . The adipose tissue small RNA transcriptome is of medium complexity , with a median of 17 species of molecules required to account for 75% of the mapped reads ( Figure 1B ) . The most highly expressed small RNAs ( Figure 2A ) have previously been associated with adipose development ( mir-143-3p [38] , mir-21-5p [12] ) , angiogenesis ( mir-126-3p [39] , mir-378a-3p [40] ) , and erythropoiesis ( mir-24-3p [41] , mir-451a [42] ) . We compared the average expression levels in adipose tissue to public human small RNA sequencing data from B-cells [43] , liver [44] , pigment cells [45] , pooled thymocytes , bone marrow , CD34+ progenitor cells [46] , lung , kidney , skeletal muscle , heart , pancreas , frontal orbital gyrus , spleen , and liver tissue [47] after processing them with our pipelines ( Materials and Methods , Table S3 ) . While seven of the ten most highly expressed small RNA genes and gene families were highly expressed in all tissues ( let7 family , mir-24-3p , mir-378a-3p , mir-21-5p ) other highly expressed small RNAs ( mir-143-3p , mir-126-3p ) were specific to adipose tissue ( average q-value of pairwise comparisons <0 . 1 , Materials and Methods ) . In total , there were 12 miRNAs with significantly higher expression ( q<0 . 1 ) compared to mean of every other tissue , and no such snoRNAs , with mir-126-3p , mir-340-3p , mir-190a , and mir-335-3p showing the strongest specificity signal ( Figure 2B ) . Next , we called variants from the RNA sequence data ( Materials and Methods ) , and found one mature miRNA and two snoRNA polymorphisms , all with independent evidence from whole genome sequencing of the UK10K cohort ( personal communication , UK10K Consortium ) ( Table S4 ) . All three found variants had relatively low ( <11% ) minor allele frequency ( MAF ) . Assuming Hardy-Weinberg equilibrium , and equal expression from both gene copies , the miRNA sequence variant represents a fraction of 7×10−5 of the 14 , 005 mature miRNA and star sequence sites that could pass our filters , consistent with previous reports of strong purifying selection in the functional small RNA regions [48] . The same regions in the UK10K project harboured 13 called polymorphic sites , 9 of which had MAF<1% . We detected one of these sites using small RNA sequencing ( MAF = 11% ) , and did not find the rest . Based on the MAF of each UK10K DNA variant , and expression levels of the small RNAs , we expected to recover one additional site ( Materials and Methods ) . While it is possible that other polymorphisms are present in sequences coding for miRNA and snoRNA products , the derived alleles were not observed on at least 10 reads in our data , and could thus not be reliably detected . In addition to genetic variants , we found three A to I RNA editing events in the mature miRNA regions ( Materials and Methods , Table S4 ) . These sites were the 7th , 8th , and 9th bases of the mature product , and edited in 25 , 18 , and 11 percent of the reads , indicating that additional variability is tolerated in the functionally important seed region . We also observed bases at the ends of mapped reads not matching the genome in line with previous reports ( [49] , Table S5 ) , but as similar discrepancies were not observed at comparable frequency in the data from other tissues , we considered them more likely to be sequencing or library preparation artefacts than true RNA modifications . As mature miRNA sequences and analogous snoRNA products function via base pair complementarity , there is selective pressure against accumulating variants in their regions . Previous reports from DNA sequence data have confirmed increased conservation of miRNA sequence compared to intronic and intergenic background , but also a more pronounced effect for more highly expressed genes . We also observed a lack of miRNAs with at least 1000 reads on average and UCSC primate conservation score of less than 0 ( Figure 3A , p<3×10−5 , chi-squared test , Materials and Methods ) . Moreover , we assessed if the variability in the expression levels is under similar influence . Indeed , we observed a lack of small RNAs with expression variance of at least 5 , and a conservation score below 0 ( Figure 3B , nominal p<3×10−4 , chi-squared test ) , suggesting that selection acts on not just average expression , but also expression variation . After analysing the variability of RNA expression levels within and between tissues , we next addressed inter-individual variation . First , we tested whether experimental confounders influenced small RNA expression variability between samples . To this end , we performed principal components analysis of log-transformed , normalised read counts ( Materials and Methods , Tables S6 and S7 ) , and associated first twenty components ( PCs ) to known covariates of sample multiplexing tag , library batch , sequencing flow cell , RNA integrity score ( RIN , [51] ) , and RNA concentration ( Materials and Methods ) . We found significant associations ( Bonferroni-corrected p<0 . 05 ) for RIN ( PC1 ) , library batches ( PC1 , 2 , 3 , 6 , 10 , 11 , 12 , 13 , 18 ) , and two multiplexing tags ( PC5 , 14 ) . As these components capture major directions of variation in the data , we included the associated covariates measured for all samples ( age , library batch and multiplexing tag ) in eventual analyses . Since it has been demonstrated that unmeasured confounders similarly have an influence on expression levels [52] , [53] , we tested whether applying the Bayesian factor analysis package PEER [54] , to account for these confounders , increases the number of discoveries . As we already corrected for 30 known covariates , including the additional inferred factors did not increase findings in downstream analyses , and were not used . To identify miRNA and snoRNA genes whose expression is driven by cis-acting genetic variation , we performed association tests between their transcript levels and SNPs within 100 kb of the transcript ( Materials and Methods ) . We found significant cis-eQTLs ( nominal p<2 . 4×10−4 , FDR<5% ) for eight of 418 miRNAs and six of 173 snoRNAs ( Table 1 ) . In comparison , 462 eQTLs were found for 27 , 499 mRNA probes in the same tissue and cohort with comparable sample size and FDR in a previous study [19] , suggesting a similar level of genetic control for mRNA and small RNA transcript levels . We validated our eQTL findings in an independent cohort of 70 human samples with array-based miRNA expression data from abdominal adipose tissue [22] . Five of the eight miRNAs with an eQTL in our study were assayed in this study , with three of them replicating ( nominal p<0 . 05 , Table S8 ) and p-values across the full set of eQTLs tested in each study concordant ( Spearman rank correlation p<8×10−4 , Figure S2 ) . However , we found no overlap between our significant cis-eQTL results and 12 significant ( p<0 . 05 from 10 , 000 permutations ) miRNA cis-eQTLs reported in human fibroblasts [21] , likely due to a different set of expressed genes and lack of replication power . As mRNA studies in larger cohorts have found only up to a third of genetic associations to be tissue-specific [19] , [55] , we also expect many of the small RNA eQTLs to have an effect in other tissues in better powered studies . The full MuTHER cohort of 776 individuals was profiled for mRNA levels from adipose tissue using the same RNA sample for the individuals in our study , as well as skin and lymphoblastoid cell lines from a separate RNA sample . Thus , we could directly assess any overlap in genetic control of transcripts of different type and across multiple tissues . We found seven of our small RNA eQTL SNPs to also be significantly associated with a nearby mRNA probe ( Table 2 ) . The mRNA transcripts were the nearest annotated transcript to the two miRNAs and two snoRNAs , but at least one and up to four annotated mRNA transcripts away from the rest of the snoRNAs . Further , in three of the eight cases , the mRNA and small RNA did not share the direction of the SNP effect . This suggests nontrivial shared genetic control , either via enhancer or promoter , or a single transcript that is spliced to form multiple genes . As our cohort has been phenotyped for DEXA-derived measurements of percentage trunk fat mass ( PTFM ) , BMI , fasting insulin , and fasting glucose ( summaries in Table S9 ) , we examined the association between small RNA expression and these obesity-related phenotypes ( Materials and Methods ) . We found 47 , 41 and 23 out of the 591 tested small RNAs to be associated with PTFM , BMI and fasting insulin respectively ( per-trait FDR<5% ) . As these traits are highly correlated ( Pearson's r>0 . 45 for all pairwise comparisons ) , there is also considerable overlap in the associated small RNAs between the traits ( Table 3 ) . Fourteen small RNAs were highly significantly associated with at least one of the phenotypes ( FDR<0 . 1% , Table 3 , miRNA targets and functional enrichment analysis [56] in Table S10 , Figure S3 , S4 ) . As a complement to the association analysis , we also contrasted small RNA gene expression levels between lean ( BMI<25; n = 45 ) and obese ( BMI>30; n = 36 ) subjects , and found that 43 small RNAs showed significant differences between the two groups ( FDR<0 . 05 , p<5 . 7×10−3 , Table S11 ) , including all significant hits from Table 3 . Four of the phenotype-associated small RNAs have previously been associated with metabolic phenotypes and/or adipogenesis . In a recent study , mir-1179 was found to be significantly associated ( FDR<5% ) with metabolic syndrome case control status , with lower expression levels in cases [22] . Here , we report similar associations between mir-1179 and obesity phenotypes with lower expression levels associated with increasing BMI , PTFM and FI , all of which are major components of metabolic syndrome . Mir-21 , here significantly associated with obesity phenotypes ( Table 3 ) , has been reported to be involved in regulation of adipogenesis and lipid metabolism through its gene targets TGFBR2 and PPARalpha respectively [57] . Furthermore , mir-21 as well as mir-146b have been reported to be expressed at higher levels in skin tissue from diabetic mice [58] , and in response to glucose stimulation in mouse adipocytes [59] , [60] . Overexpression of mir-29 isoforms in mouse adipocytes resulted in an insulin resistant phenotype [59] . In a recent study , carried out in mouse islets , isoforms of mir-29 were found to contribute to the beta-cell-specific silencing of MCT1 ( SLC16A1 ) expression required for appropriate insulin secretion [61] . In our study , mir-29b-2-5p was significantly associated BMI and fasting insulin , but not PTFM . Immune processes have previously been found to be enriched among mRNAs associated with metabolic phenotypes [17] , [62] , and mir-146a involved in inflammatory processes [63] and innate immunity [64] was here found to be associated with PTFM , BMI and insulin . We overlapped the SNPs for our 14 significant cis-eQTLs ( cis-SNPs ) , with SNPs that are directly associated with obesity-related phenotypes in published genome wide association study ( GWAS ) data [65] . Four of the cis-SNPs were associated ( nominal p<0 . 05 ) with body mass index ( BMI ) [65] , one each with waist-hip-ratio adjusted for BMI ( WHRadjBMI ) [65] , low density lipoprotein ( LDL ) high density lipoprotein ( HDL ) , and total cholesterol ( TC ) , and none with triglycerides ( TG ) [66] ( Table 4 ) . While on the whole , none of the cis-SNPs were genome-wide significant in the GWAS data , they were significantly enriched for nominally significant ( p<0 . 05 ) SNPs in the BMI GWAS results ( [65] , binomial p = 0 . 007 ) , indicating either their pleiotropic effect , or metabolic trait regulation through small RNA expression levels . rs2440129 was nominally significant in the BMI GWAS lookup [65] , while mir-195-3p was significantly associated with both rs2440129 in cis ( FDR<5% , p<2 . 4×10−5 ) , as well as BMI ( FDR<5% , p<3 . 9×10−3 ) and PTFM ( FDR<5% , p<4 . 1×10−3 ) , suggesting a mechanism for the rs2440129 association . Rs6658641 has a significant ( FDR<5% , p<1 . 6×10−4 ) cis association with mir-197-3p in our data ( Table 1 ) , GNAI3 mRNA in three tissues ( Table 2 ) , as well as nominally significant associations to metabolic traits in GWAS . As mir-197 has been reported to regulate the expression of tumour suppressor gene FUS1 [67] and to be upregulated in type two diabetes patients [32] , it is plausible that the effect of rs6658641 genotype on downstream expression and metabolic traits is mediated via the miRNA expression level . miRNA genes are either processed from intronic mRNA sequence , or transcribed from endogenous promoters [68] . A single miRNA promoter can give rise to a transcript that includes a cluster of miRNAs that are then individually cleaved [69] . We tested whether pairwise correlations between expression levels of miRNAs in the same cluster ( defined by Saini et al . [68] to be within a 10 kb block ) are larger than those between random miRNAs , and found significant enrichment of positive correlation ( Materials and Methods , Figure S5 ) . The median of median pairwise correlations between cluster member expression levels was 0 . 37 , compared to 0 . 03 of random miRNA sets of same size ( p<10−8 , Mann-Whitney U test ) . On the other hand , we found little evidence for relation between miRNA expression level and expression of its nearest mRNA probe . The distribution of correlation coefficients was centered on zero , without a heavy tail of positive correlation ( Figure S6 ) , a statistically significant difference to distribution of random small RNA-mRNA pairs ( p>0 . 37 , Mann-Whitney U test ) , or a trend for higher correlation for less distant probes . This shows that mRNA transcript levels are not good predictors of intronic miRNA levels in our dataset , and suggests that more miRNAs are expressed from an endogenous promoter than commonly appreciated , in line with recent findings [69] , [70] , [71] . One of the two modes of miRNA action is directly regulating the transcript level via influencing the stability of the transcript , or direct cleavage [72] . To test whether variability in the miRNA expression levels is related to variability in its target mRNA expression , we calculated correlations between miRNA expression levels and their validated mRNA targets from miRecords [73] or predicted mRNA targets from tarBase v5 [74] both with and without accounting for experimental confounders in mRNA and miRNA data sets ( Materials and Methods ) . To our surprise , we found that the average correlation between miRNA expression levels and their 522 validated targets was −0 . 012 , and their 194 , 205 predicted targets −0 . 004 . While these averages are statistically significantly less than 0 ( one-sample t-test p<0 . 05 and 10−5 respectively ) , they indicate no strong enrichment of extreme negative correlations compared to random miRNA-mRNA pairs ( Figure S7 ) . We also tested whether the miRNA seed sequences are overrepresented in the 3′ UTR regions of the mRNA expression levels most negatively correlated to the miRNA using Sylamer ( [75] , Methods ) . Again , we found no evidence for significant enrichment ( all q-values>0 . 5 ) . This suggests that at a genome-wide level inter-individual variation of small RNA expression levels in our reference cohort does not have a detectably large effect on mRNA expression . The mature miRNA is processed from a double-stranded RNA hairpin by the Dicer RNAse [72] , with the other arm assumed to be degraded [76] . The basis for choosing one of the hairpin arms as a mature product , and the extent to which the alternate arm ( the less commonly observed product , previously also referred to as the star sequence ) is functional , are not well understood [77] , [78] , [79] . To assess the extent of expression of both arms , and the variability of the relative expression ratio , we quantified the expression level of the alternate arm for 63 miRNAs . Other miRNA genes had only one arm detectably expressed , and only eight out of the 63 alternate arms were expressed at average level of at least 250 reads per sample . For seven miRNA genes , the alternate arm was on average more highly expressed compared to the mature product according to miRBase ( Figure S8 ) . Looking at variation between individuals , we found 12 mature sequence expression levels to be significantly correlated with their alternate arm sequence expression level ( |Spearman's rho|>0 . 4 , nominal p<2×10−5 ) . For mir-186 and mir-29a , high abundance of the alternate arm sequence was indicative of low mature sequence levels , suggesting mutually exclusive selection of the arms . As the arm choice is suggested to be influenced by the nearby RNA context [78] , [80] , we tested for whether DNA variants in the region are correlated with the relative abundance of sequence from the two arms . We found SNP rs13174179 to be associated with the expression difference of miR-378 arms ( nominal p = 5 . 3×10−4 , FDR<10% ) .
In spite of the medium complexity of the small RNA transcriptome , we quantified close to 1 , 000 different small RNA species . The highly expressed small RNAs fell into two categories in terms of inter-tissue variability - adipose-specific , and ubiquitously expressed microRNAs , corroborating previous observations [81] , [82] . We confirmed that small RNA sequences have low genetic variability . This finding was especially pronounced for small RNAs highly expressed in the tissue we assayed , as only three derived alleles and three editing events were found . Additional genetic variants have been seen using DNA sequencing methods , but their potential functional impact remains to be assessed in other tissues where the genes are expressed above background level . Purifying selection acting on highly expressed as well as highly variable small RNAs was evident from their high conservation throughout the mammalian lineage , reiterating the importance of these functional molecules . Unexpectedly , some of the largest sources of variability in our data were due to the experimental protocol . The barcoding method used in this study , whereby the indexing tag and the unknown RNA are sequenced in the same read , caused a bias in terms of the profile of small RNAs that were captured . This could be addressed by using a generic 5′ adaptor and one that incorporates the indexing tag via PCR , such that the RNA sequence and the indexing tag are determined in separate reads , or performing the reverse transcription step directly on the flow cell [83] . Similar issues with tag bias have been observed and addressed in recent work published after the experiments reported here were carried out [84] , [85] . Additional limitations for the library preparation were the quality and quantity of the starting material . Although not always feasible in a clinical situation , every attempt should be made to ensure that the quality of the total RNA is of a very high standard ( minimum RIN of 8 ) , and it is subject to minimal handling and freeze/thaw cycles prior to library construction . These considerations forced us to employ statistical methods to account for batch effects due to multiplexing tags , and to drop 37 samples from our initial design due to poor RNA quality . Differences in sample preparation and sequencing platform introduce technical variation that biases and reduces the power of direct comparisons between small RNA sequencing studies [86] . We limited such confounding effects on our assessment of small RNA expression tissue specificity by using only Illumina short read data from other studies , and treating their raw reads in an identical manner to our samples . While we do not expect this to fully mitigate the problem , we do not expect that the residual bias produces the reported large differences between tissues . These considerations do not affect the rest of our analysis , for which the small RNA and mRNA data were collected from the same RNA samples , and genotyping and phenotyping were performed on the same individuals . Another important issue for comparing RNA levels between samples and finding genetic associations was mapping bias due to sequence variants . Previously uncharacterised polymorphisms resulted in fewer reads mapped to samples with derived alleles , which also created a significant eQTL at a known linked SNP . We recommend projects using small RNA sequencing to employ our technique of including known genetic variation in the reference sequence , and to use an ambiguity aware aligner , such as NovoAlign , to avoid such pitfalls . Correcting for these technical issues , we were able to explore the biological causes of small RNA expression variation . We found genetic associations at a rate comparable with mRNA transcripts , and replicated them in an independent cohort . Unexpectedly , we found eight cases of a locus genotype influencing expression levels of a nearby mRNA and a nearby small RNA , where in four of these cases the two were unlikely to share a transcript as they were separated by at least one additional transcribed region . This highlights that cis , or proximal signal does not have to be contained to the near vicinity of the transcript , and that distal regulatory sites are shared between multiple genes . We also looked for coordinated transcription by direct correlation of nearby transcripts . Small RNAs are known to be expressed in clusters from a shared promoter , as well as cleaved from intronic RNA sequence [68] . While we found support for increased correlation between miRNAs from the same cluster , we did not see a global signal for correlation between intronic miRNAs and their nearest mRNA probe expression . Previous results have shown a strong relationship between average tissue mRNA expression level and the intronic miRNA expression [81] , but our results suggest the additional variability around the average level is not as tightly linked , possibly due to an independent promoter of the miRNA , or additional postprocessing regulation of the spliced mRNA transcript . Finally , phenotypic and environmental differences can and do elicit changes in the transcriptome . To this end , we found 51 small RNA genes whose expression level is significantly associated with metabolic phenotypes available for our cohort . Given the strength of the observed signal , it is not possible without additional information to distinguish between causal , reactive , and common cause models for the relationship between the expression and phenotype traits . Studies in mouse models and human cohorts have shown that environmental factors , such as diet , can influence the expression of both mRNA [87] as well as small RNA [88] in adipose tissue . We used fat biopsies taken from individuals who had been instructed to fast the day of the biopsy to control for potential confounding effect of the daily food consumption , but long term dietary behaviour was not available for these samples and thus could not be analysed . Modelling potential hidden causes of variation in the expression data did not increase the number of discoveries , suggesting that even if the environmental factors were observed , they could not be accounted for in a simple linear manner . Despite this , we can not infer in general that the phenotypic variability is due to changes in small RNA expression . In some cases however , previous findings suggest a plausible regulatory effect of small RNAs on phenotypes as highlighted in the results . The MuTHER cohort was set up with the aim to assess heritability of gene expression in different tissues using twins . However , as using highly related subjects reduces the power to map eQTLs using association , we focused our resources on unrelated individuals in the clinically relevant adipose tissue for which related phenotype data and an eQTL replication cohort were available . Analysing multiple tissues , or employing a co-twin design to provide heritability estimates and immediate replication of the results could be followed up from this pilot study . A major goal of this study was to assess the effect of naturally occurring variation in miRNA expression levels on the mRNA levels . However , we found no evidence for miRNA expression variation to be correlated with target mRNA variation . This negative result cannot be due to the amount of noise in our data alone , as we could successfully detect genetic effects and phenotype correlations . Thus , the strength of association between natural variation of miRNA expression and variation in their target mRNA expression is limited to a smaller scale than that of genetic control or downstream effects of global metabolic phenotypes . This lack of tight target regulation supports the growing body of evidence [22] that quantitative variation of small RNA expression within a tissue does not have even a moderately sized effect on its target mRNA levels , and is consistent with a primary role of miRNAs being to buffer mRNA levels , for example to a random fluctuations of transcriptional regulators . The small effect size of drastic miRNA level perturbation via knockdown , transfection , or overexpression of a single miRNA on its target mRNA expression levels has already been shown in several recent studies in human cell lines . For example , the median log2 expression level change of the top 150 TargetScan conserved targets was 0 . 096 ( 6 . 9% ) for mir-29 knockdown in fetal lung fibroblasts [89] , 0 . 131 ( 9 . 5% ) for mir-145 transfection of MB-231 breast cancer cells [90] , 0 . 173 ( 12 . 7% ) for mir-30 overexpression in melanoma cell lines [91] , and 0 . 465 ( 38 . 0% ) for mir-7 overexpression in A549 cancer cells [92] . Thus , even for these extreme perturbations of miRNA levels , the observed effects on the target mRNAs are not pronounced . It is therefore not surprising that the naturally occurring inter-individual variation also does not have a large effect . For the first time , we were able to assess the expression variation of both microRNA arms . We found that while the alternative arms ( star sequences ) are not highly expressed in general , there are several of them that are not degraded , and are expressed at appreciable levels . We also observed examples of high mature miRNA expression being correlated with low expression of the alternate arm , and a relatively strong genetic signal for arm choice of one miRNA .
The unrelated individuals included in this study are part of the MuTHER study of Caucasian females ( median age 58 ) recruited from the UK Adult Twin Registry ( TwinsUK , [34] ) . Punch biopsies ( 8 mm ) were taken from a relatively photo-protected area adjacent and inferior to the umbilicus , subcutaneous adipose tissue was dissected followed by DNA and RNA extraction as described in [20] . For inclusion in this study the requirements were that the individuals were not under hormone replacement therapy , and did not have confirmed Diabetes Mellitus Type 2 . Subjects were instructed to fast on the day of the biopsy to avoid potential biases due to food consumption . We used genotypes obtained , filtered and imputed to HapMap2 as described in [20] . The previously published gene expression values [20] were obtained using the Illumina Human HT-12 V3 BeadChips , followed by filtering and normalisation , and are available at the ArrayExpress [93] ( www . ebi . ac . uk/arrayexpress ) under accession number E-TABM-1140 . Metabolic phenotypes were measured at the same time point as the biopsies and were collected as previously described , including BMI [94] , DEXA measurements of percentage trunk fat mass , fasting glucose [95] and fasting insulin [96] . Only samples with good quality total RNA ( no visible degradation in BioAnalyzer profile and RIN scores in excess of 6 . 7 ) were selected for small RNA isolation . Low molecular weight RNA ( <40 nucleotides ) was size-selected from between 0 . 5 to 1 . 0 µg total RNA using a flashPAGE Fractionator ( Ambion , Austin , TX , USA ) . The recovered small RNAs were first ligated to the Illumina v1 . 5 small RNA 3′ adaptor ( Illumina , Inc . , San Diego , CA , USA ) using T4 RNA ligase 2- truncated ( New England Biolabs , Ipswich , MA , USA ) . This was followed by a second ligation , using T4 RNA ligase 1 , to one of twelve modified Illumina SRA 5′ adaptors , each with a six-base index tag at the 3′ end ( Sigma-Aldrich , Haverhill , UK ) . Both ligation steps were performed according to the Illumina v1 . 5 protocol . The 5′ and 3′ adaptor-ligated small RNAs were immediately reverse transcribed , amplified and size-selected as described in the Illumina v1 . 5 protocol . The completed cDNA libraries were pooled ( 12 libraries per pool ) in equimolar amounts and were sequenced using 37 base reads on the Illumina GAII platform . Raw sequencing data was obtained in FASTQ format , and processed with R [97] ( Bioconductor [98] , [99] , Biostrings and ShortRead [98] , [99] packages ) and python scripts . We first assigned the raw reads to their corresponding multiplexing tags . For this , we calculated the edit distance of the first six bases to all 12 index tag sequences used in the study , considering 0 . 25 as the distance between N and any other base . Reads with edit distances of at least 2 . 75 to all tags were discarded as well as those with the same minimum edit distance to more than one tag . The remaining reads were assigned to the library corresponding to the shortest edit distance , and their first six bases were removed before proceeding . The next step consisted of locating and trimming sequences matching the small RNA 3′ adaptor using the trimLRPatterns function and allowing for mismatches of up to 20% of the alignment length . The first 12 bases of the 3′ adaptor sequence were allowed to align to any location within the short reads , and if no alignment was found , a shortened adaptor sequence was realigned iteratively by removing one base from the 3′ end and anchoring the alignment to the 3′ end of the short reads . To further clean up the short reads to help avoid ambiguous mappings , any window of five bases with at least three Ns was located and the read was trimmed starting at the position of the first N . Any occurrence of an N within two bases of the 3′ end of the read was also trimmed . The reads were then low-complexity filtered to remove those with > = 90% of a single base . After all filtering steps , reads with less than 16 bases were discarded . All remaining read sequences should correspond to short RNA molecules present in the samples , and length histograms were produced to confirm the enrichment of a miRNA peak around 22 bases . Accurate quantification of small RNA molecule counts from read data is challenging due to genetic variation in the sequence , ambiguities in read mapping , and frequent contamination by large numbers of adaptor dimers . To solve these problems , we used a multi-stage mapping approach to exclude contaminating molecules that could be due to the library preparation kit , prioritise alignments to known small RNAs , and take genetic variation into account . First , we aligned the known small RNA molecule sequences against the human reference genome ( NCBI build 37 ) , retrieved all known variants in the mapped regions from the UK10K sequencing data ( July 2011 release , personal communication , UK10K Consortium ) , and created an individual sequence of each small RNA , with variable bases denoted with the corresponding IUPAC ambiguity codes . We included all mapped regions for RNAs that mapped to more than one genomic location . We then created five synthetic reference genomes ( all with the ambiguous bases at variable sites ) corresponding to: We mapped reads to these references using BWA [100] ( bwa aln -n 2 -o 1 ) and with novoalign v . 2 . 07 . 11 ( http://www . novocraft . com , parameters -h 60 60 -t30 -s -m -l 16 -R 0 -r A 30 ) . The latter is aware of sequence ambiguities , but the former was more sensitive at detecting reads aligning to the contaminating sequences . We also tested Bowtie [101] , but did not use it due to the inability of the tested version to handle indels . For the mapping calls , we excluded reads mapping to contaminating sequence with either method . For both aligners , we then took all the alignments in the highest stratum of references ( miRNA>ncRNA>pseudogenes>genome ) , and picked the ones with the smallest edit distance . Conservatively , we only retained alignments of a read if both aligners agreed on all the aligned locations . For reads mapping genome-wide , but not any known ncRNAs , we created the set of uncharacterised RNA loci covered by at least one read in at least one sample without gaps , and assigned reads to their corresponding uncharacterised loci . Finally , we quantified the expression level of each ncRNA and unannotated locus by counting the number of reads aligning to it . If a read mapped between k alternative sequences or loci in one reference , we added 1/k to the count of each . We trimmed the data matrix to contain only RNAs that were observed at least 1000 times across all individuals , or at least 100 times in a single individual . We discarded individuals with less than 500 , 000 mapped reads . This retained 131 individuals , including 129 individuals with at least 800 , 000 reads , and 119 individuals with at least 1 , 500 , 000 reads . This multi-stage approach excludes mapping contaminating sequences to the reference , avoids allelic imbalance due to ability to map by incorporating information on genetic variation , and resolves potential mapping ambiguities to reflect our belief of how small RNA molecules are generated . To use the read counts quantitatively , we normalised the data to have a comparable total number of reads for each individual . We estimated a size factor s for library j as the median inflation factor across all genes: sj = mediang ( ngj/GMj ( ngj ) ) , where GM stands for the geometric mean , and ngj is the read count of gene g for individual j as recommended by Anders and Huber [102] . For further analyses , we used the log2-transformed corrected values log2 ( ngj/sj ) to account for heteroskedasticity in the data . We downloaded UCSC genome browser [50] tracks for DNASE hypersensitivity sites , and ENSEMBL gene structures for human genome version 37 , and calculated their total length , as well as overlap with the loci giving rise to unannotated small RNA molecules . We calculated the significance of the enrichment of unannotated regions in the track from the probability of observing at least as many overlaps of the 701 unannotated regions given the frequency of bases covered by each track using a standard binomial test . We tested for significance of the correlation coefficient r between covariates and principal components of the raw read count data as well as log-transformed and normalised data by calculating a statistic t = r ( ( 1−r2 ) /129 ) −0 . 5 , and calculating the ( two-tailed ) probability of observing at least as extreme a value , and Bonferroni correcting for 131 tests ( one for each PC ) . We called the correlation significant , if the corrected p-value was less than 0 . 05 , corresponding to |r|>0 . 31 . For each sample , we created sorted BAM files from the alignment output , and called segregating sites using Samtools v . 0 . 1 . 12 ( samtools pileup –vcf ) [103] . We then combined the list of all called variable sites across all samples with sites from the UK10K project ( June 2011 release , personal communication , UK10K Consortium ) , created pileup files at them for each sample ( samtools pileup –l sites . tab –f ref . fa ) , and combined all the information into a single table giving the number of times each nucleotide was observed in every sample for each site . The sites were filtered to have information from at least 20 samples , have at least one sample with at least 10 observed alleles , and have at least one sample with at least 20% non-reference allele frequency . We further discarded three sites as likely false positives – one had 23 observed non-reference alleles , with 16 in one sample and no DNA evidence ( see below ) , and the other two were variants in the last base of the mature miRNA , consistent with a modified degradation product . For validating the genotypes using genome sequencing data , we constructed DNA read pileup files at same sites for 40 of our samples sequenced in the UK10K cohort . We called a site to be a DNA polymorphism , if it had at least five DNA sequencing reads supporting the non-reference allele . An A to I edit was called if there were no more than two DNA sequencing reads with a G allele , and both A and G alleles were observed at least 90% of the samples , implying extreme deviation from Hardy-Weinberg equilibrium . We applied the MapMi pipeline [36] to find potential novel miRNA loci . We retrieved the sequences of unannotated genomic regions , calculated their corresponding RNA secondary structure using RNAfold [104] , and applied the MapMi classifier to obtain a structure score s . We calculated the self-containment score c of hairpins with s>35 as described in [105] , and retained hairpins satisfying s*c>35 . We then mapped all reads to the filtered candidate hairpins using bowtie , allowing for zero mismatches , and manually assessed the structural characteristics , genomic context , and alignment pileup shape for each candidate hairpin . Mammalian conservation scores were downloaded from the UCSC genome browser [50] . A chi-square test with one degree of freedom was used to test the deviation of the fraction of highly expressed ( average log-scale expression>10 ) unconserved ( conservation score <0 ) genes from expectation . Similar test was used for highly variable ( log scale variability>5 ) unconserved genes . We subtracted off the linear fit of sample covariates ( library batch and multiplexing tag ) , from the log-transformed , normalised data , and calculated the Pearson correlation coefficient between the residual expression levels and other small RNAs , miRNA star expression levels , and mRNA probe expression levels from [20] . For mRNA levels , we used both raw measurements , as well as residuals after correcting for global variance components using PEER . We also tested for correlation with uncorrected expression levels , and using linear models as described below , but found no additional enrichment of statistical signal . miRNA binding specificity is controlled through binding of its seed region ( bases 1–8 of the mature miRNA ) with seed complementary regions ( SCRs ) in the 3′ UTR of mRNAs . Binding is enhanced if a SCR matches the first seed nucleotide with an adenosine , irrespective of the seed nucleotide [72] . As the strongest statistical associations have been reported for regions of length 6 , 7 and 8 ( the full region ) , we combined analyses for these seed lengths . For each miRNA , we ordered probes and their associated 3′ UTRs by correlation of probe expression values to the miRNA , with correlations calculated in four different ways as described above . For each possible 8-nucleotide sequence s8 ending in an adenosine , we considered its middle 6-mer s6 , the two constituent 7-mers s7 , 1 and s7 , 2 and the 8-mer s8 itself as the seeds . For each of these four seeds s and given n , we used Sylamer 08-185 [75] to calculate a hypergeometric p-value pn ( s ) to assess the extent to which the number of their SCR incidences in the top n of the ordered 3′ UTRs deviated from the expected . Potential nucleotide composition biases were accounted for using third order Markov correction ( flag -m 4 ) . The seed enrichment score for s8 was calculated as maxn ( −log10 pn ( s6 ) −log10 pn ( s7 , 1 ) −log10 pn ( s7 , 2 ) −log10 pn ( s8 ) ) using a grid of values for n . For each miRNA that produced a ranked list of probes , the null distribution of observed scores was estimated by fitting an extreme value distribution for all calculated adenosine-ending 8-mer scores using the R function fgev from the evd package . For the miRNA used to generate the list , the significance of its influence on the mRNA expression was evaluated by testing its seed enrichment score against the estimated null . q-values were calculated for the ordered list of miRNA p-values . We selected series GSE18651 , GSE19737 , GSE27718 , GSE14507 from the Gene Expression Omnibus [106] in which a particular miRNA was directly perturbed , either by knockdown or overexpression . We downloaded the normalized expression data using Bioconductor package GEOquery [107] , and performed differential expression analysis using limma [108] to sort the genes according to fold-change in response to the perturbation . We first validated that the mRNA expression changes actually represent the direct effect of a miRNA on its targets . To do so we used Sylamer [75] to search for enrichment of seed-matches in the 3′UTR sequences in the appropriate portion of the genelist , i . e . in knockdown experiments targets should be up-regulated , upon overexpression targets should be down-regulated . We then obtained the targets of each miRNA according to TargetScan v5 [109] , and calculated their median fold-change in the corresponding experiment . We tested different sets of targets , prioritizing by evolutionary conservation ( PCT ) or by context-score , and selecting the 150 targets with the best scores . In all cases the median fold-change of these target sets was quite low , representing changes of 5–38% . Selecting more targets led to a reduction in the median fold-change . We also calculated the median fold-change of all possible targets , taking the full set of transcripts with at least a 7mer seed-match in their 3′UTR . These larger sets had the lowest median fold-changes , representing a 2–8% change in expression . All this confirms the notion that miRNAs do not act as on-off switches on the majority of their targets . Even in experiments that dramatically alter miRNA abundance , the average effect upon targets is modest . Associations between snoRNA and miRNA expression and mean genotypes ( expected minor allele count under IMPUTE posterior probabilities , MAF>5% , IMPUTE info value>0 . 8 ) or phenotypes were tested using a linear model implemented in R [97] . Cis-eQTL analysis was limited to SNPs located within 100 kB either side of the transcript . The linear model was adjusted for age , multiplex tag and library batch . The significance of the genotype or phenotype effect was calculated from the Chi-square distribution with 1 degree of freedom using −2log ( likelihood ratio ) as the test statistic . False discovery rate ( FDR ) was calculated using the qvalue package implemented in R 2 . 11 [97] . Corrections for multiple testing were done using q-values to control the false discovery rate ( FDR ) at 5% . To calculate the FDR , the associations between the 591 small RNAs and all the cis-located SNPs for each small RNA , were considered . To test for difference in small RNA expression between obese ( BMI>30 ) and lean ( BMI<25 ) individuals we treated BMI , for subjects falling into one of the two BMI groups , as a binary categorical variable . Linear models where fitted with small RNA expression level as response variable and the lean/obese categorical variable as the predictor while adjusting for relevant covariates ( age , library batch , multiplex tag ) . Significance of the effect size estimates of the lean/obese predictor was determined by a likelihood ratio test , and FDR was calculated using the qvalue package .
|
Genetic information is transmitted to the cell only through RNA molecules . A special class of RNAs is comprised of the small ( up to 30 nucleotide ) ones , known to be potent regulators of various cellular processes . At the same time , they have not been as widely studied as messenger RNAs—we do not know how much variation in their sequence and expression level occurs naturally in human populations or how this variability influences other traits . We measured small RNA levels and genetic variability in fat tissue from 131 individuals by high-throughput sequencing . We could associate the expression levels with genetic background of the individuals , as well as changes in metabolic traits . Surprisingly , we found no large scale influence of small RNA variation on mRNA levels , their main regulatory target . Overall , our study is the first to give a quantitative picture of the naturally occurring variation in these important regulatory molecules in human fat tissue .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"population",
"genetics",
"molecular",
"genetics",
"gene",
"expression",
"biology",
"biophysics",
"molecular",
"biology",
"physics",
"rna",
"nucleic",
"acids",
"genetics",
"computational",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics",
"human",
"genetics"
] |
2012
|
Extent, Causes, and Consequences of Small RNA Expression Variation in Human Adipose Tissue
|
Quantifying the dynamics of intrahost HIV-1 sequence evolution is one means of uncovering information about the interaction between HIV-1 and the host immune system . In the chronic phase of infection , common dynamics of sequence divergence and diversity have been reported . We developed an HIV-1 sequence evolution model that simulated the effects of mutation and fitness of sequence variants . The amount of evolution was described by the distance from the founder strain , and fitness was described by the number of offspring a parent sequence produces . Analysis of the model suggested that the previously observed saturation of divergence and decrease of diversity in later stages of infection can be explained by a decrease in the proportion of offspring that are mutants as the distance from the founder strain increases rather than due to an increase of viral fitness . The prediction of the model was examined by performing phylogenetic analysis to estimate the change in the rate of evolution during infection . In agreement with our modeling , in 13 out of 15 patients ( followed for 3–12 years ) we found that the rate of intrahost HIV-1 evolution was not constant but rather slowed down at a rate correlated with the rate of CD4+ T-cell decline . The correlation between the dynamics of the evolutionary rate and the rate of CD4+ T-cell decline , coupled with our HIV-1 sequence evolution model , explains previously conflicting observations of the relationships between the rate of HIV-1 quasispecies evolution and disease progression .
Within an HIV-1 infected individual , the HIV-1 population evolves under host immune response selection pressures [1]–[3] . Development of genetic diversity within the host results from a high virus replication error frequency ( 3 . 4×10−5 mutations site−1generation−1 [4] ) coupled with an in vivo virus production rate exceeding 1010 virions per day [5] . Both diversifying and purifying selection impact the evolution of HIV-1 sequences . In the absence of antiretroviral drug treatment , HIV-1 must balance the preservation of important life cycle functions with the ability to escape host immune surveillance . The interaction between the HIV-1 population and the host is revealed in the following observations: First , an increase of fitness during the course of chronic infection has been demonstrated by comparing the replication rate of virus genomes isolated at early times following infection with that of later viruses [6] . Second , although CD8+ T-lymphocytes restrain virus replication in HIV-1 infection , escapes from both CD8+ T-cell responses and neutralizing antibodies are well documented [7]–[9] . Studies on CD8+ T-cell response to autologous virus Env , Gag , and Tat proteins observed variation at epitope-containing sites in the HIV-1 population [10] , [11] . Such variation implies escape from CD8+ T-cell responses . Furthermore , changes in N-linked glycosylation sites in Env have been observed in viruses that escape antibody neutralization [12] . Two measures have been used to describe HIV-1 evolution quantitatively , diversity , the genetic variation at a given time , and divergence , the genetic distance to a reference point , usually the founder virus . While several studies have investigated these measures , a detailed study carried out by Shankarappa et al . followed 9 patients longitudinally over 10–15 years [13] . They found that in the first phase of the asymptomatic period , both viral divergence and diversity increased linearly in the C2-V5 region of env . In a second phase , the viral population continued to diverge from the founder strain at the same rate , while diversity started to plateau or constrict . In the final phase , divergence stabilized and diversity declined . The decline of diversity was associated with the emergence of viruses using the CXCR4 coreceptors , expressed on both memory and naive cells , more so on naive T cells [14]–[17] . The stabilization of nonsynonymous divergence was reported to be more pronounced than the synonymous divergence at the late stage of infection [18] , [19] , suggesting reduced immune selective pressure . The rate of intrahost HIV-1 sequence evolution has been correlated with the progression of the disease , which shows a considerable variability among patients ( from a few months to 20 or more years ) . Several studies have found an inverse relationship between the rate of viral diversification and the host disease progression rate [1] , [2] , [18] , [20]–[22] , while others have not [23] , [24] . In addition , it has been suggested that the level of genetic diversity that can be controlled by the host immune system is limited , and that exceeding a diversity threshold may be a key factor for disease progression [25] . More recently , Lemey et al . found an association between the synonymous substitution rate of HIV-1 and disease progression parameters [19] . Subjects with moderate disease progression from Shankarappa et al . [13] displayed a faster rate of synonymous substitutions in comparison to subjects with slow disease progression . It was speculated that a longer viral generation time may be responsible for a slower rate of synonymous substitutions and slower disease progression . To unify all these observations , i . e . , the universal intrahost dynamics of divergence and diversity and the contradicting observations between the rate of disease progression and the rate of intrahost evolution , here we propose a simple sequence evolution model that includes a mutation probability and a fitness value of sequence variants . The model accurately described HIV-1 sequence evolution within a patient , reflecting the dynamics of divergence and diversity over the infection by suggesting a slowdown of the evolutionary rate as disease progresses . We then measured the dynamics of intrahost HIV-1 sequence evolution from 15 previously followed patients and linked the change in the evolutionary rate to the dynamics of the CD4+ T-cell count . Deciphering the dynamics of intrahost HIV-1 quasispecies evolution allowed us to explain previously reported contrasting relationships between the speed of HIV-1 quasispecies evolution and disease progression .
To interpret the common dynamics of the divergence and diversity within a host in the chronic phase of HIV-1 infection , we developed a sequence evolution model where each viral sequence is represented by its distance to the founder strain , d . In this model , the number of sequences , N ( d , t ) , at a distance d from the founder strain at time t is dictated by two factors: 1 ) the fitness , F ( d ) , defined as the total number of offspring sequences from sequence d generated per unit time; and 2 ) the probability , M ( d ) , of sequence d to evolve to sequence d+1 per unit time . Here we assume that the unit of time is chosen such that the probability of evolving to distances greater than d+1 in one time unit is negligible . As shown in Figure 1A , at time 0 , the total number of copies of virus 0 is N ( 0 , 0 ) . At time 1 , the total number of offspring sequences from virus 0 is F ( 0 ) N ( 0 , 0 ) . Out of this total number of offspring , the number of mutant sequences is F ( 0 ) N ( 0 , 0 ) M ( 0 ) , and the number of non-mutant sequences is F ( 0 ) N ( 0 , 0 ) ( 1−M ( 0 ) ) . Hence M ( d ) denotes the proportion of offspring that are mutants . In general form , this process is expressed as ( 1 ) This model simulates the growth of the true genetic distance over time . Rather than a simple Hamming distance , which for finite sequence lengths cannot grow at a constant rate , the genetic distance we emulate is the distance realistic substitution models attempt to estimate [26]–[28] . In our later tree analyses of real data , we have used a general-time-reversible model that includes rate variation across sites that has been shown to realistically describe HIV-1 nucleotide evolution [26] , [29] . We show below that the evolutionary rate of the 15 patients we analyze is approximately 10−3 per site per month . Since we analyze about a 600 nucleotide region of the HIV-1 env gene , this implies that we expect less than one substitution per month . Thus , a time unit of approximately one month is appropriate to analyze this data . Thus , our model is not following all the point mutations that can occur due to reverse transcription but rather simulates the growth of the true genetic distance from the founder in the presence of selection . In reality , multiple variants can exist at the same distance from the founder strain . In our model those variants have the same identification index , d , the distance from the founder strain and this implies that we need a measure of diversity that does not rely directly on sequence information but rather on the distribution of genetic distances d . Since an approximately constant number of sequences were sampled at all time points [13] , we consider the normalized distribution of the distances at time t , i . e . , ( 2 ) The divergence Ddivergence ( t ) , i . e . , the average number of nucleotide substitutions that accumulated along the branch from the founder strain as a function of time [13] , is measured by the mean value of d , i . e . , ( 3 ) ( Figure 1B ) . The diversity Ddiversity ( t ) , measured by Shankarappa et al . [13] as the average pairwise nucleic acid distance between all sequences at time t , is here measured with the standard deviation of P ( d , t ) as ( 4 ) ( Figure 1B ) . In our model , because we do not discriminate between the variants at the same distance , we measure the level of diversity with the level of spread in the distance from the founder strain . This measure is an approximation made for consistency with our modeling approach . To examine this approximation , we calculated the measure of genetic diversity at the nucleotide level used by Shankarappa et al . and the measure of the standard deviation of the distance distribution from the founder strain , Eq . ( 4 ) , for the same data in [13] . The two measures were found to be proportional to each other ( Figure 2 ) . We assumed that the probability of mutation varied as a function of the distance from the original strain , d , according to M ( d ) = m if d≤ds , m ( dmax−d ) / ( dmax−ds ) if ds<d<dmax , and M ( d ) = 0 if d≥dmax , where m is a constant , ds is the starting point ( distance ) for the decline of M ( d ) , and dmax is the distance at which M ( d ) = 0 ( left panel in Figure 1C ) . The profile of M ( d ) directly reflects the retardation in the rate of sequence evolution as the virus evolves further from the founder strain . To observe the effect of the profile of M ( d ) ( left panel in Figure 1C ) on the macroscopic evolution patterns , we first fixed the fitness as a constant , F ( d ) = f . Figure 3 shows the fit of the model to the dynamics of divergence and diversity of patients S-P1—S-P11 [13] . The fit of the model is summarized in Table 1 . The method of calculating divergence and diversity dynamics is provided in Materials and Methods . Encouragingly , our model successfully quantified the dynamics of the divergence and the diversity based on first a constant evolutionary rate , then followed by a decline of the evolutionary rate ( left panel in Figure 1C ) . To further investigate the relationship between the profile of M ( d ) and the dynamics of divergence and diversity , we studied two special cases of M ( d ) in greater detail . Submodel 1 is defined by M ( d ) equal to a constant , m ( middle panel in Figure 1C ) . For a constant fitness , F ( d ) = f , the normalized distribution of the distance at time t , P ( d , t ) , satisfies ( 5 ) with P ( d , 0 ) = 1 if d = 0 and P ( d , 0 ) = 0 otherwise . The generating function , , satisfies the following equation , ( 6 ) with F ( z , 0 ) = 1 . The solution of Eq . ( 6 ) is given by F ( z , t ) = em ( z−1 ) t . This implies that P ( d , t ) is a Poisson distribution , P ( d , t ) = e−mt ( mt ) d/d ! . The mean of this distribution gives the divergence , Ddivergence ( t ) = mt , and the standard deviation gives the diversity , . In this special case , both divergence and diversity increase as a function of time rather than saturate or decrease at later time points . Thus , assuming a constant fitness and constant evolutionary rate over the period of chronic infection fails to describe well the simultaneous intrahost dynamics of divergence and diversity . The fit of submodel 1 to the divergence and the diversity dynamics in all 9 patients is summarized in Table 1 . In submodel 2 , we set ds = 0 , resulting in a linear decrease of the probability of accumulated mutations per unit time , given by M ( d ) = ( 1−d/dmax ) for d≤dmax and M ( d ) = 0 for d>dmax ( right panel in Figure 1C ) . In this case , from Eqs . ( 1 ) and ( 2 ) , the dynamics of the evolution is summarized as ( 7 ) for d≤dmax and ∂P ( d , t ) /∂t = 0 for d>dmax . Now the generating function satisfies ( 8 ) with F ( z , 0 ) = 1 from P ( d , 0 ) = 1 when d = 0 and P ( d , 0 ) = 0 otherwise . This equation can be solved using the method of characteristics . Let z and t be the functions of s . Then F ( z , t ) = F ( z ( s ) , t ( s ) ) = F ( s ) and ( 9 ) If we choose the characteristic curve such that ( 10 ) with t ( s = 0 ) = 0 , we have t = s . By comparing Eq . ( 8 ) with ( 9 ) , we obtain ( 11 ) By integrating Eq . ( 11 ) , we have ( 12 ) Along this characteristic curve , by inserting Eqs . ( 8 ) , ( 10 ) and ( 11 ) into Eq . ( 9 ) , we obtain ( 13 ) By integrating Eq . ( 13 ) , we obtain ( 14 ) where we have used the initial condition of F ( z , 0 ) = 1 . Since s = t and from Eq . ( 12 ) , ( 15 ) If we insert Eq . ( 15 ) into Eq . ( 14 ) , we obtain the solution of Eq . ( 8 ) , ( 16 ) Then P ( d , t ) , the coefficient of zd of Eq . ( 16 ) , is given as a binofmial distribution , . Hence , the divergence as a function of time is measured by the mean of this binomial distribution , , and the diversity is given by the standard deviation of P ( d , t ) , . In submodel 2 , the divergence first grows linearly and then saturates , and the diversity first increases and later decreases , which captures the saturation of divergence and the decline of the diversity at later stages of HIV-1 infection . The fit of submodel 2 to the divergence and the diversity dynamics of all 9 subjects is summarized in Table 1 . Comparing the full model and the two special cases using Akaike's information criterion ( AIC ) showed that in all the patients the full model fitted best except for S-P5 , S-P7 , and S-P9 ( Table 1 ) . Submodels 1 and 2 are interesting to consider because they are simpler and have analytical solutions . Comparing the two submodels to each other showed smaller or equal sum of squared errors for submodel 2 in all subjects except S-P3 ( Table 1 ) . One extra parameter in submodel 2 , however , resulted in larger AIC values than for submodel 1 in all subjects except S-P6 . Despite this , we prefer submodel 2 because it qualitatively captures the decrease of the diversity at the later stage of HIV-1 infection . We next studied the impact of viral fitness on the dynamics of the divergence and diversity . Recent ex vivo experimental data have suggested that the replication rate of viruses sampled at a later stage of HIV-1 infection is greater than that of viruses at an early stage of infection [6] . Therefore , we tested how fitness would affect the viral evolutionary pattern at the sequence level . First , we let fitness grow linearly as a function of the distance from the founder strain , i . e . , F ( d ) = f1+f2d , in both submodels 1 and 2 . In the range of f2/f1 from 0 to 10 , we do not find any qualitative change in the patterns of divergence and diversity with time in either submodels 1 or 2 ( Figure 4 ) . This showed that an overall increase in fitness over the disease progression did not have a large effect on the diversity and divergence dynamics . Second , we studied the case where the fitness is reduced after a given distance , F ( d ) = f for d≤dc and F ( d ) = f′ for d>dc where f′ is less than f . Here the proportion of offsprings that are mutants is constant for all viruses , M ( d ) = 0 . 5 . We found that reduced fitness for viruses with a distance greater than dc reproduced the observed patterns for divergence and diversity . Figure 5 displays the calculated dynamics of divergence and diversity when we reduce the fitness of viruses having a distance greater than 50 mutations to 50% of the fitness of viruses with a distance less than 50 mutations . Although the profile of reduced fitness for the viruses after a given distance qualitatively explain the common dynamics of divergence and diversity , the reduction in the fitness of a virus population at later stages does not seem realistic considering the observation of increased fitness over the course of infection [6] . Finally , we investigated the case where only certain types of viruses may evolve to have a greater level of fitness . This situation has been described for emerging CXCR4-using viruses later in disease progression , and was found to correlate with a decline of diversity [13] . To simulate the outcome of emerging CXCR4-using viruses , potentially with greater level of fitness since they have a greater target cell range than CCR5-using viruses by infecting naïve CD4+ T-cells , we assigned a greater level of fitness to a fraction , α , of viruses having a distance larger than a critical value dc . This process is expressed as ( 17 ) where Fhigh ( d ) = Fhigh for d≥dc and Fhigh ( d ) = F otherwise . In this way , a proportion of viruses , 1−α , have fitness F and a proportion α have fitness Fhigh when d>dc . When d≤dc , the fitness is given by a constant F . We here chose dc = 50 mutations for the following reason . As we will show below , the average overall evolutionary rate for the 15 studied patients was estimated at approximately 10−3 nucleotide substitutions per site per month . This corresponds to 0 . 012 substitutions per site per year . With around 600 nucleotides in the dataset [13] , 50 mutations corresponds to the mutations one expects to accumulate during ∼7 years . Thus , we chose the emergence of CXCR4-using viruses at d = 50 from this calibration since usually X4 viruses appear at later stages of infection . Figure 6 plots the dynamics of divergence and diversity by changing the fraction ( α ) of X4 viruses that have a 50% increase of fitness at distance dc = 50 mutations . As we increase the value of α , we observe an increase in divergence , then a transient rapid increase followed again by the inital slope of linear increase . The emergence and persistency of X4 viruses in the population leads to a rapid increase of diversity followed by a decline of diversity . Then at the final stage , diversity starts to increase again . This trend is robust to both the amount of fitness increase and the value of dc . For example , when we chose dc = 30 , the transient rapid increases in the divergence and diversity still occur , but were shifted to 4 . 2 years . An initial rapid increase both in diversity and divergence due to the emergence of more fit virus is not compatible with the in vivo measurements from HIV-1 infected patients ( Figure 3 ) . Overall , these simulations suggest that the probability profile of the evolutionary rate , M ( d ) , rather than the fitness profile , F ( d ) , is the main component in our model that determines realistic within-patient HIV-1 evolution . To test the prediction made by the model , i . e . , a slowdown of the evolutionary rate as virus population evolves further from the founder strain , we calculated the rate of HIV-1 sequence evolution in consecutive windows over a maximum likelihood ( ML ) tree from each patient , starting from the root ( see Materials and Methods ) . We used longitudinal sequence samples for 15 patients from two independent studies [13] , [22] . As an example , Figure 7A shows the tree describing the HIV-1 evolution in patient S-P6 . Figure 7B shows the resulting evolutionary rate as a function of the distance from the root for all patients . Interestingly , the rate is not constant but rather displays a dynamic behavior as HIV-1 evolves . In agreement with our model predictions , 13 out of the 15 patients showed a decline of the evolutionary rate as the sequence population evolved further from its founder strain . The same dynamic behavior was observed using other window sizes ( Δ = 0 . 06 for the Shankarappa data and Δ = 0 . 03 for the Wolinsky data ) . Thus , the observed decline of the evolutionary rate was robust to the size of the window . In Figure 7B , we also plotted the evolutionary profile obtained by a fit to the divergence and diversity dynamics with the full model . The dynamics of the evolutionary rate calculated from the maximum likelihood tree was reasonably consistent with that obtained by a model fit to the divergence and diversity dynamics for each patient . Sometimes we observed negative evolutionary rates in some patients when the distance from the root was large , mostly in later stages of infection ( Figure 7B ) when the sequence population hardly evolves any more . As a consequence some sequence variants may have a smaller distance from the founder stochastically , and if enough of such variants are detected then a negative evolutionary rate will be apparent . Also , the apparent negative rate of evolution may be due to the emergence of less evolved strains from latent reservoirs at later sampling time points . When the rate of change of the evolutionary rate was compared to the rate of change of CD4+ T-cell counts ( Figure 8A ) , a significant correlation ( r = 0 . 68 , P = 0 . 0014 ) was observed ( Figure 8B ) . In the initial interval where CD4+ T-cell counts were relatively stable ( to the left of the dashed bar in Figure 8A ) , the evolutionary rate stayed relatively stable too . As CD4+ T-cell counts decreased and disease progressed in the patients the evolutionary rate slowed down . However , if one compares the overall ( average ) evolutionary rate from the whole study period ( as defined by Eq . ( 20 ) in Materials and Methods ) , not its slope , with the disease progression rate , no clear correlation was seen ( Figure 8B inset ) . The overall evolutionary rate of 15 patients was 10 . 4±3 . 14×10−4 substitutions per site per month . Note that increased or stable viral RNA counts rather than contraction in viral loads were observed in 7 patients under antiretroviral therapy in [13] . Thus , the decrease in the rate of evolution seems not to be associated with the onset of therapy . We estimated overall synonymous and nonsynonymous evolutionary rates across maximum likelihood trees based on synonymous and nonsynonymous changes only using HyPhy [30] . Similar to the overall total substitution rate , we found that neither synonymous nor nonsynonymous overall evolutionary rates correlated with the disease progression rate . For progressors with progression time less than seven years ( S-P1 , S-P5 , S-P6 , S-P7 , and S-P8 ) , the average synonymous and nonsynonymous evolutionary rates were estimated at 6 . 6±3 . 5×10−4 and 12±5×10−4 substitutions per site per month , respectively . For slow disease progressors with progression time greater than seven years ( S-P2 , S-P3 , S-P7 , S-P9 and SP-11 ) , the average synonymous and nonsynonymous evolutionary rates were estimated at 6 . 8±2 . 3×10−4 and 13±4 . 5×10−4 substitutions per site per month , respectively . Lemey et al . reported lower overall synonymous evolutionary rates for these same slow disease progressors [19] . These contradictory observations may be explained by the use of different methods in the estimation of the overall evolutionary rates . While Lemey et al . used codon substitution models with a Bayesian relaxed clock model , we estimated the overall synonymous and nonsynonymous evolutionary rates in separate maximum likelihood trees based on synonymous and nonsynonymous changes [30] to allow for detecting rate changes across the trees . A common finding with Lemey et al . is that they also reported higher nonsynonymous rates ( 8 . 2±3 . 0×10−4 ) than synonymous rates ( 3 . 8±1 . 9×10−4 ) . Importantly , although the overall synonymous evolutionary rate did not correlate with the disease progression rate in our calculations , we found that both synonymous and nonsynonymous evolutionary rates decline as disease progresses in 7 and 8 out of 9 patients in Ref . [13] , respectively ( Figure 9 ) . It is well known that HIV-1 recombines during its evolution . Therefore , we investigated whether recombination could have obscured our estimates of the evolutionary rates . All patient populations showed some signal for recombination ( Table 2 ) . This signal was , however , strongly correlated to the degree of homoplasy ( r = 0 . 91 ) . The homoplasy also grew with number of sequences per patient ( R = 0 . 92 ) , and all patients showed departures from neutral evolution , suggesting stochastic effects as well as selective environments rather than recombination . Most importantly , all our ML trees showed a clear time order of how sequences had been sampled through time ( Figure 6 ) , and additional trees calculated using SplitsTree showed that if recombination had occurred , then mostly samples taken close in time had been involved ( data not shown ) . Thus , although difficult to exactly quantify , recombination had no large effect on our estimates of the evolutionary rate .
The objective of this study was to develop a sequence evolution model and use it to investigate the relationship between nucleotide substitutions and disease progression within HIV-1 infected patients . In particular , we focused on explaining the pattern in which divergence from the founder increases linearly with time since infection and then saturates , whereas sequence diversity increases and ultimately declines . With these aims we developed a sequence evolution model , fitted the model to the divergence and diversity dynamics , and investigated two previously described datasets with rich HIV-1 nucleotide sequence data and CD4+ T-cell counts over time . Two important conclusions could be drawn from this study . First , we found that a model in which the survival of HIV-1 mutants was dictated by the distance from the founder strain accurately simulated HIV-1 within-patient evolution . This model could realistically simulate previously observed patterns of HIV-1 nucleotide sequence diversity and divergence over time by introducing an initially constant evolutionary rate later followed by a decline of the rate . Second , the evolutionary rate of HIV-1 within a patient follows the decline of the CD4+ T-cell count over time . Thus , the evolutionary rate of HIV-1 is not constant over time , but rather evolves in a dynamic way . This dynamic feature provides an explanation for previously conflicting observations of the relationship between the rate of HIV-1 quasispecies evolution and disease progression . Three factors may contribute to the decrease of HIV-1's evolutionary rate as a function of disease progression . First , a decrease in the number of target cells of HIV-1 may increase the effective viral generation time . At the late stage of infection , the overall CD4+ T-cell count drops rapidly while viral load increases [13] . Lymph node immunohistologic alterations in HIV-1 patients as well as progression to a burned-out lymph node accompanying end-point lymphocyte depletion in SIV have been reported [31] , [32] . Rapid loss of CD4+ T-cells in parallel with viral load increase might suggest that the proportion of infected cells out of total CD4+ T cell population is escalated as disease progresses . Our observation of a positive correlation between the slope of the evolutionary rate decrease and the slope of the CD4+ T cell count decline supports this view ( Figure 8 ) . Furthermore , the dynamics of the synonymous substitution rate shows qualitatively a similar pattern as the dynamics of the total evolutionary rate . Thus it follows that a decrease of the synonymous rate in most patients suggests an elongation of the effective viral generation time . In agreement , it was recently suggested that a slower rate of synonymous substitutions in patients with slower progression to AIDS was indicative of longer viral generation times [19] . Second , a weakening of immune selection pressure , as measured by the CD4+ T-cell count , may lower the observed evolutionary rate ( Figure 8 ) . Calculating the evolutionary rate in windows across a tree allowed us to detect a clear correlation between the slope of the evolutionary rate and the slope of the CD4+ T-cell count . Hence , deceleration of HIV-1 sequence evolution occurs in response to decreased immune selection . Not surprisingly , HIV-1 intrahost sequence evolution follows a principle of quantitative genetics where the response to selection is directly proportional to the intensity of selection [33] . If this scenario is operating , then the nonsynonymous evolutionary rate should decrease with disease progression . Here , we found that both the nonsynonymous and synonymous evolutionary rates decreased as disease progressed , supporting this scenario in addition to the first explanation . Thus , the decrease in the evolutionary rate at later stage of infection relates both to amino acid changing and non-changing nucleotide substitutions . Third , an increase of the viral fitness at later stages of infection may reduce further accumulation of mutations , finding a local fitness maximum in the rugged fitness landscape . A correlation between the decline of diversity and the emergence of viruses using the CXCR4 coreceptor was reported in Ref . [13] . The surface expression of the HIV-1 coreceptors CCR5 and CXCR4 on CD4+ T-cells is differentially expressed on memory versus naïve T cells . A chemokine receptor CXCR4 is expressed on both memory and naive cells , although at greater levels on naive T-cells [14]–[17] . It has been reported that naïve T-cells are indeed infected and may act as an important viral reservoir in patients with CXCR4-using viruses [34] . Interestingly , our modeling revealed that the emergence of a fitter virus population , using CXCR4 , resulted in a rapid increase both in divergence and diversity followed by the initial slope of linear increase of divergence and decline of diversity if the probability of mutations is a constant for all viruses . Viral escape from neutralizing antibodies [8] , [12] , [35] and CD8+ T-cell responses [7] , [10] , [36] , [37] suggest that , within a host , the HIV-1 sequence population is evolving in a dynamic environment of immune pressures . One of the selection forces controlling the evolution of env is escape from antibody neutralization . For instance , changes in N-linked glycosylation sites in env have been observed in viruses that escape antibody neutralization [12] . Also , as shown by an antibody neutralization assay , the virus population at a specific time point is neutralized more strongly with antibodies sampled at a later time point [8] . Interestingly , in Table 1 of reference [8] , we observed that antibodies generated at later time points had a lower neutralizing capacity than those generated earlier during infection . For example , the maximum strength of neutralization against virus sampled at month 0 occurred with antibodies ( plasma ) sampled at month 12 . Virus sampled 6 months later had a lower neutralizing titer with antibodies sampled at month 18 , and the neutralization strength decreased as disease progressed . This observation suggests a weakening of the immune selection pressure during chronic infection . Furthermore , apparent decrease of CD8+ T-cell levels in HIV-1 chronic infection , as well as the exhaustion of CD8+ T-cells as mediated by the PD-1 molecule [38] , both imply diminishing CD8+ T-cell responses over disease progression . Recent observations of selective depletion of high-avidity HIV-1 specific CD8+ T-cells after early HIV-1 infection also implies a lessening of CD8+ T-cell responses [39] . Thus , when the diversifying selection pressure on Env from the immune system weakens new escape mutations are not beneficial , and thus the probability to establish new mutations decreases . The immune pressure selects and removes all virus variants it can detect , while those escaping are an increasingly diverse set during chronic infection . When the immune system pressure fails in late stage disease this pressure to diversify is released and as a result , a relatively homogeneous sequence population is observed . Previous studies have suggested an inverse relationship between disease progression and evolutionary rate based on the observation of enhanced viral escape under strong immune selection in slow progressors [2] , [22] . Also , slower genetic diversification has been associated with rapid CD4+ T-cell decline [1] , [20] , [21] , [40] . However , others have reported a positive relationship , suggesting that the evolutionary rate may be low in nonprogressors due to that immune selection may suppress emerging virus with potentially high fitness [23] , [24] . To resolve these conflicting observations , we have shown that the rate of HIV-1 env evolution does not remain constant within a single infected individual , and thus simply correlating the average rate of evolution with disease progression may be misleading . Indeed , this may explain the contradictory results previously published . Thus , rather than using average rates , we show that the dynamics of the evolutionary rate reflects the rate of disease progression . In addition to the 13 out of 15 patients in Figure 7 , 3 out of 6 rapid progressors in Ref . [23] show a decrease in the evolutionary rate when their CD4 cells rapidly deplete , while 3 non-progressors display a stable evolutionary rate . Our estimates of the evolutionary rate were based on maximum likelihood trees calculated using realistic evolutionary substitution models [2] , [30] , [41] . However , these trees implicitly assume that no recombination has occurred , an assumption that may be violated by HIV-1 [42]–[45] . Detecting recombination among closely related HIV-1 sequences within a patient is difficult due to other evolutionary mechanisms causing a high degree of homoplasy ( parallel and convergent mutations in different lineages ) , potentially misleading the analysis . Indeed , most of the patient sequence sets in this study suggest some degree of intra-population recombination strongly correlated to the degree of homoplasy in the dataset ( r = 0 . 91 ) . The recombination rates are estimated under the assumption that just a single mutation has caused each polymorphism within a group , and that there is no selection [46] . Because these assumptions are violated by HIV-1 env V3 we evaluated the potential recombination signal . It is well known that the env V3 region is under positive selection [47] , which may lead to convergent evolution on some residues , explaining some of the homoplasy . In our data both synonymous/nonsynonymous mutation ratios and Tajima's D statistic [48] indicated departures from neutrality ( Table 2 ) . Importantly , previous studies have shown that recombination and selection rates may confound each other [49] , [50] . Also , it is clear that the homoplasy increases as more sequences are investigated ( Table 2 ) . Thus , although difficult to exactly quantify , part of the detected recombination signal in our data could be explained by stochastic effects and convergent evolution . This potential recombination was also analyzed using SplitsTree [51] . Importantly , that analysis showed that if recombination had occurred , it mostly involved sequences collected closely in time . Therefore , the recombination in our data could only have affected our rate estimates mildly . Most important , and in agreement with previous publications using these data ( e . g . , [13] , [19] ) , all our trees displayed a clear time-order of the samples ( Figure 7 ) , which would have been impossible if recombination had had a strong effect . Similarly , if ancestral ( archival/latent ) virus reemerged at later time points , we would have lost the time-order in the trees . In conclusion , neither recombination nor reemerging viruses could have had a strong effect on our rate estimates . Williamson et al . [18] obtained maximum likelihood estimates for the mean divergence rate and the divergence stop time in each Shankarappa patient for the nonsynonymous and synonymous changes . They observed a strong relationship between the time of disease progression and the time of divergence stabilization only for nonsynonymous sites . The evolution profile in [18] corresponds to a constant evolutionary rate before time τ followed by zero evolutionary rate after τ . This kind of evolutionary profile can be imposed in our model by introducing the evolutionary profile depending on the time rather than the distance from the founder strain , M ( t ) . Then Eq . ( 1 ) is modified as ( 18 ) The probability of mutation is a non-zero constant before τ and zero after τ , M ( t ) = m for t≤τ and M ( t ) = 0 for t>τ . We fix the fitness as a constant , F ( d ) = f . Figure 10 shows that not only divergence but also diversity saturates after τ . Since the evolution of the total population stops at τ , divergence and diversity stay constant afterwards . Hence , we can conclude that this alternative model , where the evolutionary profile depends on time , does not capture the decline of diversity at later stages of infection . Similar to HIV-1 , one study on intrahost sequence evolution in hepatitis C virus ( HCV ) reported that the diversity increased over time in non-progressors [52] . In contrast , progressors to end-stage liver disease showed that diversity in the hypervariable region I of E1/E2 env narrowed over time . We expect that the slowing down of the rate of HCV evolution also occurs as disease progresses , resulting in less diversity . In conclusion , we observed that the evolutionary rate of the HIV-1 slows down in 13 of 15 patients from two independent previous studies [13] , [22] . The rate of change in the evolutionary rate is correlated with the slope of CD4+ T-cell decline , dissolving previously reported conflicting observations of the relationships between the rate of HIV-1 evolution and disease progression . Our HIV-1 evolutionary model successfully captured the saturation of divergence and the decrease of diversity observed in the later stages of infection . In our model these effects are mostly attributed to a decrease in the proportion of offspring that are mutants in the population as the distance from the founder strain increases .
We analyzed sequence data from two independent studies , the env C2-V5 region from nine patients [13] and the V3–V5 region from another six patients [22] . Briefly , sequences from the first nine patients were collected over their entire disease progression . The follow-up time varied between 6 to 12 years , at which time seven had developed AIDS and seven of the patients received antiretroviral treatment [13] . The other six patients were followed for 3 to 10 years [22] . Three of these patients received antiretroviral therapy 2–5 years after infection . All HIV-1 sequences were downloaded from the HIV database ( GenBank Accession numbers AF137629-AF138163 , AF138166-AF138263 , AF138305-AF138703 and U35894-U36185 ) . Sequences were aligned using Se-Al [53] . Trees were created using enhanced and parallelized versions of fastDNAml and Rates [54] , [55] , that fit a general-time-reversible substitution model ( RevML ) and site rate specific rates ( RevRates ) in an iterative way [41] , [56] . Briefly , a candidate tree topology was created assuming uniform site rates and an initial random estimate of nucleotide frequencies and transition rates . RevML proceeds in a heuristic and piecewise way , starting from a small set of sequences and building up the tree topology and branch lengths while making placement decisions that maximize the tree likelihood score , similar to a stepwise addition algorithm . The resulting tree then constrains per-site rate optimization of tree likelihood as a function of global estimates of baseline nucleotide frequency and transition rates . These estimates are fit using the conjugate gradient algorithm in the RevRates program . A second RevML run was then performed using these estimates and in turn another rate estimation procedure refined from the second tree . A final tree was estimated using the twice-refined global and local site rates . Each of the trees in the refinement procedure was independently estimated from the global and site local rate parameters . Trees based on synonymous and non-synonymous substitutions , respectively , were inferred using HyPhy with optimized MG94xREV models [30] . This model uses a codon-based substitution model ( MG94 ) that considers substitutions involving non-stop codons , augmented with a general-time-reversible nucleotide substitution model ( REV ) to include the heterogeneity in nucleotide frequencies and substitution rates [57] . The total number of changes per codon is decomposed into synonymous and nonsynonymous changes according to the universal coding table . Separate synonymous and nonsynonymous rates are then fitted to each branch of the tree . Prior to model fitting and tree reconstruction , alignments were codon corrected using the HyPhy SeqAlignment procedure ( with the HIV-1 25% scoring matrix ) . We used SITES [46] and SplitsTree [51] to investigate potential recombination signals in each patient set of sequences , PAUP* [58] to estimate the amount of homoplasy , SNAP [41] to estimate average synonymous/non-synonymous rates , and Tajima's D to estimate neutral evolution[48] . The rate of evolution was calculated in consecutive windows over a maximum likelihood ( ML ) tree from each patient , starting from the root . The distance to the root for all taxa in each window [d , d+Δ] was calculated from the tree ( Figure 7A ) , and the resulting evolutionary rate was estimated as ( 19 ) where di ( dj ) is the distance from the root of sequence i ( j ) at sampling time point ti ( tj ) . Here d ̅ is the average distance from the root over all the sequences within the window , [d , d+Δ] . The window size was Δ = 0 . 09 for the Shankarappa data set and Δ = 0 . 02 for the Wolinsky data set . The average evolutionary rate over the entire sampling period from a patient was calculated as ( 20 ) by averaging the rate of evolution over the sequences in reference to the founder strains which are sampled at the earliest time point in each subject . Here , NF is the total number of founder strains and Ns is the total number of sequences in a patient .
|
Saturation of sequence divergence and a decline of diversity in later stages of infection have been commonly observed during HIV-1 infection , although the length of the time to acquired immunodeficiency syndrome ( AIDS ) is highly variable among patients . To explain this common feature , we developed a simple sequence evolution model with two main components: ( i ) fitness , the number of offspring produced , and ( ii ) the proportion of offspring that are mutants . Assuming a decrease in the proportion of offspring that are mutants as virus variants evolve further from the founder strain , we were able to fit the universal trends of divergence and diversity . In contrast , neither the model with gradual increase of fitness nor the model with rapid emergence of virus variants with greater fitness explained the dynamics of divergence and diversity . The prediction of the model was confirmed in the majority of longitudinally followed patients; the rate of HIV-1 evolution was stationary before disease progresses; however , the rate slowed down at a rate correlated with the rate of immune cell decline . Deciphering dynamic correlation between the rate of HIV-1 evolution and the kinetics of immune cell level united previous conflicting observations of the relationships between the rate of HIV-1 evolution and disease progression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"computational",
"biology/evolutionary",
"modeling",
"infectious",
"diseases/hiv",
"infection",
"and",
"aids"
] |
2008
|
Dynamic Correlation between Intrahost HIV-1 Quasispecies Evolution and Disease Progression
|
Dendritic cells ( DCs ) are phagocytes that are highly specialized for antigen presentation . Heterogeneous populations of macrophages and DCs form a phagocyte network inside the red pulp ( RP ) of the spleen , which is a major site for the control of blood-borne infections such as malaria . However , the dynamics of splenic DCs during Plasmodium infections are poorly understood , limiting our knowledge regarding their protective role in malaria . Here , we used in vivo experimental approaches that enabled us to deplete or visualize DCs in order to clarify these issues . To elucidate the roles of DCs and marginal zone macrophages in the protection against blood-stage malaria , we infected DTx ( diphtheria toxin ) -treated C57BL/6 . CD11c-DTR mice , as well as C57BL/6 mice treated with low doses of clodronate liposomes ( ClLip ) , with Plasmodium chabaudi AS ( Pc ) parasites . The first evidence suggesting that DCs could contribute directly to parasite clearance was an early effect of the DTx treatment , but not of the ClLip treatment , in parasitemia control . DCs were also required for CD4+ T cell responses during infection . The phagocytosis of infected red blood cells ( iRBCs ) by splenic DCs was analyzed by confocal intravital microscopy , as well as by flow cytometry and immunofluorescence , at three distinct phases of Pc malaria: at the first encounter , at pre-crisis concomitant with parasitemia growth and at crisis when the parasitemia decline coincides with spleen closure . In vivo and ex vivo imaging of the spleen revealed that DCs actively phagocytize iRBCs and interact with CD4+ T cells both in T cell-rich areas and in the RP . Subcapsular RP DCs were highly efficient in the recognition and capture of iRBCs during pre-crisis , while complete DC maturation was only achieved during crisis . These findings indicate that , beyond their classical role in antigen presentation , DCs also contribute to the direct elimination of iRBCs during acute Plasmodium infection .
The spleen is a primary site for the control of blood-borne infectious diseases in humans and rodents [1] , [2] . Although splenic phagocytic activity has been well documented in vitro[3]-[5] and ex vivo [3] , [6]-[8] , few studies have reported on the in vivo three-dimensional ( 3D ) interactions between splenic phagocytes and pathogens [9] . Addressing this issue is particularly important in the case of malaria , a disease characterized by splenic involvement that is critical for controlling blood-stage Plasmodium parasites [10] . In recent years , confocal intravital microscopy ( CIVM ) [11] has been used to study host-pathogen interactions during infectious diseases caused by viruses [12] , [13] , bacteria [14] and protozoan parasites [15] . For example , CIVM revealed important aspects of the Plasmodium life cycle [16] , [17] . Other works described Plasmodium-induced immune responses inside the placenta [18] and the dermis using fluorescent stereomicroscopy [19] . A single publication reported the movements of Plasmodium-infected red blood cells ( iRBCs ) inside the spleen [20] . However , no in vivo study has addressed the interactions between blood-stage Plasmodium parasites and the splenic immune system . Splenectomized patients with acute Plasmodium falciparum infections have an impaired ability to remove parasites from circulation [21] , similar to splenectomized mice infected with the blood-stages of Plasmodium chabaudi ( Pc ) [22] . In humans and mice , the phagocytosis of iRBCs or free merozoites by splenic phagocytes begins soon after infection and helps to control the parasitemia and induce the lymphocyte response [23] , [24] . This occurs primarily inside the red pulp ( RP ) and the marginal zone ( MZ ) of the spleen [23] , [24] , where a complex phagocyte network is formed by heterogeneous populations of macrophages and dendritic cells ( DCs ) [25] , [26] . In an effort to characterize the role of splenic phagocytes in Pc malaria , a recent study identified migrating monocytes as major participants in the clearance of iRBCs [8] . However , previous studies that quantified the ex vivo phagocytosis of iRBCs by flow cytometry reported low percentages of splenic phagocytes containing Pc remnants [8] , [27] . This observation is not fully compatible with the notion that the role of the spleen is of the utmost importance in parasite control . DCs are phagocytes that are highly specialized in presenting antigens to T cells [28] . Splenic DCs are efficient antigen presenting cells ( APCs ) during the massive T and B cell responses to acute Pc malaria [29]-[31] . Within the first week of Pc infection , splenic DCs up-regulate the expression of major histocompatibility complex ( MHC ) and costimulatory molecules , secrete pro-inflammatory cytokines , and stimulate T cell proliferation and IFN-γ production [32]-[34] . Nevertheless , it is still unclear whether DCs are unique in their ability to initiate CD4+ T cell responses to Pc blood-stages in the spleen , as observed in Plasmodium berghei ( Pb ) malaria [35] . Moreover , many details concerning the dynamics of splenic DCs in malaria remain unknown , limiting our understanding of the involvement of these cells in the protective immune response . After taking on antigens , immature DCs lose the ability to phagocytize and migrate towards T cell-rich areas to initiate the adaptive immune response [28] . Thus , it would be expected that DCs leave the RP soon after phagocytizing iRBCs or free merozoites and no longer contribute to parasite clearance , although this is as yet only a supposition . In this study , we took advantage of experimental approaches that enabled us to deplete or visualize splenic DCs in vivo to clarify these issues . The in vivo depletion of phagocytes clearly demonstrated that DCs are key participants in the early control of the blood stage of infection with Pc and Plasmodium yoelii ( Py ) iRBCs , as well as the blood stage of infection with Pb sporozoites . The phagocytosis of Pc iRBCs by splenic DCs was analyzed by CIVM , as well as by flow cytometry and immunofluorescence , in three distinct situations: at the first encounter , at a pre-crisis phase concomitant with parasitemia growth and at a crisis phase , when parasitemia has dramatically dropped and changes in the splenic architecture have culminated in spleen closure [36] . CIVM allowed us to visualize the phagocytosis of Pc iRBCs by the RP DC network , the movement dynamics and morphological changes of DCs and the interaction between DCs and CD4+ T cells at the different phases of acute Pc malaria . To our knowledge , this is the first description of the in vivo interaction between Plasmodium iRBCs and the splenic immune system .
To evaluate whether DCs are important for the early control of blood-stage Pc malaria , C57BL/6 . CD11c-DTR ( B6 . CD11c-DTR ) mice were treated with diphtheria toxin ( DTx ) . The great majority of splenic CD11c+I-A+ cells were eliminated in DTx-treated B6 . CD11c-DTR mice ( Fig . 1A ) . No effect was observed on F4/80+ RP macrophages , but the already small population of MARCO/MOMA-1+ MZ macrophages was depleted ( S1 Fig . ) . Starting in the earliest days of infection , DTx-treated B6 . CD11c-DTR mice had higher parasitemia ( Fig . 1B ) and weight loss ( Fig . 1C ) in comparison to their PBS-treated counterparts , leading to an accumulated mortality of 75% of mice on day 15 p . i . ( Fig . 1D ) . On day 4 p . i . , DTx-treated B6 . CD11c-DTR mice had reduced numbers of CD4+ T cells per spleen ( Fig . 1E ) . DTx treatment also completely abrogated the CD4+ T cell proliferation and IFN-γ production in vitro in response to iRBCs ( Fig . 1F ) . None of these effects were observed in DTx-treated C57BL/6 ( B6 ) mice ( Figs . 1 and S1 ) . Furthermore , the selective elimination of MZ macrophages by treating B6 mice with a low dose of clodronate liposomes ( ClLip ) did not affect the course of parasitemia , IFN-γ production by splenic CD4+ T cells or mouse survival ( S2 Fig . ) . Similarly to what was observed for the Pc parasite , DTx treatment in B6 . CD11c-DTR mice exacerbated Py malaria from the beginning of infection ( S3A–S3C Fig . ) . The role of DCs in the early control of parasitemia was also evaluated in B6 and B6 . CD11c-DTR mice that were treated with DTx on day 2 p . i . with Pb sporozoites . DTx-treated B6 . CD11c-DTR mice presented with higher parasitemias ( S3D–S3E Fig . ) . In this case , however , DTx treatment prolonged the survival of infected B6 . CD11c-DTR mice by protecting them from cerebral malaria ( S3F Fig . ) . To investigate whether splenic DCs phagocytize iRBCs in recently infected mice , we analyzed the interaction between YFP+ cells and mCherry-Pc iRBCs in the subcapsular RP of C57BL/6 . CD11c-YFP ( B6 . CD11c-YFP ) mice using CIVM [26] . Mice were infected by i . v . administration of mature iRBCs ( >95% late trophozoites/schizonts ) , as these cells are known to be recognized and phagocytized by DCs [37] . In naïve mice , YFP+ cells were non-motile and actively extended protrusions and dendrites ( S1 Video ) . At 15 min p . i . , mCherry-Pc iRBCs were present in the subcapsular RP ( Fig . 2A , S2 Video ) . CIVM 3D animations showed mCherry-Pc iRBC remnants inside YFP+ cells ( yellow spots of merged mCherry/YFP-3D signal; Fig . 2B , S3 Video ) . At this time , 16% of YFP+ cells contained mCherry-Pc fragments ( Fig . 2C ) . We also observed several mCherry-Pc iRBCs trapped by YFP+ cells without visible signs of internalization ( Fig . 2A , S4 Video ) . Thus , a substantial proportion of subcapsular RP YFP+ cells trapped or internalized iRBCs soon after Pc infection . These cells were not activated , as indicated by small YFP+ cell volume and sphericity ( Fig . 2D ) . The phagocytic activity of splenic DCs from recently infected B6 mice was also analyzed ex vivo by immunofluorescence and flow cytometry . Immunofluorescence revealed approximately 5% CD11c pixels that were colocalized with GFP pixels in those spleens ( Fig . 3A and 3B ) . The majority of GFP-Pc iRBCs were trapped inside the RP and MZ ( Fig . 3B ) . Nearly 2% of CD11c+ cells internalized Cell Tracer Violet ( CTV ) -Pc parasites ( 4 × 104 CTV+CD11c+ cells/spleen ) , as revealed by flow cytometry ( Fig . 3C ) . Comparable data were obtained with Green Fluorescent Protein ( GFP ) -Pc iRBCs ( S1 Table ) . This phagocytic activity was not restricted to a DC subtype , as subsets of CD11c+ cells co-expressing CD11b , CD8 , B220 or CD4 were CTV+ ( S4A Fig . ) . Considering the numbers of cells per spleen , CD11b+CD11c+ cells were responsible for most of the parasite clearance carried out by CD11c+ cells in recently infected mice ( S4B Fig . ) . Although 61% of YFP+ cells in recently infected B6 . CD11c-YFP mice had a DC phenotype , expressing CD11c and MHC class II ( I-A ) but not F4/80 , 20% displayed the phenotype of F4/80+ RP macrophages ( S5 Fig . ) . Therefore , we also analyzed the phagocytic activity of the YFP+ cell subsets by CIVM and flow cytometry . With injection of a fluorescent anti-F4/80 mAb into mice , CIVM revealed that 17% of cells in the subcapsular RP YFP+ cell population were F4/80+ soon after infection ( Fig . 2E and 2F , S5 Video ) . Approximately 15% of F4/80+YFP+ and F4/80-YFP+ cells internalized Cell Tracker Red CMTPX ( CMTPX ) -Pc parasites ( Fig . 2G ) , but only 20% of the CMTPX+YFP+ cells were F4/80+ ( Fig . 2H ) . Flow cytometry analysis of the YFP+ cell subsets showed that a proportion of CD11c+ and F4/80+ cells was CTV+ in B6 . CD11c-YFP mice that were recently infected with CTV-Pc iRBCs ( Fig . 3D ) . The CD11c+ cells made up 63% of the CTV+YFP+ cell population ( 4 . 5 × 104 CTV+CD11c+YFP+ cells/spleen ) , while 37% of CTV+YFP+ cells expressed F4/80 ( 2 . 5 × 104 CTV+F4/80+YFP+ cells/spleen ) ( Fig . 3E and 3F ) . Next , we evaluated the dynamics of splenic DCs during early Pc malaria . At 12 h p . i . , the subcapsular RP YFP+ cells from B6 . CD11c-YFP mice displayed higher speed and displacement ( Fig . 4A ) . This enhanced motility of YFP+ cells correlated with their migration towards CD4+ T cell-rich areas . This was evident in immunofluorescences , at 2 h and 24 h p . i . , by the presence of yellow areas of merged FITC/PE signal ( Fig . 4B ) and higher percentages of CD11c-CD4 pixel colocalization ( Fig . 4C ) . We also adoptively transferred CD4+ T cells expressing Cyan Fluorescent Protein ( CFP ) into B6 . CD11c-YFP mice to evaluate the interaction of subcapsular RP DCs with CD4+ T cells during early Pc malaria . In naïve mice , most CFP+CD4+ cells made transient contacts with YFP+ cells ( Fig . 4D , S6 Video ) , and CFP+CD4+ cells were actively moving inside spleen ( Fig . 4E ) . At 24 h p . i . , CFP+CD4+ cells contacted YFP+ cells more stably ( Fig . 4D , S7 Video ) , as indicated by a decrease in CFP+CD4+ cell speed and an increase in arrest coefficient ( Fig . 4E ) . To investigate whether splenic DCs have a direct role in parasite clearance during pre-crisis , we analyzed the interactions between splenic DCs and iRBCs after five days of infection in vivo and ex vivo . This possibility was suggested by our data showing that , on day 5 p . i . , splenic DCs had an enhanced expression of the phagocytic receptor FcγRI ( S6A–S6B Fig . ) . Notably , we visualized many mCherry-Pc iRBCs inside the subcapsular RP , and YFP+ cells displayed intense phagocytic activity ( Fig . 5A , S8 Video ) . The presence of intense vacuolization in these DCs was also clear , and we observed some YFP+ cells ( containing iRBC remnants from previous internalization events ) phagocytizing mCherry-Pc iRBCs ( Fig . 5A , S9 Video ) . CIVM 3D animations confirmed the internalization of mCherry-Pc parasites by YFP+ cells ( Fig . 5B , S10 Video ) . This phenomenon was observed in 45% of the YFP+ cells ( Fig . 5C ) . At five days p . i . , YFP+ cells were activated and displayed higher cell volume and lower cell sphericity than those from recently infected mice ( Fig . 5D; S1 Table ) . On day 5 p . i . , the CD11c+ cells also expressed higher levels of MHC class II , CD80 and CD86 compared to those from naïve mice ( S6C–S6D Fig . ) . Immunofluorescence corroborated the significant role of splenic DCs in the widespread iRBC phagocytosis observed during pre-crisis . The percentages of CD11c pixels that colocalized with GFP pixels reached up to 40% in spleens from B6 mice on day 5 p . i . ( Fig . 6A and 6B ) . Flow cytometry confirmed that splenic DCs were able to phagocytize iRBCs during pre-crisis . When mature CTV-Pc iRBCs were i . v . injected into B6 mice on day 5 p . i . , approximately 4% of splenic DCs were CTV+ ( 1 . 4 × 105 CTV+CD11c+ cells/spleen ) ( Fig . 6C and 6D ) . Phagocytic activity was not restricted to a particular DC subtype , as a proportion of all subsets studied internalized iRBCs during pre-crisis ( S4A Fig . ) . However , CTV+ CD11b+CD11c+ and CD8+CD11c+ cell numbers were significantly higher per spleen than those of other DC subsets ( S4B Fig . ) . In addition , on day 5 p . i . , 10% of CD11c+ cells from mice infected with GFP-Pc iRBCs were GFP+ ( 4 × 105 CTV+CD11c+ cells/spleen ) ( Fig . 6E and 6F ) . Comparatively , we observed substantially higher activation and phagocytic activity both in vivo and ex vivo in the splenic DCs during pre-crisis ( S1 Table ) . Furthermore , a significantly higher frequency of iRBC uptake was detected using CIVM in comparison with flow cytometry . Notably , flow cytometry analysis of splenic YFP+ cells from B6 . CD11c-YFP mice during pre-crisis showed a sharp reduction in the percentages of F4/80+ cells so that the great majority of the YFP+ cell population presented with a classical DC phenotype ( S5 Fig . ) . Moreover , a large fraction of CD11c+YFP+ cells in these mice expressed higher levels of MHC class II molecules in comparison to those in recently infected B6 . CD11c-YFP mice . This observation was confirmed by CIVM , which revealed a reduction of F4/80+YFP+ cells in the subcapsular RP of B6 . CD11c-YFP mice on day 5 p . i . ( Fig . 5E and 5F , S11 Video ) . Due to the incremental number of CD11c+ cells in the YFP+ cell population , almost all of the phagocytic activity of YFP+ cells was imputed to DCs during pre-crisis , as observed by CIVM ( Fig . 5G and 5H ) and by flow cytometry ( Fig . 6G , 6H and 6I ) . During the crisis phase of acute Pc malaria , profound modifications in the splenic architecture occur , resulting in RP closure [36] . Therefore , we extended our study into this phase of the disease . CIVM revealed only occasional mCherry-Pc iRBCs trapped by subcapsular RP YFP+ cells in B6 . CD11c-YFP mice on day 8 p . i . ( Fig . 7A and 7B , S12 Video ) , and yellow spots of merged mCherry/YFP-3D signal were infrequent ( Fig . 7C ) . At that same time point , YFP+ cell volumes were smaller than during pre-crisis ( Fig . 7D , S1 Table ) . YFP+ cell sphericity was reduced in mice on days 5 and 8 p . i . compared with naïve mice ( Fig . 7D , S1 Table ) . Flow cytometry also revealed poor phagocytosis by splenic DCs , a process that was investigated both when mice were re-infected i . v . with mature CTV-Pc iRBCs and when mice were i . p . infected with GFP-Pc iRBCs ( Fig . 7E , 7F , 7G and 7H ) . These data indicate that splenic DCs could be primarily involved in antigen presentation rather than in phagocytosis during crisis , as CD11c+ cells expressed high levels of MHC class II and CD80 on day 8 p . i . ( S6C–S6D Fig . ) .
The depletion of phagocytes in vivo allowed us to clearly demonstrate the key role of DCs in the protection against experimental blood-stage malaria . Abundant CD11c expression is a well-known marker for DCs , which are primary targets of DTx treatment in B6 . CD11c-DTR mice [38] . Nevertheless , MZ macrophages are also depleted in DTx-treated B6 . CD11c-DTR mice due to ectopic expression of the DTx receptor transgene [39] . The role of DCs was established in our study by comparing the disease progression in DTx-treated B6 . CD11c-DTR mice and in B6 mice treated with a low dose of ClLip , which selectively depletes MZ macrophages within splenic phagocytes [39] , [40] . The significant contribution of DCs in the control of Pc malaria was suggested by data showing the worsening of the disease in DTx-treated B6 . CD11c-DTR mice , while the elimination of MZ macrophages by the ClLip treatment did not alter the course of infection in B6 mice . Our data also showed that splenic DCs are required for CD4+ T cell proliferation and IFN-γ production during Pc infection . The complete abrogation of these responses in DTx-treated B6 . CD11c-DTR mice , but not in ClLip-treated B6 mice , demonstrated that other splenic phagocytes such as MZ and RP macrophages did not replace DCs in the initiation of CD4+ T cell responses to Pc infection . Our first evidence suggesting that DCs could directly contribute to parasite clearance was the effect of DC depletion on the increase of parasitemia and the reduction of body weight during the first days of blood-stage Pc and Py malaria . DCs were also required to control the early parasitemia following infection with Pb sporozoites . The early protective role of DCs could not be completely attributed to the need for these cells to activate T cells , which take longer to produce IFN-γ and induce antibody secretion during experimental malaria . The splenocytes obtained four and five days after Pc infection still require further stimulation with iRBCs in vitro to differentiate into effector cells [41] , [42] , while the ex vivo production of IFN-γ and antibodies coincides with the drop of parasitemia a week after infection [42] , [43] . Using in vivo and ex vivo approaches , we unequivocally demonstrated here that the subcapsular RP DCs recognize and phagocytize mature iRBCs during the first encounter and pre-crisis , while spleen closure coincides with limited Pc phagocytosis by DCs during crisis . Although the splenic DCs are thought to be a major DC population in intimate contact with the bloodstream , these cells may act together with other DCs outside the spleen to clear Plasmodium parasites . This idea is supported by studies in splenectomized mice showing that other reticuloendothelial organs , such as the liver , effectively substitute for the phagocytic functions of the spleen in protecting against Pc malaria [22] , [44] . In fact , hepatic CD11c+ DCs are also capable of internalizing iRBCs in the liver sinusoids during acute Pc infection [45] . CIVM allowed us to visualize the interaction between subcapsular RP DCs and iRBCs in great detail . In naïve mice , these cells actively extended protrusions and dendrites , as previously shown [26] . Soon after infection , we observed iRBCs being trapped by DCs that had a non-activated phenotype . The majority of these cells showed a classical DC phenotype , but a proportion of them exhibited strong labeling for F4/80 , a marker of RP macrophages that is also expressed by a subset of DCs in the skin [46] . Another study reporting a similar observation concluded that , based on their dendritic morphology , subcapsular RP F4/80+YFP+ cells represent a subset of peripheral tissue DCs [26] . Although we did not visualize phagocytosis of iRBCs in recently infected mice , the detection of Pc remnants inside subcapsular RP DCs suggests that iRBC uptake had occurred . In fact , parasite antigen presentation is likely to occur soon after Pc infection , similar to the process observed during L . monocytogenes infection [47] . During the first day p . i . , subcapsular RP DCs displayed high motility and made stable contacts with CD4+ T cells . DCs also migrated rapidly to T cell-rich areas following Pc infection , a process that might involve chemokine signaling as suggested by studies in CCR7-knockout mice [48] . Here , for the first time , we observed the phagocytosis of iRBCs during pre-crisis in vivo . This occurred in a large number of subcapsular RP DCs , such that up to half of this population presented with Pc remnants . The great majority of these cells had a classical DC phenotype , which was characterized by negative staining for F4/80 and high expression of both MHC class II and costimulatory molecules . It is notable that these cells displayed an activated phenotype . Even if most subcapsular RP DCs during pre-crisis are immature cells that recently migrated to the spleen [49] , it is expected that DC activation leads to their maturation and consequent blockade of phagocytic activity , allowing the cellular machinery to be restructured for antigen presentation [28] . In agreement with our data , a previous report determined that the peak of in vitro iRBC uptake by splenic DCs occurred at five days p . i . , in parallel with the increase in the expression of MHC class II and costimulatory molecules [34] . In both studies , the phagocytic activity was not restricted to a particular DC subset . Our ex vivo data implicate CD11b+ and CD8+ DCs in most of the parasite clearance imputed to splenic DCs in mice both soon after infection and at the pre-crisis phase . Consistent with the immune response to acute Pc malaria , the CD11b+ and CD8+ DC subsets are known to be specifically involved in antigen presentation to CD4+ T cells and IL-12 production , respectively [50] , [51] . Furthermore , both subsets of DCs are able to induce IFN-γ production by parasite-specific T cells during Pc infection [29] . Another important observation during pre-crisis was a sharp decline in the population of F4/80+YFP+ cells , a phenomenon that also occurred to splenic F4/80+ macrophages after the parasitemia peak ( unpublished data ) . Because DCs have a higher turnover than F4/80+ macrophages [47] , a possible explanation for our results is that a proportion of these phagocytes died after ingesting Pc parasites and only DCs were rapidly replaced . This process would substitute F4/80+ macrophages , a resident RP population that is primarily required to maintain tissue homeostasis [52] , to inflammatory phagocytes . An alternative explanation is the down-regulation of the F4/80 molecule due to macrophage activation as reported during mycobacterial infection [53] . The F4/80+YFP+ cells could also have migrated to other locations such as the splenic T cell-rich areas . During crisis , the down-regulation of the phagocytic function of splenic DCs coincided with the period of spleen closure . This was demonstrated here by in vivo images showing a few iRBCs in the subcapsular RP at eight days p . i . , when parasitemias were even higher than at five days p . i . . The decline in iRBC uptake was also associated with the maximum expression of MHC class II and CD80 molecules by splenic DCs , which indicates that complete DC maturation was only achieved during crisis . This idea is corroborated by a previous study that reported a decrease to baseline levels of the in vitro uptake of the iRBCs by splenic DCs at day 8 p . i . [34] . Thus , in addition to spleen closure and the subsequent blockade of iRBC entry inside the RP , splenic DCs seem to lose the ability to phagocytize parasites , while concomitantly increasing their ability to present cognate antigens . This is an interesting observation because , during crisis , most of the lymphocytes that are activated during early Pc infection undergo apoptosis [54] , [55] . Thus , it is possible that mature DCs are required to expand and differentiate the few remaining T cells , giving rise to the memory response to malaria [56] , [57] . The quantification of iRBC phagocytosis ex vivo by flow cytometry yielded substantially lower percentages of Pc+ DCs compared with in vivo data obtained by CIVM . This discrepancy may result from differences in the fluorescence detection thresholds of CIVM and flow cytometry , the DC subpopulations examined by these techniques ( subcapsular RP DCs or total splenic DCs , respectively ) or the fluorochrome labeling of the iRBCs ( mCherry , GFP , CTV or CMTPX ) . Another possible explanation for the low detection of iRBC uptake by flow cytometry is the rapid iRBC degradation or fluorochrome quenching [8] , such that Pc remnants were only identified inside DCs shortly after phagocytosis . Previously , low frequencies of iRBC uptake were also detected by flow cytometry in migrating monocytes [8] , [27] . Immunofluorescence confirmed that splenic DCs , particularly those localized inside the RP and MZ , play a major role in the clearance of iRBCs during acute Pc infection . Although this technique did not efficiently discriminate single cells , the percentages of CD11c-GFP pixel co-localization were comparable to those of Pc+ DCs obtained by CIVM . The in vivo approaches used in this study indicate that , beyond the classical role of DCs in antigen presentation , these cells also contribute to the direct elimination of iRBCs during acute Plasmodium infection . For several days after Pc infection , subcapsular RP DCs were highly efficient in the recognition and capture of iRBCs . Complete DC maturation appeared to be achieved only during crisis when restructuring of the spleen might facilitate the development of the acquired immunity . Taking into account the specifics of different parasite-host interactions , we speculate whether our findings in mouse models could be applied to human malaria . The adhesion of P . falciparum iRBCs to human monocyte-derived DCs through the scavenger receptor CD36 has been shown to inhibit DC maturation and subsequently reduce their capacity to activate T cells [58] . This observation was interpreted as the impairment of the DC function during P . falciparum infection . However , our data showing the induction of FcγRI in splenic DCs during pre-crisis open the possibility that recognition of opsonized iRBCs through this receptor can overcome the down-regulatory activity of CD36 signaling . Thus , the opposite effects of malaria on DC function could be related to the different activation profiles of DCs , which are greatly influenced by the surrounding tissue microenvironment , rather than other factors previously discussed such as different species and strains of hosts and parasites [59] . Together , our data add novel information to this area of immunology and demonstrate that in vivo imaging may help to unravel the mechanisms underlying protective immunity against malaria .
Six- to eight-week-old B6 , B6 . CD11c-DTR [28] , B6 . CFP [60] and B6 . CD11c-YFP mice [61] were bred under specific pathogen-free conditions at the Animal Facilities of Instituto Gulbenkian de Ciência ( IGC ) , Instituto de Ciências Biomédicas at the Universidade de São Paulo ( ICB-USP ) or Institut de Transgénose Orléans-Villejuif . Pc ( AS strain ) , Py ( XL strain ) and mCherry-Pc were maintained previously as described [62] , [63] . GFP-Pc parasites were selected by treatment with pyrimethamine ( Sigma-Aldrich , USA ) [64] . The Instituto de Medicina Molecular at the Universidade de Lisboa provided Anopheles stephensi mosquitoes infected with Pb ( ANKA strain ) . Mice were infected intraperitoneally ( i . p . ) with 1 × 106 iRBCs ( blood from infected mice ) , and intravenously ( i . v . ) with 1 × 108 iRBCs or 1 × 103 sporozoites . Purified iRBCs were used where specified . The iRBCs were obtained during a period of the circadian cycle in which mature stages predominated ( >95% late trophozoites/schizonts ) . All procedures were in accordance with the national regulations of Conselho Nacional de Saúde and Colégio Brasileiro em Experimentação Animal ( COBEA ) and Federation of European Laboratory Animal Science Associations ( FELASA ) . The protocols were approved by the Comissão de Ética no Uso de Animais ( CEUA ) of ICB-USP , São Paulo , Brazil under permit numbers 0036/2007 and 0174/2011 , and by FELASA under permit number AO10/2010 . To deplete CD11c+ cells , B6 . CD11c-DTR mice were injected i . p . with a single dose of 2 ng/g body weight of DTx ( Sigma-Aldrich ) 24 h before iRBC infection or 48 h after sporozoite infection . This dose is half of the one previously established to deplete CD11c+ cells [65] and it was used to reduce drug toxicity . To deplete MARCO+/MOMA1+ cells , B6 mice were injected i . v . with 8 . 5 µg/g body weight of ClLip 24 h before infection [40] . Phosphate buffered saline ( PBS ) or PBS liposomes ( PBSLip ) were injected as controls . The procedures to obtain ClLip and PBSLip were described elsewhere [66] . Blood from infected B6 mice was resuspended in 1 ml PBS , pipetted over 5 ml of 74% Percoll ( GE Healthcare , USA ) and centrifuged ( 2500 x g , acceleration/break 5/0 ) for 30 min at room temperature ( RT ) . The top cell layers were collected and washed with complete RPMI 1640 medium ( supplemented with 10% heat-inactivated fetal calf serum , 100 U/ml penicillin , 100 µg/ml streptomycin , 50 µM 2-mercaptoethanol , 2 mM L-glutamine and 1 mM sodium pyruvate; Life Technologies , USA ) . Purified iRBCs ( >95% purity ) were stained with CTV or CMTPX , following the manufacturer’s instructions ( Life Technologies ) . B6 . CD11c-YFP mice infected with mCherry-Pc iRBCs were deeply anesthetized i . p . with 55 ng/g body weight of ketamine ( Imalgene 1000 , Merial , USA ) and 0 . 85 ng/g body weight of xylazine ( Rompun 2% , Bayer , Germany ) . Spleens were externalized by a 1 cm incision just below the ribcage . Mice were placed above a metal plate with a coverslip and immobilized without disrupting the vasculature or splenic connective tissue . Live imaging was carried out with an Eclipse Ti microscope ( Nikon Instruments Inc . , Japan ) equipped with an Andor Revolution XD system ( Andor Technology , UK ) , a Yokogawa CSU-X1 spinning disk unit ( Andor Technology ) , a 20x PLAN APO VC objective ( Nikon Instruments Inc . ) and a 1 . 5x auxiliary magnification system ( Nikon Instruments Inc . ) . Data were processed with MicroManager 1 . 2 ( General Public License , NIH , USA ) . For each movie , 28 µm Z-sections with 4 µm Z-steps were acquired for 30 min . Imaris X64 7 . 0 . 0 . ( Andor Technology ) was used to edit images and to determine the percentage of mCherry+YFP+ cells , as well as the CD11c+ cell volume and sphericity . In other cases , B6 . CD11c-YFP mice were adoptively transferred with 5 × 106 splenic CD4+ T cells from B6 . CFP mice ( purified by FACS sorting using a FACSAria device; BD Biosciences ) . These mice were infected as described above and processed 24 h later . Imaris was used to edit images and to determine CD11c+ cell speed and displacement , as well as the coefficients of CFP+CD4+ T cell speed and arrest . B6 . CD11c-YFP mice infected with CMTPX-Pc iRBCs were injected i . v . with PE-conjugated anti-F4/80 mAbs ( 200 ng/g body weight ) and deeply anesthetized to externalize the spleen as described above . Live imaging was carried out with a Zeiss LSM 780-NLO confocal microscope ( Zeiss , Germany ) . Data were processed with Zen 2012 software ( Zeiss , Germany ) . In each movie , 28 µm Z-sections with 2 µm Z-steps were acquired for 30 min . Imaris was used to edit images and to determine the percentages of CMTPX+ cells . Mice were sacrificed and PBS-perfused to remove circulating iRBCs . Spleens were harvested , and the remaining RBCs were lysed with ACK lysis buffer . Splenocytes ( 1 × 106 ) were stained with fluorescent monoclonal antibodies ( mAbs ) against CD3 , CD4 , CD11c , CD69 , CD11b , CD80 , CD86 , I-Ab , B220 , CD36 , CD64 ( FcγRI ) , DX5 and Ter119 ( BD Biosciences , USA ) , F4/80 ( eBiosciences , USA ) , and MOMA-1 and MARCO ( Abcam , UK ) . Cells were analyzed by flow cytometry ( FACSCanto; BD Biosciences ) with FlowJo 9 . 5 . 3 . ( Tree Star Inc . , USA ) . Splenocytes ( 3 × 107 ) were resuspended in 1 ml PBS with 0 . 1% BSA ( bovine serum albumin; Sigma-Aldrich ) and stained with CFSE ( carboxyfluorescein succinimidyl ester; Life Technologies ) at a final concentration of 5 μM for 20 min at 37°C . Cells ( 1 × 106 ) were cultured in complete RPMI 1640 medium for 72 h at 37°C with 5% CO2 in the presence of iRBCs ( 3 × 106 ) . Cells were then stained with fluorescent mAbs against CD3 and CD4 , and proliferation was assessed by flow cytometry . IFN-γ was quantified in the supernatants using the OptEIA IFN-γ kit ( BD Biosciences ) . GFP-Pc iRBC-infected B6 mice were sacrificed and PBS-perfused . Spleens were removed and frozen in Tissue-Tek OCT ( Sakura Fineteck , Japan ) . Sections 8 µm thick were cut with a CM3050S Cryostat ( Leica , USA ) and fixed with 1% paraformaldehyde ( Alfa Aesar , USA ) for 30 min at RT . Sections were incubated with anti-CD16/CD32 mAb ( Fc block; BD Biosciences ) for 30 min followed by incubation in a humidified dark chamber with fluorescent mAbs against CD11c , CD19 , CD3 , CD4 ( BD Biosciences ) and MOMA-1 ( Abcam ) for 2 h at RT . Sections were then stained for 5 min with 0 . 5 μg/ml DAPI ( 4' , 6-diamidino-2-phenylindole; Sigma-Aldrich ) , washed with PBS and mounted with Fluoromount-G ( Southern Biotechnologies , USA ) . Images were acquired with a DMRA2 fluorescence microscope ( Leica ) and MetaMorph software ( Molecular Devices Inc . , USA ) . Image analysis was performed with Photoshop CS4 ( Adobe Inc . , USA ) . Percentages of CD11c-GFP/CD11c-CD4 pixel colocalization and of GFP pixel distribution in the spleen were calculated using FIJI for Windows 64-bit ( Colocalization threshold and Mixture Modeling Thresholding plugins , respectively; General Public License , NIH , USA ) . Results were analyzed with Prism 5 software ( Graph Pad ) using ANOVA or Student’s t-tests . The existence of a normal distribution was confirmed using the Kolmogorov-Smirnov test . Differences were considered statistically significant at p < 0 . 05 .
|
Malaria is a significant health issue , particularly in the tropical and subtropical regions of the world . The red pulp ( RP ) of the spleen is a major site for the control of blood-borne infections such as malaria . Macrophages and dendritic cells ( DCs ) form a complex phagocyte network inside the splenic RP . DCs are usually thought of as highly efficient antigen-presenting cells that play an essential role in the activation of adaptive immunity . However , the direct role of DCs in the clearance of pathogens is still unclear . To clarify these issues , we took advantage of in vivo experimental approaches that enabled us to deplete or visualize DCs . The depletion of phagocytes demonstrated that DCs are key participants in the protection against blood stages of experimental malaria . Using confocal intravital microscopy , we observed that splenic RP DCs efficiently recognized and phagocytized infected erythrocytes during acute infection . We also showed that splenic DCs were crucial for the CD4+ T cell response to infection , but full DC maturation was achieved only after the peak of parasitemia . This study help to elucidate the protective mechanisms against Plasmodium parasites , and it shows that in vivo imaging is a reliable tool to evaluate iRBC phagocytosis during experimental malaria .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
In Vivo Approaches Reveal a Key Role for DCs in CD4+ T Cell Activation and Parasite Clearance during the Acute Phase of Experimental Blood-Stage Malaria
|
Organisms in the wild develop with varying food availability . During periods of nutritional scarcity , development may slow or arrest until conditions improve . The ability to modulate developmental programs in response to poor nutritional conditions requires a means of sensing the changing nutritional environment and limiting tissue growth . The mechanisms by which organisms accomplish this adaptation are not well understood . We sought to study this question by examining the effects of nutrient deprivation on Caenorhabditis elegans development during the late larval stages , L3 and L4 , a period of extensive tissue growth and morphogenesis . By removing animals from food at different times , we show here that specific checkpoints exist in the early L3 and early L4 stages that systemically arrest the development of diverse tissues and cellular processes . These checkpoints occur once in each larval stage after molting and prior to initiation of the subsequent molting cycle . DAF-2 , the insulin/insulin-like growth factor receptor , regulates passage through the L3 and L4 checkpoints in response to nutrition . The FOXO transcription factor DAF-16 , a major target of insulin-like signaling , functions cell-nonautonomously in the hypodermis ( skin ) to arrest developmental upon nutrient removal . The effects of DAF-16 on progression through the L3 and L4 stages are mediated by DAF-9 , a cytochrome P450 ortholog involved in the production of C . elegans steroid hormones . Our results identify a novel mode of C . elegans growth in which development progresses from one checkpoint to the next . At each checkpoint , nutritional conditions determine whether animals remain arrested or continue development to the next checkpoint .
The development of multicellular organisms requires the coordinated differentiation and morphogenesis of multiple cell types that interact to form functional tissues and organs . In favorable environmental conditions , development proceeds in a largely stereotyped pattern . When faced with adverse conditions , tissue growth may slow or arrest until the environment improves [1]–[3] . The most critical environmental factor that regulates development is nutrient availability . Organisms can modulate growth programs in response to changing nutritional conditions [4] , although the mechanisms through which organisms sense changes in nutrient availability and alter diverse cellular processes in a coordinated manner are incompletely understood . The nematode Caenorhabditis elegans is a powerful model for understanding the effects of nutrition on development due to its short life cycle ( 3–4 days from embryo to adult ) , simple cellular make-up , and highly stereotyped development . The postembryonic development of C . elegans entails passage through four larval stages ( L1–L4 ) that are separated by molts , before reaching reproductive adulthood . Two alternative pathways of development exist in C . elegans: continuous passage through the four larval stages , or entry into an L3 dauer stage , a growth-arrested state characterized by altered body morphology , elevated stress resistance , and prolonged survival [3] . Entry into dauer is initiated late in the L1 stage in response to unfavorable environmental conditions , in particular high population density , high temperature , and reduced nutrient availability [5] . Additional points of arrest in response to poor nutritional conditions have been identified early in the C . elegans life cycle and in adults . Animals that hatch in the absence of food undergo L1 arrest [6] , [7] , and animals reared from hatching on a limited supply of heat-killed bacteria arrest in the L2 stage [8] . Finally , adult C . elegans arrest embryo production and shrink their germlines following removal of food [9] , [10] . Studies on dauer and L1 arrest have revealed critical roles for the insulin/insulin-like growth factor ( IGF ) signaling pathway in sensing the nutritional environment and regulating entry into arrest [7] , [11] , [12] . In C . elegans , insulin-like peptides are generated during feeding and signal through DAF-2 , the insulin/IGF receptor . Activation of DAF-2 leads to the phosphorylation and cytoplasmic sequestration of DAF-16 , a forkhead box O ( FOXO ) transcription factor . During conditions of low nutrition , the DAF-2-mediated phosphorylation of DAF-16 is reduced , allowing DAF-16 to enter the nucleus and transcriptionally regulate genes implicated in developmental arrest [7] , [13]–[16] . Mutant animals with reduced daf-2 function may arrest in the L1 stage or form dauers constitutively ( Daf-c phenotype ) [11] , whereas daf-16 null mutants continue development past the wild type timing of L1 arrest and are defective in dauer formation ( Daf-d phenotype ) [2] , [12] . In worms , insects , and mammals , insulin-like signaling affects the production of steroid hormones , lipophilic molecules that bind to nuclear hormone receptors and induce cellular responses [17] . In C . elegans , bypassing dauer formation requires the bile-acid like steroid hormone dafachronic acid ( DA ) [18] . DAF-9 , a cytochrome P450 ortholog , is required for the production of DAs , and daf-9 null animals are Daf-c [19]–[21] . The effects of DAF-9 on dauer formation are mediated by DAF-12 , a nuclear hormone receptor that binds DAs [18] . The steroid hormone pathway functions downstream of insulin-like signaling during dauer formation , as daf-12 Daf-d alleles suppress the Daf-c phenotype of daf-2 mutants [11] . Despite extensive work on dauer and other arrests , questions remain about the response of C . elegans to nutritional scarcity . Among these are whether the arrests in L1 , L2 , dauer , and the adult represent unique periods of the life cycle during which animals are sensitive to their nutritional environment , or if arrest can also occur at other times . It is also not known whether bypassing the opportunity to form a dauer leads to continuous development to adulthood , or whether additional opportunities exist to arrest development when faced with nutrient deprivation . Finally , the mechanisms through which numerous tissues and cellular processes are able to coordinately arrest in response to nutrient withdrawal are not well understood . We sought to address these questions by examining the response of C . elegans to nutrient deprivation during the late larval stages ( L3 and L4 ) , after the opportunity to form a dauer has been passed . Several tissues that contribute to the reproductive system undergo differentiation , growth , and morphogenesis during the L3 and L4 stages , making this period an ideal time to determine how ongoing developmental processes are affected by nutrient deprivation . By removing animals from food at different times , we show that specific checkpoints exist in the early part of the L3 and L4 stages that restrict progression through the larval stage and systemically arrest the development of diverse tissues and cellular processes . Insulin-like signaling regulates the response to nutrient deprivation in the L3 and L4 stages through cell-nonautonomous DAF-16 activity in the hypodermis ( skin ) , and functions to suppress DAF-9–mediated signaling activity . Our results identify a mode of metazoan growth in which development proceeds from checkpoint to checkpoint . At these checkpoints , nutritional conditions determine whether animals remain in an arrested state or continue development to the next checkpoint .
To study the effects of nutrient deprivation on tissue development during the late larval stages , we first focused on the hermaphrodite vulva , which develops through a stereotyped pattern of cell specification , cell division , and morphogenesis during the L3 and L4 stages ( Fig . 1A ) [22] . The vulva derives from three epidermal cells , P5 . p–P7 . p , which are specified early in the L3 stage to either the 1° vulval precursor cell ( VPC ) fate ( P6 . p ) or the 2° VPC fate ( P5 . p and P7 . p ) ( Fig . 1A ) . The VPCs undergo three rounds of cell division in the L3 stage to generate 22 cells , which differentiate into seven vulval subtypes , vulA–vulF . In the L4 stage , the 22 vulval cells undergo morphogenetic processes that include invagination , migration , and cell-cell fusion ( Fig . 1A ) [22] . Proper development of the vulva requires the uterine anchor cell ( AC ) , which invades in the mid L3 stage across basement membranes separating the uterine and vulval epithelia to form a connection between the tissues [23] . The AC remains at the vulval apex after invasion until fusing with surrounding uterine cells in the mid L4 stage ( Fig . 1A ) . We examined the effects of nutrient deprivation on vulval development by growing a synchronized population to late in the L2 stage , prior to the onset of vulval formation , and removing animals from food ( Fig . 1B ) . Part of the population was returned to food to serve as controls , with the remainder maintained in M9 , a buffer lacking a carbon source . In addition to the vulva , we also assessed the onset of molting ( observable by cuticle covering the mouth; see Fig . 1A , bottom left ) , which serves as a marker for the transition between larval stages . Results of the experiment are depicted graphically in Fig . 1B; raw data and results of triplicate assays are in Fig . S1 . The control group that was returned to food progressed through the stages of vulval development with the predicted timing . The group that remained deprived of food molted into the L3 stage and uniformly arrested prior to the first VPC divisions . No VPC divisions were observed after 10 days in the absence of food ( Fig . 1B ) . Arrested animals were phenotypically distinct from dauer larvae , which arrest after molting into a specialized L3 dauer stage ( Fig . S2 ) . When animals were returned to food after 8 d , 97 . 5% of the population ( n = 200 ) continued development to adulthood , demonstrating that animals retain the capacity to resume development upon re-introduction of food . The median survival of animals under the experimental conditions used was 11 . 7±1 . 2 d ( n = 3 trials ) . In C . elegans , removal of the germline either through ablation or genetic mutation extends lifespan and maintains adult somatic tissues for a longer duration in a youthful state [24]–[26] . This suggests the possibility of a soma-germline tradeoff in which resources are allocated away from germ cells to somatic tissues . We asked if the absence of a germline could promote the continued development of somatic tissues by growing glp-1 ( e2144 ) mutants , which do not proliferate germline progenitor cells when reared at 25°C , to late in the L2 stage and removing them from food . We found no difference in the timing of arrest , as all glp-1 ( e2144 ) animals ( n = 100 ) arrested prior to VPC divisions . These results demonstrate that the absence of the germline does not alter the timing of somatic tissue arrest in response to nutrient removal . In addition to cell divisions , fate specification of the vulval cells was also examined in L3-arrested animals . Vulval fates are specified between the late L2 and early L3 stages , when an inductive LIN-3/EGF signal from the AC and lateral LIN-12/NOTCH signaling between VPCs combine to specify the 1° fate in P6 . p and 2° fates in P5 . p and P7 . p [27] , [28] . To determine the state of VPC specification in arrested animals , we examined a marker of 1° fate , egl-17>GFP [29] , and a marker of 2° fate , lip-1>NLS-GFP [30] ( see Materials and Methods for description of transgene nomenclature ) . All arrested animals expressed egl-17>GFP exclusively in P6 . p , and 93% of animals expressed lip-1>NLS-GFP at elevated levels in P5 . p and P7 . p ( n = 30 per assay; Fig . 1C ) , demonstrating that arrest occurred after 1° and 2° VPC specification . This contrasts with dauer larvae , which are not stably specified to a VPC fate [31] , [32] . Coupled with the absence of VPC divisions , these results suggest that , when removed from food late in the L2 stage , vulval development arrests early in the L3 stage in a manner that is distinct from dauer arrest . The uniform arrest of vulval development early in the L3 stage suggested that a specific developmental checkpoint existed at this time . To determine if this was the case , we asked whether vulval development could arrest at later times in the L3 stage . A synchronized population was grown for 28 h to the mid L3 stage and removed from food . At the time of food removal , 84% had undergone one VPC division , indicating that they had bypassed the L3 arrest point ( Fig . 2A; Fig . S1 ) . Animals removed from food continued through the L3 stage , molted into L4 , and arrested in L4 after completion of VPC divisions ( Fig . 2A ) . After 48 h in the absence of food , 94% of the population was arrested after VPC divisions in the L4 stage . The remaining 6% of animals was arrested in the L3 stage prior to VPC divisions , and likely represent the youngest members of the population that failed to bypass early L3 arrest . No animals were identified at intermediate stages between the two arrest points , demonstrating that bypass of the L3 arrest point led to invariant progression to the L4 arrest point ( Fig . 2A ) . We examined the effect of the germline on L4 arrest by removing glp-1 ( e2144 ) mutants grown at 25°C from food in the mid L3 stage , and found that all animals arrested in early L4 similar to wild type ( n = 100 ) . The median survival of populations removed in the mid L3 stage was 11 . 0 d ( n = 3 trials ) . All L4-arrested animals completed vulval cell divisions ( n = 30 ) , suggesting that arrest in vulval formation could occur at a precise developmental time rather than in a variable manner . To test this notion , we first examined the reporter gene egl-17>GFP , which is expressed in 1° VPC progeny early in the L4 stage and shifts to 2° VPC progeny in mid L4 ( Fig . 1A ) . Expression of egl-17>GFP was exclusively in 1° VPC progeny in arrested animals ( Fig . 2B ) , supporting the hypothesis of a precise timing of arrest early in the L4 stage . A second marker for L4 stage timing in vulval development is cell-cell fusions . Fusions occur between homotypic cells ( i . e . , vulA with vulA ) , starting with vulA cells shortly after terminal cell divisions and continuing two hours later with vulC cells ( Fig . S3 ) [33] . Examination of a strain expressing GFP-tagged AJM-1 , an apical-membrane–localized protein that delineates the boundaries of vulval cells [33] , [34] , showed that all L4-arrested animals had undergone vulA fusions but not vulC fusions ( Fig . S3 ) . Importantly , the same timing of arrest between vulA and vulC fusions occurred in 97% of the population when animals were grown for an additional four hours prior to removal from food ( n = 30 per assay; Fig . S3 ) , demonstrating that the timing of arrest in vulval development in the L4 stage is largely independent of feeding duration . Based on the nutrient removal experiments , we conclude that specific checkpoints exist early in the L3 and L4 larval stages that arrest vulval development at precise developmental times . Only a single checkpoint on vulval development was identified in the L3 stage , and we wanted to determine whether this was also the case with the L4 stage . Animals were grown on food to the early-to-mid L4 stage and developmental progression examined following food removal ( Fig . 2C; Fig . S1 ) . After 48 h in the absence of food , 96% of the population had progressed into adulthood , as evidenced by eversion of the vulva ( Fig . 2D ) , with the remaining animals arrested early in the L4 stage , and no animals at intermediate times ( Fig . 2C ) . Arrest occurred in nearly all adult animals ( 99/100 ) prior to oogenesis . Taken together , the nutrient removal assays show that arrest in C . elegans vulval development occurs only at precise checkpoints early in the L3 and L4 stages , and that passage through one checkpoint leads invariantly to progression through the larval stage to the next checkpoint . There are two alternative possibilities for the timing of arrests observed in vulval formation in the L3 and L4 stages . The first is of a tissue-autonomous program in which arrests occur only at specific times in the developmental program , either prior to cell divisions ( early L3 ) or after cell divisions ( early L4 ) . The second is of a global timing mechanism that arrests vulval development at precise times early in each larval stage . To determine which of these was correct , we examined hbl-1 ( ve18 ) mutant animals , which have precocious VPC divisions that occur as early as the late L2 stage ( Fig . 3A ) [35] . We hypothesized that if the vulval cells were regulated by an autonomous program , then shifting the time of cell divisions relative to the L3 larval stage would not affect the all-or-none pattern of cell divisions . If instead a global timing mechanism directed the arrest of vulval development , then cell divisions would be predicted to arrest upon reaching the L3 larval stage checkpoint . Results of the experiment show that after removal from food late in the L2 stage , P6 . p divisions continued but stopped prior to completion ( Fig . 3B–C ) . Only 43% of the population was arrested either prior to or after cell divisions , with the remainder at intermediate stages of division ( Fig . 3B ) . These results support the idea of a global timing mechanism that acts on vulval development to arrest it at specific times early in the larval stage . The experiments on vulval development suggested that checkpoints exist at particular points in the larval stage . We wanted to explore this question in more detail by examining progression through the larval stage in the absence of food . Each C . elegans larval stage comprises a period of foraging for food that lasts for several hours , followed by an approximately two-hour period of lethargus during which pharyngeal pumping stops and animals do not feed . At the end of lethargus , C . elegans undergo molting , the detachment ( apolysis ) and shedding ( ecdysis ) of the existing cuticle [36] . We first asked if animals removed from food during the period of foraging underwent lethargus , and found that both the onset and duration of lethargus were similar to a control population that was maintained on food . Further , animals exited lethargus and resumed pharyngeal pumping for at least 24 h after removal from food ( Fig . 4A ) . These results show that lethargus , a key feature of the larval stage , is maintained in the absence of food . We next examined how nutrient deprivation affected the molting cycle , the oscillatory pattern of gene expression and cuticle replacement that occurs in each larval stage . Cuticle components are synthesized starting in the mid-larval stage and deposited underneath the existing cuticle , which is shed at the end of the larval stage [37] , [38] . The timing of the checkpoints in the early part of larval stage suggested that arrest could occur after molting and prior to new cuticle synthesis . We first asked whether this was the case by examining the execution of the molt following removal of food . We found that all L3- and L4-arrested animals completed ecdysis ( Fig . 4B ) , although 17% of adult-arrested animals remained attached to the L4 cuticle after 48 h in the absence of food ( n = 100 animals per assay ) . It is possible that larger animals may not be able to fully shed cuticle in the absence of sufficient feeding . Despite this defect , animals were viable and resumed pharyngeal pumping , with the L4 cuticle remaining attached only in the tail region . These results demonstrate that molting is successfully executed in most instances upon passage through a checkpoint . We then explored the second part of our hypothesis that arrest occurred prior to new cuticle synthesis . To do this , we examined the expression pattern of mlt-10 , a gene required for proper execution of the molt [38] , [39] . mlt-10 mRNA increases in the mid-larval stage at the time of new cuticle synthesis , peaks during the molt , and declines upon completion of molting . A destabilized reporter gene , mlt-10>GFP-PEST , recapitulates this oscillatory mRNA expression pattern and serves as a marker for progression through the larval stage [38] , [39] . A population of mlt-10>GFP-PEST–expressing animals was removed from food late in the L2 stage and a portion of the population returned to food to serve as controls . The fed and nutrient-deprived groups were then examined for reporter gene expression over an 8 h period . Results show that expression levels were similar in the two groups as they molted and entered the L3 stage ( Fig . 4C ) . Approximately 4 h after molting , the control group increased gene expression , indicating initiation of the L3 molting cycle . The nutrient-deprived group failed to increase expression , however , demonstrating that it had arrested prior to initiation of the L3 molting cycle ( Fig . 4C ) . Similar results were observed when animals were removed from food late in the L3 stage ( data not shown ) . The loss of mlt-10>GFP-PEST expression was unlikely to be due to general transcriptional silencing during nutrient deprivation , as past research has shown that several transcriptional reporters similarly tagged with PEST motifs maintain expression during L1 arrest [40] . Collectively , these results demonstrate that C . elegans arrest development during the L3 and L4 stages at a specific point after molting and prior to new cuticle synthesis . Our results identified nutrient-sensitive developmental checkpoints in the early part of the L3 and L4 larval stages . We sought to determine the amount of feeding required to pass the checkpoints . To achieve the greatest degree of synchronization and most accurate measurement of timing , individual animals were isolated during ecdysis , the final 10–15 minutes of molting that precede foraging [36] . Animals undergoing ecdysis were either removed from food or allowed to feed for additional 30 min intervals ( Fig . S4 ) . Feeding for 30–60 min after ecdysis was required for most animals to pass the L3 and L4 checkpoints within 24 h after food removal , and 90 minutes of feeding resulted in all animals passing the checkpoints ( Fig . S4 ) . We conclude that a sufficient duration of feeding is required after molting to advance past the L3 and L4 stage checkpoints . The insulin-like signaling pathway is a key regulator of growth in response to nutrition [41] . We wanted to determine if insulin-like signaling regulated arrest in the L3 and L4 stages following nutrient removal . We first asked if daf-16 , a FOXO transcription factor that is a major target of insulin-like signaling and is required for the proper timing of L1 arrest and dauer formation [7] , [12] , played a role in L3 and L4 arrest . Animals with the null mutation daf-16 ( mu86 ) were removed from food late in the L2 stage , and the developmental stage assessed over time by examination of the vulva and molt . The pattern of growth by 8 h after food removal was similar to wild type: all animals had molted into L3 and 97% were in the early L3 stage ( Fig . 5A; raw data and results of replicate assays are in Fig . S5 ) . By 24 h after removal from food , however , 63% of the population had progressed past the L3 checkpoint ( compared with 0% of wild type animals removed from food at a similar time; Fig . 1B ) , ultimately arresting early in the L4 stage ( Fig . 5A ) . A second experiment was performed with daf-16 ( mu86 ) animals removed from food late in the L3 stage . Again , the absence of daf-16 caused animals to bypass arrest , with 72% of the population progressing to adulthood after 48 h ( Fig . 5B; Fig . S5 ) . The time in the larval stage at which daf-16 ( mu86 ) animals arrested was similar to wild type , based on the absence of VPC divisions in L3-arrested animals , the completion of divisions in L4-arrested animals , and no animals at intermediate stages of division ( n = 30; Fig . 5C ) . In the presence of food , DAF-16 activity is inhibited by a signaling pathway downstream of DAF-2 , the insulin/IGF receptor . We hypothesized that animals with reduced DAF-2 function would require a longer duration of feeding to inhibit DAF-16 activity and progress through the L3 and L4 larval stages . To test this hypothesis , we examined the L3 and L4 development of a temperature-sensitive daf-2 mutant , daf-2 ( e1370 ) , which is Daf-c at 25°C but develops to adulthood at 15°C [11] . Animals were grown at the permissive temperature of 15°C to the mid-L2 stage , bypassing the opportunity to form a dauer , and shifted to the restrictive temperature of 25°C for an additional 24 h feeding ( Fig . 5D ) . Following this regimen , 25% of the population was in the early L3 stage and 42% was in early L4 stage . In contrast , a control wild type population had progressed to the L4/adult molt or beyond ( Fig . 5D ) . The high proportion of the population in the early L3 and early L4 stages suggests that prolonged pausing at the checkpoints may be a factor in the delayed development of daf-2 ( e1370 ) animals . This delayed development required the presence of daf-16 , as daf-16 ( mu86 ) ; daf-2 ( e170 ) double mutant animals advanced through the L3 and L4 stages at a rate comparable to wild type ( Fig . 5D ) . Taken together , the results of the daf-16 and daf-2 experiments demonstrate a role for the insulin-like signaling pathway in regulating progression through the L3 and L4 developmental arrest checkpoints in response to nutritional conditions . Previous work has shown that DAF-16 functions cell-nonautonomously to regulate multiple physiological processes , including dauer formation , lifespan extension , germline proliferation , and metabolism [42]–[44] . We asked if DAF-16 similarly functioned cell-nonautonomously to regulate L3 and L4 arrest . Plasmids that contained daf-16 cDNA tagged at the N-terminus with GFP and expressed under the daf-16 or tissue-specific promoters [42] ( Fig . 6A ) were injected into daf-16 ( mu86 ) ; unc-119 ( ed4 ) double mutant animals along with an unc-119 rescue plasmid . Animals harboring an extrachromosomal array of the plasmids were identified by rescue of the unc-119 ( ed4 ) movement defect and validated by examination of GFP expression ( Fig . S6 ) . When expressed from the daf-16 promoter , GFP::DAF-16 protein localized to neurons , hypodermis , intestine , and body wall muscles ( Fig . S6 ) , and rescued the daf-16 ( mu86 ) phenotype , in which animals bypass L3 arrest and continue development to the L4 stage ( Fig . 6B ) . GFP::DAF-16 expressed from tissue-specific promoters for neurons ( unc-119 and unc-115 ) , muscle ( myo-3 ) , and intestine ( ges-1 ) [42] failed to rescue the daf-16 ( mu86 ) bypass phenotype . Only GFP::DAF-16 expression from a hypodermis-specific promoter ( col-12 ) significantly rescued the daf-16 ( mu86 ) phenotype ( Fig . 6B ) . Similar results were obtained in assays examining bypass of L4 arrest ( data not shown ) . The lower efficiency of rescue by col-12>GFP::DAF-16 compared to daf-16>GFP::DAF-16 could be due to reduced expression of the transgene following removal from food: col-12>GFP::DAF-16 expression decreased 71% following 2 d in the absence of food , whereas daf-16>GFP::DAF-16 expression increased more than twofold during this time ( Fig . S7 ) . As in other assays , vulval development was used as the primary marker for developmental stage . Animals that were rescued for the L3 bypass phenotype by daf-16>GFP::DAF-16 or col-12>GFP::DAF-16 did not have detectable GFP::DAF-16 expression in the vulva , consistent with cell-nonautonomous DAF-16 activity regulating L3 and L4 development . To complement these studies , we carried out tissue-specific RNAi of daf-16 . Reducing daf-16 specifically in the hypodermis reproduced the phenotype of systemic loss of daf-16 . Targeted reduction of daf-16 in the intestine or muscle did not alter sensitivity to the removal from food ( Fig . 6C ) . We were unable to reduce daf-16 specifically in neurons because of the lower sensitivity of this tissue to RNAi [45] , and the inability to directly target neurons by RNAi without off-target effects in the hypodermis [46] . Taken together , the results of daf-16 tissue-specific rescue and RNAi experiments suggest that the hypodermis is a key site of action for the insulin-like signaling pathway in responding to nutritional conditions during the L3 and L4 stages . Our results do not rule out the possibility that DAF-16 functions in other tissues to also regulate L3 and L4 development , either through modulation of hypodermal DAF-16 function or through independent pathways that synergize with hypodermal DAF-16 . Previous studies have shown that DAF-16 can function in multiple tissues to regulate dauer formation and metabolism [42] , [43] , and such a situation could also occur in regulating passage through the L3 and L4 larval stages . The ability of DAF-16 to affect tissue development cell-nonautonomously implicated the presence of pathways that signal systemically . One such candidate is DAF-9–mediated steroid hormone signaling , which is downstream of insulin-like signaling during dauer formation [11] , [19] , [21] . A key site of action for DAF-9 during larval development is the hypodermis [19] , [21] , [47] , suggesting that it could similarly function downstream of insulin-like signaling during the L3 and L4 stages . To test this possibility , we depleted daf-9 by dsRNA feeding in daf-16 ( mu86 ) animals , and assessed the response to nutrient removal in the L3 and L4 stages . We hypothesized that , if nuclear DAF-16 inhibited L3 and L4 stage progressions through inhibition of DAF-9–mediated steroid hormone signaling , then the bypass of arrest observed in daf-16 null animals would be suppressed by reduction of daf-9 . Consistent with this hypothesis , daf-9 dsRNA-fed daf-16 ( mu86 ) animals had a 2 . 6-fold reduction of bypassing L3 arrest and a 1 . 9-fold reduction of bypassing L4 arrest compared to empty vector controls ( Fig . 7A ) . These results support the idea that the insulin-like signaling pathway regulates DAF-9–mediated steroid hormone production during the L3 and L4 stages . Since DAF-9 appeared to be involved in generating hormonal signals that promoted progression through the L3 and L4 larval stages , we asked whether increasing the levels of DAF-9 would lead to bypass of arrest in a manner akin to daf-16 null animals . This was tested by examining the response to nutrient removal of a strain overexpressing functional daf-9::GFP ( dhIs64 ) [19] . When removed from food late in the L2 stage , daf-9–overexpressing animals bypassed arrest at high levels , with 90% of the population progressed beyond the early L3 stage after 24 h in the absence of food ( Fig . 7B; Fig . S5 ) . In contrast to the phenotype of daf-16 ( mu86 ) , which paused at the L3 checkpoint before bypassing it ( Fig . 5A ) , daf-9–overexpressing animals continued past the checkpoint with minimal pausing ( Fig . 7B ) . Further , whereas daf-16 ( mu86 ) bypassed only one arrest point ( Fig . 5A–B ) , a portion of the daf-9-overexpressing population advanced through both the L3 and L4 arrest points and reached adulthood ( Fig . 7B ) . Thus , the bypass of arrest caused by overexpression of daf-9 is more rapid and robust than that caused by loss of daf-16 . daf-9–overexpressing animals that progressed to adulthood were typically surrounded by undetached cuticle ( 41/50 animals ) ; in some cases both the L3 and L4 cuticles remained attached ( Fig . 7C ) . The inability to shed cuticle surrounding the mouth led to lethality in a portion of the population within 24 h of food removal ( Fig . 7C ) . In contrast , neither wild type nor daf-16 null animals showed such rapid death . These findings demonstrate that overexpression of daf-9 , which forces animals through larval stages in the absence of food , has deleterious effects on the execution of the molt . Our finding that daf-9 overexpression causes continued development in the absence of food were somewhat surprising since a previous study showed that hypodermal DAF-9::GFP expression is sharply reduced during nutrient deprivation [19] . We examined the expression of hypodermal DAF-9::GFP following removal from food late in the L2 stage , and indeed found a reduction in expression over time ( Fig . 7D ) . Expression persisted at low levels in most animals for at least 8 h after removal from food , however , and was visible in some animals even after 24 h ( Fig . 7D ) . These results suggest that , when expressed at elevated levels , enough DAF-9 protein remains in the hypodermis to promote continued larval stage progressions in the absence of food . It is also possible that the second site of DAF-9 expression during larval development , the two neuronal XXX cells , also contribute to the bypass of arrest , as expression is not reduced in those cells following food removal [19] . Collectively , our results offer evidence that DAF-9 promotes passage through the L3 and L4 developmental arrest checkpoints . DAF-9 is required for the synthesis of dafachronic acids ( DAs ) , steroid hormones that bind to the nuclear hormone receptor DAF-12 to promote bypass of dauer formation [18] . In the absence of DAs , DAF-12 regulates entry into dauer though its DNA-binding activity [48] . Because our results showed that genes that regulate dauer formation—daf-2 , daf-16 , and daf-9—also regulate later larval development , we asked if daf-12 similarly had a role in regulating progression through the L3 and L4 stages downstream of daf-9 . We first examined the response to nutrient removal of daf-12 ( rh61rh411 ) , a null mutant that has a Daf-d phenotype [48] . In contrast to daf-16 ( mu86 ) Daf-d mutants , which bypass L3 arrest over 60% of the time ( Fig . 5B ) , no daf-12 null mutants bypassed L3 arrest after 48 h in the absence of food ( n = 54 ) . We also performed epistasis experiments by generating daf-12 ( rh61rh411 ) ; daf-16 ( mu86 ) double mutant animals . The bypass phenotype of daf-16 ( mu86 ) was not suppressed by the loss of daf-12 function , as 71% of double mutant animals bypassed the L3 checkpoint after 48 h in the absence of food ( n = 68 ) , similar to the percentage of daf-16 ( mu86 ) animals that bypass arrest ( Fig . 5B ) . These results suggest that daf-16 functions independently of daf-12 in regulating L3 stage progression . We also asked whether the phenotype of daf-9 overexpression was suppressed by a null allele of daf-12 . When removed from food late in the L2 stage , DAF-9::GFP; daf-12 ( rh61rh411 ) animals still bypassed arrest to a high degree: 87% of the population had bypassed L3 arrest by 24 h ( n = 61 ) , similar to the 90% observed in DAF-9::GFP animals ( Fig . 7B ) . These results provide evidence that DAF-12 does not regulate L3 and L4 stage progressions , and that DAF-9 promotes bypass of the L3 and L4 checkpoints through a different downstream effector . Our experiments with vulval formation and the molting cycle showed that checkpoints are present early in the L3 and L4 stages that limit continued development . We took advantage of the fact that several additional tissues undergo developmental processes in the L3 and L4 stages to determine if other tissues are also arrested at the checkpoints . We first examined the uterine AC , which becomes polarized early in the L3 stage , when F-actin and actin regulators localize to the basal , invasive cell membrane [49] . The AC breaches the basement membrane in the mid L3 stage , before fusing with the surrounding uterine cells in the mid L4 stage ( Fig . 1A ) . Examination of a marker of AC polarization , the F-actin probe cdh-3>mCherry::moesinABD , showed that animals removed from food late in the L2 stage arrested early in the L3 stage with polarized ACs , but in no instance did invasion occur ( n = 100; Fig . 8A ) . When removed from food in the mid L3 stage , AC invasion occurred in all animals , yet in no instance did the AC fuse with the surrounding uterine cells ( n = 100; Fig . 2B ) . These results suggest that the developmental program of the AC , similar to that the vulval cells and molting cycle , arrests at the early L3 and early L4 checkpoints . We next examined the two sex myoblast ( SM ) cells , which divide three times between the mid L3 and early L4 stages , followed by short-range migrations of terminally divided progeny cells in the L4 stage ( Fig . 8B ) . When animals were removed from food late in the L2 stage , no SM cell divisions were observed after 48 h , indicative of arrest early in the L3 stage . When removed from food in the mid L3 stage , SM cell divisions initiated in all animals and typically divided twice , although in some instances fewer or more cell divisions were observed ( Fig . 8B ) . Animals that were grown on food to late in the L3 stage and arrested in the L4 stage with completed cell divisions did not undergo short-range migrations ( Fig . 8B ) , demonstrating that the L4 morphogenetic program did not advance past the early L4 checkpoint . Although cell divisions of the SM cells are not as tightly regulated as the vulval cells , these results suggest that development of the SM cells is also under the control of the L3 and L4 checkpoints . We then examined the seam cells , stem cells that divide during the L1–L3 molts , generating an anterior daughter that fuses with the surrounding hypodermal syncytium and a posterior daughter that retains stem-like properties ( Fig . 8C ) . Animals that were removed from food in the L3 stage showed variability in the timing of seam cell arrest . Some seam cells failed to divide; others divided but anterior daughters did not fuse; and in the most advanced animals , daughter cells fused but the adherens junctions that connect seam cells did not re-form ( Fig . 8C ) . Thus , removal of animals from food in the L3 stage causes the arrest of a several aspects of the seam cell division program prior to reaching the L4 checkpoint . We finally looked at elongation of the gonad , which occurs in a continuous manner from the L2 to L4 stages . When animals were removed from food in mid L3 to cause arrested in the early L4 stage , gonad arm elongation was 35% shorter than in a fed control population of early L4 animals ( Fig . 8D ) , indicating that gonadal elongation arrested prior to the L4 checkpoint . Taken together , these results show that diverse cellular processes arrest following removal of food . Although some tissues had a variable pattern of arrest , in no instance did development continue past the early L3 and early L4 stages , demonstrating that the checkpoints limit tissue development in a systemic manner .
Previous work on L1 arrest , dauer , and adult reproductive diapause have shown that , in response to unfavorable nutritional conditions , cellular processes can arrest for extended durations and resume upon re-feeding [3] , [7] , [9] , [10] , [16] . The nature of the response to nutrient deprivation at other times in development had not been characterized . By focusing initially on the vulva , which has a stereotyped pattern of development during the L3 and L4 larval stages , we show here that checkpoints are present in the early part of the L3 and L4 stages that arrest tissues throughout the organism . The timing of arrest reflects a specific point in the larval stage after molting and prior to initiation of the subsequent molting cycle . A connection between nutrition and the molting cycle has been described in other ecdysozoans [50] , [51] . In insects , for instance , molting to a new larval instar occurs only after a sufficient duration of feeding and attainment of a critical weight [50] . It has been speculated that similar nutritional factors impinge on the endocrine signals that trigger the onset of the molting cycle in C . elegans [38] . Our results provide support for this model of nutritional control of molting cycle commitment . The response to nutrient deprivation in the L3 and L4 stages is systemic in nature and causes the arrest of multiple tissues and cellular processes . Although all tissues arrested prior to or at the checkpoints , vulval formation and the molting cycle were unique in arresting within very narrow developmental windows and in a uniform manner throughout a population . These tissues may have been under selection to arrest in such a precise way . A properly formed vulva is necessary for mating and egg-laying , and a robust developmental program with minimal variation is important for reproductive fitness [52] . For the molting cycle , the inability to shed cuticle surrounding the head during ecdysis causes rapid lethality , as observed in daf-9–overexpressing animals , making it incumbent to execute the molt . A previous study showed that the buccal cavity , which comprises the anterior-most portion of the pharynx and constrains the amount of food consumed with each pumping event , grows only during molts and not between them , which is thought to increase the amount of food that can be consumed during the larval stage [53] . Certain tissues therefore have distinct patterns of growth that are unique to their functions in development and reproduction . We show that the insulin-like signaling pathway regulates that connection between nutritional conditions and progression past the L3 and L4 checkpoints . In wild type animals , feeding of 30–60 minutes is required after molting to attain a sufficient threshold for bypassing the larval stage checkpoints . Perturbations of key genes in the insulin-like signaling pathway alter the duration of feeding required to bypass arrest . Reduction in the function of daf-2 , the insulin/IGF receptor , increases the amount of feeding , such that animals pause at the L3 and L4 checkpoints and have delayed development through the L3 and L4 stages . Loss of function of daf-16 , a FOXO transcription factor that is a key target of the insulin-like signaling pathway [54] , decreases the amount of feeding required to bypass the checkpoints . The bypass of arrest caused by loss of daf-16 is partially suppressed by reduced expression of daf-9 , a cytochrome P450 ortholog required for the production of certain C . elegans steroid hormones [18] , [55] . This result suggests that DAF-16 regulates arrest in the L3 and L4 stages at least in part through inhibition of steroid hormone signaling . DAF-16 has been shown to inhibit daf-9 expression during cholesterol starvation , an unfavorable growth environment that causes larval arrest [32] , [56] . It is possible that late larval stage arrest caused by nutrient deprivation also involves DAF-16 inhibition of daf-9 expression . A key site of action for DAF-16 in regulating L3 and L4 arrest is the hypodermis , where daf-9 is also expressed during larval development [47] , further suggesting that DAF-16 regulates daf-9 expression . Collectively , these findings support a model in which DAF-16 inhibits daf-9 expression , and possibly other genes involved in steroid hormone production , to limit progression through the L3 and L4 stages ( Fig . 9 ) . The level of DAF-9 protein is a key determinant of L3 and L4 stage progressions in the absence of food , which is demonstrated by the striking ability of overexpressed DAF-9 to promote continued development through one or two larval stages . The hormonal signaling pathway that functions downstream of DAF-9 in the L3 and L4 stages appears to be different from the pathway that regulates dauer formation . During the dauer decision , the DAF-9 biosynthetic pathway produces dafachronic acids ( DAs ) , which bind to the nuclear hormone receptor DAF-12 to promote bypass of dauer [18] . Our experiments with a daf-12 null mutant failed to show a similar role for DAF-12 in the L3 and L4 stages . Consistent with these results , we have also found that supplementation of M9 buffer with DAs does not promote continued development past the checkpoints after food removal ( Schindler & Sherwood , unpublished observations ) , further implicating a mode of hormone signaling during the L3 and L4 larval stages that is distinct from that during dauer formation . A key implication from these results is that wild type C . elegans arrest development in the L3 and L4 stages despite possessing a sufficient amount of nutrients to continue further development . This is demonstrated by the ability of animals lacking daf-16 or overexpressing daf-9 to bypass one or even two arrest points and progress through the larval stages in the absence of food . Developmental arrest in wild type animals therefore reflects a decision to halt larval stage progressions rather than a lack of available resources to sustain further development . Continued progression in the absence of food appears to have deleterious consequences , as exemplified by the death and molting defects observed in daf-9–overexpressing animals . Limiting progression through the larval stages when nutritional conditions are poor may allow resources to be conserved for survival and tissue homeostasis during prolonged periods of starvation . This scenario of sensing the environment and arresting development in response to unfavorable conditions also occurs during the C . elegans dauer decision [3] . Both non-dauer and dauer arrest are regulated by insulin-like and DAF-9 signaling pathways , and studies comparing gene expression in dauer and starved animals have revealed overlap between the two types of arrest [57] , [58] . From an evolutionary perspective , it is intriguing to speculate that dauer formation , a nematode-specific developmental diapause , evolved from pathways of starvation-induced arrest that are conserved among metazoans . Two types of growth have been described for ecdysozoans: continuous , with growth occurring throughout the course of development; and saltational , with growth occurring only at distinct times [53] . In organisms with rigid exoskeletons , growth occurs only at molts , an example of saltational growth . C . elegans , with flexible exoskeletons , grow in a continuous manner through the larval stages [53] , [59] . By manipulating the nutritional environment , we show that C . elegans growth has an additional saltational aspect to it , with distinct checkpoints present in the early part of the larval stage . At each checkpoint the nutritional environment informs a systemic decision to either proceed through the larval stage or to remain arrested ( Fig . 9 ) . Two key pathways that regulate this developmental decision—insulin-like and steroid hormone signaling—are present throughout metazoans [60] , [61] , suggesting that the mode of growth control described in this work could be conserved . A greater understanding of the mechanisms of growth control could provide insight into aging and metabolic diseases , which are linked to the dysregulation of developmental pathways important for growth [62] , [63] . Our work in C . elegans demonstrates a type of saltational growth from checkpoint to checkpoint that may similarly regulate development and physiology in other species .
Nematodes were reared at 20°C on NGM plates seeded with OP50 E . coli using standard procedures . In the text and figures we refer to linked DNA sequences that code for a single fusion protein using a ( :: ) annotation . For designating linkage to a promoter we use a ( > ) symbol . The wild type strain N2 and the following mutant strains and transgenes used were: dhIs64[daf-9::GFP] , qyEx262[unc-119>GFP::daf-16] , qyEx263[daf-16>GFP::daf-16] , qyEx264[myo-2>GFP::daf-16] , qyEx266[GFP::daf-16] , qyEx267[ges-1>GFP::daf-16] , qyEx268[unc-115>GFP::daf-16] , qyEx292[col-12>GFP::daf-16] , kbIs7[nhx-2>rde-1] , kzIs9[lin-26>rde-1] , kzIs20[hlh-1>rde-1]; LG I: daf-16 ( mu86 ) , ayIs4[egl-17>GFP] , syIs78[ajm-1::GFP]; LGII: qyIs17[zmp-1>mcherry]; LG III: daf-2 ( e1370 ) , unc-119 ( ed4 ) ; glp-1 ( e2144 ) ; zhIs4[lip-1>NLS-GFP]; LG IV: mgIs49[mlt-10>GFP-PEST] , ayIs7[hlh-8::GFP] , qyIs10[lam-1::GFP]; LG V: rde-1 ( ne219 ) , qyIs50[cdh-3>mCherry::moesinABD]; LG X: hbl-1 ( ve18 ) , qyIs7[lam-1::GFP]; daf-12 ( rh61rh411 ) . Images were acquired using either a Zeiss AxioImager A1 microscope with a 10× , 20× , or 100× plan-apochromat objective and a Zeiss AxioCam MR charge-coupled device camera , controlled by Zeiss Axiovision software ( Zeiss Microimaging , Inc . , Thornwood , NJ ) , or with a Yokogawa spinning disk confocal microscope mounted on a Zeiss AxioImager A1 microscope using iVision software ( Biovision Technologies , Exton , PA ) . Images were processed in ImageJ ( NIH Image ) and Photoshop CS6 ( Adobe Systems Inc . , San Jose , CA ) . Z-stack projections were generated using IMARIS 6 . 0 ( Bitplane , Inc . , Saint Paul , MN ) . Quantification of fluorescence intensity was performed on images acquired at identical exposure settings using ImageJ . For quantifying GFP::DAF-16 in the hypodermis , the fluorescence intensity in four nuclei ( excluding nucleoli ) were averaged per animal . All measurement of nuclear GFP::DAF-16 were taken within 5 min of removal from food to minimize relocalization of DAF-16 into the nucleus . For quantifying DAF-9::GFP , a contiguous area of the hypodermal syncytium that excluded nuclei was measured in a region below the pharynx . Populations containing gravid adults were hypochlorite treated to release embryos , which hatched in M9 buffer and arrested in L1 . The duration of L1 arrest did not exceed 20 h . Populations of L1-arrested animals were plated onto NGM plates seeded with OP50 E . coli that covered at least half the plate to minimize the duration of wandering away from food . Maximum population density was 2500 animals/60 mm dish . Animals were reared at 20°C unless indicated otherwise . For removal from food late in the L2 stage , populations were monitored starting at 22 h post-plating . The assessment of developmental age was made by observation of the gonad ( which grows through the L2 stage ) and the molt ( which covers the mouth during the time of molting , see Fig . 1A ) . Unless a specific duration of growth is indicated , animals were removed from food when the oldest members of the population were molting , and the remaining members were in the late L2 stage , based on gonad size . Populations that contained greater than 5% L3 animals were not used . N2 and daf-12 ( rh61rh411 ) populations typically developed in a synchronized manner; hbl-1 ( ve18 ) , daf-16 ( mu86 ) , and daf-9::GFP , populations grew more variably and had a wider spread of developmental ages at the time of food removal . To remove food , 1 ml M9 was added to each plate and gently rocked to dislodge worms with minimal removal of E . coli . Animals were transferred to low-retention Eppendorf tubes and centrifuged for 1 min at 500× g , a speed at which C . elegans sank to the bottom but E . coli remained largely in suspension . Liquid was aspirated , and an additional 1 ml M9 added for 6 total washes . Tests of supernatants found that bacteria were removed by the third wash , based on the inability of the supernatant to form colonies on LB plates . Animals were placed in M9 buffer at 0 . 5–1 animal/µl in 25-ml glass conical tubes and rotated in a roller drum ( New Brunswick Scientific , Enfield , CT ) at ambient temperature ( 22°C ) . For visualization of ecdysis , animals were removed from food late in the L2 stage , anesthetized in levamisole , mounted on agar pads with sealed cover slips , and maintained for 24 h in a humidified chamber . Developmental stages from L3 to young adult were assessed using the progression of the vulva ( see Fig . 1A ) and the molt . The two processes occurred synchronously in both fed and nutrient-deprived animals . Statistical significance of differences in arrest response was determined by two-tailed Fisher's exact test . For tissue-specific daf-16 rescue experiments , percentages of L3- and L4-arrested animals were determined for each promoter-driven GFP::daf-16 strains and compared to the promoterless GFP::daf-16 strain ( qyEx266 ) . For tissue-specific daf-16 dsRNA feeding , percentages of L3- and L4-arrested animals were compared between animals fed either daf-16 dsRNA or vector control . A similar comparison was made in daf-16 ( mu86 ) animals fed either daf-9 dsRNA or vector control . All assays were repeated in triplicate with n≥50 animals per assay . Populations of wild type animals were removed from food in either late in the L2 or in the middle of the L3 stage and starved in M9 buffer at an approximate population density of 1 animal/µl . Every 2 d , an aliquot of media containing at least 50 animals was removed using Rainex-coated tips to prevent adherence to the plastic , and plated onto NGM plates . After liquid was absorbed into the plate , animals were determined to be alive if they moved or dead if they did not move upon tail poke . Dead animals had a characteristic rod-like appearance . The median survival was determined as the first day at which 50% of the population was dead , for n = 3 trials . To test recovery from nutrient deprivation , early L3 arrested animals were plated onto NGM+OP50 after 8 d in the absence of food . After 72 h at 20°C , the population was scored for fertile and nonfertile adults . Synchronous populations were grown to the L2/L3 or L3/L4 molts . Animals in ecdysis were isolated by the appearance of detached cuticle separated from the body , which was most apparent in the head and tail regions . Individual animals were transferred onto NGM+OP50 plates for 30′–90′ further feeding or placed directly in M9 buffer . Animals were maintained in the absence of food for 24 h and the developmental stage assessed . Transgenic strains expressing promoter-driven daf-16 cDNA fused at the N-terminus with GFP were generated by injection of the following plasmids: pNL205 ( promoterless ) , pNL206 ( unc-119 promoter ) , pNL209 ( daf-16 promoter ) , pNL212 ( myo-3 promoter ) , pNL213 ( ges-1 promoter ) , pNL216 ( unc-115 promoter ) , and pAS10 ( col-12 promoter ) . With the exception of pAS10 , plasmids were gifts of the Kenyon lab and are described elsewhere [42] . pAS10 was generated by PCR amplification of 1 . 1 kb of col-12 promoter 5′ to the start site , which was cloned into the SnaBI restriction sites in pNL205 . GFP::daf-16 plasmids were injected at 100 ng/µl into daf-16 ( mu86 ) ; unc-119 ( ed4 ) adults with 50 ng/µl unc-119 ( + ) plasmid . Animals carrying extrachromosomal arrays were isolated by rescue of the unc-119 locomotion defect and the expression of GFP::DAF-16 validated . Although unc-115 has been reported to express in both neurons and hypodermis [64] , expression was only detected in neurons . With the exception of qyEx266 ( expressing pNL205 ) , which did not possess a gene promoter for GFP::DAF-16 , and qyEx292 ( expressing pAS10 ) , which sometimes had undetectable or minimal GFP expression in the absence of food , animals carrying the array were identified in nutrient removal assays by GFP expression . qyEx266 and qyEx292 animals were plated on NGM plates lacking food , and those that moved freely ( indicating presence of the unc-119 rescue array ) were selected for analysis . The generation and validation of strains sensitive to RNAi in the hypodermis ( NR222 , rde-1 ( ne219 ) ; lin-26>rde-1 ) ; muscle ( NR350 , rde-1 ( ne219 ) ; hlh-1>rde-1 ) ; and intestine ( VP303 , rde-1 ( ne219 ) ; nhx-2>rde-1 ) are described elsewhere [30] , [65] . Strains were fed either daf-16 or L4440 ( vector control ) dsRNA starting from L1 arrest , grown to late in the L2 stage , and removed from food . After 2 d in the absence of food , the developmental stage was scored . To assess AC polarization , the average fluorescence intensity was determined from three-pixel-wide linescans drawn along either the basal or apicolateral membranes of Z-stack projections . To determine the movement of SM cell progeny , the distance between the nuclei of the two inner cells from among the four cells on each lateral half were measured . A similar measurement was made to determine the distance between the two outer nuclei . Gonad length was measured from the vulva to the distal end . All measurements were made using ImageJ software .
|
Organisms in the wild often face long periods in which food is scarce . This may occur due to seasonal effects , loss of territory , or changes in predator-to-prey ratio . During periods of scarcity , organisms undergo adaptations to conserve resources and prolong survival . When nutrient deprivation occurs during development , physical growth and maturation to adulthood is delayed . These effects are also observed in malnourished individuals , who are smaller and reach puberty at later ages . Developmental arrest in response to nutrient scarcity requires a means of sensing changing nutrient conditions and coordinating an organism-wide response . How this occurs is not well understood . We assessed the developmental response to nutrient withdrawal in the nematode worm Caenorhabditis elegans . By removing food in the late larval stages , a period of extensive tissue formation , we have uncovered previously unknown checkpoints that occur at precise times in development . Diverse tissues and cellular processes arrest at the checkpoints . Insulin-like signaling and steroid hormone signaling regulate tissue arrest following nutrient withdrawal . These pathways are conserved in mammals and are linked to growth processes and diseases . Given that the pathways that respond to nutrition are conserved in animals , it is possible that similar checkpoints may also be important in human development .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"motility",
"growth",
"control",
"ecology",
"and",
"environmental",
"sciences",
"animal",
"genetics",
"medicine",
"and",
"health",
"sciences",
"gene",
"function",
"developmental",
"biology",
"model",
"organisms",
"organism",
"development",
"nutrition",
"molecular",
"genetics",
"morphogenesis",
"research",
"and",
"analysis",
"methods",
"gene",
"expression",
"organogenesis",
"aging",
"cell",
"biology",
"cell",
"migration",
"genetics",
"biology",
"and",
"life",
"sciences"
] |
2014
|
Identification of Late Larval Stage Developmental Checkpoints in Caenorhabditis elegans Regulated by Insulin/IGF and Steroid Hormone Signaling Pathways
|
The molecular chaperone Hsp90 regulates the folding of diverse signal transducers in all eukaryotes , profoundly affecting cellular circuitry . In fungi , Hsp90 influences development , drug resistance , and evolution . Hsp90 interacts with ∼10% of the proteome in the model yeast Saccharomyces cerevisiae , while only two interactions have been identified in Candida albicans , the leading fungal pathogen of humans . Utilizing a chemical genomic approach , we mapped the C . albicans Hsp90 interaction network under diverse stress conditions . The chaperone network is environmentally contingent , and most of the 226 genetic interactors are important for growth only under specific conditions , suggesting that they operate downstream of Hsp90 , as with the MAPK Hog1 . Few interactors are important for growth in many environments , and these are poised to operate upstream of Hsp90 , as with the protein kinase CK2 and the transcription factor Ahr1 . We establish environmental contingency in the first chaperone network of a fungal pathogen , novel effectors upstream and downstream of Hsp90 , and network rewiring over evolutionary time .
Hsp90 is an essential and highly conserved molecular chaperone in all eukaryotes that specializes in folding metastable client proteins , many of which are signal transducers [1] , [2] . Together with co-chaperones , Hsp90 interacts dynamically with client proteins regulating their stability and activation . Hsp90 function is subject to complex regulation by post-translational modification , including phosphorylation and acetylation , and depends upon an ATP binding and hydrolysis cycle [3] , [4] . Hsp90 is generally expressed at much higher levels than required for basal function , however , environmental stress can induce global problems in protein folding and thereby overwhelm Hsp90's functional capacity [5] . As an environmentally contingent hub of protein homeostasis and regulatory circuitry , Hsp90 has profound effects on biology , disease , and evolution . Hsp90 modulates the phenotypic effects of genetic variation in an environmentally responsive manner [6] , [7] , [8] , [9] , influencing ∼20% of observed natural genetic variation and serving both to maintain phenotypic robustness and promote diversification [7] . Hsp90's broad influence on current genomes in part reflects its extensive connectivity in interaction networks and its profound impact on cellular circuitry . A global analysis of the Hsp90 chaperone network has thus far only been achieved in S . cerevisiae . Systematic proteomic and genomic methods have been applied to map physical , genetic , and chemical-genetic interactions , revealing that Hsp90 interacts with ∼10% of the proteome [10] , [11] . In addition to identifying known co-chaperones and client proteins , these network analyses identified new Hsp90 client proteins as well as novel co-factors that link Hsp90 with chromatin remodeling and epigenetic gene regulation . A subsequent chemical-genetic screen identified distinct Hsp90 interactions at elevated temperatures , suggesting that specialized chaperone functions mediate responses to environmental stress [12] . While there are numerous conserved client proteins between S . cerevisiae and other eukaryotes , there is evidence for plasticity in the Hsp90 chaperone machine , with differences in co-chaperones even between S . cerevisiae and another model yeast , Schizosaccharomyces pombe [13] . The extent to which the Hsp90 chaperone network has been rewired over evolutionary time remains unknown . Fungi provide not only the most powerful eukaryotic genetic model systems , but also a major threat to human health , and Hsp90 holds great promise as a therapeutic target [14] , [15] . Invasive fungal infections are a leading cause of mortality among immunocompromised individuals , including those with cancer and HIV [16] . Treatment of fungal infections is hampered by the limited number of antifungal drugs , host toxicity , and the emergence of drug resistance [14] , [17] . We previously established that Hsp90 regulates the emergence and maintenance of resistance to the most widely deployed classes of antifungal drugs in the clinic , the azoles and echinocandins [6] , [15] , [18] , [19] . Compromising Hsp90 function can transform antifungals from ineffective to highly efficacious in combating otherwise lethal infections caused by the most prevalent fungal pathogens of humans , Candida albicans and Aspergillus fumigatus [15] . In C . albicans , Hsp90 regulates not only drug resistance , but also morphogenesis and virulence [20] . Despite the therapeutic potential of targeting Hsp90 , mouse model studies revealed that Hsp90 inhibitors in clinical development as anti-cancer agents have toxicity in the context of an acute fungal infection [15] , motivating the search for fungal-selective Hsp90 inhibitors and fungal-specific components of the Hsp90 chaperone network . C . albicans provides the ideal fungal pathogen with which to dissect the Hsp90 chaperone network given its clinical relevance and its dependence on Hsp90 for drug resistance , virulence , and temperature-dependent morphogenesis . Candida species account for 88% of all hospital-acquired fungal infections [16] . C . albicans is the leading fungal pathogen of humans worldwide with mortality rates approaching 50% , and is the fourth most common cause of hospital-acquired infections [16] , [21] , [22] . Hsp90 regulates resistance to both azoles and echinocandins by stabilizing the protein phosphatase calcineurin and the terminal mitogen-activated protein kinase ( MAPK ) in the Pkc1 cell wall integrity pathway , Mkc1 [18] , [19] . To date , these are the only two Hsp90 interactors identified in a fungal pathogen . Although genetic analyses with C . albicans have been hampered by its obligate diploid state and lack of a complete sexual cycle , recently homozygous mutant libraries have been developed to enable systematic screens and genomic analyses [23] , [24] , [25] , [26] . Here , we mapped the first Hsp90 genetic interaction network in a fungal pathogen . We conducted a chemical-genetic screen with the first C . albicans homozygous transposon insertion mutant library containing 1 , 248 strains and covering ∼10% of the genome . Growth was scored under standard conditions as well as five stress conditions , both in the presence and absence of the Hsp90 inhibitor geldanamycin . Hypersensitivity to geldanamycin is indicative of an Hsp90 genetic interaction . The resulting network of interactions was extensively contingent on the environment . Most of the 226 genetic interactors were identified as important for growth only under specific conditions , suggesting that they operate downstream of Hsp90 . Consistent with this model , Hsp90 depletion led to reduction in protein levels of several candidate interacting kinases , including Hog1 for which protein levels were reduced and stress-induced activation was abolished . Only a few genetic interactors were identified in many of the screens , and these likely operate upstream of Hsp90 . Consistent with this model , interactors identified in five of the six screens include the regulatory subunits of casein kinase CK2 , which governed function of the Hsp90 chaperone machine , and the transcription factor Ahr1 , which promoted HSP90 expression . The C . albicans Hsp90 genetic interaction network has been rewired relative to its S . cerevisiae counterpart , with a small but significant set of interactions conserved . Thus , we establish environmental contingency in the first Hsp90 chaperone network of a fungal pathogen , novel effectors upstream and downstream of Hsp90 , and rewiring over evolutionary time .
We conducted a chemical genetic screen employing a stationary liquid assay with the first C . albicans homozygous mutant library that covers ∼10% of the genome ( 661 genes ) [23] , [25] . To identify Hsp90 genetic interactors , the library was screened for mutants hypersensitive to pharmacological inhibition of Hsp90 with geldanamycin , which binds with high affinity to Hsp90's unusual ATP binding pocket and thereby blocks ATP-dependent chaperone function [27] . The screen was conducted under standard growth conditions ( 37°C ) , general stress conditions ( elevated temperature of 41°C or osmotic stress exerted by sodium chloride ( NaCl ) ) , and specific stress conditions exerted by drugs targeting the endoplasmic reticulum ( nucleoside antibiotic tunicamycin ) , the cell wall ( echinocandin caspofungin ) , or the cell membrane ( azole fluconazole ) ( Figure S1A , Table 1 ) . Following incubation , growth was monitored by optical density and normalized relative to the geldanamycin-free control ( Figure S1 ) . Control strains with different geldanamycin sensitivities were included in each screen ( Figure S1B ) . The six screens yielded a total of 226 distinct Hsp90 genetic interactors ( Figure 1 and Table S1 ) . These were displayed as a network showing relationships between interactors and the screen conditions in which they were identified ( Figure 1 ) . The number of interactions differed widely between screens , as did the extent of overlap ( Table 1 ) . The cell wall stress screen ( caspofungin ) revealed the most interactions with 73 in total , 52 of which were unique; this screen was negatively correlated with all of the other screens ( Figure 1 ) , a unique pattern . Most genetic interactions identified were specific to one or two conditions , and very few were common to four or more conditions tested . Only nine interactions were identified in at least four screens; four of these were identified in at least five screens ( AHR1 , CKB1 , CKB2 , and HOS2 ) and only one in all six screens ( HOS2 ) . Next , we tested the Hsp90 genetic interactors for enrichment of gene ontology ( GO ) gene function categories relative to the composition of the library . C . albicans Hsp90 genetic interactors were enriched for macromolecular complexes ( P = 0 . 005 ) , protein complexes ( P = 0 . 005 ) , protein modification processes ( P = 0 . 018 ) , biopolymer modification ( P = 0 . 018 ) , and post-translational protein modifications ( P = 0 . 018 ) . Kinases comprised 34 of the 226 Hsp90 genetic interactors identified ( Figure 2 ) , and were enriched from 10% of the library to 15% of the genetic interactors; 29 of the kinases were specific to one or two screens . The high temperature and tunicamycin screens identified the largest number of kinases , with seven each and three shared between them . Screen-specific GO enrichment varied ( Table 1 ) , consistent with the distinct suite of Hsp90 interactors identified as important for growth in the specific conditions . Next , we examined whether our approach to map Hsp90 genetic interactions could also reveal proteins with functional dependence on Hsp90 . Reassuringly , our screens identified both of the established C . albicans Hsp90 client proteins , Cna1 and Mkc1 . Hsp90 physically interacts with and stabilizes the catalytic subunit of the protein phosphatase calcineurin , Cna1 , such that depletion of Hsp90 leads to depletion of calcineurin , blocking calcineurin-dependent stress responses [19] . Hsp90 also stabilizes the MAPK Mkc1 , such that depletion of Hsp90 leads to depletion of Mkc1 , blocking downstream stress responses [18] . CNA1 and MKC1 were identified as Hsp90 interactors in our tunicamycin screen , consistent with their role in mediating responses to endoplasmic reticulum stress [28] , [29] . To determine if we could identify novel functional dependence of a C . albicans protein on Hsp90 based on our dataset we turned to Hog1 . Hog1 is a MAPK involved in osmoregulation that was identified as an Hsp90 interactor in our high temperature screen , and connections between Hog1 and Hsp90 have been established in other eukaryotes . In S . cerevisiae , Hog1 interacts with Hsp90 and co-chaperone Cdc37 , which facilitates Hsp90's kinase specificity , and mutation of Cdc37 leads to reduced Hog1 levels and impaired downstream stress responses [30] . The mammalian homolog of Hog1 , p38 , is an Hsp90 client and interacts with Hsp90 via Cdc37 [31]; inhibition of Hsp90 leads to autoactivation of p38 , suggesting that the Hsp90-Cdc37 complex functions as a negative regulator of p38 in mammalian cells as opposed to its role as a positive regulator in S . cerevisiae . We tested the impact of Hsp90 depletion on levels of both total Hog1 protein and activated dually phosphorylated Hog1 in response to osmotic stress induced by exposure to hydrogen peroxide ( Figure 3A ) . To deplete Hsp90 , we used a strain with its only HSP90 allele driven by a doxycycline-repressible promoter ( tetO-HSP90/hsp90Δ ) . In the presence of doxycycline , Hsp90 levels were depleted in the tetO-HSP90/hsp90Δ strain and not the wild type . Depletion of Hsp90 led to a ∼60% reduction in the levels of total Hog1 and abolished stress-induced Hog1 activation ( Figure 3A ) . Transcript levels of HOG1 were reduced by ∼30% upon Hsp90 depletion ( Figure S2 ) , suggesting that Hsp90 affects expression as well as activation , or stability of the activated form , of this MAPK . These results confirm that our chemical genetic screen can identify client proteins with functional dependence on Hsp90 . Given that our chemical genetic screens identify Hsp90 genetic interactions based on importance for growth under distinct environmental conditions , we hypothesized that low-connectivity interactors act downstream of Hsp90 to mediate specific responses , whereas high-connectivity interactors act upstream of Hsp90 to regulate its function or expression . To test this hypothesis , we determined the impact of Hsp90 depletion on three high- and five low-connectivity interactors ( Figure 3B ) . The high-connectivity interactors ( HOS2 , CKB1 , and CKB2 ) were identified in five or six conditions and the low-connectivity interactors ( CKA1 , MKK2 , CMK1 , CDR1 , and HOG1 ) in up to three . With the exception of Hog1 , which was discussed above , all candidate interactors were TAP-tagged to monitor protein levels upon Hsp90 depletion . Consistent with our hypothesis , depletion of Hsp90 caused greater reduction of protein levels for low-connectivity interactors than high-connectivity interactors ( P = 0 . 0430 ) ; protein levels of four of the five low-connectivity interactors were reduced by greater than 25% , while none of the high-connectivity interactors exhibited this magnitude of reduction ( Figure 3C ) . To distinguish more indirect effects on gene expression from effects on protein stability , we monitored transcript levels for all interactors that showed substantial reduction in protein levels . Of the five low-connectivity interactors tested , only one had significantly reduced transcript levels upon Hsp90 depletion , HOG1 ( Figure S2 ) . Both established C . albicans client proteins that require Hsp90 for stability , Cna1 [19] and Mkc1 [18] , were also low-connectivity interactors . These findings suggest that low-connectivity interactors depend on Hsp90 for stability or expression while high-connectivity interactors do not . A striking observation was that two of the high- ( CKB1 and CKB2 ) and one of the low-connectivity interactors ( CKA1 ) are subunits of protein kinase CK2 . CK2 is a serine/threonine protein kinase and phosphorylates many substrates including yeast and human Hsp90 , thereby regulating its function [32] . Like many kinases that phosphorylate Hsp90 , CK2 is also an Hsp90 client in mammalian cells [33] , suggesting that feedback loops might enable kinases to modulate their chaperoning and activation . In C . albicans , depletion of Hsp90 leads to reduced levels of both Cka1 and Cka2 catalytic subunits . Cka1 was identified as low-connectivity in our screens and Cka2 was not identified although it was present in the library ( Figure 3B , 3C ) . Hsp90 depletion did lead to reduced levels of Cka2 protein ( Figure 3C ) and CKA2 transcript ( Figure S2 ) . Protein levels of the two high-connectivity regulatory subunits , Ckb1 and Ckb2 , remained relatively stable despite Hsp90 depletion ( Figure 3C ) . These findings suggest that the catalytic CK2 subunits may be more dependent upon Hsp90 than the regulatory subunits , and that the regulatory subunits may act upstream to modify Hsp90 function . If the regulatory subunits function with the catalytic subunits to phosphorylate Hsp90 , one would expect the catalytic subunits to have been high-connectivity interactors; that they were not could be due to their partial redundancy [34] , [35] . To test this , we constructed a strain lacking Cka1 and in which Cka2 could be depleted by doxycycline-mediated transcriptional repression ( cka1Δ/cka1Δ tetO-CKA2/cka2Δ , Figure S3 ) . We repeated all six screens with the CK2 transposon mutants , clean deletion mutants , and our catalytic subunit depletion strain . The deletion mutants phenocopied the transposon mutants , confirming that the screens were reproducible and the original phenotypes of the transposon mutants were valid ( Figure 3D ) . Further , depletion of the CK2 catalytic subunits conferred hypersensitivity to geldanamycin in all six screens ( Figure 3D ) , confirming that the catalytic subunits would indeed have been high-connectivity interactors if not for their redundancy . Thus , we establish important functional connections between Hsp90 and CK2 in C . albicans . Next , we tested whether CK2 phosphorylates threonine and serine residues of Hsp90 and Cdc37 in C . albicans . In S . cerevisiae Hsp90 is phosphorylated on at least 11 residues [36] , including threonine 22 by CK2 [32] . CK2 also phosphorylates the Hsp90 co-chaperone Cdc37 , which is critical for proper binding to kinases and for their stability [37] . To determine if CK2 phosphorylates Hsp90 and/or Cdc37 in C . albicans , we immunoprecipitated either Hsp90 or Cdc37 from the wild type and mutants lacking CK2 components and monitored levels of threonine and serine phosphorylation relative to total immunocprecipitated Hsp90 or Cdc37 protein . Threonine phosphorylation of Hsp90 was reduced by 90% in the ckb1Δ/ckb1Δ mutant and serine phosphorylation was reduced by 68% ( Figure 4A ) . That phosphorylation was not reduced in the catalytic subunit mutants could be due to their partial redundancy [34] , [35] , consistent with our screen results ( Figure 3D ) . That phosphorylation was not reduced in the ckb2Δ/ckb2Δ mutant might suggest that Ckb2 directs phosphorylation of fewer threonine or serine residues than Ckb1 under these conditions , or that it plays a more important role in phosphorylation of other targets . Consistent with the latter possibility , threonine and serine phosphorylation of Cdc37 was largely abolished in the ckb2Δ/ckb2Δ mutant ( Figure 4A ) . Threonine phosphorylation of Cdc37 was also reduced by greater than 90% in the ckb1Δ/ckb1Δ mutant , while serine phosphorylation was reduced by 44% ( Figure 4A ) . Complementation of the ckb1Δ/ckb1Δ and ckb2Δ/ckb2Δ mutants with wild-type alleles of CKB1 or CKB2 restored phosphorylation of Hsp90 and Cdc37 ( Figure S4A and data not shown ) . Thus , C . albicans CK2 regulates serine and threonine phosphorylation of both Hsp90 and Cdc37 . Given that phosphorylation of Hsp90 [32] or its co-chaperone Cdc37 [37] can affect stability and function of target kinase client proteins in S . cerevisiae , we assessed the impact of deletion of each of the four C . albicans CK2 subunits on levels of Hsp90 , Cdc37 , and target kinase Hog1 , as well as on Hog1 activation . Cells were grown in standard conditions or with a short burst of oxidative stress in order to monitor Hog1 activation via phosphorylation . Consistent with our hypothesis that the high-connectivity regulatory subunits of CK2 function upstream of Hsp90 , protein levels of Hsp90 , Cdc37 , and Hog1 were reduced substantially in the ckb1Δ/ckb1Δ and ckb2Δ/ckb2Δ mutants ( Figure 4B ) . Interestingly , there was a stress-dependent difference in the impact of CK2 subunit deletion . During standard growth conditions , the only change in proteins levels greater than 25% was that Cdc37 levels were reduced by 71% in the ckb2Δ/ckb2Δ mutant ( Figure 4B , left panel ) . In response to oxidative stress , Hsp90 levels were reduced in both the ckb1Δ/ckb1Δ ( 38% ) and ckb2Δ/ckb2Δ ( 29% ) mutants ( Figure 4B , right panel ) . Hog1 levels were also reduced in response to oxidative stress in both the ckb1Δ/ckb1Δ ( 34% ) and ckb2Δ/ckb2Δ ( 35% ) mutants ( Figure 4B , right panel ) . Hog1 activation was not abolished ( Figure 4B ) , unlike with depletion of Hsp90 ( Figure 3A ) . Complementation of key mutants with a wild-type allele of CKB1 or CKB2 fully restored Hsp90 and Cdc37 protein levels , and partially restored Hog1 protein levels ( Figure S4B ) , confirming that the observed effects are indeed due to the specific gene deletions . Thus , modification of the Hsp90-Cdc37 complex via the CK2 regulatory subunits is stress-dependent , such that Ckb2 is required for Cdc37 levels in the absence of stress , while Ckb1 and Ckb2 are both required for Hsp90 and Hog1 levels during oxidative stress . If the regulatory subunits of CK2 regulate function of the Hsp90-Cdc37 complex and downstream clients such as Hog1 , then one would predict that deletion of these CK2 subunits would phenocopy deletion of Hog1 . To test this , we monitored growth during osmotic stress exerted by sorbitol , given Hog1's importance for osmotic stress responses . We found that the ckb1Δ/ckb1Δ , ckb2Δ/ckb2Δ and hog1Δ/hog1Δ mutants were equally hypersensitive to high osmolarity ( Figure 4C ) . Complementation of the mutants with a wild-type allele of CKB1 , CKB2 , or HOG1 restored high osmolarity growth , confirming that the phenotypes observed are a consequence of the specific gene deletions ( Figure S4C ) . Taken together , our results support the model that function of the Hsp90-Cdc37 chaperone complex is modulated in a stress-dependent manner by the high-connectivity interactors Ckb1 and Ckb2 , thereby affecting target kinases ( Figure 4D ) . To date , the most extensive studies of the Hsp90 chaperone network have been carried out in S . cerevisiae [11] , [12] . Chaperone networks have been examined in the protozoan parasite Plasmodium falciparum [38] , but comparative analysis was limited due to high protein interaction network divergence from other eukaryotes [39] , [40] . As of yet , comparative analysis of Hsp90 genetic interaction networks has not been feasible due to the lack of large-scale interaction data in species other than S . cerevisiae . Comparison of our C . albicans Hsp90 genetic interaction set with those from S . cerevisiae genetic screens [11] , [12] revealed a small but significant overlap ( Hypergeometric , P = 0 . 004 ) , despite their highly similar co-chaperone machineries ( Figure S5 ) . The C . albicans library screened contains insertions in 428 genes that have homologs in S . cerevisiae , 59 of which are genetic interactors in S . cerevisiae; of these 59 , 30 were unique to S . cerevisiae ( Figure S5 ) . The 29 Hsp90 interactors that were conserved out of 171 C . albicans Hsp90 interactors that have a S . cerevisiae homolog indicate that only ∼17% of C . albicans Hsp90 genetic interactions are conserved . Thus , the chaperone network has been rewired considerably over evolutionary time . The conserved interactors were distributed throughout the C . albicans Hsp90 network but differences in the extent of conservation were observed in different stress conditions . While ∼25% of Hsp90 genetic interactions identified in the cell membrane stress screen ( fluconazole ) were conserved , less than 10% of the those from the cell wall stress screen ( caspofungin ) were conserved ( Figure S5 , left insert ) . Despite the low level of conservation , both S . cerevisiae and C . albicans networks exhibited similar responses to elevated temperature in that they both maintained a large fraction of their interactions ( roughly a half [12] and a third , respectively ) ( Figure S5 , right insert ) . While the particular interactions involved may differ , similar proportions of the genome remains associated with Hsp90 during high temperature growth in C . albicans and S . cerevisiae . We tested whether the conserved Hsp90 genetic interactors might reflect dependence of the corresponding proteins on Hsp90 . We monitored the impact of Hsp90 depletion on protein and transcript levels of three conserved genetic interactors ( CKB2 , CKA1 , and CDR1 ) . Cka1 protein levels were reduced by >25% and Cdr1 levels by >50% upon depletion of Hsp90 ( Figure 3 ) , and transcript levels of both CKA1 and CDR1 remained unchanged , suggesting that some but not all of the conserved Hsp90 genetic interactions may also reflect a physical interaction . Given the considerable rewiring of the Hsp90 chaperone network , we sought to characterize a novel Hsp90 interactor in C . albicans . We focused on AHR1 , as it was identified as an Hsp90 genetic interactor in five out of our six screens . This zinc finger transcription factor binds to target promoters to regulate transcription of genes involved in adherence , morphogenesis , and virulence [24] , [41] . Consistent with the expectation that high-connectivity interactors function upstream of Hsp90 , we found that HSP90 transcript levels were reduced in an ahr1Δ/ahr1Δ mutant ( Figure 5A , t-test , P = 0 . 0298 ) . Strikingly , deletion of AHR1 phenocopies compromise of Hsp90 function leading to filamentation in rich medium at 30°C , canonical conditions for yeast growth ( Figure 5B ) . Complementation with a wild-type allele of AHR1 restores wild-type HSP90 transcript levels and morphology ( Figure 5 ) . Thus , Ahr1 is a novel high-connectivity C . albicans Hsp90 interactor that influences HSP90 expression and morphogenesis , a trait of central importance for virulence .
Our results establish the first Hsp90 chaperone network of a fungal pathogen , novel effectors upstream and downstream of Hsp90 , environmental contingency in the network , and network rewiring over evolutionary time . Based on our chemical genetic screen with the first C . albicans homozygous transposon insertion mutant library [23] , [25] , Hsp90 interacts with ∼4% of the genome ( Figure 1 ) , including many kinases ( Figure 2 ) . The proportion of the genome that interacts with Hsp90 is expected to increase upon screening additional mutants or stress conditions . The chaperone network is environmentally contingent , and most of the 226 genetic interactors are important for growth only under specific conditions , suggesting that they operate downstream of Hsp90 , as with Hog1 ( Figure 3A and Figure 6 ) . Few genetic interactors are important for growth in many environments , and these are poised to operate upstream of Hsp90 to regulate its function or expression , as with the protein kinase CK2 and transcription factor Ahr1 ( Figure 4 , Figure 5 , and Figure 6 ) . The C . albicans Hsp90 genetic interaction network is rewired relative to its S . cerevisiae counterpart ( Figure S5 ) , emphasizing the importance of dissecting the chaperone network in the pathogen to elucidate circuitry through which Hsp90 regulates key traits important for virulence . Our study provides the first unbiased analysis of C . albicans Hsp90 interactors , and a glimpse of the circuitry through which Hsp90 governs drug resistance , morphogenesis , and virulence . Prior to this work , only two Hsp90 interactors were identified in C . albicans , the protein phosphatase calcineurin [19] and MAPK Mkc1 [18] . Both are Hsp90 genetic interactors in our tunicamycin screen ( Figure 1 and Figure 2 ) , consistent with their role in mediating responses to endoplasmic reticulum stress [28] , [29] , and validating that the genetic interactors we identify here also include physical interactors with functional dependence on Hsp90 . The MAPK Hog1 , an Hsp90 genetic interactor in our high temperature growth screen ( Figure 1 and Figure 2 ) , has been previously connected with Hsp90 in other eukaryotes . In S . cerevisiae and mammalian cells , Hog1/p38 interacts with Hsp90 via the co-chaperone Cdc37 . In S . cerevisiae , Hog1 levels decrease upon compromising Hsp90-Cdc37 function and canonical ( stress-induced ) levels of Hog1 phosphorylation are reduced by ∼20% [30] . In mammalian cells , Hsp90-Cdc37 is dispensable for canonical activation of p38 and inhibition of Hsp90 leads to auto-activation of p38 [31] . In C . albicans , depletion of Hsp90 reduces Hog1 protein by ∼60% and abolishes stress-induced Hog1 activation ( Figure 3A ) , suggesting that regulation of Hog1 activation in C . albicans is similar to that in S . cerevisiae but perhaps more dependent upon Hsp90 . Hog1 is itself a global regulator of the C . albicans proteome induced in response to diverse stresses [42] , and mutants lacking Hog1 display hypersensitivity to stress , altered morphogenesis , attenuated virulence in mouse models of systemic disease , and enhanced vulnerability to killing by phagocytes [43] , [44] . Our identification of Ahr1 as a novel regulator of HSP90 expression and morphogenesis ( Figure 5 ) further validates that our chaperone network reveals novel regulators through which Hsp90 governs stress response , drug resistance , morphogenesis , and virulence . There are distinct sets of Hsp90 genetic interactions under different conditions , establishing that the network is environmentally contingent and suggesting specialized Hsp90 functions in mediating responses to specific stresses . We reasoned that interactors that are low connectivity in the network , identified in only one to three screens , likely function downstream of Hsp90 to regulate cellular processes important for growth in specific environments . Indeed , depletion of Hsp90 causes greater reduction of protein levels for low-connectivity than high-connectivity interactors; protein levels for four ( Hog1 , Cka1 , Mkk2 , and Cdr1 ) out of the five low-connectivity interactors tested are reduced by greater than 25% ( Figure 3C ) ; out of these four only one showed a significant reduction in transcript levels ( Figure S2 ) , suggesting that low-connectivity interactors depend upon Hsp90 for stability or expression . The level of reduction of low-connectivity interactors upon Hsp90 depletion ranges from 28% to 61% , suggesting that additional factors contribute to their stability . The one low-connectivity interactor that does not show reduced protein levels upon Hsp90 depletion , a calmodulin-dependent kinase Cmk1 , could still rely on Hsp90 for activation rather than stability , consistent with the finding that Cmk1 interacts with Hsp90 in the fungal pathogen Sporothrix schenckii [45] , or alternatively could function in a pathway with which Hsp90 interacts . Thus , connectivity in the network can reveal functional properties of Hsp90 interactors in terms of the environmental conditions for which they enable adaptive responses . Interactors that are high connectivity in the network are likely to function upstream of Hsp90 and thereby regulate its function or expression , impacting on growth in diverse conditions . None of the three high-connectivity interactors tested ( Hos2 , Ckb1 , and Ckb2 ) show reduced proteins levels upon Hsp90 depletion ( Figure 3C ) , consistent with the hypothesis that they function upstream of Hsp90 . Since three out of the four protein kinase CK2 subunits interact genetically with Hsp90 in our screens , we tested the hypothesis that high-connectivity interactors regulate Hsp90 function by focusing on the high-connectivity CK2 regulatory subunits , Ckb1 and Ckb2 . CK2 phosphorylates many cellular targets , including a conserved threonine in Hsp90 of S . cerevisiae ( T22 ) and mammalian cells ( T36 in hHsp90α ) [32] . CK2-dependent phosphorylation of Hsp90 modulates chaperone activity , affecting the stability and function of diverse clients . CK2 also phosphorylates Cdc37 , a prerequisite for proper binding to kinases and for their stability [37] . There is feedback such the Hsp90-Cdc37 chaperone also binds CK2 thereby promoting its stability and activation [33] . We provide the first evidence for a functional relationship between CK2 and Hsp90 or Cdc37 in C . albicans . Hsp90 serine and threonine phosphorylation is dramatically reduced in the ckb1Δ/ckb1Δ mutant ( Figure 4 ) . Cdc37 serine and threonine phosphorylation is also reduced in the ckb1Δ/ckb1Δ mutant , and is largely abolished in the ckb2Δ/ckb2Δ mutant ( Figure 4 ) . Redundancy of the CK2 catalytic subunits could explain why Hsp90 phosphorylation was not reduced in these mutants , consistent with our screen results ( Figure 3D ) . Cdc37 levels depend on Ckb2 during standard growth , while Hsp90 levels depend on both Ckb1 and Ckb2 during oxidative stress ( Figure 4 ) . The ckb1Δ/ckb1Δ and ckb2Δ/ckb2Δ mutants have reduced levels of the Hsp90-Cdc37 target kinase Hog1 and phenocopy a hog1Δ/hog1Δ mutant in terms of hypersensitivity to oxidative stress ( Figure 4 ) , supporting the model that the high-connectivity CK2 regulatory subunits influence Hsp90-Cdc37 function . Taken together , our study reveals CK2 as the first regulator of C . albicans Hsp90 function and suggests that additional high connectivity interactors might also serve to regulate Hsp90 function or expression , as with Ahr1 ( Figure 5 ) . Although Hsp90 is highly conserved , the Hsp90 chaperone network has been rewired over evolutionary time . Akin to the S . cerevisiae Hsp90 network , the C . albicans network prominently featured kinases , which were enriched from 10% in the library to 15% in the network . Enrichment for transcription factors [11] , however , was not detected in C . albicans , despite transcription factors being well represented in the library . This may be due to network rewiring since the species diverged , as only ∼17% of the genetic interaction network is conserved ( Figure S5 ) , suggesting network remodeling during adaptation to specific ecological niches . Rewiring in response to selective pressure could also explain the large set of 73 Hsp90 interactors important for growth during cell wall stress ( caspofungin ) , 52 of which are specialized for that stress , given our tested set of conditions ( Figure 1 , Table 1 ) . Notably , Hsp90 governs survival in response to echinocandin-induced cellular stress in C . albicans , but not in S . cerevisiae [19] . Further , signaling pathways governing cell wall integrity pathways have been rewired between C . albicans and S . cerevisiae [23] . The fungal cell wall is essential for viability of fungal cells and is an elaborate structure , components of which are recognized by the vigilant cadre of immune cells in the human host [46] . As a commensal and opportunistic pathogen , C . albicans is likely to harbor circuitry orchestrating cell wall structure that was subject to strong selection in response to challenge by the host immune system . Indeed C . albicans can evade immune recognition and attack by masking its β-glucan [47] . The finding of evolutionary reconfiguration of the Hsp90 chaperone network in a fungal pathogen motivates future studies to map the chaperone network in diverse eukaryotic pathogens in which Hsp90 has been implicated in governing drug resistance , development , or virulence , such as the fungal pathogen Aspergillus fumigatus and the protozoan parasites Plasmodium falciparum and Trypanosoma evansi [15] , [48] . Identifying pathogen-specific components of the Hsp90 chaperone network offers great therapeutic potential for the development of inhibitors to minimize host toxicity and cripple diverse eukaryotic pathogens .
The C . albicans transposon insertion mutant library was generously provided by Aaron Mitchell ( Carnegie Mellon University ) with additional plates obtained from the Fungal Genetics Stock Center and pinned onto YPD agar plates ( 1% yeast extract , 2% peptone , 2% dextrose , 2% agar ) . Strains were inoculated in 100 µl RPMI-1640 pH 7 ( 10 . 4 g/l RPMI-1640 , 3 . 5% MOPS , 2% glucose , 20 mg/ml histidine , 80 mg/ml uridine ) , sealed with Adhesive Plate Seals ( Thermo Scientific ) and incubated overnight at 37°C while shaking at 200 rpm . Cells were then diluted twice . First , 1∶1 , 000 using the VP 408 96 Pin Multi-Blot Replicator ( VP Scientific ) in 1× phosphate buffered saline ( PBS ) . Second , the PBS – Candida mixture was diluted 1∶10 in a total volume of 200 µl RPMI-1640 , RPMI with 3 µM geldanamycin , RPMI with stressor ( Table 1 ) , and RPMI with 3 µM geldanamycin and stressor in flat bottom 96-well plates . Plates were incubated at 37°C for between two and nine days , depending on the stressor ( Table 1 ) . Following incubation , optical densities ( ODs ) were measured at λ = 600 nm . ODs were recorded for RPMI alone , RPMI with geldanamycin , RPMI with stressor , and RPMI with geldanamycin and stressor , and normalized . The normalized values were transformed into heat maps , which represent growth as a function of color using Java TreeView 1 . 1 . 3 . [49] . For each library plate , the RPMI alone was compared with the geldanamycin and the stress alone plates and the combination of both . A genetic interactor was defined as a mutant that responded with a severe growth defect or death to the combination of geldanamycin and stressor when neither geldanamycin alone nor the stressor alone impaired growth of the mutant . Genetic interactors were scored depending on the number of mutants available for a particular ORF: ‘1→0’ indicates that only one mutant was available and that mutant was severely hypersensitive to geldanamycin; ‘2→0’ indicates that both available mutants were severely hypersensitive to geldanamycin; and ‘1/2→0’ indicates that one of two available mutants was severely hypersensitive to geldanamycin with no growth and the other had impaired growth in the presence of geldanamycin compared to the wild-type strain Day 286 . In the rare cases that more than two mutants for a particular gene were present in the library , at least two mutants had to exhibit severe hypersensitivity to geldanamycin to be scored as a genetic interaction . Networks of 226 global interactions and 34 kinase interactions were visualized with Cytoscape [50] and the layout manually improved for readability and clarity . A Fisher's Exact Test followed by a correction for multiple testing ( empirical resampling ) was used to identify GO terms that were enriched in the complete data set , in the different screens , and in the genetic interaction sets that were either unique to C . albicans , to S . cerevisiae , or shared by both . The GO enrichment analysis was performed with FuncAssociate [51] ( download date April 13 , 2011 ) using the program's default parameters on C . albicans gene lists and GO terms . All C . albicans-specific analyses were performed against a background of 661 genes in the library . When comparing C . albicans with S . cerevisiae , gene orthology information was obtained from the Candida Genome Database ( http://www . candidagenome . org/ ) . All strains used here that were not part of the library ( Table S2 ) were maintained in cryo-culture at −80°C in 25% glycerol . Genes of candidate interactors were tagged with the tandem affinity purification ( TAP ) tag [52] in the wild-type strain SN95 and its derivative CaLC1411 using a PCR-based strategy [53] . The tag and a selectable marker ( ARG4 ) were PCR amplified from pLC573 ( pFA-TAP-ARG4 [53] ) , 200 to 400 µl of PCR product were ethanol precipitated , dissolved in 50 µl sterile water and transformed into C . albicans using standard protocols . Oligonucleotides used in this study are listed in Table S3 . Correct genomic integration was verified using appropriate primer pairs that anneal ∼500 bp up or down stream from both insertion junctions together with primers oLC1593 ( TAP-R ) and oLC1594 ( ARG4-F ) that target the TAP tag and the selectable marker . The same TAP tagging strategy using pLC572 ( pFA-TAP-HIS1 [53] ) was employed to tag CDC37 and HSP90 in the CK2 subunit deletion mutants . Details regarding TAP tagging of HOS2 , complementation of CK2 subunits , and required strains and plasmids can be found in Text S1 . C . albicans CaLC239 ( SN95 ) and CaLC1411 ( tetO-HSP90-hsp90Δ ) with and without TAP tagged interactors ( Table S2 ) were grown overnight at 37°C in RPMI-1640 while shaking at 200 rpm . Stationary phase cultures were split , adjusted to an OD600 of 0 . 2 and one culture was treated with 10 µg/ml doxycycline ( BD Biosciences ) , while the other was left untreated . After 24 hours of incubation at 37°C in RPMI-1640 , cultures were re-adjusted to OD600 of 0 . 2 with and without doxycycline and grown to mid-log phase ( OD600 0 . 6–0 . 8 ) . Between 15 and 50 ml were harvested from each culture , centrifuged for 10 minutes at 3000 rpm at 4°C , and washed once with ice-cold 1×PBS . Pellets were resuspended in 200 µl lysis buffer ( 50 mM Hepes pH 7 . 5 , 150 mM NaCl , 5 mM EDTA , 1% Triton X100 , protease inhibitor cocktail ( Roche Diagnostics ) ) together with acid-washed glass beads and cells were mechanically disrupted by bead-beating for 3 minutes . To test for Hog1 activation via phosphorylation , proteins were extracted as described by LaFayette et al . [18] . Whole cell protein samples were diluted 1∶10 in water and subjected to Bradford analysis to determine protein concentrations . For separation by 8% or 10% SDS-PAGE , protein concentrations were adjusted in 6× Laemmli buffer and lysis buffer , between 1 µg and 50 µg loaded ( Table S4 ) and separated at constant 120 Volts . Tub1 and Act1 served as loading controls . Details on antibodies used are provided in Table S4 . To purify Hsp90 and Cdc37 , Hsp90-TAP and Cdc37-TAP were immunoprecipitated with anti-IgG agarose as described by Singh et al . [19] , with two modifications: phosphatase inhibitors were added to the lysis buffer ( PhosStop , Roche Diagnostics ) and protein samples were incubated with IgG-agarose for 2 . 5 hours at 4°C . After the final wash , 40 µl of 2× Laemmli buffer was added and 3 µl and 30 µl of the protein samples were separated by 10% SDS-PAGE for hybridization with CaHsp90 , anti-TAP , PhosphoThreonine Q7 , and PhosphoSerine Q5 antibodies , respectively . Following separation , proteins were wet-transferred to a PVDF membrane ( Bio-Rad Laboratories , Inc . ) over night at 4°C and 30 V . Cdr1 was transferred for an additional hour at 100 V . Membranes were blocked for one hour in 1×PBS-T ( 1×PBS with 0 . 1% Tween 20 with 5% skimmed milk , washed for 5 minutes in 1×PBS-T and probed with the respective antibody . Primary antibodies were dissolved in 1×PBS-T , 2 . 5% skimmed milk and 0 . 003% sodium azide . All primary antibodies , except p38 MAPK and PhosphoThreonine Q7 , and PhosphoSerine Q5 , were left on the membrane for one hour at room temperature . p38 , PhosphoThreonine , and PhosphoSerine antibodies were incubated over night at 4°C . Primary antibodies were washed off twice with 1× PBS-T for ten minutes and the membrane probed for one hour with secondary antibody , dissolved in 1× PBS-T and 5% milk . The secondary antibody was washed off twice with 1× PBS-T for five minutes and once with 1× PBS for five minutes . PhosphoThreonine Q7 and PhosphoSerine Q5 hybridizations were conducted according to the manufacturers instructions . Following exposure and development , films were scanned and protein levels compared using ImageJ ( http://imagej . nih . gov/ij/index . html ) . To monitor gene expression changes in response to HSP90 depletion , strains SN95 , CaLC1411 , SN152 , CaLC2114 , and CaLC2115 were cultured as described above in preparation for protein extraction . To measure CKA1 and CKA2 expression levels , overnight cultures were diluted to an OD600 of 0 . 2 with or without 20 µg/ml doxycycline , grown for 24 hours and diluted again and cultured to mid-log phase . Upon reaching mid-log phase , RNA was then isolated using the QIAGEN RNeasy kit and cDNA synthesis was performed using the AffinityScript cDNA synthesis kit ( Stratagene ) . PCR was carried out using the SYBR Green JumpStart Taq ReadyMix ( Sigma-Aldrich ) with the following cycle conditions: 94°C for 2 minutes , and 94°C for 15 seconds , 60°C for 1 minute , 72°C for 1 minute , for 40 cycles . All reactions were done in triplicate using the following primer pairs: GPD1 ( oLC752/753 ) , HSP90 ( oLC754/755 ) , HOG1 ( oLC1968/1969 ) , CKA1 ( oLC1964/1965 ) , CKA2 ( oLC1966/1967 ) , MKK2 ( oLC1970/1971 ) , CDR1 ( oLC1972/1973 ) ( Table S3 ) . Data were analyzed in the StepOne analysis software ( Applied Biosystems ) . Strains SN152 ( WT ) , CaLC2114 ( ahr1Δ/ahr1Δ ) and CaLC2115 ( ahr1Δ/ahr1Δ::AHR1 ) were grown for 24 hours in YPD at 30°C with and without 10 µM GdA while shaking at 200 rpm . Cells were then imaged using Differential Interference Contrast microscopy using a Zeiss Axio Imager . MI microscope and images analyzed with Axiovision software ( Carl Zeiss , Inc . ) . To determine if protein levels differed significantly between high- and low-connectivity interactors , an unpaired t-test with Welch's correction was carried out . Expression level differences in low-connectivity interactors in response to Hsp90 depletion were evaluated with a one-way ANOVA with Bonferroni correction . Differences in HSP90 expression levels were assessed using paired t-tests . All analyses were done using GraphPad Prism 4 . 0 .
|
Hsp90 is an essential and conserved molecular chaperone in eukaryotes that assists with folding diverse proteins , especially regulators of cellular signaling . By activating signaling in response to environmental cues , Hsp90 has a profound impact on myriad aspects of biology . In fungi , Hsp90 influences development , drug resistance , and evolution . In the model yeast Saccharomyces cerevisiae , Hsp90 interacts with ∼10% of proteins . In the leading human fungal pathogen , Candida albicans , only two interactions have been identified . We conducted a chemical genetic screen to elucidate the C . albicans Hsp90 interaction network under diverse stress conditions . The majority of the 226 genetic interactors are important for growth under specific conditions , suggesting that they act downstream of Hsp90 and that the network is environmentally contingent . For example , the kinase Hog1 depends upon Hsp90 for activation . Only a few interactors are important for growth in many conditions , suggesting that they act upstream of Hsp90 . For example , the protein kinase CK2 regulates function of the Hsp90 chaperone machine and the transcription factor Ahr1 governs HSP90 expression . Thus , we identify novel effectors upstream and downstream of Hsp90 , and establish the first chaperone network of a fungal pathogen , with evidence for environmental contingency and network rewiring over evolutionary time .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetics",
"biology",
"genomics",
"evolutionary",
"biology",
"microbiology",
"genetics",
"and",
"genomics"
] |
2012
|
Mapping the Hsp90 Genetic Interaction Network in Candida albicans Reveals Environmental Contingency and Rewired Circuitry
|
How genetic programs generate cell-intrinsic forces to shape embryos is actively studied , but less so how tissue-scale physical forces impact morphogenesis . Here we address the role of the latter during axis extension , using Drosophila germband extension ( GBE ) as a model . We found previously that cells elongate in the anteroposterior ( AP ) axis in the extending germband , suggesting that an extrinsic tensile force contributed to body axis extension . Here we further characterized the AP cell elongation patterns during GBE , by tracking cells and quantifying their apical cell deformation over time . AP cell elongation forms a gradient culminating at the posterior of the embryo , consistent with an AP-oriented tensile force propagating from there . To identify the morphogenetic movements that could be the source of this extrinsic force , we mapped gastrulation movements temporally using light sheet microscopy to image whole Drosophila embryos . We found that both mesoderm and endoderm invaginations are synchronous with the onset of GBE . The AP cell elongation gradient remains when mesoderm invagination is blocked but is abolished in the absence of endoderm invagination . This suggested that endoderm invagination is the source of the tensile force . We next looked for evidence of this force in a simplified system without polarized cell intercalation , in acellular embryos . Using Particle Image Velocimetry , we identify posteriorwards Myosin II flows towards the presumptive posterior endoderm , which still undergoes apical constriction in acellular embryos as in wildtype . We probed this posterior region using laser ablation and showed that tension is increased in the AP orientation , compared to dorsoventral orientation or to either orientations more anteriorly in the embryo . We propose that apical constriction leading to endoderm invagination is the source of the extrinsic force contributing to germband extension . This highlights the importance of physical interactions between tissues during morphogenesis .
During development , many tissues extend in one orientation while narrowing in the orthogonal one . These so-called convergence and extension movements elongate the anteroposterior axis in bilateral animals during gastrulation , where they have been most studied [1–4] . Defects in convergence and extension movements at gastrulation have been linked to neural tube defects in mouse and human embryos [5] . Convergence and extension movements are also important later in embryo morphogenesis , for example for the elongation of the cochlear tube [6] , the kidney tubules [7] , and the limb and jaw cartilages [2] . Intracellular forces are key in convergence and extension and in most cases studied , drive polarized cell rearrangements [1 , 2] . These require planar polarization of proteins at cell membranes [3 , 8] . Planar polarization of actomyosin was first shown in Drosophila germband extension ( GBE ) to result in the selective shortening of dorsoventrally ( DV ) oriented cell contacts [9 , 10] . The cell biology of this process has since been extensively characterized , and planar polarization of several other components including Bazooka ( the homologue of Par-3 ) and E-cadherin have been found to be required for active cell rearrangements [11–20] . These polarities are controlled by the anteroposterior ( AP ) segmentation cascade in Drosophila , the most downstream genes being the pair-rule genes , encoding transcription factors such as Even-skipped and Runt [9 , 10 , 21] . Recent work has found that a combinatorial code of Toll-like receptors is required for transducing the AP positional information from these transcription factors into the planar polarities required for polarized cell intercalation [22] . Recently , actomyosin-driven shortening of cell contacts has also been shown to be essential for convergence and extension movements in vertebrates [7 , 23–25] . However , cell-autonomous behaviors might not be sufficient to fully explain axis elongation [26] . Stresses generated by neighboring morphogenetic movements or by the constrained geometry of the embryo could contribute to axis extension [27–29] . Evidence for extrinsic forces influencing tissue elongation has been reported: in Caenorhabditis elegans , body wall muscle contractions guide embryonic elongation [30]; in Drosophila oogenesis , the traction force produced by the follicle rotation is required for egg chamber elongation [31]; in the Drosophila developing wing , the contraction of the hinge produces a tensile stress that orients the cell behaviours required for wing blade elongation [32 , 33] . In the Drosophila embryo , we found previously that in addition to polarized cell intercalation , AP cell elongation contributes to GBE [34] . These cell shape changes are not a consequence of cell rearrangements: in the absence of polarized cell intercalation , the germband cells elongate even more in AP , a behavior most parsimoniously explained by an extrinsic tensile force acting on the tissue [34] . This gives us the opportunity to investigate how extrinsic factors can contribute to axis extension . Here , we search for the source of the extrinsic force acting on the germband by measuring the deformation of cells as a function of time , in the absence and presence of other morphogenetic movements . We find that blocking posterior endoderm invagination abolishes AP cell elongation . Furthermore , we present evidence that apical constriction leading to invagination of the posterior endoderm primordium produces a tensile force propagating from the posterior of the embryo . We conclude that this gastrulation movement at the posterior produces an AP tensile force contributing to the elongation of the main axis in Drosophila .
We analyzed apical cell shape changes using custom-made algorithms as previously [34 , 35] . We imaged embryos labeled with the junctional marker ubi-DE-cad-GFP on their ventral side by confocal time-lapse microscopy , acquiring images every 30 s at 20 . 5 ± 1°C , starting movies around morphological stage six and finishing around stage eight ( Fig 1A , 1A’ , 1C and 1C’ ) . We segmented apical cell contours based on the fluorescent signal and linked cells in time , storing the coordinates of the centroid of each cell and of a polygon describing its outline , at each timepoint ( Fig 1D and 1D’ ) . To measure the cell shape changes , our algorithms consider small cell neighborhoods consisting of a central cell surrounded by one ring of its immediate neighbors ( Fig 1B ) . Cell shapes for this neighborhood are measured by fitting an ellipse to each cell: strain rates are calculated over a 2 min window ( ±2 timepoints , see Fig 1B ) . To analyze specifically the AP component of cell shape change ( the component that will contribute to axis extension ) , the strain rates were projected onto the AP embryonic axis . In our summary plots , we call this strain rate “AP cell length change , ” expressed in proportion per minute ( pp/min ) ( Fig 1E–1F’ ) and shorten it to “AP cell elongation” in the text thereafter . Note that from our measures of strain rates , we can also extract DV cell elongation and cell area change ( see below ) . To consider only the deformation of cells from the germband ( the tissue undergoing convergence and extension ) , we excluded any tracks from mesoderm and mesectoderm cells ( Fig 1D and 1D’ ) . These methods allow us to examine the patterns of AP cell elongation in living embryos , which we proposed to be a signature of an extrinsic force contributing to axis extension [34] . We had previously analyzed AP cell elongation in field of views that included the cephalic furrow as an anterior landmark ( the cephalic furrow forms between the head and the germband ) [34] ( Fig 1A , 1A’ and 1C ) . These views show the anteriormost region of the ventral side of the germband and are thereafter called “anterior views” for simplicity . When visualizing AP cell elongation as a function of time and position along the AP axis in spatiotemporal heat maps , we noticed that the signal was higher towards the posterior edge of the field of view [34] ( average for five movies , Fig 1E; individual movies , S1A Fig; tracking information , S1C and S1C’ Fig ) . This prompted us to image the ventral side of embryos more posteriorly , using the tail end of the embryo ( as detected in apical optical sections ) as a posterior landmark ( Fig 1A’ and 1C’ ) . Plotting spatiotemporal maps for these “posterior views” revealed that AP cell elongation becomes even stronger closer to the posterior tip of the embryo ( average for four movies , Fig 1E’; individual movies , S1B Fig; tracking information , S1D and S1D’ Fig; example S1 Movie ) . Indeed , although AP cell elongation peaks around 10 min after GBE onset in both views , the magnitude is doubled in posterior views: 0 . 04 pp/min ( average for four movies , Fig 1F’ ) compared to 0 . 02 pp/min in anterior ones ( average for five movies , Fig 1F ) . Note that to be able to make fair comparisons between anterior and posterior views , we removed the tracks of ectodermal cells deformed by the cephalic furrow in anterior views ( purple shaded region in Fig 1C , resulting tracks in Fig 1D ) , since these unrelated cell deformations would otherwise contribute to our measure of total AP cell elongation , as they did in our previous study [34] . All anterior views presented in this paper have been reanalyzed with this exclusion . We estimated that in wild-type embryos , the two fields of view overlapped by about 80 microns ( Fig 1A’ ) , and we concluded that the patterns of AP cell elongation detected in posterior views fully included the patterns seen in anterior views ( Fig 1E and 1E’ ) . The AP cell elongation patterns appeared to form a gradient increasing from the anterior to the posterior . To ascertain this , we examined a short period around the peak of AP cell elongation , from 7 . 5 min to 12 . 5 min after GBE onset ( Fig 2 ) . This confirmed that AP cell elongation increased steeply towards the posterior of the embryo ( Fig 2A–2D ) , forming a gradient over a distance of about 150 μm in posterior views ( average for four movies , Fig 2D ) . Although the gradient is clearest in posterior views , some AP gradation was already detectable in anterior views ( average for five movies , Fig 2C ) , consistent with the notion that we are visualizing the beginning of the gradient in anterior views . In posterior views , we also looked at snapshots of the gradient earlier in GBE , at 2 . 5 , 5 , and 7 . 5 min: the gradient was at first shallow and confined to the more posterior part of the field of view; it then expanded towards the anterior and became steeper with time ( Fig 2E ) . These results suggested that a tensile stress deformed the tissue from a posterior source , starting at the onset of GBE and propagating towards the anterior of the embryo over time . We also analyzed cell area change in addition to AP cell elongation ( S2A and S2A’ Fig ) . When passively responding to planar extrinsic forces , cell apical areas are expected to change in opposite ways depending on whether cells are compressed or pulled: when pulled , cell areas should increase; in contrast , when compressed , cell areas should decrease . We had already noted in our previous study that AP cell elongation was accompanied by an increase in cell area in anterior views , supporting the idea that the germband was experiencing a planar tensile stress [34] ( S2A Fig ) . This trend is even clearer for the posterior views: the patterns of AP cell elongation are matched by patterns of cell area increase , suggesting that the germband cells elongated in response to a tensile rather than compressive stress ( compare S2A’ Fig with Fig 1E’ ) . Note that in our analyses , we can observe changes in only the two planar axes defining the apical cell areas , but we expect the third axis , the cell length in Z , to increase or decrease in response to planar stress to keep the cell volume constant [37 , 38] . Around the onset of GBE , the germband cells are also subjected to a pull in the perpendicular direction , along DV , in response to the invagination of the mesoderm on the ventral side of the embryo [34] ( mesoderm invaginates through a ventral furrow visible in Fig 1A’ , 1C and 1C’ ) . In both anterior and posterior views , we found that DV elongation of ectodermal cells in response to mesoderm invagination have patterns completely distinct from the AP cell elongation patterns that we are focusing on in this study: first , they are most prominent close to GBE onset and have disappeared by 10 min into GBE ( whereas the AP cell elongation patterns peak just after 10 min ) , and second , they occur uniformly along the AP axis of the embryo ( whereas the AP cell elongation patterns occur in a posterior gradient ) ( S2B–S2C” Fig ) . Note that AP and DV cell elongation patterns are both accompanied by an increase in cell area ( S2A and S2A’ Fig ) , consistent with the idea that they are both the consequence of tensile forces . We concluded that germband cells are subjected to two independent tensile forces , one in the DV direction caused by mesoderm invagination ( see also below ) , and one in the AP direction coming from the posterior of the embryo . Together , our analysis of wild-type Drosophila embryos indicated that AP cell elongation formed an AP gradient consistent with a stress propagating from the posterior . We asked next what the origin of this tensile force was . A stress propagating from the posterior seemed at odds with our previous model suggesting a role for mesoderm invagination in generating AP patterns of cell elongation [34] . This model was based on the analysis of anterior views , where we had previously found that AP cell elongation contributing to axis extension was reduced in twist ( twi ) mutants , which are defective for mesoderm invagination . Although we had proposed at the time that mesoderm invagination might contribute to the extrinsic tensile force deforming the germband , it was difficult to formulate a model for how it could do so [29 , 34] . We reanalyzed the data from anterior views after exclusion of the region deformed by the cephalic furrow ( see above ) . We confirmed our previous results: in anterior views , AP cell elongation was significantly reduced in twi mutants compared to wild type ( average for five movies , Fig 3A and 3A’; individual movies , S3A Fig; example S2 Movie ) . Next , we acquired new movies imaging the posterior ventral side of the embryo , using the posterior end of the imaged embryo as a landmark , as before for wild type . To our surprise , we found robust AP cell elongation in posterior views of twi embryos , with no statistical difference between the rate of AP cell length change between these mutant embryos and wild type ( average for three movies , Fig 3B and 3B’; individual movies , S3B Fig ) . Elongating cells tended to increase in area in these posterior views , suggesting that they elongated in response to a tensile stress , as in wild type ( S3C’ Fig ) . Note that in these cell area plots , the cell area increase in response to mesoderm invagination is absent ( 0 to 5–7 min ) , in posterior as in anterior views , demonstrating that the embryos we imaged are indeed defective for mesoderm invagination ( compare S2A Fig with S3C Fig , and S2A’ Fig with S3C’ Fig ) . Further demonstrating this , DV cell elongation is gone in anterior and posterior views of twi mutant embryos ( S3D , S3D” , S3E and S3E” Fig , compare with S2B , S2B” , S2C and S2C” Fig ) . This shows that whereas the early DV stretch of ectodermal cells is gone as expected in twi mutants ( because there is no mesoderm invagination to pull the ectoderm in DV ) , the AP stretch of ectodermal cells is still present in posterior views ( S3E and S3E’ Fig ) . This confirmed that DV and AP cell elongation were produced by two independent tensile forces , and that mesoderm invagination caused DV cell elongation in the germband . Refuting our previous model [34] , this also indicated that mesoderm invagination did not cause the AP cell elongation contributing to GBE . As before for wild type , we examined the gradient of AP cell elongation between 7 . 5 and 12 . 5 mins and confirmed that there is a significant difference with wild type for anterior views but no clear statistical difference when comparing posterior views ( Fig 3C–3G” ) . This discrepancy suggested that the relative position of anterior and posterior fields of view are different in wild-type and twi mutants , leading to the detection of the AP cell elongation gradient in posterior views , but not in anterior views , in twi mutants . This is likely to be the result of several factors , one of which might be a difference in curvature on the ventral side of the embryo between the two genotypes . Indeed , we find that the outlines of twi embryos are less curved than wild-type ones in anterior views , and the embryos are wider , consistent with the notion that twi embryos are flatter ( S4 Fig ) . A flatter ventral surface in twi mutants would make the posterior views more posteriorly located in twi mutants , because the position of the posterior landmark we use ( the tip of the embryo in optical sections ) will be influenced by curvature . A flatter surface could be a direct consequence of the failure of mesoderm invagination and the absence of a keel-like shape in twi mutants . Absence of invaginating mesoderm could not only affect the curvature of the embryo , but also change its mechanical properties and , for example , make it flatten more under a coverslip during imaging . Both factors would make the anterior and posterior fields of view further apart in twi mutants compared to wild type . We concluded that a gradient of AP cell elongation was present in twi mutants and grossly similar to wild type in posterior views , showing that an event other than mesoderm invagination must be responsible for the AP extrinsic force deforming the germband . We reasoned that candidates for generating a tensile stress at the posterior would be morphogenetic movements taking place at , or just before , the onset of GBE , because germband cells start to elongate in AP from the beginning of GBE [34] ( Fig 1E and 1E’ ) . To identify such events , we measured the timings of gastrulation movements relative to the start of GBE ( Fig 4 ) . Because some movements take place on the ventral surface ( mesoderm invagination ) and others on the dorsal surface ( endoderm invagination , dorsal folding ) ( Fig 4A , see also Fig 1A and 1A’ ) , we used light sheet microscopy ( SPIM , selective plane illumination microscopy ) to image the whole embryo volume through developmental time [39] . We labelled the cells with plasma membrane markers such as Spider-GFP and Resille-GFP and took timepoints every 30 sec ( at 28–30°C ) . We examined three wild-type movies and three twi mutants defective for mesoderm invagination ( Fig 4B ) . We mapped the onset of GBE by identifying the first posteriorward displacement of ventral cells ( Fig 4C and 4C’ , and S3 Movie ) and used the corresponding time-point as time zero for all the movies . To check that the development rates of all embryos imaged were comparable , we used patterned mitoses in the head as a developmental timer ( Fig 4D– 4D” ) [40] . We found that these mitoses start at 8 . 5 min , 10 . 5 min , and 11 . 5 min after GBE onset in the three wild-type movies and at 10 . 5 min , 11 min , and 12 min in the three twi movies ( Fig 4B ) . This showed that there were no obvious differences in development rates between embryos and illustrates the temporal reproducibility of Drosophila early development . Next , we mapped the timings of morphogenetic movements visible in the movies ( Fig 4A ) ( for a review of the anatomy of these movements , see [29] ) . We concluded that the two morphogenetic movements most synchronous with GBE onset were mesoderm and posterior endoderm invaginations ( Fig 4B ) . We mapped the onset of posterior endoderm invagination ( also called posterior midgut invagination ) by identifying in which movie frame the cells initiated apical constriction at the posterior of the embryo ( Fig 1E and 1E’ and S3 Movie ) . Posterior midgut invagination preceded GBE by −3 . 5 , −2 , and −1 . 5 min in the three wild type , and by −2 , −1 . 5 , and −0 . 5 min in the three twi mutant embryos ( Fig 4B ) . To map a clearly identifiable step of mesoderm invagination , we recorded the timepoint when the right and left sides of the mesoderm first met to begin forming the internal mesodermal tube , thereafter called “mesoderm sealing” ( Fig 4F and 4F’ and S4 Movie ) . The times relative to the onset of axis extension were −0 . 5 min , −0 . 5 min , and +0 . 5 min for the 3 wild type movies ( twi embryos fail to form a mesodermal tube ) ( Fig 4B ) . This confirms a remarkable synchrony between mesoderm sealing and GBE onset , which we had noted before [34] ( see Discussion ) . We also looked at morphogenetic movements that occur on the dorsal side of the embryo . Dorsal folding occurs in two stereotyped locations under the control of the AP patterning system [41] . Although these folds start forming just before the onset of axis extension in wild type embryos , they initiate after GBE onset in two out of three twi embryos ( Fig 4G and 4G’ ) . Since AP cell elongation at the posterior end of the embryo are already high in twi mutants at GBE onset ( Fig 3B ) , this suggests that the dorsal folds are not initiating these ( although they could later contribute ) . Other dorsal movements include a dorsal contraction ( Fig 4H , 4H’ and 4B ) and the onset of amnioserosa cell flattening [42] . These occur respectively too early and too late , relative to the onset of GBE , to be key influences . We conclude from this temporal mapping of morphogenetic movements that both mesoderm sealing and endoderm invagination are synchronous with the onset of GBE . Since we have refuted a role of mesoderm invagination in producing the gradient of AP cell elongation contributing to axis extension ( see previous section ) , posterior endoderm invagination was the main candidate to generate a tensile stress during GBE . To test a role of posterior endoderm invagination in AP cell elongation during axis extension , we examined folded gastrulation ( fog ) and torso-like ( tsl ) mutants that abolish endoderm invagination . Fog is a zygotic gene required for the apical constriction of the endoderm primordium cells arranged in a disc at the posterior , which leads to posterior midgut invagination [43] . The expression of fog in the posterior midgut primordium requires the zygotic gap genes huckebein and tailless , which themselves require the activity of the maternal gene tsl , an upstream component of the terminal patterning system [44] . In anterior views , no obvious AP cell elongation gradient was detected at the onset of GBE in fog mutants ( compare S5B Fig with Fig 1E ) . However , fog mutants proved problematic to analyze because their extending germband form ectopic folds ( arrows in S5A and S5B Fig and S7 Movie ) . These folds occur because the posterior end of the germband does not move away in these mutants , but polarized active cell intercalations still elongate the germband [43] . Folding stretched the germ-band cells locally and produced strong AP cell elongation , as seen on spatiotemporal maps from approximately 7 min after GBE onset ( S5B Fig ) . As a consequence , the total AP cell elongation could not be meaningfully compared between wild-type and fog mutants . To prevent folding , we analyzed one of the mutants that abolishes posterior midgut invagination , tsl , in combination with a mutant abolishing most of the active polarized cell intercalations in the trunk , Kruppel ( Kr ) [21 , 34] . To ask if tsl was required for the gradient of AP cell elongation , we compared Kr single mutants with these Kr; tsl double mutants . In posterior fields of views , AP cell elongations are slightly higher in Kr compared to wild type ( Fig 5A ) . This was expected , since AP cell elongation increases in the absence of cell intercalation , presumably because in wild type , polarized cell intercalation acts to release some of the tensile stress in the germband [34] . The patterns of AP cell shape changes are , however , comparable in both genotypes ( average for three movies , Fig 5B , compare with Fig 1E’; individual movies S5C Fig; example S5 Movie ) . As in wild type , the AP cell shape changes are accompanied by an increase in cell area , consistent with a tensile rather than compressive stress ( Fig 5H; individual movies in S5D Fig ) . In double mutants Kr; tsl however , very little AP cell length change was detected ( average for three movies , Fig 5C and 5D; individual movies S5E Fig; example S6 Movie ) . Note that the residual AP cell length change detected on the averaged spatiotemporal map ( Fig 5C ) was mainly present in one of the three individual movies ( krtslCL040713 , S5E Fig ) , and this signal was not accompanied by an increase in cell area , as would be expected for a tissue under tensile stress ( S5F Fig ) . Consistent with this , there was no significant increase in cell area detected in double mutants Kr; tsl in the other two movies or in the averaged data ( S5 Fig and Fig 5I ) . This indicated that the ectodermal cells in the posterior region of Kr; tsl mutants embryos were not under tensile stress . We also examined in more detail the AP cell elongation gradient around its peak ( from 7 . 5 to 12 . 5 min ) , in Kr versus Kr; tsl mutants . The steep gradient of AP cell elongation was abolished in Kr; tsl mutants ( Fig 5E–5G ) . We concluded that posterior midgut invagination was required for the AP cell elongation contributing to axis extension in Drosophila . To understand more precisely how posterior endoderm invagination could produce a stress that in turn leads to a gradient of AP cell elongation , we analyzed a simplified system , in the form of acellular mutant embryos . Several mutations are known that produce embryos , which fail to cellularize . In one such mutant , an endoderm-like invagination is still visible on the dorsal side of the embryo , suggesting that apical constriction of the endoderm primordium still occurs in acellular embryos [45] . Consistent with this notion , another acellular mutant was shown recently to undergo apical constriction of the mesoderm primordium , albeit at a slower rate ( about 60% of the wild type ) [46] . To confirm that apical constriction also happened for the endoderm primordium , we made movies of these acellular mutants expressing sqh-GFP [46] , to visualize the actomyosin cytoskeleton ( sqh encodes the non-muscle Myosin II Regulatory Light Chain ) ( S8 Movie ) . We observed a concentration of Myosin II in the region where apical constriction would normally occur in the presumptive posterior endoderm , close to where the pole cells ( PC ) are attached , in both live and fixed embryos ( Fig 6A and 6B , and S6D’ Fig ) . We find that the acellular embryos go through the initial steps of wild-type endoderm invagination [43] , with first the formation of a flattened plate at the posterior ( S6D’ Fig ) , then constriction of the embryo’s surface leading to some degree of invagination ( Fig 6A–6C ) ( see also Fig 3a in [46] ) . In live embryos , we noticed that the concentration of Myosin II at the posterior is accompanied by flows of Myosin II towards it ( S8 Movie , top panel ) . This suggested that apical constriction of the presumptive endoderm surface could exert a tensile stress on the surrounding apical surface of the embryo . We also saw flows towards the ventral region , presumably in response to apical constriction of the presumptive mesoderm . We confirmed the direction of these flows by tracking the Myosin II signal at the surface of the acellular embryos using Particle Imaging Velocimetry ( PIV ) . In our example movie showing the whole lateral surface of the presumptive germband , we can clearly see by PIV both ventralward ( towards mesoderm ) and posteriorward ( towards posterior endoderm ) flows of Myosin II signal ( S8 Movie , bottom panel ) . To confirm the existence of posterior flows , we acquired more movies of the posterior end of the embryo and visualized the flows by PIV . We found that all embryos analyzed showed posteriorward flows towards the presumptive posterior endoderm ( n = 8 , 2 examples in Fig 6A’–6B’ ) . To understand better how the Myosin II flows relate to the surface membranes of the acellular embryos , we compared the localization of Myosin II with those of the E-cadherin complexes . Just before gastrulation movements started , E-cadherin and Myosin II colocalized in a hexagonal-like pattern ( estimated stage 5; S6A , S6A’ , S6C and S6C’ Fig ) . These presumably correspond to the regions of the surface membrane that , in wild-type embryos , would normally invaginate and become furrow canals encircling each syncytial nucleus ( for example , see [47] ) . Once gastrulation movements started in acellular embryos , this relatively regular organization became disrupted: E-cadherin and Myosin II still colocalized but now formed a meshwork at the surface of the embryo ( estimated stage 7; S6B , S6B’ , S6D and S6D’ Fig ) . Since E-cadherin complexes are presumably associated with membranes , we infer that Myosin II flows that we observe track the movement of surface membranes in these embryos . The presence of posteriorward flows of Myosin II signal in acellular embryos suggested that apical constriction of the presumptive endoderm was able to pull the apical surface behind it and could generate an AP tensile stress , which in wild-type embryos could contribute to axis extension . We reasoned that acellular embryos provided an excellent system in which to physically probe this tension , since it is unlikely to exhibit more complex morphogenetic behaviours such as polarized cell intercalation , and so we could rule out a contribution of the latter to measured tensions . No planar polarization of Myosin II was recognizable in these embryos , confirming that the apical surface of the embryo was unlikely to undergo intercalation-like movements ( S6A–S6D’ Fig ) . To directly test our hypothesis that apical constriction of the endoderm primordium generated a tensile stress at the posterior , we carried out line ablations at the surface of the embryo . Using a near-infrared laser , we made 20 micron-long incisions oriented parallel to the AP or DV embryonic axes and at the posterior or the anterior of the presumptive germband , on the lateral side of sqh-GFP-labelled acellular embryos ( Fig 6D and S6E Fig ) . If , as we proposed , a tensile stress propagated from the posterior endoderm , we predicted that the DV cuts at the posterior should show a faster relaxation than any of the other three types of cuts . We used fine-grained PIV to track the movement of the Myosin II network , as a proxy for surface motion , and measured the velocities of recoil in a small region around the cuts , subtracting the velocity of that region before the cut to correct for translation ( see Materials and Methods ) ( S6F–S6G’ Fig ) . We found that , as predicted , the average relaxation velocity of the DV cuts at the posterior was significantly higher than for any of the other cuts ( Fig 6E ) . At the posterior , there was a clear anisotropy in the relaxation velocities , the DV-oriented cuts relaxing much faster than the AP-oriented cuts , whereas at the anterior , there was no statistically significant anisotropy . This provided evidence for an increased AP-oriented tension at the posterior of the embryo , in response to apical constriction leading to invagination of the endoderm primordium in acellular embryos .
We have investigated the origin of the patterns of planar cell shape changes that we hypothesized previously were the signature of an extrinsic force acting during Drosophila axis extension [34] . We showed that the AP-oriented elongation of cell apices contributing to GBE are strongest at the posterior end of the embryo and decrease gradually towards the anterior . AP cell elongation is accompanied by an increase in cell area , suggesting that this gradient of cell shape change arises in response to a planar tensile stress coming from the tail end of the embryo . We found that the patterns of AP cell elongation and cell area increase are eliminated in the absence of posterior endoderm invagination ( but not mesoderm invagination ) , suggesting that this morphogenetic movement is the source of the extrinsic force deforming the germband . We show that in acellular embryos , the cortical Myosin II meshwork flows towards the contracting posterior endoderm region , and that this is accompanied by an increased tension at the posterior . We conclude from these experiments that the apical constriction and invagination of the posterior endoderm primordium generates a tensile stress propagating to the germband and causing the AP cell elongation gradient that contributes to Drosophila axis extension ( Fig 7 ) . We can think of two alternative explanations that could challenge this conclusion . First , AP cell elongations could be a secondary consequence of active cell intercalation . However , in AP patterning mutants such as Kr , where active polarized cell intercalation is diminished , AP cell elongation is increased rather than decreased [34] . This indicates that active cell intercalation ( and AP patterning ) is not required for AP cell elongation . Also , cell intercalation rates are high throughout the trunk [34] , whereas AP cell elongation is found in a gradient culminating at the posterior ( this paper ) . Therefore , these differing spatial patterns suggest that these two cell behaviours have independent origins . Also in acellular embryos , we observe posteriorward flows of the apical cortex associated with increased tension at the posterior , in absence of polarized cell intercalation . Together , this argues that polarized cell intercalation is not responsible for the gradient of AP cell elongation we observe . Another possibility is that AP cell elongations are cell-autonomous , that is the result of an active spreading of the germband cells under the control of a genetic program . AP patterning is not required ( see above ) , and the other patterning systems known to operate in the early embryo are the DV and terminal systems [44] . The observed gradient of AP cell elongation is orthogonal to the DV patterning axis and extends outside the terminal domain , so it cannot be explained simply by the activity of either of these systems . We conclude that the most parsimonious explanation is that the AP cell elongation patterns we observe are passive cell behaviours that occur in response to mechanical stresses . We have found that the AP cell elongation gradient is still present in twi mutants in posterior views , refuting our previous model for a role of the mesoderm in producing these cell shape changes , which was based on analyzing anterior views [34] . We think that the source of the discrepancy is that the anterior and posterior views we imaged are further apart in twi mutants compared to wild-type , which means that the AP cell elongation gradient was mostly missed in twi anterior views . We identify at least one factor , curvature , to explain this difference . The difficulty in registering fields of view between these two genotypes precludes a more detailed comparison of the AP cell elongation gradient . Therefore , we cannot rule out a subtle contribution of mesoderm invagination to GBE . For example , mesoderm invagination , by changing the shape and perhaps the mechanical properties of the germband , might affect how the stress from endoderm invagination propagates throughout the ectoderm . This has some support from the analysis of the AP cell elongation gradient’s slope at specific timepoints , which appear shallower in twi mutant ( see for example timepoint 7 . 5 min in Fig 3G” ) . To be able to compare the gradient of AP cell elongation between the two genotypes , we will need to perform apical cell deformation analysis in whole embryo movies such as the SPIM movies presented in this paper , in order to circumvent the problem of registering fields of view . Our experiments identify the endoderm primordium as a source of tensile force . Using acellular embryos allowed us to explore how mechanical stresses could be produced by the posterior endoderm . Although they do not have cells , these mutant embryos are able to undergo the initial steps leading to both mesoderm [46] and endoderm invagination ( this study ) . The apical surfaces of the embryo corresponding to the mesoderm and endoderm primordia are seen to enrich Myosin II , contract , and begin to invaginate ( [46] , this study ) , as in wild-type embryos [48] . A rigorous quantitative analysis on the mesoderm has demonstrated that the apical forces of constriction are transmitted to the underlying cytoplasm deep in the tissue and are sufficient to promote invagination , showing that cell individualization is dispensable , at least for the initial phases of invagination [46] . Our qualitative study suggests that the forces generated by apical constriction are also transmitted in the plane at the surface of acellular embryos . Using PIV , we visualized surface flows of Myosin II towards the mesoderm and endoderm primordia . Our laser ablation experiments indicate that the flows towards the endoderm primordium are accompanied by an increase in tension at the cortical surface of the acellular embryo . This suggests that apical cell constrictions of the endoderm primordium and the beginning of invagination are able to produce planar forces that pull the adjoining apical surfaces of the germband . How do stresses transmitted from the apical cortex of constricting endodermal cells translate into a gradient of AP cell elongation in the elongating germband ? Epithelial cells of the germband are thought to be connected mechanically to each other through the actomyosin cytoskeleton interacting with components of the apical adherens junctions such as the E-cadherin complexes [29 , 49] . Thus , tensile stresses caused by apical constriction should propagate through tissues and can conceivably result in mechanically stretching cells over some distance . We find here that germband cells elongate in AP over a distance of at least 150 μm from the site of endoderm constriction ( See Fig 1E’ ) . The gradation in AP cell elongation in response to endoderm invagination that we observe might be explained by friction or resistance from the cellular environment . These would prevent forces being instantaneously propagated throughout the germband . Since the germband tissue has to go around the posterior tip of the embryo to elongate , geometry might also have an impact on how forces are transmitted . Finally , we cannot exclude that spatial variation in stiffness of germband cells along the AP axis could cause them to respond differently to mechanical stress . Endoderm and mesoderm invagination are both triggered by apical constrictions powered by apical networks of actomyosin [48] . We previously detected a stretch of the ectodermal cells in DV behind the invaginating mesoderm [34] . We confirm this in this paper , showing that DV elongation of the germband cells occurs for the first 5–7 min of GBE in wild-type . This is abolished in twi mutants in which mesoderm invagination is defective . Thus , germband cells are subjected to two independent tensile forces: one in the DV direction ( around the onset of GBE ) caused by mesoderm invagination , and another in the AP direction ( during early GBE ) , caused by posterior endoderm invagination . Together , these observations show that the epithelial cells in the germband can respond passively to tensile stress generated in adjacent tissues apically constricting and invaginating , by stretching along the direction of stress . The directionality of apical cell elongation is strongly constrained to AP for the patterns linked to endoderm invagination and to DV for those linked to mesoderm invagination . Indeed , the patterns of AP cell length change caused by endoderm invagination are not accompanied by much change in DV cell length and vice versa for mesoderm invagination ( Compare S2C’ and S2C”Fig ) . Since both AP and DV cell elongation patterns are accompanied by an increase in cell area ( S2A’Fig ) , this implies that the germband cells must shorten their z-axis if they are to maintain a constant cell volume . The maintenance of a constant cell volume throughout gastrulation appears likely , based on recent measurements [37 , 38] . We cannot access the Z dimension with our analyses of apical cell surface deformation and so verifying that cells do shorten along their z-axis will require tracking and analyzing cell shape changes in 3-D . We had shown previously that the AP cell elongation patterns that we are observing in the germband contribute to axis extension [34] . This was shown by measuring strain rates ( deformation ) for the whole tissue and decomposing these into the strain rates caused by the cell length change and the strain rates caused by polarized cell intercalation [34 , 35] . We found that although the predominant behavior extending the germband is polarized cell intercalation , AP cell length changes are contributing significantly ( about one-third of the total deformation ) early in GBE . A question that remains is why AP cell elongation is temporally limited to early GBE , peaking around 10 min after the onset of GBE ( Fig 1F and 1F’ ) . In fact , AP cell elongation ceases rather abruptly at around 15 min after GBE onset ( Fig 1E’ ) . SPIM movies indicate that this developmental time ( taking into account the difference in temperature for the acquisition of these movies , see Materials and Methods ) corresponds to when the posterior midgut invagination becomes cup-shaped and appears to drop down from the surface of the embryo ( S3 Movie; see schematics in Fig 4A and Fig 7 ) . A possibility is that force generation from endoderm invagination ceases at this time , perhaps because apical constrictions in the primordium are completed . Alternatively , the presence or not of AP cell elongation in the germband could be a function of the balance between how much the actively elongating germband can push and how much endoderm invagination can pull . In other words , early , the pull from endoderm invagination might be stronger than the push from the extending germband , causing a stress in the germband tissue , which manifests as AP cell elongation . Late , the push from GBE versus the pull from endoderm invagination might be balanced: germband cells would not experience stress anymore and would cease to elongate . In addition to producing cell shape changes contributing to axis extension , does the endoderm-generated tensile force have other roles in axis extension ? The posterior pole of the embryo does not move dorsally in fog and tsl mutants and is associated with a buckling of the germband [43] . A possible interpretation of this phenotype is that the actively extending germband cannot intrinsically “push” round the corner ( or displace presumptive endoderm ) . So endoderm invagination may have the role of guiding the germband around the posterior tip to overcome the obstacles posed by the surrounding tissues and the embryo geometry . The tensile stress from the endoderm might also facilitate polarized cell intercalation . Whereas DV-shortening of junctions is known to be caused by the intrinsic activity of the actomyosin cytoskeleton , it remains unclear how the AP-oriented nascent junctions elongate at the end of intercalation [9–19 , 22 , 26] . A possibility is that the extrinsic tensile force from the endoderm facilitates this AP junctional elongation either by directly exerting tension on the junctions or by indirectly “making space” for cells to intercalate , or in other words by displacing the boundary ahead of the self-deforming tissue [50] . It is also possible that an AP tensile stress could contribute to the nonreversibility of cell intercalation . Finally , it is remarkable that three morphogenetic movements principally driven by cell-autonomous behaviours , GBE by polarized cell intercalation , and mesoderm and endoderm invagination by apical constriction , are happening so synchronously ( Fig 4B ) . Furthermore , these movements are controlled by three distinct patterning systems: AP , DV , and terminal , respectively , that are understood to function independently of each other at these early stages [44] . It is unclear how the embryo can synchronize these three movements so precisely . One possibility is that there is a yet-undiscovered genetic cross talk between these pathways . However , our findings suggest an alternative explanation , that mechanical coupling between the invaginations of gastrulation and axis extension helps this synchronization . In vertebrate embryos , convergence and extension movements also happen at the same time as other morphogenetic deformations , for example epiboly [3] or neurulation [5] , so understanding how morphogenetic movements interact is going to be important to fully understand how embryos are shaped .
Transgenic strains were spider-GFP , resille-GFP [51] , ubi-DE-cad-GFP [52] and sqh-GFP[53] . Mutant alleles were Kr [1] , twi [1] , tsl [4] , the tsl deficiency Df[3R]ED6076 ( Flybase and Bloomington Stock Centre ) , and the acellular mutant characterized in [46] . Anterior movies are taken from [34] . Posterior movies were acquired as follows: late stage five embryos labeled with ubi-DE-cad-GFP were imaged ventrally every 30 sec at 20 . 5 ± 1°C , using a spinning disc confocal . Cell tracking , cell shape , and cell area analyses were performed as before using custom software ( oTracks ) written in IDL [34 , 35] . Best-fit ellipses are used to represent cell shapes and to calculate cell deformation . For statistics , we used a mixed-effects model as before [34] . Late stage five embryos labeled with spider-GFP and/or resille-GFP were mounted in 1 . 5% low melting point agarose and imaged using mSPIM [39] . Embryos were rotated to image four perpendicular views , which were reconstructed into a whole embryo image stack post-acquisition [54] . Image stacks were acquired every 30 sec at 28–30°C for 60 min . Reconstructed movies of three wild-type and three twi mutants were viewed in 4-D in custom software ( Browser and Tracer ) written in Interactive Data Language ( IDL , Exelis ) [55] to map timings of morphogenetic movements . Scatter graphs were plotted in Prism ( GraphPad ) . PIV was performed to visualize Myosin II flows at the embryo scale and also at a smaller scale to analyze relaxation of the tissue after laser ablation in acellular embryos . Further details on the Materials and Methods are in S1 Text .
|
Embryos change shape dramatically during development . The genetic programs that drive the active behavior of cells underlying these changes are well understood , but little is known about how movements of neighboring tissues influence the shaping of a given tissue . We address this question for the anteroposterior elongation of the body axis ( germband ) of Drosophila embryos . We had previously shown that during elongation , the germband cells stretch along the anteroposterior axis , in addition to undergoing active rearrangements; this suggested that extrinsic tensile forces might be at play . In the current study we find that the start of main body elongation is synchronous with the invagination of both the mesoderm and the endoderm . We analyze mutants and find that cell stretching disappears in embryos lacking endoderm invagination but remains in those lacking mesoderm invagination . We then measure tension using laser ablation in acellular embryos that lack active cell rearrangements in the germband but undergo the initial stages of endoderm invagination . We find that tension is higher in the anteroposterior direction close to the invaginating endoderm . Our results indicate that endoderm invagination generates a tensile force that is transmitted to the germband , and contributes to its elongation . This study reveals how tissues interact during embryo morphogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Mechanical Coupling between Endoderm Invagination and Axis Extension in Drosophila
|
Pneumonic plague is a fatal disease caused by Yersinia pestis that is associated with a delayed immune response in the lungs . Because neutrophils are the first immune cells recruited to sites of infection , we investigated the mechanisms responsible for their delayed homing to the lung . During the first 24 hr after pulmonary infection with a fully virulent Y . pestis strain , no significant changes were observed in the lungs in the levels of neutrophils infiltrate , expression of adhesion molecules , or the expression of the major neutrophil chemoattractants keratinocyte cell-derived chemokine ( KC ) , macrophage inflammatory protein 2 ( MIP-2 ) and granulocyte colony stimulating factor ( G-CSF ) . In contrast , early induction of chemokines , rapid neutrophil infiltration and a reduced bacterial burden were observed in the lungs of mice infected with an avirulent Y . pestis strain . In vitro infection of lung-derived cell-lines with a YopJ mutant revealed the involvement of YopJ in the inhibition of chemoattractants expression . However , the recruitment of neutrophils to the lungs of mice infected with the mutant was still delayed and associated with rapid bacterial propagation and mortality . Interestingly , whereas KC , MIP-2 and G-CSF mRNA levels in the lungs were up-regulated early after infection with the mutant , their protein levels remained constant , suggesting that Y . pestis may employ additional mechanisms to suppress early chemoattractants induction in the lung . It therefore seems that prevention of the early influx of neutrophils to the lungs is of major importance for Y . pestis virulence . Indeed , pulmonary instillation of KC and MIP-2 to G-CSF-treated mice infected with Y . pestis led to rapid homing of neutrophils to the lung followed by a reduction in bacterial counts at 24 hr post-infection and improved survival rates . These observations shed new light on the virulence mechanisms of Y . pestis during pneumonic plague , and have implications for the development of novel therapies against this pathogen .
The recruitment of neutrophils is a fundamental component of the initial phase of the innate immune response to bacterial lung infections , as demonstrated by the selective depletion of neutrophils and the consequences on pathogen clearance from the lungs [1] . In response to infection , neutrophils are mobilized from the bone marrow ( BM ) , resulting in a rise in circulating neutrophils in the blood within a few hours after infection [2 , 3] . The robust expression of G-CSF modulates the production of neutrophils to meet the increased need of the host during infection [4] . Circulating neutrophils migrate to the infection site along a chemotactic gradient of potent chemoattractants , such as KC ( CXCL1 or IL-8 in humans ) and MIP-2 ( CXCL2 ) , produced at the infection site [5 , 6] . To allow circulating neutrophils to cross the vascular wall and arrive at the site of infection , multiple adhesion molecules are induced on endothelial cells adjacent to the inflamed tissue . E- and P-selectins are known to be involved in the initial attachment of neutrophils to the endothelium as well as their rolling behavior . Intracellular adhesion molecule 1 ( ICAM-1 ) and vascular cell adhesion molecule 1 ( VCAM-1 ) mediate the subsequent step of tight adhesion to the endothelium , allowing neutrophils to transmigrate to the site of inflammation [7] . After migration , neutrophils phagocytose and digest the invading pathogen and produce pro-inflammatory cytokines [8] , thereby serving a beneficial role for the host . However , their excessive and uncontrolled activity may also cause severe damage to the host [9 , 10] . The important role of neutrophils in protecting the host against infection with respiratory pathogens has been investigated primarily with regard to pathogens such as Pseudomonas aeruginosa [11] , Legionella pneumophila [12] , Klebsiella pneumonia [13] and Yersinia pestis [14 , 15] . Y . pestis gained notoriety as the causative agent of plague [16] . Inhalation of Y . pestis droplets or aerosols leads to the development of primary pneumonic plague , which is a rapidly progressing fatal disease with the capability of spreading from person to person [17 , 18] . These characteristics also led to the recognition of Y . pestis as a potential biological threat agent [19] . Recent in vivo studies in animal models of pneumonic plague have revealed the biphasic nature of the progression of this disease [20–22] . The observed initial delay in the recruitment of immune cells and neutrophils in particular to the lungs of Y . pestis-infected mice is correlated with the limited up-regulation of multiple inflammatory cytokines and chemokines . Additionally , pulmonary infection with Y . pestis creates a permissive environment for the proliferation of other avirulent bacterial species [23] . Bacterial pathogens have developed a variety of mechanisms to inhibit immune cell functions as a means to disarm the host defense . For example , Y . pestis utilizes the pCD1-encoded type III secretion system ( TTSS ) composed of a secretory apparatus , chaperones and several translocated effectors ( Yops ) to disable the early innate immune response [24–26] and the activity of neutrophils in particular [27 , 28] . Recently , neutrophils were found to be an important cellular target of Y . pestis Yop secretion during the early stage of pneumonic plague [29] . While the lack of an adequate early immune response in the lungs during pneumonic plague is well described , the cascade of neutrophil recruitment from the circulation into the lungs during pneumonic plague and the identification of Y . pestis virulence factors involved in suppressing this process have yet to be fully elucidated . We previously reported that an early immune response is initiated by bone-marrow ( BM ) cells after airway infection of mice with a fully virulent Y . pestis strain , causing rapid mobilization of neutrophils from the BM to the blood circulation [30] . In the present study , we analyzed the interference of Y . pestis with the recruitment of neutrophils from the circulation to the lungs of infected mice . Our observations indicate that in the lungs of infected mice , the induction of the major neutrophil chemoattractants KC , MIP-2 and G-CSF as well as the leukocytes adhesion molecules E-selectin , P-selectin , ICAM-1 and VCAM-1 is delayed . In addition , we describe the role of YopJ in preventing the induction of the chemoattractants mRNA at the early stage of disease progression . Finally , we demonstrate that early attraction of neutrophils to the infected lung by intranasal installation of exogenous chemoattractants improves bacterial clearance as well as survival rate .
Studies in animal models of pneumonic plague have revealed the biphasic nature of the progression of this disease . The early phase of the disease , which takes place during the first 24–36 hours post infection ( hpi ) , involves a limited pro-inflammatory response in the lung , whereas the later phase of disease progression ( 48–72 hpi ) is associated with an excessive pro-inflammatory response [20–22] . To further characterize the early innate immune response during pneumonic plague , C57BL/6 mice were exposed i . n . to a lethal dose of 1x105 cfu ( 100 LD50 ) of the highly virulent Y . pestis strain Kim53 , which typically kills mice within 72 hpi ( S1 Fig ) . As neutrophils are one of the first innate immune cells recruited to the site of infection , we measured the levels of neutrophils infiltrating into the lungs at the early time point of 24 hpi . No significant change was observed in the absolute number or percentage of neutrophils at this time point in comparison to naïve mice ( Fig 1A and 1B ) . Because the metalloproteinases MMP8 and MMP9 are produced and released by neutrophils while combating invading pathogens [31] , we measured the levels of MMPs in lung extracts at the early time point of 24 hpi . Consistent with the limited infiltration of neutrophils into the lungs , the expression of MMP8 ( Fig 1C and 1D ) and MMP9 ( Fig 1E and 1F ) in the lungs of infected mice did not change in comparison to naïve mice . These findings indicate that the early pulmonary innate response , associated with neutrophils homing to the lung , is impaired after infection with the virulent Y . pestis strain . Moreover , the increased number of live Y . pestis bacilli detected in the lungs at 24–48 hpi ( Fig 1G ) suggests that the delayed influx of neutrophils into the lungs allowed the pathogen to rapidly proliferate and overwhelm the innate immune system . As the disease progressed into the excessive pro-inflammatory state at 48 hpi , massive infiltration of neutrophils into the lungs was observed ( Fig 1A and 1B ) together with a dramatic increase in the expression of both MMP8 and MMP9 ( Fig 1C–1F ) . Evidently , this late intensive pulmonary immune response was unable to prevent the pathogen from propagating to high levels in the lung tissue ( Fig 1G ) . This phenomenon was in contrast to the kinetics of neutrophil infiltration after infection with an equivalent infective dose of the avirulent Y . pestis strain Kim53Δ70Δ10 , which lacks the pCD1 and pPCP1 plasmids that carry essential virulence factors including the TTSS and Pla protease ( Fig 1A and 1B ) . The rapid elevation in neutrophil counts in the lungs of mice infected with the avirulent Y . pestis strain was accompanied by a significant up-regulation of the levels of MMP8 and MMP9 expression ( Fig 1C–1F ) and by a significant decrease in bacterial loads in the lungs at 24 hpi ( Fig 1G ) . Notably , this early recruitment of neutrophils to the lungs was transient , as depicted by the return of neutrophil numbers in the lungs to their basal level by 48 hpi ( Fig 1A and 1B ) . As a result , the levels of MMP8 and MMP9 expression were also reduced ( Fig 1C–1F ) . We previously demonstrated that i . n . infection of mice with the virulent Y . pestis strain is sensed by the BM compartment early after infection , resulting in the subsequent release of neutrophils to the blood by 12–24 hpi [30] . The observed delay in the homing of neutrophils from the circulation to the lungs motivated us to study the impairment of this pathway during the progression of pneumonic plague . Because the chemoattractants KC , MIP-2 and G-CSF are of central importance for the recruitment of neutrophils to infected organs [32] , their levels in the blood of mice infected with the virulent Y . pestis strain were measured by ELISA . As shown , the levels of KC , MIP-2 and G-CSF at the early stage of 24 hpi were comparable to those observed in naïve mice , and these levels increased significantly only during the late stage of infection at 48 hpi ( Fig 2A ) . This result suggests that although neutrophils are released from the BM to the blood early after airway infection , their ability to navigate towards the infected lungs is impaired . To better understand the limited ability of infected lungs to induce neutrophil chemotaxis and infiltration , we performed a Transwell-migration assay of naïve BM-derived neutrophils towards lung supernatants obtained from mice at several time points after infection with Y . pestis Kim53 . Only lung supernatants obtained from mice at late stages of disease progression e . g . , 48 hpi , demonstrated the potential to induce in vitro Transwell-migration of naïve neutrophils ( Fig 2B ) , implying that at this time point , the lungs are enriched with chemotactic factors that facilitate neutrophil migration . Next , we measured the levels of mRNA and protein expression of KC , MIP-2 and G-CSF in lung extracts and BALF at 12 , 24 and 48 hpi with Kim53 . The levels of mRNA ( red line ) and protein ( blue line ) of KC , MIP-2 and G-CSF were significantly increased in the lungs only at 48 hpi ( Fig 2C–2E ) . These observations are in line with previous reports describing the delayed pulmonary pro-inflammatory response to Y . pestis infection using various animal models of pneumonic plague [20–22 , 33] . In contrast , infection with the avirulent Y . pestis strain Kim53Δ70Δ10 was characterized by early and moderate induction of the mRNA levels of KC , MIP-2 and G-CSF in the lungs at 12 hpi , accompanied by elevated protein levels in the BALF ( Fig 2F and 2H ) . The recruitment of neutrophils to infected tissue is a complex process dependent on orchestrated and tightly regulated communication between neutrophils and endothelial cells , resulting in a multistep adhesion cascade . This process includes the initial attachment of the neutrophils to the endothelium , rolling along the endothelial surface and arrest at the final destination to allow complete transmigration [34] . This sequence of coordinated and transient interactions relies on the synchronized expression of several adhesion molecules by endothelial cells adjacent to the site of inflammation . Hence , we decided to examine the expression levels of four molecules that participate in two different stages of neutrophil transmigration: E-selectin , P-selectin , ICAM-1 and VCAM-1 [7] . Quantitative PCR analysis of lung mRNA obtained from mice infected with the virulent Kim53 strain revealed a delay in the up-regulation of the expression of all four adhesion molecules during the first 24 hpi ( Fig 3A ) . Again , this delayed expression was in contrast to the early induction of adhesion molecule mRNA in the lungs of mice infected with the avirulent strain Kim53Δ70Δ10 ( Fig 3B ) . Together , these data suggest that the rapid propagation of Y . pestis in the lungs during the early stages of pneumonic plague results from an impaired innate immune response associated with delayed neutrophil infiltration to the lung , presumably due to a combined delay in the expression of chemotactic signals and adhesion molecules by lung resident cells . The virulence characteristics of Y . pestis are mostly attributed to the pCD1 plasmid that encodes the TTSS and its six effectors proteins , termed Yops . During interactions with host target cells , these proteins are transported into the cytosol of the host cell via a needle-like apparatus . Together , the translocated Yops target the phagocytic machinery and deregulate signaling pathways , resulting in a reduced immune response by the host [35–37] . As described , infection of mice with the avirulent strain Kim53Δ70Δ10 that does not express the entire TTSS and the Pla protease , was characterized by the early induction of neutrophil chemoattractant mRNA and protein in the lungs ( Fig 2F and 2H ) , and by the early recruitment of neutrophils to the lungs ( Fig 1A and 1B ) . We suspected that one of the Yop effectors may be involved in the early suppression of the up-regulation of chemoattractants expression in the lungs after exposure to the virulent Y . pestis strain . To decipher which Yop was responsible for the early inhibition of KC , MIP-2 and G-CSF up-regulation , we used two different Yop-null derivatives of the fully virulent Y . pestis strain Kim53 , and we performed a series of in vitro infection experiments using alveolar-derived macrophages ( MH-S ) and lung epithelial ( TC-1 ) cell lines . We used the avirulent Y . pestis strain Kim53Δ70Δ10 and the wild-type Y . pestis strains as controls in these experiments . As shown , infection with 50 MOI of the Kim53ΔYopJ strain , but not with the Kim53 derivative lacking YopH , resulted in increased levels of KC and MIP-2 mRNA and protein in MH-S cells ( Fig 4A and 4B ) and of MIP-2 and G-CSF mRNA and protein in TC-1 cells ( Fig 4C and 4D ) . The induction of these chemokines production by the cell lines following infection with the YopJ mutant was similar to infection with the avirulent strain Kim53Δ70Δ10 . Additionally , infection of both cell lines with a YopJ mutant of the Y . pestis EV76 vaccine strain and with an EV76 derivative lacking the pCD1 plasmid led to induction of chemoattractants mRNA , whereas infection with EV76 and other Yop-null mutants including EV76ΔYopE , EV76ΔYopK and EV76ΔYopH did not ( S2 Fig ) . These results point to the involvement of YopJ in the regulation of KC , MIP-2 and G-CSF expression by lung-derived cells in vitro . Because YopJ activity was associated with suppression of KC , MIP-2 and G-CSF expression in Y . pestis-infected alveolar macrophages and epithelial cells , we infected C57BL/6 mice i . n . with a dose of 1x105 cfu of Kim53ΔYopJ and monitored the disease progression in the lung . All infected mice succumbed within 4 days of infection ( S1 Fig ) . Bacterial counts in the lungs were elevated to 1x109 cfu by 48 hpi ( Fig 5A ) , and no significant change was observed in neutrophil numbers or percentage in the lungs after the first 24 hpi ( Fig 5B and 5C ) . Massive infiltration of neutrophils to the lungs was apparent only at 48 hpi ( Fig 5B and 5C ) , and the delayed kinetics of neutrophil influx to the lungs following i . n . infection with Kim53ΔYopJ and bacterial propagation resembled the kinetic responses observed following infection with the wild-type Kim53 strain ( Fig 1 ) . Taking into account the observed involvement of YopJ in preventing KC , MIP-2 and G-CSF up-regulation by lung resident cells in vitro ( Fig 4 ) , we further evaluated the changes in mRNA and protein levels of these chemokines in lung extracts and BALF of Kim53ΔYopJ-infected mice . Surprisingly , we observed that while the mRNA levels of these chemokines were significantly elevated at 12–24 hpi , their protein levels remained constant and relatively low during this time frame . A significant increase in the protein levels of G-CSF , KC and MIP-2 was detected only at 48 hpi with Kim53ΔYopJ ( Fig 5D and 5F ) . Notably , significantly higher levels of G-CSF , KC and MIP-2 mRNA were measured at 48 hpi in the lungs of mice infected with the YopJ mutant compared to those of mice infected with the wild-type Y . pestis strain ( Fig 5G ) . This difference may result from earlier induction of chemoattractants mRNA following infection with the YopJ mutant , leading to the accumulation of higher levels of chemoattractants mRNA at later stages of disease progression . The data indicate that YopJ mediates the delayed up-regulation of chemoattractant mRNA expression in the lungs of Y . pestis-infected mice . However , unlike in the in vitro infection system we utilized , in the absence of YopJ , the early elevation of chemoattractant mRNA levels did not yield a subsequent increase in the protein levels . These observations suggests that additional virulence mechanisms involving host-pathogen interactions play a role in modulating the early expression of chemoattractants in a complex multicellular organ such as the lungs , thereby preventing the early recruitment of neutrophils to the infected lung . The early recruitment of neutrophils to the lungs appears to be of central importance for the defense against pneumonic plague . Due to the fact that some key players in this process ( e . g . KC , MIP-2 and G-CSF ) are targeted by the bacterium early after the infection , we tested the potential of chemokine therapy for early neutrophil recruitment to the lung . The treatment included subcutaneous administration of G-CSF to synchronize and overload the circulation with newly formed neutrophils , combined with i . n . instillation of KC and MIP-2 to guide the neutrophils and stimulate their recruitment and homing to the infected lungs ( Fig 6A ) . We first examined the potential of this treatment regimen to promote neutrophil recruitment to the lungs of naïve mice . As depicted in Fig 6B , treatment with G-CSF alone for 5 consecutive days significantly increased the numbers of neutrophils in the blood but not in the lung , whereas the combined treatment ( GKM ) , which included an additional intranasal administration of KC and MIP-2 ( 1 μg/mouse each ) , led to a significant accumulation of neutrophils in the lungs ( Fig 6B ) . This treatment was not associated with deleterious effects on animal morbidity or mortality ( S1 Fig ) . Next , we assessed the ability of GKM treatment to induce the early recruitment of neutrophils to the lungs of mice infected i . n . with the virulent Y . pestis strain . Similar to naïve mice , the percentage of neutrophils measured at 24 hpi in the blood of Y . pestis-infected mice treated for 5 consecutive days with G-CSF alone ( starting 3 days before the infection ) was elevated in comparison to the percentage in control-treated mice , whereas the percentage of lung neutrophils did not change ( Fig 6C , G-CSF ) . In contrast , a rapid increase in the percentage and total number of neutrophils was detected at 24 hpi in the lungs of GKM-treated mice that received KC and MIP-2 at 6 hpi ( Fig 6C and 6D , GKM ) . Injection of the specific anti-neutrophil antibody anti-Ly-6G into GKM-treated mice resulted in a reduction in the influx of neutrophils to the lungs by nearly 30% , verifying the specificity of the response with regard to the involvement of neutrophils ( Fig 6C and 6D , GKM+αLy-6G ) . We further assessed whether the early influx of neutrophils to the lungs of GKM-treated mice led to induction of the MMP8 and MMP9 metalloproteinases . Indeed , their expression was significantly higher in lung extracts obtained from GKM-treated mice as compared to control-treated mice ( Fig 6E ) . Again , injection of anti-Ly-6G to GKM-treated mice lowered the expression of MMPs ( Fig 6E ) , indicating that these early recruited neutrophils are in an active state once they reach the lung . To determine whether early recruited neutrophils to the lungs are able to clear Y . pestis , we measured the bacterial burden in the lungs of GKM-treated mice at 24 hr following infection with 1x105 cfu ( 100LD50 ) of the virulent Kim53 strain . A substantial reduction of almost a thousand fold was observed for the load of Y . pestis in the lung in comparison to untreated mice ( Fig 7A ) . This beneficial effect of GKM treatment was mediated by neutrophils , as the injection of anti-Ly-6G neutralizing antibody to GKM-treated mice decreased bacterial clearance , reflecting the importance of neutrophils for lung defense against Y . pestis ( Fig 7A ) . Furthermore , analysis of the relationships between the numbers of neutrophils and bacterial loads in the lungs of GKM-treated versus sham-treated mice at the early time point of 24 hpi , indicated that under this treatment levels of neutrophils greater than 1–2×106 were associated with effective bacterial clearance ( Fig 7B ) . Following the demonstration of early migration of neutrophils to the lung by the treatment with the recombinant proteins and the pronounced effect on Y . pestis propagation , it was interesting to evaluate this treatment in animals exposed to a lethal challenge . Relatively high protection level of 60% was observed in GKM-treated mice that were exposed i . n . to a dose of 2x103 cfu of the virulent Kim53 strain , whereas only 10% of the sham-treated mice survived this infection ( Fig 7C ) .
The ability of bacterial pathogens to prevent the early recruitment of neutrophils to infected organs provides an obvious advantage during infection because these cells , with their various antimicrobial capabilities , would otherwise kill the pathogen . Y . pestis , the causative agent of plague , exploits a variety of mechanisms for evading and coping with the host immune response during the early stages of infection . Accumulating evidence based on studies in various animal models of pneumonic plague indicates that following airway infection with Y . pestis , the early induction of a pro-inflammatory immune response in the lungs as well as the recruitment of neutrophils to the lungs are delayed [20–22] . We previously showed that the immune response is initiated by BM cells early after i . n . infection of mice with a fully virulent Y . pestis strain , causing rapid modulation of the BM CXCR4-SDF-1 axis and prompt mobilization of neutrophils into the circulation within 12–24 hpi [30] . These observations raised intriguing questions , namely , at what stage of neutrophil recruitment to the lungs does the pathogen interfere and which Y . pestis virulence factor is involved in this process . In this study , we further analyzed the mechanisms involved in the late homing of neutrophils to the lungs following i . n . infection of mice with the fully virulent Y . pestis strain Kim53 . Our results clearly indicate that the influx of neutrophils to the lungs in Y . pestis-infected mice is delayed . In addition , the delayed recruitment of neutrophils to the lungs is associated with a significant increase in bacterial burden . Cytokines and chemokines act in a coordinated manner to mobilize and recruit neutrophils to the site of inflammation . Because production of these factors represents the first step in the neutrophil recruitment process , we monitored the expression of several chemokines critical for the chemoattraction of neutrophils , in the lungs and plasma of infected mice during disease progression . Up-regulation of G-CSF , KC and MIP-2 in the plasma of infected mice was delayed until the late stages of disease progression ( i . e . , 48 hpi ) , consistent with the absence of a pro-inflammatory response in the lungs at the early stage of disease progression [20–22] . In addition , the levels of CXCR2 ( KC and MIP-2 receptor ) , did not change at the first 24 hpi on circulating neutrophils in the blood ( S3 Fig ) . Using a Transwell-migration assay , we found that lung extracts from the early stage of disease progression were incapable of inducing neutrophil migration , in contrast to lung extracts obtained from mice at later stages of the disease . This result suggests that the intrinsic induction of chemoattractant production in lung resident cells early after Y . pestis infection is inhibited . Moreover , the delayed induction of KC , MIP-2 and G-CSF mRNA and protein in the lungs of Kim53-infected mice corroborates this observation . The expression of adhesion molecules is up-regulated on endothelial cells located at the site of inflammation [38] . Circulating neutrophils that egress from the BM undergo E- and P-selectin-mediated rolling along the endothelial surface , followed by firm attachment via ICAM-1 and VCAM-1 [39] . In addition to the delay in chemokine up-regulation , the ability of Kim53-infected lung cells to support neutrophil transmigration into the lungs also appears to be impaired at the early stage of infection . This is due to the delayed induction of the adhesion molecules E- and P-selectin as well as ICAM-1 and VCAM-1 . Because chemokines are involved in the expression of adhesion molecules on capillary endothelia [1] , the delayed induction of adhesion molecules in Y . pestis-infected lungs might result from delayed expression of the KC and MIP-2 chemokines . Alternatively , direct interaction of Y . pestis with endothelial cells might affect the expression of these molecules , as demonstrated for ICAM-1 during in vitro infection of human umbilical vein endothelial cells ( HUVECs ) with the related enteropathogen Y . enterocolitica [40] . The virulence of pathogenic Yersinia strains is mostly attributed to the TTSS and its effector proteins which are used by the pathogen to subvert early innate immune responses [26 , 41] . In striking contrast to the impaired innate immune response in the lungs of mice infected i . n . with the virulent Y . pestis strain Kim53 , rapid and moderate induction of the expression of chemokines and adhesion molecules followed by an influx of neutrophils to the lungs was observed early after pulmonary infection of mice with the avirulent Y . pestis strain ( Kim53Δ70Δ10 ) that lacks the TTSS and Pla protease virulence factors . Consequently , this prompt response was associated with effective bacterial clearance from the lungs . TTSS Yop effectors of pathogenic Yersinia species are known for their ability to suppress the induction of innate immune responses in various types of mammalian cells through disruption of the target cell signaling network . For example , YopE has been reported to inhibit the production of IL-8 ( the human homologue of KC ) in Y . pseudotuberculosis-infected HeLa cells [42 , 43] , and YopJ-dependent suppression of TNFα secretion has been reported in Yersinia-infected macrophages [44–46] . Human bronchial epithelial cells co-transfected with cDNAs encoding Y . pseudotuberculosis YopJ also exhibited reduced transcription of IL8 , RANTES and ICAM-1 in a promoter activity assay [47] . Moreover , the YopH effector was shown to inhibit the expression of monocyte chemoattractant protein 1 ( MCP-1 ) in macrophages infected with Y . enterocolitica [48] and to suppress early pro-inflammatory cytokines in the lungs during pneumonic plague [49] . We investigated the involvement of YopJ and YopH in the modulation of KC , MIP-2 and G-CSF expression by alveolar macrophages ( MH-S ) and lung epithelial cell lines ( TC-1 ) following infection of the cells with Y . pestis Kim53 YopJ and YopH deficient mutants . The in vitro results indicated that inhibition of KC , MIP-2 and G-CSF mRNA and protein expression was mediated by the YopJ effector in a similar manner that was shown after infection with the avirulent Y . pestis strain—Kim53Δ70Δ10 , which lacks the entire TTSS . YopJ ( YopP in Y . enterocolitica ) belongs to a family of proteases related to the ubiquitin-like protein proteases [50] , and YopJ was shown to be a deubiquitinating cysteine protease capable of removing ubiquitin moieties from IκBα , thereby inhibiting its proteasomal degradation and leading to the down-regulation of NF-κB function [51] . In addition , YopJ was shown to acetylate Ser/Thr residues in the activation loop of MAPK kinases ( MKKs ) and IκB kinases ( IKKs ) , thereby preventing their activation by phosphorylation [52 , 53] . This acetyltransferase activity of YopJ may well account for its ability to inhibit MAPK pathways and NF-κB activation . Interestingly , expression of the KC and MIP-2 genes is known to be tightly regulated by the NF-κB transcription factor [54 , 55] , and NF-κB activation in lung epithelial cells was shown to be important for the migration of neutrophils to the lungs [56] . Therefore , YopJ-mediated inhibition of NF-κB activity may be involved in the suppression of KC and MIP-2 expression in lung-derived cell lines infected with Y . pestis . Studies on the involvement of YopJ in Y . pestis virulence have indicated that this effector is not essential for virulence in various rodent models of plague [57–59] . Similar results were obtained in our study using C57BL/6 mice infected i . n . with the Kim53ΔYopJ mutant . Close examination of the expression of neutrophil chemoattractants in the lungs of Kim53ΔYopJ-infected mice revealed that while the mRNA levels of KC , MIP-2 and G-CSF were induced in the lungs early after infection , the proteins levels were up-regulated only at the late stage of the disease , i . e . , 48 hpi . In line , the influx and homing of neutrophils to the lungs of Kim53ΔYopJ-infected mice was also delayed to late stage of the disease as observed following exposure to the wild-type Y . pestis strain . These intriguing data may indicate that Y . pestis affects the early recruitment of neutrophils to the lungs by regulating the expression of neutrophil chemoattractants at both the transcriptional level via the YopJ effector and at the posttranscriptional level by another , yet unknown virulence factor . One possible candidate is the YopH tyrosine phosphatase that was shown to suppress early pro-inflammatory cytokine up-regulation in the lungs of Y . pestis-infected mice; this factor was also shown to be essential for Y . pestis virulence in the mouse model of pneumonic plague [49] . Another candidate is the Pla protease that was shown to be important for Y . pestis proliferation in the lungs [60] and for degradation of Fas ligand to manipulate host cell death and inflammation [61] . The apparent discrepancy between the in vitro and in vivo infection systems suggests a different mode of chemokines expression in the absence of YopJ . While YopJ depletion resulted in augmented mRNA and protein expression levels of KC , MIP-2 and G-CSF by infected macrophages and epithelial cell lines in vitro , lungs obtained from infected mice exhibited changes only in the mRNA and not in the protein levels of these chemokines early after infection . We assume that these differences emphasize the major impact of the alveolar niche and its various resident cells on the complex virulent mechanisms generated by Y . pestis . Furthermore , the lack of host defense components ( such as other white blood cells , immunoglobulins , complement , cytokines , defensins and more ) in in vitro models , may have a great influence on the virulence quality of the pathogen . Because we previously showed that neutrophils egressed from the BM at 12–24 hr after i . n . infection with Y . pestis [30] , our current findings point to the delayed induction of chemokines as the major reason for the inability of circulating neutrophils to rapidly infiltrate the site of infection in the lungs . To test this hypothesis , we administered recombinant KC and MIP-2 by i . n . instillation into the lungs of Y . pestis-infected mice that were pre-treated systemically with G-CSF to synchronize and overload the circulation with primed neutrophils [62 , 63] . Whereas systemic pre-treatment of mice with G-CSF alone prior to pulmonary infection with Y . pestis did not lead to early recruitment of neutrophils to the lungs , additional i . n . administration of KC and MIP-2 several hours after the infection resulted in rapid mobilization of neutrophils to the lungs . Furthermore , the recruitment of neutrophils to the lungs of treated mice was accompanied by an increase in neutrophil-associated MMP8 and MMP9 expression and induction of the expression of the adhesion molecules E- and P-selectin in the lungs ( S4 Fig ) . The early influx of neutrophils to the lungs of Y . pestis-infected mice led to a rapid and significant reduction of nearly 1 , 000-fold in the average bacterial cfu in the lungs of mice infected with a dose of 100 LD50 of the virulent Y . pestis Kim53 strain . Moreover , this treatment was also successful in improving the survival rates of mice following i . n . exposure to a lethal dose of 2 LD50 of the fully virulent Y . pestis strain . Injection of treated mice with neutralizing anti-Ly-6G antibodies reduced the percentages of neutrophils by 30% and diminished the beneficial antibacterial outcome of treatment as well as the expression of MMPs , highlighting the contribution of recruited neutrophils to lung defense against Y . pestis . Collectively , our results indicate that Y . pestis-mediated interference with the early induction of G-CSF , KC and MIP-2 expression by lung resident cells and the recruitment of neutrophils to the lungs are important for the manifestation of Y . pestis virulence during pneumonic plague . These observations are consistent with previous reports demonstrating ( a ) the delayed expression of pro-inflammatory cytokines and chemokines during pneumonic plague [20–22] and ( b ) the importance of neutrophils for defense against Y . pestis pulmonary infection [15] . Notably , Goldman and his colleagues have recently reported that the number of neutrophils in the lungs of Y . pestis-infected mice increases rapidly within 24 hr post intranasal infection [29] . Differences in the experimental systems may account for this discrepancy , as this group infected the mice with an inoculation dose of 106 cfu of the CO92 strain pre-grown at 37°C , whereas in our experiments mice were infected with a lower dose of 105 cfu of the Kim53 strain pre-grown at 28°C . Directed guidance of immune cells to the lungs following i . n . administration of recombinant proteins has become an important tool for understanding the lung defense mechanisms against bacterial infections . Intranasal administration of exogenous KC ameliorated B . pertussis-mediated inhibition of neutrophil recruitment to the lungs in infected mice [64] . In addition , the recruitment of neutrophils to the lungs of S . pneumonia-infected mice was observed following i . n . administration of IL-12 and was associated with increased levels of KC , decreased bacterial burden and improved survival [65] . The early arrival of neutrophils to sites of infection may also influence the outcome of disease progression by regulating other types of immune cells . Bi Y et al . recently reported that production of IL-17A by neutrophils coordinates the antimicrobial activity of neutrophils and macrophages against Y . pestis infection during pneumonic plague [66] . In addition , the secretion of neutrophil-derived granule proteins and antibacterial peptides was associated with the migration of inflammatory monocytes to the site of infection [67 , 68] . The therapeutic potential of the early recruitment of neutrophils to the lungs may also be attributed to the involvement of these cells in disease resolution and anti-inflammatory processes . Neutrophils have recently been shown to play an important role in the repair of damaged tissue through the expression of MMP9 , which degrades intracellular matrix ( ICM ) components , thus promoting the removal of damage-associated molecular pattern ( DAMP ) -containing ICM proteins released from damaged cells ( Reviewed in [69] ) . Taken together , this study highlights ( a ) the complex virulence mechanisms employed by Y . pestis to minimize its early encounter with neutrophils in the lungs following airway infection and ( b ) the beneficial effect of modulating neutrophil chemotaxis into the lungs at the early stage of Y . pestis infection by treatment with chemoattractants . This therapeutic approach could be useful for improving treatments against plague as well as other pathogens that suppress the recruitment of neutrophils to sites of infection . The existence of antibiotic-resistant Y . pestis strains [70] further emphasizes the importance of modulating the host defense for the treatment of plague infections .
This study was carried out in strict accordance with the recommendations for the Care and Use of Laboratory Animals of the National Institute of Health . All animal experiments were performed in accordance with Israeli law and were approved by the Ethics Committee for animal experiments at the Israel Institute for Biological Research ( Permit Numbers: IACUC-IIBR M-07-2012 , IACUC-IIBR M-28-2013 ) . During the experiments , the mice were monitored daily . Humane endpoints were used in our survival studies . Mice exhibiting loss of the righting reflex were euthanized by cervical dislocation . Analgesics were not used , as they may have affected the experimental outcomes of the studies . The following Y . pestis strains were used in this study: the Y . pestis virulent strain Kimberley53 ( Kim53 ) [71] , the avirulent Kim53Δ70Δ10 strain that is spontaneously cured for pPCP1 and pCD1 [72] , Kim53 deleted for YopJ ( Kim53ΔYopJ ) [73] and the Kim53 deleted for YopH ( Kim53ΔYopH ) ( Constructed by Zauberman A , replacing the yopH gene with a kanamycin resistance cassette as described in [73] ) . The Y . pestis vaccine strain EV76 [71] , EV76 spontaneously cured for pCD1 ( EV76Δp70 ) [74] , EV76ΔYopJ [46] , and the EV76 deleted mutants EV76ΔYopK , EV76ΔYopE , EV76ΔYopH ( Constructed by Zauberman A . , replacing the yopK , yopE and yopH genes with a kanamycin resistance cassette as described in [73] ) . The strains were routinely grown on brain heart infusion agar ( BHIA , BD , MD USA ) for 48 hr at 28°C . The Y . pestis Yop-deleted strains were grown on BHIA supplemented with 100 μg/ml kanamycin ( Sigma-Aldrich , Israel ) . Deletion mutagenesis of the Y . pestis Kim53 and EV76 strains was performed by replacing the central region of the genes with a kanamycin resistance cassette ( Pharmacia ) by homologous recombination . The protocol used was based on previously established methodologies [75 , 76] . The linear PCR fragment in which kanamycin sequences were flanked by yop sequences was electroporated into Y . pestis bacteria expressing the λ phage red system from pKOBEG::sacB ( generous gift from Dr . E . Carniel ) . Electroporation was performed in 10% glycerol and 10% PEG-8000 ( Sigma-Aldrich , Israel ) , and the bacteria were incubated in HIB for 2 h at 28°C . Transformants were selected on BHIA containing 50 μg/ml kanamycin , and then the pKOBEG::sacB plasmid was removed from the bacteria by growing the bacteria on BHIA supplemented with 10% sucrose . The expected knockout phenotype was verified by PCR and Western blot analyses . Female C57BL/6 mice ( 6–10 weeks old ) were purchased from Harlan Laboratories ( Rehovot , Israel ) and maintained under defined flora conditions at the animal facilities of the Israel Institute for Biological Research . The i . n . infections were performed as described previously [77] . Briefly , bacterial colonies were harvested and diluted in heart infusion broth ( HIB ) ( BD , USA ) supplemented with 0 . 2% xylose and 2 . 5 mM CaCl2 ( Sigma-Aldrich , Israel ) to an OD660 of 0 . 01 and grown for 22 h at 28°C in a shaker ( 100 rpm ) . At the end of the incubation period , the cultures were washed , diluted in PBS solution to the required infectious dose and quantified by counting colony forming units ( cfu ) after plating and incubating on BHIA plates ( 48 hr at 28°C ) . Prior to infection , mice were anesthetized with a mixture of 0 . 5% ketamine HCl and 0 . 1% xylazine and then infected i . n . with 35 μl/mouse of the bacterial suspension , whereas naïve mice were instilled i . n . with PBS only . The intranasal LD50 of the Kim53 strain under these conditions is 1 , 100 cfu . LD50 values were calculated according to the method described by Reed and Muench [78] . Three days prior to infection , mice received daily subcutaneous injections of recombinant G-CSF ( rhG-CSF 300 μg/kg/Nupogen 48 MU/0 . 5 ml , Roche Applied Science ) for 5 consecutive days . Six hours after i . n . infection with Y . pestis Kim53 , mice were euthanized , and 1 μg of each recombinant KC and MIP-2 ( recombinant MCXCL1/KC , recombinant MCXCL2/MIP-2 , R&D Systems ) , diluted in 25 μl of PBS , or 25 μl of PBS alone ( sham ) were instilled i . n . Mice were either sacrificed and analyzed 24 hpi or followed to observe the rates of morbidity and mortality . For the depletion of neutrophils , 100 μg purified anti-Ly-6G antibody clone 1A8 ( Biolegend , USA ) diluted in 300 μl PBS was administered intraperitoneally twice at 24 hr prior to infection and 3 hpi . The murine alveolar macrophage cell line MH-S was obtained from ATCC . TC-1 is a tumor cell line derived from primary lung epithelial cells of C57BL/6 mice[79] . This cell line was a kind gift from the laboratory of Prof . T . C . Wu ( Johns Hopkins University ) . Both cell-lines were grown in RPMI 1640 medium supplemented with 10 mM HEPES , 2 mM L-glutamine , 1 mM sodium pyruvate , 0 . 1 mM non-essential amino-acids and 10% fetal bovine serum . Cell cultures were maintained at 37°C with 5% CO2 . Cell infection studies were performed as previously described [73] . Briefly , bacteria were grown by shaking ( 150 rpm ) for 22 h at 28°C in HIB . The resulting cultures were diluted in HIB medium to OD660 0 . 05 and allowed to grow for 3 h at 37°C ( 100 rpm ) . Bacteria were harvested , washed once and re-suspended in complete RPMI supplemented with 10% fetal calf serum and added to the cells at a multiplicity of infection ( MOI ) of 50 . Bacteria were adhered onto the cells by centrifugation at 130 g for 5 min followed by incubation for an additional 1 h at 37°C and 5% CO2 . Gentamicin was then added to the cultures to a final concentration of 50 mg/ml , and the cultures were incubated for an additional 4 h before using the cells for RNA extraction and RT-PCR analysis and the media for ELISA . To prepare lung cell suspensions , mice were euthanized , and blood was withdrawn from the heart using a heparinized syringe . Lungs were then removed and placed on a 70-μm nylon cell strainer ( BD Falcon , USA ) dipped in 2 ml PBS containing 1% protease inhibitor cocktail ( Sigma-Aldrich , Israel ) . Cell suspensions were pelleted at 260 g for 10 min at 4°C , fixed in 4% paraformaldehyde in PBS for 1 h at room temperature and washed twice in flow cytometry buffer . Neutrophils ( CD11b+/Gr-1high ) were stained with PerCP-Cy5 . 5-conjugated anti-mouse CD11b antibody ( clone M1/70 ) ( eBioscience , USA ) and APC-conjugated anti-mouse Ly-6G ( Gr-1 ) antibody ( clone RB6-8C5 ) ( eBioscience , USA ) . CXCR2 staining was performed with PE-conjugated anti-CXCR2 antibodies ( clone 242216 ) ( R&D , USA ) . Cells were stained using standard protocols with appropriate matched isotype control antibodies . The analysis was performed using a FACSCalibur flow cytometer with CellQuest Pro software ( BD Bioscience , USA ) . Lung cell suspensions were prepared as previously described . Total RNA was extracted using Tri-reagent ( Sigma-Aldrich , Israel ) according to the manufacturer’s instructions . Two micrograms of total RNA were reverse-transcribed using Moloney murine leukemia virus reverse transcriptase and oligo-dT primers ( Promega , USA ) . Quantitative PCR analysis was performed using an ABI 7500 instrument ( Applied Biosystems , USA ) with SYBR green PCR master mix ( Applied Biosystems , USA ) . The fold change in the quantity of gene transcripts was measured and compared to the hypoxanthine phosphoribosyl transferase ( HPRT ) gene using the comparative ( -2ΔΔCt ) method . Forty cycles of PCR amplification were performed in duplicate for each primer set . Primer sequences used are listed in Table 1 . Blood was collected and centrifuged at 260 g for 10 min , and the plasma was collected , filtered and stored at -70°C . Bronchoalveolar lavage fluid ( BALF ) was collected by exposing the trachea and injecting , and then removing twice , a total of 1 ml PBS containing 1% protease inhibitor cocktail ( Sigma-Aldrich , Israel ) . BALF was then filtered and stored at -70°C . Before analysis , samples were centrifuged again at 13 , 000 g for 5 min . The levels of KC , MIP-2 and G-CSF in the plasma and BALF and the levels of MMP9 in the BALF , were measured by enzyme-linked immunosorbent assay ( ELISA ) according to the manufacturer’s protocol ( R&D Systems , MN , USA ) . Gelatin zymography for MMP9 activity was performed as previously described [80] . Briefly , whole lung supernatants were mixed with a non-reducing sample buffer , and an equal amount was loaded onto 10% SDS-polyacrylamide gels co-polymerized with 1 mg/ml gelatin derived from porcine skin ( Sigma-aldrich , Israel ) . After electrophoresis , the gels were washed for 30 min in Triton X-100 , followed by 3 washes with H2O and incubation at 37°C for 16 hr in developing buffer . The gels were then stained with SeeBand Forte ( Gene Bio-Application Ltd ) until clear bands appeared , indicating the presence of MMP9 activity . Conditioned media of HT-1080 cells secreting MMP9 served as a control . Transwell-migration assays were performed as previously described [81] . Briefly , total BM cells were extracted from the femur and tibias of naïve mice and suspended in complete RPMI media . Prior to the migration assay , the total BM cells were labeled with the neutrophil markers Gr-1 and CD11b to evaluate the levels of neutrophils . The cells were counted , and 250 , 000 cells/100 μl were allowed to migrate towards a total of 600 μl media containing 150 μl of lung supernatant through 24-well filters with a pore size of 5 μm ( Corning , NY , USA ) at 37°C for 3 hr . The cells were then collected and counted using flow cytometry . In parallel , a portion of the migrated cells was labeled with Gr-1 and CD11b to evaluate the number of migrated cells . The percentage of migrated cells was then calculated by comparing the number of neutrophils before and after migration . Bronchoalveolar lavage fluid ( BALF ) was collected as previously described , and equal amount of BALF samples were subjected to 10% SDS-PAGE followed by immunoblot with anti MMP8 polyclonal antibody ( Proteintech cat:17874-1-AP ) . Statistical significance was determined using the nonparametric Mann-Whitney test . A Kaplan-Meier survival estimate of treated and non-treated mice ( of at least 10 mice per group ) was also performed . Calculations were made using GraphPad Prism software .
|
The pathogen Yersinia pestis is the causative agent of pneumonic plague , as well as a potential bioweapon . The nature of this disease involves an initial non-inflammatory phase where the influx of neutrophils to the lungs is suppressed , allowing bacterial propagation in this organ . Using the mouse model of pneumonic plague , we demonstrate that the early expression of neutrophil chemoattractants and adhesion molecules in the lungs is delayed concomitant with a delayed recruitment of neutrophils to the lung . We also show that the Y . pestis virulence factor YopJ is involved in the early suppression of chemoattractants mRNA expression in the lung early after infection , but it seems that additional Y . pestis factors interfere with the protein synthesis of these chemoattractants . Indeed , administration of recombinant KC and MIP-2 to the infected lung of G-CSF treated mice restored the early neutrophil influx to the lungs , leading to a significant reduction in bacterial burden . The treatment has also proved efficacious in reducing mortality . This study highlights the complex virulence mechanisms employed by Y . pestis to diminish the early homing of neutrophils to the lungs thereby allowing bacterial propagation and disease progression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Circumventing Y. pestis Virulence by Early Recruitment of Neutrophils to the Lungs during Pneumonic Plague
|
Cryptosporidiosis is a common cause of infectious diarrhea in young children worldwide , and is a significant contributor to under-five mortality . Current treatment options are limited in young children . In this study , we describe the natural history of Cryptosporidium spp . infection in a birth cohort of children in Bangladesh and evaluate for association with malnutrition . This is a longitudinal birth cohort study of 392 slum-dwelling Bangladeshi children followed over the first two years of life from 2008 to 2014 . Children were monitored for diarrheal disease , and stool was tested for intestinal protozoa . Anthropometric measurements were taken at 3-month intervals . A subset of Cryptosporidium positive stools were genotyped for species and revealed that C . hominis was isolated from over 90% of samples . In the first two years of life , 77% of children experienced at least one infection with Cryptosporidium spp . Non-diarrheal infection ( 67% ) was more common than diarrheal infection ( 6 . 3% ) although 27% of children had both types of infection . Extreme poverty was associated with higher rates of infection ( chi-square , 49 . 7% vs 33 . 3% , p = 0 . 006 ) . Malnutrition was common in this cohort , 56% of children had stunted growth by age two . Children with Cryptosporidium spp . infection had a greater than 2-fold increased risk of severe stunting at age two compared to uninfected children ( odds ratio 2 . 69 , 95% CI 1 . 17 , 6 . 15 , p = 0 . 019 ) independent of sex , income , maternal body-mass index , maternal education and weight for age adjusted z ( WAZ ) score at birth . Cryptosporidium infection is common ( 77% ) in this cohort of slum-dwelling Bangladeshi children , and both non-diarrheal and diarrheal infections are significantly associated with a child’s growth at 2 years of age .
Diarrheal disease is the second leading cause of death in children under age five worldwide , with cryptosporidiosis estimated to be second only to rotavirus as the leading cause of moderate-to-severe diarrhea [1–3] . Cryptosporidium spp . are enteric protozoa , with 26 described species , but C . hominis and C . parvum most commonly infect humans [4 , 5] . Infection is characterized by profuse , watery diarrhea . Disease is self-limited in immune-competent adults , but can be associated with fulminant disease in immunocompromised patients and young children . Cryptosporidium infection has been associated with longer duration of diarrhea and 2–3 times higher mortality in young children [6 , 7] . Risk factors for cryptosporidiosis relate to the host , environment , and the species of the parasite . Immune-compromised adults , including those with HIV/AIDS or on immunosuppressive drugs , are at increased risk [8–10] . Young children , especially those with malnutrition , are more vulnerable , presumably due to lack of acquired immunity , however the biological mechanism is not clear [11–16] . In a child cohort in Bangladesh , almost 40% of children were infected with Cryptosporidium in the first year of life [15] . Breastfeeding and breast milk IgA have been identified as protective factors [15–17] . Host genetic susceptibility is also implicated as Cryptosporidium infection has been associated with HLA class II alleles and polymorphisms in the mannose binding lectin gene [18–19] . Environmental risk factors can be species specific , and include prolonged contact with domestic animals [20–22] , overcrowded living conditions [23–26] and household contact with young children [12 , 22 , 27] . Sporadic epidemics have been reported in relation to contaminated water sources [28] . Livestock serve as an environmental reservoir for C . parvum , and transmission has been reported after direct contact with animals or drinking water contaminated by human or animal waste [20] . In contrast , humans are the only major reservoir for C . hominis , and transmission is related to person-to-person contact , thus urban settings and overcrowding have been associated with C . hominis [29] . In contrast to other enteric diseases , household income has not been reported to be a protective factor , and in one study was even associated with increased risk of infection [29 , 30] . Malnutrition has been identified as a risk factor for infection , and may potentiate adverse impact from infection [14] . Furthermore , studies from Brazil and Peru have noted short-term growth faltering after infection [31 , 32] . The relationship between cryptosporidiosis and malnutrition is complex and poorly understood . In this study we describe the natural history of cryptosporidiosis in a peri-urban slum community near Dhaka , Bangladesh , over the first two years of life , a critical period for childhood growth and development [33] . Additionally , we identify risk factors in an endemic region , describe genetic diversity of the parasite , and test for contribution of infection on growth faltering in this population .
All stool samples were tested for Cryptosporidium species by real-time polymerase chain reaction . DNA Extraction was performed by a modified QiaAmp stool DNA extraction protocol which incorporates a three-minute bead-beating step to lyse Cryptosporidium oocysts ( Qiagen , Valencia , CA ) [34] . Cryptosporidium positive samples were detected using an assay previously described by our group that targets the Cryptosporidium Oocyst Wall Protein ( COWP ) [35] . COWP positive samples were further genotyped by the polymorphic region within the gp60 gene using primers and conditions previously described [36 , 37] but with the modification that the amplifications were done using the MyFi ( Bioline , Taunton , MA ) with an activation 94°C for 5 min followed by 40 cycles in the primary PCR ( 94°C , 30 sec; 45°C , 55 sec; 72°C , 60 sec ) with a final extension of 10min at 72°C . In the secondary PCR the cycling conditions [35] were ( 94°C , 30 sec; 55°C , 30 sec; 72°C , 30 sec ) with a final extension of 10 min at 72°C . Phusion high fidelity polymerase ( ThermoFisher Scientific Inc , Waltham , MA ) was used in a final PCR to add sequencing primer binding sites [38] . As necessary for the Phusion enzyme the activation step was at 98°C , 20 sec was followed by 34 cycles ( 98°C , 10 sec; 60°C , 20 sec; 72°C , 20 sec ) with a final extension of 10min at 72°C . The QIAquick PCR purification kit was then used as per manufacturer’s instructions ( Qiagen , Valencia , CA ) to purify the amplicon . This was then sequenced by a contract research organization ( GENEWIZ , South Plainfield , NJ ) using standard protocols and the sequencing primers IS5 ( AATGATACGGCGACCACCGA ) or IS6 ( CAAGCAGAAGACGGCATACGA ) . The resulting sequences were then trimmed and aligned to the gp60 reference sequence ( Genbank Accession Numbers HQ631408 , AY738187 , AY738192 , AY738184 and AF440638 ) using the Geneious Program ( R7 ) ( Biomatters , NZ ) . Consensus phylogeny was inferred from 500 bootstrap replicates to build a neighbor-joining consenus tree , and based on the Tamura-Nei model , the Nearest Neighbor method and the Geneious Program ( R7 ) . Branches that produced in fewer than 50% of the bootstrap phylogenies were collapsed . Differences in demographic factors between infected and uninfected children were assessed using two-sample t-test and chi-square according to exposure status ( no Cryptosporidium infection , any type of Cryptosporidium infection , diarrheal Cryptosporidium infection , and exclusively non-diarrheal Cryptosporidium infection ) . Family monthly income ( expressed in Bangladeshi Taka or BDT ) below 6000 BDT/month was defined as “extreme poverty” based on the World Bank’s definition of less than 1 . 25 international dollars per person per day [39] . Anthropometric measures ( height-for-age adjusted z-score or HAZ; weight-for-age adjusted z-score or WAZ; weight-for-height adjusted z-score or WHZ ) were evaluated both as continuous and categorical variables . HAZ score at 24 months was used as the outcome in the final analyses evaluating nutrition , as HAZ is most representative of chronic malnutrition [40] . For all analyses evaluating malnutrition , we excluded children who fell into the bottom 2 . 3% of HAZ scores at birth ( HAZ < -3 . 49 ) , per WHO Global Database on Child Growth and Malnutrition guidelines [40] . Based on our cohort’s distribution of anthropometric indices , we classified children into four categories: 1 ) HAZ > -1; HAZ < = -1 and > -2 ( mild stunting ) ; HAZ < = -2 and >-3 ( moderate stunting ) ; and HAZ < = -3 ( severe stunting ) . To determine the time to first diarrheal and time to first asymptomatic Cryptosporidium infection , we performed Kaplan Meier survival analysis separately . All children who completed 24 months in the study were included , and those children without a Cryptosporidium infection within the first 24 months were censored at this time point . The probability of growth impairment at 24 months of age was evaluated using univariate and multivariable logistic regression with the categorized HAZ at 24 months as a polynomial response . Potential confounding variables including sex , family income , maternal body-mass index , maternal education , and WAZ at birth were adjusted in the multivariable analysis . Statistical significance was considered if p <0 . 05 . Analyses were performed in Stata v . 10 ( Statacorp , USA ) . Comparison of Cryptosporidium quantitation in stool ( threshold cycle ) to severity of clinical infection was performed using the Kruskal Wallis test . The study was approved by the Institutional Review Board of the University of Virginia and the Research and Ethical Review Committees of the International Centre for Diarrhoeal Disease Research , Bangladesh ( icddr , b ) . Informed written consent was obtained from parents or guardians for the participation of their child in the study .
Table 1 summarizes demographic characteristics of children who became infected with Cryptosporidium spp . during the first 24 months of life . A higher proportion of Cryptosporidium-infected children came from extreme poverty ( monthly income <6000 BDT/month ) ( chi-square p = 0 . 006 ) . We found no association between increased risk of Cryptosporidium infection and HAZ at birth ( p = 0 . 89 ) , presence of an animal in the house ( chi-square , p = 0 . 57 ) , or family size ( chi-square , p = 0 . 44 ) . Seventeen percent of children ( 68/392 ) met W . H . O . guidelines for moderate stunting ( height-for-age adjusted z-score less than -2 ) at birth . By age two , 29 . 6% , 35 . 7% , and 21 . 1% of children met W . H . O . criteria for mild , moderate , and severe stunting , respectively . Over the first two years of life , the mean height-for-age adjusted z-score in this cohort fell consistently below the W . H . O . reference population . Fig 3 shows the steady decline in HAZ from birth to 24 months of age . Fifty-percent of children had at least one non-diarrheal Cryptosporidium infection by 16 months of age , and 25% of children had a symptomatic infection by 2 years of age . The hazard ratio of an asymptomatic infection did not differ by sex ( HR = 1 . 12 , 95% CI 0 . 89 . 1 . 42 , p-value = 0 . 315 ) but was decreased for individuals with higher family income ( HR = 0 . 74 , 95% CI 0 . 58 , 0 . 93 , p-value = 0 . 011 ) . The hazard ratio remained consistent for the symptomatic diarrheal infections . There was no significant difference in time to first asymptomatic or diarrheal infection ( Fig 4 ) . However , children who had both diarrheal and asymptomatic infections during the 24-month follow up period were infected at an earlier age ( HR = 1 . 74 , 95% CI 1 . 34 , 2 . 27 , p-value < 0 . 0001 ) . A subset of diarrheal and surveillance stool samples ( n = 238 ) testing positive for Cryptosporidium spp . were further typed by gp-60 . C . hominis was the sole Cryptosporidium species in 92 . 4% of samples ( n = 220 ) and C . parvum was the sole species identified in 3 . 4% ( n = 8 ) . In only five cases did we observe a mixed infection of C . hominis and C . parvum . C . hominis positive samples were further subtyped and the distribution of gp 60 alleles found in our study population was 1a ( 17 . 3% ) ; 1b ( 21 . 4% ) ; 1d ( 13 . 9% ) ; 1e ( 40 . 5% ) ; 1f ( 7 . 0% ) . Of these sequences , one hundred and one samples were of sufficient quality to subtype by multiple sequence alignment with GenBank reference sequences and phylogenetic analysis ( Fig 5 ) . In contrast to other molecular epidemiologic studies there was no gp60 subtype diversity within our population ( Ia , A14R1; Ib , A9G3R2; Id , A15G1; Ie , A11G3T3; If , A13G1 ) ( 32 ) . Using logistic regression , we observed that children with linear growth faltering at 24 months had a greater than 2-fold increased odds of experiencing any type of Cryptosporidium spp . infection during the first two years of life compared to non-stunted children ( Table 2 ) . The associated risk of Cryptosporidium spp . infection increased with the severity of stunting , and children with severe stunting at 24 months had a 2 . 69 times increased odds of Cryptosporidium infection ( Table 2 ) . The association of stunting and linear growth faltering at 24 months was present for both non-diarrheal “asymptomatic” and symptomatic Cryptosporidium spp . infections ( Table 2 ) . Stunting was associated with increased odds of infection even after adjusting for income , gender , maternal BMI , maternal education , days of exclusive breastfeeding , and nutritional status at birth . When we considered HAZ at 24 months as a linear variable , the association with Cryptosporidium was further supported in an adjusted model ( linear regression , b = -0 . 33 , 95% CI -0 . 57 , -0 . 08 , p = 0 . 0009 ) . Furthermore , there was additive impact of each additional infection on HAZ at 24 months ( linear regression , b = -0 . 18 , 95% CI -0 . 24 , -0 . 026 , p = 0 . 02 ) . Among Cryptosporidium spp . infected children , the mean 24-month HAZ was significantly lower in children from households with lower monthly income ( t-test , p = 0 . 0053 ) . Among those from higher income households , Cryptosporidium spp . infected children had lower 24-month HAZ scores compared to uninfected children ( t-test , p = 0 . 016 ) . Additionally , there was no association between number of diarrheal episodes over the first 24 months of life and 24-month HAZ score ( S1 Table ) .
In an urban slum in Bangladesh , we followed 392 children from birth to age two years of life . In this cohort , extreme poverty and malnutrition were common , affecting almost half of all households and half of enrolled children . Fifty-six percent of children met W . H . O criteria for moderate stunting at age two , which is higher than the reported rate of under-five stunting across South Asia ( 40 . 7% ) [40] . Cryptosporidium spp . infection affected 77% of children in this cohort . Interestingly , we identified a larger number of asymptomatic infections than diarrheal infections . Potentially , this is due to consistent monthly surveillance sample collection and use of qPCR for diagnosis , rather than microscopy or antigen detection . In a birth cohort in India , Sarkar et al also reported a higher rate of asymptomatic than diarrheal Cryptosporidium infection [12] . In the current study , poverty was associated with Cryptosporidium infection , with children in households with more income less likely to have cryptosporidiosis . Based on gp-60 genotyping , we identified a predominance of C . hominis isolates in this cohort , which is consistent with other reports from the South Asian subcontinent [41–44] . Previously described risk factors for cryptosporidiosis in children include close contact with domesticated animals [20–22] , crowded living conditions [12 , 25 , 26] , and malnutrition [12 , 14] . We did not find the same association between domesticated animals and infection , likely because C . hominis predominated in our study population , and spread is anthroponotic . Additionally , we hypothesize that lower household income is related to overcrowding , and overcrowding is associated specifically with C . hominis infection [22] . Our study was not able to measure level of overcrowding in households , but median household size was 5 , with a range of 2 to 18 persons per home , and this figure was not significantly different between groups . We did evaluate malnutrition and subsequent Cryptosporidium infection , however , there was no significant association . This was likely due to our study design that enrolled children from birth . We excluded children with extreme stunting at birth , as we were interested in controlling for peri-natal factors that may have led to stunting at birth and not potential maternal or prenatal factors . In non-birth cohort designed studies it would be difficult to differentiate between children stunted at birth and those who developed stunting perinatally . One of the most significant findings of this study was the predisposition towards linear growth faltering that occurred in Cryptosporidium spp infected children . While malnutrition at birth did not predispose to Cryptosporidium spp infection , children who had at least one Cryptosporidium spp infection in the first two years of life had significantly worse nutritional status at 24 months , independent of income and maternal factors , suggesting that Cryptosporidium spp infection is associated with downstream growth faltering . Notably , both diarrheal and non-diarrheal infections were associated with subsequent stunting . This is supported by prior studies from Peru that have shown that children with asymptomatic and symptomatic Cryptosporidium spp infection had less weight gain in the first month of infection [32] , but in contrast to the Peruvian studies , we found that children infected by Cryptosporidium even after 6 months of age , do not have “catch up growth” and once infected , are on a trajectory to growth stunting [45] . Our findings suggest that even a single Cryptosporidium spp infection at any point in the first two years of life , whether diarrheal or non-diarrheal , can be detrimental to a child’s physical development , resulting in impaired growth at age two . Therefore , we propose that malnutrition , rather than diarrhea , should be considered the most important outcome of Cryptosporidium spp infection in children . We have shown that non-diarrheal Cryptosporidium spp infection is widely prevalent in this cohort . The mechanism between non-diarrheal infection and malnutrition requires further study . Cryptosporidium infection has been associated with increased inflammation of the gut and loss of villus architecture [46] and murine models suggest that immune signaling in the gut may be disrupted resulting in enteropathy and poor growth [47] . Our study is limited in that we did not assess for multiple enteric pathogens . However , the aim of our study was to describe the total burden of Cryptosporidium infection in this cohort , rather than ascribe etiologies of diarrhea . Previous studies in this area have found that children may carry four or more pathogens in any given stool specimen [48] . It is possible that the presence of other enteric infections , or even other disease processes that were not evaluated in this study ( e . g . acute respiratory infections ) contribute to stunting . However , in this study we did not find a relationship between total burden of diarrhea and HAZ at 24 months . Therefore , we would argue that it is the presence of enteric infection , rather than the phenotype of diarrhea , that is contributing to stunting . Based on our findings , future studies of cryptosporidiosis should aim to further study genotypic differences . We have demonstrated that in our cohort with C . hominis predominance , risk factors for infection are significantly different than in other populations with potentially different species . This may also impact means of transmission of infection . And beyond the species level , there may be additional clinical and immunologic differences between different subtypes of C . hominis that have not yet been described . Further studies of Cryptosporidium spp . genotyping will be important for informing strategies for prevention and treatment . We have demonstrated that poverty , malnutrition , and Cryptosporidium spp . infection remain intricately connected . Worldwide , an estimated 178 million children under 5 suffer from stunted growth [40] and stunted growth in the first two years of life leads to irreversible damage , contributing to poor cognitive development , poor educational performance , and reduced earning potential in adulthood , trapping individuals in a lifetime of poverty [49 , 50] . Therefore , in populations like the Mirpur cohort , where cryptosporidiosis is found in 80% of children less than two years of age , tackling strategies for interrupting spread of infection , vaccination , and treatment , may have a staggering impact on human potential . Elimination of cryptosporidiosis may be one important step towards improving the condition of impoverished children around the world .
|
Diarrheal disease is a leading cause of death in young children worldwide . Cryptosporidium species are responsible for a large proportion of global burden of diarrhea . This study describes the natural history of cryptosporidiosis in a birth cohort of impoverished Bangladeshi children . Children were enrolled at birth and monitored for diarrhea twice a week for two years . Stool samples were tested for enteric protozoa . Children in this cohort had significant rates of malnutrition compared to the W . H . O . reference population , and extreme poverty was common . A majority of children were infected with Cryptosporidium spp , and we found that children who had at least one infection with Cryptosporidium spp during the two year follow up period were significantly more likely to have growth faltering by age 24 months . Cryptosporidiosis is a common infection in this cohort , and is associated with poverty and reduced growth during the first two years of life .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"children",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"pathogens",
"microbiology",
"cryptosporidium",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"pulmonology",
"pediatrics",
"diarrhea",
"age",
"groups",
"protozoans",
"signs",
"and",
"symptoms",
"nutrition",
"gastroenterology",
"and",
"hepatology",
"pediatric",
"infections",
"cryptosporidiosis",
"malnutrition",
"bacterial",
"pathogens",
"families",
"cardiobacterium",
"hominis",
"medical",
"microbiology",
"microbial",
"pathogens",
"people",
"and",
"places",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"organisms"
] |
2016
|
Natural History of Cryptosporidiosis in a Longitudinal Study of Slum-Dwelling Bangladeshi Children: Association with Severe Malnutrition
|
Mutants lacking the ψ ( HolD ) subunit of the Escherichia coli DNA Polymerase III holoenzyme ( Pol III HE ) have poor viability , but a residual growth allows the isolation of spontaneous suppressor mutations that restore ΔholD mutant viability . Here we describe the isolation and characterization of two suppressor mutations in the trkA and trkE genes , involved in the main E . coli potassium import system . Viability of ΔholD trk mutants is abolished on media with low or high K+ concentrations , where alternative K+ import systems are activated , and is restored on low K+ concentrations by the inactivation of the alternative Kdp system . These findings show that the ΔholD mutant is rescued by a decrease in K+ import . The effect of trk inactivation is additive with the previously identified ΔholD suppressor mutation lexAind that blocks the SOS response indicating an SOS-independent mechanism of suppression . Accordingly , although lagging-strand synthesis is still perturbed in holD trkA mutants , the trkA mutation allows HolD-less Pol III HE to resist increased levels of the SOS-induced bypass polymerase DinB . trk inactivation is also partially additive with an ssb gene duplication , proposed to stabilize HolD-less Pol III HE by a modification of the single-stranded DNA binding protein ( SSB ) binding mode . We propose that lowering the intracellular K+ concentration stabilizes HolD-less Pol III HE on DNA by increasing electrostatic interactions between Pol III HE subunits , or between Pol III and DNA , directly or through a modification of the SSB binding mode; these three modes of action are not exclusive and could be additive . To our knowledge , the holD mutant provides the first example of an essential protein-DNA interaction that strongly depends on K+ import in vivo .
Chromosome replication is performed in Escherichia coli by a replicase called the DNA Polymerase III holoenzyme ( Pol III HE ) and composed of 9 different polypeptides [1] . DNA synthesis is realized by the core polymerase , composed of a polymerase subunit ( α , encoded by dnaE ) , associated with a proof-reading activity ( ε , encoded by dnaQ ) and a stabilizing subunit ( HolE ) . Each Pol III HE contains two core polymerases , one for the continuously synthesized leading-strand , one for the lagging-strand synthesized as 1–2 kilobase ( kb ) Okazaki fragments . The presence of an additional spare one to form a trimeric polymerase was proposed and is still debated [2–4] . The lagging-strand template is transiently single-stranded during Okazaki fragment synthesis and covered by single-stranded DNA binding proteins ( SSB ) . The stability of each core polymerase on DNA is ensured by its interaction with a polymerase clamp , in E . coli this is a β-dimer ( encoded by dnaN ) which encircles the DNA and is structurally homologous to PCNA in eukaryotes [5] . The β-clamp is loaded onto DNA by a clamp loader complex , functionally homologous to the RFC complex in eukaryotes , which , in addition to loading the β-clamp , ensures the cohesion of the replisome by interacting with the three core polymerases and with the DNA helicase . In E . coli the clamp loader complex is composed of three τ subunits ( encoded by dnaX ) , the δ and δ’ subunits ( encoded by holA and holB , respectively ) , and a heterodimeric complex of χ and ψ proteins , encoded by holC and holD respectively [1 , 2] . The ψχ complex forms a bridge between the τ3δδ’pentamer and SSB . Actually , ψ interacts with τ and χ , which itself interacts with SSB [6–9] . In addition , the ψ-τ interaction favors the assembly of the τ3δδ’pentamer at limiting δδ’concentrations , stabilizes an ATP-activated DNA-high affinity conformation of the clamp loader , and thus facilitates the clamp loading reaction in vitro [10–13] . Most Pol III HE subunits are essential for growth , with the notable exception of HolE and DnaQ [14] . Inactivating holC is not lethal but impairs growth , particularly at a high temperature; growth of holC mutants can be significantly improved if the induction of the SOS response , a set of repair genes induced by DNA damage or replication impairment , is prevented [15 , 16] . Inactivating holD is strongly deleterious in all growth conditions; however , a residual growth on minimal medium at 30°C facilitates the selection of suppressor mutations . We previously reported that ΔholD mutant growth is improved by mutations that inactivate the SOS response and more specifically by the inactivation of dinB and polB genes , encoding the SOS-induced bypass polymerases Pol IV and Pol II , respectively [17] . We proposed that increased ( SOS-induced ) levels of these polymerases compete with HolD-less Pol III HE and destabilize its interaction with DNA . The viability of holD and holC mutants is also restored by a duplication of the ssb gene , which doubles the intra-cellular level of SSB proteins [16] . We proposed that increasing intracellular concentration of SSB favors the SSB-DNA binding mode where each SSB tetramer binds 35 nucleotides , which stabilizes Pol III bound to DNA and bypasses the need for HolD . In the present study , we describe the isolation of holD suppressor mutations that affect K+ import . K+ is the major intracellular cation in E . coli , present at concentrations that vary from 100–150 to 500–600 mM , which is higher than that in the extracellular medium [18–20] . In the growth medium used here , about 45–50% of K+ ions are expected to be “free” and balance charges of small diffusible anions , including glutamate , and about 50–55% are thought to be “bound” and serve to balance charge on macromolecule anions , proteins and nucleic acids [21] . K+ glutamate is the intracellular ionic compound that ensures turgor and osmolarity . K+ is imported in E . coli by several independent systems [22 , 23] , [24] ( Table 1 ) . The major one is composed of TrkA , TrkH or TrkG , and TrkE [22 , 25] . These four genes are unlinked and represent two separate pathways , TrkAEG and TrkAEH . A second system depends on one protein , TrkD later called Kup [22 , 26] . Trk proteins and Kup have a low affinity for K+ ( Km 1 . 5 mM ) , and Trk is the main active pathway when K+ concentration in the growth medium is 5 mM or more; Kup activity could only be studied in mutants lacking the other K+ import systems [27 , 28] . At low K+ concentrations the kdpFABC operon is induced by a two component system , KdpD and KdpE; Kdp is a high affinity K+ import system ( Km 2 μM ) , active below 5 mM K+ [29 , 30] , [31] . Finally , at high K+ concentrations the triple trk kup kdp mutant imports K+ with a system called TrkF , which is a combination of several minor nonspecific ( called “illicit” ) transport pathways [32] . In this work , we show that trkA and trkE mutations restore the ΔholD mutant growth . The viability of ΔholD Δtrk mutants requires the Kdp and the TrkF K+ import pathways to be of low or negligible activity . We propose that HolD-less Pol III HE can replicate E . coli chromosomes when K+ intracellular concentration is affected , owing to its stabilization on DNA by improved electrostatic interactions .
As inactivation of the holD gene prevents E . coli growth , ΔholD mutants were constructed and propagated in the presence of pAM-holD , a plasmid that carries the holD wild-type gene and replicates only in the presence of IPTG . Suppressors of the ΔholD growth defects can be obtained by growing ΔholD [pAM-holD] cells in the absence of IPTG and selecting for plasmid-less fast growing clones [16 , 17] . Four such ΔholD fast growing colonies were isolated on minimum medium ( M9 ) at 37°C in an MG1655 background ( JJC6376 to JJC6379 ) ( unless otherwise indicated , all minimal media used in this work contain 0 . 4% glucose and 0 . 2% casamino acids ) . The SOS response is induced in the ΔholD mutant and a sfiA mutant was used to prevent cell division blockage by the SOS-induced SfiA protein . Suppressor mutations were also isolated in a ΔholD sfiA::Mu strain , and two fast growing colonies obtained on M9 at 30°C were studied ( JJC6389 and JJC6390 ) . Interestingly , whole genome sequencing identified the presence of a mutation affecting the trk pathway of K+ import in two of these six clones . These mutations were confirmed by re-sequencing the genes of interest . JJC6377 carries a 84 base pair ( bp ) in-frame deletion in the trkA gene , from nucleotides 513 to 597 , with 9 bp microhomology at the junction ( called hereafter ΔtrkA84 , S1 Fig ) . It also carries a large duplication , from about 3 648 000 to around 4 167 000 . It has to be noted that a holD mutation was shown to increase the frequency of recombination events between repeated sequences , which may account for the recurrent presence of deletions and duplications in suppressed holD mutants [16 , 33] . , JJC6389 carries a trkE point mutation at position 767 changing glutamine 255 to proline . This was the only sequence modification identified in this strain . The suppressor mutations identified in the four other sequenced genomes will be described in future publications . The two ΔholD clones affected for K+ import formed colonies on LB at 30°C , 37°C , and 42°C overnight . On M9 , the strain carrying ΔtrkA84 formed colonies at all temperatures , while the strain mutated in trkE formed colonies at 30°C and 37°C only ( Fig 1 , the two original suppressed clones are called ΔholD trkA sup and ΔholD trkE sup ) . However , the spontaneous suppressor strain carrying ΔtrkA84 also carries a large duplication in addition to the ΔtrkA84 mutation , which could improve its viability , whereas the trkEQ255P mutation was the sole genome modification identified in JJC6389 . Co-transduction of ΔtrkA84 with an adjacent marker ( zhd-3082::Tn10 ) in a ΔholD [pAM-holD] or in a ΔholD sfiA [pAM-holD] background yielded clones that , after plasmid loss , were able to grow on LB at all temperatures and on M9 at 30°C and 37°C , and were only slightly more viable than the trkEQ255P suppressed strain at 37°C ( Fig 1 ΔholD ΔtrkA84 , ΔholD ΔtrkA84 sfiA ) . A ΔtrkA::Cm mutant lacking the entire trkA sequence was constructed by gene replacement and behaved as the ΔtrkA84 deletion ( Fig 1 ΔholD trkA::Cm ) , suggesting that the ΔtrkA84 deletion is a null allele . The use of a null allele of trkE showed that the inactivation of trkE or trkA restores holD mutant growth to the same extent , as expected since these two genes belong to the same pathway . The trkEQ255P mutation was slightly less efficient than the ΔtrkE deletion in restoring holD mutant growth at 37°C , suggesting a residual activity of the mutant protein . We conclude that inactivation of the trk pathway of K+ import restores growth of the ΔholD mutant on LB at all temperatures and on M9 at 30°C and 37°C . To check that the ΔtrkA84 , ΔtrkA::Cm , ΔtrkE::kan and trkEQ255P mutations allow ΔholD mutant growth by affecting K+ import , we tested the effect of different K+ concentrations in the external medium on the viability of the different ΔholD trk mutants . In K+-limiting conditions the kdpFABC operon is induced by the regulatory proteins KdpD and KdpE [30] . As shown in Figs 2 and S2 , ΔholD ΔtrkA84 , ΔholD ΔtrkE and ΔholD trkEQ255P did not form colonies on minimal medium containing 0 . 2 mM or 1 mM K+ ( called MK0 . 2 and MK1 , respectively ) , although , as expected , they could grow on the same medium containing 22 mM K+ ( MK22 ) , the K+ concentration in M9 . Lethality of ΔholD trk mutants on plates containing low K+ concentrations may result from the activation of the kdp operon , which promotes active K+ uptake below 5 mM in the external medium [30] . Inactivation of kdpA restored the growth of ΔholD ΔtrkA84 and ΔholD ΔtrkE mutants on MK0 . 2 and MK1 ( ΔholD ΔtrkA84 Δkdp , ΔholD ΔtrkE Δkdp , Figs 2 and S2 ) , indicating that the lethality of holD trk mutants at low K+ concentrations results from the activity of Kdp . However , inactivation of kdp in a Trk+ ΔholD did not restore viability at low K+ concentrations , with the exception of a partial growth at 30°C on MK0 . 2 ( ΔholD Δkdp Figs 2 and S2 ) . Because the Trk system is constitutively expressed and has a high Vmax it is still partly active even on MK0 . 2 ( the growth rate of the kdp mutant is only 20% lower in 0 . 2 mM than in 5 mM K+ [18 , 34] ) . On MK0 . 2 , the remaining activity of Trk in the holD kdp mutant is responsible for the growth defect , but this activity is limited , allowing a weak but significant residual growth at 30°C . Finally , it has to be noted that the K+ concentration in our LB was measured and found to be 10 . 9 ± 0 . 74 mM , but as shown below other components than K+ also play a role in the growth of ΔholD Δtrk mutants on LB . The Trk system is the main K+ import system in E . coli . Therefore , inactivating trkA might decrease K+ intracellular concentration in growing cells . As shown in Table 2 , intracellular K+ concentration was significantly decreased in trkA and holD trkA mutants compared to wild-type cells . This result suggests that a 12–17% decrease in intracellular K+ is sufficient to promote HolD-less Pol III stabilization on DNA . In conclusion , when ΔholD trkA ( or holD trkE ) cells are grown on 22 mM K+ ( M9 or MK22 ) , K+ uptake is impaired and intracellular K+ concentration decreases to a level that rescues the ΔholD mutant . At low K+ concentrations ( 0 . 2 mM and 1 mM ) the activation of the kdp operon increases K+ uptake to a level that prevents the growth of ΔholD ΔtrkA and ΔholD ΔtrkE mutants . By preventing Kdp-mediated K+ import , kdp gene inactivation restores growth of the ΔholD ΔtrkA Δkdp and ΔholD ΔtrkE Δkdp mutants . We conclude that viability of the ΔholD mutant can be restored by decreasing K+ import . Mutants lacking all three K+ import systems ( trk , kdp , kup ) require a high K+ concentration for growth and rely on multiple minor K+ import activities called TrkF [32] . To test whether TrkF activity prevents growth of the holD trkA and holD trkE mutants on medium containing a high level of K+ , colony formation of the ΔtrkA , ΔholD ΔtrkA , ΔholD ΔtrkE and ΔholD trkEQ255P mutants was compared on synthetic medium containing 22 mM ( M9 , MK22 ) , 115 mM ( MK115 ) or 150 mM ( MMA ) K+ . As shown in Figs 2 and S2 , a high concentration of K+ prevented growth of the ΔholD trk mutants; therefore activating K+ import by a high concentration of K+ in the growth medium is lethal to the ΔholD trk mutants . It was not possible to inactivate TrkF , which is not a defined locus but a combination of several minor pathways [32] . Growth of the ΔholD ΔtrkA mutant was compared in different liquid media ( Fig 3 ) . For these experiments , the ΔholD ΔtrkA [pAM-holD] mutant was grown overnight at 30°C to saturation in LB , M9 , MK1 or MK115 medium without IPTG , diluted to OD 0 . 002 in the same medium , and grown at 37°C; growth was monitored by plating appropriate dilutions on M9 . As shown in Fig 3A , growth was rapid in LB , slower in M9 , and stopped after two or three generations in MK1 and MK115 , in agreement with the lack of colony formation at these K+ concentrations . Generation times were calculated from the slope of the best fit straight line during exponential growth ( Fig 3B , generation times could not be calculated for the ΔholD ΔtrkA mutant grown in MK1 or MK115 owing to the rapid growth arrest ) . The growth rate of the single ΔtrkA mutant was similar in M9 , MK1 and MK115 , and not significantly different from wild-type ( 30 min ) , while the generation time of ΔholD ΔtrkA cells in M9 was nearly 50% longer ( 43 min ) . Surprisingly , the generation time of ΔholD ΔtrkA cells was similar to that of wild-type and ΔtrkA single mutant in LB ( 22–23 min ) , in agreement with overnight colony formation on LB , confirming that the rescue of the holD mutant by trkA is very efficient in LB . The kup gene , originally called trkD , is constitutively expressed and active in the same growth conditions as the Trk system , but Kup has a low level of activity [22 , 26] . A Δkup mutant was used to test whether the Kup pathway plays a role in the viability of ΔholD and ΔholD ΔtrkA mutants at different K+ concentrations . Inactivating kup did not rescue the ΔholD mutant on M9 ( S3A Fig ) , in agreement with the idea that Trk is the main K+ import pathway under these conditions . It did not affect the growth of the ΔholD ΔtrkA mutant at any K+ concentration tested ( S3A Fig ) , including in 10 . 9 mM K+ , the K+ concentration in our LB . These results are in agreement with the idea that Trk is the major K+ import system on M9 , and that the Kdp K+ import system is activated at low K+ concentrations . They also suggest that at high K+ concentrations the poor viability of ΔholD trkA cells is caused by TrkF activity , since Kup and TrkF are the only active pathways at high K+ concentrations in a trk mutant , and the phenotype of the holD trk mutant is not affected by Kup inactivation . Surprisingly , at 42°C ΔholD trkA mutants formed colonies overnight ( ON ) on LB but not on minimal medium ( Fig 1 ) . This result was unexpected since the number of replication forks per cell is increased in rich medium compared to minimal medium , which is expected to disfavor a mutant that lacks a Pol III HE subunit . Furthermore , LB contains 10 . 9 mM K+ , a concentration that does not allow growth of the ΔholD trkA mutant at 42°C in synthetic medium ( S3A Fig ) . Consistent with the idea that a low intracellular K+ concentration restores viability by stabilizing HolD-less Pol III HE-DNA complexes , the lethality on M9 at 42°C could result from a destabilization of replication complexes by temperature . To understand why this destabilization is not observed in LB , we tested the three LB components individually ( tryptone , yeast extract and NaCl ) . As shown on Fig 4 , adding yeast extract to M9 casamino acids medium allowed growth of the ΔholD ΔtrkA and ΔholD trkEQ255P mutants at 42°C . Tryptone is the product of casein hydrolysis by trypsin , while casamino acids are the product of acidic hydrolysis of casein . Consequently , tryptone contains mainly peptides while casamino acids contain mainly free amino acids . Replacing casamino acids by tryptone in M9 improved growth of holD trk mutants , although colonies were heterogeneous in size , possibly because growth remains slow , favoring the appearance of new suppressor mutations . Therefore , although yeast extract is more efficient than tryptone , each of these two components can improve growth at 42°C in M9 . Note that the presence of casamino acids increases growth rate , but does not affect the plating efficiency of holD trkA cells at any temperature ( S3B Fig ) . The salts in M9 are 92 . 5 mM Na+ and 22 mM K+ , while LB is 171 mM NaCl ( 10 g/l ) and 10 . 9 mM K+ . The ΔholD ΔtrkA mutants formed slow growing colonies and ΔholD trkEQ255P did not grow at 42°C on low salt LB ( 0 . 5 g/l NaCl , 8 . 5 mM ) ( Fig 4 ) . As on M9 at 37°C , the ΔholD trkEQ255P mutant was more impaired than the ΔholD ΔtrkA mutant , presumably because of a residual activity of the mutated TrkE protein . Nevertheless , the high Na+ concentration in LB improves growth of both ΔholD ΔtrkA and ΔholD trkEQ255P mutants at 42°C . In conclusion , all three components of LB , and particularly yeast extract in the presence of high Na+ , participate to the viability of the ΔholD trk mutants at 42°C . ψ ( HolD ) plays a dual role in the clamp loader complex: its interaction with τ ( DnaX ) stabilizes the complex , and its interaction with χ ( HolC ) connects clamp loading and Okazaki fragment synthesis through the χ-SSB interaction . Accordingly , a ΔholC mutation , which lacks only the clamp loader-SSB interaction , is less deleterious than the ΔholD mutation , particularly at 30°C [16] ( Figs 1 and 5 ) . The ΔtrkA mutation improved the ΔholC mutant growth at 30°C , 37°C , and 42°C on LB , and at 30°C and 37° on M9 . Results were variable on M9 at 42°C , with either no growth of the ΔholC ΔtrkA mutant or , as in the example shown in Fig 5 , appearance of new suppressor mutations . ΔholC ΔholD ΔtrkA mutant viability was similar to that of the ΔholD ΔtrkA mutant , showing that suppression of the ΔholD growth defects by trkA inactivation does not require χ ( Fig 5 ) . The effects of K+ concentration on the suppression of the holC mutant growth defects by trk inactivation were tested ( S4 Fig ) . At 30°C , presumably owing to a significant growth of the ΔholC mutant , the ΔtrkA mutation improved growth at all K+ concentrations , forming smaller colonies only on MK1 . Therefore K+ import by Kdp or TrkF did not prevent ΔholC ΔtrkA mutant growth . However , at 37°C where growth of the ΔholC mutant is severely impaired , rescue by the ΔtrkA mutation was efficient only on 22 mM K+ , while on low and high K+ concentrations colonies were highly heterogeneous , indicating that improvement of residual growth mainly favored the appearance of new suppressor mutations . As expected , rescue of ΔholC ΔholD mutant was weak or null at low or high K+ concentrations compared to rescue at 22 mM ( S4 Fig ) . Therefore , a defect in K+ import allows chromosome replication in the absence of either ψ , or χ or both . We previously showed that growth of the ΔholD mutant is improved upon inactivation of SOS by a lexAind or a recF mutation [17] . These suppressor mutations were more efficient in the AB1157 context than in the context used here , MG1655 , where ΔholD lexAind and ΔholD recF viability was mainly improved at 37°C ( [16] ( Fig 6 ) . We tested whether the trk mutations act through a decrease of SOS induction . Measures of SOS induction using a sfiA::lacZ fusion showed that the SOS response was induced in the ΔholD ΔtrkA mutant as in the ΔholD mutant ( Table 3 ) . SOS induction was RecF-dependent indicating that it results from the accumulation of single-stranded DNA gaps in the ΔholD ΔtrkA mutant as in the ΔholD mutant . Furthermore , preventing SOS induction and trk inactivation showed an additive effect on the viability of the holD mutant , as ΔholD ΔtrkA lexAind and ΔholD ΔtrkA recF mutants were viable on M9 at 42°C ( Fig 6 ) . Rescue of the ΔholD lexAind and ΔholD recF mutants by ΔtrkA was only efficient on 22 mM K+ and was not observed at low or high K+ concentrations , indicating that it requires alternative K+ import systems to be of low or negligible activity ( S5 Fig ) . We conclude from these experiments that ΔtrkA did not restore the viability of the ΔholD mutant by preventing RecF-dependent SOS induction , and that the ΔholD ΔtrkA mutant still accumulates single-stranded DNA gaps during replication . If these gaps result from a delay in the use of RNA primers for the synthesis of Okazaki fragments , the strain might be sensitive to an excess of RNase H , which could destroy the RNA primers prior to their elongation . The results in Table 4 shows that the presence of a 20 copy plasmid that expresses RNase H prevented the formation of ΔholD ΔtrkA colonies , although it did not affect the growth of HolD+ cells , suggesting that single-stranded DNA gaps are , at least in part , caused by a defect in RNA primer elongation . We previously showed that two SOS-induced proteins play an important role in the ΔholD mutant growth defects , the bypass polymerases DinB ( Pol IV ) and Pol II , and we proposed that these SOS-induced proteins are deleterious in the holD mutant because they compete efficiently with HolD-less Pol III for β-clamps binding at replication forks [17] . Accordingly , the over-production of DinB with a deletion of the last 5 amino acids ( DinBΔC5 ) , which fails to bind DnaN [35] , was not lethal in a ΔholD ssb-duplicated strain [16] . The amount of DinB is expected to be similar in holD trkA , where DinB is 5- to 8-fold over-expressed owing to SOS-induction , and in holD trkA lexAind [pGB-DinB] , where it is not SOS-induced but expressed from a 8–10 copy pSC101 vector [16 , 36] . pGB2-DinB could be introduced by transformation in the ΔholD ΔtrkA lexAind mutant , in which SOS is inactivated , ( Fig 7 ) , however it slightly slowed down growth , as cells harboring pGB2-DinB formed smaller colonies than those harboring the vector pGB2 or the control plasmid pGB-DinBΔC5 . In the ΔholD ΔtrkA mutant , where the SOS response is induced , pGB2 and pGB-DinBΔC5 could be introduced while pGB-DinB could not ( Fig 7 ) . We conclude that the ΔtrkA mutation restores growth by preventing SOS-induced DinB proteins from destabilizing the HolD-less Pol III HE , but ΔholD ΔtrkA [pGB-DinB] cells remain sensitive to the large excess of DinB resulting from the combination of plasmid-borne expression and SOS induction . The ΔholD mutant is rescued by a duplication of the ssb gene , and we proposed that this duplication allows a modification of the SSB-binding mode , from 65 nucleotides to 35 nucleotides bound by an SSB tetramer , which stabilizes HolD-less Pol III on DNA [16] . Low salts favor the ( SSB ) 35 DNA binding mode in vitro ( [37] and references therein ) , and to test whether the decrease in intracellular concentration of K+ in trk mutants suppressed the ΔholD mutant defects through a modification of the SSB binding mode , the ssb duplication was introduced in the ΔholD ΔtrkA mutant . The ssb gene duplication showed an additive effect with ΔtrkA mutation , allowing growth of the ΔholD ΔtrkA argE::ssb triple mutant on M9 at 42°C ( Fig 6 ) . Rescue of the ΔholD argE::ssb mutants by ΔtrkA was only efficient on 22 mM K+ and was not observed at low or high K+ concentrations , as expected for the role of intracellular K+ concentration in this phenomenon ( S5 Fig ) . The additive effects of ssb gene duplication and trkA inactivation on ΔholD viability at 42°C suggests that ΔtrkA does not act only by modifying the SSB binding mode . In fact , variations of temperature from 25°C to 37°C only modestly affect the SSB binding mode in vitro [38] , 42°C was not tested ) , which disfavors the hypothesis that in vivo the ssb gene duplication promotes the ( SSB ) 35 binding mode at 37°C and not at 42°C . If indeed in strains carrying the ssb duplication SSB binds DNA in the 35 base-pair mode at 42°C as at 37°C , the lethality of the holD argE::ssb mutant at 42°C results from a lack of stabilization of HolD-less Pol III by ( SSB ) 35 at this temperature , and , in turn , the viability of the holD argE::ssb trkA triple mutant at 42°C suggests an effect of the trkA mutation on Pol III stability . However , we cannot exclude that trk mutations could promote a shift to the ( SSB ) 35 at 30°C and 37°C on their own , and at 42°C in cells that carry the ssb gene duplication . Because these two modes of action are not exclusive and may have cumulative effects , the trkA mutation could stabilize Pol III both by enhancing directly its binding to DNA and by promoting the replication-favorable ( SSB ) 35 binding mode , as discussed below .
The finding that decreasing the intracellular K+ concentration rescues the ΔholD mutant is in line with the idea that ψ ( HolD ) plays an important role in vivo in the stabilization of Pol III HE on DNA during replication . In vitro , ψ increases the affinity of τ for δδ’ in the core loading clamp pentameric complex , stabilizes the ATP-activated clamp loader complex conformational state , and increases the affinity of the clamp loader complex for the primer-template junction , thus favoring clamp loading [10–12] . Therefore ψ improves both protein-protein and protein-DNA interactions . The association of Pol III HE with DNA is intrinsically salt-dependent and strongly depends on the anion used for the reaction , with glutamate , the most physiological anion , protecting against high salt destabilizing effects [39] . Our results show that an intrinsically unstable Pol III HE can be stabilized by decreasing intracellular K+ concentration . However , our results argue that clamp loading during lagging-strand synthesis is still defective in the ΔholD ΔtrkA mutant , since the SOS response remains highly induced , and RNA primers remain sensitive to an excess of RNaseH . Furthermore , the inactivation of trk improves the viability of the ΔholC mutant , although in vitro clamp-loader activity is mainly stimulated by ψ alone and the presence of χ only weakly affects the reaction [11] . Therefore , we propose that lowering K+ import restores the viability of the ΔholD mutant mainly by improving leading-strand replication , by limiting replication arrest or Pol III HE disassembly after arrest , or by facilitating replication restart . A duplication of the ssb gene also rescues the holD mutant , and we hypothesized that this rescue results from the promotion of the ( SSB ) 35 binding mode by a two-fold increase in intracellular SSB protein concentration . Because in vitro a shift from the ( SSB ) 65 to the ( SSB ) 35 binding mode can be promoted by increasing SSB concentration or decreasing salt concentration [37] , it is conceivable that the 12–17% decrease in intracellular K+ concentration observed in trk mutants promotes the formation of ( SSB ) 35 , which , in turn , stabilizes the HolD-less Pol III on DNA . In conclusion , the trkA mutation could allow chromosome replication either by a direct effect on Pol III , if affecting potassium import increases electrostatic interactions between Pol III subunits and/or between Pol III and DNA , or indirectly , if it favors a shift to the SSB binding mode to ( SSB ) 35 . These two models are not exclusive and the trk mutation could affect Pol III stability on DNA by cumulative effects of a direct and an SSB-mediated stabilization . Furthermore , the observation that the difference in K+ intracellular concentration between wild-type and trkA mutants is only 12–17% supports the idea that viability of holD trkA mutant might result from a combination of potassium concentration effects . Inhibition of SOS induction by a lexAind or a recF mutation restores full viability to ΔholD ΔtrkA cells , including at high temperatures . This observation suggests that the remaining growth defects of ΔholD ΔtrkA cells are caused by SOS-induced polymerases and that the combination of a more stable Pol III and less abundant Pol II and DinB competitors is sufficient to restore full viability . Altogether , we conclude that electrostatic interactions are crucial for replisome stability in vivo and can be improved beyond the wild-type level by decreasing K+ import . K+ glutamate is the natural solute in E . coli , and is the major intracellular ionic osmolyte [20] . K+ intracellular concentration is regulated after hyper- or hypo-osmotic shocks by changes in the amount of water and by the action of several K+ import and efflux systems . In spite of a strong effect of salt concentration on protein-DNA interactions in vitro , K+ intracellular concentration can largely increase in vivo without affecting lac operator-repressor or phage λ RNA polymerase-promoter interactions [20] . Following a hyper-osmotic shock only some specific promoters are affected by the increased intracellular concentration of K+ glutamate: genes involved in osmotic protection are induced and ribosomal transcription is decreased [40]; effects of K+ transporters on the virulence of pathogenic bacteria were also reported [41 , 42] . To account for the lack of effect of hyper-osmotic conditions on operator-repressor binding and promoter activity , it was proposed that increased intracellular K+ concentrations trigger a decrease in free cytoplasmic water , which enhances molecular crowding and thereby compensates for the destabilizing effect of the original K+ concentration increase [43–45] . Here the use of a mutant where the replication complex is intrinsically unstable on DNA allows us to show that protein-DNA interactions and possibly protein-protein interactions can be increased by lowering K+ import below the physiological wild-type level . We would like to propose the following hypotheses to account for the strong effect on viability in spite of a relatively weak K+ concentration decrease ( i ) either the replication machinery is highly sensitive to weak variations of intracellular K+ concentrations , for example because the effects on several replication proteins such as Pol III and SSB are cumulative as discussed above , ( ii ) or the 80 mM decrease in total K+ concentration that we observe in the holD trk mutants compared to wild-type affects only K+ ions free for exchange , leading to a 30% reduction of the potassium ions that affect DNA-protein interactions effectively , ( iii ) or trk and kdp mutations exert secondary effects on ions other than K+ that also control protein-DNA interactions . Strikingly , stabilization of Pol III HE on DNA , reflected by the ΔholD ΔtrkA mutant viability , is particularly efficient in LB . The effects of LB are likely the result of a combination of multiple factors , including the presence of molecules such as glycine betaine and glutamate that stabilize protein-DNA complexes [39 , 46] . However , only ΔholD mutants that lack the trk import system formed colonies on LB at 42°C , and not ΔholD mutants that lack SOS induction or carry an ssb gene duplication . Therefore , whatever the compounds that favor growth in LB , they are active in the holD mutant when combined with a limited K+ import . Replication fork arrest is a recognized source of genome rearrangements in all organisms , and any replication defect can have severe consequences [47–50] . The identification of factors that improve replication fork stability in perturbed conditions is therefore crucial . Furthermore , theoretically mutations that affect protein-protein and protein-nucleic acid interactions in processes other than replication could also be suppressed by limiting K+ import . Our work underlines the influence of chemical intracellular composition on essential processes .
Strains , plasmids and oligonucleotides used in this work are described in S1 Table . Genes were inactivated by recombineering as described in [51] using DY330 [52] . Mutations were transferred by P1 transduction . Antibiotics were used at the following concentrations: kanamycin ( Kan ) 50 μg/ml , chloramphenicol ( Cm ) 20 μg/ml , tetracycline ( Tet ) 15 μg/ml , ampicillin ( Ap ) 100 μg/ml , spectinomycin ( Spc ) 60 μg/ml . All minimum media used in this work contain 0 . 4% glucose , 0 . 2% casamino acids and 1 mg/L thiamine , except M9 tryptone medium which contains 0 . 4% tryptone instead of casamino acids . LB broth ( Miller ) is from Sigma , yeast extract , tryptone and casamino acids are from Difco . M9 is Na2HPO4 42 mM , KH2PO4 22 mM , NaCl 8 . 5 mM , NH4Cl 18 . 7 mM , MgSO4 1 mM , CaCl2 0 . 1 mM [53]; MMA is K2HPO4 60 . 3 mM , KH2PO4 33 mM , ( NH4 ) 2SO4 7 . 6 mM , Na Citrate 1 . 7 mM , MgSO4 1 mM [53]; MK115 is K2HPO4 46 mM , KH2PO4 23 mM , ( NH4 ) 2SO4 8 mM , Na Citrate 1 mM , MgSO4 0 . 4 mM , FeSO4 6 μM [29]; MK0 is Na2HPO4 46 mM , NaH2PO4 23 mM , ( NH4 ) 2SO4 8 mM , Mg SO4 0 . 4 mM , FeSO4 6 μM . Different amounts of MMA 2X or MK115 2X were added to MK0 for MK0 . 2 , MK1 and MK22 , to adjust to 0 . 2 , 1 and 22 mM K+ respectively [29] . Strains containing pAM-holD were routinely grown in M9 containing 500 mM IPTG and 60 μg/ml spectinomycin at 37°C . pAM-holD ( or pAM-holCD ) were cured prior to each experiment by growing cells in the absence of IPTG , and plasmid-less colonies were isolated on M9 [16 , 17] . We determined that less than 10% cells in the culture contain pAM-holD and less than 1% had acquired a suppressor mutation . Because of the high frequency of appearance of suppressor mutations , all new holD derivatives were constructed in the presence of pAM-holD . All mutations introduced by P1 transduction were verified by PCR , and all mutations constructed by recombineering were verified by PCR and sequencing . lexAInd and recF mutations were checked by measuring UV sensitivity . For spot assays , plasmid-less colonies formed in three days on M9 at 30°C were suspended in M9 or MK0 salts . Serial 10-fold dilutions were performed and 5 μl of dilutions 10−1 to 10−5 were spotted on different media . Pictures of LB plates incubated at 37°C and 42°C were taken after 24 h incubation , for all holD mutants , pictures of LB plates incubated at 30°C and of minimum medium plates were taken after 2 days; for HolD+ strains for pictures of minimal medium plates incubated at 37°C were taken after 24 h incubation . All strains were tested at least three times independently . For growth curves , cultures of wild-type ( JJC1392 ) , trkA ( JJC6800 ) and holD trkA [pAM-holD] ( JJC6898 ) strains were grown overnight at 30°C in LB , M9 , MK1 , MK115 medium . Cells were diluted to O . D . 0 . 002 in the same medium and further grown at 37°C for 7 hours . This protocol was chosen because it allows overnight cultures and the subsequent growth curves to be performed in the same medium , without medium shift , and a direct comparison with the same protocol of viable ( holD trkA in M9 and LB ) and lethal ( holD trkA in MK1 and MK115 ) growth conditions . The number of colony forming units per ml of culture ( cfu/ml ) was determined by plating appropriate dilutions on M9 and incubating plates at 30°C . The average percentage of plasmid-less cells , determined by plating appropriate dilutions on M9 with spectinomycin and IPTG , was independent of the medium , in average 74% after overnight propagation and 96% at the end of the growth curve . To verify that the holD trkA cultures did not acquire additional suppressor mutations during growth in M9 , appropriate dilutions were also plated at 42°C , to check that cells were thermosensitive as expected . Chromosomal DNA was extracted using Sigma GenElute bacterial genomic DNA kit . 5 μg of DNA were used to generate a genomic library according to Illumina's protocol . The libraries and the sequencing were performed by the High-throughput Sequencing facility of the I2BC ( http://www . i2bc . paris-saclay . fr/spip . php ? article399&lang=en , CNRS , Gif-sur-Yvette , France ) . Genomic DNA libraries were made with the ‘Nextera DNA library preparation kit’ ( Illumina ) following the manufacturer’s recommendations . Library quality was assessed on an Agilent Bioanalyzer 2100 , using an Agilent High Sensitivity DNA Kit ( Agilent technologies ) . Libraries were pooled in equimolar proportions . Paired-end 2x250 bp reads were generated on an Illumina MiSeq instrument , using a MiSeq Reagent kit V2 ( 500 cycles ) ( Illumina ) , with an expected depth of 217X . Reads from mutant genome were aligned on the Escherichia coli K12 MG1655 genome using Illumina's package CASAVA 1 . 8 . 2 . The point mutation and the small indels were detected also using Illumina's package CASAVA 1 . 8 . 2 and the large indels with profil visualisation and Blast ( Basic Local Alignment Search Tool ) . Cells grown in M9 at 37°C until OD650 = 0 . 4 were cooled in ice , harvested by centrifugation , and washed three times in cold hyper-tonic medium: 1 . mM Tris-Cl ( pH 8 ) , 1 mM MgSO4 , and 0 . 4 M glucose [31] . Pellets were dried overnight at 56°C . Dry pellets were weighted , digested in 2 ml of HNO3 >68% ( 20 min at 80°C and 1h at 120°C ) , diluted 50 fold in H2O , and K+ was measured by flame spectrophotometry using a Varian AA240FS spectrophotometer and a range of 0 . 1 to 5 mg/L K+ standard solutions . β-galactosidase assays for measures of SOS induction were performed as described previously [16 , 53] . Since isolated JJC6545 and JJC7058 colonies could not be cultivated owing to the growth advantage of suppressor mutations , pAM-holD containing clones were grown overnight in M9 lacking IPTG and diluted 50 fold in M9 for the experiment . Cultures were tested for the loss of pAM-holD and for containing less than 1% suppressor mutations .
|
Replication polymerases are responsible for genome duplication; they are ubiquitous and show high levels of functional and structural conservation across all species . The HolC-HolD ( χψ ) complex is a component of the replicative polymerase in the model bacteria Escherichia coli , and is crucial for normal growth . We isolated suppressor mutations that restore the viability of the holD mutant and we found that inactivating the Trk system , responsible for the main pathway of potassium import , renders the entire χψ complex dispensable for growth . Activation of alternative pathways of potassium import abolishes the suppression . The viability of the holD trk mutant is due in large part to a better capacity of the χψ-less polymerase to compete with other polymerases . Potassium glutamate is the major intracellular ionic osmolyte in E . coli , and we propose that mutations that affect potassium concentration in vivo stabilize the χψ-less polymerase by increasing electrostatic interactions between the different polymerase subunits and between polymerase and DNA . Stabilized by the lower intracellular potassium concentration , the χψ-less polymerase becomes functional in trk mutants , to a level that permits cell growth with this defective polymerase . Our results imply that K+ import can play an important role in the stability of protein complexes on DNA .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"protein",
"interactions",
"dna-binding",
"proteins",
"cloning",
"mutation",
"polymerases",
"dna",
"replication",
"gene",
"types",
"dna",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"proteins",
"molecular",
"evolution",
"molecular",
"biology",
"biochemistry",
"point",
"mutation",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"gene",
"duplication",
"evolutionary",
"biology",
"suppressor",
"genes"
] |
2016
|
Mutations Affecting Potassium Import Restore the Viability of the Escherichia coli DNA Polymerase III holD Mutant
|
Solitary cysticercus granuloma ( SCG ) is the commonest form of neurocysticercosis in the Indian subcontinent and in travelers . Several different treatment options exist for SCG . We conducted a Bayesian network meta-analysis of randomized clinical trials ( RCTs ) to identify the best treatment option to prevent seizure recurrence and promote lesion resolution for patients with SCG . PubMed , EMBASE and the Cochrane Library databases ( up to June 1 , 2015 ) were searched for RCTs that compared any anthelmintics or corticosteroids , alone or in combination , with placebo or head to head and reported on seizure recurrence and lesion resolution in patients with SCG . A total of 14 RCTs ( 1277 patients ) were included in the quantitative analysis focusing on four different treatment options . A Bayesian network model computing odds ratios ( OR ) with 95% credible intervals ( CrI ) and probability of being best ( Pbest ) was used to compare all interventions simultaneously . Albendazole and corticosteroids combination therapy was the only regimen that significantly decreased the risk of seizure recurrence compared with conservative treatment ( OR 0 . 32 , 95% CrI 0 . 10–0 . 93 , Pbest 73 . 3% ) . Albendazole and corticosteroids alone or in combination were all efficacious in hastening granuloma resolution , but the combined therapy remained the best option based on probability analysis ( OR 3 . 05 , 95% CrI 1 . 24–7 . 95 , Pbest 53 . 9% ) . The superiority of the combination therapy changed little in RCTs with different follow-up durations and in sensitivity analyses . The limitations of this study include high risk of bias and short follow-up duration in most studies . Dual therapy of albendazole and corticosteroids was the most efficacious regimen that could prevent seizure recurrence and promote lesion resolution in a follow-up period of around one year . It should be recommended for the management of SCG until more high-quality evidence is available .
Neurocysticercosis ( NCC ) , a parasitic disease of the nervous system caused by Taenia solium ( pork tapeworm ) , is a leading cause of acquired epilepsy worldwide [1 , 2] . The disease is widely prevalent around the world , and has pleomorphic clinical and radiologic manifestations [1] . Solitary cysticercus granuloma ( SCG ) , presenting as a single small enhancing lesion , is found in ~20% of NCC cases in endemic areas , and is the commonest type of NCC in the Indian subcontinent as well as in travelers of industrialized countries returning from endemic zones [3 , 4] . SCG has traditionally been considered the degenerating form of long-established vesicular cyst that cannot maintain immune evasion and thus is under the host’s immune attack . A recent hypothesis proposes that SCG represents fresh infection that is rapidly detected and destroyed by the host’s immune system . [5] Treatment might be different for patients with live and degenerative/dead parasite . While there is sufficient information in support of the use of the combination of anthelmintics and corticosteroids in patients with viable cystic parenchymal NCC [6–10] , the treatment of SCG has not been optimally defined [11] . Besides , the recent American Academy of Neurology ( AAN ) evidence-based guideline on NCC didn’t address management issues of different types of lesion independently [12] . Currently , the overall evidence from randomized clinical trials ( RCTs ) on drug therapy for SCG consists of comparisons between the combination of anthelmintics and corticosteroids therapy , anthelmintics therapy alone , corticosteroids therapy alone and conservative treatment ( limited to treatment of symptoms ) , such as antiepileptic drugs alone without anthelmintics or corticosteroids . Several pairwise meta-analyses have evaluated the independent efficacies of anthelmintics and of corticosteroids [9 , 13 , 14] . However , multiple different regimens have never been compared with each other simultaneously . The network of evidence can be better examined in a mixed treatment comparison framework with Bayesian method [15 , 16] . This approach fully respects randomization , accounts for the correlation of multiple observations within the same trial , and allows estimation of relative efficacies of different drugs and their combination . Here , we systematically reviewed and analyzed RCTs on drug therapy for SCG and conducted a Bayesian network meta-analysis to determine the effect of different therapies on seizure control and on radiological resolution of the disease .
We searched the electronic databases of PubMed , EMBASE and the Cochrane Library ( from inception until June 1 , 2015 ) without restrictions on language or publication date . The logic combinations of the following terms were searched in the Title/Abstract: cysticercosis , neurocysticercosis , solitary cysticercus granuloma , single small enhancing computed tomographic lesion , cysticidal , anticysticercal , anthelmintic , albendazole , praziquantel , corticosteroid , steroid , prednisolone , methylprednisolone , and dexamethasone . The computer retrieval was supplemented by manual search of reference lists of identified studies and ( systematic ) reviews on neurocysticercosis . The identified citations were initially screened at the title and abstract level , and then retrieved as full-text copies if they reported potentially relevant studies . To be included in the analysis , studies had to be randomized clinical trials ( RCTs ) that included patients with new onset seizures and diagnosed with SCG based on clinical and imaging studies according to the accepted criteria [18] . All studies compared the efficacy of anthelmintics ( albendazole and/or praziquantel ) or corticosteroids , or both , head to head or with placebo or no drugs . Studies were excluded if they compared different dosages or durations of the same medication , if only patients with cystic or multiple enhancing lesions were included , and if none of the quantitative outcomes of interest ( see below ) were reported . Studies using concomitant drugs , such as antiepileptic drugs ( AEDs ) were not excluded if balanced among the trial arms . When more than one report describing the same study were published , the one with the most recent or complete data was used for meta-analysis . Two researchers independently reviewed the studies with disagreements in eligibility , methodological quality or data extraction resolved through discussion and consensus . Data were collected for each eligible RCT on study characteristics , patient characteristics , and outcome results . The goal of this study was to compare the efficacies of different drug therapies in the following aspects: seizure recurrence , defined as the occurrence of one or more convulsions after the beginning of treatment , and lesion resolution , defined as complete disappearance of the granuloma with no residual scar , calcification or perilesional edema on imaging studies , by MRI or CT scan . If a study reported outcomes at multiple time points , only data from the most recent follow-up were extracted for analysis . The methodological quality of the included RCTs was appraised using the Cochrane Collaboration’s tool for assessing risk of bias [19] , which consists of seven items: sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessors; incomplete outcome data; selective outcome reporting; and other bias . Blinding and incomplete outcome data were assessed separately for the two primary outcomes . The overall risk of bias of a trial was expressed as low , moderate , or high . Therapeutic interventions were included in quantitative analyses if they had been studied in at least two trials . We conducted Bayesian network meta-analysis using the binomial likelihood model for multi-arm trials [20 , 21] , given the outcomes were dichotomous and included multi-arm trials . Our model adopted random effects because it is the most appropriate and conservative analysis to account for variance among trials . The Markov Chains Monte Carlo method was used for analysis . Three Markov chains ran simultaneously with different initial values . 150 , 000 simulations were generated for each of the three sets of initial values , with the first 50 , 000 discarded to avoid the influence of initial unstable values . The convergence was assessed with trace plots and the Brooks-Gelman-Rubin statistic . The odd ratios ( OR ) were estimated from the median of the posterior distribution and the accompanying 95% credible intervals ( CrI ) , which can be interpreted in the same manner as the conventional 95% confidence interval ( CI ) . For comparison , the estimates from just the head-to-head evidence for each pair of comparison were also worked out with the Mantel-Haenszel method of the conventional pairwise meta-analysis . Furthermore , for each outcome , we estimated the probability that each treatment regimen was the most , the second , the third , and the least efficacious , based on their ranks in each iteration of Markov chain . These probability values were used for generating cumulative probability plots and calculating the Surface Under the Cumulative RAnking curve ( SUCRA ) , with 1 representing the best treatment and 0 the worst [22] . We examined the validity of the network models by evaluating three of their important characteristics . The goodness of model fit was measured by the posterior mean of the residual deviance , which should be close to the data points when the model can provide adequate fit . Heterogeneity was defined as the variability of the results across trials . It was estimated from the posterior median between-study variance τ2 , with τ2 < 0 . 04 indicating a low level of heterogeneity and τ2 > 0 . 40 a high level [23] . Consistency , defined as agreement between direct and indirect sources of evidence , was first assessed visually by comparing the results of network meta-analysis and pairwise meta-analysis , and then tested statistically by calculating the ratio of two odds ratios ( RoR ) from direct and indirect evidence in each closed loop in the network of interventions . RoR values close to 1 mean that the two sources are in agreement [24] . The considerable variation in follow-up duration among the included RCTs and the fact that probability of both primary outcomes are related with time [25] did not allow calculation of the absolute rate difference and number needed to treat for each intervention by using the baseline rates across the conservative treatment arms . Considering that evidence may be different from RCTs with different follow-up duration , we performed meta-regression analysis with follow-up duration ( ≤ 6 months versus ≥ 9 months for seizure recurrence , 3 months versus 6 months for lesion resolution ) as an interaction [26] . We calculated the subgroup interaction term β and checked whether its 95% credible interval included the possibility of no interaction . We performed further sensitivity analysis by sequentially removing one study at a time from the overall dataset . A post hoc analysis was performed to compare different treatments on the risk of residual calcification during the evolution of SCG lesions . Assessment of publication bias using the funnel plots was precluded by the small number of studies included in the meta-analysis . Conventional pairwise meta-analysis was performed with Review Manager 5 . 3 . 3 ( Cochrane Collaboration , Nordic Cochrane Centre , Denmark ) . Network meta-analysis including meta-regression analysis was performed with winBUGS 1 . 4 . 3 ( MRC Biostatistics Unit , Cambridge , UK ) . Test for consistency was conducted with Stata 12 . 0 ( StataCorp LP , College Station , TX ) .
Fig 1 is a flow chart of the study and summarizes the process of trial selection . Twenty articles reporting 16 RCTs met the inclusion criteria [27–42] . The included RCTs covered six different treatment regimens for SCG: albendazole ( evaluated in 5 trials ) , praziquantel ( 1 trial ) , corticosteroids ( 9 trials ) , albendazole plus corticosteroids ( 6 trials ) , albendazole and praziquantel plus corticosteroids ( 1 trial ) , and conservative treatment ( 11 trials ) . The two praziquantel-containing regimens were evaluated only in one trial , so that they and the corresponding trials were not suitable for the network meta-analysis . The main features of the RCTs included in the quantitative analysis are summarized in Table 1 . Fourteen trials involving 1 , 277 randomized patients were included . All the included RCTs were conducted in India where cysticercosis is highly endemic . The proportion of women ranged from 29 . 7% to 47 . 2% , and the mean age of patients at baseline ranged from 7 . 4 to 24 years . Each participant in each RCT , but two , was diagnosed with a solitary enhancing lesion . In one RCT [31] , only 70 . 8% of the patients had a single enhancing lesion while the others carried two or more lesions . However , it was possible to extract the data of patients with single lesions , thus allowing the inclusion in the analysis of data of only these patients . In another trial [32] , some patients ( 18% ) had two rather than one enhancing lesions and the outcome data could not be separated . We decided to include this trial given the small proportion of patients with two lesions . All trials had two arms , except one in which the three active treatments were compared directly with each other [40] . The dosage of albendazole and corticosteroids were similar across the trials , but the duration of treatment varied from 3 to 28 days . Antiepileptic drugs were used in all trials . The follow-up period ranged from 6 to 18 months for seizure recurrence and from 2 to 6 months for lesion resolution . There was high risk of selection bias in most studies because they used random number tables to generate random number sequences with no or unclear method of allocation concealment . The performance bias was high in more than half of the studies due to lack of blinding of participants . Blinding of seizure assessment was unclear or non-existent in those studies too , but for the assessment of lesion resolution , blinding was generally well maintained . S1 Fig shows the assessment process of the risk of bias of the studies included in this meta-analysis . Thirteen RCTs were used for the quantitative analysis of seizure recurrence . The network diagram for this outcome is presented in S2 ( A ) Fig . Network meta-analysis showed that albendazole plus corticosteroids was the only treatment protocol that significantly decreased the recurrence of seizure during the follow-up period compared with conservative treatment ( OR 0 . 32 , 95% CrI 0 . 10–0 . 93 , Figs 2 and 3 ) . The results were similar in the only direct comparison RCT that evaluated albendazole plus corticosteroid versus conservative treatment ( OR 0 . 31 , 95% CI 0 . 11–0 . 89 ) [32] . The risk reduction for corticosteroids alone was marginal outside the level of significance ( 0 . 46 , 0 . 19–1 . 01 ) , and the efficacy of albendazole alone did not even approach statistical significance ( 0 . 66 , 0 . 22–2 . 17 ) . While there were no significant differences among the three active treatments , the ranking probabilities and cumulative probability plots indicated that the combination of albendazole and corticosteroids was superior to either treatment alone ( Fig 3 ) . The combination therapy had the greatest probability of being the best treatment ( Pbest 73 . 3% ) , and the SUCRA values were 0 . 884 , 0 . 637 , and 0 . 388 for albendazole plus corticosteroid , corticosteroid , and albendazole , respectively . A test of subgroup interaction between RCTs with follow-up period of ≥9 months and those with ≤ 6 months was not statistically significant ( subgroup interaction term β 0 . 05 , -1 . 73–1 . 77 ) , adding support to the conclusion that the combination of the two groups of RCTs was not inappropriate . The fit of model was good with the posterior mean of the residual deviance of 26 . 77 , compared with 27 data points . However , the estimated between-study variance was 0 . 54 ( 0 . 03–2 . 52 ) , which is potentially considerable with its uncertainty caused by the relatively small number of studies . Visual inspection of the results from pairwise and network meta analyses showed obvious inconsistency between direct and indirect estimates for the contrast albendazole versus corticosteroids , and this was confirmed by a large RoR value ( 3 . 15 ) of the corresponding loop in the network ( S3 ( A ) Fig ) . Since only one RCT [40] supplied direct evidence for the comparison , we investigated the inconsistency by removing this trial in a sensitivity analysis . The result is presented in S1 Table . The combination of albendazole plus corticosteroid ( 0 . 31 , 0 . 11–0 . 76 ) and corticosteroid alone ( 0 . 37 , 0 . 17–0 . 68 ) both significantly reduced the risk of seizure recurrence , with combination therapy being the better one in probability analysis ( Pbest 67 . 1% versus 31 . 8% , SUCRA 0 . 882 versus 0 . 767 ) . The probability analysis suggested that the efficacy of albendazole monotherapy was even worse than conservative treatment although there was no significant difference between the two ( OR conservative treatment versus albendazole 0 . 73 , 0 . 23–2 . 23 ) . Note that the pooled estimates of network meta-analysis generally overlapped with the results of conventional pairwise meta-analysis ( when available ) and that the estimated between-study variance decreased from 0 . 54 to 0 . 16 . The outcome of lesion resolution was analyzed in all 14 RCTs . The network diagram is presented in S2 ( B ) Fig . In the network meta-analysis , compared with conservative treatment , the efficacy of albendazole plus corticosteroids combination therapy in inducing resolution of SCG was the highest ( 3 . 05 , 1 . 24–7 . 95 ) , followed by albendazole alone ( 2 . 63 , 1 . 61–6 . 34 ) , and corticosteroids alone ( 2 . 32 , 1 . 20–4 . 75 , Figs 2 and 3 ) . The same order was also identified in conventional pairwise meta-analysis but the confidence intervals were wider and only the efficacy of corticosteroid therapy reached statistical significance . The differences between the three treatments were not conclusive . Nevertheless , the combination of albendazole plus corticosteroid ( Pbest 53 . 9% , SUCRA 0 . 789 ) was more likely the best treatment for this outcome in probability analysis , compared with the monotherapy of albendazole ( 33 . 9% , 0 . 659 ) and corticosteroid ( 12 . 1% , 0 . 541 ) ( Fig 3 ) . The posterior mean residual deviance was close to the number of data points ( 29 . 71 compared with 29 ) and thus the model fit was adequate . Heterogeneity was high ( between-study variance 0 . 41 ) but acceptable . RoR values all close to 1 demonstrated no significant inconsistency between direct and indirect evidence for any of the pairwise treatment comparisons ( S3 ( B ) Fig ) . The effect of interaction between the trials with 3-month follow-up and those with 6-month was insignificant ( subgroup interaction term β 0 . 40 , -0 . 84 to 1 . 75 ) , although the point estimate was positive , suggesting that the efficacy of the combination therapy in promoting lesion resolution could be more obvious in short follow-up . This is reasonable since many SCGs resolve spontaneously with time [25] . Sensitivity analyses by sequentially removing one study at a time yielded largely the same results . During review of the literature , 8 studies were identified that included data describing the frequency of residual calcification on follow-up imaging [27–29 , 31 , 34 , 35 , 41 , 42] . Since calcific residue is one of the major predictors for future seizure recurrence [25] , we did an additional post hoc analysis to evaluate the effect of different therapies on reducing the risk of residual calcification of SCG . Only pairwise meta-analysis was conducted because a closed loop for network meta-analysis was not formed . All pooled ORs are close to 1 with wide 95% CIs ( S4 Fig ) , indicating that none of regimens showed a better effect on reducing the risk of residual calcification compared with others .
SCG is the commonest form of NCC seen in India and high-income countries , and is also found in about 20% of NCC cases elsewhere [3–5] . Although the granuloma shows spontaneous resolution with time , complete resolution can take anywhere from a few weeks to several years [43] . Since 1993 when albendazole was first shown to hasten the resolution of long persistent SCG [44] , several clinical trials have been conducted to evaluate the effects of albendazole and other treatment options . Based on a Bayesian network of 14 RCTs that included 1277 patients , the results of the present meta-analysis showed that the combination therapy of albendazole and corticosteroids for SCG reduced the risk of seizure recurrence by two thirds and tripled the odds of lesion resolution during a short follow-up period of around one year , compared with conservative treatment . Although the differences in the beneficial effects of the combination therapy of albendazole plus corticosteroids compared with either treatment alone did not reach statistical significance , the combined therapy consistently showed higher probabilities of being at superior ranking positions for both outcomes . The albendazole and corticosteroids monotherapies showed similar and significant efficacy in promoting lesion resolution , but their benefits failed to translate into better seizure outcome during the follow-up . The superiority of the combination therapy was robust and changed little in trials with different follow-up durations and in sensitivity analyses . Previous meta-analyses [9 , 10 , 13] reported that anthelmintic therapy with albendazole improved the seizure-free rate and hastened the resolution of the granuloma . However , these analyses combined clinical trials with different comparison groups ( anthelmintics versus conservative treatment , anthelmintics versus corticosteroids , combination of anthelmintics and corticosteroids versus conservative treatment , and combination of anthelmintics and corticosteroids versus corticosteroids ) making it impossible to determine the efficacy of anthelmintics itself and of the combination therapy . In fact , only three studies directly compared albendazole alone with placebo or no drugs , and the pooled estimates showed borderline significant improvement in lesion resolution and no difference in seizure outcome . Our network meta-analysis confirmed that albendazole alone did not improve the seizure-free rate although more lesions showed radiological resolution . The efficacy of corticosteroids alone in the treatment of SCG was evaluated in two pairwise meta-analyses with inconsistent results [13 , 14] . Both studies used the same set of trials , yet Otte et al . [13] in their meta-analysis incorrectly extracted the data from the trial by Kishore et al . [34] . They misused the data of the placebo arm for the prednisolone arm and that of prednisolone for the placebo , thus yielding pooled effects that fell just short of statistical significance . In our study , we also used the pairwise meta-analysis , which confirmed the benefits of corticosteroids monotherapy for both outcomes . These benefits remained significant in network meta-analysis , although the corticosteroids monotherapy tended to be inferior to the combination therapy and albendazole monotherapy in promoting lesion resolution and inferior to the combination therapy in preventing seizure recurrence . Based on RCTs and pairwise meta-analysis , an expert consensus on diagnostic and therapeutic schemes for SCG recommended a short course ( 1–2 weeks ) of albendazole with or without corticosteroids be prescribed soon after the first seizure [11] . Our study suggests that albendazole alone may not be effective on seizure control , and that the combination therapy of albendazole and corticosteroids should be initiated to bring the most benefit for patients with SCG . The observed effects of albendazole and corticosteroids are supported by their mechanisms of actions and the histopathology of the granulomatous lesion . The cysticercus granuloma consists of a dying parasite surrounded by fibrosis , angiogenesis and infiltration of inflammatory cells [45] . The parasite or its parts are still present , offering a target for the anthelmintics to act on . The attack on parasite accelerates its destruction and leads to a faster and more efficient lesion resolution , but at the same time hastens the release of parasitic antigens and exacerbates local inflammation [46] . The study by Robinson et al . demonstrated that substance P produced within cysticercosis granulomas is capable of inducing seizure activity [47] . The anti-inflammatory and immunosuppressive properties of corticosteroids seem to reduce the generation of the seizure-inducing mediators , limit the inflammatory damage to neural tissue and control perilesional edema . Corticosteroids also interact with albendazole by reducing the elimination rate of albendazole sulfoxide , the active component of albendazole , thus increasing its plasma concentrations [48] . The clinical synergism between albendazole and corticosteroids results in better seizure control as well as early resolution of the granulomatous lesion . However , because the analyzed clinical trials do not provide information on the timing of seizure recurrence in relation to drug administration , it is not clear whether the favorable seizure outcome achieved by the combination therapy is the result of a reduced likelihood of seizure activity during and shortly after the administration of albendazole and corticosteroids or due to a more sustained effect . Our study has several limitations . First , the majority of RCTs included in the analysis were at high risk of bias mainly because of inadequate allocation concealment and blinding . Only four studies were considered to have low-to-moderate risk of bias for the two outcomes , respectively , so that sensitivity analyses with only high quality studies were not possible . Second , all included RCTs were conducted in India . It is not certain whether the conclusions of this study apply to other populations . Third , under each class of treatment , there were variations in dosage and duration of the drugs used . This might have introduced some heterogeneity into the network meta-analysis . Treating them as different regimens , however , would not be feasible due to the insufficient number of studies to form a well connected network . The optimal match of dosages and durations of albendazole and corticosteroids needs further research . Fourth , the duration of the follow-up period varied among the included RCTs , making it another source of heterogeneity . Previous meta-analyses have tried some resolutions to the problem , such as performing separate meta-analyses at different time points of follow-up [13] or extracting data in the form of number of events per person-years observed [49] . In fact , estimating person-year of follow-up in these trials is very imprecise , and to our knowledge , there are currently no suitable methods that allow inclusion of all time points in a network meta-analysis . Here we explored the effects of differences in the duration of the follow-up period by meta-regression . The follow-up period was found not to significantly influence the results . Nevertheless , the average duration of the follow-up period of the included RCTs was generally short . Currently we cannot make firm conclusions on the effects of therapies more than one year after treatment . Future studies should focus on the efficacy of treatment in long-term seizure recurrence and granuloma resolution . Finally , limited data were available for two praziquantel-containing regimens to include in the analysis . In one trial [38] , 26 patients were assigned to receive single-day praziquantel therapy or no therapy . Complete resolution was found in 78% ( 11 out of 14 ) and 50% ( 6 out of 12 ) of patients , respectively . Another trial compared the combination of albendazole , praziquantel and prednisolone with the combination therapy of albendazole and prednisolone [33] . After 6-month follow-up , complete lesion resolution was observed in 72% ( 38 out of 53 ) of patients of the praziquantel-treated group , versus 52% ( 26 out of 50 ) of the control group . The differences were not statistically significant in both studies . Although a previous meta-analysis showed that praziquantel might be less effective than albendazole in the treatment of NCC [49] , the two anthelmintics have different mechanisms of action and have synergistic effects when used in combination [50] . More data are required before praziquantel can be added to the combination of albendazole and corticosteroids therapy for the treatment of SCG . Despite the above limitations , based on the comprehensive review and robust statistical method , our network meta-analysis provides a complete picture for the efficacy of different management options for patients with SCG . The combination of albendazole and corticosteroids performs better than other therapies in reducing seizure recurrence and promote lesion resolution during a follow-up period of around one year . Until more direct active comparisons are available , it should be recommended for the treatment of SCG .
|
Neurocysticercosis is an infection of the central nervous system by the larva of Taenia solium ( pork tapeworm ) . It is a leading cause of epilepsy in the world . The disease takes many different forms , each with different optimal treatment . In this study , we focused on the treatment of solitary cysticercus granuloma ( SCG ) , previous evidence on which is inconclusive . Since many different regimens have been compared in clinical trials of SCG , we conducted a network meta-analysis . This method is powerful as it can analyze quantitatively all of the data from all comparisons together . The result can tell us how different treatments perform against each other and how treatments should be ranked . The outcomes of our meta-analysis suggest that the combination of albendazole and corticosteroids is the most efficacious regimen to control seizures in affected patients and to promote the total disappearance of the lesion , compared with albendazole alone , corticosteroids alone , and conservative treatment .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"diagnostic",
"radiology",
"pathology",
"and",
"laboratory",
"medicine",
"granulomas",
"immunology",
"corticosteroid",
"therapy",
"database",
"searching",
"neuroscience",
"physiological",
"processes",
"mathematics",
"signs",
"and",
"symptoms",
"statistics",
"(mathematics)",
"pharmaceutics",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"imaging",
"techniques",
"animal",
"cells",
"mathematical",
"and",
"statistical",
"techniques",
"lesions",
"steroid",
"therapy",
"tomography",
"computed",
"axial",
"tomography",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"cell",
"biology",
"meta-analysis",
"physiology",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"drug",
"therapy",
"calcification",
"statistical",
"methods"
] |
2016
|
Albendazole and Corticosteroids for the Treatment of Solitary Cysticercus Granuloma: A Network Meta-analysis
|
The causative agent of gonorrhea , Neisseria gonorrhoeae , bears retractable filamentous appendages called type IV pili ( Tfp ) . Tfp are used by many pathogenic and nonpathogenic bacteria to carry out a number of vital functions , including DNA uptake , twitching motility ( crawling over surfaces ) , and attachment to host cells . In N . gonorrhoeae , Tfp binding to epithelial cells and the mechanical forces associated with this binding stimulate signaling cascades and gene expression that enhance infection . Retraction of a single Tfp filament generates forces of 50–100 piconewtons , but nothing is known , thus far , on the retraction force ability of multiple Tfp filaments , even though each bacterium expresses multiple Tfp and multiple bacteria interact during infection . We designed a micropillar assay system to measure Tfp retraction forces . This system consists of an array of force sensors made of elastic pillars that allow quantification of retraction forces from adherent N . gonorrhoeae bacteria . Electron microscopy and fluorescence microscopy were used in combination with this novel assay to assess the structures of Tfp . We show that Tfp can form bundles , which contain up to 8–10 Tfp filaments , that act as coordinated retractable units with forces up to 10 times greater than single filament retraction forces . Furthermore , single filament retraction forces are transient , whereas bundled filaments produce retraction forces that can be sustained . Alterations of noncovalent protein–protein interactions between Tfp can inhibit both bundle formation and high-amplitude retraction forces . Retraction forces build over time through the recruitment and bundling of multiple Tfp that pull cooperatively to generate forces in the nanonewton range . We propose that Tfp retraction can be synchronized through bundling , that Tfp bundle retraction can generate forces in the nanonewton range in vivo , and that such high forces could affect infection .
Type IV pili ( Tfp ) are bacterial appendages with important biological functions , including motility , DNA transformation , and virulence [1] . Among the most-studied Tfp are those elaborated by N . gonorrhoeae , which display Tfp over its entire surface . The N . gonorrhoeae Tfp is a helical polymer of pilin [2] subunits ∼6 nm wide and several micrometers long [3] . Tfp enable N . gonorrhoeae , as well as other bacteria , to attach to and pull on host cells or other substrates [4–6] . Cycles of Tfp extension , substrate binding , and retraction enable N . gonorrhoeae to crawl ( twitching motility ) [7 , 8] . Interest in the role of Tfp in infection was renewed by recent studies linking Tfp retraction force to the induction of epithelial cell responses [9 , 10] . The motor protein PilT [11 , 12] is essential for pilus retraction , and measurements using optical tweezers indicate that retraction of a single Tfp filament by a single PilT motor generates forces of 50–100 piconewtons ( pN ) [13] . These assays were limited to measuring retraction forces from individual diplococci . During infection of human tissue culture cells , N . gonorrhoeae aggregate into microcolonies via retracting Tfp [14 , 15] . Because the retraction forces exerted by the bacteria have been shown to play a key role in their interaction with host cells [9 , 10] , it is important to determine the range of forces that can be exerted by bacteria under infection conditions . Given the force limitation of optical tweezers , which cannot measure forces above 200 pN , and the potential of multiple Tfp contributing to retraction forces , we developed a new methodology to measure high retraction forces . Here , we demonstrate the ability of N . gonorrhoeae to impose tremendous forces ( nanonewton [nN] range ) on its substrate and allow us a more precise and comprehensive view of the physical role of Tfp in N . gonorrhoeae pathogenesis .
To measure high Tfp retraction forces from N . gonorrhoeae , we developed a new assay based on elastic micropillars by modifying previously published methods [16 , 17] . Our force sensor consists of an array of evenly spaced micropillars made of an elastic hydrogel ( Figure 1A and 1B , and Video S1 , to see a video of the device ) . A pulling force exerted on a pillar , such as that imposed by an attached pilus fiber , causes a displacement of the pillar tip . This displacement can be correlated to the force applied on the pillar after proper calibrations . Modifying the elastic properties of the gel , we were able to vary the stiffness of the pillars from 100 to 500 pN/μm . This allowed us to record forces in the pN to nanonewton ( nN ) range . Bacteria were seeded on these force sensors in Dulbecco's modified Eagle's medium ( DMEM ) tissue culture medium , allowed to interact with each other as in an infection assay , and the displacement of the pillars adjacent to the bacteria was recorded . The motion of the pillars revealed two types of pulling behavior from the bacteria . As reported previously [13] , we observed transient retraction events—or pulls—lasting anywhere from a few tenths of a second to several seconds , with a magnitude <100 pN ( 1 , trace β in Figure 1C ) . We also recorded much longer retraction events lasting from a few seconds to several hours , which exerted much higher forces of between 200 pN and 1 nN ( 2 , trace β in Figure 1C ) . Until now , retraction events of this magnitude have never been seen in N . gonorrhoeae . In all cases , only a few pillars were observed to be moving at a given time relative to stationary neighboring pillars ( trace α in Figure 1C ) , showing that these retraction events were specific to particular pillars . Tfp retraction forces under those conditions were up to 10 times higher than those recorded for a single Tfp filament . We next asked if there was a regular pattern associated with these high-amplitude forces . We used two methods for recording the forces exerted on a pillar ( see Figure S1 for details ) . We used either video rate ( 30 Hz ) imaging to capture the motion of a pillar's tip ( dynamic studies ) , or recording the deflection of the pillars by taking a picture at the base and at the tip of the pillar in fixed samples ( static studies ) . Dynamic studies allowed us to monitor accurately short-lived pulling events , whereas the static studies allowed us to view long-term pulling events . Histograms of both types of recordings revealed a low force peak that was characteristic of single Tfp retraction events ( 70 ± 20 pN for static studies , and 40 ± 20 pN for dynamic studies ) ( Figure 1D and 1F , respectively ) . The two types of force recordings differed in the percentage of higher force measurements as , for instance , the maximum force ( ∼1 nN ) events represent ∼1% of the measurements in the static studies and only 0 . 1% in the dynamical studies ( Figure 1D and 1F , respectively ) . In the force histograms from the fixed samples , we observed a number of peaks with values that were roughly multiples of a single filament retraction force ( Figure 1D ) . These peaks indicated that higher force retractions may involve the simultaneous pulling of 2 , 3 , 4 or more Tfp . A careful examination of rare dynamic events lends additional support to this possibility ( Figure S2 ) . In most cases , force increased in a step-wise fashion , commonly at increments of 70–100 pN . Why have such high forces ( nN ) not been recorded before ? One possible explanation is that previous assays were conducted in medium containing a high concentration of bovine serum albumin ( BSA; 1 mg/ml ) [7 , 13] . In many cases , BSA has the potential to prevent nonspecific interactions between proteins , and this could apply to proteinaceous assemblies such as the Tfp . To test the effect of BSA on Tfp retraction , we conducted our measurements in the presence of BSA . Under those conditions , we obtained very different histograms ( Figure 1E ) . We observed a primary peak in the histogram that was much higher than the primary peak recorded in the absence of BSA , and we recorded subsequent peaks that were much smaller than those measured in the absence of BSA . Thus , low-force retraction events were more numerous in the presence of BSA . Furthermore , the maximum forces measured in the presence of BSA never reached the levels observed in the medium without BSA either in static or dynamic measurements ( Figure 1F and 1G ) . Thus , the addition of BSA to the assay medium inhibits bacteria from pulling with high forces . Because of the possibility that Tfp could break as well as bundle [18–20] , we used thin section and scanning electron microscopy ( SEM ) to reveal differences in the Tfp structure of N . gonorrhoeae microcolonies incubated in DMEM with or without BSA . Single Tfp filaments were present on bacteria incubated with and without BSA , as judged by thin-section EM . However , in the absence of BSA , Tfp bundled along their long axis to form rope-like structures ( Figure 2A and 2C ) . Using SEM , bundles were seen emanating from single diplococci as well as from junctions between different cells . Single Tfp were not apparent in the SEM studies . Thus , Tfp bundling can be prevented by the addition of BSA to the culture medium . These differences in bundling phenotypes were also observed in microcolonies immunostained with a monoclonal antibody against Tfp ( Figure 2B ) . In untreated medium ( no BSA ) , microcolony Tfp staining gave an intense fluorescence at the interstices of the bacteria that was suggestive of bundles . In contrast , Tfp fluorescence in BSA-treated medium yielded microcolony Tfp staining that was diffuse ( Figure 2B ) . Interestingly , Tfp retraction was apparently necessary for bundle formation , since Tfp from PilT-null mutant microcolonies had a similar appearance to wild-type ( WT ) microcolonies assayed in BSA-containing medium ( Figure 2A–2C ) . We have thus established a direct correlation between Tfp bundling and the generation of high retraction forces . The occurrence of bundled Tfp have been reported previously in N . gonorrhoeae [20] and N . meningitidis [21] . To understand further the process of Tfp bundle formation , we promoted bundling by adding soluble polylysine ( 30 μg/ml ) to the incubation medium . Polylysine is a polymer that bears multiple positive charges that can be used to promote the binding between negatively charged entities . This treatment yielded thicker , ropelike Tfp structures ( Figure 2A ) . The microcolonies were smaller and the bacteria within them were less tightly packed , compared with untreated samples ( Biais and Sheetz , unpublished data ) . After 3 h of incubation , we noted an almost complete cessation of retraction events in the polylysine-treated samples ( Video S4 ) , compared with WT ( Video S2 ) and WT+BSA ( Video S3 ) samples . Thus , the larger Tfp aggregates that were formed in the presence of polylysine appeared unable to retract . Upon closer examination by thin-section microscopy , we observed that these Tfp structures consisted of multiple aggregates of 8–10 Tfp bundles ( Figure 3A–3C ) . The high density and ordered structure of these bundles enabled us to easily quantify the number of Tfp within a bundle . In bacteria incubated in DMEM alone , the Tfp bundles appeared less ordered ( Figures 2C and 3D ) , and we estimated that each bundle contained up to 8–10 Tfp . Thus , 8–10 Tfp filaments can bundle together to form high force–generating structures ( up to forces 8–10 times the force of a single filament retraction force ) . In this regard , it is interesting to note that N . meningitidis Tfp bundles , some containing 8–10 filaments , are hypothesized to promote bacteria–host cells interactions [21–23] . The pattern of force generation suggests that Tfp bundling occurred after an initial anchoring of a single Tfp filament to a surface . The fact that Tfp retraction forces generally increased over time implies that Tfp bundling increased over time as well . Short-term ( up to a minute ) , stepwise increases in the pulling forces ( Figure S2A , S2B , and S2D ) are indicative of the multiple filament nature of the bundles but do not capture the long-term ( ∼hours ) increase in force generation . Upon careful long-term observation , we detected build-up in pulling force at early attachment sites over extended time ( Figure 4A ) , suggesting that Tfp bundling occurred successively over that time . This did not occur at sites adjacent to these early attachment sites . The early , low-force retractions were transitory , and subsequent higher force retractions always occurred on those initial contact sites rather than on adjacent pillars in the few instances where we could measure the entire force history of a high-force pulling event . This shows that N . gonorrhoeae are able to build strong and lasting contacts on early adhesion sites . The sum of our observations supports the following model for the generation of high forces through the formation of Tfp bundles ( Figure 4B ) . When a Tfp filament undergoes a cycle of extension and retraction , there are two likely outcomes: either the retraction encounters resistance from the substrate or not . When a single Tfp adheres to a substrate and is not fully retracted , it remains extended . When a second Tfp associates with this extended Tfp ( the force of retraction possibly bringing them close to one another ) , the two can form a retractable doublet that can generate twice the force of a single retractable Tfp . If a third Tfp contacts the doublet , a working triplet is formed . This cooperative mechanism proceeds until a bundle of 8–10 Tfp is formed . Using a new micropillar-based assay to measure force , we found that N . gonorrhoeae Tfp filaments form higher-order structures that retract with forces up to 8–10 times higher than previously measured for a single Tfp filament . That a Tfp bundle retracts indicates that retraction of multiple Tfp filaments is coordinated . These studies raise important questions concerning the biochemical interactions among Tfp retraction motors that allow such cooperation [24] , as well as the factors that limit bundle size . Our study lends strong support to the notion that the Tfp retraction motor , PilT , is the strongest biological motor known to date . These very strong motors apparently use cooperation as well as relative irreversibility [25] to generate such forces . Those high forces generated by the Tfp can redefine the role of mechanical forces in bacteria–host interactions and perhaps also bacteria–bacteria interactions . The cooperative nature of the mechanisms at play in N . gonorrhoeae Tfp constitutes a new paradigm for the generation of high forces in the biological realm .
N . gonorrhoeae strains MS11 and MS11pilT [26] were used for all experiments and were maintained on GCB agar plus Kellogg's supplements at 37 °C and 5% CO2 and passed daily . Piliation and Opa phenotypes were monitored microscopically by assessing colony morphology . Only piliated , Opa–nonexpressing bacteria were used . The pillars were molded from microfabricated wafers containing evenly spaced holes [17] . To ease the demolding , the wafers were silanized with tridecafluoro-trichlorosilane in vapor phase and then plasma cleaned . A droplet of polyacrylamide mix was then sandwiched between an activated coverslip that retained the gel [16] and a piece of wafer bearing the microfabricated holes . After 15 min of curing , the gel was detached from the wafer in 50 mM HEPES . Polyacrylamide to bis-acrylamide concentrations of 20% acrylamide to 0 . 2% bis and of 20% acrylamide to 1% bis , with the wafers used for this study , resulted , respectively , in pillars with a stiffness constant of 100 ± 30 pN/μm and 500 ± 100 pN/μm , as measured with optical and magnetic tweezers . The analysis of the dynamic motion of the top of the pillars was performed by a home-written correlation plug-in for ImageJ ( NIH ) with a resolution of 0 . 3 pixel . The analysis of the static images was performed using the Manual Tracking plugin ( Fabrice Cordelières , Institut Curie ) with a precision of 1 pixel . The surface of the gel was coated with poly-L-lysine ( 30 μg/ml , Sigma ) using the bifunctional chemical compound sulfo-SANPAH ( Pierce ) to enable the Tfp to stick to the pillars . Bacteria were harvested from 14–16-h old agar plates and dispersed in GCB liquid medium at a concentration of 5 × 108 bacteria/ml . A 100-μl volume of that dispersion was added to a 35-mm well of a six-well plate containing a polylysine-coated coverslip or a coverslip with the pillars containing 2 ml of tissue culture medium ( DMEM , DMEM + 1mg/ml of BSA or DMEM + 30 μg/ml of poly-L-lysine ) . The samples were incubated for 3 h at 37 °C , 5% CO2 , then processed for electron microscopy , fluorescence microscopy , and/or static analysis of Tfp retraction . To analyze the dynamics of Tfp retraction , 1 to 2 μl of the bacteria dispersion was added to 260 μl of relevant medium and sealed between two coverslips spaced approximately 0 . 5 mm apart . Subsequently , movies were taken at video rate ( 30 Hz ) on an inverted microscope whose temperature was maintained at 37 °C . All experiments were performed at pH 7 . 4 . All optical microscope images were obtained on conventional inverted microscopes ( either Olympus IX 71 or Zeiss Axiovert 100 ) . Samples processed for fluorescent microscopy were fixed with 3 . 7% formaldehyde in phosphate-buffered saline ( PBS ) , blocked with a solution of 0 . 2% fish gelatin in PBS , and incubated with monoclonal antibody 10H5 . 1 . 1 that recognizes the conserved ( SMI ) domain of pilin [27] . All samples were stained with an Alexa-488–conjugated secondary anti-mouse antibody . Samples for SEM were fixed with 3 . 7% formaldehyde in PBS for 1 h , then dehydrated with successive baths in 50%–100% ethanol . They were critical point dried and coated with gold-palladium . Samples for thin-section EM were fixed with 2% glutaraldehyde in PBS for 1 h . They were sequentially exposed to osmium tetroxyde ( 1% ) , tannic acid ( 1% ) , uranyl acetate ( 1% ) , dehydrated with successive baths in 50%–100% ethanol , and finally embedded in an embedding resin . Areas of interest were then glued on a chuck and cut with a microtome .
|
Type IV pili are filamentous appendages borne by a large number of pathogenic and nonpathogenic bacteria . They play crucial roles in basic microbial processes such as surface motility , virulence , and DNA exchange . Neisseria gonorrhoeae , the causative agent of gonorrhea , can extend and retract these long , thin threads—around 6 nm in diameter and up to 30 μm long—to explore and pull on the environment . The retraction of one N . gonorrhoeae pilus filament can exert forces of 50–100 piconewtons , or roughly 10 , 000 times the bacterium's bodyweight . The bacteria can exert those forces on human cells that they infect , and force has been shown to be an important parameter in their infectivity . We use a micropillar assay system to show that N . gonorrhoeae cells can exert even higher forces by forming bundles of 8–10 filaments that act as coordinated retractable units . The bacteria can thus achieve forces in the nanonewton range ( or 100 , 000 times their bodyweight ) making them the strongest microscale elements known to date . This study demonstrates the power and cooperativity of pilus nanomotors and opens new territories for the exploration of force-mediated bacteria–host-cell interactions .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"biology",
"microbiology",
"biophysics"
] |
2008
|
Cooperative Retraction of Bundled Type IV Pili Enables Nanonewton Force Generation
|
We report a computational approach that integrates structural bioinformatics , molecular modelling and systems biology to construct a drug-target network on a structural proteome-wide scale . The approach has been applied to the genome of Mycobacterium tuberculosis ( M . tb ) , the causative agent of one of today's most widely spread infectious diseases . The resulting drug-target interaction network for all structurally characterized approved drugs bound to putative M . tb receptors , we refer to as the ‘TB-drugome’ . The TB-drugome reveals that approximately one-third of the drugs examined have the potential to be repositioned to treat tuberculosis and that many currently unexploited M . tb receptors may be chemically druggable and could serve as novel anti-tubercular targets . Furthermore , a detailed analysis of the TB-drugome has shed new light on the controversial issues surrounding drug-target networks [1]–[3] . Indeed , our results support the idea that drug-target networks are inherently modular , and further that any observed randomness is mainly caused by biased target coverage . The TB-drugome ( http://funsite . sdsc . edu/drugome/TB ) has the potential to be a valuable resource in the development of safe and efficient anti-tubercular drugs . More generally the methodology may be applied to other pathogens of interest with results improving as more of their structural proteomes are determined through the continued efforts of structural biology/genomics .
The construction and analysis of molecular interaction networks provides a powerful means to understand the complexity of biological systems and to reveal hidden relationships between drugs , genes , proteins , and diseases . In particular , the study of drug-target networks may facilitate an improved understanding of the principles of polypharmacology and hence improved rational drug design [2] . In recent years , several computational methodologies have been developed to predict drug-target networks based on ligand chemistry [4]–[6] , phenotypic changes resulting from drug perturbation [7]–[9] , or a combination of chemical features of drugs and sequence features of protein targets [10]–[12] . Extensive experimental and computational evaluation has proven that these methods are valuable for drug repurposing and side effect prediction . However , these methods are biased towards known drug-target pairs , which are mainly derived from well-established human target classes such as G-protein coupled receptors ( GPCRs ) , which only cover a small portion of the human proteome . The lack of a broad spectrum of drug-target pairs is more severe in pathogens than it is in human . For example , amongst the 3 , 999 proteins encoded by the Mycobacterium tuberculosis ( M . tb ) genome , only nine ( cmaA1 , cyp51 , embA , embB , embC , folK , InhA , katG and rpoC ) have been pharmaceutically investigated [13] . Thus , drug-target networks that are constructed from only existing drug targets are retrospective , and less capable of discovering new druggable targets and predicting off-target profiles of new compounds on a proteome-wide scale . In addition , the incompleteness of drug-target data poses questions as to whether or not the topology of drug-target networks is inherently modular or random [1] . It is important to construct and analyze a proteome-wide drug-target network that includes not only the primary targets , but also all of the potential off-targets of the drugs in the network . Such a network , if available , would provide unparalleled opportunities for mapping a comprehensive drug-target space and understanding the molecular basis of drug efficacy , side- effects and drug resistance , thereby providing the foundation for the rational design of polypharmacological ( multi-target ) drugs . For anti-infectious drug discovery , where pharmaceutically investigated targets only represent a small portion of the whole pathogen's proteome , it is more challenging to establish a proteome-wide drug-target network . The linkage of drugs to less exploited proteins such as virulence factors , transport proteins and transcription factors will greatly expand the repository of anti-infectious drug targets and provide new solutions for combating multi-drug and extensively drug resistant pathogens , and for repurposing existing drugs for new uses . Structural bioinformatics provides an alternative and complementary way to extend drug-target networks to less characterized proteins on a proteome-wide scale . The structural coverage of a given pathogen proteome is usually much larger than the pharmaceutical target coverage . In the case of the M . tb proteome , there are 284 unique proteins in the RCSB Protein Data Bank ( PDB ) [14] ( as of November 5 , 2009 ) , which is more than 30 times the number of existing pharmaceutical targets for M . tb . By taking reliable homology models into consideration , it is possible to increase the structural coverage of the M . tb proteome to approximately 43% . By taking advantage of this structural information , we have developed an integrated structural bioinformatics , molecular modelling and systems biology method to construct and analyze a drug-target interaction network , to discover novel druggable targets , and to propose new drug repositioning strategies . Our method is based on the comparison of the binding sites of existing drugs approved for human use against the entire structural proteome of the pathogen under investigation , in order to relate these drugs to new targets . For each identified drug-target pair , the atomic details of the interaction are studied using protein-ligand docking . If the protein is in a metabolic network model , the phenotype change resulting from the drug perturbation is further investigated using flux balance analysis ( FBA ) of the metabolic network . This strategy has been applied to study several selected drug targets , and proven , both computationally and experimentally , to be a useful tool in drug repositioning [15] , side effect prediction [16] , [17] , and polypharmacological target discovery [18] . In this paper , we extend this methodology to the construction of a proteome-wide drug-target network . Compared with existing methods that are either ligand or target centric , our method provides a framework to correlate the molecular basis of protein-ligand interactions to the systemic behavior of organisms . The proteome-wide and multi-scale view of target and drug space may shed new light on unsolved issues related to drug-target networks , and facilitate a systematic drug discovery process , which concurrently takes into account the disease mechanism and druggability of targets , the drug-likeness and ADMET properties of chemical compounds , and the genetic dispositions of individuals . Ultimately it may help to reduce the high attrition rate during drug discovery and development . The continuing emergence of M . tb strains resistant to all existing , affordable drug treatments means that the development of novel , effective and inexpensive drugs is an urgent priority . However , conventional drug discovery is a time-consuming and expensive process that is poorly equipped in the battle against tuberculosis . In this study , we apply our integrated approach in constructing the drug-target network of M . tb , which we refer to as the ‘TB-drugome’ . Using the TB-drugome we first attempt to characterize all drug-target interactions ( i . e . , the polypharmacological space ) of the M . tb proteome and to shed new light on controversial issues surrounding drug-target networks [1]–[3] . It has been argued that drug-target networks are similar to random networks , and that the observed modularity in drug-target networks may simply be the result of missing links between drugs and targets [1] . Our results support the idea that drug-target networks are inherently modular , and further that any observed randomness is mainly caused by biased target coverage . Then we introduce a new concept , the target chemical druggability index ( TCDI ) , which we use to determine the chemical druggability and prioritization of a protein as a drug target , and to characterize the potential of a drug as a polypharmacological lead compound . The TB-drugome reveals not only that many existing drugs show the potential to be repositioned to treat tuberculosis , but also that many currently unexploited M . tb proteins may be highly druggable and could therefore serve as novel anti-tubercular targets . The TB-drugome is publically available ( http://funsite . sdsc . edu/drugome/TB ) and has the potential to be a valuable resource for the development of safe and efficient anti-tubercular drugs . Structural biology and structural genomics efforts continue to increase the structural coverage of the M . tb proteome [19]–[21] , as well as those of other pathogens . This will improve the robustness of the TB-drugome and facilitate the application of this methodology to other pathogens . We hope that the application of the drugome concept will revitalize our way of thinking about how drug discovery is approached , something which is urgently needed if we wish to succeed in this on-going battle against multi-drug and extensively drug resistant infectious diseases .
A total of 274 different drugs approved for human use in the United States and Europe were identified in the RCSB Protein Data Bank ( PDB ) [14] . While the majority of these drugs were only co-crystallized with a single protein structure , many drugs were co-crystallized with more than one structure , bringing the total number of drug binding sites in the PDB to 962 ( see the Supporting Information , Table S1 ) . Many of these structures were derived from the same protein in different source organisms , and so the number of binding sites per drug is not a good indicator of drug promiscuity . In order to overcome this issue , the number of unique proteins co-crystallized with each drug was determined ( Figure 1 ) . While the vast majority of the drugs ( 194/274 ) had only been co-crystallized with a single protein , several had been co-crystallized with a number of different proteins , often from completely different folds . With a total of 11 , 9 , 8 and 7 different binding sites , the drugs niacinamide , acarbose , alitretinoin and indomethacin , respectively , were co-crystallized with the greatest number of different proteins . The distribution of the drug connections of co-crystallized proteins is close to a power-law distribution ( Supporting Information , Figure S1 ) . However , most of the proteins are only co-crystallized with a single drug . Only five proteins are co-crystallized with two drugs , and no proteins are co-crystallized with more than two drugs . It is not clear whether or not target connections in the PDB are scale-free . The TB-drugome , a structural proteome-wide drug-target network of M . tb , was constructed by associating the putative ligand binding sites of M . tb proteins with the known binding sites of approved drugs for which structural information about the target was available . The premise is that two entirely unrelated proteins can bind similar ligands if they share similar ligand binding sites . In this way , a M . tb protein can be connected to a drug through the drug's target , irrespective of whether that protein target is from human or another organism . The binding site comparison software SMAP [22]–[24] , was used for this purpose in an all-drug-against-all-target manner ( see the Methods section ) . In a previous benchmark study , SMAP outperformed most of the existing ligand binding site comparison algorithms [22] , [24] . Moreover , several predictions from SMAP have been experimentally validated [15] , [18] , [25] . Thus SMAP has proven a useful tool to identify the off-targets of existing drugs . The resulting TB-drugome network is shown in Figure 2 and consists of M . tb proteins ( blue circles ) connected to drugs ( red circles ) , where a single connection indicates binding site similarity between any of the structures of the connected M . tb protein , and any of the binding sites of the connected drug . The TB-drugome is highly connected , indicating that many binding site similarities were observed between M . tb proteins and drug targets , even though those proteins had different overall structures . The number of edges in the TB-drugome network depends on the confidence level of the prediction . To determine the SMAP P-value threshold that gives a balanced false positive and negative rate in the TB-drugome , the average connectivity of the drugs was plotted against the SMAP P-value . A turning point in the curve exists for a SMAP P-value of 1 . 0e-5 ( Figure 3 ) , i . e . , the connectivity of the drugs changes only slightly with a SMAP P-value of less than 1 . 0e-5 , but rapidly increases when the P-value is greater than 1 . 0e-5 . The use of a SMAP P-value threshold greater than 1 . 0e-5 will therefore reduce the false negative rate , but dramatically increase the false positive rate when detecting similar ligand binding sites . Thus , a SMAP P-value of 1 . 0e-5 was selected as a threshold for network construction , and was used throughout this study . Based on the previous SMAP benchmark study [22] , [24] , the false positive rate is approximately 5% when the SMAP P-value is close to 1 . 0e-5 . Thus , it is estimated that the false positive rate of connections is approximately 5% in the TB-drugome . In the TB-drugome , 123 of the 274 drugs are connected to 447 of the 1 , 730 proteins ( 284 PDB structures plus 1 , 446 homology models ) . Thus , it is estimated that around 40% of these 274 approved drugs , or their associated compound libraries , may be active against around 25% of the M . tb structural proteome , greatly expanding the existing anti-tubercular drug-target space . Unlike other drug-target networks [1]–[3] , the TB-drugome is not biased towards certain gene families . The largest family in the TB-drugome is cytochrome P450 , which consists of 20 proteins , comprising approximately 4 . 5% of the connected proteins and 1% of all proteins in the TB-drugome , respectively . The average degree of drug connectivity in the TB-drugome is 12 . 1 , which is almost twice the predicted value of 6 . 3 for drug-target networks [1] . Despite the high degree of drug connectivity , the modularity of the network is maintained , as shown in the next section . Reliable and unbiased drug-target networks may shed new light on the controversial issues surrounding the underlying topological structure of drug-target networks . It has been argued that drug-target networks are similar to random networks , and that the observed modularity in drug-target networks may simply be the result of missing links between drugs and targets [1] . Topological analysis of the TB-drugome provides evidence for the modularity of drug-target networks . Although the average connectivity of drugs is much higher ( Figure 3 ) than that predicted for a drug-target network in which the targets are pharmaceutically annotated [1] , the distribution of target connectivities follows a power-law distribution regardless of P-value threshold ( Figure 4A and Table 1 ) . That is , most targets have few connections , but a small number of targets are connected to a large number of drugs . This is also true for drug connectivity ( Supporting Information , Figure S2 and Table S6 ) . This observation strongly suggests that proteome-scale drug-target networks are not random . This scale-free property is not sensitive to the systematic noise introduced by the increased number of false positive edges that result from an increase in the P-value threshold , indicating that the connections between proteins and drugs are not completely random . The connections reflect the underlying evolutionary , geometric and physicochemical relationships between the M . tb proteins and the drug targets . In contrast , if the edges in the network were random , this scale-free property would break down ( Figure 4B and Table 1 ) . Similarly , the false negative rate also has little effect on the topology of the network since the power-law distribution remains consistent when the number of false negatives is increased as a result of decreasing the P-value threshold . Besides being scale-free , the TB-drugome network is modular , as measured by the clustering coefficient . As shown in Table 2 , the clustering coefficients of both the targets and the drugs are almost twice those of the corresponding random networks . Moreover , since there is no significant change in the clustering coefficient when using different SMAP P-value thresholds to define the network connectivity , this implies that an underlying architecture exists in the TB-drugome . The modularity of the TB-drugome is also measured by the largest connected component ( nLCC ) . The nLCC values for M . tb targets and drugs are 0 . 93 and 0 . 84 , respectively , compared to nLCC values of 0 . 97 and 1 . 0 , respectively , for a random network ( Supporting Information , Figure S3 ) . This modularity becomes more obvious for high confidence networks that are derived from P-value thresholds of 1 . 0e-6 and 1 . 0e-7 . Since the 274 structurally characterized drugs only comprise around 20% of all approved drugs , it is interesting to investigate the effects of increasing drug structural coverage on the properties of the network . To address this question , we randomly selected a subset of the 274 structurally characterized drugs to see how the structural coverage of drug-target complexes affects the power-law distribution and the clustering coefficient of the network . Even when the number of drug-target complexes drops to 20% of the total number present in the PDB , there are no significant changes in the network properties of the TB-drugome ( Supporting Information , Figure S4 , Table S7 and S8 ) . Thus , it is expected that the scale-free properties and modularity observed in the TB-drugome will not be affected by an increase in the number of drug-target complex structures . One factor that may contribute to the randomness of existing ligand-based drug-target networks is target bias towards several gene families , for instance , G-protein coupled receptors . Proteins in the same gene family tend to cluster together; therefore , if such gene families dominate a network , then a large nLCC value is easily obtained and the power law distribution breaks down . It is to be expected that the topological properties of drug-target networks will change once extended to include the entire proteome . The current incompleteness of drug-target networks is not only due to the missing links between drugs and targets , but also due to the biased and limited coverage of target space . However , as this coverage improves we anticipate that power-law behaviour will be preserved . To our knowledge , there are currently only nine M . tb proteins that have been validated as drug targets; cmaA1 ( Rv3392c ) , cyp51 ( Rv0764c ) , embA ( Rv3794 ) , embB ( Rv3795 ) , embC ( Rv3793 ) , folK ( Rv3606c ) , InhA ( Rv1484 ) , katG ( Rv1908c ) and rpoC ( Rv0668 ) [13] . According to the TB-drugome there are numerous other drug targets yet to be exploited . An important question in drug discovery is whether or not a biologically validated target is able to bind drug-like molecules with high affinity , i . e . , whether or not the target is chemically druggable . Although chemical druggability can be predicted from the ligand binding site of a protein [26] , there is still a big gap between identifying lead compounds and developing safe drugs . Analysis of the TB-drugome not only provides molecular insights into chemical druggability , but also suggests existing drugs that could either be directly repurposed or act as lead compounds . Here we introduce a new Target Chemical Druggability Index ( TCDI ) , which is orthogonal to biological essentiality , and directly links target and drug space . After the drug target has been biologically validated as an essential gene , the TCDI may be applied to determine if it is a suitable candidate for medicinal chemistry efforts . The TCDI is determined by the number of unique drugs ( those with a 2D Tanimoto coefficient to one another of less than 0 . 75 ) that are connected to a protein in the TB-drugome . In this way , it is possible to prioritize the chemically druggable targets on a proteome-wide scale . In the TB-drugome , there are 165 proteins with a TCDI of greater than 2 . Those proteins with a TCDI of greater than 8 are listed in Table 3 . Since most of these proteins have not been pharmaceutically investigated , their propensity to bind drug-like molecules should be determined experimentally . Although gene essentiality is not necessarily correlated with chemical druggability , it is interesting to investigate whether or not those proteins with a large TCDI are crucial for bacterial survival or virulence . If a gene is both essential and chemically druggable , it will be an ideal target for drug development . The biological roles of these proteins were determined primarily from the literature . Since several of the proteins listed in Table 3 are involved in metabolism , it is possible to investigate the effects of their knockout by carrying out flux balance analysis ( FBA ) using a proteome-wide network model of M . tb metabolism . The GSMN-TB model [27] was selected to simulate in vivo conditions , while the iNJ661 model [28] was selected to simulate in vitro conditions . Most of the proteins in Table 3 with known functions are essential for bacterial survival , as predicted by metabolic simulation , or validated by experiments . The top ranked protein , Rv3676 , encodes the cAMP receptor protein/fumarate and nitrate reductase ( CRP/FNR ) transcriptional regulator . Members of the CRP/FNR class of transcriptional regulators respond to environmental conditions associated with low oxygen stress and starvation , and may play an important role in reactivating dormant bacilli . The importance of the M . tb CRP/FNR transcriptional regulator has been demonstrated through knockout studies . Indeed , deletion of this gene is known to cause growth defects in laboratory medium , in bone marrow derived macrophages and in a mouse model of tuberculosis [29] . 22 unique drugs are predicted to be potential lead compounds targeting CRP/FNR . As shown in Figure 5 , besides the known cAMP binding site ( site A ) , a second binding site ( site B ) is identified in the DNA binding domain . This finding provides opportunities to design drug conjugates or combination therapies to inhibit this protein . The M . tb protein with the second highest TCDI is InhA ( enoyl-acyl carrier protein reductase ) , which is actually the target of the front-line anti-tubercular agent isoniazid [30] . As a pro-drug , the therapeutic effect of isoniazid depends on its conjugation with the NAD co-factor . The development of isoniazid-resistant M . tb strains has promoted the discovery of a number of direct inhibitors of InhA [31] . Most of the predicted drug binding sites are located in the substrate binding site of InhA , and therefore serve as potential leads for direct InhA inhibitors . Indeed , the prediction that InhA can be directly inhibited by an existing drug has already been experimentally validated . Both an in vitro bacterial growth study and an enzyme kinetic assay supported our previous in silico prediction that Comtan , a drug used in the treatment of Parkinson's disease , could potentially be repurposed to target InhA directly [15] . Thus the prediction that InhA is a highly druggable target is supported by existing experimental data , in addition to common clinical practice . There are a number of M . tb proteins that , although not predicted to be essential , may play important roles in the host-pathogen interaction . The protein with the third highest TCDI is Rv1264 , a class III adenylyl cyclase which synthesizes cAMP from ATP in response to sensing the mildly acidic pH of the host macrophage phagosome . Biochemical studies of Rv1264 have suggested that it may be crucial for the M . tb host-pathogen interaction , thereby highlighting it as another potentially interesting drug target [32] . The predicted drug binding site is located in the dimerization interface of the regulatory domain ( Supporting Information , Figure S5 ) . Since dimerization is critical for the function of adenylyl cyclase , it is speculated that the inhibition of its association may disrupt its function [33] . Other proteins that are involved in the host-pathogen interaction include Rv2413c [34] , narL [35] , [36] , and lprG [37] . A new strategy emerging to combat drug resistant pathogens is to target the pathways involved in host-pathogen interactions [38] . The identification of druggable targets that contribute towards pathogenicity ( e . g . , proteins involved in two-component regulatory systems [39] , [40] ) and the host-pathogen interface may present new opportunities for the discovery of novel therapeutics effective against tuberculosis . Several other non-essential genes may contribute to drug resistance mechanisms exhibited by M . tb . For example , Rv1272c is an efflux pump that detoxifies antibiotics . Combination therapy using antibiotics mixed with efflux pump inhibitors could therefore be a practical solution for increasing the efficacy of antibiotics [41] . In addition , the TB-drugome may provide clues about the biological roles of proteins with unknown functions . Since Rv0856 is predicted to bind to antibiotics such as minocycline and rifampin , it is possible that this protein is involved in the detoxification of these antibiotics . The TB-drugome reveals that , of the 274 different drugs investigated , 92 drugs could potentially inhibit more than one M . tb protein . This is advantageous both in terms of drug effectiveness and preventing the development of drug resistance . Indeed , large-scale functional genomics studies in model organisms have shown that the vast majority of single-gene knockouts actually exhibit little or no effect on phenotype [42] . The concept of ‘synthetic lethality’ - genes that are not essential individually , but are essential in combination - uncovers a whole new plethora of drug targets that may have been overlooked due to their non-essentiality in individual gene knockout studies . Synthetic lethality explains the success of several multi-target anti-infectives that have been discovered serendipitously over the years , including D-cycloserine , beta-lactam antibiotics , fosfomycin and fluoroquinolone antibiotics [43] . Furthermore , inhibition of two or more proteins that are essential individually is advantageous from a drug resistance perspective . Indeed , while pathogens are able to rapidly acquire resistance to single target agents through mutations in the target protein , it is much more difficult to acquire resistance to multiple target agents , since a mutation in one of the essential target proteins would not confer any selective advantage over the wildtype [44] . Some drugs in the TB-drugome have the potential to inhibit a large number of different M . tb proteins simultaneously . It is important to note that there are two types of connections in the TB-drugome; those that involve proteins belonging to the same fold , and those that involve proteins belonging to different folds . The detection of functional relationships between proteins belonging to the same fold is considered to be a trivial task because it can be achieved by simply using conventional sequence and structure comparison tools . It is more interesting and novel to relate proteins across fold space , i . e . , when the primary drug target and its off-target ( s ) do not share similar global structures . Such cross-fold connections constitute around 60% of all connections in the TB-drugome ( see Tables S2 and S3 in the Supporting Information for a full list of cross-fold connections in the TB-drugome ) . The 15 most highly cross-fold connected drugs are listed in Table 4 , along with the names of the solved M . tb proteins to which they are connected . With 98 cross-fold connections , alitretinoin , a drug used to treat cutaneous lesions in patients with Kaposi's sarcoma , is the most highly connected drug . The solved M . tb proteins to which it is connected include bioD , InhA and purN , all of which are predicted to be essential in vivo by a metabolic network reconstruction of M . tb [27] . With 63 different cross-fold connections , levothyroxine , a drug used to treat hypothyroidism , is the second most highly connected drug . Further investigation revealed that it was the structure of levothyroxine bound in the binding site of serum albumin that was determined to be significantly similar to many of the 63 different M . tb binding sites . This is interesting because , as a non-specific binder of steroid hormones and a transport protein for various fatty acids , serum albumin is known to be a highly promiscuous protein [45] . While it is not necessarily a useful result for the purposes of this study , the fact that SMAP is able to detect similarities between the binding site of serum albumin and the binding sites of multiple other proteins at least provides some validation that it is working correctly . Serum albumin also accounts for all 24 connections between the drug propofol and various different M . tb proteins . Note that although serum albumin is also listed as an intended target of methotrexate , this drug has not actually been cocrystallized with serum albumin in the PDB , and so this does not account for its high connectivity . The front-line anti-tubercular agent rifampin is listed as the fifth most highly connected drug in Table 4 . The structure of its known M . tb target , DNA-directed RNA polymerase ( rpoC ) has not been solved , therefore explaining why it is not listed as a potential target in Table 4 . However , a suitable homology model of rpoC was identified in ModBase , based on RNA polymerase from the eubacterium Thermus thermophilus . The fact that rifampin has connections with six other solved M . tb proteins in Table 4 suggests that it may be mediating some of its anti-tubercular effects through proteins other than its known target , rpoC . A recent study showed that rifampin is able to bind to the NAD binding site of ADP-ribosyl transferase [46] , which is ranked highly at 24/962 with a SMAP P-value of 4 . 32e-4 . Rifampin is predicted to bind to the NAD and FAD binding sites of InhA and lpdA , respectively . Both of these predictions are supported by the compound association listed in the TDR target database [47] . Since the off-targets of rifampin may be involved in drug metabolism and detoxification , the proteome-wide identification of off-targets may provide molecular insight into the understanding of drug resistance mechanisms . A literature search of the M . tb proteins listed in Table 4 reveals that most of them are potentially novel targets for the development of anti-tubercular therapeutics . For instance , aroF ( chorismate synthase ) , aroG ( chorismate mutase ) and aroK ( shikimate kinase ) are attractive targets because they are all involved in the shikimate pathway , which is both essential for the viability of M . tb , and absent from humans [48] . LppX is a lipoprotein required for the translocation of complex lipids to the outer membrane , and disruption of the lppX gene has been shown to result in attenuation of virulence of the tubercle bacillus [49] . Another protein that is essential for the pathogenesis and virulence of M . tb is the sigma factor sigC , which controls the environment dependent regulation of transcription [50] . A potential target against M . tb persistence is the universal stress protein , TB31 . 7 , which is required for the entry of the tubercle bacillus into the chronic phase of infection in the host [51] . These are merely a few examples of the many potentially interesting M . tb targets listed in Table 4 . Furthermore , there are likely to be many more attractive targets in the form of homology models , which have not been investigated here . Since many of the genes encoding the M . tb proteins listed in Table 4 are involved in metabolism , it is possible to investigate the effects of their knockout using a proteome-scale network model of M . tb metabolism . The GSMN-TB model [27] was selected for this purpose due to its ability to simulate in vivo conditions . Those genes that were present in the GSMN-TB model , and whose knockout could therefore be simulated , are underlined in Table 4 . Those genes whose knockout resulted in a maximal theoretical growth rate of zero or close to zero were considered essential and have been highlighted in bold . All of the drugs in Table 4 , with the exception of amantadine and lopinavir , are predicted to potentially inhibit one or more essential metabolic proteins with solved structures . In particular , the anti-HIV therapeutic ritonavir could potentially inhibit a total of five different essential proteins involved in metabolism; accD5 ( propionyl-CoA carboxylase ) , aroK ( shikimate kinase ) , fabH ( 3-oxoacyl- ( acyl carrier protein ) synthase III ) , panC ( pantoate—beta-alanine ligase ) and serA1 ( D-3-phosphoglycerate dehydrogenase ) . Amantadine has connections to homology models only and so was excluded from this study . Although lopinavir may not inhibit any essential metabolic proteins , some of the proteins that it could potentially inhibit may be interesting anti-tubercular targets . For instance , pknG , a eukaryotic-type protein kinase , has been shown to support the survival of mycobacteria in host cells [52] , and is required for the intrinsic resistance of mycobacterial species to multiple antibiotics [53] . In addition , the GSMN-TB model was used to simulate multiple gene knockouts and therefore the effect of a single drug inhibiting multiple metabolic proteins simultaneously . For each drug ( excluding amantadine and lopinavir ) , the combined knockout of all metabolic genes listed in Table 4 resulted in zero or close to zero biomass ( except for the case of levothyroxine , where combined inhibition of bioD ( essential ) and thyX ( non-essential ) resulted in growth ) . More studies are required to verify this prediction . If all members of a set of proteins can bind to the same set of multiple drugs , this set of proteins could provide interesting targets for polypharmacological drugs . Such polypharmacological drug targets can be derived from the TB-drugome . Indeed , several multi-drug-multi-target clusters are distinguishable within the drug-target matrix shown in Figure 6 . The three largest clusters are the cytochrome P450s ( CYP ) , protein kinases ( PKN ) , and polyrenyl-diphosphate/polyrenyl synthases ( GRC ) . As promiscuous metabolizing enzymes , the cytochrome P450s bind to multiple drugs , while the protein kinases and polyrenyl-diphosphate/polyrenyl synthases bind human protein kinase inhibitors and farnesyl-diphosphate synthase inhibitors , respectively . Although this result is not surprising , the fact that similar drugs and similar targets are clustered together provides further validation of the TB-drugome . An interesting cluster is ilvG ( acetolactate synthase ) , asd ( aspartate-semialdehyde dehydrogenase ) , fadE13 ( acyl-CoA dehydrogenase ) and Rv0037c ( MFS-type transporter ) , all of which are predicted to bind to HIV-1 protease inhibitors . There is a major problem with coincidence of HIV and tuberculosis in sub-Saharan Africa . Indeed , HIV and tuberculosis form a deadly combination , each accelerating the other's progress . Since HIV weakens the immune system , HIV-positive individuals are much more susceptible to developing an active form of tuberculosis and becoming infectious [54] . Co-administration of existing anti-TB and anti-HIV therapeutics is undesirable due to adverse side-effects . Therefore , the finding that an anti-HIV therapeutic can actually be used to treat both HIV and TB simultaneously would be of great interest . It is also worth noting that five of the top 15 most highly connected drugs in the TB-drugome , which are listed in Table 4 , are also HIV-1 protease inhibitors .
All existing drug-target networks have been constructed from annotated drug-target pairs or predicted based on the chemical properties of the ligands from these drug-target pairs . As a result they only include the limited number of human drug targets that have been pharmaceutically investigated , i . e . , a small , highly biased subset of the human proteome . The lack of a broad spectrum of drug-target pairs is more severe in pathogens than it is in human . For example , among the 3 , 999 proteins encoded in the M . tb genome , only nine proteins ( cmaA1 , cyp51 , embA , embB , embC , folK , InhA , katG and rpoC ) have been pharmaceutically investigated [13] . Conventional methods can only build a drug-target network based on these nine proteins and their associated ligands . Thus , they cannot generate a comprehensive drug-target network like the TB-drugome . The chemical systems biology strategy applied in this paper provides a complementary approach to constructing a structural proteome-wide drug-target network . To our knowledge , the TB-drugome is the first drug-target network that covers this many proteins in the TB structural proteome and all drugs that have been structurally characterized . The TB-drugome includes 50 times more proteins than the existing TB targets , and more than 100 drugs that have not been investigated for tuberculosis treatment . Compared with existing methods that require information about drug-target pairs , one of the unique features of the TB-drugome is that the relationship between two proteins can be established by their ligand binding site similarity , independent of their associated ligands . This feature not only greatly extends target coverage to those proteins with unknown or less characterized ligands , but also includes drugs that may not necessarily be used to target TB proteins directly . Thus , the resulting network is more complete and less biased . Since the TB-drugome includes a large number of poorly characterized or uncharacterized proteins , it may provide greater insight into the progressive drug discovery process than existing drug-target networks . Indeed , it may aid the discovery of novel druggable targets that have not been explored previously , guide medicinal chemists to design compounds with desirable specificity to avoid unwanted side effects , and promote the rational design of polypharmacological drugs by selecting multiple suitable targets . Coincident with recent efforts involving screening compound libraries of existing human drug targets to treat anti-infectious diseases [25] , [55] , [56] , our method will be particularly useful in genome-wide compound profiling , lead generation from existing drug-like molecules , and identifying the molecular targets of active compounds . It is not feasible to achieve such goals using existing drug-target networks in cases where the actual molecular targets or their ligands are unknown . Notwithstanding , there are major limitations in the methodologies applied in this study . Firstly , the structural coverage of the M . tb proteome is limited . Currently only 7 . 2% of M . tb proteins have solved structures in the PDB . The use of reliable homology models increases the structural coverage to around 43% . However , each homology model consists of only a single chain rather than the entire biological unit , which could be a multi-polypeptide chain complex . As a result interesting binding sites located in the interface between the chains may be missed . Similarly , only around 20% of all drugs approved for human use have actually been solved with a protein target structure in the PDB . Coverage of drug space can be increased by using crystal structures or homology models of proteins that are known targets of approved drugs , but for which there are no structures with the drugs bound . For instance , the additional inclusion of homology models of GPCRs would double the number of targets . Two proteins may bind to similar ligands even though their binding pockets may have varied geometrical and physicochemical properties . Such proteins may be sequence homologues , have similar structures , or belong to entirely different folds . For the first two scenarios , SMAP is more sensitive than conventional sequence and structural comparison methods in detecting ligand binding site similarity [24] . For the third scenario , SMAP takes into account residue mutations and geometrical variances within the binding site , therefore making it a sensitive algorithm for ligand binding site similarity searches . However , a fraction of true positives may still be missed in all three scenarios . Despite the existence of false negatives in the drug-target network , the TB-drugome has generated abundant testable hypotheses . From the point of view of real-life applications , it may be more important to reduce the false positive rate than to reduce the false negative rate . The further construction of reliable proteome-wide drug-target networks will benefit from the integration of diverse techniques such as ligand-centric methods [4] , [57] and omics data such as gene expression profiles in response to drugs [9] . The integration of multiple data resources will not only increase the coverage of the network , but also the confidence of the predictions made , through the use of consensus results . Aside from the false negatives that result from the limited structural coverage of the M . tb proteome and the completely different ligand binding poses , ligand binding site similarity is necessary but not sufficient to determine the cross-reactivity between two proteins for a specific ligand . The chemical nature of the ligand also determines off-target binding . Although off-target predictions based on similar ligand binding sites are invaluable for the progressive design of selective or polypharmacological drugs , they may result in false positive connections between proteins and existing drugs . Thus , the TCDI may be not correlated with the docking scores . The direct assessment of protein-drug interactions using protein-ligand docking may solve part of this problem , but success is not guaranteed due to the inaccuracy of docking scoring functions . While free energy calculations based on molecular dynamics simulations may improve the prediction of protein-ligand interactions , they are computationally intensive and currently impractical on a proteome-wide scale . It remains a significant challenge to develop new methodologies for accurate and efficient protein-ligand docking and free energy calculation for the prediction of drug off-targets on a proteome-wide scale . It has been argued that drug-target networks are not modular but random [1] . Drug-target networks constructed by linking all drug-target pairs from annotated chemical libraries or computationally predicted results are limited and biased . Mestres et al . discovered that the topologies of drug-target networks are implicitly dependent on drug properties and target families [58] . Consequently , given the biased coverage of target families , the topological properties observed in drug-target networks may not necessarily reflect the inherent properties observed in proteome-wide protein-ligand interaction networks . Here we suggest that modularity does exist in our structural proteome-wide drug-target network , and that it follows a power-law distribution . Any observed randomness appears to result from the biased coverage of drug targets . The power-law distribution has been observed in many biological networks including protein-protein interaction networks [59] and metabolic networks [60] , [61] . Recently , it has been found that interaction networks between proteins and their endogenous ligands follow a power-law distribution [62] . Such connectivity distributions also appear in other man-made networks , such as the World Wide Web and social networks . The preferential attachment principle [63] , [64] , which has been tested in social networks , can be applied to biological networks [62] , [65] according to the evolutionary history of ligands and proteins . These studies have shown that evolutionarily ancient ligands and proteins tend to have more connections . It follows that the local structures of the binding site and the core fragments of the ligand are more conserved than global structures and sequences [66] . In the case of protein-ligand interaction networks , the structural basis behind their power-law distribution and scale-free nature could be the modularity of protein-ligand binding sites , the modular arrangement of chemical fragments [67] , [68] , and the flexibility of both ligand [69] and protein structures [70] . By studying the characterizing descriptors for ligands and small molecules , Ji et al . found that polar molecular surface area , H-bond donor counts , H-bond acceptor counts and partition coefficients are key factors that can be used to discriminate hub ligands from others [62] . There are two aspects of target druggability; biological and chemical . From a biological point of view , druggability is conventionally based upon multiple criteria such as gene essentiality , conservation across kingdoms , protein-protein interactions , redundancy among pathways , endogenous metabolite distributions , and coupling between metabolic , regulatory and signalling pathways . However , a biologically druggable essential gene is not necessarily chemically druggable because it may be difficult to design a drug-like molecule to bind it with high affinity and specificity . Thus , biologically validated drug targets need to be linked to their chemical space as early as possible in order to determine their chemical druggability . Although chemical druggability can be predicted from the ligand binding site of a protein [26] , there is still a big gap between identifying lead compounds and developing safe drugs . The Target Chemical Druggability Index ( TCDI ) proposed here is intended to bridge the target validation process and medicinal chemistry efforts to select targets that are both essential ( as determined from other resources or methodologies ) and appropriate for use in the design of drug-like molecules . If the functional site of a single protein is connected to , and could therefore potentially be inhibited by , one or more approved drugs , this is a strong indication that this protein may be chemically druggable . Moreover , if a protein has a high TCDI , this implies that any new ligand found will likely occupy the chemically constrained space of approved drugs , as opposed to the essentially unlimited chemical space , and this could benefit drug discovery in many ways . Firstly , it could narrow down the infinite chemical space needed for high-throughput screening to identify lead compounds . Secondly , it provides information about the ligand binding site , which is critical for rational drug design . Thirdly , it may reduce medicinal chemistry efforts to optimize the lead compound as a drug candidate . Finally , and perhaps most critical in this new era of drug discovery , it offers more opportunities to design polypharmacological drugs , which may not only improve drug efficacy and combat drug resistance , but also minimize human side effects . By taking gene essentiality data , chemical druggability information , ligand binding site information , and the ligand coverage of drug space into account simultaneously , the significant time and costs associated with anti-infectious drug discovery and development could be significantly reduced . A search of the TDR target database [47] reveals that there are no chemical compounds associated with any M . tb proteins with a high TCDI other than InhA . Thus , the TB-drugome provides abundant testable hypotheses for the development of new anti-tubercular therapeutics . It is expected that the discovery of a drug candidate by the targeted screening of these drugs will require a fraction of the time and costs associated with conventional high-throughput screening . Even if a drug shows weak activity in an initial assay , the assay can be extended to include the large number of analogues of that drug that have already been synthesized and tested . In this way , it may be possible to discover a potent compound that weakly inhibits the primary drug target , but strongly binds to the M . tb target . Such a strategy has been successfully applied to repurpose a library of protein kinase inhibitors to target bacterial biotin carboxylase [25] . Conventional drug discovery and development proceeds as a linear process from target identification and validation , to lead discovery and optimization , to preclinical and clinical trials . It is estimated that more than 90% of drug candidates fail during the late stages of drug development , mainly due to poor efficacy or safety [71] . If information were available about disease mechanisms , target druggability , the chemical space of the target , the pharmacokinetics and dynamic properties of drug candidates , and their potential off-targets that may result in unwanted side effects ( or sometimes a desirable therapeutic effect ) , then their consideration in drug development would help to optimize resource allocation and improve productivity in the pharmaceutical industry [72] . Proteome-wide multi-scale drug-target interaction networks help here by providing a resource to unify disease , target , and chemical space , thereby allowing the simultaneous assessment of target essentiality , target druggability , drug design feasibility , chemical availability , compound toxicity , and individual drug response . In the context of anti-infectious drug discovery , network analysis can be used to identify critical nodes in molecular networks which could represent novel drug targets [28] as illustrated here . Moreover , it is believed that druggability and essentiality are best assessed at the binding site level rather than the global sequence or structural level [43] . Thus , the integration of ligand binding site characterization with systems biology is critical for target identification and prioritization . Even if druggability can be assessed by analyzing the ligand binding site of the target , there is still a huge gap between identifying hit compounds and producing drug candidates . Moreover , the drug candidate may not be safe for human use due to undesirable ADME properties or unwanted off-target effects . By bridging target and drug space , drug-target interaction networks provide invaluable information about the use of existing drugs as lead compounds . In an ideal situation , the drug can be repositioned directly to target the intended target in the pathogen , hence promising a solution to reduce both the time and costs associated with drug development [73] . Since the drugs have already been approved for human use , it is possible to bypass toxicological and pharmacokinetic assessments , which together contribute approximately 40% of the overall cost of bringing a new drug to market . Newly identified drug indications can be evaluated relatively quickly in phase II clinical trials , which typically only take two years and cost around $17 million [74] , [75] . The continuing emergence of M . tb strains that are resistant to all existing affordable drug treatments means that the development of novel , effective and inexpensive drugs is an urgent priority [15] , [76] . However , current drug discovery methods appear inadequate in the battle against infectious diseases such as tuberculosis [74] . Drug repositioning provides a promising solution to reduce both the time and costs associated with drug development [73] . We have developed a computational approach to compare the binding sites of a subset of existing drugs approved for human use against the entire M . tb structural proteome . In this way , it is possible to identify putative new targets of existing drugs within the M . tb proteome , providing the basis for their repositioning to treat tuberculosis . Our drug-target interaction network , the TB-drugome , revealed not only that many existing drugs show the potential to be repositioned to treat tuberculosis , but also that some drugs show the potential to be multi-target inhibitors . This is beneficial since multi-target therapy is thought to be more effective than single-target therapy when treating infectious diseases [77] . In addition , the TB-drugome suggests that a large number of M . tb proteins are potentially druggable and could therefore serve as novel drug targets in the fight against tuberculosis . We provide the TB-drugome ( http://funsite . sdsc . edu/drugome/TB ) for analysis by others .
There are 3 , 996 proteins in the M . tb proteome , 284 of which have solved structures in the RCSB PDB ( November 5 , 2009 ) . Although this approximates to only 7 . 2% structural coverage of the M . tb proteome , it is worth noting that there is likely to be a strong bias towards those targets being relevant to drug discovery . There are multiple structures available for many of these proteins ( i . e . , a single protein may have been solved with a number of different ligands ) , bringing the total of solved M . tb structures to 749 ( November 5 , 2009 ) ( see Table S4 in the Supporting Information for further details ) . It was decided that all 749 of these structures should be used in this study , since a single protein may exhibit multiple binding modes and such information would be missed if only a single structure was chosen to represent each of the 284 proteins . It is important to note that the whole biological unit , rather than a single chain of each structure was used in the case of experimental structures so as to capture ligand binding sites at the interface between polypeptide chains . ModBase [78] , a database of annotated comparative protein structure models , contains homology models for the entire M . tb proteome . However , since they are derived from an automated pipeline , it is likely that some of these models may contain significant errors . Each model in ModBase has been assigned a score corresponding to its reliability , which is derived from statistical potentials . A model is predicted to be reliable if its model score is greater than 0 . 7 and its ModPipe Protein Quality Score ( MPQS ) is greater than 1 . 1 ( http://modbase . compbio . ucsf . edu/modbase/modbase_help . html ) . By employing these thresholds , it is possible to discard unreliable models . ModBase was found to contain ‘reliable’ homology models for a total of 1 , 446 unsolved M . tb proteins ( see Table S5 in the Supporting Information for further details ) . Through the additional use of these reliable homology models , the structural coverage of the M . tb proteome was increased to around 43% . However , only a single chain of each homology model was available , rather than the entire biological unit . Drugs approved for human use in the United States and Europe are listed in the U . S . Food and Drug Administration ( FDA ) Orange Book ( http://www . accessdata . fda . gov/scripts/cder/ob/default . cfm ) and by the European Medicines Agency ( EMEA ) ( http://www . emea . europa . eu/htms/human/epar/a . htm ) , respectively . The names of the active ingredients of these drugs were extracted and mapped to compounds in three databases; PubChem ( http://pubchem . ncbi . nlm . nih . gov/ ) , DrugBank [13] , [79] ( http://www . drugbank . ca/ ) and ChEBI ( http://www . ebi . ac . uk/chebi/ ) . After removing all nutraceuticals and prodrugs , InChI keys were used to map the remaining compounds to protein crystal structures in the PDB . Non-protein crystal structures such as DNA , RNA and ribosomes were excluded . 274 different drugs were identified bound to a total of 962 different protein binding sites ( November 30 , 2009 ) . A full list of the approved drug binding sites used in this study is provided in the Supporting Information , Table S1 . Xie et al . recently developed the ligand binding site comparison software SMAP [22] , which is based on a sequence order independent profile-profile alignment ( SOIPPA ) algorithm [24] . Firstly , the protein structure is characterized by a geometric potential; a shape descriptor that is analogous to surface electrostatic potential , but which uses a reduced C-alpha only structural representation of the protein . It has been shown that both the location and the boundary of the ligand binding site can be accurately predicted using the geometric potential [23] . The reduced representation of the protein structure makes the algorithm tolerant to protein flexibility and experimental uncertainty; thus SMAP can be applied to low-resolution structures and homology models . Secondly , two protein structures are aligned , independent of sequence order , using a fast , maximum weighted sub-graph ( MWSG ) algorithm [80] , [81] . The MWSG finds the most similar local structures in the spirit of local sequence alignment . Finally , the aligned surface patches are ranked by a scoring function that combines evolutionary , geometric and physical information . The statistical significance of the binding site similarity is then rapidly computed using a unified statistical model derived from an extreme value distribution [22] . The SMAP software was used to compare the binding sites of the 749 M . tb protein structures plus 1 , 446 homology models ( a total of 2 , 195 protein structures ) with the 962 binding sites of 274 approved drugs , in an all-against-all manner . While the binding sites of the approved drugs were already defined by the bound ligand , the entire protein surface of each of the 2 , 195 M . tb protein structures was scanned in order to identify alternative binding sites . For each pairwise comparison , a P-value representing the significance of the binding site similarity was calculated . FATCAT ( Flexible structure AlignmenT by Chaining Aligned fragment pairs allowing Twists ) [82] is a program for the flexible comparison of protein structures . It optimizes the alignment between two structures , whilst minimizing the number of rigid body movements ( twists ) around pivot points introduced in the reference structure . In addition to the optimal structural alignment , FATCAT reports the statistical significance of the structural similarity , measured as a P-value . In order to identify pairs of similar binding sites that were from proteins with dissimilar global structures ( i . e . , cross-fold connections ) , the first chain of each PDB file was aligned using FATCAT , and those pairs with a significant P-value of less than 0 . 05 were discarded . yEd Graph Editor from yWorks ( http://www . yworks . com/en/products_yed_about . html ) was used to visualize the drug-target interaction network . M . tb protein names were taken from the NCBI Entrez protein database ( http://www . ncbi . nlm . nih . gov/protein ) , to avoid inconsistencies in the naming of proteins in the PDB . GSMN-TB [27] , a web-based genome-scale network model of M . tb metabolism was used to carry out flux balance analysis ( FBA ) computations . The GSMN-TB model contains 739 metabolites and 726 genes that are involved in 849 unique reactions . For those M . tb genes of interest that were also present in the GSMN-TB model , the single gene knockout tool was used to run essentiality prediction under conditions optimized for in vivo growth . If the resulting maximal theoretical growth rate was zero or close to zero , then a gene was predicted to be essential , whereas if it was the same as wildtype ( 0 . 050191 mmol/g DW/h ) , it was predicted to be non-essential . In order to simulate multiple gene knockouts , the reactions in which these genes were involved were constrained by setting their upper and lower bound values to zero . Note that this was only carried out for those reactions that could not be carried out by any other genes , i . e . , those that were entirely dependent on the gene of interest . iNJ661 [28] is another genome-scale metabolic reconstruction of M . tb that contains 828 metabolites and 661 genes which are involved in 939 reactions . In order to determine in vitro essentiality we used the COBRA Toolbox [83] to perform single gene deletions on the iNJ661 model grown in Middlebrook 7H9 media . Again , genes were predicted to be essential if the maximal theoretical growth rate resulting from their deletion was zero or close to zero . For those pairs of interest , molecular docking was used to predict the binding pose and affinity of the drug molecule to the M . tb protein . eHiTS Lightning [84] was selected due to its fast speed , relatively high accuracy and ease of automation for large-scale docking studies . Since SMAP had aligned the drug binding site with the M . tb protein binding site , the aligned coordinates of the drug molecule were used to define the search space for docking that drug into the M . tb protein . The aligned drug molecule was used as the clip file with a default search space of 10Å3 . As recommended by the manual , the eHiTS accuracy level was set to 6 ( default = 3 ) , in order to increase the accuracy of the predicted binding poses . Following all docking , the binding pose with the lowest estimated binding affinity was selected for further investigation . For those proteins with cofactors ( e . g . , InhA has an NAD cofactor ) , the cofactor was added as the last residue in the protein structure prior to docking . The drug-target interaction network can be represented as a graph . The number of targets or drugs against their connectivity in the graph can be fitted to a power-law distribution , where: y and x are the number of targets or drugs and their connectivity , respectively , and α and k are two fitted parameters . A protein graph was constructed for the drug-target network . Nodes represented proteins and edges were formed between two protein nodes if they were connected to the same drug . Then the fraction of the largest connected component ( nLCC ) of the protein was computed by dividing the number of proteins in the largest single linkage cluster by the total number of proteins in the graph . The nLCC values of drugs can be computed in a similar manner . Protein and drug binding profiles in the TB-drugome were hierarchically clustered using GenePattern 2 . 0 [85] . The distance between the profiles was measured using the city block distance . The 2D fingerprint similarity of drugs was computed using OpenBabel 2 . 1 . 1 ( http://openbabel . org ) .
|
The worldwide increase in multi-drug resistant TB poses a great threat to human health and highlights the need to identify new anti-tubercular agents . We have developed a computational strategy to link the structural proteome of Mycobacterium tuberculosis , the causative agent of tuberculosis , to all structurally characterized approved drugs , and hence construct a proteome-wide drug-target network – the TB-drugome . The TB-drugome has the potential to be a valuable resource in the development of safe and efficient anti-tubercular drugs . More generally , the proteome-wide and multi-scale view of target and drug space may facilitate a systematic drug discovery process , which concurrently takes into account the disease mechanism and druggability of targets , the drug-likeness and ADMET properties of chemical compounds , and the genetic dispositions of individuals . Ultimately it may help to reduce the high attrition rate in drug development through a better understanding of drug-receptor interactions on a large scale .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"computational",
"biology/macromolecular",
"structure",
"analysis",
"computational",
"biology/systems",
"biology",
"biotechnology/small",
"molecule",
"chemistry"
] |
2010
|
The Mycobacterium tuberculosis Drugome and Its Polypharmacological Implications
|
To fight infections , rare T cells must quickly home to appropriate lymph nodes ( LNs ) , and reliably localize the antigen ( Ag ) within them . The first challenge calls for rapid trafficking between LNs , whereas the second may require extensive search within each LN . Here we combine simulations and experimental data to investigate which features of random T cell migration within and between LNs allow meeting these two conflicting demands . Our model indicates that integrating signals from multiple random encounters with Ag-presenting cells permits reliable detection of even low-dose Ag , and predicts a kinetic feature of cognate T cell arrest in LNs that we confirm using intravital two-photon data . Furthermore , we obtain the most reliable retention if T cells transit through LNs stochastically , which may explain the long and widely distributed LN dwell times observed in vivo . Finally , we demonstrate that random migration , both between and within LNs , allows recruiting the majority of cognate precursors within a few days for various realistic infection scenarios . Thus , the combination of two-scale stochastic migration and signal integration is an efficient and robust strategy for T cell immune surveillance .
Pathogens are enormously diverse . They differ in tissue localization , epitope expression , virulence , and many other factors . Still , our immune system has to swiftly cope with invading pathogens to ensure our survival . Intriguing evidence from rather different infection models like influenza ( a local infection of the respiratory tract ) , dermal herpes simplex , and listeriosis ( a systemic infection ) shows that the immune system manages to activate a majority of the Ag-specific T cell precursors within just a few days [1] , [2] . How can this remarkable efficiency and robustness be achieved ? A key component of our immune system's defense strategy is to keep T cells and other lymphocytes constantly mobile . Because the T cell repertoire needs to be both specific and diverse , each T cell recognizes only a few epitopes . Conversely , only very few T cells – in mice , as little as 20–200 [3]–[5] – can respond to any given Ag . To avoid that local pathogen intrusions go unnoticed , T cells search for Ag proactively by migrating between and within different organs and tissues . Lymphocyte migration between tissues has been studied for decades , notably from the 1960s to the 1980s [6] , whereas cell migration within tissue has become amenable to experiments only recently with the advent of two-photon imaging [7] , [8] . Here , we combine classic and recent data about T cell migration on both scales into a common model . Our goal is to pinpoint the key aspects of T cell trafficking that help the immune system respond firmly and rapidly against many different pathogens . Several previous modeling studies have addressed individual aspects of T cell migration in their own right , many of them spurred by pioneering intravital two-photon experiments that surprisingly showed lymphocyte migration in LNs to be random-walk-like [9] , [10] . These models have provided insights into stop-and-go T cell motion [11] , the relationship between LN transit time and LN structure [12] , [13] , and the time needed for T cells to find dendritic cells ( DCs ) presenting cognate Ag [11] , [14] , [15] . Fewer models have addressed LN migration between organs [16]–[19] , and only recently have the first models combined between-organ migration with a simple representation of T cell priming in LNs as an exponential decay process [20] , [21] . From two-photon imaging , we know however that T cell priming in LNs follows a more complex three-phase timecourse [22] , [23] . Here we combine existing hypotheses on T cell priming to build a general kinetic model of T cell retention in LNs . Fitting our model against imaging data suggests that T cells in LNs can integrate Ag signals on a timescale of hours , which might help to detect even low-dose Ag reliably . Moreover , we combine the priming kinetics with an explicit model of T cell migration within and between LNs , blood and spleen to ask how two-scale migration and priming interact and affect each other . Specifically , we study the impact of signal integration on the trade-off between fast recirculation and thorough Ag search [20] , [21] , and ask why in vivo LN transit times are so broadly distributed . Finally , we show that the fast T cell recruitment observed in vivo for various infections [1] , [2] can indeed be explained by two-scale stochastic migration .
T cell priming in LNs can occur in 3 distinct phases [22] , [24] , [25]: In phase I , the T cell remains motile and establishes serial brief contacts ( lasting a few minutes ) with Ag-bearing DCs until , in phase II , the cell comes to a halt and establishes a stable DC contact ( lasting hours ) . Ultimately , in phase III , it detaches from the DC , migrates away , and starts proliferating . T cells upregulate CD69 during phase I [22] , suggesting that the brief contacts are immunologically productive and allow to integrate Ag signals from several DCs before committing to retention . Alternatively , given that Ag dose may vary among DCs , T cells in phase I could simply fail to find a DC with a high enough dose for retention , thus the brief contacts might represent unsuccessful retention attempts that do not contribute to reaching phase II . The latter hypothesis has been termed probabilistic priming [26] . Despite recent advances that allow to combine intravital cell tracking with in situ cytometry [27]–[29] , demonstrating that signal integration occurs in vivo remains difficult because phase I lasts for several hours , and it is currently infeasible to track single cells that long in intravital imaging experiments [30] . We therefore used a mathematical model to derive testable predictions from the signal integration and probabilistic priming hypotheses ( Figure 1A; Methods ) , and tested these predictions against in vivo two-photon data . Because T cell migration through LN tissue resembles a persistent random walk [13] , [31] , we considered waiting times between DC encounters to be exponentially distributed , a simplification that has been used and validated in a similar sphere model of T cell random walk in LNs [15] . With probabilistic priming , the waiting time is interpreted as the time required to find a new DC , as multiple encounters with the same DC that does not present enough peptide do not contribute to retention . For each new contact , the DC presents a sufficient amount of peptide with a probability that depends on the Ag . For example , at a 1/8 probability , retention occurs on average after 8 unique contacts . With signal integration , multiple contacts with the same DC ( or different DCs ) do contribute to retention , which occurs after an Ag-dependent number of contacts . For persistent random walks in a relatively large three-dimensional structure like a LN , one can expect roughly 2/3rd of all contacts to be unique due to Polya's recurrence theorem [32] . This expectation has been confirmed by a detailed agent-based model of T cell–DC contacts in lymph nodes [11] . Hence , at the same “true” underlying DC contact rate , the effective contact rate of probabilistic priming is about 2/3rd that of signal integration . In the rest of this paper , we only refer to the effective contact rate for each model . With both priming models , the time until retention ( duration of phase I ) is stochastic due to the waiting times , and the variance of this duration differs between the models . For instance , when comparing simulation trajectories of both models at ( on average ) 8 required contacts and the same effective contact rate , retention typically starts earlier , but completes later with probabilistic priming than with signal integration ( Figure 1B ) . In other words , probabilistic priming implies gradual retention , whereas signal integration leads to a switch-like timecourse ( Figure 1C , blue lines ) . This observation is independent of the contact rate , which equally affects the time scaling of both models . However , at higher Ag doses , the difference between the 2 priming models is much smaller ( Figure 1C , red lines ) , because signal integration becomes less relevant when retention can occur after 1 or a few contacts . Nevertheless , this basic effect implies that signal integration completes retention of an entire Ag-specific cell population faster as well as more reliably than probabilistic priming . The switch-like retention kinetics predicted by signal integration ( Figure 1C ) provide a testable prediction that can be confirmed or rejected by experiments . To determine the retention kinetics of real T cells , we applied a “FACS-like” motility analysis [27] to a set of ∼22 , 000 T cell tracks extracted from 38 two-photon videos from a previous study [25] where naive Ag-specific and control T cells were imaged at different time points after synchronized entry into popliteal LNs containing peptide-pulsed DCs ( Figure 2A ) . In these experiments , varying doses of 2 peptides were used that differed only in the terminal MHC anchor residue ( “M-peptide” with high MHC affinity , or “C-peptide” with low affinity [25] ) . We estimated the fraction of retained cells in each video by “gating” T cells on the motility coefficient estimated from their track ( Figure 2B; Methods ) , which confirmed that retention increased over time for the Ag-specific but not for the control cells ( Figure 2C ) . At high Ag doses , most T cells were retained early on ( Figure 2C , M-peptide and C-peptide ) . They should therefore have entered phase II after only a few contacts , making it difficult to assess whether signal integration took place ( cf . Figure 1C ) . However , at low Ag doses ( Figure 2C , M-peptide and C-peptide ) , retention kinetics were indeed markedly switch-like , as predicted by our model . Specifically , most cell retention occurred at 4–5 h after cell transfer , whereas most retention should occur shortly after LN entry with probabilistic priming . These data suggest that cells integrate signals in vivo during phase I on a timescale of hours , which governs the onset of phase II . To more precisely quantify the level of support that our data lends to the signal integration hypothesis , we employed statistical model selection starting from a general model that accommodates both signal integration and probabilistic priming ( Methods ) . The 3 parameters of this general model are as follows . First , T cells encounter DCs at a fixed rate . Second , there is a peptide dose-dependent success probability for each contact , with “success” meaning that a cognate signal is transmitted . Third , there is a dose-dependent number of successful contacts required for T cell retention . Because the C-peptide has a very short half-life of 2 . 4 h on the MHC molecule compared to 6 h for the M-peptide [25] , implying that T cell priming might stop within the 8 h time frame of interest , we first analyzed the M-peptide data only . Specifically , we fitted the general model to six M-peptide datasets comprising 3 different Ag doses ( Figure 3 ) . Each dataset was recorded in an independent experiment and consisted of 2 or 3 two-photon videos imaged at different times upon cell entry . We constrained the underlying DC encounter rate to be equal across all videos , as the number of injected DCs was constant . Further details on the fitting procedure can be found in the Methods . The general model ( top row of Figure 3 ) gave an acceptable fit and showed good agreement among different experiments with the same Ag dose . Next , we created 2 restricted versions of our general model by disabling signal integration or probabilistic priming , which leads to the “pure” signal integration and probabilistic priming models shown in Figure 1C . By comparing the Bayesian information criterion ( BIC ) score of each pure model fit ( Figure 3 , middle and bottom rows ) to the general model fit ( Figure 3 , top row ) , we assessed the relative importance of each priming mechanism in the general model . The general model fits the data best in terms of BIC , suggesting that both priming mechanisms are required for explaining the data . However , the BIC score ( misfit ) increased by 24 . 6 when signal integration was disabled ( purely probabilistic priming ) but only by 1 . 41 when probabilistic priming was disabled ( pure signal integration ) . Applying a common interpretation scale for BIC [33] , this indicates that the evidence for probabilistic priming is quite weak ( ) whereas the evidence for signal integration is very strong ( ) . Similar results were obtained when fitting the models to the M-peptide and C-peptide data combined ( to general model: 0 . 6 without probabilistic priming and 20 . 2 without signal integration ) . However , the model fit to the C-peptide data was considerably poorer ( not shown ) , probably due to the rapid peptide loss which the model does not take into account . Overall , our statistical analysis lends further support to the hypothesis that T cells integrate signals from DCs they encounter . For high Ag doses this is difficult to distinguish from probabilistic priming because only few interactions lead to T cell retention , yet at low Ag doses the signal integration is clearly detectable . To study the interplay between priming within LNs and trafficking between LNs , we designed a stochastic two-scale model of T cell trafficking between secondary lymphoid organs ( SLOs ) in mice ( Figure 4 ) , similar to previous models [17] , [20] , [21] but anatomically more explicit . In the new model , cells in the blood home to T cell zones in LNs and splenic white pulp ( Figure 4A ) . We represent the T cell areas in the LN paracortex as three-dimensional spheres ( Figure 4B ) , which in silico cells enter in the center and then migrate randomly until reaching the surface . The sphere center represents a high endothelial venule , and the surface represents cortical sinusoids as well as subcapsular and medullary sinuses . In contrast , splenic T cell areas ( periarteriolar lymphoid sheaths , or PALS ) are cord-like structures around central arterioles in the white pulp , which T cells are thought to access via so-called marginal zone bridging channels [34] . In our model we represent the PALS as a cylinder with small apertures on both sides , with 1 aperture being used for entry and the other for exit ( Figure 4C ) . The length of this cylinder is irrelevant for our purpose , because movement along the cylindrical main axis does not bring the cell closer to or further away from an exit site . A quantitatively reasonable parametrization of this model is necessary to make reliable predictions . Adopting data from a previous meta-analysis of several migration experiments [17] , we first set the total entry rates from blood to spleen to , and from blood to all LNs combined to . To determine the entry rates into and egress rates from the individual LNs , we analyzed raw data from a recent study [35] where adoptively transferred cells were counted in LNs at various time points after injection and LN entry blockade to estimate entry and egress rates . We found a strong correlation between LN entry rate and LN size ( Figure 5A ) . However , egress from peripheral LNs was not significantly faster than egress from the substantially larger mesenteric LNs ( Figure 5B , C ) , thus there is no evidence for a relationship between LN size and egress rate . This observation is consistent with the fact that large LNs are often composed of several individual lobes or compartments that each have their own entry and egress structures . We incorporated these findings by using a single sphere to represent small LNs , and multiple spheres to represent larger LNs . These multiple spheres could for example be viewed as different lobes of an inguinal LN , or as the individual LNs that form the mesenteric LN . In our model we use 30 LNs , similar to mice [36] . We represent large peripheral LNs , i . e . , brachial , axillary , and inguinal LNs , by 2 spheres each , and the mesenteric LN by 4 spheres . All other LNs ( e . g . , the popliteal LN ) are represented by 1 sphere . We thus have 39 spheres in total . By distributing the T cells leaving the blood evenly across the spheres , and using the same sphere diameter for all LNs , we achieve that entry rate is correlated to LN size but egress rate is not . The transit time through the spheres is determined by the random walk motility coefficient and the sphere radius . We set the motility coefficient to , an estimate that we previously obtained from two-photon data [13] , and set the sphere radius to a value that yields a physiologic average transit time of 13 . 5 h [17] , [35] . Similarly , we set the geometric parameters of the spleen cylinder ( Figure 4D ) to values that lead to an average transit time of 6 h [37] . With these parameters ( Table 1 ) , the model accurately predicts a blood residence time of 25 min and a realistic distribution of T cells across SLOs ( about 74% in LNs , 23% in spleen ) and blood ( 3%; Table 2 ) . Despite the simple structure of our model , its quantitative predictions ( Table 2 ) suggest that our simulations provide a reasonable reproduction of the kinetics of T cell migration . We emphasize that this is largely achieved by setting the parameters to values reported in previous studies or derived from our own data , rather than by parameter fitting; sphere diameter and cylinder aperture angles were set to obtain realistic transit times , but the values used are anatomically reasonable . We therefore proceed to use this model to study the interplay between within-LN priming kinetics and between-LN migration kinetics . Previous models [20] , [21] suggested that T cell trafficking strategies have evolved subject to a trade-off: Frequently recirculating cells arrive more rapidly at relevant SLOs upon infection , but reliable Ag detection may require long dwell times within SLOs . It may seem that this conflict could be solved by letting T cells transit rapidly through non-infected LNs and keeping them longer in infected LNs . However , classic data shows that after a brief ( <1d ) initial “shutdown” period [38] , [39] , T cell egress from infected LNs is fully restored [40]–[42] , perhaps to avoid that infected LNs clog up with irrelevant T cells . Hence , the baseline LN dwell time of T cells has to be long enough to ensure reliable retention of Ag specific cells , and short enough to ensure rapid arrival at infected LNs . Aiming to quantify this trade-off in our model , we combined our simulations of migration between SLOs and priming within LNs . For simplicity , we used a hypothetical infection where Ag dose and quality , as well as DC encounter rates were kept constant over time , similar to earlier models [21] . Each in silico cell was followed until successful retention in an Ag-bearing LN ( Figure 6A ) . To quantify the efficiency of specific LN transit times , we also performed simulations where we let in silico cells spend fixed times in each LN instead of searching an exit via random walk ( deterministic LN transit; Figure 6B ) . In each scenario , efficiency was assessed by determining the average time taken from Ag appearance until T cell retention in a LN . In the following , we refer to this time period as the capture time to emphasize the difference to the within-LN retention time studied above . For instance , in simulations of an infection where the Ag is present in 25% of the LN spheres , combined with signal integration priming at an 8 h phase I ( like in Figure 1B ) , a realistic LN transit time of 12 h balances well between rapid arrival and robust retention , and leads to an expected capture time of ∼4d . In contrast , we obtain capture times of ∼6d for a transit time of 24 h and ∼9d for a transit time of 6 h ( Figure 6C ) . Even though this quantitative prediction is based on our simplified hypothetical infection , it is intriguing to observe that the most efficient range of transit times predicted by our model is similar to physiological transit times [35] . To generalize these results , we analytically determined how the capture time depends on the migration and priming parameters ( Methods ) , and derived equations to calculate the “optimal” LN transit time that would lead to the fastest detection of a given Ag . The results show that the disadvantage of overly long transit times is hardly affected by signal integration ( Figure 6C , 24 h and beyond ) . However , very short transit times can be extremely detrimental with signal integration at low Ag dose ( Figure 6C , 6 h and below ) , as most cells then exit before retention starts . With probabilistic priming , this effect is less severe as at least some cells are still retained early on . In summary , signal integration has different implications for T cell trafficking between organs than probabilistic priming . Rapid LN transit in particular appears much less favorable when taking signal integration into account . Together with the need for T cells to receive “survival signals” by self-pMHC [21] , this may explain why LN transit times are not shorter in vivo . At first sight , the prediction that LN transit time is important for rapid Ag detection appears inconsistent with the fact that T cell dwell times in LNs are widely distributed [35] , [37] . If rapid capture depends on proper LN transit timing , then should evolution not have settled for a more tight control of the LN transit ? For instance , it has been hypothesized that T cells migrate from the deep paracortex towards egress sites in a directed fashion [12] , [43] , [44] . Such a mechanism could facilitate a more precise timing of the LN transit . Indeed , if we make all cells transit every LN in the “optimal” 11 . 7 h instead of transiting randomly in our previous simulation ( infection in 25% of the LN spheres , 8 h phase I , signal integration ) , the capture time slightly decreases from ∼5 . 5d to ∼4d – in other words , precisely timed LN transit can lead to 1 . 4-fold faster Ag detection in this setting . However , adjusting LN transit to optimally detect a given Ag dose comes at a price with respect to detection at other Ag doses: when performing simulations where the LN transit time was kept constant but the Ag parameters were varied , we found that T cells with deterministic LN transit were not well equipped to deal with low Ag doses ( Figure 6D ) . To systematically assess the potential benefits and risks of LN transit optimization , we considered Ag doses with 2 to 10 required contacts ( leading to a 2–10 h phase I ) . For each of these settings , we computed the optimal LN transit time based on our analytical solution of the model ( Methods ) . As expected , in silico cells that stay in each visited LN for exactly this optimal time detect the Ag faster than cells that transit LNs stochastically ( e . g . , for 8 contacts , 1 . 38-fold faster ) . However , testing how fast Ags at other doses would be detected by deterministic LN transit ( e . g . , we exposed cells that transit deterministically in 11 . 7 h , which is optimal for 8 contacts , to Ag doses requiring 2 , 4 , 6 and 10 contacts ) showed that the risk of LN transit optimization , in terms of slower detection of “unexpected” Ag doses , can be orders of magnitude larger than the best possible gain ( Figure 6E ) . In reality , Ag abundance , dose , and quality will vary considerably across infections and over time . Therefore , whereas letting T cells roam freely through LNs provides robust protection against different infection scenarios , the potential speedup gained for some pathogens by tightly controlling LN transit seems to be dwarfed by the large potential risk . These findings offer an explanation for the broad distribution of in vivo LN residence times [35] . For many infections , the immune system is able to recruit almost all Ag-specific T cells into the immune response within a few days [1] , [2] , which is similar to the capture times predicted by our simulations . However , these simulations were based on a hypothetical infection where Ag was instantly and constantly available , and the draining area was kept constant . Aiming for a more realistic infection simulation , we integrated data on spatial and temporal Ag availability for real infections and tested whether our model predictions are consistent with the efficient recruitment observed in vivo . Because computation of the capture time would require detailed kinetic information on Ag dose and density as well as DC frequency and quality within LNs , we focused on the time required to arrive at a relevant SLO ( Figure 7A ) . Unprimed Ag-specific cells do not egress from Ag-bearing SLOs in relevant numbers [40]–[42] . The capture time should therefore be just a few hours above the arrival time . First , we simulated arrival of Ag-specific T cells at the spleen , and found that ∼95% of all cells arrive within the first 3d ( Figure 7B , solid line ) . We compared this prediction to data for blood-borne Listeria monocytogenes infections . Listeria enters the spleen almost instantly [45] , and T cell priming then occurs mainly during the first 3–4d [46] ( Figure 7B , colored rectangle ) . The percentage of naive T cells that are recruited into immune responses to various different infections has been estimated using cellular barcoding [1] , [47] . For intravenously administered Listeria [1] , about 95% of all Ag-specific T cells took part in the immune response ( Figure 7B , gray bar ) . Given the arrival speed predicted by our model , a priming time window of 3 . 5d would suffice to support recruitment of >90% of all Ag-specific T cells even if the immune response were only formed in the spleen . Therefore , due to the large throughput of the spleen , random circulation appears sufficient for a swift response initiation against blood-borne pathogens such as Listeria . Searching for Ag is more difficult for local infections , when Ag is only available in the draining LNs ( dLNs ) near the site of infection . Still , when applying the barcoding approach to an influenza infection in the lung , ∼2/3 of all precursors were found to be recruited into the immune response [1] . We simulated T cell arrival for different numbers of dLNs , considering T cell homing from the blood to LNs to be uniformly distributed among 30 equally sized LNs . In this setting ( Figure 7C ) , we found that 6–9 dLNs were required to ensure arrival of 2/3 of all T cells within a typical 5d priming period of influenza [48] . This prediction is consistent with the number of LNs that drain the lung in mice – 2–8 mediastinal , 1 bronchial , and 2–4 deep cervical LNs [36] . Hence , also for local infections , randomly circulating cells can arrive at appropriate SLOs fast enough to support recruiting a majority of all Ag-specific T cells within a few days , as long as the T cell entry rate into dLNs is at least ∼20% of the T cell entry rate into all LNs . Strikingly , it has been shown that even for an HSV-1 infection in the footpads with only 2 draining popliteal LNs , ∼2/3 of the Ag-specific T cells disappear from the circulation within 4d [2] . In contrast to the influenza infection , this result can no longer be explained by stochastic circulation alone – in that case , not even ∼20% of in silico cells arrive at 2 dLNs within 4d ( Figure 7C ) . However , it is known that inflammation-driven vascular growth can rapidly and massively increase cell flux into dLNs [49] , [50] . For instance , following HSV-2 infection in mice , influx increases ∼9-fold within 4d [50] . When we increase the dLN entry rate in our model accordingly ( Figure 7D ) , a single dLN accumulates almost half as many T cells as the spleen within 3–4d , and the predicted cell disappearance from non-dLNs ( Figure 7E ) closely matches experimental estimates for the HSV-1 footpad injection [2] . In summary , even though our model lacks many features of T cell migration that might potentially further accelerate Ag detection and T cell removal , mere inclusion of stochastic recirculation ( ensuring rapid arrival ) and signal integration ( ensuring reliable retention ) on their own were already sufficient to explain the efficient T cell recruitment of the listeriosis and influenza case study . Only for the highly localized HSV-1 infection it was necessary to take an additional migration feature into account , namely the increased entry rate into inflamed LNs .
It might appear implausible that a vital function like detection of foreign Ag would depend on aimlessly wandering cells [51] . Yet , our two-scale modeling of T cell migration showed that the combination of random walk within tissue with stochastic migration between tissues is overall a very efficient and robust strategy to bring Ag-specific T cells to the correct location . Alteration of this simple migration pattern only seems necessary for local infections with very few dLNs , in which case a local increase of the dLN entry rate suffices . In fact , it turned out that stochastic migration can be superior to tightly controlled migration: Optimization of the T cell transit through LNs for most rapid detection of a particular pathogen with specific replication and Ag presentation kinetics would leave the immune system vulnerable to other pathogens , whereas stochastic transit provides far more robust protection at only slightly slower Ag detection . These results align well with the general observation that random search strategies can be very effective [52] . We aimed to base the organ representations in our model ( Figure 4 ) as much as possible on available information on the anatomical structure of lymphoid tissue . This is in contrast to other models of T cell migration [20] , [21] which instead use an exponential distribution to model egress . The main difference is that our models lead to an initial “lag period” during which no cells exit from the organ , because they need a minimum time to reach egress structures . Such a lag period might be beneficial because it prevents premature egress . However , the difference to exponential egress is not very big ( Figure 4E ) , so our results would remain similar if we had we used a rate equation instead of explicit organ representations . Similarly , the precise shape of the compartments does not play a very big role , e . g . almost identical results are obtained when one uses a sphere to model the spleen instead of a cylinder ( not shown ) . However , we found it reassuring to observe that realistic anatomical structures combined with realistic T cell motility lead to realistic transit times . Our migration model does of course not capture the full complexity of T cell migration in our immune system . For this reason , we started our validation using data obtained in a carefully controlled experiment , where many of these complexities are absent . In real infections , the kinetic signature of signal integration that we aimed to detect would likely be obscured by other factors . For instance , even though statistical analysis of T cell migration in the absence of Ag does not reveal any directed migration , there could still be some directional migration hidden in the data [13] , . In the presence of Ag , a weak directional bias has indeed been observed in LNs where productive interactions between CD4 T cells and DCs have already been established [55] . Such biased migration may act in conjunction with signal integration to achieve T cell retention even faster [56] . Moreover , during real infections , T cells arriving early or late at the same LN may encounter very different priming parameters . Given these complexities , we focused on the arrival kinetics when we compared our simulations to priming data for real infections ( Figure 7 ) , and stopped these simulations after arrival . Therefore , we expect the benefit of stochastic migration for robust Ag detection to be even larger in reality than our model predicts ( Figure 6 ) , given that T cells will encounter greater varieties of Ag quantity and quality in vivo than in our simulations . T cell retention in LNs is thought to be mediated by upregulation of CD69 , which blocks S1P-driven egress from LNs [57] . In other words , by upregulating CD69 , a T cell “commits” to staying in the current LN rather than egressing and searching for Ag elsewhere . In our simulations ( Figures 1 and 6 ) , we used the onset of long-lasting stable contacts in phase II as an indicator of T cell retention . However , for low Ag dose , CD69 induction can occur already in phase I [22] . As a consequence , the capture time for e . g . an Ag with an 8 h phase I might in fact be shorter than predicted by our model ( Figure 6C ) . Nevertheless , because we have shown that the implications of our simulations hold within a large range of within-dLN retention times , this possibility does not affect our qualitative conclusions . Moreover , it has recently been shown that in some circumstances , effector responses develop without phase II [58] . Importantly , this finding does not affect our conclusion that the onset of phase II in the data we analyzed was determined by signal integration during phase I . Our study focused on 2 theories that explain the occurrence of short contacts at low Ag dose at the T cell level , i . e . , signal integration and probabilistic priming . We found the purely probabilistic retention model , where cells do not accumulate signals from multiple interactions [14] , [26] , difficult to reconcile with our data . However , a further possible explanation for the transition from phase I to phase II [22] , [25] could be that this is dictated by the DCs instead , e . g . , as a result of ongoing DC maturation [59] , [60] . Detailed information on the progress of these proposed changes at the DC level over time would be necessary to allow us to test this third hypothesis . Because such information is currently lacking , it is not possible to distinguish the DC-driven retention model from signal integration or probabilistic priming models . For the data analyzed here , however , it is hard to argue that differences in DC maturation account for the different retention kinetics , because the only change between experiments was the peptide dose , which is not known to affect DC maturation . Our modeling results for local infections with few dLNs suggest that increased blood flow to the dLNs might be indispensable to combat such infections . This increased blood flow , and the resulting dramatic dLN enlargement , are achieved by remodeling of the central LN feeding arteriole [50] , [61] . Still , even the blood flow through the enlarged arteriole amounts to only a small percentage of the cardiac blood output , and therefore it may still seem baffling how such large fractions of all T cells can arrive at the dLNs so quickly ( Figure 7D , E ) . This finding is more easily understood when the relation between the speed of blood flow and the blood residence time is taken into account . In rodents , T cells remain in the blood for about half an hour [62] , [63] . Because the cardiac output of a rodent sums up to the total blood volume within just a few seconds [64] , a T cell in the blood can circulate many times through the whole body before entering an organ . Therefore , many T cells that come in close proximity of a given LN still end up homing elsewhere . Increased blood flow through the central feeding arteriole thus simply recruits a larger fraction of those lymphocytes that are passing by anyway , and a major global redistribution of the cardiac output is not required to achieve an increased entry rate into dLNs . Although we focused on mouse data , the basic principles of our model are applicable to other species as well , including humans: The basic routes of lymphocyte recirculation described in rodents are similar to those in many vertebrates . Therefore , our qualitative conclusions likely generalize to other species . For example , also in humans with about 550 LNs we expect increased blood flow to dLNs to be extremely important for localized infections with few dLNs , and our finding that near-complete T cell retention is achievable more quickly by signal integration than by probabilistic priming is independent of the migration between SLOs . However , in species other than rodents there is still too little data on T cell migration on both scales to allow for a comprehensive quantitative analysis as we performed in this paper . In a similar vein , it was recently shown that LN dwell times differ considerably between CD8 and CD4 T cells [35] . Future work could address whether these differences might reflect different migration strategies given that these T cell subsets have very different tasks and are exposed to Ag in different contexts and interact with each other in a consequential manner . In summary , we have presented a model of T cell immune surveillance as a two-scale stochastic search and compared the predictions of our model to various experimental findings . Even for local infections with very few dLNs , random migration between SLOs combined with a nonspecific increase of the dLN entry rate enables rare Ag-specific T cells to arrive at dLNs within a few days . Within dLNs , highly reliable retention of randomly migrating T cells can be achieved within a few hours even at low Ag densities owing to the integration of information from multiple cognate DC contacts . Overall , the two-scale stochastic migration pattern of T cells appears to be a remarkably efficient and robust solution to the needle-in-a-haystack problem of recruiting rare T cells into immune responses .
Cells were tracked using Volocity software , and statistical analysis of the cell tracks was performed using custom-written software . Tracks shorter than 2 minutes were removed from the analysis . Motility coefficients of 3D cell tracks were estimated as ( 1 ) where is the duration of the track , is the th of positions in the track , and is the distance of the th position of the track to its middle position . For tracks of even length we use ( 2 ) The derivation of Equation 1 is straightforward if one considers the T-cell migration as a Brownian motion . Note that in this manner we probably underestimate the motility coefficients of short T-cell tracks [13] . However , for our analysis this bias is acceptable because we do not use the actual motility coefficient values nor do we directly compare motility coefficients of cell tracks of different length . To classify cells in a given video as retained or non-retained , we first analyze the combined set of all control cell tracks from the same experiment ( several videos imaged on the same day ) . Let denote the motility coefficients and track durations ( in video frames ) of these control cells , respectively . We first compute the weighted median of the , which is the median of the sequence in which each is repeated times . The threshold to define a retained cell is then set to . Now let denote motility coefficients and track durations of a set of Ag-specific cells . We estimate the fraction of retained cells as ( 3 ) that is , the combined duration of all tracks with a motility coefficient below divided by the combined duration of all tracks . This way of computing corrects for the fact that non-retained cells will have on average shorter tracks than retained cells , and therefore makes it possible to view as an estimate of the fraction of retained cells simultaneously visible in the video . To obtain the data for the model fitting , we consider 3 time windows of 20 min per video of 60 min length . Tracks that cross the boundary of a time window are split accordingly . Moreover , we consider the time point of each video relative to the time point of LN entry ( ∼1 h after injection ) rather than relative to the time point of injection , because priming can only start after LN entry . From videos imaged directly after infection , we estimated that entry occurred on average 1 h after injection . Let denote the fraction of retained cells in a time window estimated as described above . We correct for “background noise” using the formula where denotes the fraction of retained cells estimated by applying the above analysis to all control cells imaged in the same experiment as the given video . In the general priming model , which combines signal integration and probabilistic priming , we consider in silico cells to be retained after they have established “successful” cognate encounters with being the mean waiting time between such encounters . Because the waiting times are exponentially distributed and independent from each other , the time to retention is Gamma distributed . Hence , the function used to fit the resulting data is , where is the probability that retention occurs before time , i . e . ( 4 ) Here denotes the usual gamma function and the lower incomplete gamma function . Note that our fit cannot identify the “true” DC contact rate and success probability , but only the rate of successful contacts : A 2 h waiting time ( ) with a 50% success rate ( ) leads to and is therefore indistinguishable from a 4 h waiting time with a 100% success rate . When fitting the general model , we allow both and to vary between independent experiments . For instance , with 6 experiments ( Figure 3 ) , the general model has in total 12 parameters . The purely probabilistic model has 6 parameters ( varies across experiments , is fixed to 1 ) and the pure signal integration model has 7 parameters ( is constant across experiments , varies ) . To account for heteroschedasticity ( the variance of the retention data near the limits 0% and 100% is lower than near 50% ) , we fit the model on a logit scale . Best fits and the corresponding BIC values are computed using GNU R . Our stochastic model follows single Ag-specific T-cells . Because their frequency in the pool of all T-cells is extremely low [3] , [4] , Ag-specific T-cells are considered to circulate independently without influencing each others' paths . Therefore , our results are independent of the number of cells . In the blood , in silico T-cells keep circulating until they encounter a random entry site into a secondary lymphoid organ ( SLO ) . The waiting times for encountering these entry sites are exponentially distributed with rate for the spleen and rate for the th LN sphere ( ; larger LNs are modelled by multiple spheres as discussed in the main text ) . The SLOs are modeled as three-dimensional objects . Specifically , the LN is represented as a sphere with radius . The transit of in silico cells through this sphere starts at the center , which represents a high endothelial venule in the LN paracortex . Cells then perform a Brownian motion with motility coefficient until reaching the sphere surface , which represents cortical and medullary sinusoids [12] . From the sphere surface , the cells move back into the blood . The spleen is represented as a cylinder of arbitrary length and radius . Cells enter the cylinder at a point on the left border , which represents immigration from the splenic marginal zone ( Figure 1 in ref . [34] ) via a marginal zone bridging channel ( MZBC ) . They then perform a Brownian motion through the cylinder with the same motility coefficient as in LNs . However , in contrast to the LN sphere , a large part of the cylinder boundary is treated as a reflecting boundary , representing the interfaces to splenic B-cell areas that ensheathe T-cell areas . Exit is only possible through an opening on the opposite side of the entry point , which represents another MZBC and has an aperture angle of . In silico cells reaching that opening are moved back to the blood . The geometrical parameters of the cylinder and sphere were set to the values shown in Table 1 . For the spleen , these values were set empirically based on anatomical considerations: A cylinder radius and an aperture angle imply that a cross-section through the cylinder ( Figure 4C ) resembles histological PALS sections taken perpendicular to the arteriole ( Figure 1 in ref . [34] ) . These choices lead to a mean residence time in the spleen of 6h , matching classical estimates [65] , [66] . For the LN , the relationship between sphere radius , mean residence time and motility coeffcient can be analytically determined . Let denote the residence time of an in silico cell with motility coefficient in a spherical organ of radius . Then the expected residence time is given by We use the estimate , which is based on two-photon data [13] , for naive T cells . Classic data indicates that in rodents , naive T cells spend on average 13 . 5 h in LNs [17] . Therefore , we set . This value is anatomically reasonable for the T-cell zone of a medium-sized murine LN . In addition to the above equation for the expected residence time , it is also possible to express the entire distribution of cell residence times in the sphere analytically [67] in terms of the infinite series ( 5 ) The transit time distributions for spleen and LNs in our model are shown in Figure 4E . The model described above is easily transformed into a Monte Carlo simulation , which allows us to generate individual cell trajectories ( e . g . Figure 4D ) to examine the fates of simulated cells . In these simulations , cells alternate between transiting the blood and transiting an SLO . We apply the kinetic Monte Carlo method [68] to the following rate equation , which describes cell movement from the blood ( B ) to LNs and spleen according to the rules set out above: ( 6 ) In brief , the kinetic Monte Carlo method works as follows . Let denote the time at which a cell last entered the blood , and let be the sum of all compartment entry rates at time . Then each organ ( the spleen or 1 of the LNs ) is chosen as the next organ to visit with probability . The cell is then moved to the chosen organ , and simulation time is increased by , with uniformly at random . The residence time of a given cell in a LN is sampled according to Equation 5 , or , in some simulations ( Figure 6 ) , is set to a constant . Transit through the spleen is explicitly simulated as described above . To avoid “synchronization” between cell trajectories , Monte Carlo simulations are initialized by putting cells in the blood and letting them circulate for 1+ weeks , with uniformly at random . After initialization , cell trajectories are recorded and the properties of interest ( e . g . , the arrival time at the spleen or at 1 of the dLNs ) are investigated . The probability that a cell is retained when passing through a dLN can be expressed as , where is the waiting time to retention ( Equation 4 ) and is the LN residence time ( either distributed according to Equation 5 or a constant ) . We note that for probabilistic priming , a closed form for exists: Let denote the residence time of an in silico cell with motility coefficient in a LN sphere with radius . Let denote the retention time for probabilistic priming with parameter . Then we have ( 7 ) where is the hyperbolic cosecant . This formula is obtained by integration . Let us now consider the capture time ( Figure 6A ) . The overall efficiency of the two-scale surveillance process can be quantified by the expectation ( lower means more efficient surveillance ) . can be determined by extending our Monte Carlo simulation: When a T cell enters a dLN , the within-dLN retention time is drawn at random according to Equation 4 . The T cell is considered to be retained if . However , for simulations with constant and an infection with a constant dLN entry rate and constant priming parameters ( Equation 4 ) , can also be determined analytically . Consider an infection starting at , and an in silico T cell that is not in a dLN at that time . Let denote the fraction of dLNs among all LNs , the average time spent in the blood and possibly the spleen between 2 consecutive visits of LNs , and the average time at which the cell first enters a LN . Then the expected capture time is given by ( 8 ) with defined as in Equation 4 . Equation 8 can be obtained as follows . Let denote the probability density function of the within-dLN retention time , and let be the associated cumulative distribution function . Let the variable denote the time at which the cell is retained counting from the time at which it entered the final dLN , i . e . , the dLN in which the cell eventually is retained . Let denote the number of unsuccessful visits to dLNs ( that did not lead to retention ) before the successful visit . Let denote the time spent in blood and/or spleen between 2 consecutive LN visits , and the time at which the cell first reaches a LN . Then the overall retention time ( counting from the start of the infection at ) can be written as Importantly , the random variables , , and are mutually independent . For this reason , the expectation of expands as follows: To shorten notation , we identify the variables and with their expectations , i . e . and . is a geometrically distributed variable with parameter . The expectation of can be obtained by noting that is a truncated version of the within-dLN activation time , i . e . , This leads to ( 9 ) from which one obtains Equation 8 by inserting a Gamma distribution for and its integral for .
|
Each of the immune system's T lymphocytes recognizes a highly specific molecular pattern , and only a very few T cells are capable of detecting any given infection . These rare cells are at first scattered throughout the body when a pathogen invades the host . To mount an immune response , they need to come together within lymphatic tissue near the infection site , and find cells that carry molecular fragments of the pathogen . Remarkably , experiments show that the immune system solves this needle-in-a-haystack problem in just a few days for various bacterial and viral infections . Aiming to understand how the immune system achieves this , we built a model that brings together classic and recent data on T cell migration . Our model explains how perpetual migration of T cells between and within lymphatic organs helps to find invading pathogens swiftly and reliably . Specifically , our results suggest that T cells can collect signals from activation-inducing cells for several hours , which allows for reliable detection of even low-profile infections . Thus , random T cell trafficking between and within lymphatic organs robustly protects against a broad range of pathogens , and comes close to an “optimal” surveillance strategy .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"blood",
"cells",
"white",
"blood",
"cells",
"immune",
"cells",
"cell",
"biology",
"animal",
"cells",
"t",
"cells",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"immunology",
"immune",
"response"
] |
2014
|
Random Migration and Signal Integration Promote Rapid and Robust T Cell Recruitment
|
UNC93B1 associates with Toll-Like Receptor ( TLR ) 3 , TLR7 and TLR9 , mediating their translocation from the endoplasmic reticulum to the endolysosome , hence allowing proper activation by nucleic acid ligands . We found that the triple deficient ‘3d’ mice , which lack functional UNC93B1 , are hyper-susceptible to infection with Toxoplasma gondii . We established that while mounting a normal systemic pro-inflammatory response , i . e . producing abundant MCP-1 , IL-6 , TNFα and IFNγ , the 3d mice were unable to control parasite replication . Nevertheless , infection of reciprocal bone marrow chimeras between wild-type and 3d mice with T . gondii demonstrated a primary role of hemopoietic cell lineages in the enhanced susceptibility of UNC93B1 mutant mice . The protective role mediated by UNC93B1 to T . gondii infection was associated with impaired IL-12 responses and delayed IFNγ by spleen cells . Notably , in macrophages infected with T . gondii , UNC93B1 accumulates on the parasitophorous vacuole . Furthermore , upon in vitro infection the rate of tachyzoite replication was enhanced in non-activated macrophages carrying mutant UNC93B1 as compared to wild type gene . Strikingly , the role of UNC93B1 on intracellular parasite growth appears to be independent of TLR function . Altogether , our results reveal a critical role for UNC93B1 on induction of IL-12/IFNγ production as well as autonomous control of Toxoplasma replication by macrophages .
Toxoplasma gondii is a widespread obligate intracellular protozoan parasite , which establishes itself in the brain and muscle tissues , persisting for life in humans and other vertebrate hosts [1] . One of the most distinctive aspects of T . gondii life cycle is the establishment of an often benign chronic infection , which is dependent on the parasite's ability to elicit a strong and persistent cell-mediated immunity [1] . Severe forms of toxoplasmosis are often seen in humans with an immature or suppressed immune system . In particular , cytokines that activate macrophage effector functions , such as IFNγ and TNFα , are critical in mediating host resistance to T . gondii infection [2] . The mammalian Toll-like receptors ( TLRs ) sense conserved molecules from all classes of microorganisms [3] , including those from protozoan parasites [4] . Studies employing MyD88−/− mice , which are deficient in the function of most TLRs ( except for TLR3 ) , suggest that TLRs are critical in many aspects of host:protozoan parasite interaction , including the initiation of the pro-inflammatory cytokine response and the expression of co-stimulatory molecules [5] , [6] , [7] . The initial activation of the innate immune system leads to the immediate activation of anti-microbial effector mechanisms . In addition , innate immune activation gives way , over time , to the development of Th1 lymphocytes and host resistance to protozoa , including T . gondii [4] . TLR2 , TLR4 , TLR9 , and TLR11 have been shown to be important cognate innate immune receptors involved in the recognition of T . gondii derived components , such as glycosilphosphatidylinositol ( GPI ) anchors [8] , CpG DNA [9] and profilin [10] , [11] . However , the deficiency of each of these TLRs , and even the loss of two TLRs , as is the case with TLR2/TLR4 double knockout mice , leads to a relatively minor phenotype after T . gondii infection , as compared to the results obtained with infected MyD88−/− mice [6] . The “endosomal TLRs” , TLR3 , TLR7 , TLR8 and TLR9 , recognize microbial RNA and DNA [12] , [13] , [14] , [15] . In addition to its well-described function in the recognition of CpG motifs , TLR9 has been shown to play an important role in the recognition of parasite DNA and host resistance to infection by several different protozoan parasites [16] , [17] , [18] . However , the combined role of nucleotide-sensing TLRs in host resistance to T . gondii has not been explored . Tabeta and colleagues [19] identified a mutant mouse line by forward genetic screening that they named “3d” , so called because of its deficiency in response to TLR3 , TLR7 and TLR9 ligands ( mouse TLR8 do not respond to single stranded RNA ) . The 3d mouse was shown to have altered UNC93B1 function , an endoplasmic reticulum protein with distant homology to an ion transporter in worms . UNC93B1 is now known to be essential for signaling through mouse TLR3 , TLR7 , and TLR9 , and the consequent production of pro-inflammatory cytokines [19] , [20] . The combined deficiency of nucleic acid-sensing TLRs results in altered host resistance to microbial infections [3] , [19] , [20] . Specifically , UNC93B1 associates and mediates the translocation of the nucleotide-sensing TLRs from the endoplasmic reticulum ( ER ) to the endolysosomal compartment , allowing their proper activation by microbial RNA and DNA [21] , [22] . Here , we show that although mounting a normal systemic pro-inflammatory response , the 3d mice are extremely susceptible to infection with T . gondii . Nevertheless , we provide evidence of a critical role of UNC93B1 in mediating IL-12 as well as early IFNγ production during acute infection with T . gondii . Its well known that active host cell invasion by T . gondii leads to formation of a high pH-parasitophorous vacuole [23] , parasite replication and parasitism , whereas passive internalization of the parasite by phagocytosis results in parasite elimination in the lysosomes [24] , [25] . We also found that in macrophages infected with T . gondii , UNC93B1 translocates to the parasitophorous vacuole , rather than to the predicted phagolysosomes . Finally , our results demonstrate that the lack of functional UNC93B1 results in enhanced tachyzoite replication in macrophages . Altogether , our experiments reveal a role for UNC93B1 on IL-12 production induced by Toxoplasma infection , as well as an unprecedented TLR-independent role for UNC93B1 on host cell control of T . gondii replication , which combined are of central importance for the in vivo resistance to infection with this intracellular protozoan parasite .
We found that the 3d mice are highly susceptible to infection with T . gondii , suggesting the possibility that the combined action of TLR3 , TLR7 and TLR9 is critical for host resistance to infection with T . gondii . However , we used as a control the double MyD88/TRIF-null mice , which are deficient in all TLR responses . While very susceptible , we found that upon infection with T . gondii the double MyD88/TRIF deficient mice were somewhat more resistant than the 3d mice ( Fig . 1A ) . While this difference was not dramatic , it was consistent , suggesting that UNC93B1 may also mediate host resistance to this parasitic infection by a novel mechanism , independent of TLR function . As measured by real-time PCR , enhanced parasite replication was observed in different peripheral organs ( i . e . spleen , liver , and lungs ) ( Fig . 1B ) , despite of unimpaired systemic IFNγ and TNFα production from 3d mice ( Figs . 2A and 2B ) . This is remarkable in view of the long established role for these cytokines in triggering macrophage effector functions and mediating host resistance to T . gondii . In agreement with the real-time PCR data , analysis of the cells collected from the peritoneal cavity of infected animals showed significant higher numbers of infected cells in 3d animals ( Figs . 1C and 1D ) . Phenotypic analysis revealed that 55–75% of the cells recruited to the peritoneal cavity of 3d animals expressed the surface marker CD11b . Of these , 75% were neutrophils ( CD11b+ , Ly6G+ , F4/80− ) and the remaining 25% were inflammatory monocytes ( CD11b+ , Ly6Chigh , Ly6G− , F4/80low ) . Although infected neutrophils were observed , they rarely contained more than two parasites , whereas four or more tachyzoites were typically observed in monocyte/machophage-like cells . Thus , enhanced susceptibility to T . gondii infection was associated with uncontrolled parasite replication within monocyte/macrophage-like cells in peritoneal cavity of 3d mice . In order to further investigate the role of nucleotide-sensing TLRs in resistance to T . gondii , we infected the TLR3−/− , TLR7−/− and TLR9−/− mice , as well as the single knockout mice deficient in each of the two main adaptors required for TLR function , TRIF and MyD88 . Our results showed that except for the MyD88−/− mice , which were very susceptible to infection , all of the other mice lineages had a similar survival curve and cyst numbers , in comparison to wild-type mice ( Fig . 1A and data not shown ) . We anticipated that the lack of endosomal TLR function would result in impaired cytokine production during infection . However , to our surprise , we found that upon infection with T . gondii , except for impaired IL-12p40 production , 3d mice showed high levels of IL-6 , MCP-1 , IFNγ , and TNFα in their sera ( Fig . 2 , A–B ) . Similarly , splenocytes from 3d mice infected with T . gondii demonstrated unimpaired ex-vivo cytokine production ( data not shown ) . While infected MyD88−/− and MyD88/TRIF double deficient mice had decreased serum levels and ex-vivo production of several cytokines ( Fig . 2B and data not shown ) , unimpaired cytokine production of IL-12p40 , IFNγ , TNFα , IL-6 and MCP-1 were confirmed in the sera from TRIF−/− ( Fig . 2C ) , TLR3−/− , TLR7−/− , and TLR9−/− mice ( Fig . S1 ) infected with T . gondii . Together , these initial results suggest that the extreme susceptibility of the 3d mouse to T . gondii infection is due to an additional function of UNC93B1 that is not related to the regulation of endosomal TLRs . T . gondii is a promiscuous parasite that infect any nucleated host cells of both hemopoietic and non-hemopoietic origin . Thus , its replication could be controlled by metabolites secreted by activated macrophages or , alternatively , directly by cytokine-induced microbicidal mechanisms triggered within infected non-phagocytic cells . To distinguish between these two basic mechanisms of cell-mediated immunity , reciprocal bone marrow chimeras were constructed between wild-type and 3d mice and their survival assessed following challenge with T . gondii . The reverse chimeras were generated employing wild type ( B6 . SJL ) and 3d mice , which hemopoietic cells express CD45 . 1 and CD45 . 2 respectively ( Fig . S2A ) . Notably , transplanted mice , which possess hemopoietic cells from 3d mice , became non-responsive to any of the TLR7 and TLR9 agonists , but sustained cytokine response ( TNFα and RANTES ) to LPS and Concanavalin A ( Fig . S2B ) . Finally , infectious challenge of reciprocal chimeras demonstrated that expression of wild type , functional UNC93B1 , in the hemopoietic , but not in the non-hemopoietic compartment was necessary for host resistance to infection with T . gondii ( Fig . 3 ) . These findings are consistent with data indicating the primary expression of UNC93B1 in murine cells from myeloid origin ( http://biogps . gnf . org/#goto=genereport&id=81622 ) . Based on the results obtained with the reverse chimeras , we decided to focus our attention to evaluate the function of lymphoid/myeloid cells from UN93B1 mutant mice . Because UNC93B1 has also been suggested to be involved on antigen presentation and T cell responses [19] , we evaluated the expansion and expression of activation markers on antigen presenting cells ( i . e . , CD11b+ and CD11c+ ) , as well as CD4+ T and CD8+ T cells from mice infected with T . gondii . CD11b+ as well as CD11c+ cells from 3d mice expressed significant amounts of activation markers , e . g . MHC class I and II , CD40 , CD80 and CD86 , which were intermediary between cells from wild type and MyD88−/− mice infected with T . gondii ( Fig . 4 ) . In addition , spleens of 3d mice infected with T . gondii contained 20–40% less CD4+ and CD8+ T cells expressing CD25 , CD69 and CD154 ( Fig . 5A ) . While these differences in wild-type vs . infected 3d mice were statistically significant , the impairment in activation of cells was far more pronounced in MyD88−/− mice ( Fig . 4 and Fig . 5A ) , whose splenocytes showed no signs of expansion after T . gondii infection ( data not shown ) . The percentage of CD4+T as well as CD8+T lymphocytes producing IFNγ was similar when comparing wild type and 3d mice ( Fig . 5B ) , whereas the total number of IFNγ -producing T cells was significantly smaller in infected 3d as compared to wild type mice ( Fig . 5C ) . Consistent with the low serum levels of IL-12 ( Fig . 2 ) and partial impairment on activation of antigen presenting cells ( CD11b+ or CD11c+ cells ) ( Fig . 4 ) , we observed a largely impaired IL-12 production by spleen cells from infected 3d mice ( Fig . 6A , top panel ) . Production of IL-12 was also severely impaired at the infection site ( Fig . 6B , top panel ) . Importantly , bone-marrow derived macrophages from 3d animals exposed to T . gondii tachyzoites in vitro also produce 50% less IL-12 than WT cells ( Fig . 6C ) . Consistent with the delayed production of IL-12 , we also observed a late IFNγ response by spleen or peritoneal exudate cells from 3d mice , which was lower on days three and five , but not on days seven or eight post-infection ( Figs . 6A and 6B , bottom panels ) . IFNγ is thought to be the most critical cytokine in controlling T . gondii replication [2] , and mediates the IL-12 role on host resistance to T . gondii [26] , [27] . Despite the fact that MyD88 knockout and MyD88/TRIF double deficient animals displayed a similar impairment on IL-12 production as the 3d mice ( Figs . 2B and 6B–C ) , and an even more dramatic defect on IFNγ production ( Fig . 2B , and 6B–C ) , the 3d mice consistently showed a more pronounced susceptibility to T . gondii infection ( Fig . 1A ) . Therefore , we believe that the delay of IFNγ production could not be solely responsible for the profound susceptibility phenotype observed on 3d mice . Thus , we sought to determine if cells from 3d mice were properly responding to IFNγ . We observed normal STAT1 phosphorylation , an essential component for IFNγ signaling , in 3d mice infected with T . gondii ( Fig . 6D ) . Furthermore , 3d macrophages responded normally to IFNγ by producing high levels of TNFα , IL-12 , and IL-6 when stimulated in vitro in combination with LPS ( Fig . S3 ) . Similarly , activation with IFNγ resulted in the enhanced expression of MHC I , MHC II and CD40 in macrophages from 3d mice ( Fig . S3 ) . Thus , we found no evidence that the response to IFNγ is affected in mice lacking functional UNC93B1 . Since we found large amounts of parasite within the peritoneal cavity of 3d mice , despite of the high levels of IFNγ , we decided to investigate the ability of macrophages expressing UNC93B1 to control tachyzoite growth in vitro . T . gondii can actively invade host cells or it can be actively internalized by professional phagocytic cells . The fate of the parasite inside the host cell depends on the way the parasite is internalized [24] , [28] . Active invasion normally results in generation of a unique organelle known as the parasitophorous vacuole , which is incompetent to fuse with lysosomes . The parasitophorous vacuole consequently has a high pH and is not stained by LysoTracker [23] . In contrast to active invasion by the parasite , phagocytosis directs the parasite to the phagolysosome , which has a low pH , and leads to elimination of the parasite . We next evaluated the association of UNC93B1 with T . gondii parasites in infected host cells . We began by generating immortalized macrophage cell lines from the UNC93B1 mouse . These lines were then genetically engineered to express either YFP-tagged wild-type UNC93B1 or YFP-tagged UNC93B1H412R ( the non-functional mutant expressed by the 3d mouse ) . Both cell lines expressed high levels of UNC93B1 ( Fig . 7A ) . Macrophage cell lines expressing the wild-type form , but not the mutated UNC93B1H412R , recovered cytokine responses to agonists for TLR9 and TLR7 , respectively ( Fig . 7B ) . In non-activated cells , UNC93B1 was observed as a resident ER protein , and did not co-localize to LysoTracker positive acidic compartments ( Fig . S4 ) . Confocal microscopy revealed that after infection , there was an enrichment of UNC93B1 around the internalized parasites in cell lines that expressed the wild-type , but not the mutated/non-functional protein ( Fig . 7C ) . The results presented in Fig . 7D show a macrophage cell containing two intracellular parasites . One of these parasites was seen in the acidic LysoTracker positive phagolysosome , and was not surrounded by UNC93B1 . Conversely , the parasite found in the parasitophorous vacuole was LysoTracker negative and was surrounded by an intense green ring , indicating a high concentration of UNC93B1-YFP around the parasitophorous vacuole , but not in the predicted phagolysosomes [22] . Immunofluorescence analysis of the parasitophorous vacuoles showed that 92% of internalized tachyzoites present in LysoTracker negative compartments , also positive for the parasite-specific protein GRA7 [29] , were surrounded by a membrane enriched with UNC93B1 . Such enrichment was never observed in acidic organelles containing phagocytosed parasites . We also tested the ability of bone marrow immortalized macrophages derived from wild type and 3d mice to control parasite replication in vitro . Our results show that upon activation IFNγ ( 10 or 100 U/ml ) macrophages from 3d mice were as efficient as the ones derived from wild type mice to control tachyzoite replication as detected by 3H-uracil uptake ( Fig . 8A ) . Notably , the levels of reactive nitrogen intermediates release as measured by nitrite levels ( Fig . S5A ) , induction of Irga6 , Irgm1 and Irgm3 ( Fig . S5B ) , and translocation of Irga6 and Igrb6 to the parasitophorous vacuole ( Fig . S5C ) , all involved on tachyzoite control by IFNγ activated macrophages , were normal in cells from 3d mice . Nevertheless , we consistently observed and enhanced replication of tachyzoites in non-activated macrophages from 3d mice as compared to the same cells derived from wild type mice . Thus , we further explore the possibility that non-activated macrophages from 3d mice are more permissive to tachyzoite growth . Unprimed immortalized macrophages from wild type , 3d and double MyD88/TRIF knockout mice were infected with T . gondii ( ME-49 strain ) expressing luciferase reporter gene at a ratio of one parasite per cell , and parasite growth evaluated 48 hs later , by measuring luciferase activity ( Fig . 8B ) . Our experiments demonstrated that parasite replication was more pronounced in cells from 3d mice ( Fig . 8B ) . In contrast , when we compared parasite replication in macrophages from wild type versus MyD88/TRIF double knockout mice , no difference was observed in parasite growth ( Fig . 8B ) . Finally , we performed a detailed analysis of parasite growth in bone marrow-derived macrophages from wild type and 3d mice , evaluating the numbers of vacuoles per cells and tachyzoites per parasitophorous vacuole ( Figs . 8C and 8D ) . The results of this experiment demonstrate a lower number of parasitophorous vacuoles containing 2 or 4 tachyzoites ( p<0 . 05 ) , and a significantly higher number of vacuoles containing 8 ( p<0 . 05 ) or 16 parasites when comparing macrophages from wild type and 3d mice . The number of parasitophorous vacuoles per infected cell was similar ( 1 . 72±0 . 22 ) in both cell types . Altogether , these results indicate that UNC93B1 mediates host cell resistance to T . gondii through a mechanism that controls parasite replication in the parasitophorous vacuole and is independent from both TLR activation and the anti-parasitic effects of IFNγ .
UNC93B1 is a critical mediator of the translocation of nucleotide-sensing TLR3 , TLR7 and TLR9 from the ER to endolysosomes [22] . Here , we tested the 3d mouse , which has a non-functional UNC93B1 [19] , to evaluate the combined role of nucleotide sensing TLRs in controlling initial activation of innate immunity and host resistance to infection with T . gondii . Despite the fact that none of single TLR3 , TLR7 or TLR9 knockout yield an altered phenotype on cytokine response or enhanced susceptibility , we found that 3d mice are extremely susceptible to infection with T . gondii . Therefore , our results raise the possibility that combined action of nucleotide sensing TLRs is critical for host resistance to T . gondii . Much to our surprise , the MyD88/TRIF null mice , which are devoid of all TLR functions and show impaired production of pro-inflammatory cytokines when infected with T . gondii , were consistently more resistant to infection than the 3d mice . Furthermore , except for IL-12 , 3d mice infected with T . gondii mounted a normal systemic pro-inflammatory response , while the MyD88/TRIF double knockouts did not , indicating that the immunological response to infection was fundamentally different . Nevertheless , animals bearing UNC93B1 mutation succumbed to infection as a result of unchecked tachyzoite replication , similar to IFNγ−/− mice [30] . Thus , while we cannot exclude that the combined action of intracellular TLR 3 , 7 and 9 contributes to host resistance against T . gondii , our hypothesis is that UNC93B1 also mediates host resistance against T . gondii through an additional mechanism , which is TLR-independent . Considering the UNC93B1 involvement on antigen presentation [19] , we first investigated whether antigen presenting cells ( APCs ) , CD4+ T , and CD8+ T lymphocytes were properly activated in 3d mice . Our results show that production of IL-12 as well as expression of activation markers by APCs was significantly impaired in 3d mice infected with T . gondii . Intriguingly , the IFNγ levels were similar in the sera , splenocyte cultures and peritoneal cavity , when comparing 3d and WT mice at day 8 post-infection with T . gondii . While the percentage of T cells producing IFNγ in splenocytes from infected 3d was comparable to infected C57BL/6 mice , the total numbers of IFNγ producing-CD4+ T as well as -CD8+ T cells were lower in de 3d mice . Regardless , mice deficient in CD8+ T cells or in the so-called transporter associated with antigen processing ( TAP-1 ) protein , while more susceptible to infection with T . gondii , often survive 30–40 days post-infection [31] , [32] . Thus , our data suggest that defective antigen cross-presentation and CD8+ T cell activation are not the primary events accounting for the extreme susceptibility of 3d mice to T . gondii infection . Importantly , our in vivo experiments suggest that UNC93B1 is an important mediator of IL-12 production during T . gondii infection . We also addressed this question in vitro and observed that exposure of macrophages from 3d or MyD88−/− mice to live ME49 tachyzoites resulted in impaired production of IL-12 , as compared to macrophages from WT mice . Since IL-12 is a key mediator of IFNγ production during T . gondii infection [26] , [27] , [33] , [34] , it is surprising that IFNγ responses , as discussed above , were close to normal in the infected 3d mice . Therefore , we performed experiments at earlier time points and observed that production of IL-12 and IFNγ was significantly impaired in the peritoneal cavity and peritoneal cavity/spleens from 3d mice at 3 and 5 days post-infection , respectively . These findings could be explained as a result of combined deficiency of nucleotide sensing-TLRs , since no phenotype is observed in each of the single TLR3 , TLR7 or TLR9 knockout mice . Consistently , experiments performed in our laboratory and elsewhere [35] demonstrate that despite of severe impairment IL-12 production in MyD88−/− mice , IFNγ is still produced at 8 days post-infection , and yet , mice are highly susceptible to ME-49 infection . Similarly , different studies demonstrate that except for IL-12 , infection with the highly virulent RH strain of T . gondii elicits elevated levels of pro-inflammatory cytokines , including IFNγ [36] , [37] . However due to the inherent ability of RH parasites to rapidly replicate and disseminate , infected animals are still unable to control parasite burden and die during the acute phase of infection [36] , [37] . Even though the observed defect on IL-12/IFNγ axis seems to be sufficient to render animals more susceptible to T . gondii , our results clearly indicate that an additional TLR independent-function mediated by UNC93B1 contributes for the extreme susceptibility of 3d mice , when compared to MyD88−/− or MyD88/TRIF double knockouts infected with T . gondii . Notably , from in vitro and in vivo experiments we had no evidence that cells from 3d mice have a defect in responding to IFNγ . Thus , we next investigated the ability of host cells from UNC93B1 mutant mice to control parasite replication . The fate of T . gondii inside the host cell relies upon the mechanism of entry . Passive internalization by phagocytosis directs parasites to the lysosomal compartment , leading to tachyzoite elimination [24] , [25] . Upon active invasion , T . gondii establishes itself in the parasitophorous vacuole [38] , [39] , [40] , which avoids fusion with lysosomes , allowing parasite survival . Interestingly , we found that UNC93B1 is recruited from the ER to the parasitophorous vacuole ( PV ) , rather than to the expected endolysosomal compartment [22] . In spite of being considered a non-fusogenic compartment , few host cell proteins are found at the membrane of parasitophorous vacuole containing tachyzoites . Specific proteins appear to be selectively recruited from the host cell plasma membrane [41] , or after host cell activation [42] . In addition , the ER is known to be in close contact with the parasitophorous vacuole membrane [43] , and fusion between the ER and parasitophorous vacuole containing live parasites has been demonstrated [44] . Notably , UNC93B1 was shown to translocate from ER to the endolysosomal compartment upon cell stimulation with TLR agonists [22] . Notwithstanding , here we show that UNC93B1 is recruited from the ER to the parasitophorous vacuoles . It is likely that the recruitment of UNC93B1 to the parasitophorous vacuole membrane occurs during the process of ER fusion , given that UNC93B1 is an ER resident protein [19] , [21] . The transfer of ER proteins to the parasitophorous vacuole membrane seems to be selective , since neither the mutant form of UNC93B1 nor TLR9 ( not shown ) were found to be enriched around the parasitophorous vacuole . Dissociation of the intracellular traffic of UNC93B1 and TLR9 has also been suggested by Ewald and co-workers [45] , as they observed that forced expression of UNC93B1 at the plasma membrane was not accompanied by relocation of TLR9 . Certainly , the process that involves selection or exclusion of specific host proteins , including UNC93B1 and nucleotide sensing TLRs , to the parasitophorous vacuole membrane , is likely to be a key event in the successful establishment of parasitism , and remains to be elucidated . Most of the mechanisms involved in the control of intracellular replication of T . gondii have been studied in IFNγ-activated macrophages . For example the downstream effects of GTPases [46] , [47] , production of reactive nitrogen intermediates [48] , tryptophan degradation in human cells [49] or autophagy [50] , [51] are all IFNγ-inducible mechanisms involved in controlling and/or killing of T . gondii replication . Markedly , we found that in vitro , macrophages from 3d mice present a normal response and are perfectly able to control tachyzoite replication when activated with IFNγ . Consistently , the production of reactive nitrogen intermediates as well as and expression and translocation of IFNγ-inducible GTPases ( i . e . Irga6 , Irgb6 , Irgm1 and Irgm3 ) or formation of autophagic vacuoles ( data not shown ) are not impaired in macrophages from 3d mice activated with IFNγ . Remarkably , our results demonstrate an uncontrolled parasite replication in macrophages from 3d mice infected in vivo with T . gondii , despite an unimpaired IFNγ response . Further , we found that in vitro the 3d mutation renders non-activated macrophages more susceptible to intracellular tachyzoite replication . Even though the difference in parasite numbers is modest , when comparing macrophages from WT and 3d mice , it may reflect large differences in vivo , where multiple rounds of parasite replication during a long period of time will result in exponential parasite growth . To support our interpretation , other studies also show that small but significant differences in parasite replication in vitro , reflects dramatic differences in parasite growth and virulence in vivo [52] , [53] . It is difficult to imagine how this phenomenon could be related to the effects of the UNC93B1 mutation on TLR signaling , since the rates of parasite replication in non-activated macrophages from MyD88/TRIF null mice were similar to that observed in the same cells derived from wild-type mice . Thus , UNC93B1 effects on parasite control appear to be independent of what are thought important immune mediators of host resistance to T . gondii , such as TLRs , IFNγ and TNFα . To establish itself inside a host cell T . gondii has to acquire metabolites from intracellular stores . Indeed , UNC93B1 is a distant ortholog to an ion transporter from Caenorhabditis elegans [54] , [55] , [56] . Despite this homology , a similar function has not been described for UNC93B1 in mammals [19] . Alternatively , nutrient can be acquired from channels present at the parasitophorous vacuole membrane , which allow free diffusion of small metabolites up to 1300 Da [57]; lipids may be acquired through the closely apposed mitochondrial and ER membranes [43]; and parasite seems to exploit the host endolysosomal system via sequestration of host organelles into invaginations present at the parasitophorous vacuole membrane [58] . Therefore , it is also possible that UNC93B1 regulates metabolite/nutrient acquisition by T . gondii tachyzoites , and hence interferes with parasite replication . In conclusion , our study reveals a critical anti-parasitic role for UNC93B1 . This role appears to involved in at least two steps: ( i ) control of IL-12 and early IFNγ response , which may be a result of combined TLR3/TLR7/TLR9 deficiency; and ( ii ) UNC93B1 enrichment in the membranes surrounding the parasitophorous vacuole containing T . gondii tachyzoites , which mediates control of parasite growth in a TLR- and IFNγ-independent manner . Altogether our results indicate that UNC93B1 plays a critical role on innate immune response and host resistance to T . gondii infection .
All experiments involving animals were in accordance with guidelines set forth by the American Association for Laboratory Animal Science ( AALAS ) . All protocols developed for this work were approved by the Institutional Animal Care and Use Committee ( IACUC ) at the University of Massachusetts Medical School . All cell culture reagents were obtained from Mediatech , unless otherwise indicated . LPS derived from Escherichia coli strain 0111:B4 was purchased from Sigma and re-extracted by phenol chloroform to remove lipopeptides as described [59] . Pam2CysSer ( Lys ) 4 was obtained from EMC Microcollections . R848 , a synthetic small molecule agonist for TLR7 was provided by 3 M Pharmaceuticals . Phosphorothioate-stabilized unmethylated DNA oligonucleotide-bearing CpG ( ODN 1826 , 5′-TCCATGACGTTCCTGACGTT-3′ ) and qPCR primers were obtained from Integrated DNA Technologies . Interferon- γ was purchased from R&D Systems . C57BL/6J ( CD45 . 2+ ) , B6 . 129SF2/J ( CD45 . 2+ ) and B6 . SJL-PtprcaPepcb/BoyJ ( CD45 . 1+ ) mice were obtained from The Jackson Laboratory . UNC93B1 mutant ( 3d ) mice were generated by Dr . Bruce Beutler at The Scripps Research Institute , La Jolla , California [19] . TLR3 , TLR7 , TLR9 , MyD88 and TRIF deficient mice were provided by Dr . Shizuo Akira ( Department of Host Defense , Osaka University , Osaka , Japan ) . Mice deficient in both MyD88 and TRIF were generated by interbreeding single knockout animals . Except for the TRIF−/− mice , which are F5 , all mice used were backcrossed to C57BL/6 background at least for 8 generations . Age ( 5–8 weeks old ) and sex matched groups of wild-type ( WT ) and knockout mice were used in all experiments . Mice were bred and housed under specific pathogen-free conditions at the University of Massachusetts Medical School animal facilities . ME-49 tachyzoites were initially obtained by inoculating brain homogenate containing cysts from C57BL/6 mice infected one month earlier onto fibroblast monolayers . After the emergence of tachyzoites , parasites were maintained in human foreskin fibroblast monolayers by weekly passages as described [60] . Luciferase and GFP expressing ME-49 and RH parasites [61] were a kind gift from Dr . Jeroen P . J . Saeij ( MIT , Cambridge , MA ) . Animals were inoculated intraperitoneally ( i . p . ) with 25 cysts obtained from brain homogenates of 6–8 weeks infected mice . Mice were monitored for survival or sacrificed at 0 , 5 and 8 days post infection in order to collect peritoneal exudate cells and fluid , blood and spleens . Samples from each mouse were individually processed and analyzed . Peritoneal exudate cells , spleen , liver , lung and brain from mice at days 0 , 3 , 5 or 7 after infection were collected and frozen in liquid nitrogen . Total DNA was extracted using the DNeasy Blood and Tissue kit ( Qiagen ) according to the manufacturer's instructions , and quantified using a NanoDrop Spectrophotometer ( Thermo Scientific ) . Primers used for amplification of Toxoplasma gondii B1 gene ( Forward: 5′- CTGGCAAATACAGGTGAAATG-3′; Reverse: 5′- GTGTACTGCGAAAATGAATCC-3′ ) were designed using the PrimerSelect application ( Lasergene software suite , DNASTAR ) . PCR reactions were setup in a final 20 µl volume using 5 ng of total tissue DNA , 200 ng of each primer and 1x of the iQ SYBR Green Supermix ( BioRad ) . Quantitative RT-PCR analysis was performed on a DNA Engine Opticon 2 Real-Time Cycler ( MJ Research ) . Specificity of amplification was assessed for each sample by melting curve analysis . Non-infected samples gave no signal . Relative quantification was performed using standard curve analysis of purified parasite DNA , and results were expressed as pg of parasite DNA per mg of total tissue DNA . Cells were stained for 30 min with conjugated antibodies against the surface markers CD3 , CD4 , CD8 , CD11b , CD25 , CD40 , CD69 , CD80 , CD86 , CD154 , MHC I or MHC II ( eBioscience ) . For intracellular measurement of cytokines splenocytes were cultivated for 4 h in presence of GolgiPlug ( BD Bioscience ) , surface stained , permeabilized and incubated with Phycoerythrin-anti- IFNγ or TNFα for 30 min . Subsequently cells were washed and analyzed by flow cytometry in an LSRII cytometer ( BD Bioscience ) . Fluorescent cell lines were analyzed without staining . Data were acquired with DIVA software ( BD Bioscience ) and analyzed with FlowJo ( Tree Star ) . Recipient B6 . SJL-PtprcaPepcb/BoyJ and 3d mice were given lethal total body irradiation ( 900 rads ) and reconstituted intravenously with 5–10 million bone marrow cells within 4 h . Marrow cell suspensions were prepared from donor tibial and femoral bones by flushing with phosphate-buffered saline using a 30-gauge needle syringe . Irradiated and reconstituted mice were given 150 mg/ml Sulfamethoxazole and 30 mg/ml N-Trimethoprim in their drinking water for 6 weeks . Thereafter , they were switched to sterile drinking water , thus ensuring that the antibiotic treatment would not affect the ensuing experimental infection with T . gondii . Mice were used for experimental infection or for analysis of chimerism 7–9 weeks after transplant . Animals showed full reconstitution of lymphoid and myeloid cell populations as determined by flow cytometric analysis ( not shown ) . Spleens were homogenized , RBC lysed ( Red Blood Cell Lysis buffer , Sigma ) , and splenocytes were re-suspended in complete RPMI medium . Cells were cultured at 5×106/well in 24-well tissue culture plates in absence of exogenous stimuli , and supernatants were collected 24 h later . Splenocyte , macrophage culture supernatants or peritoneal exudates were assayed for pro-inflammatory cytokines with DuoSet ELISA kits from R&D Systems according to the manufacturer's instructions . Reactive nitrogen intermediates were measured by the Griess reaction of nitrites accumulated in the supernatants [62] with chemicals from Sigma . Cytokines present in mouse serum were assayed using the BD Cytometric Bead Assay ( CBA ) Mouse Inflammation kit according to the manufacturer's instructions . Splenocytes collected from either non-infected animals or from mice infected for 5 or 8 days were lysed , the extract re-suspended in Laemmli sample buffer [63] and boiled for 5 min at 95°C . Samples were separated by 10% SDS-PAGE and were transferred onto nitrocellulose membranes [64] . Blots were incubated with mouse monoclonal antibodies against STAT-1 ( C-terminus ) or STAT-1P ( BD Biosciences ) and subsequently incubated with HRP-conjugated anti-mouse IgG ( Bio-Rad ) . Membranes were then incubated with HRP substrate ( enhanced chemiluminescence substrate; Amersham Biosciences ) and developed by exposure to film ( Hyperfilm; Amersham Biosciences ) . Bone marrow-derived macrophages ( BMDMs ) were isolated as described [65] , and were cultured in RPMI medium supplemented with 25 mM Hepes , 10 mM L-glutamine , 100 U/ml Penicillin-Streptomycin , 50 µM 2-mercaptoethanol ( Sigma ) and 10% FCS ( Hyclone ) . Immortalized macrophage cell lines were generated as described [66] , [67] . Briefly , primary bone marrow cells were incubated in L929 mouse fibroblast-conditioned medium for 3–4 days for the induction of macrophage differentiation . Subsequently , cells were infected with J2 recombinant retrovirus carrying v-myc and v-raf ( mil ) oncogenes [68] . Growth factors were removed from the culture medium and cells were maintained until they were growing in the absence of conditioning medium . Macrophages phenotype was verified by surface expression of the markers CD11b ( M1/70 , BD Pharmigen ) and F4/80 ( BM8 , eBioscience ) as well as a range of functional parameters , including responsiveness to Toll-like ligands . Macrophage cell lines were generated from wild-type ( C57BL/6 ) , 3d and TLR9-deficent mice . Cells were treated with the indicated stimuli or infected in a multiplicity of infection ( MOI ) of 5 , and supernatants were collected 24 h later . Murine UNC93B1 ( BC018388 ) was C-terminally fused with YFP and cloned into pcDNA3 ( Invitrogen ) . The point mutant UNC93B1 ( H412R ) was generated by sequential PCR with primers carrying the point mutation CAC ( His ) to CGC ( Arg ) . YFP-tagged UNC93B1 wild type and H412R were then cloned into a retroviral vector that was modified from the original pCLXSN backbone from Imgenex ( Fig . S6 ) . All of the constructs were verified by sequencing . Recombinant retroviruses were produced as described [69] . Briefly , human embryonic kidney ( HEK293T ) cells were co-transfected with the vectors encoding YFP-tagged UNC93B1 and plasmids carrying the retroviral gag-pol genes and the envelope protein VSV-G using the GeneJuice transfection reagent ( Novagen ) according to the manufacturer's instructions . The virus containing supernatants were filtered and used to infect immortalized 3d or TLR9-deficient cells . ME-49 tachyzoites were stained with the Cell Tracker Red CMTPX ( Invitrogen ) according to the manufacturer's instruction . Immortalized macrophages were infected in a multiplicity of infection ( MOI ) of 3 . After 2 h , unbound parasites were washed off and live cells were imaged by confocal microscopy at 37°C . Acidic intracellular compartments were stained with the acidophilic lysomotropic dye LysoTracker Blue ( Invitrogen ) . To study the kinetics of parasite growth , BMDMs were infected with GFP-expressing ME-49 tachyzoites in a MOI of 1 . After 32 hours cells were fixed and processed for immunofluorescent staining as described below . Number of GRA7-positive vacuoles per cell , and of parasites per vacuole , were evaluated in 10 randomly selected microscopic fields , and at least 100 vacuoles were counted per sample . In order to quantify number of vacuoles positive for interferon-induced GTPases ( IRGs ) , immortalized macrophages were induced overnight with 200 U/ml IFNγ before infection . Cells were fixed and processed for immunofluorescence . Intracellular parasites were identified by GRA7 staining , and the percentage of IRG-positive vacuoles was determined after analysis of 200 vacuoles . Murine immortalized macrophages were stimulated with IFNγ ( BD Pharmigen ) at 10–100 U/ml as indicated for 24 h prior to infection while control cultures were left untreated . Cells were then infected with ME-49 tachyzoites at a multiplicity of infection of 1 . After 24 h of incubation , cultures were labeled with 1 µCi/well [3H]-uracil for additional 24 h . The amount of incorporated uracil was determined by liquid scintillation counting [70] . Alternatively , immortalized macrophages were infected with luciferase-expressing parasites . After 48 h samples were lysed , cell lysates were mixed with a Luciferase reporter assay system substrate ( Promega ) and luciferase units calculated by normalizing the raw luminescence values to the background from non-infected cells . Cells were fixed with paraformaldehyde 4% for 15 min at room temperature , washed with PBS , permeabilized with saponin 0 . 1% in PBS for 10 min and incubated with primary antibodies for 1 h . Subsequently cells were washed and incubated with Alexa conjugated secondary antibodies for an additional 1 h . Host and parasite DNA was stained with DAPI . Slides were mounted with Gel-mount anti-fading reagent ( EMS ) and analyzed by confocal microscopy . Primary antibodies used: α-GRA7 antiserum ( a kind gift from Dr . George Yap , New Jersey Medical School , Newark , NJ ) , rabbit α-Irga6 antiserum 165°3 [71] , rabbit anti-Irgb6 antiserum 141/1 ( raised against bacterial purified full length protein , unpublished ) , α-Irgm1 ( Santa Cruz Biotechnology ) , and α-Irgm3 ( BD Pharmigen ) . Secondary antibodies: Alexa 633 goat α-rabbit , Alexa 488 donkey α-goat and Alexa 488 goat α-mouse conjugated antibodies ( Invitrogen ) . We used an inverted Axiovert 100-M microscope equipped with a Zeiss LSM 510 META scanning unit and a 1 . 4 NA 63x plan apochromat objective ( Zeiss ) , and an inverted Leica LSM TSC SP2 AOBS . Cells were cultured on glass-bottom 35-mm tissue-culture dishes ( Matek ) . Dual or triple color images were acquired by consecutive scanning with only one laser line active per scan to avoid cross-excitation . The statistical significance of the differences in the means of experimental groups was determined by one-way or two-way ANOVA analysis and Bonferroni post-test using GraphPad Prism 5 . 0a Software .
|
One third of the human population in the world is chronically infected with Toxoplasma gondii . While the majority of infected individuals are asymptomatic , toxoplasmosis is a major cause of congenital disease , abortion , and a life-threatening opportunistic disease in immunocompromised individuals . Early activation of the innate immune system and cytokine production by myeloid cells is required for establishment of protective immunity to T . gondii infection . In mice , a mutation in the UNC93B1 gene abolishes signaling via the intracellular innate immune receptors , namely Toll-like receptors ( TLR ) 3 , 7 and 9 , thus , named triple-deficiency ( 3d ) mice . Our results demonstrate that the hyper-susceptibility of 3d mice to T . gondii infection is associated with impaired IL-12 production , delayed IFNγ production , and uncontrolled parasite replication in macrophages . Overall , our study reveals a critical role for UNC93B1 in the immunological control of T . gondii infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"infectious",
"diseases/protozoal",
"infections",
"immunology/immunity",
"to",
"infections",
"microbiology/parasitology",
"immunology/innate",
"immunity"
] |
2010
|
UNC93B1 Mediates Host Resistance to Infection with Toxoplasma gondii
|
Early bacterial surface colonization is not a random process wherein cells arbitrarily attach to surfaces and grow; but rather , attachment events , movement and cellular interactions induce non-random spatial organization . We have only begun to understand how the apparent self-organization affects the fitness of the population . A key factor contributing to fitness is the tradeoff between solitary-planktonic and aggregated surface-attached biofilm lifestyles . Though planktonic cells typically grow faster , bacteria in aggregates are more resistant to stress such as desiccation , antibiotics and predation . Here we ask if and to what extent informed surface-attachments improve fitness during early surface colonization under periodic stress conditions . We use an individual-based modeling approach to simulate foraging planktonic cells colonizing a surface under alternating wet-dry cycles . Such cycles are common in the largest terrestrial microbial habitats–soil , roots , and leaf surfaces–that are not constantly saturated with water and experience daily periods of desiccation stress . We compared different surface-attachment strategies , and analyzed the emerging spatio-temporal dynamics of surface colonization and population yield as a measure of fitness . We demonstrate that a simple strategy of preferential attachment ( PA ) , biased to dense sites , carries a large fitness advantage over any random attachment across a broad range of environmental conditions–particularly under periodic stress .
Early bacterial surface colonization that takes place prior to the development of mature biofilm , is a critical stage during which cells attempt to establish a sustainable population [1] . There is growing evidence that this is not a random process wherein cells arbitrarily attach to surfaces and grow to form microcolonies . Relocation and detachment of cells were shown to play a major role in bacterial colonization on leaves [2 , 3] . A “rich get richer” process was observed during the surface colonization of P . aeruginosa , where some aggregates ( cell clusters ) are enriched by biased surface movement of cells controlled by a web of secreted polysaccharide [4] , and surface-attachment of planktonic cells was shown to be biased toward lower distances to previously attached cells [5] . Thus , early surface colonization seems to be a self-organized process , resulting from the behavior of individual-cells . Yet , we know very little of how such individual-cell behavior and the emergent self-organization affect the fitness of the population . A key factor in overall population fitness is the inherent tradeoff between the planktonic and biofilm lifestyles [6–9] . Planktonic cells typically grow faster by eluding the costly production of extracellular polymeric substances ( EPS ) that is common in biofilms [10 , 11] and by avoiding the high cell density and limited diffusion within aggregates [12] . Aggregates on the other hand , provide protection from various stresses including desiccation , antibiotics , and predation [13–18] ( Fig 1A ) . Often , protection from stress within aggregates is collective and is a function of the size of the aggregate , with a minimal size required to achieve protection . For example , protection from desiccation on plant leaf surfaces was shown to be gained in aggregates above ~100 cells [19] ( Fig 1B ) . The decision of a planktonic cell to attach to the surface and change its lifestyle is a function of several factors including environmental cues , surface properties , and its physiological state [20–22] . These individual cells’ decisions play an important role in their proliferation and survival , as a consequence of the aforementioned tradeoff . Since individual-cell behavior modulates aggregate formation dynamics , it affects the fitness of the collective through self-organization . Therefore , the mechanisms governing these decisions are expected to be under strong selection [7 , 8 , 23 , 24] . In many microbial habitats stress conditions are not continuous but vary with time , often in a periodic manner . The largest terrestrial microbial habitats–soil , plant root and leaf surfaces–are not constantly saturated with water , and experience daily recurring wetting and drying dynamics [25 , 26] . Cells on plant leaf surfaces and shallow water of marine environments are periodically exposed to strong UV radiation at mid-day [27] . Bacterial grazing in the marine environment can be periodic over the diel cycle [28] . Human microbiota are likely exposed to daily variation in immune system activity [29] or in antibiotic levels [30] . Thus , periodic stress is likely very prevalent during bacterial surface colonization . Temporal variation in stress levels often favors a division of labor between cell phenotypes , i . e . reproductive specialists and survival specialists ( e . g . planktonic and biofilm cells , respectively ) [9] . The tradeoff of planktonic and biofilm phenotypes has been studied in several contexts [7] , including the role of life-style transitions in periodically fluctuating environments [31] , and both analytic and individual-based models were used to explore the mechanisms that optimize fitness and balance the population composition [9 , 32] . Because aggregate size plays a role in survival , it can be beneficial to accelerate the development of aggregates , especially when growth rate is too low to support the formation of protected aggregates in the intervals between stress periods . Since clonal growth rate is constrained by cell physiology and environmental conditions including nutrient availability , an alternative way to accelerate aggregate growth is by recruitment of other cells; either by movement of nearby , already surface-attached , cells ( as observed by Zhao et . al [4] ) , or by attachment of foraging planktonic cells to , or near , the aggregate . We term such self-organizing mechanisms that enrich or promote aggregation Aggregate-Enrichment Mechanisms ( AEMs ) . One such mechanism is preferential attachment ( PA ) , which we use as a general term for attachment pattern that is biased towards attachment to existing aggregates , and thus depends on the spatial organization of sessile bacteria on the surface . A PA process is a general stochastic growth process wherein individuals join groups in a system in a “biased” non-random way . PA has been shown to be common in many real-world complex systems such as networks where new nodes are preferentially attached to nodes with high connectivity [33] . PA has been demonstrated to explain the degree distribution of scale-free networks including many biological and social networks [34] . We hypothesize that PA plays an important role in early bacterial surface colonization under environments that select for collective protection , through the self-organization induced by biased individual surface-attachment decisions . Here we ask if and to what extent PA strategies can provide fitness advantage in environments with periodic stress . We use an individual-based modeling approach [35 , 36] to simulate early bacterial surface colonization under diel cycles of alternating wet and dry periods , confronting cells with periodic desiccation stress . Our simulations allowed us to compare different PA strategies with random strategies , to analyze the emerging spatio-temporal dynamics of surface colonization , and evaluate population yield as a measure of fitness .
To study the fitness advantage of PA under periodic stress , we developed an individual-based model of bacterial surface colonization in fluctuating hydration conditions . The model consists of two layers: the fluid and the surface inhabited by planktonic and sessile ( i . e . biofilm ) bacterial cells , respectively ( Fig 1C ) . Planktonic cells are motile while sessile cells are not ( except when passively moved ) . Nutrients are assumed to diffuse into the system from a constant concentration source . Cells grow and divide in accordance with the available nutrient levels and the cells' state , die as a function of stress level and cell-density , and may attach to the surface or detach from the surface ( Fig 1D ) . Cells' growth parameters are typical for environmental bacteria that colonize the leaf surface [37] ( Tables 1 and 2 , Eq 1 ) . Importantly , our model employs the tradeoff between growth and survival that is associated with the planktonic and biofilm lifestyles ( Fig 1 ) . Here , sessile cells devote part of their resources to the production of EPS and thus grow slower than planktonic cells ( see Methods ) . The effect of stress is modeled by the hourly probability of cell death ( S ) that is a function of both the local cell density and the hydration conditions . S of solitary cells and cells within small aggregates is low ( SB ) during wet periods and high during the dry periods ( SL ) . Cells in larger aggregates ( densities >QL ) gain protection and thus have low death probability ( SB ) during both wet and dry periods . To account for nutrient deprivation and toxin buildup at the centers of established biofilms , S increases to SH at densities higher than QH ( Fig 1E , Tables 1 and 2 , Eq 2 ) . To model attachment , we used a probability function which describes the chances of planktonic cell attachment . The attachment probability per time step ( A ) of a planktonic cell depends on the local cell density of surface-attached bacteria , defined as the number of sessile cells in 10μm-radius neighborhood . Cells that do not enact PA , and rely on random attachment ( RA ) alone , have a constant ARA attachment probability . The attachment probability function for cells with PA is a step-like function , as described in Fig 1F . Other probability functions were tested , but did not show significant advantage over the step-like function ( See S1 Fig ) . When a planktonic cell with PA encounters local sessile-cell densities above QPA , it has a higher attachment probability ( APA > ARA ) . Last , in our model , detachment ( biofilm-to-planktonic transition ) rate is constant and equals D = 0 . 01 [h−1] ( Tables 1 and 2 , Eqs 3 and 4 ) . Our simulation begins with 100 planktonic cells and an empty surface . A simulation lasts five diel cycles , with alternating 12-hour long wet and dry periods . Fitness is assessed by the final population size ( population yield ) after the last dry period . Typical overall population dynamics , with RA at a rate equal to the detachment rate ARA = D , are shown in Fig 2A . After the initial growth the population oscillates periodically , in accordance with the inbound nutrient flux and the stress-induced death rate , determined by hydration conditions and the spatial organization of cells on the surface . To study the impact of PA strategies on the dynamics of surface colonization , we compared simulations with a range of PA strategies ( represented by different threshold values QPA ) as well as simulations with RA with a range of ARA rates . We then compared the mean population size , i . e . yield , at the end of a five diel cycles simulation , as a measure for fitness . Remarkably , we find that under the simulated conditions , there are PA strategies that confer a significant fitness advantage ( i . e . a higher yield ) over any random attachment rate . Moreover , the optimal strategy is at an intermediate QPA value , and fine tuning of QPA has a large effect on fitness ( Fig 2B ) . Snapshots of the final simulation state for the optimal PA and RA simulations are shown in Fig 2C and 2D . Two representative simulations are shown in S1 Video . Different QL values ( size required for protection ) had an impact on the value of the optimal attachment threshold ( the higher the QL , the higher the optimal QPA ) but the overall picture remained similar ( Fig 3 ) . To examine the dynamics of the surface colonization process and the mechanism that confers fitness advantage to PA over RA , we tracked the population partition to planktonic and biofilm sub-populations and the distribution of aggregate sizes of the biofilm sub-population . We first compared RA simulations with various attachment rates ARA ( Fig 4D–4F ) . Even with the highest ARA , where most of the cells were sessile , no large stress-protected aggregates formed ( Fig 4F ) . This is in contrast with the optimal PA strategy that produced large stress-protected aggregates ( Fig 4B ) . These large aggregates were established on day 1 , though not reaching the size required for protection ( i . e . QL = 40 cells ) until day 2 . Notably , during day 2 these aggregates grew massively and reached a size of hundreds of cells–way over the required protection-size–and persisted until the end of the simulation . To understand how large aggregates form and persist under the optimal PA strategy , we analyzed the lineage composition of all aggregates . Aggregates that were enriched by attachments of planktonic cells are expected to be composed of many lineages , as opposed to clonally growing aggregates which are all descendants of a single founder cell . Large aggregates formed in PA simulations with the optimal QPA value are indeed composed of many lineages ( Fig 4B ) , in contrast to the smaller unprotected aggregates formed by RA simulations , mostly composed of the founder cell progeny ( Fig 4D–4F ) . An in-depth comparison between PA with the optimal attachment threshold ( QPA = 12 ) and PA with a low QPA value ( QPA = 4 ) , revealed that under a low threshold more composite aggregates that did not reach protected density were generated . Some protected aggregates were formed , but they were much smaller and less stable than with the optimal QPA . The total number of aggregates for QPA = 4 is higher , but since a smaller fraction of cells belongs to large protected aggregates , the total population yield is lower ( Fig 4A ) . On the other hand , higher thresholds QPA = 20 did not produce protected aggregates at all ( Fig 4C ) . We also analyzed the frequency of attachment events per local cell density ( Fig 5 ) . With RA most attachments are to vacant locales ( 0 . 79±0 . 01 , 0 . 89±0 . 01 , 0 . 99±0 . 01 mean±SE for high , balanced and low attachment rates , respectively ) . With low QPA most attachments are to very small aggregates and are thus inefficient in generating protected aggregates . PA facilitated the formation of protected aggregates , but their growth and contribution to survival are restricted without a sufficient flux of attaching cells . With optimal QPA attachments below QL are to larger and fewer aggregates , some of them eventually yield protection; and attachments above QL enrich already protected aggregates . With high QPA there are hardly any density-dependent attachments . To examine the ecological relevance of the PA mechanism , and to find under what conditions PA confers fitness advantage , we performed a series of simulations that compared PA and RA at a range of nutrient concentrations ( Nc ) and a range of death probabilities of unprotected cells ( SL ) ( Fig 6 ) . We first ran RA simulations with a range of attachment rate values ( ARA ) . The emerging picture is described in Fig 6A: there is an area of the phase plane that led to extinction , and the rest of the plane is roughly partitioned into an area where lower attachment rates ( preference for planktonic lifestyle , area in Green , Fig 6A ) are superior , and an area where higher attachment rates ( preference for biofilm lifestyle , area in Red , Fig 6A ) are superior . Next , we ran the same series of simulations , this time with PA with a range of QPA values , and compared the results of the optimal PA with the best RA strategy of the same point in the parameter space . We found that there exists a large area on the phase plane where PA is advantageous ( area in Blue , Fig 6B ) . For a given SL , increasing values of Nc results in transitions in the outcome . At low SL values ( ≤ 0 . 6 [h−1] ) , a first transition between extinction and survival occurs when Nc is high enough to replenish the planktonic sub-population after a dry period ( Fig 6B , zone c ) . In zone c , PA and RA do equally well . A second transition occurs when Nc is high enough to enable the optimal PA to outperform any RA by enriching the nascent aggregates and granting them protection ( Fig 6B , zone b ) . At high SL values ( > 0 . 6 [h−1] ) unprotected survival is prohibited , thus high enough Nc values allow only PA strategies to survive ( Fig 6B , zone a ) , thereby extending the Hutchinsonian niche [41] beyond the niche occupied by RA . Notably , in nearly all cases there exists some threshold QPA which can confer PA at least the same fitness as the best RA . What happens if stress is constant and not periodic ? To answer this question , we repeated the analysis under constant stress conditions . We found that the region of the phase plane where PA is better than RA is smaller and the mean advantage that PA confers is lower compared to periodic stress conditions: under constant stress , the most significant relative advantage of PA occurs where Nc ≥ 30[g/m3] and SL ≤ 0 . 26[h−1] , and the mean relative advantage of PA in this region is 1 . 8 ± 0 . 1; while under periodic stress , the mean relative advantage for the analogous conditions ( i . e . equivalent survival chances for unprotected cells , SL ≤ 0 . 45[h−1] ) is 3 ± 0 . 25 ( mean±SE ) ( Fig 6C ) . In addition , under stronger constant stress SL ≥ 0 . 33[h−1] , PA does not lead to survival and does not extend the Hutchinsonian niche of RA . This highlights the advantage of PA strategies over RA under periodic , rather than constant , stress . Last , we asked how changes in the duration of the wet and dry periods over the diel cycle affect our results . We found that , under the studied conditions , PA confers advantage over all range of wet period lengths ( 4h to 20h per day ) . Moreover , we find that the optimal QPA changes with the length of the wet period: the longer the wet period the higher is the optimal QPA ( Fig 7 ) . This is because as the duration of the wet period increases , larger aggregates are formed clonally , and a higher QPA allocates less planktonic cells to the aggregates that will likely not reach the protected size .
In this work we used individual-based modeling to study early bacterial surface colonization under periodic stress . Importantly , we reflected in our simulations the tradeoff between growth and survival of planktonic and biofilm lifestyles . We show how surface-attachment strategies of individual cells influence self-organization during colonization on a surface and most importantly how this in turn impacts fitness . We find that across a wide range of conditions under periodic stress , a simple strategy , such as preferential attachment ( PA ) to existing aggregates above a given size , carries a large fitness advantage over any random attachment ( RA ) strategy . The improvement in fitness is achieved by a more optimal partition between the planktonic and biofilm sub-populations and precise modulation of the aggregate-size distribution dynamics , both controlled by the attachment decisions of individual-cells . The early phases of bacterial surface colonization have a crucial spatio-temporal aspect . When only a small fraction of the surface is colonized and cells are not uniformly distributed over the surface , the time and location of surface-attachment of planktonic cells affects the fitness and survivability of the individual cell , as well as the nearby population . PA modulates the aggregate-size distribution via the attachment threshold QPA . The allocation of planktonic cell attachments to aggregate growth can be optimized by fine-tuning the QPA . The optimal QPA results with the largest part of the population protected from stress by carrying aggregates beyond the critical size ( Fig 8 ) . The temporal periodicity of stress plays a major role in the relative advantage of PA over RA . Under constant stress , environments harsh enough to attenuate aggregate growth throughout the colonization process will subsequently render PA ineffective . Without the temporal variation of stress , PA does not extend the Hutchinsonian niche of the system . It is important to note that PA is not expected to be beneficial under all conditions and all combinations of variables and parameters . The parameter space in our model is huge and it is unfeasible to thoroughly scan it . We demonstrate that under some reasonable assumptions and environmentally relevant conditions , PA can provide fitness advantage over RA . Generally , PA can be a result of both passive and active mechanisms . An example of a passive mechanism is the reported biased surface-movement toward dense areas controlled by the sticky trails of polysaccharide [4] . Active mechanisms may involve chemotactic movement toward aggregates [42] or informed attachment decisions like the one we modeled here . Nevertheless , other–both passive and active–mechanisms can equally well lead to a PA process where aggregates are enriched in a rich-get-richer manner and thus may confer fitness advantage in environments that select for collective protection . In our model we assumed that information on local cell density is available to foraging planktonic cells . There are several possible mechanisms that can facilitate the attachment of planktonic cells to denser locations . Quorum-sensing ( QS ) signals in early surface colonization on natural unsaturated surfaces such as plant root and leaf surfaces are highly localized , and quorum size can be surprisingly small: even as low as 10 cells [43 , 44] . Other sensing systems , such as peptidoglycan sensing by bacteria , can serve as indicators of cell proximity [45] . Adherence to Psl trails [4] is another AEM that increases the probability of cells’ irreversible attachment to existing aggregates , biased towards larger established aggregates . In this study we considered homogenous populations ( i . e . all cells are from a single species and employ the same strategy ) and we measured fitness by population yield . This allowed us to quantify the effect that individual-cell behavior has on the collective fitness , that would have been disrupted ( because of interactions between competing strategies ) if we chose to rely on evaluating fitness based on competition simulations that are commonly used to estimate relative fitness and to study evolutionary dynamics [46] . Preliminary results of competition simulations between populations that employ different PA or RA strategies showed various interesting dynamics , including commonly known evolutionary scenarios such as co-existence , cooperation , exploitation , cheating , and invasion . For example , populations with the optimal PA strategy can be exploited by invaders with non-optimal higher PA threshold , that can join established protected aggregates later and benefit form a longer planktonic phase at a higher growth rate . The protected aggregates of the optimal PA strategy have a lower relatedness in comparison to the mostly clonal aggregates created by RA and high threshold PA strategies ( Fig 4 and S2 Fig ) , and thus PA is not expected to be an Evolutionary Stable Strategy [47 , 48] . These intriguing initial results demonstrate the richness of the studied system and open the way for further research which was beyond the scope of the current study . The present study demonstrates the impact that PA may have on bacterial fitness during early surface colonization , particularly under periodic stress . The general principles of the model are applicable to different stresses such as desiccation , antibiotics and predation . Our modeling results , together with evidence for PA processes in bacterial surface colonization in both experimental and natural systems [4 , 5] , call for further study of PA and similar processes and their contribution to collective protection in microbial ecosystems .
Our Individual Based Model is based on the Repast Platform [49] . The simulation domain is a two-dimensional 1mm by 1mm square that is comprised of a bulk liquid phase and a surface phase . The individuals are bacterial cells , either planktonic or sessile . Planktonic cells move at random . Sessile bacteria are static and move only by physical shoving to avoid overlaps , with aggregates consisting of a single layer of cells . A single nutrient resource is consumed by the cells , and replenished by diffusion into the domain from an infinite source . The cells’ growth rate , actions , and death probability are determined by their state and environmental conditions , i . e . nutrient concentration , stress , and local surface density , as described in Tables 1 and 2 . All simulations begin with 100 planktonic cells , and a nutrient concentration in equilibrium with the source . For a detailed description of the model , see S1 Text .
|
A vast portion of bacterial life on Earth takes place on surfaces . In many of these surfaces cells collectively organize into biofilms that are known to provide them protection from various environmental stresses . Early bacterial colonization of surfaces , prior to the development of mature biofilm , is a critical stage during which cells attempt to establish a sustainable population . It is not a random process wherein cells arbitrarily attach to surfaces and grow to form micro-colonies . Rather , surface-attachments , movement and cellular interactions take place to yield non-random organization . Using computer simulations , based on individual-based modeling , we demonstrate that simple attachment strategies , where planktonic cells preferentially attach to existing surface-attached aggregates , may confer fitness advantage over random attachment . The advantage of preferential attachment is particularly substantial under periodic stress–a common characteristic of many natural microbial habitats . This is due to a more efficient recruitment of planktonic cells that accelerates the formation of stress-protected aggregates .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"bacteriology",
"biofilms",
"cell",
"death",
"invertebrates",
"plant",
"anatomy",
"cell",
"processes",
"microbiology",
"animals",
"plant",
"science",
"mathematics",
"population",
"biology",
"probability",
"density",
"leaves",
"biophysics",
"probability",
"theory",
"physics",
"population",
"metrics",
"bacterial",
"biofilms",
"population",
"size",
"eukaryota",
"cell",
"biology",
"plankton",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"organisms",
"biophysical",
"simulations"
] |
2019
|
Bacterial surface colonization, preferential attachment and fitness under periodic stress
|
Small RNAs—including piRNAs , miRNAs , and endogenous siRNAs—bind Argonaute proteins to form RNA silencing complexes that target coding genes , transposons , and aberrant RNAs . To assess the requirements for endogenous siRNA formation and activity in Caenorhabditis elegans , we developed a GFP-based sensor for the endogenous siRNA 22G siR-1 , one of a set of abundant siRNAs processed from a precursor RNA mapping to the X chromosome , the X-cluster . Silencing of the sensor is also dependent on the partially complementary , unlinked 26G siR-O7 siRNA . We show that 26G siR-O7 acts in trans to initiate 22G siRNA formation from the X-cluster . The presence of several mispairs between 26G siR-O7 and the X-cluster mRNA , as well as mutagenesis of the siRNA sensor , indicates that siRNA target recognition is permissive to a degree of mispairing . From a candidate reverse genetic screen , we identified several factors required for 22G siR-1 activity , including the chromatin factors mes-4 and gfl-1 , the Argonaute ergo-1 , and the 3′ methyltransferase henn-1 . Quantitative RT–PCR of small RNAs in a henn-1 mutant and deep sequencing of methylated small RNAs indicate that siRNAs and piRNAs that associate with PIWI clade Argonautes are methylated by HENN-1 , while siRNAs and miRNAs that associate with non-PIWI clade Argonautes are not . Thus , PIWI-class Argonaute proteins are specifically adapted to associate with methylated small RNAs in C . elegans .
MicroRNAs ( miRNAs ) , PIWI-interacting RNAs ( piRNAs ) and small interfering RNAs ( siRNAs ) are distinct classes of ∼20–30 nt regulatory RNAs . Each acts as a guide to direct an Argonaute-containing effector complex to target mRNAs [1] . The features required for small RNA-target interactions and the regulatory outcomes of these interactions are largely dictated by the Argonaute cofactor . There are three distinct clades within the Argonaute family [2] . miRNAs associate with Argonautes in the AGO clade [1] , [3] , whereas piRNAs associate with members of the PIWI clade [1] , [4] , [5] . siRNAs associate with PIWIs and AGOs in a variety of eukaryotes as well as several Argonautes in the expansive WAGO clade found only in nematodes [1] , [2] , [6]–[10] . Most eukaryotes contain multiple classes of small RNAs and Argonaute cofactors and thus require specialized mechanisms for sorting small RNAs and their target transcripts into the proper pathways [1] . Small RNA duplex structure , 5′ nt identity and length are important determinants for sorting small RNAs into specific effector complexes , although these features alone fail to account for some interactions [1] . In C . elegans , piRNAs ( also called 21U RNAs ) are broadly distributed throughout the genome but derive primarily from two clusters on chromosome IV [11] . They are almost exclusively 21 nt and contain a 5′U [11] . At least some piRNAs are modified at their 3′ ends , presumably by 2′-O-methylation [10] , [11] . The PIWIs PRG-1 and PRG-2 are the only proteins that have been shown to function in the C . elegans piRNA pathway . The specific roles of piRNAs in development are unclear , but mutations in prg-1 cause developmental defects including failure in spermatogenesis , abnormal germline development and sterility at elevated temperatures [4] , [5] , [12] . The only validated target of the piRNA pathway is the Tc3 DNA transposon family [4] , [5] . Increased Tc3 transposition may partially account for the defects observed in prg-1 mutants . Endogenous siRNAs are processed from thousands of distinct loci , including transposons , pseudogenes and protein coding genes [7] , [13] . There are two types of endogenous siRNAs in C . elegans: 22G siRNAs which are 22 nt and bear a 5′ triphosphorylated guanine and 26G siRNAs which are 26 nt and bear a 5′ monophosphorylated guanine [14] . Processing of 26G , but not 22G siRNAs , requires the endoribonuclease Dicer [9] , [10] , [15]–[17] . Cleavage by Dicer generates RNAs containing 5′ monophosphates , whereas the nascent transcripts of RNA dependent RNA polymerases ( RdRPs ) are predicted to bear 5′ triphosphorylated nucleotides; this may account for the difference in 5′ phosphorylation state between 26G and 22G siRNAs . In addition to differences at their 5′ ends , siRNAs also differ at their 3′ ends , with a subset presumably having a 2′-O-methyl group [10] , [11] . Both 26G and 22G siRNA formation requires an RNA-dependent RNA Polymerase , but it is unclear if the nascent RdRP product is further processed to accommodate association with the ∼20 to 30 nt cleft of an Argonaute . 26G siRNAs function as primary siRNAs to initiate formation of the more abundant secondary 22G siRNAs from target transcripts; however , the majority of 22G siRNAs are processed independent of a 26G siRNA trigger [8] , [10] , [18] . 26G and 22G siRNAs can be further classified by their Argonaute binding partners . 26G siRNAs associate with the AGO clade Argonautes ALG-3 and ALG-4 during sperm development or with the PIWI clade Argonaute ERGO-1 during embryo development [8]–[10] . 22G siRNAs associate with either CSR-1 to direct chromosome segregation or WAGO-1-WAGO-11 to guide RNA silencing [7] , [19] , [20] . At least a subset of 22G siRNAs also associate with the Argonaute NRDE-3 to block RNA polymerase II activity at target loci within the nucleus [21] , [22] . To identify the requirements for routing transcripts into RNA silencing pathways , we developed a GFP based sensor for endogenous siRNA activity in C . elegans . The responses of the siRNA sensor indicate that a single siRNA target site is sufficient to route a transcript into an RNA silencing pathway involving NRDE-3 . Mutagenesis of the sensor siRNA target site revealed that siRNA target recognition and silencing of the sensor is permissive to some degree of mispairing . Additionally , we identify an endogenous gene that is targeted in trans by a partially complementary 26G siRNA to trigger 22G siRNA formation . Finally , from a candidate RNAi screen for gene inactivations that results in desilencing of the siRNA sensor , we identified the C . elegans HEN1 ortholog henn-1 . Together with Billi et al . [23] and Kamminga et al . [24] , we show that henn-1 is required for proper accumulation of both piRNAs and siRNAs that associate with PIWIs , but not for miRNAs and siRNAs that associate with AGO or WAGO clade Argonautes .
To identify the requirements for siRNA directed RNA silencing , we developed a GFP based sensor for endogenous siRNA activity in C . elegans . The siRNA sensor ubl-1::GFP-siR-1-sensor contains a single target site for an abundant endogenous siRNA , 22G siR-1 , embedded in the 3′ UTR of ubiquitin-like1 ( ubl-1 ) and expressed under the control of the ubl-1 promoter , which is presumably active in all tissues throughout development ( Figure 1A ) . A control construct , ubl-1::GFP , lacks the siRNA target site , but is otherwise identical ( Figure 1A ) . Each construct was introduced into C . elegans using Mos1-mediated single copy insertion [25] . GFP expression was ubiquitous in C . elegans containing the control , which lacks the 22G siR-1 target site , but was nearly absent in C . elegans containing the reporter with the 22G siR-1 sensor element in the 3′ UTR ( Figure 1B ) . 22G siR-1 is derived from a cluster of 22G siRNAs on the X chromosome ( termed the X-cluster [26] ) that are dependent on ERGO-1 class 26G siRNA pathway components for their formation [10] . Thus , silencing of the siRNA sensor was predicted to require ergo-1 and other factors essential for ERGO-1 class 26G siRNA activity , as well as factors required for 22G siRNA formation and activity . To test this , RNAi against ergo-1 and several other validated and suspected RNAi factors was done in ubl-1::GFP-siR-1-sensor-transgenic C . elegans . GFP expression was derepressed in C . elegans containing the siRNA sensor when treated with RNAi against ergo-1 and each of the other validated factors tested [7] , [9] , [10] , [13] , [27] ( Figure 1C and Table 1 ) . RNAi against several other factors implicated in RNAi [28] , including the chromatin factors mes-4 and gfl-1 , the ubiquitin ligase ncl-1 , the transcription elongation regulators tcer-1 and R03D7 . 4 and the spliceosome factor rnp-2 , also derepressed the siRNA sensor ( Table 1 ) . RNAi against many of the factors analyzed , including mutator ( mut ) class genes , causes desilencing of multicopy array based transgenes [29]; conceivably , the siRNA sensor , although a single copy transgene , is reporting on this phenomenon . However , loss of eri-6 or ergo-1 activity enhances silencing of tandem array transgenes and would therefore be expected to decrease GFP expression if the siRNA sensor was reporting on transgene desilencing [30] . In fact , eri-6 and ergo-1 were two of the strongest derepressors of the siRNA sensor , indicating that it is not reporting on transgene desilencing ( Table 1 and Figure 1C ) . These results indicate that the genetic requirements for silencing the siRNA sensor reflect those of endogenous siRNA targets . We assessed GFP mRNA and protein levels to identify the mode by which the siRNA sensor is silenced . GFP mRNA levels were much lower in the siRNA sensor strain than in the control strain , as determined by RNA blot assay ( Figure 1D ) . RNAi against ergo-1 in C . elegans containing the siRNA sensor caused substantial increases in both GFP mRNA and protein levels ( Figure 1E ) . GFP protein and mRNA levels were proportionally elevated ∼8 fold in siRNA sensor-transgenic C . elegans treated with ergo-1 RNAi relative to control RNAi ( p = 0 . 00003 and p<0 . 00001 , respectively; Figure 1F ) , indicating that translational repression does not substantially contribute to GFP silencing . To determine if silencing of the siRNA sensor occurs cotranscriptionally via the nuclear RNAi pathway involving the Argonaute NRDE-3 , we introduced the siRNA sensor or the control transgene into nrde-3 mutant C . elegans . GFP expression from the siRNA sensor in the nrde-3 mutant was derepressed to a level comparable to that of the control transgene , while GFP expression from the control transgene was unchanged between wild type and nrde-3 mutants ( Figure 1G and Figure S1 ) . Thus , NRDE-3-mediated cotranscriptional gene silencing is the primary mode by which the siRNA sensor is silenced . Exogenous RNAi is initiated by low abundance primary siRNAs that recruit RdRPs and other factors to trigger formation of more abundant secondary siRNAs [31]–[33] . Endogenous ERGO-1 class 26G primary siRNAs are also expressed at relatively low levels compared to secondary 22G siRNAs derived from the same loci . Thus , an important role of at least some classes of siRNAs is to trigger siRNA amplification and spreading outside of the primary siRNA target site . To determine if 22G siRNAs trigger production of siRNAs in the genomic vicinity of the initial target site , we deep sequenced small RNAs from C . elegans containing either the ubl-1::GFP or ubl-1::GFP-siR-1-sensor transgene . siRNAs derived from both the control and siRNA sensor transgene were predominantly 22 nt and contained 5′G ( Figure 2A ) . The normalized siRNA levels ( reads per million total small RNA reads ) derived from the GFP mRNA were indistinguishable between the control and siRNA sensor strains ( Figure 2B ) . siRNAs were uniformly distributed across both transgenes and were derived exclusively from coding and vector sequence and not from the ubl-1 5′ and 3′ untranslated regions ( Figure 2C and 2D ) . Although a large peak was observed at the siRNA target site of the sensor , it likely corresponds to 22G siR-1 and its derivatives originating from the endogenous X-cluster siRNA locus , as the levels of 22G siR-1 were identical between control- and siRNA sensor-transgenic C . elegans ( Figure 2D , inset blot panel ) . These results suggest that , unlike primary exogenous siRNAs and endogenous 26G siRNAs , 22G siRNAs that function in the nrde-3 pathway do not trigger siRNA amplification or spreading outside of the siRNA target site . Furthermore , that siRNAs were formed from the GFP control construct that lacks an siRNA target site suggests that , even when introduced as single copies , transgenes are still subjected to siRNA surveillance . The degree of sequence complementarity required for target recognition by miRNAs is relatively well characterized . Near perfect complementarity is required in the seed sequence ( positions 2–8 of the miRNA , relative to its 5′ end ) , but generally not in the central or 3′ regions [34] . However , little is known about the requirements for siRNA target recognition , particularly in C . elegans . To determine the sequence requirements for target recognition of the siRNA sensor by 22G siR-1 , the target site was mutated to contain 1–3 mispairs or a single deletion or insertion , relative to 22G siR-1 , at various positions along the target sequence ( Figure 3A ) . When introduced into C . elegans , mutations in the sensor that prevented or interfered with basepairing at the 5′ end of 22G siR-1 ( ubl-1::GFP-siR-1-sensor-1-3sub , -4-5sub , and -4del ) , which includes the region analogous to the seed sequence of miRNAs , resulted in GFP expression similar to what was observed the control that lacks an siRNA target site ( Figure 3A and 3B ) , indicating that near perfect complementarity is required between the 5′ end of an siRNA and its target for efficient silencing . Argonaute catalyzed endonucleolytic cleavage typically occurs between positions 10 and 11 on the target mRNA , relative to the 5′ end of the small RNA guide; mispairs at or near these positions inhibits cleavage [35] . We were unable to detect cleavage within the siRNA target site of the endogenous 22G siR-1 target transcript using 5′ RACE ( Figure S2 ) . Furthermore , most Argonautes that associate with 22G siRNAs in C . elegans , including NRDE-3 , lack the conserved RNase H residues required for catalytic activity [2] . However , when we mutated positions 9–11 ( ubl-1::GFP-siR-1-sensor-9-11sub ) we did observe a modest increase in GFP expression from the siRNA sensor transgene ( Figure 3A and 3B ) , indicating that these positions do play a role in siRNA target recognition . Basepairing in the region 3′ of the bulged nucleotides of a miRNA , at positions 12–17 , can enhance miRNA target recognition [34] , suggesting that these positions could play an important role in target recognition by siRNAs . Three mispairs introduced at positions 12–14 of the siRNA target site of ubl-1::GFP-siR-1-sensor ( ubl-1::GFP-siR-1-sensor-12-14sub ) resulted in derepression of the siRNA sensor to a level similar to that of the control ( Figure 3A and 3B ) . When we introduced a single mispair at position 13 we did not observe an increase in the levels of GFP expression ( Figure 3A and 3B ) . Deletion of the paired nucleotide at position 13 ( ubl-1::GFP-siR-1-sensor-13del ) , which would require the siRNA to loop out to accommodate binding to the 3′ end of the siRNA , resulted in only a very modest increase in GFP expression from the siRNA sensor . Introduction of a single nucleotide at position 13 ( ubl-1::GFP-siR-1-sensor-13ins ) , which would require the mRNA to loop out at a single position somewhere between positions 13–15 to facilitate pairing with the 3′ end of the siRNA , caused partial derepression of the siRNA sensor ( Figure 3A and 3B ) . Finally , to determine if pairing at the 3′ terminus of the siRNA is required for target recognition , we introduced three mispairs at positions 20–22 ( ubl-1::GFP-siR-1-sensor-20-22sub ) of the siRNA target site within the siRNA sensor ( Figure 3A and 3B ) . GFP expression from the ubl-1::GFP-siR-1-sensor-20-22sub transgene was similar to that of the wild type siRNA sensor , indicating that basepairing at these positions is not essential for siRNA target recognition ( Figure 3A and 3B ) . Although this study does not provide a comprehensive analysis of siRNA target recognition requirements , it demonstrates that a certain degree of mispairing is permissible for siRNA target recognition in C . elegans . 22G siR-1 and other 22G siRNAs derived from the X-cluster are dependent on the 26G siRNA pathway components , although the locus itself does not produce 26G siRNAs [10] . The X-cluster locus is unannotated but inspection of mRNA deep sequencing data [36] indicates that siRNAs are derived from an ∼5 kb transcript produced directly upstream of an annotated coding gene , however , the annotated gene itself lacks evidence for transcription ( Figure 4A ) . 22G siR-1 is the most abundant siRNA produced from the locus and is processed from a motif that is repeated multiple times within the cluster ( Figure 4A ) . Given our finding that siRNAs do not require perfect complementarity for target recognition , we hypothesized that 22G siRNA formation from the X-cluster is initiated by a 26G siRNA derived from a distinct gene . To search for such an siRNA trigger , we aligned 26G siRNAs identified in a deep sequencing library enriched for ERGO-1 class 26G siRNAs [37] to the X-cluster transcript . We identified a 26G siRNA , 26G siR-O7 , derived from the gene K02E2 . 11 that aligns with >69% nt complementarity at seven positions within the X-cluster region ( Figure 4B and 4C ) . Aside from the 26G siR-O7 sequence , K02E2 . 11 does not share significant similarity to the X-cluster region . Interestingly , 26G siR-O7 aligns to the same repeated motif that gives rise to 22G siR-1 and shares perfect complementarity between positions 1–10 and 14–19 , aside from 2 G∶U pairs , and is mispaired at positions 11–13 , relative to the 5′ end of the siRNA ( Figure 4C ) . If 26G siR-O7 is indeed required for siRNA formation from the X-cluster , deleting its genomic locus should result in loss of 22G siR-1 . To test this , we generated a partial deletion of the gene K02E2 . 11 , that includes the sequence that gives rise to 26G siR-O7 , using Mos1-mediated deletion [38] ( Figure 4B ) . As predicted , the K02E2 . 11 deletion resulted in complete loss of 26G siR-O7 as well as 22G siR-1 , but not other 26G or 22G siRNAs ( Figure 4D and 4E ) . When introduced into the siRNA sensor strain , the K02E2 . 11 deletion resulted in derepression of GFP fluorescence but did not affect GFP fluorescence from the control strain that lacks an siRNA target site ( Figure 4F ) . Thus , we conclude that 26G siR-O7 triggers 22G siRNA formation from the X-cluster , indicating that endogenous siRNAs can act in trans to regulate endogenous genes . Because of the similarity between the 22G siR-1 target site within the siRNA sensor and the 26G siR-O7 target sites within the X-cluster , conceivably 26G siR-O7 could directly target the siRNA sensor ( Figure 4G ) . To rule out this possibility we introduced the siRNA sensor or the control transgene into either an rde-2/mut-8 or rrf-1 mutant . The rde-2 mutation does not affect 26G siRNA levels , in particular 26G siR-O7 , but it does result in a substantial , although not complete , loss of 22G siR-1 [13] ( Figure S3 ) . rrf-1 is an RNA-dependent RNA polymerase ( RdRP ) that produces 22G siRNAs , but it is not required for 26G siRNA formation [7] , [9] . An rrf-1 mutation by itself does not result in complete loss of 22G siRNAs due to redundancy with the RdRP ego-1 [7] . When introduced into either an rde-2 or rrf-1 mutant , GFP fluorescence from the siRNA sensor was substantially elevated relative to wild type , while GFP fluorescence from the control transgene was indistinguishable between rde-2 or rrf-1 mutants and wild type ( Figure 4H ) . Furthermore , as described above , NRDE-3 , which associates specifically with 22G siRNAs [21] , is also required to silence the siRNA sensor ( Figure 1G ) . Thus , although we cannot entirely rule out a modest or temporal primary contribution of 26G siR-O7 , our data indicates that the siRNA sensor directly reports on 22G siRNA activity and indirectly on 26G siRNA activity . In C . elegans , piRNAs and at least a subset of 26G siRNAs are modified at their 3′ ends , presumably by 2′-O-methylation , a common modification to small RNAs [39]–[44] . An ortholog of the 3′ methyltransferase HEN1 required for small RNA methylation [39] has not been described in C . elegans . The protein encoded by C02F5 . 6 is the only C . elegans gene with significant homology to Arabidopsis ( p = ∼5×10−20 ) and Drosophila ( p = ∼2×10−17 ) HEN1 proteins and is thus a likely ortholog . To determine if C02F5 . 6 is required for siRNA function , C . elegans containing the ubl-1::GFP-siR-1-sensor transgene were treated with RNAi against C02F5 . 6 ( hereafter referred to as henn-1 , where the extra n in the name indicates that it is the nematode ortholog of HEN1 ) . When treated with henn-1 RNAi , a modest increase in GFP fluorescence was observed in C . elegans containing the siRNA sensor transgene , but not in C . elegans containing the control transgene that lacks an siRNA target site ( Table 1 and Figure 5A ) . henn-1 RNAi resulted in a modest increase in GFP protein levels in the siRNA sensor strain but not in the control strain ( Figure 5B; data shown for one of three biological replicates ) . When introduced into a strain containing a mutation in henn-1 ( pk2295 ) that presumably results in a truncated protein due to a premature stop codon [45] , the siRNA sensor yielded GFP protein and fluorescence levels similar to C . elegans containing the control transgene ( Figure 5C and 5D; data shown for one of three biological replicates ) . These results suggest that henn-1 is required for the activity of 22G siR-1 , although possibly by affecting 26G siR-O7 , the 26G siRNA that triggers 22G siR-1 formation . HEN1 is required for the stability of siRNAs in Arabidopsis and Drosophila [42] , [46] . To determine if henn-1 is required for the accumulation of piRNAs , miRNAs or siRNAs , RNA blot and qRT-PCR assays were done on RNA isolated from embryo , L4 larval and adult stage C . elegans . We also assessed by qRT-PCR the levels of several siRNA and one piRNA target mRNAs . In embryos , the level of the piRNA 21UR-2921 was substantially reduced in henn-1 mutants , relative to wild type C . elegans ( Figure 6A; data shown for one of three biological replicates ) . As determined by qRT-PCR , the levels of three other piRNAs ( 21UR-1 , 21UR-3442 and 21UR-3502 ) were reduced by ∼60–80% in henn-1 mutants , relative to wild type ( p<0 . 0002; Figure 6B ) . The requirement for henn-1 in piRNA stabilization is likely dependent on the developmental stage , as the levels of 21UR-1 were only modestly reduced in adults and unaffected in L4 stage henn-1 mutants , relative to wild type ( Figure S4 ) . The levels of two ERGO-1 class 26G siRNAs , 26G siR-O1 derived from C40A11 . 10 and 26G siR-O2 derived from E01G4 . 7 , were depleted by ∼72% ( p<0 . 00001 ) and 45% ( p = 0 . 03 ) , respectively , in henn-1 mutants , relative to wild type ( Figure 6A and 6B ) . Modest reductions in 26G siR-O1 and 26G siR-O2 levels were also observed in adult staged C . elegans ( Figure S5 ) . We also observed a modest reduction in the levels of 26G siR-O7 in henn-1 mutants , as determined by RNA blot assays ( Figure 6A; data shown for one of three biological replicates ) . The levels of 22G siR-1 , which is dependent on ergo-1 and 26G siR-O7 for its formation , were depleted by ∼80% in henn-1 , relative to wild type ( p<0 . 00001; Figure 6A and 6B ) . An ergo-1-dependent 22G siRNA derived from E01G4 . 5 was also depleted in henn-1 mutants ( Figure S5 ) . In contrast , the levels of a 22G siRNA derived from fkb-8 , which is not downstream of 26G siRNAs , were indistinguishable between henn-1 and wild type ( Figure 6A ) . We also examined miR-35 and miR-58 using RNA blot assays . The levels of both miRNAs were unchanged between henn-1 mutant and wild type C . elegans ( Figure 6A; data shown for one of three biological replicates ) . Consistent with the reduced levels of ERGO-1 class 26G siRNAs , the levels of three ERGO-1 class 26G siRNA target mRNAs , C40A11 . 10 , E01G4 . 7 and E01G4 . 5 , were elevated ∼2–3 fold in henn-1 mutants , relative to wild type ( p<0 . 0008; Figure 6C ) . The levels of two transposon mRNAs analyzed , Tc1 and Tc3 , were unchanged in henn-1 mutants ( p>0 . 8; Figure 6C ) . Both Tc1 and Tc3 are targets of 22G siRNAs that are not dependent on 26G siRNAs . However , Tc3 is also the only validated piRNA target and its levels are modestly elevated in the absence of piRNAs [4] , [5] . That henn-1 mutants did not display elevated levels of Tc3 was somewhat puzzling . It is possible that there is residual activity of piRNAs in the absence of henn-1 , which is consistent with the incomplete loss of piRNAs in henn-1 mutants . In henn-1 mutant L4 larvae , which are enriched for ALG-3/4 class 26G siRNAs , the levels of three miRNAs ( miR-1 , miR-35 and miR-58 ) and an ALG-3/4 class 26G siRNA ( 26G siR-S5 ) derived from ssp-16 were each indistinguishable from wild type ( Figure 6D and 6E ) . In contrast , 22G siR-1 , which is expressed throughout development , was depleted similar to what was observed in embryos ( Figure 6E ) . The levels of three ALG-3/4 target mRNAs , C04G2 . 8 , ssp-16 and ZC168 . 6 , were modestly depleted in henn-1 mutants in two independent experiments ( Figure 6F ) . Mutations in prg-1 , the PIWI Argonaute that associates with piRNAs , result in reduced fertility , particularly at 25°C [4] , [5] . To determine if henn-1 mutants also display defects associated with reduced piRNA activity , the brood sizes of wild type and henn-1 mutants grown at either 20°C or 25°C were measured . At 20°C , a modest , but significant reduction in brood size was observed in henn-1 mutants ( p<0 . 00001; Figure 6G ) . At 25°C , henn-1 mutants were nearly sterile , whereas wild type animals had only a modest reduction in brood size relative to those grown at 20°C ( Figure 6G ) . The reduced fertility of henn-1 mutants is likely caused by defects in piRNA activity and not ERGO-1 class 26G siRNA activity because ergo-1 mutants do not display obvious fertility defects [10] . Taken together , these results suggest that henn-1 is specifically required for the accumulation and activity of piRNAs , ERGO-1 class 26G siRNAs and ergo-1-dependent 22G siRNAs . The reduction in ergo-1-dependent 22G siRNAs in henn-1 mutants could be an indirect effect caused by reduced levels of the ERGO-1 class 26G siRNAs that trigger their formation . To comprehensively identify methylated small RNAs in C . elegans and to determine if henn-1 is specifically required for methylated small RNAs , we deep sequenced both β-eliminated and untreated small RNAs isolated from wild type C . elegans . β-elimination is a chemical treatment that removes the 3′ nucleotide of RNAs that contain a 2′-OH but not those that contain a 2′-O-methyl at the 3′ end , and leaves behind a 2′-P at the 3′ end which is incompatible with adapter ligation [47] . Thus , β-elimination can be used to enrich for methylated small RNAs in deep sequencing libraries [48] . Nearly every annotated piRNA was enriched and nearly every miRNA was depleted in the β-eliminated library , relative to the non-treated library ( Figure 7A ) . ERGO-1 class 26G siRNAs were enriched in the β-eliminated library , whereas ALG-3/4 class 26G siRNAs were depleted ( Figure 7A ) . The levels of normalized reads corresponding to piRNAs and ERGO-1 class 26G siRNAs were ∼10 fold greater in the β-eliminated library relative to the non-treated library ( Figure 7B ) . Each of the other classes of small RNAs was depleted in the β-eliminated library ( Figure 7B ) . 22G siR-1 yielded ∼1270 normalized reads ( reads per million total ) in the non-treated library and ∼257 normalized reads in the β-eliminated library , amounting to an ∼80% depletion of 22G siR-1 following β-elimination , indicating that 22G siR-1 is not methylated and thus indirectly affected by mutations in henn-1 ( Figure S6 ) . Interestingly , the methylated small RNAs , that is , piRNAs and ERGO-1 class 26G siRNAs , associate exclusively with Argonautes that are in the PIWI clade , while all other small RNAs in C . elegans are not methylated and associate with AGO and WAGO clade Argonautes ( Figure 7C ) . Therefore , we conclude that HENN-1 specifically methylates small RNAs that associate with PIWIs in C . elegans . In Drosophila , small RNAs that interact with perfect complementarity to target RNAs are subjected to trimming ( 3′-5′ shortening ) and tailing ( untemplated nucleotide additions ) which marks them for degradation [49] . 3′ end methylation protects small RNAs from trimming and tailing in Drosophila and Arabidopsis [46] , [49] . Each class of siRNAs in C . elegans interacts with perfect or near perfect complementarity to their targets , whereas miRNAs generally interact with only partial complementarity , particularly at the 3′ end . It is unclear how piRNAs interact with their targets in C . elegans . We assessed which classes of small RNAs are tailed and trimmed in C . elegans by analyzing our deep sequencing libraries . miRNAs and piRNAs displayed relatively low proportions of trimmed and tailed sequences ( Figure 7D ) . In contrast , each class of siRNAs showed relatively high proportions of trimmed and tailed sequences , although CSR-1 class 22G siRNAs and both classes of 26G siRNAs displayed the highest proportions ( Figure 7D ) . Uridylation of certain siRNAs promotes their association with CSR-1 , which at least partially explains the high levels of trimming and tailing observed for this class of siRNAs [20] . It is interesting that although ERGO-1 class 26G siRNAs are presumably methylated , they are still subject to trimming and tailing at levels similar to the non-methylated ALG-3/4 class 26G siRNAs ( Figure 7D ) .
We developed a GFP-based sensor for endogenous siRNA activity in C . elegans . Using the siRNA sensor , we determined that endogenous 22G siRNAs , at least those that are dependent on nrde-3 , do not trigger siRNA amplification or spreading from the target site and that a certain degree of mispairing is permissible for effective siRNA target recognition . We also show that 22G siRNA formation from an endogenous mRNA is initiated by a trans active 26G siRNA . This phenomenon is reminiscent of the trans-acting siRNA pathway in plants and the miR-243 pathway in C . elegans , in which one or more miRNAs or siRNAs trigger siRNA amplification from a distinct mRNA [50]–[53] . These findings are important to our understanding of RNA silencing pathways for two reasons . First , that endogenous siRNAs require only partial complementarity to their targets suggests that the hundreds of thousands of endogenous siRNAs in C . elegans have a multitude of potential targets distinct from the genes from which they are processed . Secondly , because our results suggest that endogenous 22G siRNAs do not trigger siRNA amplification , the effects of off targeting may be negligible for all but the most abundant 22G siRNAs , as well as the 26G siRNAs . From a candidate screen for endogenous siRNA factors , we identified a requirement for the C . elegans HEN1 ortholog henn-1 in a specific endogenous siRNA pathway . Small RNA analysis in henn-1 mutants and deep sequencing of methylated small RNAs revealed that ERGO-1 class 26G siRNAs and piRNAs are both methylated by HENN-1 . Secondary 22G siRNAs that depend on ERGO-1 class 26G siRNAs also require henn-1 , albeit indirectly , for their biogenesis . In Drosophila , small RNA methylation prevents degradation of small RNAs perfectly basepaired to their targets [49] . It is somewhat puzzling that although all siRNAs share perfect complementarity to their targets in C . elegans one class requires methylation but the others do not . One possibility is that only ERGO-1 class 26G siRNA and piRNAs actually interact perfectly with their targets . Perhaps the 3′ ends small RNAs are more easily liberated from the PIWI PAZ domains than from the AGO or WAGO PAZ domains , which accommodate the 3′ ends of small RNAs [54]–[56] , to interact with their targets . In this model , PIWI-associated methylated small RNAs bound at their 3′ ends to target mRNAs would be protected by the 3′-2′-O-methyl group , while AGO- and WAGO-associated small RNAs would remain anchored to the PAZ domain and therefore inaccessible to nucleases . This might also explain why trimming and tailing levels are similar for ERGO-1 and ALG-3/4 class 26G siRNAs – both are equally protected , but by different means . Perhaps in the absence of HENN-1 , ERGO-1 class 26G siRNAs would be hyper trimmed and tailed . Given that only small RNAs that associate with PIWIs require henn-1 , we propose that PIWIs are specifically adapted to associate with 3′-2′-O-methylated small RNAs and perhaps also with HENN-1 in C . elegans . An intriguing , but highly speculative possibility is that methylation is used as a sorting determinant to direct certain small RNA-Argonaute interactions . In vitro , the PAZ domains of the human PIWI clade Argonautes Hili and Hiwi preferentially bind methylated small RNAs , whereas the PAZ domain of a human AGO clade Argonaute Ago1 preferentially binds small RNAs lacking a 3′-2′-O-methyl group [56] , [57] . In animals , PIWIs associate with methylated small RNAs , while non-PIWI clade Argonautes associate with non-methylated small RNAs , with one exception: methylated siRNAs in Drosophila associate with the AGO clade Argonaute Ago2 [42] . In C . elegans , methylation of ERGO-1 class 26G siRNAs may prevent them from associating with ALG-3 and ALG-4 and lack of methylation on ALG-3/4 class 26G siRNAs may in turn prevent them from associating with ERGO-1 . This model does conflict somewhat with findings in Drosophila that small RNAs are methylated only when bound to their Argonaute binding partner [42] , but one could imagine that other features of the small RNA tag it for methylation before Argonaute loading and then upon loading methylation occurs . The presence or absence of methylation would then dictate whether or not the 3′ end of the small RNA is stabilized within the Argonaute PAZ domain or if the small RNA is discarded . ERGO-1 class 26G siRNAs function during oogenesis and trigger formation of 22G siRNAs that persists into adulthood [9] , [10] , [18] , while piRNAs function during germline and sperm development [4] , [5] , [12] . Therefore , henn-1 is likely to have important roles in RNA silencing pathways throughout C . elegans development . It will be important to learn why henn-1 effects only specific siRNA pathways and why its activity seems to be dispensable for piRNA stabilization except at specific developmental stages .
The ubl-1 upstream and downstream regulatory sequences were amplified from N2 genomic DNA using Phusion polymerase ( Finnzymes ) and the primers attB1-ubl-1p F and attB4-ubl-1p R or attB3-ubl-1u F and attB2-ubl-1u R . GFP was PCR amplified from plasmid DNA with the primers attB4r-GFP F and attB3r-GFP R . The 22G siR-1 target site was introduced by PCR into the ubl-1 3′ UTR using the primers X-motif-ubl-1u F and attB2-ubl-1u R . 22G siR-1 target site mutations were introduced by PCR using various forward primers in combination with attB2-ubl-1u R ( Table S1 ) . To generate the K02E2 . 11 mosDEL construct an ∼2 . 4 kb sequence of homology to K02E2 . 11 and sequence immediately downstream was PCR amplified from N2 genomic DNA using the primers attB1-K02E2 . 11 LH F and attB4-K02E2 . 11 LH R . A 2 kb sequence adjacent to the Mos1 insertion site in ttTi18384 was PCR amplified with attB3-K02E2 . 11 RH F and attB2-K02E2 . 11 RH R from genomic N2 DNA . The unc-119 rescue transgene was amplified from C . briggsae genomic DNA using attB4r-Cbr-unc-119 F and attB3r-Cbr-unc-119 R . PCR products were cloned into pDONR entry vectors using Gateway BP recombination ( Invitrogen ) . Entry vectors were recombined into pCFJ178 or pCFJ151 modified to contain Gateway Pro LR recombination sites ( pCMP2 and pCMP1 , respectively ) . Constructs were sequence verified for accuracy . GFP constructs were introduced into C . elegans strain EG5003 using Mos1-mediated single copy insertion [25] . The K02E2 . 11 knockout construct was introduced into IE18384 , which carries the Mos1 insertion ttTi18384 , using Mos1-mediated deletion [38] . The henn-1 mutant strain , NL4415 , contains the pk2295 allele; the rrf-1 mutant strain , NL2098 , contains the pk1417 allele; and the rde-2 mutant strain , NL3531 , contains the pk1657 allele [45] . The nrde-3 mutant strain , WM156 , contains the tm1116 allele . Each of the strains developed in this study are listed in Table S2 . All primer sequences are listed in Table S1 . GFP antibody ( Invitrogen , A-11122 and A-11034 ) and DAPI staining were done as described [58] . All imaging was done on a Zeiss AxioImager . Z1 Microscope . RNA was isolated from synchronized embryos , L4 larvae or adult C . elegans using Trizol ( Invitrogen ) followed by chloroform extraction and isopropanol precipitation . RNA samples were normalized to 1 . 0 or 2 . 0 ug/ul prior to blot assays , qRT-PCR assays and deep sequencing . Protein was extracted from synchronized L4 larvae using Laemmli buffer and normalized by Actin and the number of animals . For small RNA Northern blots , 10 ug total RNA was separated on 17% denaturing polyacrylamide gels , transferred to positively charged Nitrocellulose membranes , crosslinked and probed with 32P-labeled LNA-modified ( siRNA and piRNA probes ) or unmodified ( miRNA probes ) DNA oligonucleotides antisense to each of the small RNAs analyzed ( Table S1 ) . For GFP mRNA blots , 2 ug total RNA was separated on denaturing 1 . 5% Agarose gels , transferred to positively charged nitrocellulose membranes , crosslinked and probed with a randomly labeled ∼450 bp GFP DNA fragment . For Western blots , proteins were resolved on 4–12% Bis-Tris SDS polyacrylamide gels , transferred to nitrocellulose membranes and probed with GFP or Actin antibodies ( Invitrogen , A-11122 and A-11034; Abcam , ab3280 ) . Protein levels were quantified on a Typhoon phosphorimager using the ImageQuant TL software ( GE Healthcare Life Sciences ) . Actin levels were used for normalization across samples . β-elimination was done as described [47] . 18–28 nt small RNAs were size selected on 17% denaturing polyacrylamide gels . Small RNAs were Tobacco Acid Phosphatase treated to reduce 5′ di- and triphosphate groups to monophosphates , ligated to 3′ and 5′ adapters and subjected to RT-PCR and gel purification of small RNA amplicons . A detailed protocol is available on request . For Illumina GAII sequencing ( ubl-1::GFP and ubl-1::GFP-siR-1-sensor libraries ) , the 5′ adapter sequences were modified to contain barcodes ( AAC and CCC , respectively ) for multiplexing two libraries into one lane of a flowcell . For Illumina HiSeq sequencing , the TruSeq small RNA PCR Indexing primers RPI1 and RPI2 were used to introduce index sequences into each library and then multiplexed into one lane of a flowcell . Small RNA sequences were parsed and mapped to either the N2 reference genome ( Wormbase release WS204 ) or ubl-1::GFP and ubl-1::GFP-siR-1-sensor transgene sequences using CASHX v . 2 . 0 and custom Perl programs [59] . Data analysis was done as described [13] . The small RNA trimming and tailing analysis was done as described [49] using annotated miRNA and piRNA sequences [4] , [60] . siRNAs were classified by their length and genomic locus [13] . Synchronized C . elegans were fed E . coli HT115 expressing dsRNA against target genes [61] , [62] beginning at L1 larval stage and scored and imaged at the L4 larval stage during the second generation of feeding at 23–25°C . Quantitative RT-PCR assays of small RNA ( TaqMan , Life Technologies ) and mRNA ( SYBR Green , Bio-Rad ) levels were done according to Life Technologies and Bio-Rad recommendations and as described [13] . For mRNA assays , rpl-32 levels were used for normalization across samples . miR-1 or miR-35 levels were use for normalization of small RNA levels after determining their levels were unchanged using Northern blot assays . TaqMan probes were validated using mutants defective for each of the small RNAs analyzed . The 2−ΔΔct method was used for comparing relative levels of small RNAs and mRNAs . 5′ RACE assays for siRNA-guided cleavage were done as described [63] . Primer and small RNA sequences are listed in Table S2 . Statistical analysis was done in R and Excel . When comparing quantitative protein data , p values were calculated using two sample t-tests . For qRT-PCR data analysis , p values were calculated using ANOVA and Tukey's HSD tests . P values for comparing wild type and henn-1 mutant brood sizes were calculated using the Mann-Whitney test . Bonferroni corrections were applied to account for multiple comparisons . Nucleic acid sequence alignments were done with ClustalW v . 2 . 1 . Argonaute protein sequences were aligned with ClustalW v . 2 . 1 using protein weight matrix Pam350 ( Dayhoff ) [64] . The phylogenetic tree was drawn with PHYLIP v . 3 . 69 . The deep sequencing data reported here is available through the Gene Expression Omnibus database , www . ncbi . nlm . nih . gov/geo , via accession number GSE35550 .
|
RNA interference ( RNAi ) is the process in which endogenous small RNA pathways are exploited by researchers to direct RNA silencing of particular genes . Plants and animals use endogenous RNA silencing pathways for protection against viruses and transposable elements and to regulate genes during development . The features that route genes into specific RNA silencing pathways are poorly understood . Furthermore , it is not clear how small RNAs identify target mRNAs and how they repress their activity . Here , we show that a single siRNA target site is sufficient to trigger gene silencing in C . elegans without requiring perfect complementarity for target recognition . We also discovered an endogenous siRNA that acts in trans to initiate siRNA amplification . Finally , we show that siRNAs and PIWI-interacting RNAs ( piRNAs ) that bind specifically to PIWI clade Argonautes are methylated by the C . elegans HEN1 ortholog HENN-1 .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"genetics",
"biology",
"genomics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
PIWI Associated siRNAs and piRNAs Specifically Require the Caenorhabditis elegans HEN1 Ortholog henn-1
|
Visceral Leishmaniasis in humans presents with fever , anemia , and splenomegaly and can be lethal if not treated . Nevertheless , the majority of Leishmania infantum-infected individuals does not manifest symptoms and remain so provided they are not immunosuppressed . In this work , the performance of different tests was evaluated to detect asymptomatic individuals who were living in Teresina , Piauí state , Brazil , an endemic area for VL . L . infantum-specific antibodies were detected by ELISA and two different rapid immunochromatographic ( IC ) diagnostic tests , Kalazar Detect and OnSite , and parasitic loads were detected by real time PCR [qPCR] . Additionally , we measured levels of the biomarkers monokine induced by IFN-γ ( MIG ) and IFN-γ-induced protein 10 ( IP-10 ) before and after stimulation of whole blood with soluble Leishmania antigen [SLA] . Kalazar Detect and OnSite detected , respectively , 76% and 64% of patients presenting with active Visceral Leishmaniasis; 50% and 57% of patients remained positive in these tests , respectively , after treatment . Of the healthy participants in the study who were living in the endemic area , only 1 . 7% were positive with both of the IC tests . On the other hand , reactivity in ELISA tests revealed that 13% of these individuals presented asymptomatic infections; among VL patients , 84% presenting with active disease were reactive in ELISA , and after treatment , 55 . 5% were seropositive . L . infantum DNA was present in the blood of 37 . 9% of infected individuals living in the endemic area , while IP-10 and MIG biomarkers were detected in 26 . 7% of them . The greatest concordance of positivity occurred between ELISA and qPCR . The association of different techniques can detect asymptomatic infections , however , more research is necessary to develop ideal biomarkers that are simple to use in the clinic and in field studies in areas endemic for Visceral Leishmaniasis .
Visceral Leishmaniasis ( VL ) is classified by the World Health Organization as a neglected tropical disease due to high mortality rates , low attention given by the public sector and high endemicity in poor regions around world [1] . At a global level , ninety percent of VL cases were reported in seven countries , which include Brazil , Ethiopia , India , Kenya , Somalia , South Sudan and Sudan [1 , 2] . Caused by Leishmania infantum [synonymous L . chagasi] in the New World or L . donovani in the Old World , VL can be classified as an anthroponotic or zoonotic disease because it is transmitted between humans and others mammals , such as dogs [2 , 3] . Furthermore , blood transfusion is known to be another form of transmission of the parasite described in endemic areas and a cause for much concern since healthy uninfected donors are similar to asymptomatic infected donors . [4–6] . Indeed , in endemic areas in Brazil most people infected with L . infantum remain asymptomatic provided they are not immunosuppressed . Dogs are the main reservoirs of the disease , being well documented that asymptomatic canine hosts are frequent [7 , 8] . However , in addition to dogs , asymptomatic humans can represent an important reservoir and contribute to the maintenance of the pathogen in the endemic area [9 , 10] . The accurate epidemiological characterization of humans infected with L . infantum , as well as the elucidation of factors that result in asymptomatic infections or in development of active disease ( VL ) is essential for monitoring endemic areas and controlling this disease , thus the relevance of identifying asymptomatic individuals . This knowledge may contribute to development of control strategies and even treatments [11] . In this context , the clinical characteristics of active VL are shared with other diseases such as typhoid fever , tuberculosis and malaria , and coinfections of all types may occur , which make the diagnosis of VL complex [12] . Direct and indirect methods are used for diagnosis of VL . Direct methods consist of detection of the parasite through direct examination , culture or PCR using samples of tissue or marrow bone aspirates . The indirect methods screen for parasite-specific antibodies and for cellular responses with the leishmanin skin test ( LST ) ; antibody–based assays are immunofluorescence ( IFAT ) , direct agglutination test ( DAT ) , ELISA , Western blot and an immunochromatographic ( IC ) test ( rk39 ) [13] . The first test capable of detecting infections in individuals without symptoms of the disease was the Montenegro LST used by Manson-Bahr , 1959 [14] but the reactant for performing it is currently not marketed in Brazil , thus requiring an alternative test for detecting asymptomatic individuals in an endemic area . A gold standard , a reliable method to detect individuals presenting with asymptomatic infections with causative agents of VL is still lacking , therefore it is difficult to determine the extent of infection rates . In addition , asymptomatic individuals alternately exhibit positive and negative reactions in different immunologically-based tests , thus making a negative test result difficult to interpret [8 , 15] . In summary , the identification , management and understanding of the biological and epidemiological significance of all categories of infected individuals has become a major challenge in programs for controlling VL as there are no validated markers yet , and the currently employed parasitological , molecular , and serological tools are not fully adequate . There is , therefore , an urgent need for new biomarkers that can identify the asymptomatic population in areas where viscerotropic species of Leishmania are endemic and which might also better assist in monitoring the success of treatment in patients with the active disease [16 , 17] . This work aims to evaluate different antibody-based diagnostic tests and the use of biomarkers produced by antigen stimulated cells , such as IFN-γ-induced protein 10 ( IP-10 , or CXCL-10 ) and monokine induced by gamma interferon ( MIG , or CXCL9 ) , for the detection of L . infantum-infected individuals without clinical manifestations of active VL in the city of Teresina , state of Piauí , Brazil , an endemic area for VL .
The participants of this study consist of patients with Visceral Leishmaniasis , before and after treatment , who were hospitalized at the Institute of Tropical Diseases Natan Portela in the city of Teresina , Piauí , Brazil , and volunteers without any symptom of the disease residing in the same city . Sample collections were carried out between August and November , 2017 . Samples obtained from healthy volunteers residing in the city of Ribeirão Preto , São Paulo , Brazil , were used as control of non-endemic area . The diagnosis of VL was confirmed by the positivity of amastigote forms in samples of bone marrow aspirate stained by Giemsa , and confirmed by cell culture in NNN medium . The project was approved by the Research Ethics Committee by the Hospital das Clínicas of the Ribeirão Preto School of Medicine of the University of São Paulo , with Ethics Presentation Certificate number 67213017 . 0 . 0000 . 5440 and opinion number 2 . 101 . 755 . The certificate for authorization for research at the Institute of Tropical Diseases Natan Portela is number AA . 901 . 1 . 009518/17-65 . All methods were performed according to the approved guidelines . A consent term was obtained from all study participants . The assays were performed using Kalazar Detect Rapid ( InBIOS International , Seattle , WA ) , according to the manufacturers' instructions: 20 μL of the sera were applied in the sample pad area from the test strip plus 3 drops of Chase Buffer , the results were read 10 minutes later and , in the positive samples , a control line and test line appear in the test area . In the OnSite Leishmania IgG/IgM Combo test ( CTK Biotech ) , 20 μL of the sera were applied to the test area followed by 2 drops of diluents; in the positive samples a control line plus “G” and/or “M” lines appear , corresponding to IgG and IgM , respectively , according to the manufacturers' instructions . Promastigote forms of L . infantum parasites ( strain MHOM / BR / 74 / PP75 ) were cultured in Schneider's medium supplemented with 2% urine , 10% fetal bovine serum , 2% L-glutamine , 100 U / ml penicillin and 100 μg/mL streptomycin . The parasites in the stationary phase were enriched on the basis of negative agglutination by peanut agglutinin , and the soluble Leishmania antigen ( SLA ) was extracted . The promastigotes were washed with phosphate buffered saline ( 1X PBS ) at 3 , 000 x g , 10 minutes , 4°C three times , re-suspended in Tris-HCl ( pH 7 . 5 ) supplemented with protease inhibitors and subjected to 5 cycles of immersion alternately in liquid nitrogen and heated water baths . The lysate was sonicated , homogenized and centrifuged at 14 , 000 x g for 5 minutes at 4°C . The 96-well plates were incubated overnight with 2 μg/ml SLA per well . The next day , they were washed using Tris plus tween 20 ( TBT ) buffer and blocked with 2 . 5% Molico milk plus tween 20 for 2 hours . The serum diluted 1:50 to 1:400 was added to the same blocking solution and incubated for 1 hour at 37°C . After washing , peroxidase conjugated protein G diluted at 1:15 , 000 in TBT buffer was added for 1 hour at 37°C . A further washing step was performed and the TMB substrate was added , the reaction was stopped with 0 . 2 N sulfuric acid and read at 450 nm in Multiskan GO , Thermo Scientific ELISA reader . For each sample , the reactivity index ( RI ) was calculated by dividing the optical density of the serum test mean by the cutoff value which was determined by the mean optical density obtained by 16 negative samples plus three times the standard deviation . Samples were considered positive if RI ≥1 . 1 and negative if IR<1 . 1 as described in Marques et al . 2017 [11] . DNA extraction was performed using 200 μl of total peripheral blood at -80°C using the DNeasy Blood & Tissue kit , Quick Start Protocol ( Qiagen , Chatsworth , CA , USA ) according to the manufacturer's instructions . The DNA concentration and purity were checked spectrophotometrically at 260 nm and 280 nm . The standard curve was constructed from the parasite DNA of a patient diagnosed with VL . The DNA of the parasites on day 5 of culture was extracted and quantified . A calculation was then made considering the size of the haploid genome of L . infantum to determine the DNA concentration corresponding to 100 , 000 copies of the DNA polymerase gene . From this concentration serial dilutions ( 105 , 104 , 103 , 102 , 101 , 100 , and 10−1 ) were performed . Quantification of L . infantum in peripheral blood was determined via qPCR , in which the hydrolysis probe technology ( TaqMan ) was used . The probe used was the one corresponding to the DNA polymerase gene , which is a single copy gene ( GenBank Access AF009147 ) according to Bretagne and colleagues [18] . The reaction was performed using forward primers , 5'-TGTCGCTTGCAGACCAGATG-3'; and reverse primers 5'-GCATCGCAGGTGTGAGCAC-3' and probed at 5'FAM-CAGCAACAACTTCGAGCCTGGCACC-3'TAMRA . The amplification was performed in duplicate in a final volume of 20 μL containing the reagents: 2 μl of DNA from the samples , 1 μl of the probe ( 2 . 5 pmol ) , 1 μl of each primer ( 10 pmol ) , 10 μl of Master Mix ( Applied Biosystems ) and 5 μl of sterile ultrapure water . After the initial denaturation of 10 minutes at 95°C , the samples were subjected to 40 cycles of amplification consisting of two steps: 15 seconds at 95°C and 1 minute at 60°C . Negative control was included in the reaction . The reaction was run on the 7500 Fast Real-Time PCR System ( Applied Biosystems ) . Samples of whole blood ( 500 μL ) were stimulated with 10 μg/ml SLA and , incubated at 37°C for 24 hours , parallel control without stimulation was , also , incubated , after incubation time , were centrifuged at 2000 x g for 10 minutes and , the supernatant was collected and stored at -20°C for further analysis of biomarkers [16] . IP-10 and MIG were quantified in this study in 50 μL of plasma from whole blood SLA-stimulated or non-SLA-stimulated using the BD Cytometric Bead Array Human Flex Set ( Becton Dickinson Biosciences , USA ) as instructed by the manufacturer . Each sample was incubated for 1 hour at room temperature with 50 μl of capture beads and , after incubation , 50 μl of detection antibody was added and incubated for 2 hours at room temperature . The data were acquired using a FACSCanto II flow cytometer and analyzed using the Flow Cytometric Analysis Program Array ( Becton Dickinson Biosciences , USA] . The results for each biomarker were expressed by the difference in plasma concentration between SLA stimulated and control in pg/mL . The concentration of the biomarkers was compared using the Mann-Whitney U test . Significance was set at p ≤0 . 05 . All calculations were performed using GraphPad Prism 7 . 0 software ( GraphPad Software , USA ) [16] .
The volunteers in all groups were 18 years old or older . The mean age of the group of infected individuals presenting with active VL ( VL ) or after treatment of VL ( AT ) was 40 years , while for all other individuals who were living in the endemic area , before testing to separate asymptomatic infections with L . infantum ( ASYMP ) from uninfected endemic controls ( EC ) the mean age was 39 years and for non-endemic uninfected controls was 32 years ( Table 1 ) . Two different IC rapid diagnostic tests , Kalazar Detect ( InBios International Seattle , WA ) and OnSite Leishmania IgG/IgM combo ( Bio Advance Diagnósticos ) , both of which employ the K39 antigen , a kinesin-related protein found in isolates from cases of active VL [19] , were used to detect asymptomatic people in the endemic area in Piauí state , Brazil , besides examining responses of individuals presenting with active VL . In addition , the tests were applied to individuals who had been treated for VL . The results presented in Table 2 show that 76 ( 19/25 ) and 64% ( 16/25 ) of individuals in the active VL group presented positivity in Kalazar Detect and OnSite Leishmania IgG tests , respectively; however , the OnSite Leishmania IgM test was not positive in any of these individuals . In treated individuals ( AT group ) , some of them years after cure , reactivity was slightly lower: 50 ( 5/10 ) and 57% ( 4/7 ) were positive in the Kalazar Detect and OnSite Leishmania IgG tests , respectively , and none were positive in the OnSite Leishmania IgM test . Among those individuals who were living exposed to L . infantum ( Asymp and EC groups ) , only 1 . 7% ( 2/115 ) were positive in both tests , while none of the samples in the EC group were positive with the OnSite IgM test . When ELISA technique was used with sera diluted at 1:200 and reacted with soluble Leishmania antigen ( SLA ) , 84% ( 21/25 ) of seropositivity was found in the active VL group , whilst 55 . 5% ( 5/9 ) of the treated individuals ( AT group ) were seropositive . Among those individuals who were living exposed to L . infantum , 13% ( 15/115 ) were positive and may be infected with L . infantum , without , however , presenting with manifest symptoms . Provided these individuals presented a reactivity index ( RI ) ≥1 . 1 , as described by Marques and colleagues [11] , in which the mean of the optical density each sample is divided by the cutoff point , which was 0 . 04 in this study ( Fig 1A ) , they are categorized as infected . Once all individuals infected with L . infantum were identified , symptomatic infections were then categorized based on the statistical difference between the levels of their different markers and those of the EC or NC groups . Thus , it is possible to differentiate asymptomatic infections ( Asymp group ) , showing RI ≥1 . 1 from active VL and AT groups . Individuals from Teresina presenting an RI <1 . 1 in ELISA were considered non-infected and were categorized as the endemic control ( EC ) group . As illustrated in Fig 1B , where data are plotted as optical densities ( O . D ) measured at a wavelength of 450 nm , levels of SLA-specific antibodies differed significantly between healthy individuals from a non-endemic area ( normal controls , NC ) and infected , asymptomatic individuals from the endemic area ( Asymp ) ( Kruskal-Wallis test , p-value < 0 . 0001 ) , and between the NC group and patients with VL , whether active VL ( VL ) or treated VL ( AT ) ( Kruskal-Wallis test , p-value <0 . 05 ) . Furthermore , individuals with disease , whether active or after treatment , presented significantly higher levels of SLA-specific antibodies than endemic controls ( EC ) , but not when compared with levels of SLA-specific antibodies seen in the Asymp group ( Kruskal-Wallis test , p-value < 0 . 0001 ) . The qPCR technique was performed to detect the presence of L . infantum in the peripheral blood from patients with active VL and other groups from the endemic area confirming , in this manner , asymptomatic individuals . We selected at random 29 individuals living in the endemic area among 115 volunteer participants in this study and 11 ( 37 . 9% ) of those were classified as being latently infected with L . infantum by means of the presence of gene copies of parasites/mL on qPCR . Additionally , 6 ( 60% ) of the 10 individuals with active VL analyzed were positive . The results depicted in Fig 2 show that the number of gene copies of Leishmania differed significantly ( Mann Whitney test , p-value < 0 . 05 ) between the Asymp and active VL groups . The slope of the reaction was -3 . 33 and the efficiency was 99 . 5%; the negative reactions were not represented . For further analysis of individuals infected with L . infantum in the population from the endemic area , the concentrations of the IP-10 or MIG biomarkers were determined after stimulation with SLA of peripheral blood obtained from 14 patients with active VL , 6 cured individuals and 30 from the 115 healthy individuals who were living exposed to L . infantum without developing any symptoms of the disease; unstimulated blood was used as control for analysis . In addition , we also sought to validate in a Brazilian population the results from a study by Ibarra-Meneses and colleagues that described these cytokines as biomarkers of infections with L . infantum and L . donovani in individuals from Spain and Bangladesh , respectively [16] . The results presented in Fig 3 show that by using both biomarkers it was possible to detect 8 ( 26 . 7% ) responsive asymptomatic individuals among 30 individuals analyzed from the endemic area; these individuals were also categorized in the Asymp group . Significant differences in quantities of MIG ( Fig 3C ) ( t test , p-value <0 . 0001 ) or IP-10 ( Fig 3G ) ( t test , p-value<0 . 0001 ) were observed between non-infected , healthy endemic controls ( EC ) and asymptomatic infected individuals ( Asymp ) . The concentrations of MIG in active VL and AT groups differed significantly ( t-tet , p = 0 . 0093 [Fig 3A] and p = 0 . 0337 [Fig 3B] , respectively ) when compared to those of endemic controls ( EC ) . No significant difference was found in active VL versus AT group ( Fig 3D ) . Concentrations of IP-10 in reactions of cells from the active VL group ( Fig 3E ) and AT group ( Fig 3F ) did not differ significantly when compared to the reactions of cells from the EC group , the same occurring between active VL and ATgroups ( Fig 3H ) .
Teresina , in Piauí state , is one of the urban areas in Brazil where the population is at risk for developing VL [20] , however Public Health authorities lack a diagnostic routine to identify asymptomatic individuals infected with L . infantum . Such a routine would be important in such a context for following HIV-positive individuals , to assess transmission force and herd immunity and to evaluate false positives for other diseases . In this study , we evaluated different methods to detect asymptomatic individuals in that endemic region . We found that the IC tests employed in this study are not efficient for identifying this group of individuals because only 1 . 7% of the healthy participants were positive . However , these tests remain as an important alternative for debilitated patients who cannot undergo bone marrow aspirate procedures: in this study positive results were found to be above 60% and 70% for the two tests in cases of active VL . We obtained satisfactory results when qPCR was used for detection of asymptomatic individuals . ELISA , in turn , determined that 13% of individuals in the healthy population living exposed to L . infantum were infected and was the test that best concurred with results of qPCR . Finally , the biomarkers IP-10 and MIG indicated that 26 . 7% of individuals from an endemic area were infected , being that 62 . 5% of these ( 5 of 8 ) concurred with ELISA and qPCR tests . IC tests used for serodiagnosis of active VL are based on rK39 , a recombinant protein derived from a specific antigen produced by Leishmania donovani complex [21] . Several studies have evaluated the efficacy of rapid tests IC for diagnosis of active VL with regard to performance , cost , acceptability and suitability [22] for both of the agents that cause VL . In this study , we report that Kalazar Detect and OnSite Leishmania IgG/IgM were able to detect 76% and 64% of patients with clinically active VL caused by L . infantum; this performance is lower as compared with other studies , which reported 91 . 2% sensitivity and 94 . 5% specificity when using OnSite Leishmania IgG/IgM [23] , and 88 . 1–100% sensitivity and 90 . 6–99 . 1% specificity when using Kalazar Detect [24–26] . It has already been reported that the performance of similar diagnostic products may differ according to geographical regions due to the fact that L . infantum might be less abundant in blood than L . donovani [27] , due to parasite antigenic diversity and/or differences in antibody concentrations . These latter differences , in turn , may be due to different age patterns , immune responses , and nutritional status of patients [28] . Indeed , the study by [23] employed a historical collection of sera from Brazilian individuals to evaluate these IC tests . These authors did not inform the periods that the sera cover , but it is safe to say that the important and relatively recent epidemiological transition that Brazil has undergone changed its population’s nutritional status , access to sanitation and many other aspects that affect performance of antibody-based responses , including performances of antibody-based diagnostic tests . Furthermore , we refer readers to the important recent study by Kityo and colleagues that elucidated the mechanism whereby different populations mount distinct responses to identical antigens according to their levels of immune activation [29] . In addition , median ages of the cohorts differ between the studies that compare these tests . This study employed OnSite Leishmania IgG/IgM and Kalazar Detect because they fulfilled the Brazilian government’s public call for purchase of the new rapid tests for VL in that country , which stipulated the minimum performance as 90% for sensitivity and specificity [23 , 30] . The fact is that in Brazil reproducibility of all IC tests for diagnosis of VL is low and these tests need further evaluation [31] . In addition to issues of performance of tests in active and treated VL in patients from different geographical areas , there is the issue of their performance in asymptomatic individuals infected with L . infantum . Active VL , but not asymptomatic infections , are accompanied by inflammation [32 , 33] and polyclonal activation of B cells [34] , responses which may facilitate targeting the k39 antigen , thus explaining why the tests based on this antigen do not detect asymptomatic infections . Furthermore , L . infantum remains in the blood years after it is first detected , which may explain the positive results of the IC test in the AT group [27 , 35] . On the other hand , asymptomatic individuals have been infected for an unknown period during which Leishmania may circulate in the peripheral blood episodically , detectable by PCR , but not necessarily inducing a humoral or cellular response [35–39] . Indeed , Kalazar Detect cannot detect asymptomatic individuals infected with L . infantum , but it is able to detect infected asymptomatic individuals in regions where L . donovani is presented [16 , 40] , as in a hyper-endemic area from Bangladesh , albeit with low sensibility: Kalazar Detect detected only one positive serum sample among 35 individuals exposed to the parasite . On the other hand , the OnSite test was able detect asymptomatic with a slightly higher sensitivity , but only in fresh serum samples and not in stored serum [41] . In Brazil , asymptomatic infection was detected in the same region of this study by PCR-based assays of blood [10] . qPCR is a complex technique that is not standardized , although it is reported to present high sensitivity and specificity , both in blood and bone marrow samples [27] . Furthermore , it is too expensive for the routine application in an endemic area due to high cost of reagents and equipment , however the method has progressed in recent years . This would require greater accessibility of reagents , miniaturization of equipment , etc . , so that routine use of the technique is possible in less favored areas [42] . A promising possibility for this to be achieved is the qPCR performed with blood dried on cellulose paper [43 , 44] . The qPCR can be an important tool to identify asymptomatic infections in endemic areas , because it is a highly sensitive test to detect significant parasitemia , to recognize of potential progressers to clinical disease , facilitate early intervention , avoid possibility of disease transmission and for genotyping of species of interest in a given geographical area [41 , 45–49] . Thus , despite the fact that PCR may be of less value as a marker for acute clinical disease in VL endemic areas , it is a good marker of infection [27] . Antibody-based detection of infections with L . infantum in endemic areas also faces limitations , but can also identify healthy infected individuals [27] . However , serological tests such as direct agglutination ( DAT ) could be useful in the field [50 , 51] and , with efforts to improve technician skills and laboratory facilities , it could become a perfect tool for detection of the asymptomatic population . DAT reativity is controlled mainly by the host’s cell-mediated immune response , which highlights the importance of researching new markers [27] . Automated ELISA tests promise to be very efficient [52] , but ELISA is considered to be a diagnostic tool of low sensitivity for detection of asymptomatic human infections with L . infantum and false-negative test results underestimate the actual rate . However , ELISA can identify asymptomatic infections as is shown in the present study and also in other studies [6 , 53 , 54] . In addition , the ELISA method using crude antigen is generally more sensitive , albeit less specific [11] . Our results show a great coincidence of positivity of ELISA when compared with qPCR and IP-10/MIG biomarkers tests for the detection of asymptomatic individuals . Since ELISA is based on a multicomponent antigen , SLA , this fact may explain why it is a more sensitive test for detecting asymptomatic individuals than single antigen-based IC tests . In this study the concentrations of the MIG biomarker were significantly higher in supernatants of blood cells responding to SLA from individuals in active VL , AT and Asymp groups , compared to those in the EC group and was higher in VL patients , followed by asymptomatic infections and lower in AT group; no significant differences were observed between active VL and AT groups . On the other hand , concentrations of the IP-10 biomarker , interestingly , were higher in supernatants of blood cells responding to SLA from individuals presenting with asymptomatic infections compared to those seen in samples from the active VL group and the AT group . Thus , chemokines MIG and , especially , IP-10 can indeed provide a sensitive means of detecting a specific T-cell response to antigen following Leishmania infections in Brazil , in the same manner as has been shown in studies conducted in regions endemic for L . infantum in Spain and for L . donovani in Bangladesh [16 , 40] . IP-10 is released in response to both type I and type II IFN and is a chemotactic factor for activated T cells and NK cells , which when thus stimulated subsequently express CXCR3 and are recruited to sites of tissue inflammation [55] . This is one mechanism of resistance to L . infantum mediated by IP-10; in addition , this cytokine also mediates protection against Leishmania by inducing nitric oxide , an important leishmanicidal mediator [56–59] . These mechanisms mediated by IP-10 may explain the higher levels of this cytokine seen in the ASYMP group , which is seems to be resistant to developing disease despite infection . These facts about this biomarker raise questions concerning the mechanisms that result in resistance to development of disease , seen in the majority of infections of humans with L . infantum . The cultures of stimulated cells are performed with circulating SLA antigen-reactive leukocytes . Transcriptomes of whole blood leukocytes have been useful not only for identifying disease-specific biomarkers [60 , 61] , but also for unraveling mechanisms of resistance and susceptibility to developing disease after infections with L . infantum [62] . In this study , IP-10 seems to indicate that the immune response is well established regarding recruitment of activated/effector T cells , thereby initiating the effector arm of Th-1 cell immunity and differentiating active VL that asymptomatic infection . In conclusion , IP-10 showed high sensitivity and specificity to identify asymptomatics in L . donovani and L . infantum areas , meaning it is a good biomarker [16] . Since there is no standard diagnostic for the detection of asymptomatic infection , the combination of different techniques to determine their real prevalence is currently the best option [16 , 63] . rK39 IC , direct agglutination and leishmanin skin tests detected , respectively , 1 . 5% , 5 . 3% and 5 . 6% of asymptomatic infection in study performed in selected villages from Libo Kemkem and Fogera districts ( Amhara State , Ethiopia ) , indeed 10 . 1% of asymptomatic infection was detected with combined use of serologic methods and leishmanin skin test [63] . It is important to also use DNA detection tests , since asymptomatic infected individuals are carriers the Leishmania infection [27] . We must point out that a previous study performed with different samples from the same the same population examined in the present study [53] found a significantly higher prevalence of asymptomatic individuals with positive cellular skin reactions to Montenegro antigen than the number found with the in vitro cellular reaction employed in the present study ( 60 . 41% vs 26 . 7% , respectively; chi-square statistic 5 . 9997; p-value 0 . 0143 ) . While this difference may reflect the transmission dynamics of two periods separated by six years , there are also important differences between the two cellular tests: the Montenegro reaction results from the global cellular response elicited by the antigen , while the in vitro cellular reaction accounts for the production of two cytokines in this study and four cytokines in the study by Ibarra-Meneses and colleagues [16] . In our previous study we also showed that , of nine cytokines evaluated , only the levels of TNF-α differed significantly between individuals with asymptomatic infections and endemic healthy , non-infected controls [53] . In view of these results we suggest that levels of TNF-α also be evaluated in futures studies employing the in vitro cellular test . In conclusion , the present study shows that qPCR can be an efficient tool to distinguish asymptomatic individuals infected with L . infantum among healthy individuals living in Teresina , Piauí , Brazil , an endemic area for VL . In addition , ELISA to detect L . infantum-reactive antibodies continues to be a standard method to use in the field . The biomarkers MIG and IP-10 showed equivalent results compared to the qPCR and ELISA , with highlights for IP-10 for indicating asymptomatic infections with L . infantum . In summary , healthy individuals living in Teresina , Brazil and exposed to L . infantum , were evaluated for subclinical infection with serological , molecular and cellular approaches: among 115 of these individuals , 15 ( 13% ) presented antibodies reacting with SLA in an ELISA; gene copies of parasites detected with qPCR were present in 11 ( 37 . 9% ) of 29 of these individuals; finally , leukocytes from 8 ( 26 . 7% ) of 30 of these individuals when treated with SLA responded by producing significant amounts of IP-10 and MIG when compared with their untreated cells . Healthy , asymptomatic individuals from endemic areas of VL were the focus of this study because they represent the majority of outcomes of infections with L . infantum and for this reason they can help elucidate the immune mechanisms that operate in this outcome . We are not proposing to use any of the tests employed in this study for routine evaluations in the clinic ( e . g . , for blood banks ) . However , we need to identify these individuals in the endemic area for the reasons stated above . This is not an easy task without employing the current alternatives . Understanding this group , may however lead to development of tests and biomarkers applicable in the clinic . So far , few studies have been focused on individuals bearing latent infections with L . infantum and the results presented here justify further investments to definitely validate biomarkers for this group .
|
Infections with Leishmania infantum occurring in humans can become clinically manifest as active visceral leishmaniasis [VL] , more frequently so in individuals with compromised immune systems and the disease can be fatal if left untreated . However , many individuals living in endemic areas who are infected through of a bite of the insect vector remain asymptomatic . Currently , other forms of transmission are being discussed , such as through blood transfusions and organ donations from asymptomatic individuals . Parallel to this knowledge , we know that dogs are reservoirs for L . infantum , but the role of asymptomatic humans as alternative reservoirs of VL is still is debated . A bone marrow aspirate is the best test to diagnose active VL in humans , but is too invasive to use in all people living in endemic areas . Therefore , less invasive tests are needed that use blood samples to detect infected persons who have not developed disease . In this regard , we evaluated different tests in order to find out which is the most effective to diagnose asymptomatic infections .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"body",
"fluids",
"tropical",
"diseases",
"geographical",
"locations",
"biomarkers",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoans",
"leishmania",
"neglected",
"tropical",
"diseases",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"south",
"america",
"immunoassays",
"protozoan",
"infections",
"brazil",
"people",
"and",
"places",
"biochemistry",
"leishmania",
"infantum",
"eukaryota",
"blood",
"anatomy",
"physiology",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2019
|
Evaluation of methods for detection of asymptomatic individuals infected with Leishmania infantum in the state of Piauí, Brazil
|
Interferon ( IFN ) -induced transmembrane protein 3 ( IFITM3 ) is a cell-intrinsic factor that limits influenza virus infections . We previously showed that IFITM3 degradation is increased by its ubiquitination , though the ubiquitin ligase responsible for this modification remained elusive . Here , we demonstrate that the E3 ubiquitin ligase NEDD4 ubiquitinates IFITM3 in cells and in vitro . This IFITM3 ubiquitination is dependent upon the presence of a PPxY motif within IFITM3 and the WW domain-containing region of NEDD4 . In NEDD4 knockout mouse embryonic fibroblasts , we observed defective IFITM3 ubiquitination and accumulation of high levels of basal IFITM3 as compared to wild type cells . Heightened IFITM3 levels significantly protected NEDD4 knockout cells from infection by influenza A and B viruses . Similarly , knockdown of NEDD4 in human lung cells resulted in an increase in steady state IFITM3 and a decrease in influenza virus infection , demonstrating a conservation of this NEDD4-dependent IFITM3 regulatory mechanism in mouse and human cells . Consistent with the known association of NEDD4 with lysosomes , we demonstrate for the first time that steady state turnover of IFITM3 occurs through the lysosomal degradation pathway . Overall , this work identifies the enzyme NEDD4 as a new therapeutic target for the prevention of influenza virus infections , and introduces a new paradigm for up-regulating cellular levels of IFITM3 independently of IFN or infection .
Interferon ( IFN ) -induced transmembrane protein 3 ( IFITM3 ) is a 15 kDa protein that restricts cellular infection by influenza virus [1 , 2 , 3] . IFITM3 is active against all strains of influenza virus that have been tested to date , regardless of serotype or species of origin [1 , 3 , 4 , 5 , 6] , and it similarly inhibits many other medically important viruses such as HIV , SARS coronavirus , and Ebola virus [1 , 5 , 7 , 8 , 9] . Confirming its importance in vivo , IFITM3 knockout mice succumb to sublethal doses of influenza virus [10 , 11] . Likewise , IFITM3 is the only known protein for which a genetic polymorphism present in a significant percentage of the human population is associated with severe influenza virus infections [10 , 12 , 13 , 14] . In the cell , IFITM3 localizes to endosomes and lysosomes [15 , 16 , 17] , and traps endocytosed virus particles within these degradative compartments by impeding the formation of the virus fusion pore [16 , 18 , 19] . Yet , even with this potent mechanism by which IFITM3 limits infections , influenza virus remains a significant health concern [20 , 21] . This may be explained by the fact that IFITM3 is present at low levels within most cells at steady state and is induced by IFNs only after infection has already been established [3 , 11 , 22] . The inability to up-regulate IFITM3 levels independently of infection or IFNs is a challenge preventing the field from harnessing the activity of IFITM3 for infection prevention . We previously showed that ubiquitination increases the rate of IFITM3 turnover within the cell [15] . A non-ubiquitinated lysine-to-alanine mutant of IFITM3 possessed enhanced antiviral activity and a longer half-life as compared to WT IFITM3 [15] . These findings indicated that inhibition of IFITM3 ubiquitination could augment the activity and/or levels of endogenous IFITM3 , thus offering a strategy for exploiting IFITM3 therapeutically or prophylactically against viral infections . The identification of the E3 ubiquitin ligase ( s ) capable of modifying IFITM3 among the more than 600 annotated E3 ligases in the human genome will be an important step toward validating this antiviral strategy . Through our work studying tyrosine phosphorylation of IFITM3 , we discovered that phosphorylation at tyrosine 20 ( Y20 ) inhibited IFITM3 ubiquitination [23] . This led us to posit that phosphorylation of Y20 may block an E3 ubiquitin ligase recognition signal . Indeed , Y20 is part of a highly conserved PPxY motif ( where P = proline , x = any amino acid , and Y = tyrosine , Fig 1A ) [24] . PPxY motifs are commonly recognized by WW ( characterized by two tryptophan residues spaced approximately 20 amino acids apart ) domains of NEDD4-family E3 ubiquitin ligases , of which there are nine family members [25] . We chose to focus first on NEDD4 , the prototypical member of this family , for several reasons: 1 ) NEDD4 and IFITM3 both have ubiquitous expression patterns while several other NEDD4-family members are tissue-specific ( BioGPS . org [26] ) , 2 ) Like IFITM3 , many of the known NEDD4 substrates are membrane proteins and are associated with endosomal and lysosomal pathways [25 , 27] , 3 ) IFITM3 and NEDD4 are both S-palmitoylated , suggesting that they may localize to similar membrane subdomains [2 , 28] , and 4 ) NEDD4 is reported to be inhibited by ISG15 [29 , 30 , 31] , an IFN-inducible protein , thus providing an intriguing model whereby IFN might induce IFITM3 expression while also inhibiting its ubiquitination . Herein , we provide results demonstrating the ability of NEDD4 to ubiquitinate IFITM3 and identify a unique role for NEDD4 in decreasing steady state IFITM3 abundance , leading to increased cellular susceptibility to influenza virus infection .
To explore the possibility that NEDD4 ubiquitinates IFITM3 , we first examined whether IFITM3 and NEDD4 are in proximity to one another within cells . We stimulated mouse embryonic fibroblasts ( MEFs ) with IFN-α to induce abundant expression of IFITM3 . By performing immunofluorescence microscopy imaging of endogenous IFITM3 and NEDD4 , we detected co-localization of these two proteins ( Fig 1B ) . Further co-localization of these proteins with endogenous LAMP1 , a lysosomal marker , indicates that NEDD4 and IFITM3 may interact at lysosomes ( Fig 1B ) . We next examined the effect of overexpressing HA-tagged human NEDD4 ( HA-NEDD4 ) on IFITM3 ubiquitination . We observed a significant increase in myc-tagged human IFITM3 ( myc-hIFITM3 ) ubiquitination when HA-NEDD4 was expressed as compared to the transfection control ( Fig 1C ) . On the contrary , no increase in IFITM3 ubiquitination was seen upon overexpression of HA-tagged human CBL-B , another E3 ubiquitin ligase that has been reported to interact with NEDD4 [32] , and that is associated with regulation of immune responses [33] ( Fig 1C ) . Additionally , we examined the effect of overexpressing FLAG-tagged human NEDD4 ( FLAG-NEDD4 ) on IFITM3 ubiquitination in comparison to FLAG-tagged human SMURF1 and SMURF2 ( FLAG-SMURF1 and FLAG-SMURF2 ) , both of which are members of the NEDD4-family of ubiquitin ligases . FLAG-NEDD4 caused an increase in IFITM3 ubiquitination while FLAG-SMURF1 and FLAG-SMURF2 were unable to robustly modify IFITM3 ( Fig 1D ) . These results demonstrate that NEDD4 possesses a degree of specificity for IFITM3 that is lacking for CBL-B and the NEDD4-family members , SMURF1 and SMURF2 . As previously mentioned , NEDD4-family ubiquitin ligases possess two to four characteristic WW domains that interact with proline-rich motifs , including PPxY motifs , on substrate proteins [34] . NEDD4 has four WW domains and IFITM3 contains a highly conserved PPxY motif within its N-terminus ( Fig 1A ) . To test whether these domains are required for IFITM3 ubiquitination , we generated an IFITM3 mutant in which each residue of the PPxY motif ( 17-PPNY-20 in IFITM3 ) was mutated to alanine ( designated 17-20A ) and utilized a FLAG-NEDD4 mutant in which its four WW domains were deleted ( designated ΔWW ) . Upon co-overexpression of FLAG-NEDD4 with myc-hIFITM3 , ubiquitination of IFITM3 was increased as expected , while the ΔWW mutant was unable to increase IFITM3 ubiquitination ( Fig 1E ) . In fact , FLAG-NEDD4-ΔWW partially decreased steady state IFITM3 ubiquitination , perhaps indicating a dominant negative effect ( Fig 1E ) . Moreover , the 17-20A mutant of IFITM3 showed less ubiquitination than WT IFITM3 and was unaffected by overexpression of NEDD4 ( Fig 1E ) . The IFITM3 PPxY motif shares its tyrosine with an overlapping YxxΦ motif known to be involved in the trafficking of IFITM3 from the plasma membrane to endosomes [23 , 35 , 36] . Thus , in order to be certain that the results we observed for the 17-20A mutant of IFITM3 was not because of interference with the YxxΦ motif , we tested additional PPxY mutants in which the two prolines were mutated to alanine ( myc-hIFITM3-P17 , 18A ) or in which the tyrosine was mutated to alanine ( myc-hIFITM3-Y20A ) . Upon co-overexpression of FLAG-NEDD4 , the ubiquitination of both of these mutants was only minimally increased as compared to the robust increase in ubiquitination of WT IFITM3 ( Fig 2A ) . Interestingly , a truncated form of IFITM3 missing its first 21 amino acids , including the PPxY motif , is prevalent in certain human populations . This variant is associated with severe influenza virus infections [10 , 13 , 14] and more rapid progression of HIV-related disease [37] . A myc-hIFITM3 construct lacking these first 21 amino acids ( Δ1–21 ) was , as expected , largely unaffected in terms of ubiquitination by overexpression of FLAG-NEDD4 ( Fig 2B ) , identifying a potentially important difference between the truncated and full-length IFITM3 proteins . Next , since NEDD4 has been shown to physically interact with the PPxY motifs of its substrate proteins [25] , we examined whether or not NEDD4 and IFITM3 co-immunoprecipitate with one another . We found that myc-hIFITM3 and FLAG-NEDD4 indeed co-immunoprecipitated with one another ( Fig 2C ) , suggesting a physical interaction . Importantly , this interaction was greatly diminished between FLAG-NEDD4 and the P17 , 18A mutant of IFITM3 ( Fig 2C ) . In sum , these results indicate that the IFITM3 PPxY motif is required for a strong interaction with and ubiquitination by NEDD4 . To determine whether a non-enzymatic activity of NEDD4 might be mediating its effect on IFITM3 ubiquitination , we tested a catalytically inactive NEDD4 point mutant . We found that this mutant was unable to increase IFITM3 ubiquitination , establishing that catalytic activity of NEDD4 is indeed required for its ability to increase IFITM3 ubiquitination ( Fig 3 ) . Since murine ( m ) IFITM3 also possesses a PPxY motif ( Fig 1A ) , we tested the ability of NEDD4 to affect mIFITM3 modification . Like myc-hIFITM3 , we observed an increase in myc-mIFITM3 ubiquitination when HA-NEDD4 was co-overexpressed and observed no effect of the catalytic mutant ( Fig 3 ) , suggesting a possible evolutionary conservation of NEDD4 modification of IFITM3 in mice and humans . These data further implicate NEDD4 as an E3 ubiquitin ligase capable of enzymatically modifying mouse and human IFITM3 . While NEDD4 overexpression experiments suggest that NEDD4 directly ubiquitinates IFITM3 ( Figs 1C , 1D and 1E , 2A and 2B and 3 ) , this effect could be indirect . We therefore tested the ability of purified NEDD4 to ubiquitinate immunoprecipitated IFITM3 in vitro in order to confirm that NEDD4 can directly modify IFITM3 . HA-hIFITM3 was incubated with purified NEDD4 , enzymatic cofactors , and ubiquitin . We then re-immunoprecipitated IFITM3 and subjected it to anti-ubiquitin western blotting . Our results show that NEDD4 is capable of robustly ubiquitinating IFITM3 in vitro ( Fig 4 ) . Additionally , we employed ubiquitin mutants that could only be added via lysine 48 ( K48 ) or lysine 63 ( K63 ) linkages in order to examine whether NEDD4 preferentially utilizes one of these polyubiquitination linkages for modifying IFITM3 . While both K48 and K63 linkages could be added to IFITM3 by NEDD4 , we observed a preference for the K48 linkage in long polyubiquitin chains , which is traditionally associated with protein degradation ( Fig 4 ) . These results are consistent with our past results using linkage-specific anti-ubiquitin antibodies , which demonstrated that while both K48 and K63 ubiquitin linkages could be detected on IFITM3 , K48 linkages are more prevalent [15] . These data are also consistent with our previous results indicating that ubiquitination of IFITM3 promotes its turnover [15] . In order to examine the effects of NEDD4 on endogenous IFITM3 , we examined NEDD4 WT and knockout ( KO ) mouse embryonic fibroblasts ( MEFs ) [38] . We also utilized KO MEFs reconstituted with NEDD4 via retroviral transduction . Remarkably , Western blotting of lysates from NEDD4 KO cells showed an increase in steady state IFITM3 levels as compared to WT cells , while NEDD4 reconstitution decreased IFITM3 to WT levels ( Fig 5A ) . To examine the requirement for NEDD4 in ubiquitinating IFITM3 , we immunoprecipitated IFITM3 from large quantities of lysate from both WT and KO cells , expecting that the immunoprecipitation reagents would be saturated , thus providing us with comparable amounts of IFITM3 for examination of ubiquitination . Indeed , IFITM3 from NEDD4 KO cells was ubiquitinated much less than IFITM3 from WT cells ( Fig 5B ) . These results demonstrate that NEDD4 is required for proper steady state ubiquitination of IFITM3 , and that the absence of NEDD4 results in cellular accumulation of unmodified IFITM3 . Given the increase in baseline IFITM3 levels , we predicted that NEDD4 KO cells would be more resistant to influenza virus infection . We observed that NEDD4 KO MEFs were in fact significantly less susceptible to infections with influenza A virus ( IAV ) subtypes H1N1 and H3N2 ( PR8 and X-31 strains , respectively ) compared to WT control cells ( Fig 5C ) . The decreased susceptibility of KO cells was returned to WT levels of infection upon NEDD4 reconstitution ( Fig 5C ) . We also verified that the enhanced resistance of NEDD4 KO cells to influenza virus infection included resistance to recently circulating strains . NEDD4 KO cells were significantly less susceptible than WT cells to infection by both influenza B virus ( IBV ) and IAV H3N2 strains isolated in 2011 ( Fig 5D ) . We also examined retrovirus pseudotyped with the vesicular stomatitis virus ( VSV ) G protein , which is also reported to be inhibited by IFITM3 [3 , 35 , 36 , 39 , 40] . As expected , the percent of NEDD4 KO cells infected with VSV G-pseudotyped virus was significantly less than WT cells ( Fig 5D ) . Sendai virus ( SeV ) , a parainfluenza virus that primarily fuses at the cell surface [41] and is thus only minimally affected by IFITM3 [4] , was also tested . Unlike IAV , IBV , and VSV G-pseudotyped retrovirus , SeV was not appreciably affected by NEDD4 KO ( Fig 5D ) . Thus , the pattern of virus restriction we observed is consistent with protection of NEDD4 KO cells by IFITM3 . To confirm that the increased resistance of NEDD4 KO cells to influenza virus infection was due to increased levels of basal IFITM3 , we knocked down IFITM3 in NEDD4 WT and KO cells for 24 hours prior to infection . Knockdown was verified through Western blotting of cell lysates prepared at the time of infection ( Fig 6A ) . Importantly , knockdown of IFITM3 in both NEDD4 WT and KO MEFs resulted in an increase in influenza virus susceptibility , and largely eliminated the resistance of NEDD4 KO cells to infection ( Fig 6B ) . Overall , these experiments demonstrate that NEDD4 promotes cellular susceptibility to influenza virus infection by decreasing levels of IFITM3 . To extend our results to more relevant human lung cells , we utilized the A549 human alveolar epithelial cell line to study the role of NEDD4 in the regulation of steady state IFITM3 levels . Knockdown of NEDD4 with siRNA in A549 cells led to a significant increase in endogenous IFITM3 compared to non-targeting control siRNA ( Fig 7A ) . As expected , NEDD4 knockdown led to a significantly greater resistance to IAV infection ( Fig 7B and 7C ) . Importantly , we found that the relationship between NEDD4 knockdown and increased IFITM3 levels was preserved in two additional human lung cell lines ( Fig 7D ) . Taken together with experiments presented in Figs 3 , 5 and 6 , these data confirm an evolutionary conservation between mice and humans in the regulation of cellular IFITM3 levels by NEDD4 . This work also identifies NEDD4 as a novel target in human cells for improving resistance to influenza virus infection independently of IFNs or adaptive immunity . The degradative pathway involved in the turnover of steady state IFITM3 has not been previously investigated . Since NEDD4 is known to associate with the endosomal and lysosomal system and to target several of its substrates for lysosomal degradation [25] , our results identifying NEDD4 as the primary ubiquitin ligase for IFITM3 would suggest that IFITM3 is degraded in lysosomes . To test this hypothesis , we utilized chloroquine and bafilomycin , which inhibit endosomal and lysosomal acidification and thus the activation of pH-dependent lysosomal proteases . We observed that treatment of A549 lung cells with these two inhibitors caused an accumulation of IFITM3 ( Fig 7E ) . Similarly , treatment with leupeptin , an inhibitor of specific lysosomal proteases resulted in a similar increase in IFITM3 levels ( Fig 7E ) . This is in contrast to the treatment of cells with the proteasomal inhibitor MG132 , which consistently caused a modest decrease in IFITM3 levels , perhaps due to up-regulation of lysosomal degradation pathways when proteasome activity is inhibited . Overall , these experiments demonstrate that , consistent with the co-localization of IFITM3 and NEDD4 at lysosomes ( Fig 1B ) , and the ubiquitination of IFITM3 by NEDD4 ( Figs 1C , 1D and 1E , 2 , 3 , 4 and 5B ) , IFITM3 is turned over by the lysosomal degradation pathway .
Our previous work established that ubiquitination promotes the turnover of IFITM3 [15] . Thus , identification of the IFITM3 ubiquitin ligase would provide a potential target for increasing IFITM3 abundance and resistance to virus infections . In our previous work studying regulation of IFITM3 endocytosis by phosphorylation , we made the serendipitous discovery that the amino acid Y20 within IFITM3 is involved in regulating IFITM3 ubiquitination [23] , which led us to identify the involvement of the IFITM3 PPxY motif in its ubiquitination by NEDD4 ( Figs 1E and 2 ) . NEDD4 knockdown or knockout in human or mouse cells , respectively , resulted in substantially greater levels of steady-state IFITM3 ( Figs 5A , 6A and 7A and 7D ) . This accumulation of unmodified IFITM3 is consistent with the observed decrease in IFITM3 ubiquitination in NEDD4 KO cells ( Fig 5B ) . An additional intriguing aspect of our finding that IFITM3 steady state levels are regulated by NEDD4 is the previously described role of the IFN effector ISG15 in inhibiting NEDD4 [29 , 30] . ISG15 is a ubiquitin-like protein that specifically binds to NEDD4 , blocking its productive interaction with Ubiquitin-E2 ligase complexes [29 , 30] . The importance of this pathway was highlighted by two independent studies demonstrating that ISG15 blocks NEDD4-mediated monoubiquitination of the VP40 matrix protein of Ebola virus , thereby inhibiting the budding of Ebola virus-like particles [29 , 30] . Importantly , several studies have implicated ISG15 as a critical antiviral effector against IAV and IBV [42 , 43 , 44] . Two studies have demonstrated conjugation of ISG15 onto the IAV NS1 protein by the E3 ligase HERC5 , and found that ISGylation of IAV NS1 antagonizes virus replication [44 , 45] . Interestingly , IBV NS1 specifically blocks human ISG15 conjugation by preventing ISG15 interaction with the ISG15 activating enzyme UbE1L , effectively counteracting its antiviral effect [43 , 46 , 47 , 48] . We posit that high levels of IFITM3 attained after IFN stimulation result from both IFITM3 gene induction , as well as increased IFITM3 protein stability as a result of ISG15 inhibition of NEDD4 . We are currently investigating this exciting potential synergistic link between ISG15 and IFITM3 . Our work demonstrates that NEDD4 is required for proper basal ubiquitination of IFITM3 ( Fig 5B ) . However , our results would also suggest that additional ubiquitin ligases are also able to modify IFITM3 , particularly when IFITM3 is present at high levels . This is supported by detection of partial ubiquitination of our various IFITM3-PPxY mutants ( Figs 1E and 2A and 2B ) and by detection of modest IFITM3 ubiquitination in NEDD4 KO cells ( Fig 5B ) . The identities of secondary ubiquitin ligases for IFITM3 are still unknown . Of particular interest are the ubiquitin ligases capable of modifying the truncated Δ1–21 splice variant of human IFITM3 , which was not significantly ubiquitinated by NEDD4 ( Fig 2B ) . Identifying the ubiquitin ligases that modify this disease-associated variant may implicate crucial differences in the stability and degradative pathways potentially underlying the defect possessed by this protein . Nonetheless , our data clearly implicate NEDD4 as the primary E3 ubiquitin ligase for IFITM3 , and demonstrate that NEDD4 is essential for maintaining low steady state IFITM3 levels . This current work is in contrast to a prior study that concluded the IFITM3 PPxY motif was not involved in regulating the levels or antiviral activity of overexpressed IFITM3 [36] . However , this previous work did not directly assess ubiquitination of IFITM3 upon mutation of the PPxY motif . Additionally , our experiences studying IFITM3 ubiquitination here and in our prior work suggest that when examining overexpressed IFITM3 constructs , ubiquitination has only subtle effects on total protein levels detected by Western blotting despite significant effects on the IFITM3 half-life13 , 20 . Thus , overexpression likely masked any effects of mutating the PPxY motif on the parameters previously tested [36] . Our study has uncovered a novel mechanism by which NEDD4 indirectly promotes cellular entry of influenza virus by decreasing IFITM3 levels ( Figs 5 , 6 and 7 ) . Although this work is the first of its kind to identify NEDD4 as a negative regulator of IFITM3 levels , NEDD4 is well described to be necessary for the replication of several important RNA viruses . For example , NEDD4 interacts with proline-rich motifs in the viral late budding domains of Ebola virus [49] , rabies virus [50] , and HIV [51] . Mono-ubiquitination of these domains promotes efficient budding and viral egress necessary for productive viral spread . However , it remains to be determined whether inhibition of NEDD4 will serve as an effective in vivo antiviral strategy , particularly since NEDD4 is a developmentally essential molecule as demonstrated by the embryonic lethality of NEDD4 KO mice [52] . Likewise , NEDD4 has been implicated in regulation of insulin-like growth factor signaling [53] , T-cell-mediated immunity [54 , 55] , and tumor suppression [56] . On the other hand , neuron- and skeletal muscle-specific NEDD4 KO mice are viable [57 , 58] , and NEDD4 is naturally inhibited by ISG15 during virus infections [29 , 30] , perhaps suggesting that short-term inhibition of NEDD4 can occur without adverse effects . Additional experimentation will be needed to answer these vital questions , and this will be aided by the development of selective NEDD4 inhibitors , which is an area of active investigation [59 , 60] . Overall , our study identifies inhibition of NEDD4 as a novel strategy for preventing infection by influenza virus and other IFITM3-sensitive viruses through the increased accumulation of the antiviral restriction factor IFITM3 .
All cell lines used in these studies ( HEK293T , A549 , NCI-H358 , NCI-H2009 , and MEFs ) were cultured in DMEM supplemented with 4 . 5 g/L D-glucose , L-glutamine , 110 mg/L sodium pyruvate , and 10% fetal bovine serum ( Thermo Scientific ) at 37°C and 5% CO2 in a humidified incubator . HEK293T and A549 cells were purchased from ATCC . NCI-H358 and NCI-H2009 cells were obtained from the ATCC and provided to us by Dr . Gustavo Leone ( The Ohio State University ) . NEDD4 WT and KO MEFs used in this study were generated by Dr . Hiroshi Kawabe ( Max Planck Institute ) [38] and were kindly provided to us by Dr . Matthew Pratt ( University of Southern California ) who also generated the retrovirally reconstituted control cell lines . For Western blotting , cells were plated for 90% confluency in 6-well plates for 24 h prior to transfection with 2 μg/well of plasmids using Lipofectamine 2000 ( Invitrogen ) . For microscopy , MEFs were plated for 50% confluency on glass coverslips in 12-well plates for 24 h prior to overnight treatment with IFN-α ( BEI Resources ) . IFITM3 constructs were expressed from the pCMV-HA or pCMV-myc vectors ( Clontech ) as described previously [2 , 4 , 23] . IFITM3 mutants were made using the QuikChange Multi site-directed mutagenesis kit ( Stratagene ) . Plasmids expressing HA-NEDD4 and HA-NEDD4-C867A were obtained from Addgene ( plasmids 27002 and 26999 , deposited by Dr . Joan Massagué , Memorial Sloan Kettering Cancer Center ) [61] , and plasmids expressing FLAG-NEDD4 , FLAG-NEDD4-ΔWW , and HA-CBL-B were kindly provided by Dr . Jian Zhang ( The Ohio State University ) . FLAG-SMURF1 and FLAG-SMURF2 were obtained from Addgene ( plasmids 11752 and 11746 , desposited by Jeff Wrana , University of Toronto ) [62] . IFITM3 knockdown in MEFs was performed using Silencer Select Ifitm3 siRNA ( Ambion , catalog no . 4390816 ) and negative control ( Ambion , catalog no . 4390844 ) . Human NEDD4 knockdown in A549 , NCI H358 , and NCI H2009 cells was performed using Dharmacon ON-TARGETplus SMARTpool Human NEDD4 ( GE Healthcare , catalog no . L-007178-00 ) and Dharmacon ON-TARGETplus Control Pool Non-targeting control ( GE Healthcare , catalog no . D-001810-10-20 ) . siRNAs were transfected into cells using Lipofectamine RNAiMax transfection reagent ( Invitrogen ) . Transfection of siRNA was performed for 24 h for mIFITM3 knockdown , and 48 h for NEDD4 knockdown . For Western blotting , cells were lysed with 1% Brij buffer ( 0 . 1 mM triethanolamine , 150 mM NaCl , 1% BrijO10 ( Sigma ) , pH 7 . 4 ) containing EDTA-free protease inhibitor mixture ( Roche ) and 25 μM MG132 ( Sigma ) . Immunoprecipitations were performed using EZview Red anti-c-myc or anti-HA affinity gel ( Sigma ) , or with Protein G Plus Agarose Suspension ( Calbiochem ) in conjunction with anti-mIFITM3 . Chloroquine , bafilomycin , and leupeptin were purchased from Sigma . Co-immunoprecipitation assays were adapted from a previously described protocol [63] . HEK293T cells were co-transfected overnight with plasmids expressing myc-hIFITM3 and FLAG-NEDD4 . Cells were washed twice with PBS , lysed on ice in Triton X-100 lysis buffer ( 50 mM Hepes , pH 7 . 5 , 150 mN NaCl , 1% Triton X-100 , 10% glycerol , 1 . 5 mM MgCl2 , 1 . 0 mM EGTA , 10 μg/mL leupeptin , 10 μg/mL aprotinin , 10 μg/mL pepstatin , and 1 mM PMSF ) for 5 min , and centrifuged at 1 , 000 x g for 5 min at 4°C . 50 μg of cell lysate was set aside for each sample in order to evaluate , via Western blotting , expression of myc-hIFITM3 , FLAG-NEDD4 , and GAPDH as a loading control . Equal concentrations of cell lysate were immunoprecipitated using 15 μL EZview Red anti-c-myc or anti-FLAG affinity gel ( Sigma ) per sample for 1 h at 4°C with gentle nutation . Immunoprecipitations were washed three times with lysis buffer and examined by Western blotting with both anti-myc and anti-FLAG for each immunoprecipitate . Western blotting was performed with anti-myc ( Developmental Studies Hybridoma Bank at the University of Iowa , deposited by Dr . J . Michael Bishop , catalog no . 9E 10 ) , anti-HA ( Clontech , catalog no . 631207 ) , anti-hIFITM3 ( Proteintech Group , catalog no . 11714-1-AP ) , anti-mIFITM3 ( Abcam , catalog no . ab65183 ) , anti-NEDD4 ( Millipore , catalog no . 07–049 ) , anti-FLAG ( Sigma , catalog no . F7425 ) , anti-actin ( Abcam , catalog no . ab3280 ) , or anti-GAPDH ( Invitrogen , catalog no . 398600 ) antibodies . All primary antibodies were used at a 1:1000 dilution . Secondary antibodies , Goat Anti-Mouse IgG , HRP conjugate ( Millipore catalog no . 12–349 ) , Goat Anti-Rabbit IgG , HRP-linked ( Cell Signaling , catalog no . 70745 ) , and Goat Anti-Mouse , IgG1 Gamma 1 Heavy Chain Specific ( SouthernBiotech , catalog no . 1070–05 , specifically used for detecting immunoprecipitated protein ubiquitination ) were all diluted at 1:20 , 000 . HEK293T cells were transfected overnight with plasmid expressing HA-hIFITM3 . Protein collected from all wells of one 6-well plate was immunoprecipitated using anti-HA affinity gel and was washed extensively . Immunoprecipitated protein on affinity gel was resuspended in PBS . 10% of the retrieved protein was used in each reaction containing 500 μM ubiquitin or ubiquitin mutants ( Boston Biochem , catalogue nos . U-100H , UM-K630 , or UM-K480 ) , 0 . 5 μM UbcH5b E2 ligase ( Boston Biochem , catalogue no . E2-622 ) , 100 nM UBE1 E1 ligase ( Boston Biochem , catalogue no . E305 ) , and 1x Ubiquitin Conjugation Reaction Buffer containing ATP ( Boston Biochem , catalogue no SK-10 ) in the presence or absence of 100 ng recombinant human NEDD4 ( Sigma , catalogue no . SRP0226 ) . Reactions were allowed to proceed at 37°C for 1 h and were stopped by boiling for 5 min . The reactions were then diluted 1:100 in ice cold 1% Brij buffer , and IFITM3 was re-immunoprecipitated at 4°C using newly added anti-HA affinity gel prior to Western blot analysis . Cells were fixed for 10 min with 3 . 7% paraformaldehyde , permeabilized with 0 . 1% Triton X-100 in PBS for 10 min , and blocked for 10 min with 2% FBS in PBS . Primary antibodies , anti-mIFITM3 ( Fragilis , Abcam , catalogue no . ab15592 ) ( 1:500 ) , anti-NEDD4 ( 1:500 ) , and anti-LAMP1 ( Santa Cruz Biotechnology , catalogue no . sc-19992 ) , and Alexa Fluor-labeled anti-mouse and anti-rabbit secondary antibodies ( Life Technologies , 1:1000 ) or anti-rat DyLight 550-labeled secondary antibody ( Abcam , catalogue no . ab96888 , 1:1000 ) were diluted in 0 . 1% Triton X-100 in PBS . Cells were treated with antibodies sequentially for 20 min at room temperature and washed five times with 0 . 1% Triton X-100 in PBS after each antibody treatment . Glass slides were mounted in ProLong Gold antifade reagent containing DAPI ( Life Technologies ) . Images were captured using a Fluoview FV10i confocal microscope ( Olympus ) . Influenza viruses A/Puerto Rico/8/1934 ( H1N1 , PR8 ) , a PR8 reassortant virus possessing the hemagglutinin and neuraminidase genes from A/Aichi/2/1968 ( H3N2 , X-31 ) , A/Victoria/361/2011 ( H3N2 ) , and B/Texas/06/2011 were propagated in 10-day embryonated chicken eggs ( purchased as day 0 eggs from Charles River Laboratories ) for 48 h at 37°C as described previously [64] . PR8 and X-31 were provided to us by Drs . Bruno Moltedo and Thomas Moran ( Mount Sinai School of Medicine ) and the 2011 virus isolates were obtained from BEI Resources sponsored by the NIH/NIAID . SeV expressing green fluorescent protein ( SeV-GFP ) [65] was generated by Dr . Dominique Garcin ( University de Geneve ) and provided to us by Dr . Mark Peeples ( Nationwide Children’s Hospital Research Institute ) . SeV-GFP was propagated in 10-day embryonated chicken eggs for 40 h at 37°C as described previously [66] . VSV G-pseudotyped retrovirus expressing green fluorescent protein was generated by transfection of the viral vector pLenti-CMV-GFP-puro ( Addgene plasmid 17448 , deposited by Dr . Eric Campeau ) [67] and packaging plasmids ( provided by Dr . Li Wu , The Ohio State University ) along with plasmid expressing VSV G into HEK293T cells . Direct inhibition of GFP production by IFITM3 was not expected since this is driven by the CMV immediate early promoter and bypasses retrovirus-specific expression machinery [68] . Media was changed 18 h post-transfection , and media containing virus was then harvested 48 h post-transfection . Virus-containing media was centrifuged at 1200 x g for 5 min , filtered with 0 . 45 μm filters , frozen , stored at -80°C , and used for infection at a dose that provided approximately 70% infection of WT MEFs . MEFs were infected with IAV PR8 , X-31 , and H3N2 2011 strains , SeV and IBV at a multiplicity of infection of 5 . 0 or 10 . 0 . MEFs were infected for 24 h , except in the case of VSV G-pseudotyped retrovirus infections , which were analyzed after 48 h of infection . A549 cells were infected with IAV strain PR8 at a multiplicity of infection of 2 . 5 for 6 h . Infected cells were washed with PBS and harvested in 0 . 25% trypsin EDTA . Cells were fixed in 3 . 7% paraformaldehyde for 10 min and permeabilized with 0 . 1% Triton X-100 for 10 min . IAV infected cells were stained with anti-influenza nucleoprotein ( Abcam , catalog no . ab20343 , 1:333 ) directly conjugated to Alexa Fluor 647 using a 100 μg antibody labeling kit ( Life Technologies ) . IBV infected cells were stained with anti-IBV nucleoprotein ( Thermo Scientific catalogue no . MA1-80712 , 1:1000 ) followed by anti-mouse secondary antibodies conjugated directly to Alexa Fluor 488 ( Life Technologies ) . Measurement of SeV and VSV G-pseudotyped retrovirus infection rates was done by detecting virus-encoded GFP . All antibodies were diluted in 0 . 1% Triton X-100 in PBS , and cells were stained for 20 min . Cells were washed three times with 0 . 1% Triton X-100 in PBS after each antibody treatment . PBS was used for final resuspension of cells for flow cytometric analysis using a FACSCanto II flow cytometer ( BD Biosciences ) . Results were analyzed using FlowJo software .
|
IFITM3 is critical for limiting the severity of influenza virus infections in humans and mice . Optimal antiviral activity of IFITM3 is achieved when it is present at high levels within cells . Our results indicate that the E3 ubiquitin ligase NEDD4 decreases baseline IFITM3 levels by ubiquitinating IFITM3 and promoting its turnover . Depleting NEDD4 from cells results in IFITM3 accumulation and greater resistance to infection by influenza viruses . Therefore , we have identified NEDD4 as a regulator of IFITM3 levels and as a novel drug target for preventing influenza virus and other IFITM3-sensitive virus infections .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
E3 Ubiquitin Ligase NEDD4 Promotes Influenza Virus Infection by Decreasing Levels of the Antiviral Protein IFITM3
|
Accumulating evidence suggests that breast cancer metastatic progression is modified by germline polymorphism , although specific modifier genes have remained largely undefined . In the current study , we employ the MMTV-PyMT transgenic mouse model and the AKXD panel of recombinant inbred mice to identify AT–rich interactive domain 4B ( Arid4b; NM_194262 ) as a breast cancer progression modifier gene . Ectopic expression of Arid4b promoted primary tumor growth in vivo as well as increased migration and invasion in vitro , and the phenotype was associated with polymorphisms identified between the AKR/J and DBA/2J alleles as predicted by our genetic analyses . Stable shRNA–mediated knockdown of Arid4b caused a significant reduction in pulmonary metastases , validating a role for Arid4b as a metastasis modifier gene . ARID4B physically interacts with the breast cancer metastasis suppressor BRMS1 , and we detected differential binding of the Arid4b alleles to histone deacetylase complex members mSIN3A and mSDS3 , suggesting that the mechanism of Arid4b action likely involves interactions with chromatin modifying complexes . Downregulation of the conserved Tpx2 gene network , which is comprised of many factors regulating cell cycle and mitotic spindle biology , was observed concomitant with loss of metastatic efficiency in Arid4b knockdown cells . Consistent with our genetic analysis and in vivo experiments in our mouse model system , ARID4B expression was also an independent predictor of distant metastasis-free survival in breast cancer patients with ER+ tumors . These studies support a causative role of ARID4B in metastatic progression of breast cancer .
Breast cancer remains the most commonly diagnosed malignancy among women in the United States [1] . Because the vast majority of breast cancer related mortality is attributable to disseminated metastatic disease , a clear need exists to identify factors that modulate breast cancer metastatic progression . In addition to acquired somatic mutations , there is accumulating evidence that the genetic background on which a tumor arises can influence disease progression [2] . Identifying and characterizing metastasis susceptibility genes would provide additional insights into the mechanisms associated with tumor dissemination and growth , leading not only to better understanding of this complex process but also ultimately to new targets and strategies for clinical intervention . Due to the complex interactions between inherited factors and somatic mutations in metastatic progression , as well as the genetic complexity of human populations , identification of inherited susceptibility genes directly in human populations is difficult . To circumvent this our laboratory has chosen to apply a systems genetics approach on a mouse model of metastatic luminal breast cancer , the FVB/N-TgN ( MMTV-PyMT ) 634Mul ( MMTV-PyMT ) transgenic model . The MMTV-PyMT transgenic mouse model , which expresses the polyoma virus middle T antigen under the control of the mouse mammary tumor virus promoter , rapidly develops tumors in approximately 100% of female mammary glands and >85% of these animals develop pulmonary metastases by 14 weeks of age . When the MMTV-PyMT model is bred onto a variety of different mouse strains , the F1 progeny display broad and strain-dependent heterogeneity in primary tumor latency , primary tumor growth rate and lung metastatic density [2] . Two strains , the highly metastatic AKR/J and poorly metastatic DBA/2J , were found to have a 20-fold difference in their metastatic capacity but no significant difference in any other measured tumor phenotype . These strains were also the progenitor strains for the AKXD recombinant inbred panel of mice , which consists of more than 20 substrains that are composites of the original parental strains AKR/J and DBA/2J . The MMTV-PyMT model was therefore bred to 18 different AKXD strains , the F1 mice were phenotyped with respect to primary tumor latency and burden and lung metastatic density , and the phenotypes were compared to haplotype maps of the AKXD strains to determine quantitative trait loci ( QTLs ) associated with mammary tumor progression [3] . Subsequently , RNA was also harvested from F1 tumors and gene expression analysis was performed to define individual genes whose expression correlated with progression [4] . In this study we have utilized these resources to identify Arid4b as a novel candidate metastasis susceptibility gene . Although the precise molecular functions of ARID4B are unknown , it has been shown to associate with the SIN3A histone deacetylase ( HDAC ) complex [5] . As predicted by the genetic linkage and gene expression data , higher expression of Arid4b is associated with more rapid tumor growth in animal models , as well as increased tumor cell motility and invasion . These effects are associated with differential binding of the AKR and DBA alleles of ARID4B to HDAC complex members mSIN3A and mSDS3 . ARID4B was also found to bind the mSIN3A-associated breast cancer metastasis suppressor protein BRMS1 . Stable shRNA-mediated knockdown of Arid4b significantly inhibited the pulmonary metastatic efficiency of orthotopic mammary tumors without inhibiting primary tumor growth . Consistent with impaired metastasis in the Arid4b knockdown lines was decreased expression of a recently described metastasis-predictive gene network [6] . High expression of ARID4B was associated with an approximately 2-fold increased risk of metastatic progression in human breast cancer patients who were lymph node negative at diagnosis . Taken together these results demonstrate a causal role for Arid4b in tumor growth and metastatic progression and suggest that mechanisms of action involve modification of epigenetic state via the mSIN3A complex and regulation of the conserved Tpx2 gene network .
Previously a cross between the highly metastatic PyMT model and the AKXD recombinant inbred ( RI ) panel was performed to map QTLs associated with inherited predisposition to developing pulmonary metastasis [3] . In addition to metastasis susceptibility loci on chromosomes 6 and 19 , linkage analysis revealed a potential peak on proximal chromosome 13 ( Figure S1 ) . In a subsequent study , gene expression analysis was also carried out on these samples to examine the effect of varying metastatic genotypes on tumor transcriptional patterns [7] . To discover potential candidate genes that may affect metastatic predisposition , correlation analysis was performed using GeneNetwork [8] to identify genes whose differential expression was highly associated with metastasis . Upon integrating the data from these two studies we found that of the top ten genes most significantly associated with metastasis in our expression correlation analysis , two also mapped to potential QTLs: Ttc9c and Arid4b . The potential role of Ttc9c was investigated and no significant differences were detected with respect to orthotopic tumor growth or metastasis of 6DT1 mouse mammary carcinoma cells stably expressing Ttc9c compared to vector control cells ( data not shown ) . Similarly , we detected no significant effects on tumor growth or metastasis when Ttc9c BAC transgenic mice were bred to the MMTV-PyMT model ( data not shown ) . The most likely explanation for why Ttc9c did not pass our validation experiments is that its initial identification in our screens was a false positive owing to its close physical proximity on chromosome 19 to the metastasis modifier gene Sipa1 [9] . Our current studies have therefore focused on Arid4b , which maps within the chromosome 13 locus and whose mRNA expression was positively associated with metastatic disease and tumor growth ( Figure 1 ) , suggesting a possible causative role as a progression modifier . To validate the potential differences in Arid4b expression between strains , microarray data from AKR and DBA normal tissues were examined [4] . Consistent with the AKXD RI results , Arid4b expression was 2 . 3-fold higher in thymus ( p = 9 . 32×10−5 , FDR = 0 . 0004 ) and 2 . 5-fold higher in bone marrow ( p = 1 . 28×10−5 , FDR = 0 . 0005 ) of DBA mice compared to AKR , suggesting that constitutional polymorphisms can influence Arid4b expression levels in normal tissues . Sequence analysis was also performed to both validate SNPs in the public database as well as identify potential new variants between the AKR and DBA alleles of Arid4b . Complete exon sequencing revealed that the DBA allele matched the consensus C57BL/6 sequence . Analysis of the AKR allele revealed numerous silent SNPs as well as polymorphisms encoding eleven amino acid substitutions , as shown in Figure 2 . Interestingly , eight of these eleven polymorphisms are located in exon 22 and their encoded substitutions are densely clustered towards the C-terminal end between amino acids 1171 and 1198 . These results are consistent with the possibility that inherited variation of Arid4b may contribute to tumor progression . Analysis of the data revealed that increased Arid4b expression and increased metastatic susceptibility were associated with the DBA rather than the AKR genotype at the chromosome 13 QTL . This result suggests that the DBA allele at this locus promoted metastatic progression relative to the AKR allele , and a series of in vitro and in vivo assays were performed to test this hypothesis . V5-tagged AKR and DBA alleles of Arid4b were ectopically expressed in the mouse mammary carcinoma cell line Met-1 , which was originally derived from tumors arising in the MMTV-PyMT transgenic model [10] . Because the Met-1 line was derived from an FVB strain background , we also sequenced the FVB allele of Arid4b and found it to be identical to the DBA and C57BL/6 alleles . Cell lines were then identified that expressed the epitope tagged constructs at levels that were only two to three-fold higher than endogenous levels as measured by QRT-PCR ( Figure S2A ) , consistent with the approximately two-fold range of Arid4b mRNA between high and low metastatic AKXD strains . Furthermore , the ectopically expressed AKR and DBA alleles were detected at approximately equal levels in our stable lines as assessed by western blots ( Figure S2B ) . Orthotopic implantation assays were then performed to examine the role of Arid4b expression in vivo ( Figure 3A ) . By four weeks post-implantation , cells expressing the DBA allele formed tumors with a 2 . 6-fold larger mass compared to control cells ( 741 mg versus 284 mg; p = 6 . 08×10∧−7 ) . The AKR allele expressing cells formed tumors with a median mass of 480 mg , which was significantly larger than control tumors ( p = 0 . 010 ) but significantly smaller than the DBA cohort ( p = 7 . 73×10∧−3 ) , consistent with our previous genetic analysis and our in vitro studies . In vitro assays were performed to address the potential affect of the Arid4b polymorphisms on tumor cell behavior . In vitro growth assays demonstrated no significant difference in proliferation between cells expressing the DBA or AKR alleles or control cells ( data not shown ) . In contrast , ectopic expression of either allele significantly increased the abilities of Met-1 cells to migrate through a porous membrane and to invade through Matrigel , compared to control cells expressing lacZ ( Figure 3B ) . Notably , Met-1 cells stably expressing the DBA allele were significantly more migratory and invasive than those expressing the AKR allele . Since both cell lines express the epitope-tagged construct at approximately the same level , these results suggest a potential functional consequence for the amino acid substitutions present between the two variants in addition to the effects associated with differential expression . Because Met-1 cells are poorly metastatic in our laboratory , and because we were unable to stably overexpress Arid4b in several more aggressive mouse breast cancer cell lines , we adopted a knockdown strategy to examine the role of Arid4b in lung metastasis in vivo . To this end , the highly metastatic 6DT1 cell line [11] was transduced with five lentiviral shRNAs targeting Arid4b , or a scrambled control , and knockdown of ARID4B protein was evaluated using western blots ( Figure 4A ) and densitometry ( Figure 4B ) to select stable shRNA lines for in vivo studies . No significant knockdown was observed using the scrambled control shRNA . Cells stably transduced with Arid4b shRNAs designated H3 and H4 expressed 81% and 85% less ARID4B protein , respectively , compared to controls , and were therefore selected for further in vivo study . Following orthotopic implantation of 10∧5 cells into the mammary fat pad we observed only slight differences in median primary tumor mass between the scrambled control , H3 , and H4 cohorts , and these data did not achieve statistical significance ( p = . 070 , Kruskal-Wallis; Figure 4C ) . In contrast , we observed a 2-fold decrease in the median number of macroscopic lung metastases in the H3 cohort ( 10 vs . 22; p = . 013 ) and a 7-fold decrease in the H4 cohort ( 3 vs . 22; p = 9 . 72×10∧−5 ) compared to controls ( Figure 4D ) . Differences in lung metastasis between the two Arid4b knockdown cohorts were not statistically significant following post hoc testing ( p = . 066 , Conover-Inman ) . These data demonstrate that ARID4B protein levels are a critical determinant of pulmonary metastatic efficiency in this model system . Previous studies demonstrated that ARID4B is a member of the mSIN3A HDAC complex and that binding to mSIN3A involves the C-terminal domain of ARID4B [5] , where the majority of the amino acid substitutions were found between the AKR and DBA variants ( Figure 2 ) . Co-IP analysis was therefore performed to examine a potential effect of the observed amino acid substitutions on ARID4B-mSIN3A binding . For these experiments V5-tagged ARID4B was transiently transfected into HEK293 cells and immunoprecipitated using an anti-V5 antibody . Binding to endogenous mSIN3A and another component of the mSIN3A complex , mSDS3 [12] , was evaluated by western blots ( Figure 5 ) . Input controls for ARID4B , mSIN3A , and mSDS3 were approximately equal as were the amounts of the two Arid4b variants immunoprecipitated; however , a marked decrease in binding to mSIN3A was observed along with diminished mSDS3 association for the DBA variant ( Figure 5A ) . Densitometry analysis revealed that binding of the DBA variant was reduced by 51% and 37% for mSIN3A and mSDS3 , respectively , compared to AKR ( Figure 5B ) . These results demonstrate a functional consequence of Arid4b polymorphisms and provide insight into one potential molecular mechanism whereby Arid4b may modulate breast cancer progression . Breast cancer metastasis suppressor 1 ( BRMS1 ) belongs to the same family of proteins as mSDS3 and is known to associate with the mSIN3A complex as well as ARID4A [13] . Because ARID4B is also known to bind mSIN3A , mSDS3 , and ARID4A [14] , we postulated that ARID4B might physically bind BRMS1 . Proteomics screens to identify BRMS1 interacting proteins also support this association: in a yeast two-hybrid screen for proteins binding full-length BRMS1 , ARID4B was the number one hit identified , and ARID4A and mSDS3 were also detected ( unpublished data ) . In a separate screen , mass spectrometry was performed to identify mSIN3A binding proteins in MCF10A human breast epithelial cells . Peptides representing endogenous ARID4B and BRMS1 were detected , providing further evidence for this interaction and demonstrating that it is not simply an artifact of supraphysiologic expression in transfected cells ( Douglas Hurst; personal communication ) . To validate this interaction we performed co-IPs using lysates from 293 cells transiently transfected with the FLAG-tagged AKR or DBA variants of ARID4B along with either HA- or myc-tagged BRMS1 . HA-BRMS1 was readily detected following pull-down of ARID4B using an anti-FLAG antibody ( Figure 6A ) . Likewise , ARID4B was efficiently co-precipitated with myc-BRMS1 ( Figure 6B ) . Unlike the associations with mSIN3A and mSDS3 however , the AKR and DBA variants of ARID4B did not exhibit differential binding to BRMS1 . One possible explanation for this observation is that BRMS1 binds to a different region of ARID4B than the polymorphic C-terminal domain that mediates binding to mSIN3A . Because little is known about the specific cellular processes regulated by Arid4b that might influence the metastatic phenotype , we performed expression microarray analysis on the 6DT1 cell lines stably expressing Arid4b shRNAs to identify genes that are differentially expressed as a function of Arid4b levels . Based on the western blot densitometry shown in Figure 4A–4B cell lines expressing hairpins H3 and H4 were chosen to represent the Arid4b knockdown cohort , and the control cohort consisted of untreated 6DT1 cells and lines expressing the scrambled control shRNA or hairpin H5 . We detected 2 , 048 unique genes whose expression was significantly different ( p<0 . 05 , ANOVA ) between the two groups and those with the greatest fold change are summarized in Table 1 . While the most highly upregulated genes function in pathways with diverse biological roles , it was noted that among the most downregulated genes were multiple factors associated with centromeres ( Cenpi , Cenpq ) , microtubule and spindle dynamics ( Kif2c , Kif4a , Sass6 ) , and cell cycle regulation ( Ccne , Cdc25c ) . Consistent with this observation were the results of pathway analysis conducted to identify biological functions impacted as a consequence of Arid4b knockdown ( Table 2 ) . The most differentially regulated processes based on gene ontology were checkpoint control and DNA repair , and processes related to centrosome , centriole , and chromosome dynamics . In examining the microarray data we noticed a striking overlap between genes downregulated in the Arid4b knockdown lines and components of the TPX2 gene network . This transcriptional network was recently identified based on expression profiling of three mouse data sets and two human breast cancer data sets [6] . The TPX2 network is tumor cell-autonomous and conserved across species , its activation is predictive of reduced distant metastasis-free survival ( DMFS ) in ER-positive patients , and the nine common hub genes in the TPX2 signature ( TPX2 , BUB1 , UBE2C , CDC20 , CCNB2 , KIF2C , BUB1B , CEP55 , CENPA ) that were conserved across all five data sets consist primarily of genes involved in microtubule and mitotic spindle function . To determine how Arid4b levels influence the activation state of the TPX2 network , the fold changes of the 311 TPX2 network genes were examined in the Arid4b knockdown lines . Compared to control cell lines , 119 network genes were significantly downregulated ( p<0 . 05 ) including Tpx2 itself and the other eight common hub genes , versus only 5 network genes upregulated ( Figure 7; high resolution available as Figure S3 ) . The downregulation of this gene network concomitant with the inhibition of metastasis observed in the Arid4b knockdown lines provides further support for the role of the TPX2 network in metastatic susceptibility and suggests that a significant portion of this network may be regulated by Arid4b . Because Arid4b was identified as a candidate gene in part based on differential expression between high and low metastatic strains of mice in the AKXD panel , and because Arid4b expression levels were associated with tumor growth and metastasis in mice as well as the activity of the metastasis-associated TPX2 network , we tested whether ARID4B expression alone correlated with human patient outcomes . A search of publically available breast cancer microarray data sets using Oncomine ( Compendia Bioscience , Ann Arbor , MI ) revealed that ARID4B expression was 2 . 3-fold higher in 40 ductal breast carcinoma samples compared to 7 normal breast tissue samples in the Richardson study [15] , confirming that high ARID4B expression is clinically associated with breast cancer ( Figure S4 ) . Analysis of a pooled breast cancer dataset using GOBO ( http://co . bmc . lu . se/gobo/ ) [16] showed that among the subgroup of patients with ER-positive tumors , the cohort with high expression of ARID4B had significantly reduced DMFS compared to the low or median ARID4B cohorts ( Figure 8 ) . Because this association was significant among patients with ER-positive tumors who were lymph node negative at the time of diagnosis ( Figure 8A ) , this finding indicated that ARID4B expression level is predictive of patient progression to metastatic disease . As determined by multivariate analysis ( Figure 8B ) , the hazard ratio compared to the high ARID4B tercile was 0 . 54 for middle ARID4B ( 95% C . I . = 0 . 33–0 . 89; p = . 015 ) and 0 . 42 for the low ARID4B tercile ( 95% C . I . = 0 . 26–0 . 70; p = 7 . 51×10∧−4 ) , indicating that patients with tumors expressing high levels of ARID4B are approximately twice as likely to develop metastatic disease . The association of ARID4B with reduced DMFS was also highly significant among ER-positive patients not receiving adjuvant therapy ( Figure 8C–8D ) , indicating that ARID4B expression level plays a significant role in the natural metastatic progression of ER-positive breast cancer in human patients and its relevance is not confined solely to our mouse model systems .
Arid4b was identified as a candidate gene of interest through linkage and expression correlation analyses , and the in vitro and in vivo data presented here provide the first direct evidence of a causal role of Arid4b in mammary tumor progression and metastasis . The initial QTL analysis revealed association with Arid4b on proximal chromosome 13 , and Arid4b was among the most highly correlated genes with the most significant p values in the subsequent eQTL analysis in the AKXD recombinant inbred panel . It was noted that although AKR/J is the more highly metastatic of the two parental strains , progression was associated with the DBA/2J allele , suggesting that the metastasis promoting influence of Arid4b is likely masked by other suppressive factors in a pure DBA/2J background . Although the AKXD recombinant inbred panel lacks the power to detect these epistatic interactions , ongoing experiments using the latest generation of recombinant inbred mice including the Collaborative Cross [17] , [18] will enable higher resolution QTL mapping and more robust systems genetics analyses going forward . Mouse Arid4b encodes a protein of 1314 amino acids that shares 89% identity and 95% similarity to the 1312 amino acid human protein . Alternate nomenclature includes breast cancer-associated antigen 1 ( BRCAA1 ) , retinoblastoma-binding protein-1-like protein-1 ( RBP1L1 ) , and mSIN3A-associated protein of 180 kDa ( SAP180 ) . Indeed , there are multiple lines of evidence implicating Arid4b in breast cancer . A ten amino acid peptide was found to represent an antigen epitope expressed in 65% of breast cancer specimens and was significantly upregulated in the sera of breast cancer patients compared to healthy donors [19] . ARID4B was also found to associate with the mSIN3A HDAC complex [5] , which is in turn known to be bound by the breast cancer associated tumor suppressor ING1 [20] , [21] , the well-characterized breast cancer metastasis suppressor BRMS1 [13] , and the ARID family homolog ARID4A/RBP1 [22] , which has also been identified as a breast cancer associated antigen [23] . Ectopic expression of Arid4b at a physiologically relevant two- to three-fold increased level resulted in a 3-fold increase in orthotopic tumor mass relative to controls for Met-1 cells expressing the DBA allele , while the AKR allele induced 1 . 9-fold larger tumors versus control cells . To our knowledge , this is the first direct evidence that Arid4b upregulation promotes tumor growth . Although Met-1 orthotopic tumors do not readily metastasize in our experience , transwell assays in vitro demonstrated that upregulation of either allele of Arid4b increased tumor cell migration and invasion , consistent with a role of Arid4b in metastatic progression , and cells expressing the DBA allele were significantly more migratory and invasive than cells expressing the AKR allele . While stable upregulation of Arid4b did not induce Met-1 cells to metastasize with any greater frequency , stable knockdown of Arid4b in the highly metastatic 6DT1 cell line did cause a dramatic reduction in pulmonary metastases , raising the possibility that ARID4B may represent a novel therapeutic target . Taken together , the results of the orthotopic implantation and transwell assays are broadly consistent with our genetic linkage and expression correlation analyses that showed an association of the DBA haplotype on chromosome 13 with metastatic progression in the MMTV-PyMT×AKXD mice , and validate a functional role of Arid4b polymorphism in modulating the breast cancer phenotype . While the molecular mechanisms of Arid4b are incompletely understood , an examination of its sequence and conserved domains provides further insight into its potential functions . Arid4b contains a nuclear localization signal ( NLS ) towards the C-terminus as well as conserved Tudor , RBB1NT , ARID/BRIGHT , and Chromo domains in the N-terminal half of the protein . The ARID domain mediates binding to DNA , although the affinity for AT-rich sequences varies among members of the Arid superfamily [24] , [25] . The RBB1NT domain is present in many Rb binding proteins including ARID4A , although it is noteworthy that unlike ARID4A , ARID4B does not contain the LCXCE motif necessary for RB binding [26] , and no interaction was observed when we attempted to co-IP ARID4B with RB ( data not shown ) ; therefore , the function of the RBB1NT domain of ARID4B remains uncertain . Tudor domains are present in many RNA binding proteins [27] and also bind methylated lysine residues on histone tails [28] . Chromo domain-containing proteins have also been shown to bind methylated lysines and mediate the recruitment of chromatin modifying complexes [29] . Because mSIN3A itself lacks intrinsic DNA binding capability , targeting of mSIN3A-associated HDAC activity depends on interactions with other transcription factors including Mad1 and KLF repressors among others [30] . The presence of putative DNA and histone binding domains in the N-terminal half of ARID4B suggest that its influence on mammary tumor progression involves directing the HDAC activity of mSIN3A complexes to chromatin . This is supported by our observations that the high and low metastatic alleles of ARID4B have a dense cluster of amino acid polymorphisms in the C-terminal domain and bind with different affinities to mSIN3A and mSDS3 , though the biochemical significance of this observation remains to be determined . Diminished expression levels or binding affinity of ARID4B may allow mSIN3A to be bound by other proteins with different DNA sequence specificity , perhaps not resulting in a global change in the abundance of any one particular histone mark but rather altering the expression of different subsets of genes . It is noteworthy that the pro-metastatic ARID4B and the metastasis suppressive BRMS1 bind each other and also to the mSIN3A complex in vitro . This observation reinforces the significance of the mSIN3A complex in metastatic progression , and it is tempting to speculate that an HDAC complex may be caught in a molecular tug-of-war between these two metastasis modifier genes . However , the mSIN3A complex is modular in nature and interacts with a great variety of transcriptional regulators [30] , [31] . Many different complexes exist , and their precise composition and function within the context of breast cancer are not well understood . Further studies will be necessary to define a role , if any , for the ARID4B-BRMS1 interaction in human disease . While ARID4B expression was not a significant predictor of DMFS across all patients in a meta-analysis of an 1 , 881 sample data set , statistical significance emerged when patients were stratified based on ER status . The observation that ARID4B is predictive of metastatic progression only in ER+ patients is consistent with the identification of Arid4b as a candidate gene in the context of the MMTV-PyMT mouse model system , in which tumors arise from a predominantly ER+ luminal epithelial cell population [32] . Loss of ER and PR is detected during progression to late carcinomas , however in a systematic analysis of gene expression profiles these late PyMT tumors clustered most closely with human luminal tumors [33] , which are ER+ . Also consistent with ARID4B promoting metastatic progression of ER+ tumors is our observation that Arid4b knockdown caused a significant downregulation of the core components of the Tpx2 gene network . The TPX2 signature was tumor cell autonomous and predictive of DMFS only in those patients who were ER+ at diagnosis , and was distinct from a CD53 network that was associated with ER-negative stromal components [6] . Polymorphisms in several other tumor cell autonomous metastasis susceptibility genes identified in our laboratory including Sipa1 , Rrp1b , and Brd4 are prognostic only in ER+ patients [9] , [34]–[37] , and stable expression of Brd4 can also differentially regulate the Tpx2 network [6] . The association of multiple metastasis susceptibility genes with a transcriptional network comprising many cell cycle and mitotic spindle checkpoint regulatory genes highlights the possibility that these cellular functions are critical determinants of metastatic efficiency . Further experiments are underway in our laboratory to determine whether upregulation of the TPX2 network is causative in promoting metastasis .
Using genomic DNA from AKR/J and DBA/2J mice as templates , PCR was performed to amplify the protein coding region , exons 2 through 24 ( Table S1 ) . PCR Products were subjected to agarose gel electrophoresis , bands isolated using the QIAquick Gel Extraction kit ( Qiagen ) according to manufacturer's recommendations , and used as templates for sequencing . All sequencing runs were performed by the DNA Sequencing and Gene Expression Core , NCI , Bethesda , MD . Genomic sequences for the AKR and DBA alleles were aligned using pairwise BLAST [38] and non-synonymous polymorphisms verified by manual comparison of chromatograms using Chromas software ( Technelysium ) . V5-tagged AKR and DBA alleles of Arid4b were generated using long range PCR with forward primer 5′-AACAAAGGTGCAGGTGAAGC-3′ and reverse primer 5′-CCTGCACTCAACTGACATTCCATTC-3′ to amplify Arid4b , and PCR products were cloned into pcDNA3 . 1/V5-His-TOPO ( Invitrogen ) . FLAG-tagged Arid4b vectors were constructed by the Protein Expression Laboratory , SAIC-Frederick , Inc . using Gateway technology ( Invitrogen ) . Briefly , the AKR or DBA allele of Arid4b was PCR amplified and cloned into entry vector pDonr-253 , then subcloned by Gateway LR recombination into pDest-737 to generate an expression construct with CMV promoter and N-terminal 3xFLAG tag . Full-length BRMS1 was epitope tagged at the N-terminus by PCR with the HA or myc tag sequence incorporated into the forward primer and cloned into pCMV or pcDNA3-hygro ( Invitrogen ) , respectively . Correct sequences of all vectors were confirmed prior to use . Met-1 cells [10] were a gift from Dr . Robert Cardiff ( University of California , Davis , CA ) . 6DT1 cells [11] were a gift from Dr . Lalage Wakefield ( NCI , NIH , Bethesda , MD ) . HEK293 cells were purchased from ATCC ( Manassas , VA ) . Cell lines were maintained in DMEM supplemented with 10% FBS , 2 mM L-glutamine , penicillin and streptomycin . Cells were confirmed to be free of mycoplasma contamination using the MycoAlert detection kit ( Lonza ) . Met-1 cells seeded onto 10 cm tissue culture plates were co-transfected with 6 µg of the appropriate V5-tagged Arid4b construct described above , or pcDNA3 . 1/V5-His-TOPO/lacZ ( Invitrogen ) as a control vector , plus 600 ng of pSuper . Retro . Puro ( Oligoengine ) as a selectable marker , using FuGENE 6 transfection reagent . Cells were selected using 1 mg/ml G418 plus 4 µg/ml puromycin and clones derived by limiting dilution . Stable upregulation of Arid4b was verified by performing QRT-PCR using forward primer 5′- GGTGAGTGGGAGCTGGTCTA-3′ and reverse primer 5′- ATAAAGGGCCCACTGAAGGT-3′ , and western blotting for endogenous and ectopically expressed ARID4B as described below . 6DT1 cells were transduced with one of five different lentiviral shRNAs targeting Arid4b ( RMM4534-NM_194262 , Open Biosystems ) or a scrambled control shRNA in the same pLKO . 1 vector . Stable cells were selected using 10 µg/ml puromycin and pooled clones were analyzed for Arid4b knockdown by western blot . Met-1 or 6DT1 stable lines were orthotopically implanted into the fourth mammary fat pad of six week old female NU/J or FVB/NJ mice using 105 cells suspended in 100 µl of PBS per animal . Primary tumors and lungs were harvested 28 days later . All experiments were performed according to the National Cancer Institute Animal Care and Use Committee guidelines . Met-1 cells stably expressing Arid4b or lacZ were seeded at 75 , 000 cells per well into invasion chambers coated with Matrigel basement membrane matrix ( 534480 , BD Biosciences ) or control chambers lacking Matrigel ( 354578 , BD Biosciences ) . After 24 hours , cells were fixed in 100% methanol , stained with crystal violet , and mounted onto glass slides using mineral oil . Cells were visualized at 400× magnification and five fields were counted for each of three experiments . HEK293 cells were transfected with the V5- or FLAG-tagged AKR or DBA allele of Arid4b , with or without HA-BRMS1 or myc-BRMS1 where appropriate , using FuGENE 6 transfection reagent . After 30 hours , cells were harvested in mild IP lysis buffer ( 25 mM Tris-HCl pH 7 . 4 , 150 mM NaCl , 1 mM EDTA , 1% NP-40 , 5% glycerol ) supplemented with protease inhibitors ( 11836170001 , Roche ) and phosphatase inhibitors ( P-5726 , Sigma ) . Protein samples were quantitated using Bradford assays . Gammabind G Sepharose beads ( 17088501 , GE Healthcare ) were washed twice in NET buffer ( 50 mM Tris pH 8 . 0 , 150 mM NaCl , 5 mM EDTA , 1% NP-40 , 0 . 5% BSA , 0 . 04% sodium azide ) supplemented with protease and phosphatase inhibitors , and resuspended to form a 50% bead slurry . Lysates were precleared by adding 40 µl of bead slurry and rotating for 30 minutes at 4°C . Samples were centrifuged at 10 , 000 rpm for 1 minute at 4°C and pre-cleared supernatant transferred to a fresh tube . Anti-V5 , anti-FLAG , or anti-myc tag antibodies was added to a final concentration of 1 . 0 µg/ml and samples rotated for 1 hour at 4°C , then 50 µl of bead slurry was added and co-IPs performed overnight at 4°C . Beads were then washed four times with NET buffer and resuspended in SDS-PAGE sample buffer . NuPAGE precast gels and buffers ( Invitrogen ) were used according to manufacturers recommendations and gels were transferred onto Immobilon-P ( Millipore ) . Membranes were blocked in TBS with 0 . 5% Tween-20 ( TBST ) plus 5% nonfat dry milk for 1 hour at room temperature then incubated with the primary antibody diluted in blocking buffer overnight at 4°C . The following primary antibodies and concentrations were used: anti-Arid4b ( 1∶3 , 000; A302-233A , Bethyl Laboratories ) , anti-V5 ( 1∶5 , 000; 37–7500 , Invitrogen ) , anti-mSIN3A ( 1∶1 , 000; sc-994 , Santa Cruz ) , anti-mSDS3 ( 1∶2 , 000; A300-235A , Bethyl Labs ) , anti-β-actin ( 1∶10 , 000; ab6276 , Abcam ) , anti-FLAG ( 1∶3 , 000; F-3165 , Sigma ) , anti-HA ( 1∶5 , 000; 11867423001 , Roche ) , anti-myc tag ( 1∶2 , 000; 2276 , Cell Signaling ) . After three washes in TBST , membranes were incubated with one of the following horseradish peroxidase conjugated secondary antibodies diluted in TBST plus 0 . 5% milk for 1 hour at room temperature: anti-mouse ( 1∶5 , 000; NA931V , GE Healthcare ) , anti-rabbit ( 1∶10 , 000; sc-2004 , Santa Cruz ) , anti-goat ( 1∶10 , 000; sc-2304 , Santa Cruz ) . Membranes were washed an additional three times in TBST and proteins detected using the Amersham ECL Plus system ( RPN2132 , GE Healthcare ) and Amersham Hyperfilm ECL ( 28906837 , GE Healthcare ) according to manufacturer's recommendations . Densitometry data were collected and analyzed using a ChemiDoc-It Imaging System and VisionWorksLS software ( UVP ) . Total RNA was isolated from pooled clones of 6DT1 Arid4b knockdown cell lines using RNeasy kits ( Qiagen ) and then arrayed on Affymetrix GeneChip Mouse Gene 1 . 0 ST arrays by the Microarray Core in the NCI Laboratory of Molecular Technology . Expression data were normalized using Partek Genomics Suite to identify genes whose expression was significantly different ( p< . 05 ) between the Arid4b normal cohort ( untreated , scrambled control , and H5 lines ) and the Arid4b knockdown cohort ( lines H3 and H4 ) . The gene list and expression values were then analyzed using Ingenuity Pathways Analysis ( Ingenuity Systems , www . ingenuity . com ) to identify differentially regulated signaling pathways and biological functions . Expression of the Tpx2 transcriptional network was visualized and figure generated using Cytoscape software [39] . Microarray data are available through the Gene Expression Omnibus under accession number GSE35731 .
|
A person's individual genetic background influences not only the likelihood of developing breast cancer , but also the likelihood of that cancer becoming metastatic . The identification of metastasis susceptibility genes using human samples is rendered impractical by the high degree of genetic diversity among people . Our laboratory's strategy is to cross genetically defined inbred mouse strains to recapitulate a degree of genetic diversity that is more readily studied . By breeding these panels of inbred mouse crosses to a mouse model of breast cancer , we can identify regions of the genome that correlate with observed phenotypic variation including metastatic density and then identify individual candidate genes . This manuscript describes the identification of Arid4b as a candidate gene of interest and the experiments we performed to validate its role in metastasis . High expression of Arid4b enhances cell migration and invasion and , conversely , knockdown of Arid4b inhibits metastasis of breast tumor cells to the lungs . The mouse gene and human ARID4B are highly conserved , and among women with ER+ tumors ARID4B expression level is predictive of which patients will progress to develop metastatic disease .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"animal",
"models",
"systems",
"biology",
"animal",
"genetics",
"cancer",
"genetics",
"model",
"organisms",
"mouse",
"genetics",
"biology",
"gene",
"networks",
"genetics",
"and",
"genomics"
] |
2012
|
Allelic Variation and Differential Expression of the mSIN3A Histone Deacetylase Complex Gene Arid4b Promote Mammary Tumor Growth and Metastasis
|
Integrated Information Theory ( IIT ) has become nowadays the most sensible general theory of consciousness . In addition to very important statements , it opens the door for an abstract ( mathematical ) formulation of the theory . Given a mechanism in a particular state , IIT identifies a conscious experience with a conceptual structure , an informational object which exists , is composed of identified parts , is informative , integrated and maximally irreducible . This paper introduces a space-time continuous version of the concept of integrated information . To this aim , a graph and a dynamical systems treatment is used to define , for a given mechanism in a state for which a dynamics is settled , an Informational Structure , which is associated to the global attractor at each time of the system . By definition , the informational structure determines all the past and future behavior of the system , possesses an informational nature and , moreover , enriches all the points of the phase space with cause-effect power by means of its associated Informational Field . A detailed description of its inner structure by invariants and connections between them allows to associate a transition probability matrix to each informational structure and to develop a measure for the level of integrated information of the system .
Dynamical Systems and Graph Theory are naturally coupled since any real phenomenon is usually described as a complex graph in which the evolution of time produces changes in specific measures on nodes or links among them [1 , 2] . In this work , the starting point is any structural network , including a parcelling of the brain , possessing an intrinsic dynamics . For brain dynamics , the collective behavior of a group of neurons can be represented as a node with a particular dynamics along time [3–6] . In general , the mathematical way to describe and characterize dynamics is by ( ordinary or partial ) differential equations ( continuous time ) [7] or difference equations ( discrete time ) [8] . Global models on brain dynamics are grounded on anatomical structural networks built under parcelling of the brain surface [9–11] . Indeed , they are based on systems of differential equations described on complex networks , which may include noise , delays , and time-dependent coefficients . Thus , the designed dynamical system models the activity of nodes connected to each other by a given adjacency matrix . A global dynamics emerges through simulated dynamics at each node , which is coupled to others as detailed in the anatomical structural network ( see , for instance , [12] for the structural networks on primate connectivity ) . Then , an empirical functional network and a simulated functional network emerge by correlation or synchronization of data on the structural network [4 , 13 , 14] , showing a similar behavior and topology after a proper fitting of the parameters in the differential equations associated to the dynamics . We take advantage of this approach to apply some of the main results on the modern theory of dynamical systems showing that , given a dynamics on a network , there exists an object , the global attractor [15–18] , determining all the asymptotic behaviour of each state of the network . The attractor exists and its nature is essentially informational , as it possesses the power to produce a curvature of the phase space enriching every point with the information on its possible past and future dynamics . The structure of the global attractors ( or attracting complex networks [19 , 20] ) , described as composed structures by invariants and connections , naturally shows that its information is structured , composed by different parts , and can be unreachable from the study of the information of its parts , so allowing for a definition of integrated information . Integrated Information Theory ( IIT , [21] ) , created by G . Tononi [22–24] starts with a phenomenological approach to the theory of consciousness . It assumes that consciousness exists and tries to describe it by defining the axioms that it satisfies . Having the axioms on hand , they serve to introduce the postulates that every physical mechanism has to obey in order to produce a conscious experience . This fact opens the door to the possibility of the mathematization of the theory by defining and describing postulates on concrete networks where a dynamics can be settled . It is then possible to define the appropriate structured dynamics which is supposed to explain a conscious experience by preserving its axioms . The IIT approach allows to represent a conscious experience and even to measure it quantitatively and qualitatively by the so called integrated information Φmax , which , at the same time indicates that , at the base of consciousness , there are essentially phenomena of causal processes of integrated information nature [25] . This fact links IIT to Information Theory and the Theory of Causality [26] . On the mathematical level , IIT approach is based on graphs consisting of logic gates and transition probabilities describing causality of consecutive discrete states on those graphs [21] . In this paper we present a continuous-time version ( see Section Results for a formal description ) related to IIT based on the theory of dynamical systems . IIT bases any particular experience on a mechanism , defined in a particular state , which possesses a well defined cause-effect power . Our starting point is given by a graph describing a mechanism and , thus , a graph is first defined . However , we focus our study on the network patterns arising from dynamical phenomena . So , a dynamical system and the associated mathematical objects ( global attractor , equilibrium points , unstable invariant sets ) have to be also defined . As a novelty in dynamical systems theory , the global attractor ( which , for the gradient case consists of the equilibria and the heteroclinic connections between them [15 , 17 , 27] ) is redefined as an object of informational nature , an Informational Structure ( IS ) . An IS is a flow-invariant object of the phase space described by a set of selected invariant global solutions of the associated dynamical system , such as stationary points ( equilibria ) and connecting orbits among them ( Fig 1 ) . This set of invariants inside the IS creates a new structure , a new complex network with the power to ascertain the dynamics ( past and future scenarios ) of natural phenomena . Every IS posseses an associated Informational Field ( IF ) , globally described from the attraction and repulsion rates on the nodes of the IS . We are able to translate the energy landscape caused by the IS and the IF into a transition probability matrix ( TPM ) to pass from one state to another within the system ( see Section Results ) . Thus , the level of information of a mechanism in a state is going to be given by the global amount of deformation of the phase space caused by the intrinsic power of the IS and IF . The geometrical characterization of ISs can provide both the quality of the related information and , in particular , the shape in which it is integrated in the whole system , allowing to measure the level of integrated information it contains . Thus , the quality of the information comes from the detailed study of informational structure , which now possesses an intrinsic dynamics and enjoys a continuous change . This structure depends on the parameters of the underlying equations and has the ability to possibly rapidly change in the response to the change of those parameters ( see Section Materials and methods ) . From this continuous approach , we are able to introduce first definitions for postulates of existence , composition , information and integration for a mechanism in a state . There is still a gap to the more elaborated formal definitions from IIT 3 . 0 [21] ( see Section Discussion ) , including the composition and exclusion postulates . However , our framework naturally leads to a study on the continuous dependence between the topology of the network and the level of integrated information for a given mechanism ( see Section Results ) .
Many real phenomena can be described by a set of key nodes and their associated connections , building a ( generically ) complex network . In this way , we can always construct an application between a real situation and an abstract graph describing its essential skeleton . An undirected graph is an ordered pair G = ( V , E ) comprising a non-empty set V of vertices ( or nodes ) together with a set E of edges joining 2-element subsets of V . The order of a graph is given by the number of nodes , and its size by the number of edges . A directed graph or digraph is a graph in which edges ( named arcs ) have orientations . We want to study the behaviour on networks of systems of evolutionary differential equations as d u d t = F ( t , u ) , ( 1 ) where F is a nonlinear map from ( t , u ) ∈ R × R N to R N . For modeling purposes , we could also add , for instance , delays , stochastic terms , or to make solution u ( t ) also depend on a subset Ω of the three dimensional space , i . e . u ( t , x ) , for x ∈ Ω ⊂ R 3 . Given an initial condition , suppose existence and uniqueness of solutions . If not , a multivalued approach could also be adapted . The phase space X ( in our case X = R N ) represents the framework in which the dynamics described by a group of transformations S ( t ) : X → X is developed . Given a phase space X we define a dynamical system on X as a family of non-linear operators { S ( t ) } t ∈ R + , S ( t ) : X → X u ∈ X , S ( t ) u ∈ X which describes the dynamics of each element u ∈ X . In particular , S ( t ) u0 = u ( t;u0 ) is the solution of the differential Eq ( 1 ) at time t with initial condition u0 . The global attractor is the central concept in dynamical system theory , since it describes all the future scenarios of a dynamical system . It is defined as follows [15–18 , 28 , 29]: A set A ⊆ X is a global attractor for {S ( t ) : t ≥ 0} if it is Observe that ( ii ) is showing a crucial property of an attractor , as supposes a set with a proper intrinsic dynamics . Moreover , ( iii ) points that this set is determining all the future dynamics on the phase space X . We say that u* ∈ X is an equilibrium point ( or stationary solution ) for the semigroup S ( t ) if S ( t ) u* = u* , for all t ≥ 0 . A stationary point is a trivial case for a global solution associated to S ( t ) , i . e . , ξ : R → X such that ξ ( t + s ) = S ( t ) ξ ( s ) for all s∈R , t∈R+ . Stationary points are the minimal invariant objects inside a global attractor . Every invariant set is a subset of the global attractor [15] . Generically , connections among invariant sets in the attractor describe its structure [27 , 30] . To this aim we need the following definitions , which also allow us to define the behaviour towards the past in a global attractor . The unstable set of an invariant set Ξ is defined by W u ( Ξ ) = { z ∈ X : there is a global solution ξ : R → X for S ( t ) satisfying ξ ( 0 ) = z and such that lim t → − ∞ dist ( ξ ( t ) , Ξ ) = 0 } . The stable set of an invariant set Ξ is defined by W s ( Ξ ) = { z ∈ X : such that lim t → + ∞ dist ( S ( t ) z , Ξ ) = 0 } . We have to think in a global attractor as a set which does not depend on initial conditions , with an intrinsic proper dynamics , composed by a set of special solutions ( global solutions ) , which are connecting particular invariants , so generating a complex directed graph . Moreover , the global attractor has the following properties [17 , 18]: The Fundamental Theorem of Dynamical Systems [33] states that every dynamical system on a compact metric space X ( the one defined on a global attractor , for instance ) has a geometrical structure described by a ( finite or countable ) number ( indexed by I ) of sets {Ei}i∈I with an intrinsic recurrent dynamics and a gradient-like dynamics outside them . In other words , when we define a dynamical system on a graph , the attractor can be always described by a ( finite or countable ) number of invariants and connections between them . The starting point in our approach is a system of connected elements where a dynamics is defined . This system is called a mechanism . Composition and exclusion postulates in IIT 3 . 0 allow to consider mechanisms given by any subset of the system . Here , we are only focusing in the mechanism given by the whole system . A global attracting network can be characterized by the amount of information it provides to the mechanism , since the nature of a global attractor is essentially informational . Indeed , the informational nature of the attractor is based on the following assertions: The state of a mechanism is given by the state of its nodes . If those nodes take real values , the state is given by a vector of real numbers . When a dynamical system is defined for a mechanism in a state , it has cause-effect power , meaning that it conditions the possible past and future states of the mechanism . We can find the cause-effect power of a mechanism in a state by looking at its ( local ) attracting invariant sets . Typically , these are stationary points or periodic orbits [32 , 39–41] , but it could also contain invariant sets with chaotic dynamics [42–44] . Following IIT , we will measure the amount of information on the structure made by these invariants , in the sense of its power to restrict the past and the future states . In IIT language , this is intrinsic information: differences ( state of a mechanism ) that make a difference ( restricted set of possible past and future states ) within a system ( see the IIT glossary in [21] ) . Now we can define an Informational Structure ( IS ) . Suppose we have a complex graph G given by N nodes and links among them . We denote by Gi every subset of G . An informational structure is a complex graph I = { G 1 , ⋯ , G M } which nodes are subgraphs Gi of G and links among them , with the following properties An IS ( see Fig 2 ) becomes a natural emerging object ( of informational nature ) from a given mechanism in a state with intrinsic cause-effect power to the past and the future . Moreover , some properties of ISs can be given:
In this section we define a preliminary version of IIT postulates: existence , composition , information , integration and exclusion . We do not try to mimic or generalize the concepts of IIT 3 . 0 ( see Section Discussion ) , where they appear in a more elaborated fashion , allowing finer computations and insights . For example , IIT considers the integrated information of each subset of a mechanism ( candidate set ) , while we always focus on the whole mechanism . Although ISs and mechanisms possess quantitative and qualitative major differences on the structure and topology of both networks , it is clear that ISs possess an strongly dependence of their mechanisms’ topology . To show this dependence ( but no determination ) between mechanisms and associated ISs , we have tried to model the continuous evolution of integrated information for simplified mechanisms . This is probably one of the virtues of our continuous approach to integrated information . In particular , we consider the cases of totally disconnected mechanisms , lattice ones , the presence of a hub , and totally connected mechanisms , showing the key functions of the topology and strength of connections with respect to integrated information . To allow the comparison of the different topologies , the reference value for αi is 1 . 6 and for γij is 0 . 1 . The behaviour of ϕ-cause and ϕ-effect when some of these parameters change is shown for the different mechanisms .
The concept of informational structure is not only related to the understanding of brain processes and their functionality but , more broadly , as an inescapable tool when analyzing dynamics in complex networks [19 , 20] . For instance , in Theoretical Ecology and Economy related to the modeling of mutualistic systems [61 , 62] , a very important subject is the study of the dependence between the topology of complex networks ( lattice systems , unconnected systems , totally connected ones , random ones , “small world” networks ) and the observed dynamics on their sets of solutions . The architecture of biodiversity [63 , 64] is thus referred as the way in which cooperative systems of plants-animals structure their connections in order to get a mechanism ( complex graph ) achieving optimal levels in robustness and life abundance . The nested organization of this kind of complex networks seems to play a key role for higher biodiversity . However , it is clear that the topology of these networks is not determining all the future dynamics [65] , which seems also to be coupled to other inputs as modularity [66] or the strength of parameters [67] . This is also a very important task in Neuroscience [54 , 68] . Informational structures associated to phenomena described by dynamics on complex networks contain all the future options for the evolution of the phenomena , as they possess all the information on the forwards global behaviour of every state of the system and the paths to arrive into them . Moreover , the continuous dependence on the parameters measuring the strength of connections and the characterization of the informational structure allow to understand the evolution of the dynamics of the system as a coherent process , whose information is structured , so giving a comprehension to the appearance of sudden bifurcations [69] . From a dynamical system point of view [44] , informational structures introduce several conceptual novelties in the standard theory . In particular , for phenomena associated with high dimensional networks or models by partial differential equations [15 , 18 , 29] . On the one hand , an informational structure has to be understood as a global attractor at each time , i . e . , we do not reach this compact set as the result of the long time behaviour of solutions , but it exists as a complex set made by selected solutions which is causing a curvature of the whole phase space at each time . Because we allow parameters to move in time , the informational structure also depends on time , so showing a continuous deformation of the phase space by time evolution . It is a real space-time process explaining all the possible forwards and internal backwards dynamics . On the other hand , the understanding of attracting networks as informational objects is also new in this framework , allowing a migration of this subject into Information Theory . It is remarkable that dynamical systems cover from trivial dynamics , as global convergence to a stationary point , to the much more richer one usually referred as chaotic dynamics [43 , 44 , 57] and dynamics of catastrophes [70] . While the first one can be found with total generality , attractors with chaotic dynamics can be only described in detail in low dimensional systems . Note that some of the classical concepts of dynamical systems have been reinterpreted . In particular , the global attractor it is understood as a set of selected solutions ( organized in invariant sets and their connections in the past and future ) which creates the informational structure . This network , moreover , produces a global transformation of the phase space ( the informational field ) enriching every point in the phase space on the crucial information on possible past states and possible future states . It is the continuous change on time of informational structures and fields what allows to talk on an informational flow for which we can analyse the postulates of IIT . Our approach deals with dynamics on a graph for which a global description of the asymptotic behaviour is posible by means of the existence of a global attractor . That is , dissipative dynamical system for which a global attractor exists . It is important to remark the Fundamental Theorem of Dynamical Systems in [33] , inspired in the work of Conley , because it gives a general characterisation of the phase space by gradient and recurrent points , so leading to a global description of the phase space as invariants and connections among them . This is the really crucial point which allows for a general framework on the kind of systems to be considered . The Lotka-Volterra cooperative model we consider is for the applications to IIT , because in this case we have a detailed description of the global attractor , which allows to make concrete computations for the level of integrated information on it . Thus , in application to L-V systems , we are considering heteroclinic channels inside the global attractor . But all the treatment in the paper can be done for every dissipative systems for which a detailed description of the global attractor is available , i . e . , this description has not be to be restricted to heteroclinic channels coming from L-V models . Gradient dynamics [15 , 27 , 29 , 31] describes the dynamics from heteroclinic connections and associated Lyapunov functionals [29 , 32] , and it naturally suits into higher order systems including infinite dimensional phase spaces [27 , 34 , 35] . Thus , although our description of informational structures associated to Lotka-Volterra systems is general enough to describe real complex networks , we think the concept may be also well adapted both for different topologies in the networks and also different kind of non-linearities defining the differential equations [51 , 52] . Actually , for comparison with data and experiments associated for real brain dynamics we certainly need to allow much more complex networks as primary mechanisms as well as different kinds of dynamical rules and nonlinearities . But , in all of them , we expect to find the existence of dynamical informational structures providing the information on the global evolution of the system . The concise geometrical description of these structures and their continuous dependence on sets of parameters is a real open problem in the dynamical systems theory . We have presented a preliminary approach to the notion of integrated information from informational structures , leading to a continuous framework for a theory inspired by IIT , in particular in order to define integrated information of a mechanism in a state . The detailed characterization of informational structures comes from strong theorems in dynamical systems theory about the characterization of global attractors . The description of causation in an informational structure by a coherent and continuous flow of integrated information is new in the literature . Indeed , the definition of TPMs between two equilibria in the IS by the strength of eigenvalues and direction of eigenvectors associated to these and intermediate equilibria produces a global analysis of the level of information contained in a particular IS . The same procedure for partitions of the associated mechanism allows a preliminary measure for integration . Moreover , the continuous dependence of the structure of ISs on parameters also opens the door to the analysis to sudden bifurcations , from an intrinsic point of view , for critical values of the parameters . It seems also very clear the close dependence between the topology of a mechanism , the actual value of the parameters and the current state with respect to its level of integrated information , so pointing for optimality for a “small world” configuration of the brain [71] . With respect to global brain dynamics , we think the concept of informational structures and informational fields could serve as the abstract objects to describe functionality and global observed dynamics now described by other concepts and methodologies as multistability [6] or metastability [72 , 73] . In [53] the notion of ghost attractors is used , which would seem to suit into our informational fields and its bifurcations under the movement of the coupling parameter in the connectivity matrix . We think it could also serve as a valuable tool in line of the perspectives developed in [5] . In [54] a detailed study of different attractor repertoires is studied for a global brain network for which a ( local , global and dynamic ) Hopfield model is defined . Note that , depending on parameters , the authors observe the bifurcation of dozens or even hundreds of stationary hyperbolic fixed points , which basis of attraction and stability properties has to be analyzed in an heuristic way by simulation of thousands of realizations related to initial conditions . The comparison with empirical data occurs at the edge of multistability . All of these phenomena could naturally enter into a broader global approach , by the study of an abstract formulation of their associated informational fields , creating a network of attractors and their connections , and their dependence on parameters to better understand transitions and bifurcations of structures . Moreover , the continuous approach for ISs possibly leads , if we are able to fit them to real data of brain activity , to a useful tool to analyze the functional connectivity dynamics [74] . The ability to mathematically represent human consciousness can help to understand the nature of conscious processes in human mind , and the dynamical systems approach may indeed be also a correct tool to start this trip towards a complementary mathematical approach to consciousness . Indeed , informational structures allow to associate the processes underlying consciousness to a huge functional and changing set of continuous flow structures . However , we think we are still far to the aim of describing a conscious experience with the actual development , which will be continued in the near future . We have introduced the postulates for IIT associated to ISs , and , in particular , we have developed a first approach for definitions of information and integration , leading to the introduction of measures as cei ( cause-effect information ) or ϕ ( integrated information of a mechanism ) . However , in order to make this approximation comparable to the latest published IIT version for discrete dynamical systems of this theory [21] , a series of additional developments are required: In IIT 3 . 0 the exclusion postulate is applied to the causes and effects of the individual mechanisms , so that only the purview that generates the maximum value of integrated information intrinsically exists as the core cause ( or core effect ) of the candidate mechanism that generates a concept . In fact , a concept is defined as a mechanism in a state that specifies a maximally irreducible cause and effect , the core cause and the core effect , and both specify what the concept is about . The core cause ( or effect ) can be any set of elements , but only the set that generates the maximum value of integrated information in the past ( or future ) exists intrinsically as such . We can see the present work as a simplified particular case in which the candidates to core cause and effect , the candidate set to be a concept and the complete system match . But in general they will be different sets . In addition , elements outside the candidate set should be treated as background conditions . On the other hand , the partitions in this work isolate one part of the system from another for any instant of time . However , in line with IIT 3 . 0 , partitions should be performed in this way: two consecutive time steps are considered ( past and present or present and future ) and the elements of the candidate mechanism in the present are joined to the elements of the candidate to core cause in the past ( or core effect in the future ) to form a set that will be separated in two , so that in each of the parts there may be elements of the two time steps or not . It is also necessary to distinguish between mechanisms for which the value of ϕ for each concept or quale sensu stricto is calculated , and systems of mechanisms for which Φ for each complex or quale sensu lato is calculated . For the later a conceptual structure or constellation of concepts is defined as the set of all the concepts generated by a system of mechanisms in a state . It is necessary to specify how the conceptual information ( CI ) and the integrated conceptual information Φ of that system and its corresponding constellation of concepts are calculated . To this aim an extended version of the earth mover’s distance ( EMD ) is used . When the exclusion postulate is applied to systems of mechanisms , the complexes are defined as maximum irreducible conceptual structures ( MICS ) , that is , local maxima of integrated conceptual information . To do this we should use unidirectional partitions with which to verify that integrated conceptual structures are formed and see if apart from the major complex , and in a non-overlapping way , minor complexes are formed . We are now working in all of these items , and hope , with a much higher computational complexity , a refinement of the results , also in relation to the topology of mechanisms and the integrated information . Note that the aim to relate ( informational ) fields and matter by a mathematical treatment is not restricted to our approach or even IIT theory . It is remarkable in this line of research the description of central neuronal systems ( and more generically , biological systems ) as a gauge theory where a lagrangian approach to model motion is the crucial adopted formalism ( [75 , 76] ) . Our techniques are different , since we move into a more classical approach given by dynamical systems related to differential equations . In this way we take advantage of the huge theory on the qualitative analysis of differential equations . In particular , the curvature in the phase space produced by the informational field , closely related to the local stability of stationary points given by their associated eigenvalues , is translated into dynamical properties in the physical space , as a gauge theory approach . But the informational field , in our case , is , for each time , a global transformation of the phase space . This characterisation of an energy landscape ( the phase space ) evolves continuously in time . It remains an interesting and outstanding challenge to relate gauge theoretic treatments based upon information geometry ( for example , the free energy principle ) ( [76] , S3 appendix , or [77] ) to our dynamical formulation; however , their common focus on information geometry ( and manifold curvature ) may admit some convergence in the future . This framework may provide a promising and fruitful perspective . Thus , our approach provides many open questions and possibilities for future work . Three key problems arise at this stage , showing the necessity for a really interdisciplinary work in the area: the necessity to define the level of information of a global attractor ( a totally new question in dynamical systems’ theory ) , which notion of information better suits to our purposes , and how this concept of information is related to causality [78] on past and future events . We need to continue the development of the dynamical systems version of IIT 3 . 0 , finding the intrinsic representation of cause-effect repertoires and a measure to assess the quantity and quality of integrated information encoded in the informational structure . We also need to represent the evolution in continuous time of this information , and to find a way to measure the strength of attraction and repulsion through measure and dimension of stable and unstable sets and their intersections . Finally , use this notion to define the measure of integrated information compliant with IIT’s Φmax and describe the dynamics of this measure , by also creating computer codes for the informational structures evolution . Informational structures are really grounded in the topology of their associated mechanisms . In a very natural way , we can combine the ( continuous ) relation between structural networks and informational structures . A huge further research is probably deserved on this subject , not only in neuroscience , but in the huge area of dynamics on complex networks . This approach also opens the door to the development of a theory on the dependence of the structure of informational structures on parameters and on the topology on the complex graphs supporting them . To this aim , we need a theory on bifurcations of invariants and a theory on bifurcation of attractors , both in the autonomous and non-autonomous cases . In order to the comparison of this approach with real data , we need to develop global models for brain dynamics where the functionality of informational structures and informational fields can be tested . We will need to use the global models of brain dynamics [4 , 5 , 13 , 74 , 79] , allowing to the use of a brain simulator [80] which correlates with functional networks , and functional connectivity dynamics as developed from real data on human brains .
|
In this paper we introduce a space-time continuous version for the level of integrated information of a network on which a dynamics is defined . The concept of integrated information comes from the IIT of consciousness . By a strict mathematical formulation , we complement the existing IIT theoretical framework from a dynamical systems perspective . In other words , we develop the bases for a continuous mathematical approach to IIT introducing a dynamical system as the driving rule of a given mechanism . We also introduce and define the concepts of Informational Structure and Informational Field as the complex network with the power to ascertain the dynamics ( past and future scenarios ) of the studied phenomena . The detailed description of an informational structure is showing the cause-effect power of a mechanism in a state and thus , a characterization of the quantity and quality of information , and the way this is integrated . We firstly introduce how network patterns arise from dynamic phenomena on networks , leading to the concept of informational structure . Then , we formally introduce the mathematical objects supporting the theory , from graphs to informational structures , throughout the integration of dynamics on graphs with a global model of differential equations . After this , we formally present some of the IIT’s postulates associated to a given mechanism . Finally , we provide the quantitative and qualitative characterization of the integrated information , and how it depends on the geometry of the mechanism .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"neuroscience",
"cognitive",
"neuroscience",
"systems",
"science",
"mathematics",
"probability",
"distribution",
"algebra",
"computer",
"and",
"information",
"sciences",
"consciousness",
"dynamical",
"systems",
"differential",
"equations",
"probability",
"theory",
"eigenvectors",
"linear",
"algebra",
"information",
"theory",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"theories",
"of",
"consciousness",
"cognitive",
"science",
"eigenvalues"
] |
2018
|
Informational structures: A dynamical system approach for integrated information
|
To generate highly specific and adapted immune responses , B cells diversify their antibody repertoire through mechanisms involving the generation of programmed DNA damage . Somatic hypermutation ( SHM ) and class switch recombination ( CSR ) are initiated by the recruitment of activation-induced cytidine deaminase ( AID ) to immunoglobulin loci and by the subsequent generation of DNA lesions , which are differentially processed to mutations during SHM or to double-stranded DNA break intermediates during CSR . The latter activate the DNA damage response and mobilize multiple DNA repair factors , including Parp1 and Parp2 , to promote DNA repair and long-range recombination . We examined the contribution of Parp3 in CSR and SHM . We find that deficiency in Parp3 results in enhanced CSR , while SHM remains unaffected . Mechanistically , this is due to increased occupancy of AID at the donor ( Sμ ) switch region . We also find evidence of increased levels of DNA damage at switch region junctions and a bias towards alternative end joining in the absence of Parp3 . We propose that Parp3 plays a CSR-specific role by controlling AID levels at switch regions during CSR .
During immune responses , B cells diversify the antibody repertoire through mechanisms involving the generation of programmed DNA damage . Somatic hypermutation ( SHM ) introduces mutations in the immunoglobulin ( Ig ) variable ( V ) region genes , thereby modifying antibody affinity for its cognate antigen [1] . Class switch recombination ( CSR ) is a long-range recombination reaction occurring between switch ( S ) regions at the immunoglobulin heavy chain ( IgH ) locus and which replaces the exons encoding the heavy chain constant region , switching the antibody isotype ( from IgM to IgG , IgA or IgE ) , generating receptors with different effector functions [2] . SHM and CSR are initiated by activation induced cytidine deaminase ( AID ) , an enzyme , which deaminates cytosines into uracils in single stranded DNA ( ssDNA ) exposed by transcription [3] . These DNA lesions are processed by proteins of the base excision repair ( BER ) and/or mismatch repair ( MMR ) pathways to generate mutations in V regions during SHM and/or double stranded DNA breaks ( DSBs ) in S regions during CSR [1 , 2] . These breaks activate the cellular DNA damage response and mobilize multiple DNA repair factors , including the Poly ( ADP ) ribose polymerases Parp1 and Parp2 [4] and APLF [5] to promote appropriate DNA repair and long-range recombination . AID-mediated DSBs are ultimately resolved through classical and alternative non-homologous end joining ( NHEJ ) [6 , 7] . Poly ( ADP ) ribose polymerases ( Parp ) catalyze the formation of linear or multi-branched polymer of ADP-ribose ( PAR ) on acceptor proteins using β-NAD as substrate . This labile and transient post-translational modification is involved in the control of numerous basic cellular processes such as DNA repair , transcription and chromatin remodeling [8–10] . Inactivation of Parp1 or Parp2 in mice leads to increased sensitivity to DNA damaging agents and to genomic instability highlighting their essential role in DNA repair and in the maintenance of genome integrity . Indeed , Parp1 and Parp2 are activated by DNA damage and act as DNA damage sensors [8–10] . We have previously shown that PAR signaling plays an important role in the resolution of AID-induced damage [4] and that Parp1 promotes DNA repair through a microhomology-mediated pathway during CSR , while Parp2 behaves as a potent translocation suppressor [4] . In spite of Parp1 involvement in BER and MMR pathways , and the possibility to be activated by post-AID deamination DNA lesions , Parp1 appears dispensable for SHM [11] . Parp1 and Parp2 were believed to be the only members of the Parp family to mediate DNA repair . Recently however , Parp3 was found to associate with many different DNA repair factors and to respond to exogenous and endogenous DSBs [5 , 12 , 13] . Indeed , its inactivation leads to a delay in DSB repair in the context of chromatin [5 , 12] . Parp3 was first described to work in concert with APLF to promote the retention of the XRCC4/DNA ligase IV complex on chromatin and accelerate DNA ligation during NHEJ in human cells [5 , 14] . In addition , we have shown that APLF participates to the resolution of AID-induced DSBs by facilitating repair of switch regions by classical NHEJ during CSR [5] . More recently , Parp3 was also found to cooperate with the Ku70-Ku80 heterodimer to limit end-resection thereby favoring accurate NHEJ [15] . As a consequence , its inactivation results in defective repair of DSBs [5 , 12 , 15] . Here , we have examined the contribution of Parp3 in the response to AID-induced DNA damage generated during SHM and CSR .
To determine whether Parp3 plays a role in CSR , we tested the intrinsic ability of Parp3-/- B lymphocytes to undergo CSR . We purified mature resting B cells from the spleen of wild-type and Parp3-/- mice [12] , labeled them with CFSE to track proliferation and cultured them in vitro under conditions known to induce CSR to precise isotypes . After 72 h in culture , cells were analyzed by flow cytometry for proliferation ( CFSE dye dilution ) and immunoglobulin ( Ig ) surface expression ( Fig 1 ) . Surprisingly , we found that the efficiency of CSR , to all isotypes tested , was increased by 20 to 30% in Parp3-/- B lymphocytes , when compared to wild-type controls ( Fig 1A and 1B ) . Importantly , this phenotype was neither due to hyper-proliferation of Parp3-/- B lymphocytes ( Fig 1A and 1C ) , nor to an enhanced expression of Parp1 or Parp2 ( Fig 1D ) and was specific , since the efficiency of CSR in Parp3-/- B cells could be substantially reduced by re-expressing Flag-tagged Parp3 ( Parp3Flag; Fig 1E ) . Importantly , overexpression of Parp3Flag in wild-type cells significantly impaired CSR ( Fig 1E ) . We conclude that Parp3 is a negative regulator of CSR , and that the enhanced CSR observed in the absence of Parp3 is not linked to increased cell proliferation or to overexpression of Parp1 or Parp2 in Parp3-deficient B cells . To determine whether Parp3-deficiency also enhances SHM , we immunized Parp3-/- and littermate control mice with NP-CGG . Ten days post-immunization , we sorted germinal center B cells from the lymph nodes of individual mice and scored the mutation frequency in the JH4 intron . We did not find significant differences in the percentage of germinal center B cells ( Fig 2A ) , mutation distribution ( Fig 2B ) , mutation frequency ( Fig 2C ) , mutation pattern ( S1 Table ) or mutations at hotspots ( S2 Table ) . The only difference observed , was a reduction in the frequency of T->A mutations ( S1 Table ) , which did not , however , impact on the overall mutation frequency of mutations at A:T base pairs ( S1 Table ) . To determine whether affinity maturation is affected by Parp3-deficiency , we analyzed the appearance of high affinity antigen-specific IgM and IgG antibodies by ELISA using NP ( 4 ) and NP ( 23 ) as coating antigens ( Figs 2D and S1 ) . Consistent with normal SHM , we found no significant differences in the ratio of NP4/NP23 binding between Parp3-/- and wild-type mice for both IgM and IgG responses ( Fig 2D ) . Note however , that we observed a trend of higher IgG NP-specific response in Parp3-/- mice , although it was not statistically significant ( S1 Fig ) . We conclude that Parp3 is dispensable for SHM and antibody affinity maturation and that Parp3 plays a CSR-specific role . CSR and SHM are transcription-dependent processes [3] , with transcription providing AID access to its substrate ( ssDNA ) and significantly contributing to its chromatin recruitment to sites of RNA polymerase II stalling via the association of AID with the transcription elongation factor Spt5 [3 , 16] . Spt5 behaves as a stalling factor in vitro [17 , 18] , and several studies have reported the correlation of Spt5 occupancy and stalled RNA polymerase II in vivo [19 , 20] . To determine whether switch region transcription is enhanced in the absence of Parp3 , we assessed the level of switch region transcripts in Parp3-/- and control B cells stimulated for 72 h under different culture conditions by quantitative real time RT-PCR ( Fig 3A ) . We did not observe significant changes in the level of donor ( Iμ-Cμ ) or acceptor ( Iγ1-Cγ1 , Iε-Cε , Iγ3-Cγ3 Iγ2b-Cγ2b , Iγ2a-Cγ2a ) switch region germline transcripts in Parp3-/- B cells , when compared to littermate controls ( Fig 3A ) . To determine whether deficiency in Parp3 results in increased levels of RNA polymerase II stalling , we assessed RNA polymerase II and Spt5 occupancies at the Sμ region by chromatin immunoprecipitation experiments coupled to quantitative PCR ( ChIP-qPCR ) on chromatin prepared from Parp3-/- and control B cells stimulated for 60 h . We did not find significant differences in the level of RNA polymerase II recruitment downstream of the JH4 exon , at the Eμ enhancer , the Iμ exon , the Sμ switch region , and at the constant Cμ region upon Parp3 deficiency ( Fig 3B ) . Similarly , Spt5 occupancy at the donor switch region was unaffected in Parp3-deficient B lymphocytes when compared to wild-type B cells ( Fig 3C ) . This suggests that deficiency in Parp3 does not lead to increased levels of RNA polymerase II stalling at the IgH locus in B cells undergoing CSR . The efficiency of SHM , CSR and AID-initiated chromosomal translocations are closely related to AID expression level [21 , 22] . To determine whether increased AID expression could be responsible for the Parp3-dependent increase in CSR efficiency observed , we measured AID mRNA and protein levels from Parp3-/- and control activated B cells by qRT-PCR and western blot , respectively ( Fig 3D and 3E ) . We did not find significant differences in the level of AID transcripts in activated Parp3-/- B cells , when compared to wild-type B cells ( Fig 3D ) . Similarly , no changes in AID protein levels were observed ( Fig 3E ) . We conclude that increased CSR upon Parp3-deficiency is not due to enhanced levels of AID expression . The subcellular localization of AID is tightly controlled by both nuclear export and cytoplasmic retention mechanisms [23] and it has been shown that changing the ratio of nuclear vs . cytoplasmic AID has a significant impact on the frequency of CSR , SHM and/or IgH/c-myc translocations [24 , 25] . To determine whether the sub-cellular localization of AID is altered in the absence of Parp3 , we performed cell fractionation experiments in Parp3-/- and control B cells cultured in vitro to undergo CSR . Consistent with the fact that SHM is not enhanced in the absence of Parp3 , we found that the fraction of nuclear AID was not changed in Parp3-/- B cells when compared to controls ( Fig 3F ) . Together , these results show that deficiency in Parp3 does not lead to changes in switch region transcription , RNA polymerase II stalling or AID's expression level and nuclear localization . Given the recently described role for Parp3 and APLF in DSB repair [5 , 12] , we focused on the resolution phase of the CSR reaction . Switch region joining proceeds through the NHEJ pathway [6 , 7] , and switch region junctions usually display little or no microhomology , indicative of the predominant usage of classical end joining mediated repair [6 , 7] . Parp3 has recently been found to favor accurate NHEJ in response to genotoxic stress while limiting end-resection mediated alternative end joining [15] . To determine how AID-induced breaks are repaired in absence of Parp3 , we amplified and sequenced Sμ-Sγ3 and Sμ-Sγ1 junctions from Parp3-deficient and control B cells ( Figs 4 and S2 ) . Similar to APLF-deficient B cells [5] , we did not find a significant reduction in the usage of blunt joining , which is characteristic of B cells deficient for the core components ( XRCC4 or DNA ligase IV ) of the NHEJ pathway [7] . Nevertheless , we found that Sμ -Sγ1 junctions derived from Parp3-/- B cells displayed increased usage of microhomology when compared to wild-type B cells ( Fig 4A , upper panel ) . While the average length of overlap ( excluding insertions ) was of 1 . 27 bp for the controls , it was increased to 2 . 35 bp for Parp3-/- B cells , a finding comparable to what is observed in APLF-/- B cells [5] . Consistent with this , we found that Sμ -Sγ3 junctions derived from Parp3-/- B cells also displayed increased usage of microhomology ( wild-type: 1 . 82 bp; Parp3-/-: 3 . 20 bp; Fig 4A , lower panel ) . The increase in microhomology was due to a significantly higher frequency of sequences bearing ≥15 bp of microhomology at the junction . This pattern is reminiscent of the junctions observed in B cells deficient for DNA ligase IV , XRCC4 , Artemis or APLF [5–7] . This suggests that Parp3 facilitates the resolution of AID-induced breaks through the classical NHEJ pathway and is consistent with the cooperation between APLF and Parp3 in mediating DNA repair through the classical NHEJ pathway [5] . Furthermore , we found a two-fold increase in the proportion of switch junctions with additional intra-switch region recombination events in Parp3-deficient B cells when compared to controls ( p = 0 . 0325; Figs 4B and 4C and S2 ) , giving rise to compound sequences of the type Sμ-Sμ-Sx , Sμ-Sx-Sx or Sμ-Sμ Sx-Sx and bearing very long microhomology at the junction ( up to 30 bp ) , inversions and/or micro-deletions ( Figs 4B–4D and S2 ) , a finding similar to what is observed in the few 53BP1-/- B cells that succeed to undergo CSR [26] . Furthermore , we found that among these compound sequences , the degree of complexity ( i . e . the number of intra-recombination events , insertions , inversions and deletions per sequence ) was five-fold higher in Parp3-/- B cells , when compared to controls ( p = 0 . 002; Figs 4C and S2 ) . Together , these results indicate that DSB resolution is altered upon Parp3-disruption , that Parp3 facilitates DSB repair through the classical NHEJ and suggests , that in the absence of Parp3 , switch regions sustain enhanced levels of AID-induced DNA damage during CSR . As Parp3-deficient switch junctions show evidence of increased AID-induced DNA damage , we hypothesized that this might be indicative of increased levels of AID binding at the IgH locus and that this could explain the enhanced CSR efficiency observed in Parp3-deficient B cells . To determine whether AID binding to the switch regions is enhanced in the absence of Parp3 , we performed ChIP-qPCR experiments using two different anti-AID antibodies on chromatin prepared from Parp3-deficient , Parp3-proficient and AID-deficient ( AIDCre/Cre ) B cells stimulated for 60h ( Fig 5A ) . As expected , we found a specific enrichment of AID at the donor Sμ region in wild-type B cells when compared to AIDCre/Cre B cells ( Fig 5A ) . Interestingly , we found that AID occupancy at the Sμ switch region was more than 2-fold enhanced in Parp3-/- B cells , when compared to controls ( Fig 5A ) . To determine whether this is due to increased AID loading kinetics at the donor switch region in Parp3-/- B cells , we analyzed AID occupancy at an earlier time point . We found that after 48h of stimulation , AID occupancy at the Sμ region was comparable between Parp3-/- and wild-type B cells ( Fig 5B ) . We conclude that the increased AID occupancy observed in Parp3-deficient B cells is not due to faster loading kinetics of AID to Sμ . This suggests that Parp3 negatively regulates AID occupancy . If this were to be the case , then Parp3-deficient B cells would be less sensitive to an AID knockdown when compared to wild-type B cells . To test this , we performed shRNA-mediated AID knockdown experiments in Parp3-/- and wild-type B cells and analyzed the capacity of transduced cells to undergo CSR ( Figs 5C and S3 ) . We found that reducing AID levels in Parp3-/- B cells only leads to a modest decrease in the frequency of IgG1+ cells whereas the decrease observed in wild-type cells is much more pronounced ( Fig 5C ) . Moreover , while AID overexpression robustly enhanced CSR in wild-type B cells ( Fig 5C ) , there was only a modest increase in Parp3-/- B cells ( Fig 5C ) . We conclude that AID binding to Sμ is enhanced in the absence of Parp3 and that Parp3-deficiency is able to counteract reduced levels of AID expression . We have previously defined Parp2 as a potent translocation suppressor [4] . To determine whether Parp3 also possesses a translocation suppressor function and whether the increased binding of AID at the IgH locus impacts on the off-targeting activity of AID , we assessed the occurrence of AID-dependent translocations occurring between the Sμ switch region in the IgH locus ( chromosome 12 ) and the 5’ region of c-myc ( chromosome 15 ) in Parp3-/- and control B cells by long range PCR and Southern blot [4] . Surprisingly , we found that the frequency of IgH/c-myc translocations was similar between Parp3-/- and wild-type B cells ( Fig 5D ) . We conclude that contrary to Parp2 [4] , Parp3 is not a translocation suppressor .
We have shown that deficiency in Parp3 results in enhanced CSR , while SHM and affinity maturation remain unaffected . Therefore , Parp3 plays a CSR-specific role and acts as a negative regulator . Mechanistically , we have shown that Parp3-deficiency results in increased binding ( recruitment , retention , or increased residence time ) of AID at the Sμ switch region in the IgH locus , which is not due to faster loading kinetics , that renders Parp3-deficient B cells less sensitive to reduced levels of AID expression and that translates into increased DNA damage at switch regions . Therefore , it appears that Parp3 negatively regulates CSR by controlling the level of AID on the chromatin . This by itself is sufficient to explain the increased efficiency of CSR observed in Parp3-deficient B cells . It is also possible that increased CSR efficiency could be due to rerouting to the microhomology-mediated pathway , although this would imply that the alternative NHEJ pathway is more robust than the classical NHEJ . In the case of APLF-deficiency , no defect in CSR is observed in spite of increased usage of microhomology , suggesting that microhomology-mediated joining is able to compensate the altered classical NHEJ and to maintain CSR at wild-type levels [5] . However , no overcompensation of the classical NHEJ by the alternative NHEJ pathway was observed . This is consistent with the potential kinetic disadvantage of alternative NHEJ relative to classical NHEJ [27–30] . This could explain why over-expression of Parp3 ( even if it promotes the joining pathway of choice for CSR ) reduces the CSR efficiency in Parp3-deficient or wild-type B cells . The Parp3 phenotype is reminiscent of previous findings showing that pharmacological inhibition of Parp activity in the I . 29μ [31] or CH12 B cell lines [4] results in enhanced CSR efficiency . In these reports however , the underlying mechanisms could not be unequivocally attributed to any of the members of the Parp family [4 , 31] and it was believed to be restricted to transformed cell lines , as the same pharmacological inhibitors did not significantly increase the efficiency of CSR in primary B cell cultures from wild-type mice [4] . Here we reconcile these findings by showing that enhanced CSR is indeed specific to a deficiency in Parp3 . The efficiency of CSR and SHM is directly correlated to the extent of AID mRNA or protein levels and haplo-insufficiency in AID results in reduced levels of SHM , CSR and IgH/c-myc translocations [21 , 32–34] . Similarly , controlling the abundance of nuclear AID leads to increased efficiency of CSR [24] . Therefore , changes in the levels of AID mRNA or protein , or in its sub-cellular localization could potentially explain the enhanced CSR observed in the absence of Parp3 . Nevertheless , we have excluded these possibilities , as we have shown that neither the level of AID mRNA or protein , nor the abundance of its nuclear fraction , are increased in the absence of Parp3 . SHM and CSR are transcription-dependent mechanisms , and their efficiency is tightly correlated to transcription levels [3] . Furthermore , genome-wide studies have shown that AID recruitment to chromatin correlates with sites of RNA polymerase II stalling and is achieved through the interaction between AID and the transcription elongation factor Spt5 [16] . We have shown that the level of switch region transcripts is not different from wild-type controls in Parp3-deficient B cells , arguing against an increase in transcription . We have also shown that RNA polymerase II occupancy and Spt5 recruitment at Sμ are not increased by Parp3-deficiency , indicating that RNA polymerase II stalling is not enhanced and at the same time showing that the effect of Parp3 in the increased occupancy of AID at the IgH locus is independent of Spt5 . Although we have not been able to show that enhanced binding of AID is also observed at the acceptor switch regions , the fact that we find complex switch region junctions with additional intra-switch recombination events implicating the acceptor switch regions , suggests that this is indeed the case . Therefore , it is possible that the Parp3-mediated control of AID occupancy on the chromatin is not restricted to Sμ and that it also applies to acceptor switch regions . We also find that despite increased occupancy of AID at Sμ region , a higher incidence of illegitimate chromosomal translocations between the IgH locus and c-myc is not observed . This could be explained by the fact that Parp3-/- B cells express normal levels of Parp2 , which we have defined as a potent translocation suppressor [4] . Even if we have not directly assessed this point , we expect Parp3-deficient B cells to be proficient for ATM and miR155 , factors known to protect against chromosomal translocations and which might contribute to preserve genomic stability upon Parp3 inactivation [34 , 35] . Alternatively , it is possible that the role of Parp3 in controlling AID level on chromatin is restricted to the switch regions at the IgH locus and that it therefore has no influence in the regulation of AID occupancy over the variable regions or at non-Ig off-targets . This would be consistent with the fact that SHM is not affected by Parp3-deficiency and with the described role for the classical NHEJ pathway in suppressing oncogenic translocations [6] . Finally , it is also possible that the activation of the p53-dependent checkpoint via ATM eliminates cells with unresolved DNA breaks and that activation of the arf/p19-mediated p53 checkpoint circumvents the accumulation of cells bearing oncogenic translocations [35] . Several non-exclusive hypotheses could be envisaged to explain the enhanced occupancy of AID at Sμ observed in Parp3-deficient B cells after 60h stimulation . First , Parp3 could directly modify AID and thereby control its eviction from the IgH locus , in a mechanism comparable to TRF1 eviction from telomeres through its PARylation by Tankyrase 1 [36] or the Parp1-mediated release of histone H1 during chromatin de-compaction [37] . Nevertheless , this is unlikely , as we failed to detect such a modification on AID in in vitro assays . Parp3 could also reduce AID residence time at Sμ . However , up-to-date , the mechanisms controlling AID residence on the IgH locus and its release from chromatin remain totally unknown and unexplored . Another possibility would be that Parp3 impacts on the activity of protein kinase A ( PKA ) . Indeed , mutation of the R1 alpha subunit of PKA leads to constitutive activation of PKA and increases the efficiency of CSR [38] , probably by phosphorylating AID at S38 , a post-translational modification that is known to promote CSR by providing a binding platform for RPA [38–41] . It is not known , however , if the constitutive activation of PKA results in increased binding of AID at switch regions . Parp3 could also have an impact on AID recruitment via GANP [42] , nevertheless this is unlikely as GANP's overexpression or deletion clearly affects the efficiency of SHM [42] , which is not the case in Parp3-deficient mice . Additionally , Parp3 could have an influence on AID activity by modulating the phosphorylation of AID at serine 3 , which negatively controls AID function , and whose mutation to alanine leads to higher mutation frequency and increased translocation occurrence in AID-deficient B cells reconstituted with AID mutant [43] . Parp3 could also impact on the activity of DNA polymerase β [44] and/or λ [45] , since their inactivation causes a higher number of single stranded DNA breaks , which can be further processed into DSBs ( mandatory intermediates for CSR ) . Nevertheless , DNA pol β and λ act downstream of AID deamination and processing of dU by uracil DNA glycosylase ( UNG ) , and contribute to some extent to somatic hypermutation . Therefore , it is unlikely that Parp3 impacts on their activity , as our analysis of SHM in Parp3-deficient mice did not reveal defects in the frequency , distribution , and mutation pattern or affinity maturation . Importantly , no mutation bias was found at G:C and A:T pairs suggesting that both phase I , directly dependent on AID and UNG activities , and phase II , responsible for lesion processing by MMR and BER factors [1] , are normal upon Parp3 inactivation . An alternative hypothesis is that Parp3 could modulate the processing of intermediates resulting from AID-mediated DNA deamination , promoting error-free repair at the expense of the error-prone processing required for CSR and SHM . Parp3 activity was recently shown to be also stimulated by 5’ phospho nicks in DNA [46] and it is possible that nicked abasic sites ( intermediates in the processing of AID induced lesions ) could potentially activate Parp3 . However , no role for Parp3 in BER or MMR pathways has been reported so far , and as discussed above , mutation pattern did not reveal any bias in the phase II of SHM , indicative of unaltered processing of such intermediates . Finally , we cannot exclude that Parp3 impacts on other factors mediating AID targeting , like KAP1 , 14-3-3 , PTBP2 or subunits of the PAF complex [47–49] . However , as AID occupancy at Sμ after 48h stimulation is comparable between wild-type and Parp3-/- B cells , this suggests that the initial recruitment of AID to Sμ is unaffected , and that most probably AID targeting is achieved normally in Parp3-/- B lymphocytes . Nevertheless , we cannot rule out the possibility that alterations in AID recruitment at Sμ occur later during the reaction . Concerning the DSB resolution phase of the CSR recombination reaction , we found that Parp3-deficiency results in altered NHEJ , with switch junctions obtained from Parp3-/- B cells displaying microhomology-mediated end-joining and increased frequency of complex switch junctions bearing insertions , inversions and micro-deletions . This is consistent with our previous findings showing that Parp3 responds to DSBs [12] and facilitates accurate NHEJ in concert with APLF and Ku70-Ku80 [5 , 15] . The usage of longer microhomology in Parp3-/- B cells resembles the alterations in DNA repair observed upon DNA ligase IV or XRCC4 inactivation ( although these result in defective CSR , contrary to Parp3 ) . Furthermore , switch junctions obtained from APLF-deficient B cells also showed microhomology-mediated repair , without strongly reducing the usage of blunt joining , a hallmark feature of cells deficient in NHEJ core components [6 , 7] . It appears then , that Parp3 and APLF are accessory factors rather than core NHEJ components [5] . This is consistent with the fact that neither Parp3 nor APLF are required for V ( D ) J recombination . In addition , a potential role for Parp3 in putative end-joining pathways that could operate during CSR [50] can not be ruled out , as Parp3 is known to interact with components of the alternative NHEJ pathway ( DNA ligase III and Parp1 [13] ) , classical NHEJ pathway ( DNA-PKcs , Ku80 , Ku70 , DNA ligase IV [13] ) and was also recently shown to promote HR [15] , showing its versatility in promoting DNA DSB repair . Finally , we cannot exclude the possibility that increased usage of microhomology in the absence of Parp3 results from a dual role for Parp3 in directly mediating DSB repair and in regulating the in situ catalytic activity ( or processivity ) of AID , as it has been recently shown that lowering the density of AID-mediated DNA deamination at switch regions increases DSB resolution by microhomology-mediated repair [51] . Overall , our results implicate Parp3 in the repair of programmed double-stranded DNA breaks induced by AID , provide further evidence for its involvement in the classical NHEJ pathway and reveal a novel negative regulation mechanism of CSR governed by the Parp3-mediated control of AID occupancy at the IgH locus .
Reagents for primary B cell stimulation include CD43-microbeads ( Miltenyi Biotec ) , LPS ( 50 μg/ml; Sigma-Aldrich ) , IFN-γ ( 100 ng/ml; Peprotech ) and IL-4 ( 5 ng/ml; Peprotech ) . NP-CGG ( 75–100 μg/mouse , Biosearch Technologies Inc . ) was used for immunization experiments . Parp3-/- [12] mice were on a B6;129 mixed background . AIDCre/Cre [52] mice were on C57BL/6 background . All mice were bred and maintained under specific pathogen-free conditions . Age-matched littermates ( 8–12 week-old ) were used in all experiments . All animal work was performed under protocols approved by the Direction des Services Vétérinaires du Bas-Rhin , France ( Authorization N° 67–343 ) . Resting splenic B cells were isolated using CD43-microbeads , stained with 5 μM CFSE and cultured for 72 h in vitro with LPS ( 50 μg/ml ) for CSR to IgG3 and IgG2b , LPS + IL-4 ( 5 ng/ml ) for CSR to IgG1 and IgE and LPS + IFN-γ ( 100 ng/ml ) for CSR to IgG2a . CSR was assayed by flow cytometry as described [4] . Primary B cells were transduced with retroviruses expressing Parp3Flag , Flag alone or AIDFlag-HA and a GFP reporter as previously described [53] . For knockdown experiments , the hairpin sequence for AID ( 5’-ACCAGTCGCCATTATAATGCAA-3’ ) was cloned into the LMP retroviral vector ( Open Biosystems ) , transductions were performed as previously described [49] . Littermate Parp3-/- and control mice were immunized by footpad injection of NP-CGG ( 75 μg/mouse ) in Freund's adjuvant . After 10 days , germinal center B cells ( B220+ Fas+ GL-7+ ) were sorted from the lymph nodes from each mouse individually . Fragments corresponding to the region downstream of rearranged JH4 exon were amplified by PCR ( see S3 Table for oligonucleotides ) , cloned and sequenced from each animal individually [54] . Sequences were analyzed for mutation with SHMTool [55] . Littermate Parp3-/- and control mice were immunized i . p . with 100 μg of NP-CGG in alum ( Pierce ) . Serum was obtained after blood coagulation and kept at -20°C . Ninety-six–well plates ( Nunc ) were coated with 5 μg/ml NP ( 23 ) -BSA to detect both low- and high-affinity antibodies or NP ( 4 ) -BSA ( Biosearch Technologies Inc . ) to detect high-affinity antibodies . Dilutions of sera were incubated overnight at 4°C . Goat anti−mouse IgM and goat anti−mouse IgG conjugated to horseradish peroxidase ( Jackson ImmunoResearch ) were incubated for 1 h at 37°C . Horseradish peroxydase activity was revealed with SigmaFast OPD substrate kit ( Sigma-Aldrich ) . Results are expressed as absorbance at 490 nm for serum diluted 1:1000 for IgM and 1:10000 for IgG ( all absorbance readings were in the linear range ) . RNA and cDNA were prepared using standard techniques . qPCR was performed using QuantiTect SYBR green PCR kit ( Qiagen ) or with Roche LightCycler 480 Probes Master mix UPL in combination with appropriate UPL probes ( see S3 Table for oligonucleotides and UPL probes ) . Approximately 3 ng of cDNA were run ( in triplicate ) and analyzed on a LightCycler 480 ( Roche ) . Transcript quantities ( mean of triplicates ± SD ) were calculated relative to standard curves and normalized to CD79b or HPRT transcripts . Gene of interest/ normalizing gene values ± SD were then normalized to the appropriate controls , all standard deviations after normalization were calculated following the rules for error propagation while calculating a ratio . Germline switch region transcripts were analyzed as described previously [4] . The protocol was adapted from Upstate-Millipore ( http://www . millipore . com/userguides/tech1/mcproto407 ) . Briefly , 2x107 stimulated B cells were cross-linked at 37°C for 10 min in 5 ml PBS/0 . 5% BSA with 1% formaldehyde . The reaction was quenched with 0 . 125 M glycine . Following lysis , chromatin was sonicated to 0 . 5–1 kb using a Covaris system ( Covaris ) . After 5x dilution in ChIP dilution buffer ( final concentrations are 0 . 21% SDS , 0 . 88% Triton X-100 , 3 mM EDTA , 23 . 4 mM Tris-HCl [pH 8 . 1] , 133 . 6 mM NaCl ) , chromatin was pre-cleared by rotating for 2 h at 4°C with 80 μl protein A/G magnetic beads ( Dynabeads , Life technologies ) . 0 . 5 to 0 . 9x106 cell equivalents were saved as input and 5 to 9x106 cell equivalents were incubated overnight with protein A/G magnetic beads that were preloaded with specific or control antibodies ( see S4 Table for antibodies used ) . Washes were performed according to the Millipore protocol . Cross-links were reversed for 4 h at 65°C in Tris-EDTA buffer with 0 . 3% ( wt/vol ) SDS and 1 mg/ml proteinase K . qPCR was performed at several locations across the IgH locus using primer pairs listed in S3 Table . Specific antibody-ChIP values ( mean of triplicate samples ± SD ) were normalized to the input control and are expressed as percent input or as fold-change relative to the control conditions . All standard deviations after normalization were calculated following the rules for error propagation while calculating a ratio . Sμ-Sγ3 and Sμ-Sγ1 switch junctions were amplified using previously described primers [56–58] ( see S3 Table for oligonucleotides ) and conditions [4] from genomic DNA prepared from 72 h-stimulated B cells . PCR products were cloned using TOPO-TA cloning kit ( Invitrogen ) and sequenced using T7 and T3 universal primers . Sequence analysis was performed using the CSRtool software ( manuscript in preparation ) . IgH/c-myc translocations were analyzed by long-range PCR and Southern blot as described [4] .
|
During infections , B cells diversify the antibodies they produce by two mechanisms: somatic hypermutation ( SHM ) and class switch recombination ( CSR ) . SHM mutates the regions encoding the antigen-binding site , generating high-affinity antibodies . CSR allows B cells to switch the class of antibody they produce ( from IgM to IgA , IgG or IgE ) , providing novel effector functions . Together , SHM and CSR establish highly specific and pathogen-adapted antibody responses . SHM and CSR are initiated by the recruitment of the activation-induced cytidine deaminase ( AID ) enzyme to antibody genes . Once recruited , AID induces DNA lesions that are processed into mutations during SHM or chromosomal DNA breaks during CSR . These breaks activate multiple DNA repair proteins and are resolved by replacing the IgM gene segments by those encoding IgA , IgG or IgE . AID carries a significant oncogenic potential that needs to be controlled to preserve genome integrity . Nevertheless , the underlying mechanisms remain poorly understood . Here we show that Poly ( ADP ) ribose polymerase 3 ( Parp3 ) , an enzyme recently implicated in DNA repair , contributes to antibody diversification by negatively regulating CSR without affecting SHM . We show that Parp3 facilitates the repair of AID-induced DNA damage and controls AID levels on chromatin . We propose that Parp3 protects antibody genes from sustained AID-dependent DNA damage .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Parp3 Negatively Regulates Immunoglobulin Class Switch Recombination
|
Human cysticercosis is a zoonotic disease causing severe health disorders and even death . While prevalence data become available worldwide , incidence rate and cumulative incidence figures are lacking , which limits the understanding of the Taenia solium epidemiology . A seroepidemiological cohort study was conducted in a south-Ecuadorian community to estimate the incidence rate of infection with and the incidence rate of exposure to T . solium based on antigen and antibody detections , respectively . The incidence rate of infection was 333 . 6 per 100 , 000 person-years ( 95% CI: [8 . 4–1 , 858] per 100 , 000 person-years ) contrasting with a higher incidence rate of exposure 13 , 370 per 100 , 000 person-years ( 95% CI: [8 , 730–19 , 591] per 100 , 000 person-years ) . The proportion of infected individuals remained low and stable during the whole study year while more than 25% of the population showed at least one antibody seroconversion/seroreversion during the same time period . Understanding the transmission of T . solium is essential to develop ad hoc cost-effective prevention and control programs . The estimates generated here may now be incorporated in epidemiological models to simulate the temporal transmission of the parasite and the effects of control interventions on its life cycle . These estimates are also of high importance to assess the disease burden since incidence data are needed to make regional and global projections of morbidity and mortality related to cysticercosis .
Human cysticercosis ( CC ) is a parasitic disease caused by the development of the metacestode larval stage of Taenia solium ( cysticercus ) in the muscles , the central nervous system ( causing neurocysticercosis ( NCC ) ) , the subcutaneous tissue and the eyes ( causing subcutaneous and ocular cysticercosis , respectively ) [1] . The life cycle of the parasite includes humans as sole definitive hosts and pigs as main intermediate hosts . Humans get infected by consumption of raw or undercooked pork infected with cysticerci , resulting in the development of an adult intestinal tapeworm ( taeniosis ) . Pigs become infected by ingestion of T . solium eggs contained in infected human feces , through coprophagic behavior or via ingestion of contaminated water or food , and develop porcine CC . Man can also act as a dead-end intermediate host by accidental ingestion of T . solium eggs [2] and develop human CC . NCC may cause severe neurological disorders and even death [3] , [4] . It is the most important parasitic disease of the central nervous system and the main cause of acquired epilepsy in T . solium endemic areas , where NCC is associated with 14 . 2 to 50% of the epilepsy cases [5] , [6] . The maintenance of the parasite life cycle is associated with poor sanitation , lack of hygiene and traditional pig rearing systems allowing free roaming of the animals . Endemic areas have been identified in Asia , Africa and Latin America [7]–[10] . In Latin America the infection has been reported in at least 18 countries and is considered a major public health problem , especially in poor rural areas [7] , [8] . The Andean region of Ecuador and neighboring countries is hyper-endemic for cysticercosis [11] . While reliable prevalence data become available worldwide , they may considerably vary depending on the diagnostic test used [12]–[14] . Several tools are available for the diagnosis of human CC , i . e . imaging and serological techniques . Serological antigen and antibody detections are valuable tools when conducting epidemiological studies , since they inform on infection with and exposure to the parasite , respectively . Taking the latter distinction into account , studies conducted in Ecuadorian endemic rural communities have shown an exposure to the parasite ranging from 25 to 40% and a proportion of infected individuals ranging from 2 . 25 to 4 . 99% [15]–[17] . However , prevalence figures do not inform on the evolution of the number of positive cases over time and estimates for human cysticercosis incidence rate and cumulative incidence are lacking , which limits the understanding of the transmission dynamics of T . solium and does not allow a precise estimation of its disease burden . García et al . ( 2001 ) [18] conducted longitudinal studies in endemic areas of Peru and Colombia and demonstrated the presence of transient antibody responses suggesting a high number of antibody seroconverted cases per year ranging from 8 to 25% of the population depending on the studied area . Through rule-based modeling , Praet et al . ( 2010 ) [16] simulated the annual antibody seroconversion rate in an endemic area of Ecuador . They estimated an annual incidence rate of exposure of people becoming seropositive of 14 per 100 person-years . On the other hand , studies estimating both incidence rate of infection and cumulative incidence are scarce [16] , [18] , [19] . Mwape et al . ( 2013 ) [20] reported an incidence rate of infection of 6 , 300 per 100 , 000 person-years in a rural community of eastern Zambia . Such estimates for Latin America are inexistent . For this reason , the present study aims at estimating the cumulative incidence and the incidence rate of human CC in an endemic area of Ecuador . A sero-epidemiological cohort study was conducted to investigate the transmission dynamics of T . solium among individuals living in a southern Ecuadorian rural community . This paper reports estimates of the incidence rates , cumulative incidences of active infection and exposure rates to T . solium and discusses the implications for the disease burden assessment and control .
The protocol used in this study was approved by the Ethical Committee of the Central University of Ecuador ( IRB 00002438 ) and by the Ethical Committee of The University Hospital of Antwerp , Belgium . Written informed consent was obtained from each individual willing to participate in the study . For participants aged less than 18 years old written informed consent was also obtained from a parent or a legal adult representative . Individuals testing positive for T . solium cysticercosis antigens were referred to the local health center for follow-up . The study was conducted in the rural parish of Sabanilla ( 4° 12′S , 80° 8′W ) belonging to the Celica canton in the Southern Ecuadorian province of Loja . The parish has 1145 inhabitants; most of them are farmers involved in activities related to agriculture and animal husbandry . The climate is semi-arid , and the altitude is 700 meters above sea level . The region is endemic for T . solium cysticercosis and presents the risk factors for the transmission of the parasite [15] , [21] . A sero-epidemiological community-based cohort study was performed . Three blood sampling rounds were organized in Sabanilla in a period of 13 months: the first sampling round took place in June 2009 ( SR1 ) , the second in November 2009 ( SR2 ) and the third one in July 2010 ( SR3 ) . Based on the three sampling rounds , three periods of time were defined as follows: a six-month period from June 2009 to November 2009 ( P1 ) , a seven-month period from November 2009 to July 2010 ( P2 ) , and a total 13-month period from June 2009 to July 2010 ( P3 ) . First , an informative meeting inviting the population to participate took place at the beginning of the study in collaboration with the local authorities . Then , a census of the population was conducted based on a door-to-door survey , including collection of information on age and sex of the inhabitants . After informed consent , all individuals older than one year willing to participate and present at the time were blood sampled . At each sampling round , 10 ml of blood was collected in dry tubes . After coagulation and centrifugation , serum was collected and stored at −20°C until analysis . Two serological diagnostic tests were performed . ( 1 ) The Enzyme Linked Immunosorbent Assay for the detection of circulating antigens of the metacestode of T . solium ( Ag-ELISA ) [22]–[24] . The sensitivity and specificity of the Ag-ELISA for detecting active infection in humans are 90% ( 95% CI: [80–99%] ) and 98% ( 95% CI: [97–99%] ) , respectively . No cross-reaction with other parasites has been reported [13] , [23] . ( 2 ) The Enzyme-Linked Immunoelectrotransfer Blot ( EITB ) for the detection of antibodies directed against seven specific T . solium metacestode glycoproteins [25] . The sensitivity and specificity of the EITB for detecting exposure to the parasite range from 97% to 98% and from 97% to 100% , respectively [13] , [25] . The antigen and antibody seroprevalence ( Ag and Ab seroprevalence ) , as based on the results of the Ag-ELISA and of the EITB , respectively , were calculated for each sampling round for the whole population and by sex . A multinomial Bayesian model adapted from Berkvens et al . ( 2006 ) [26] was used to estimate the true prevalence of T . solium larval infections for each sampling round based on the antigen seroprevalence data and on prior information on the test characteristics ( sensitivity and specificity of the Ag-ELISA ) . Prior information was extracted from the available literature [13] . A uniform distribution with lower and upper limits of 0 . 80 and 1 . 00 , and 0 . 97 and 1 . 00 were used to constrain the sensitivity and the specificity of the test , respectively . The analysis was conducted in WinBUGS and R [27] , [28] . Three chains , 20 , 000 iterations , following a burn-in of 5 , 000 were used to assess the convergence of the results . Criteria assessing the fit between prior information and the seroprevalence data were evaluated , i . e . the Bayesian p-value ( Bayesp ) , the Deviance Information Criterion ( DIC ) and the number of parameter effectively estimated by the model ( pD ) [26] , [28] . First , proportion of change to antigen seropositivity/seronegativity ( change to Ag seropositivity/seronegativity ) and proportion of antibody seroconversion and seroreversion ( Ab seroconversion and seroreversion ) were calculated to characterize the transmission dynamics of the disease . Seroconversion is defined as the change from a negative to a positive serological test result between 2 sampling rounds; the opposite is defined as seroreversion [29] . The proportion of Ab seroconversion and the proportion of change to Ag seropositivity reflect the cumulative incidence for a defined time period . They were calculated by dividing the number of new cases by the number of susceptible individuals ( having a negative test result at the previous sampling round ) during a given time . The proportion of Ab seroreversion and the proportion of change to Ag seronegativity were calculated by dividing the number of positive tests that turned negative by the number of positive tests at the previous sampling round . The incidence rate of infection with the larval stage of T . solium and the incidence rate of exposure to T . solium eggs were also calculated based on the results of the antigen and antibody detection tests , respectively . The incidence rate was calculated as the number of new ( change from seronegativity to seropositivity ) cases in a defined time period divided by the number of person-time units at risk during the time-period . Yearly incidence rates were multiplied by 100 , 000 to be expressed by 100 , 000 person-years [30] , [31] . The person-time unit represents one person for a defined period of time . The latter was calculated as described in Ngowi et al . ( 2008 ) assuming that the infection occurs uniformly over time and considering halfway the period between two sampling rounds [32] . For example , if a person is followed up for six months and does not seroconvert during this time , this person will contribute 0 . 5 person-year to the person-time at risk . If a person that is followed up for the same period but seroconverts during that period , this person will contribute 0 . 25 person-year to the person-time at risk . Yearly incidence rates were calculated based on this calculation method . Ninety-five % exact Poisson confidence intervals were calculated using the epitools package in R for all incidence rates [33] . Data were entered in Excel 2010 ( Microsoft Office 2010 ) . Statistical analyses were performed in Stata ( Stata Corp . , College Station , TX ) and in R: [34] Fisher exact test was used to compare ( 1 ) Ag/Ab seroprevalence between sex within each sampling round and ( 2 ) Ag seroprevalence with Ab seroprevalence within each sampling rounds . Also , McNemar test was performed to compare the sero-Ag and sero-Ab prevalence between rounds . Multivariate logistic regression analysis was used to study the association between sero-Ag/Ab prevalence and age and sex , and this for the three samplings rounds . The significance level was set at 0 . 05 . Fisher exact test was used to compare ( 1 ) the proportion of Ab seroconversion with the proportion of antibody seroreversion and the proportion of change to Ag seropositivity with the proportion of change to Ag seronegativity within periods , ( 2 ) the proportions of Ab seroconversion/seroreversion and the proportions of Ag change to seropositivity/seronegativity between sexes , ( 3 ) the proportions of Ab seroconversion/seroreversion and the proportions of change to Ag seropositivity/seronegativity between periods . In addition , a change point analysis was used to compare the proportion of Ab seroconversion with the proportion of Ab seroreversion in function of age . The change point analysis classifies the population into 2 age groups at different age points ( 10 , 20 , 30 , 40 , 50 , 60 , 70 , 80 years old ) . The Fisher exact test was then used on both age groups in order to identify any change of significance when comparing the proportions of Ab seroconversion and seroreversion [16] , [20] , [35] . The significance level was set at 0 . 05 for all statistical analyses .
The Ag and Ab seroprevalence for each sampling round for the whole population and by sex are presented in Table 1 . The prevalence adjusted for misclassification error of T . solium larval infections was 0 . 7% ( 95% Credibility Interval ( CI ) : [0 . 03–1 . 75] ) , 0 . 7% ( 95% CI: [0 . 03–2 . 00] ) and 1 . 1% ( 95% CI: [0 . 05–2 . 84] ) for SR1 , SR2 and SR3 , respectively . All except one Ag positive individuals were also Ab positive in the 3 sampling rounds . Fisher exact test did not reveal any significant difference of Ag and Ab seroprevalence between sexes . McNemar test did not reveal any significant difference of Ag and Ab seroprevalence between rounds . Ab seroprevalence was significantly higher than Ag seroprevalence within each sampling round . Multivariate logistic regression analysis showed a significant positive correlation between Ab seroprevalence and age . Table 2 shows the proportions of antigen and antibody seropositive and/or seronegative individuals who participated in all 3 sampling rounds and whose sera were available for both tests ( n = 277 ) . Only one individual changed to antigen seropositivity status throughout the entire study period . Eighteen percent of this restricted population remained antibody positive throughout the entire study period while about 20% of the individuals showed at least 1 change of antibody positivity status . The overall incidence rate of human T . solium larval infection based on -antigen detection was 333 . 6 per 100 , 000 person-years ( 95% exact Poisson CI: [8 . 4–1 , 858] per 100 , 000 person-years ) . The overall incidence rate of exposure to T . solium based on antibody detection was 13 , 370 per 100 , 000 person-years ( 95% exact Poisson CI: [8 , 730–19 , 591] per 100 , 000 person-years ) . Ag proportion of changes to seropositivity/seronegativity and proportion of Ab seroconversion/seroreversion by period are represented in Table 3 . Incidence rates estimates for individuals who participated in at least two of the sampling rounds are given in Table 4 . Fisher exact test did not show any difference of proportion of Ab seroconversion/seroreversion and proportions of change to Ag seropositivity/seronegativity between sexes , nor between periods . Proportion of Ab seroreversion was significantly higher than proportion of Ab seroconversion for each period ( Figure 2 ) . The change point analysis showed that the proportion of Ab seroreversion was significantly higher than proportion of Ab seroconversion until the age of 30 years . After this change point , the difference was not significant ( Figure 3 ) .
This is the first study reporting cumulative incidence and incidence rate figures of human T . solium larval infection in Latin America . The overall incidence rate of infection in the endemic rural community of Sabanilla , was 333 . 6 per 100 , 000 person-years ( 95% exact Poisson CI: [8 . 4–1 , 858] per 100 , 000 person-years ) , which suggests that less than 1% of the population becomes infected yearly with the parasite . In contrast , the incidence rate of exposure to T . solium is much higher: about 14% of the population has a yearly contact with the parasite . The latter estimates are in line with observed and simulated antibody seroconversion rates ranging from 8 to 25% in Peru , Colombia and Ecuador [16] , [18] . Proportions of change to Ag seropositivity/seronegativity and Ab seroconversion and seroreversion were identical in males and females indicating that both genders get equally infected with/are equally exposed to the parasite . Moreover , these proportions did not significantly vary in time ( one year period ) . On the other hand , proportion of Ab seroreversion was significantly higher than proportion of seroconversion Ab for each period and a change point analysis showed that proportion of Ab seroreversion was significantly higher than proportion of Ab seroconversion until the age of 30 years . After this change point , the difference was not significant . These results corroborate the findings of Praet et al . ( 2010 ) [16] suggesting a higher proportion of seroreversion before the age of 40 years due to a higher number of primary immune responses before this age . In other words , individuals will serorevert more rapidly before the age of 30–40 years because primary humoral response is shorter and weaker than secondary response . Thus , the proportion of seroreversions depends on the immunological status of the individuals . The dynamics of infection and exposure in the population , represented by the proportions of antigen and antibody results from the individuals who participated in all 3 sampling rounds , showed that the proportion of infected individuals remains low and stable during the whole study year , while the proportion of exposed individuals is remarkably higher . Of note is the high level of serological status variation with more than 20% of the population showing at least one antibody seroconversion/seroreversion during the year . Together with the prevalence estimates presented by period , these longitudinal data corroborates the findings of other studies conducted in Latin America highlighting a high prevalence of exposure to the parasite but a low prevalence of active infections . This contrast between exposure and infection may be linked to an effective resistance to the parasite acquired through long-term exposure of the population . In addition , these results confirm the occurrence of transient antibody responses in individuals living in T . solium endemic areas and suggest exposure to the parasite without infection or mild infections that are aborted by the natural immunity of the individual [20] . Mwape et al . ( 2013 ) [20] conducted a similar community-based longitudinal study in the Eastern Province of Zambia . While a much higher incidence rate was observed in the African endemic area , similar higher proportion of Ab seroreversion than Ab seroconversion and the presence of transient antibody responses were described . Further studies are needed to unravel the difference of parasite transmission patterns in different epidemiological settings . Specifically , research should focus on identifying the causes for differences in infection levels . In this context , the identification of tapeworm carriers in a community should be based on improved methods , because sensitivity and specificity of conventional coprological methods are low [36] . Our study has limitations that are mainly due to the inhabitant proportion of participation . Although 967 individuals from a total of 1045 inhabitants provided at least one blood sample , 396 provided only one sample and another 283 provided two blood samples . Compliance in participating in all 3 sampling rounds was of 288 volunteers despite extensive information sessions prior to the sampling procedure . The main reason for irregular participation was the absence of the individuals for professional duties . Reduced participation can have an impact on the precision of the estimation of the incidence rate , however , the participation on the three sampling rounds is still representative of the total population and all the participants selected for the incidence rate estimation match all the selection criteria for this calculation ( n = 288 ( 27 . 6% ( 95% CI: [24 . 9–30 . 4] ) ) ) . Another limitation of the study is the limited number of samplings and the relatively long sampling intervals depending on logistic , economic and ethical constraints . In other words , much more information could have been produced if more sampling had been organized at shorter time intervals , i . e . the positivity status of the participants would have been more accurately monitored over time . The incidence rate estimation for both infection and exposure is likely to be lower with increasing intervals between samplings: a proportion of the new infections may be undetected and the time of occurrence of a new infection overestimated . A quicker detection of new infections will result in a decrease of the number of person-years at risk and consequently in higher estimates of the incidence rate . Finally , even though the tests used in this study have shown high sensitivity and specificity , false positive and negative individuals may bias the prevalence and the incidence rate estimates . Bayesian estimation of infection with T . solium larva prevalence has been used to estimate the true prevalence of infection with an exposure to T . solium . The Bayesian estimation corrects the apparent prevalence at , but does not allow to know the true infection status at the individual level . Consequently , it does not allow to estimating the true incidence rate . In conclusion , the present study underlines the importance of conducting longitudinal serological follow-up allowing generating incidence rather than prevalence data to fully understand the transmission dynamics of the infection and to avoid under/overestimation of the occurrence of the parasite . Similar cohort studies assessing the effect of risk factors such as development of immunity and behavioral factors should be conducted to identify all the parameters that may influence parasite transmission . Understanding the transmission dynamics of T . solium is essential to develop ad hoc cost-effective prevention and control programs . The estimates generated here may now be incorporated in epidemiological models to simulate the temporal transmission of the parasite and the effects of control interventions on its life cycle [19] . These estimates are also of high importance to assess the burden of T . solium cysticercosis since incidence data are needed to make regional and global projections of morbidity and mortality related to cysticercosis . To this end , the link between the incidence rate of infection and health outcomes related to human cysticercosis , such as epilepsy and chronic headache , as well as the case-fatality ratio still need to be estimated .
|
Human cysticercosis is a neglected zoonotic parasitic disease causing severe health disorders such as epilepsy and even death . Cysticercosis is related to poverty , inadequate hygiene conditions and traditional pig farming . The present study describes the dynamic nature of human Taenia solium larval infections in an Ecuadorian endemic community . In this study we report for the first time incidence rate and cumulative incidence figures of human T . solium larval infections in Latin America . The simultaneous use of antibody and antigen serological detections allowed estimating both parasite exposure and infection rates , respectively . While about 13% of the inhabitants were exposed to T . solium eggs , less than 1% of the population became yearly infected with the parasite . This contrast between exposure and infection may be linked to an effective resistance to the parasite acquired through long-term exposure of the population and differs from the African situation , where much higher levels of infection have been observed . These estimates are of high importance to understand the epidemiology of T . solium in order to develop ad hoc cost-effective prevention and control programs . They are also essential to assess the burden of T . solium cysticercosis since longitudinal data are needed to make regional and global projections of morbidity and mortality related to cysticercosis .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"taeniasis",
"infectious",
"diseases",
"helminth",
"infections",
"veterinary",
"diseases",
"zoonoses",
"medicine",
"and",
"health",
"sciences",
"neurocysticercosis",
"epidemiology",
"cysticercosis",
"neglected",
"tropical",
"diseases",
"biology",
"and",
"life",
"sciences",
"tropical",
"diseases",
"parasitic",
"diseases",
"veterinary",
"science"
] |
2014
|
Incidence of Human Taenia solium Larval Infections in an Ecuadorian Endemic Area: Implications for Disease Burden Assessment and Control
|
In the bloodstream of mammalian hosts Trypanosoma brucei undergoes well-characterised density-dependent growth control and developmental adaptation for transmission . This involves the differentiation from proliferative , morphologically ‘slender’ forms to quiescent ‘stumpy’ forms that preferentially infect the tsetse fly vector . Another important livestock trypanosome , Trypanosoma congolense , also undergoes density-dependent cell-cycle arrest although this is not linked to obvious morphological transformation . Here we have compared the gene expression profile of T . brucei and T . congolense during the ascending phase of the parasitaemia and at peak parasitaemia in mice , analysing species and developmental differences between proliferating and cell-cycle arrested forms . Despite underlying conservation of their quorum sensing signalling pathway , each species exhibits distinct profiles of gene regulation when analysed by orthogroup and cell surface phylome profiling . This analysis of peak parasitaemia T . congolense provides the first molecular signatures of potential developmental competence , assisting life cycle developmental studies in these important livestock parasites . Furthermore , comparison with T . brucei identifies candidate molecules from each species that may be important for their survival in the mammalian host , transmission or distinct tropism in the tsetse vector .
African trypanosomes are responsible for Human African Trypanosomiasis ( HAT ) and Animal African Trypanosomiasis ( AAT ) in sub Saharan Africa[1] . The human disease is caused by two sub-species of Trypanosoma brucei , T . b . rhodesiense and T . b . gambiense , whilst Trypanosoma brucei brucei is responsible only for animal disease because it is sensitive to the trypanolytic component of human serum[2] . However , T . b . brucei is not the major cause of AAT , this being predominantly caused by infection with two alternative trypanosome species , Trypanosoma congolense and Trypanosoma vivax[3] . In combination , AAT threatens approximately 55 million cattle in sub-Saharan Africa and despite considerable costs to agriculture of prophylaxis against trypanosomiasis , approximately 3 million animals succumb each year[4] . Trypanosoma congolense is considered the most important agent of disease , with Trypanosoma brucei brucei comprising only a minority of infections . Although African trypanosomiasis is caused by different trypanosome species , these species are all transmitted by the tsetse fly[5] . This arthropod disease vector transmits the parasites by feeding on the blood of infected mammalian hosts , after which the trypanosomes undergo life cycle development in the fly before being transmitted to a new mammalian host . Although all three trypanosome species ( T . brucei , T . congolense , T . vivax ) have tsetse flies as their vector , the developmental progression of the different parasites differs within the arthropod[6] . For Trypanosoma brucei , the parasites enter the midgut , where they multiply as procyclic forms , before migration to the salivary glands where attached epimastigote and then metacyclic forms develop , the latter being mammal infective . For Trypanosoma congolense , there is also multiplication in the fly midgut as procyclic forms before the parasites develop in to long trypomastigotes and migrate via the proventriculus and foregut to the proboscis and cibarium where the transition to epimastigote forms and then metacyclic forms occurs [7] . Contrasting with the other trypanosome species , Trypanosoma vivax does not develop in the fly midgut , but rather matures in the proboscis forming epimastigote and infective metacyclic forms in the mouthparts . As a consequence of this less elaborate developmental path , Trypanosoma vivax can also exhibit mechanical transmission by tsetse flies and other biting arthropods , this having assisted the spread of this parasite outside the tsetse belt and in South America [3] . For Trypanosoma brucei and Trypanosoma congolense , the development of the parasite within the tsetse fly vector is characterised by their expression of stage-specific surface proteins , with EP and GPEET procyclin characterising early and late procyclic forms in the tsetse midgut for T . brucei[8] , and epimastigote forms expressing the BARP surface coat[9] . For T . congolense , GARP was originally described as a marker for the midgut forms[10] , but subsequent analysis has identified a procyclin-related protein encoded in the T . congolense genome which seems to be functionally equivalent to EP/GPEET procyclin[11 , 12] and grouped with T . brucei procyclins in a cell surface phylome analysis ( i . e . the evolutionary relatedness between predicted cell surface proteins in different trypanosome species ) . One characteristic of Trypanosoma brucei is that it exhibits clear developmental preadaptation for transmission in the mammalian host . Specifically , the parasite proliferates as a morphologically slender form until parasite numbers increase , at which point a density dependent quorum sensing ( QS ) mechanism causes differentiation to a non-proliferative , stumpy form[13] . The stumpy form is characterised by altered morphology[14] , cell cycle arrest in G1/G0[15 , 16] , a reduced anterior free flagellum , expression of the stumpy specific protein PAD1 [17]and appearance of an elaborated mitochondrion . The latter represents the preparation for metabolic adaptation of the parasite to the tsetse midgut , since in the bloodstream energy generation is via the use of blood glucose in glycolysis , whereas in the vector oxidative phosphorylation and the utilisation of proline as an energy source predominates[18 , 19] . Critically , stumpy forms are adapted for transmission[20 , 21] , exhibiting efficient and synchronous differentiation to procyclic forms when exposed to a cis-aconitate , mild acid or protease stimulus , of which the latter two conditions are lethal to slender forms[22] . The generation of transmissible stumpy forms in the bloodstream is stimulated by a soluble stumpy induction factor ( SIF ) [23] that is currently unidentified , although the signalling pathway through which the signal is transduced in the trypanosome has been characterised in some detail[24 , 25] . Whilst the production of stumpy forms in T . brucei is characteristic of that species , Trypanosoma congolense has also been shown to undergo a related phenomenon albeit without obvious morphological transformation[26] . Thus , T . congolense in the mammalian bloodstream exhibit proliferation control in response to parasite numbers , accumulating in G1/G0 when parasite numbers exceed approximately 8x107/ml . Their genome also encodes orthologues of T . brucei QS signalling pathway components , and at least one of these ( TcHYP2; TcIL3000 . 0 . 19510 ) can complement a T . brucei null mutant to restore stumpy formation in that species . This raises the question of whether there is a distinct developmental form of Trypanosoma congolense equivalent to the stumpy form of T . brucei . Here , we have analysed the transcriptome of T . congolense in the ascending phase of its parasitaemia and at the peak of parasitaemia when the proliferation of the parasites is reduced by an accumulation in G1/G0 . We have then compared these with the transcriptomes of T . brucei slender and stumpy forms , respectively . This has defined molecular characteristics of the peak parasitaemia T . congolense forms and highlighted molecules that may adapt the parasite for chronicity in the bloodstream or preadapt the parasite for its developmental path in the tsetse fly . The identification of enriched molecules at the peak of parasitaemia may provide morphology-independent markers for a possible transmissible form of T . congolense .
All animal experiments were carried out after local ethical approval at the University of Edinburgh Animal Welfare and Ethical Review Body ( approval number PL02-12 ) and were approved under the United Kingdom Government Home office licence P262AE604 to satisfy requirements of the United Kingdom Animals ( Scientific Procedures ) Act 1986 . T . congolense IL3000 and T . brucei EATRO 1125 AnTat1 . 1 . 90:13 parasites were used for infections . T . congolense IL3000 was derived from the ILC-49 strain that was isolated from a cow in the Trans Mara , Kenya [27] . T . brucei EATRO 1125 AnTat1 . 1 . 90:13 were provided by Professor Markus Engstler and Professor Michael Boshart . For generation of T . b . brucei RNA-seq samples , mice were treated with 25mg/ml cyclophosphamide at least two hours prior to infection . For generation of T . congolense RNA-seq samples cyclophosphamide was not used . Infections were usually monitored daily from day 3 post-infection by tail snip . On the final day of an experiment , total blood was harvested from mice by cardiac puncture using a Microlance 0 . 6 x 25mm needle and a 2ml syringe containing 250μl of 2% sodium citrate . Trypanosomes were purified from whole blood by passage through a DE52 column ( Whatman anion exchange cellulose , Z742600 ) at pH 7 . 8 . Cells were smeared onto slides , left to dry and fixed in cold methanol , and these slides were used for cell cycle analysis . Slides were initially rehydrated for 5 minutes in 1x PBS then 30μl of a DAPI working dilution ( 10μg/ml in PBS ) was applied to the smears . Slides were incubated in a humidity chamber for 2 minutes and were then washed for 5 minutes in 1x PBS . Slides were then mounted with 40μl Mowiol containing 2 . 5% DABCO ( 1 , 4-diazabicyclo[2 . 2 . 2]octane ) and were left to dry overnight at room temperature in the dark , before storage at 4°C . Slides were analysed on a Zeiss Axioskop 2 plus for KN configuration counts , and QCapture software ( QImaging ) was used for image capture . RNA samples were prepared using the Qiagen RNeasy kit ( Qiagen , 74106 ) according to the manufacturer’s instructions . Briefly no more than 5x107 cells were pelleted and resuspended in 594μl RLT buffer with 6μl β-mercaptoethanol . Samples were stored at -80°C before processing according to the protocol ‘Purification of Total RNA from Animal cells using spin technology’ . RNA samples were resuspended in 10μl RNase free water per 107 cells used to prepare the sample . The concentration and purity of the resultant RNA was measured on a Nanodrop spectrophotometer . RNA samples were stored at -80°C . Northern blotting was carried out using digoxigenin labelled riboprobes as described in [28] . Protein samples were prepared and analysed for PAD1 expression according to[28] . RNA-Seq was carried out by BGI Hong Kong using a Truseq library preparation protocol with polyA selection , and sequenced on a HiSeq2000 platform with 90bp paired end reads . Quality control on generated data was performed using Fast QC v0 . 10 . 0 ( http://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) and paired end reads were trimmed using cutadapt v1 . 9 . 1 [ref: https://cutadapt . readthedocs . io/en/stable/] before alignment of the reads to the T . congolense genome ( ftp://ftp . sanger . ac . uk/pub/project/pathogens/gff3/CURRENT/ , with sequences and CDS gene model coordinates obtained 18 February 2015 ) , or T . brucei genome ( ftp://ftp . sanger . ac . uk/pub/project/pathogens/gff3/CURRENT/ , with sequences and CDS gene model coordinates obtained 18 February 2015 ) . Alignment was carried out using Bowtie 2 v2 . 2 . 7 ( http://bowtie-bio . sourceforge . net/bowtie2/index . shtml [29] , and was not restricted to annotated genes . Read counts were normalised to reads per kb/ map ( rpkm ) , in order to account for gene size and the different read depths of the replicates . A joint quantile threshold cut-off of 10% was applied across the relevant replicate rpkms to remove the lowest numbers of reads from each group ( e . g . ascending , peak ) . A group-wise comparison using the R/Bioconductor software package limma [30]was then performed between the ascending parasitaemia and peak parasitaemia replicates; significance values were corrected for multiple testing ( adjusted P values ) using eBayes . Phylome data were as described by[11]; the family member sets were sourced from http://www . genedb . org/Page/trypanosoma_surface_phylome . OrthoMCL [31] clusters were obtained from GeneDB [in 2015] . The “TrypanosomA” ortho group was generated using: T . cruzi , T . vivax , T . congolense , T . brucei 927 and T . brucei gambiense . The “TrypanosomE” ortho group was generated using: T . cruzi , T . vivax , T . congolense , T . brucei 927 , T . brucei gambiense , L . braziliensis , L . major , L . infantum , L . donovani , L . mexicana . Data is available at GEO via the link: https://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE114813 . All graphs were produced using GraphPad Prism version 6 ( GraphPad Software , La Jolla , California , USA , www . graphpad . com ) . Microscope images were generated ( i . e . cropping , brightness , contrast , overlay , scale bar application ) using Image J 64 [32] .
The 804 transcripts that showed significant changes between peak and ascending parasitaemia were grouped by the description of their protein product . The category that was most represented for transcripts more abundant at peak parasitaemia was ‘VSG’ , followed by ‘hypothetical protein’ ( S2a Fig ) . The predominant categories for transcripts with reduced abundance at peak parasitaemia were ‘hypothetical protein’ , followed by ‘other’ ( S2b Fig ) , this including transcripts predicted to be associated with proliferation ( e . g . paraflagellar rod protein , DNA polymerase catalytic subunits , and a putative cell division protein kinase ) and three putative amino acid transporters . To assist with functional assignment for those transcripts differentially regulated between ascending and peak parasitaemia , a BLASTP search with the encoded T . congolense protein sequences was carried out against T . b . brucei 927 proteins and the output was interrogated for genes known to be linked to differentiation . Searching for the stumpy specific marker PAD1 , revealed a putative orthologue , TcIL3000 . 0 . 18180 but this was not significantly more abundant in peak relative to ascending parasitaemia . However , this molecule was also related to at least 2 other T . congolense predicted proteins ( TcIL3000 . 7 . 5000 , TcIL3000 . 0 . 02160 ) similar to the PAD gene family in T . brucei , comprising members of cell surface phylome family 58 [11] . Assignment of reads to differentially expressed members of the family might suppress any obvious elevation in peak parasitaemia samples . Moreover , TcIL3000 . 0 . 18180 was itself quite abundant in ascending phase T . congolense , which may reflect an early elevation of its expression prior to peak parasitaemia similar to the expression of PAD1 mRNA ( but not protein ) in intermediate form T . brucei [33] . To interrogate those transcripts most upregulated ( i . e . at least 4-fold ) at peak parasitaemia in T . congolense , T . brucei orthologues of the gene cohort ( where available ) were analysed for a loss-of-fitness phenotype following RNAi at any T . brucei life cycle stage , but particularly differentiation[34] , and also cell surface phylome family membership[11] . Seven genes in the cohort of 65 elevated transcripts generated a defect in differentiation as determined by genome-wide RNAi fitness screening . Of these , a putative UDP-Gal or UDP-GlcNAc-dependent glycosyltransferase ( orthologue Tb927 . 8 . 7140 ) exhibited a fitness deficit restricted only to differentiation , as opposed to bloodstream or procyclic form growth . More strikingly , however , of the strongly elevated transcripts , 46 transcripts ( or the predicted T . brucei orthologue ) were annotated as encoding cell surface phylome members ( 35 transcripts when those belonging to surface family groups described as VSG were excluded ) . Of these remaining 46 transcripts , 29 transcripts belong to orthogroup OG5_127658 [31] , an orthogroup being defined as genes derived from a single gene in the last common ancestor of the species groups . This orthogroup incorporates proteins of cell surface phylome group 15 , which is shared between T . brucei and T . congolense but whose membership is considerably expanded in T . congolense . This family contains ESAG6 and 7 proteins associated with transferrin uptake in T . brucei ( see later ) . The enrichment of cell surface phylome family 15 in T . congolense prompted us to explore the developmental regulation of other cell surface phylome members , and particularly those restricted only to T . congolense ( Fig 3 ) . Analysis of the expression profile of members of these T . congolense species-specific family groups ( families 17 , 18 , 20 , 21 , 22 ) demonstrated that members of families 17 , 18 and 21 were downregulated in T . congolense at peak parasitaemia , albeit to different levels and with varying levels of statistical support . In contrast , every identified representative of Family 22 was elevated at peak parasitaemia ( with a range between logFC 0 . 19–2 . 60 ) ( Fig 3 ) , this also being observed when gene-specific reads were considered ( S3 Fig ) . Family 22 comprises a VSG-related protein family of 175–186 amino acids whose open reading frame is always closely adjacent to , or within the 3’UTR , of a VSG gene . The predicted protein sequence of Family 22 members is not related to other known proteins or protein features and could also represent a ncRNA . Similar to Family 22 , all Family 20 members were also upregulated at peak parasitaemia , although the overall level of regulation was more modest . Interestingly , Family 12 members , which comprise the T . brucei procyclins and the T . congolense related molecules ( TcIL3000 . 0 . 53640; TcIL3000 . 0 . 02860; procyclin like ) were not upregulated in the T . congolense peak parasitaemia samples; nor were GARP or CESP family members ( cell surface phylome family 50 ) . Among the T . congolense transcripts at least 4-fold more abundant at peak relative to ascending parasitaemia ( adj p<0 . 05 ) there were 23 proteins described as ‘hypothetical’ . The amino acid sequences for these 23 hypothetical proteins were searched using the InterPro search tool to identify potential domains of interest ( S2 Table ) . Seven of these proteins were predicted to have one transmembrane domain by the Phobius tool[35] , and 3 of these predictions were also supported by TMHMM[36] . Further , 3 of the hypothetical proteins were predicted to have more than one transmembrane domain by Phobius , and in some cases this was supported by predictions by TMHMM . None of these hypothetical proteins were described as part of the Trypanosoma cell surface phylome[11] , and so they may localize to internal membranes . Among the enriched transcripts encoding hypothetical proteins was also TcIL3000_0_60190 ( logFC 2 . 156 ) , annotated as a dicer-like protein . In fact , this molecule forms a member of orthogroup OG5_133097 which incorporates a family of proteins largely comprised of imperfect 12 amino acid repeats . There are 25 members of this family in T . congolense , out of 33 in the orthogroup , no other members being found in other kinetoplastids . The protein encoded by TcIL3000_0_60190 contains a predicted signal peptide , N-terminal transmembrane domains ( 2 ) and BLASTP similarity to adhesins , adherence factor and cell agglutination proteins ( S4 Fig ) . Interestingly , all members of this orthogroup identified in our datasets exhibited transcript elevation at peak parasitaemia , though many fell below the significance threshold ( Fig 4 ) . To act as comparator for the T . congolense ascending and peak parasitaemia RNA profiles , transcriptome datasets were generated from T . brucei slender and stumpy form parasites isolated from mouse infections ( Fig 5 ) . Thus , parasites from infections with T . b . brucei EATRO 1125 AnTat 1 . 1 90 . 13 were collected at either ascending parasitaemia when parasites were of predominantly slender morphology , or at peak parasitaemia when parasites were of stumpy morphology . Triplicates were prepared for each developmental form , with the stumpy RNA samples and two of the slender RNA samples each derived from an individual T . b . brucei EATRO 1125 AnTat 1 . 1 90 . 13 infection . The third slender RNA sample was derived from a pool of 4 infections to generate sufficient material for RNA-Seq analysis . Matching the analysis in T . congolense infections , the cell cycle status of the parasite isolates was also analysed , with proliferative slender cells expected to be enriched for the 2K1N/2K2N configuration with respect to G1/G0-arrested stumpy forms ( the parasitaemias and associated cell cycle data for each sample are shown in S1 Fig ) . Fig 5A and 5B confirms that the parasites used to prepare the stumpy RNA samples were enriched in 1K1N , whereas those parasites used to prepare slender RNA samples were proliferative with a greater proportion of cells that had either a 2K1N or 2K2N configuration ( Fig 5A and 5B ) . Fig 5C shows the isolated total RNA analysed by Northern blotting using a probe targeting PAD1 , confirming that the levels of the transcript were higher in the stumpy RNA samples than in the slender RNA samples ( Fig 5C ) , although some transcript was detected perhaps reflecting the beginning of development to intermediate forms in the isolated slender cells ( intermediate forms express as much PAD1 mRNA as stumpy forms;[37] ) . Further validation of the developmental status was obtained by western blotting using a PAD1 specific antibody ( Fig 5D ) , which confirmed that the levels of PAD1 protein expression were considerably higher in the protein samples from the parasites used to isolate stumpy RNA than in control protein samples from slender T . b . brucei EATRO 1125 AnTat 1 . 1 90 . 13 parasites cultured in vitro . An equivalent analysis of the material used to generate slender RNA for RNA-Seq analysis could not be carried out due to the low parasitaemia for these samples , but morphological analysis confirmed the slender and stumpy morphotypes for parasites from which each slender and stumpy RNA sample was derived . ( Fig 5E ) . Matching the analysis of T . congolense ascending and peak parasitaemia samples , slender and stumpy RNA-Seq reads were processed to generate Log2FC values comparing developmental forms . In total , 5028 genes demonstrated significant fold changes in abundance ( adj . P-value <0 . 05 ) ( Fig 6A and 6B ) . Of these , 3096 transcripts were increased in stumpy relative to slender parasites , and 1932 transcripts were decreased in stumpy relative to slender parasites ( S3 Table ) . When changes of less than 2-fold magnitude were excluded , 421 transcripts were at least 2-fold more abundant in stumpy relative to slender parasites , and 501 transcripts were at least 2-fold less abundant in stumpy relative to slender parasites . This contrasts with the absence of transcripts downregulated at least 2-fold in peak versus ascending parasitaemia T . congolense samples . Transcripts were further sorted to exclude those annotated as VSG or with no protein product assigned . In comparison to T . congolense , relatively few regulated transcripts were annotated as VSG and , of those transcripts regulated at least 2-fold that were described as VSG , all were decreased in abundance in stumpy relative to slender parasites . The 3096 transcripts significantly increased , or 1932 transcripts significantly decreased , in stumpy relative to slender form parasites , were sorted by their protein product description ( S5 Fig ) . The most common category of differing transcript abundance between stumpy and slender forms was ‘hypothetical proteins’; there are 3460 genes annotated to encode ‘hypothetical proteins’ out of 9729 protein coding genes in the genome overall . Other well-represented categories included protein kinases , phosphatases , and RNA-binding proteins . An increase in abundance of transcripts encoding transporter proteins , and particularly amino acid transporters , was seen whereas glucose transporters were downregulated . There was also an upregulation of procyclin and BARP surface protein transcripts consistent with the preparation for transmission . Additionally , consistent with changing metabolic requirements on transmission , transcripts described as mitochondrial carrier proteins , or as NADH-ubiquinone oxidoreductase ( part of the mitochondrial electron transfer chain in procyclic forms , [38] ) , were also increased in abundance in the stumpy form , as would be expected for this life cycle stage with a more elaborated mitochondrion than slender forms [14] . As previously noted , the ESAG9 family were significantly increased in abundance in stumpy forms [39 , 40] . Similarly , the protein-associated with differentiation 1 ( PAD1 ) transcript , alongside those of PAD2 and 3 , were found to be of increased abundance in stumpy forms . However , the fold change in PAD1 was not as high as PAD2 , probably due to some PAD1 elevation due to its early expression in intermediate forms present in the isolated ‘Slender’ material used . Also matching expectation , increased abundance of the PTP1-interacting protein , TbPIP39 , transcript was observed in the stumpy-enriched data set [41] . In a study of early proteomic changes made during commitment to differentiation to the procyclic form [42] TbPIP39 protein abundance increased 3 hours after induction of differentiation , alongside a trans-sialidase and eukaryotic initiation factors . The increased abundance of these transcripts in stumpy forms is likely a necessary preadaptation for the rapid proteomic changes required after transmission . With respect to transcripts downregulated in stumpy forms , these included flagellar protein , histone , and beta-tubulin transcripts , consistent with the growth-arrested status of the quiescent transmission stage . Additionally , transcripts for the haptoglobin-haemoglobin receptor ( Tb927 . 6 . 440 ) were reduced in abundance in the stumpy form as expected[43] . In combination , these data confirmed that the slender and stumpy RNA-Seq samples corresponded to the known characteristics of these developmental forms , validating them as a suitable comparator for T . congolense ascending and peak parasitaemia forms . To compare regulated processes between peak parasitaemia forms from T . brucei and T . congolense , we specifically analysed the expression of genes involved in several cellular functions that are linked to developmental regulation and cellular quiescence in T . brucei . Namely , we explored the regulation of transcripts involved in glycolysis , cell proliferation ( histone , tubulin , paraflagellar rod transcripts ) , kinetochore proteins ( KKT transcripts; [44] ) , amino acid transporters and RNA binding proteins ( including pumillio proteins , zinc finger proteins and mitochondrial RNA binding proteins ) . Except glycolysis , where the specific enzymes involved were analysed , in each case protein product descriptions were used as the search term and the expression profile of the cohort plotted with respect to log2 FC between either stumpy and slender forms of T . brucei , or between peak parasitaemia and ascending parasitaemia T . congolense . Fig 7 shows the expression profiles exhibited in each transcript cohort , which demonstrated a greater regulatory trend in stumpy forms of T . brucei than in peak parasitaemia T . congolense with respect to their proliferating ascending parasitaemia counterparts . Thus , most enzymes involved in glycolysis were down regulated in stumpy forms consistent with their quiescence , whereas no change between peak and ascending parasitaemia was seen in T . congolense . Similarly , mRNA markers of cell proliferation , such as histone , paraflagellar rod protein and alpha and beta tubulin transcripts were strongly downregulated in stumpy forms but these were largely unchanged in the T . congolense peak parasitaemia samples . The recently identified and divergent kinetochore proteins also showed distinct regulation in stumpy forms compared to slender forms , and this was less pronounced in T . congolense . Contrasting with the down regulation of the aforementioned transcripts in quiescent stumpy forms , transcripts annotated as amino acid transporters were predominantly upregulated whereas in T . congolense no general increase was observed . The regulatory pumillio class of RNA binding protein mRNAs were also upregulated in stumpy forms , perhaps associated with their role in translational regulation , as were those annotated ‘mitochondrial RNA binding protein’ , however an equivalent regulation was not observed in T . congolense . Zinc finger proteins , known to be important in several development events in T . brucei were not consistently upregulated as a group in either T . brucei stumpy forms or T . congolense peak samples , although two transcripts were highly elevated: TcIL3000_0_11070 in T . congolense ( Log2 FC 3 . 56; adj p value = 0 . 057 ) and Tb927 . 5 . 810 ( ZC3H11 ) in T . brucei ( Log2 FC 2 . 01; adj p value = 0 . 001 ) . To expand the comparison of the T . brucei and T . congolense data sets based on biological function , the transcripts identified by RNA-Seq were placed into orthologue groups based on OrthoMCL [31]and regulation of orthogroup members was compared between species . When the Log2FC values for T . congolense ‘peak relative to ascending’ parasitaemia were plotted against orthologue cluster , certain clusters had multiple members with pronounced increases in abundance at peak parasitaemia . These orthologous groups did not display the same trend in the T . brucei ‘stumpy relative to slender’ plot ( Fig 8a; S4 Table ) and many of the T . congolense peak parasitaemia-enriched orthologous groups that contained hypothetical proteins were T . congolense specific or enriched ( S5 Table ) . This included members of the aforementioned FAM22 cell surface family[11] . Also upregulated , as examples , were orthogroups 1736 ( GRESAG4 ) , 4428 ( ESAG3 ) , 306 ( cysteine rich acidic membrane proteins; comprising hexamer repeat sequences ) , 6700 ( SecA-DEAD-like domain containing protein ) , and 9522 ( glycosyl transferase proteins ) . Similarly , the analysis for ‘stumpy versus slender’ enriched T . brucei transcripts identified the T . brucei specific ESAG9 orthogroup ( Cell surface phylome family 2 ) , plus procyclin ( Cell surface phylome family 12 ) and hypothetical protein containing orthogroups . To directly compare the regulation of orthologue clusters shared between the two trypanosome species , the mean log fold change ( Log2FC ) from ascending to peak T . congolense parasitaemia was calculated for each orthogroup and compared with the same analysis in T . brucei . Focusing on those orthogroups for which the mean T . congolense fold change in abundance was greater than two , in most cases the mean change in abundance for stumpy relative to slender forms was also positive for these orthologue groups , but in only one case ( cluster 3122 ) was the mean change in abundance of comparable magnitude between the T . brucei and T . congolense data sets ( Fig 8 ) . Cluster 3122 members are described as conserved hypothetical proteins , and RNAi-mediated knock down of the T . brucei orthologue generates a loss-of-fitness on differentiation from bloodstream to procyclic forms[34] . An InterPro search using the T . congolense and T . brucei sequences from cluster 3122 did not reveal any predicted domains . Finally , for the orthologue clusters shared between T . brucei and T . congolense , we plotted the differences between the mean Log2FC values for T . congolense ‘peak versus ascending’ data and T . brucei ‘stumpy versus slender’ data ( Fig 9 ) . Clusters with the largest differences between T . congolense and T . brucei are labelled with the cluster ID and summarised in S6 Table . Beyond orthogroups involved in cell proliferation ( histones , PFR , etc . , as highlighted earlier ) the most pronounced differentially-regulated orthogroup between T . brucei and T . congolense was the ‘Transferrin receptor ( TFR ) - like’ orthogroup highlighted earlier as significantly enriched in peak parasitaemia T . congolense . This orthogroup has many more members in the T . congolense genome than in T . brucei and includes ESAG6 like ( Surface Phylome Family 15; 45 genes ) and PAG like ( Surface Phylome Family 14; 31 genes ) groups . Analysis of the expression of each type , revealed that it is specifically the ESAG6 TFR family that is upregulated as a group , and that the PAG like TFR group are largely unregulated between ascending and peak parasitaemia parasites ( Fig 10 , S3 Fig ) . Since the transferrin receptor of T . brucei is a heterodimer , with one of the components possessing a GPI anchor modification , we predicted the presence or absence of this modification on those transcripts upregulated in peak parasitaemia T . congolense using both the big-PI [45] and PredGPI [46] predictive algorithms . This revealed that approximately 45% of those transcripts encoding TFR were predicted to have a GPI anchor modification whereas 55% were not . This predicts that at peak parasitaemia transcripts encoding a functional heterodimer were upregulated , rather than one or other constituent components of the functional receptor .
The developmental cycles of different African trypanosome species differ dramatically in their tsetse fly vector , with T . brucei undergoing maturation in the salivary glands after proliferation in the midgut , whereas T . congolense matures in the proboscis or cibarium of the fly . Both species however , initially occupy the tsetse midgut and it would be expected that similar adaptations for transmission may be shared by these different parasite species . A key adaptation for transmission occurs in the bloodstream of mammalian hosts , whereby T . brucei generates an alternative developmental stage , the stumpy form , that is less sensitive to protease exposure or pH fluctuations , and which undergoes efficient differentiation to tsetse midgut procyclic forms in vivo and in culture . Trypanosoma congolense , by contrast , does not develop a morphologically-distinct stumpy form but has been shown recently to undergo cell-cycle arrest when at peak parasitaemia , reminiscent of the cell quiescence exhibited by mature stumpy forms . T . congolense also conserves the machinery identified in T . brucei as needed for density-dependent differentiation in the bloodstream [28] . In the work described here we sought to compare the gene expression changes between proliferating and arrested forms of each parasite species ( slender/stumpy for T . brucei; ascending/peak parasitaemia for T . congolense ) to identify common or distinct gene expression patterns in response to the density of each species in the bloodstream of their mammalian host . With respect to T . brucei , our ex vivo datasets do not compare well with those of a recent study using in vitro grown parasites when considering transcripts that show altered expression between slender and stumpy forms . This may reflect differences between ex vivo and in vitro derived parasites , mRNA isolation methods and/or inter-laboratory variation [47] . Overall our analyses highlight that T . brucei undergoes more extreme adaptation , with the downregulation of transcripts linked to cellular proliferation and metabolism and the upregulation of molecules required in the tsetse midgut after transmission . Although T . congolense also exhibits reduced proliferation at elevated parasite density , these parasites show much less evidence of progression to cellular quiescence , matching their absence of morphological transformation . Nonetheless , the parasites do not simply exhibit reduced proliferation; rather they significantly up-regulate transcripts encoding several specific surface protein families upon their development at peak parasitaemia . This indicates that , although monomorphic , these parasites exhibit development within the bloodstream and that this is associated with the expression of specific surface molecules that may promote infection chronicity or assist parasite survival in the mammalian host , or during their transmission to tsetse flies . Analysis of the cell surface phylome of different trypanosome species highlights the existence of 5 non-VSG protein families that are specific to T . congolense . In addition , there are four protein families shared with T . brucei ( but not T . vivax ) of which two , Family 14 and Family 15 are expanded in T . congolense . Of the T . congolense specific surface protein families , Families 17 , 18 and 21 did not change in expression as a cohort between ascending and peak parasitaemia , but Family 20 and particularly Family 22 members were upregulated during the development to the peak parasitaemia form , accounting for both reads assigned randomly between members for specific gene families and uniquely mapped reads ( S3 Fig ) . Similarly , of those families shared with T . brucei , the expanded Family 15 surface proteins were upregulated , contrasting with the Family 14 cohort despite both Families being annotated as transferrin binding . Thus , elevated expression of Family 15 , 20 and 22 members defines the peak parasitaemia forms of T . congolense . Of the upregulated surface protein families , Family 20 represents a family of small ( ~107 amino acid ) proteins that show no homology to other proteins and are of unknown function . Family 22 , in contrast , represents a much larger gene family with over 100 representatives , these encoding proteins of approximately 180 amino acids . The genes occupy an interesting genomic location , being found directly downstream of sub-telomeric VSG gene sequences and potentially positioned within the 3’UTR of these genes . As with Family 20 members , however , the predicted protein sequences do not provide any indication of the likely function of the family , although a subset do contain a predicted signal peptide , suggesting surface expression of at least some members . In contrast , Family 15 transcripts encode ESAG6 related transferrin receptor proteins . The Transferrin receptor ESAG6/7 is encoded as an integral component of the T . brucei VSG gene expression site and the proteins form a heterodimer to allow iron uptake . The trypanosome transferrin receptor proteins are predicted to be derived from VSG sequence , and this has diverged in T . congolense to an ESAG6 related family ( represented by Family 15 ) and a further family related to the procyclin associated genes ( represented by Family 14 ) . Of these two families , only Family 15 was upregulated in T . congolense peak parasitaemia parasites , with isoforms predicted to be modified by GPI addition and those without this modification upregulated , suggesting regulation of functional heterodimers . Their upregulation in T . congolense at peak parasitaemia perhaps relates to an increased need to scavenge transferrin at high parasite numbers , or when stressed at peak parasitaemia , although the peak parasitaemia levels in T . congolense were equivalent to the parasite levels for the isolated T . brucei stumpy samples . Sensitivity to iron depletion may be relevant for T . congolense in chronic livestock infections where trypanosome infection can be characterised by anaemia and limited iron availability as an anti-pathogen defence mechanism [48] . Beyond the described cell surface phylome members , other transcripts upregulated at peak parasitaemia in T . congolense were of potential interest . These included representatives of orthogroup OG5_133097 , all members of which were upregulated , although only one was upregulated more than 4 fold ( adj p<0 . 05 ) . These transcripts are predicted to encode a family of proteins with imperfect 12 amino acid repeats comprising the majority of the protein sequence . Although functionally uncharacterised , BLAST searches highlighted similarity to proteins involved in cell adhesion or agglutination . This is intriguing with respect to T . congolense , which is characterised by binding to culture flasks when grown as bloodstream forms . Indeed , the parasites adhere in the vasculature and in the proboscis during life in the blood or tsetse fly . Hence the expression of these proteins at peak parasitaemia may promote transmission by causing the parasites to accumulate , for example , in the skin . Alternatively , these proteins could contribute to virulence in the mammalian host through enhanced adhesion and perhaps sequestration of parasites . Exploration of these potential functions awaits the development of robust tools for reverse genetic analysis in T . congolense bloodstream forms . In comparison to T . brucei slender to stumpy transitions , the changes in T . congolense were relatively modest . This reflects the descriptions of these parasites in the literature where a morphological transformation is not obvious and the mitochondrial activity of the parasite is already well developed in the bloodstream , unlike in T . brucei where the slender form has a much simpler and underdeveloped mitochondrion than in stumpy or tsetse fly forms[14] . Recently , an analysis by flow cytometry of the parasites used in this study , T . congolense IL3000 , confirmed the absence of a clear morphological transition at peak parasitaemia despite their cell cycle arrest [28] . Whilst this could reflect the particular stock used , which has been subject to long term passage in rodents , other more recent field isolates also show a lack of clear morphological transformation , suggesting that monomorphism is a general feature of these parasites , rather than a laboratory adaptation . In the absence of morphological transformation , the question of whether peak parasitaemia T . congolense represents a distinct developmental form , like the stumpy form , becomes relevant . A number of lines of evidence suggest that there is developmental adaptation . Firstly , the parasites conserve the signalling pathway for quorum sensing exploited in T . brucei and at least one component is able to functionally complement a T . brucei null mutant to restore stumpy formation [28] . Secondly , although the downregulation of transcripts observed with stumpy form quiescence was less obvious in T . congolense , specific subsets of molecules were strongly upregulated in peak parasitaemia stages , arguing for an active developmental progression rather than a passive cell-cycle arrest only . Finally , T . congolense isolated at peak parasitaemia may be more able to infect tsetse flies than those during the ascending phase [49] though this has not been the observation in all studies[50] . All of these observations argue for a specific adaptation for transmission in T . congolense and their lack of morphological adaptation makes it even more important than in T . brucei that specific molecular markers for that form are identified . The transcriptome datasets generated here provide not only potential molecular signatures of preadaptation but also molecular tools that can help to dissect the transmission biology of T . congolense and how this compares with the alternative developmental fates of T . brucei and T . vivax . This complements recent transcriptome studies derived for tsetse-inhabiting stages for both T . congolense [51] and T . brucei [52 , 53] . Furthermore , in bacterial systems quorum-sensing regulates expression of virulence factors [54] , thus some of the transcripts differentially regulated at low versus high parasite density could represent novel virulence factors in T . congolense infection .
|
Animal African trypanosomiases are important diseases of livestock in sub-Saharan Africa . Two of the responsible parasite species are Trypanosoma brucei and Trypanosoma congolense , both being blood-borne parasites transmitted by tsetse flies . In T . brucei there is a well-characterised developmental event in the bloodstream that prepares the parasite for tsetse transmission—the generation of morphologically stumpy forms . In contrast , Trypanosoma congolense does not undergo the same obvious morphological event , but does respond to parasite density in the mammalian bloodstream by accumulating as a cell cycle arrested form . This prompted us to explore the adaptations of T . congolense in response to cell density in blood and to compare this with T . brucei . The datasets generated , and their analysis , represent a first detailed transcriptional profile for T . congolense and also a new high-resolution analysis of the developmental forms of T . brucei in a mammalian host . Critically , the analysis also carefully characterised the biological material used for RNA-seq analysis with respect to cell cycle status , morphology and the expression ( in the case of T . brucei ) of PAD1 –a molecular marker for stumpy forms . The manuscript highlights clear differences in the developmental adaptation of each parasite species , with T . congolense showing less extreme adaptation at peak parasitaemia than T . brucei . Nonetheless , several predicted surface protein families in T . congolense are strongly upregulated at high parasite density in the bloodstream , which may represent adaptations for their transmission or survival .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"parasitic",
"cell",
"cycles",
"cell",
"cycle",
"and",
"cell",
"division",
"trypanosoma",
"congolense",
"cell",
"processes",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"trypanosoma",
"brucei",
"developmental",
"biology",
"protozoans",
"proteins",
"life",
"cycles",
"biochemistry",
"trypanosoma",
"eukaryota",
"cell",
"biology",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"trypanosoma",
"brucei",
"gambiense",
"organisms",
"parasitic",
"life",
"cycles"
] |
2018
|
A gene expression comparison of Trypanosoma brucei and Trypanosoma congolense in the bloodstream of the mammalian host reveals species-specific adaptations to density-dependent development
|
As the brain matures , its responses become optimized . Behavioral measures show this through improved accuracy and decreased trial-to-trial variability . The question remains whether the supporting brain dynamics show a similar decrease in variability . We examined the relation between variability in single trial evoked electrical activity of the brain ( measured with EEG ) and performance of a face memory task in children ( 8–15 y ) and young adults ( 20–33 y ) . Behaviorally , children showed slower , more variable response times ( RT ) , and less accurate recognition than adults . However , brain signal variability increased with age , and showed strong negative correlations with intrasubject RT variability and positive correlations with accuracy . Thus , maturation appears to lead to a brain with greater functional variability , which is indicative of enhanced neural complexity . This variability may reflect a broader repertoire of metastable brain states and more fluid transitions among them that enable optimum responses . Our results suggest that the moment-to-moment variability in brain activity may be a critical index of the cognitive capacity of the brain .
During neurodevelopment , behavioural performance tends to improve in terms of speed and accuracy . This improvement usually entails a decrease in trial-to-trial variability as performance approaches ceiling e . g . [1] , [2] . A fundamental question is whether the neuronal dynamics that mediate behaviour show a similar decrease in variability . There are two arguments that furnish opposite predictions . The first is that neurodevelopmental trajectories will converge on optimal structure and dynamics , suggesting that trial-to-trial variations in evoked neuronal responses will decrease with age . This argument is particularly attractive in view of known neural pruning that accompanies normal brain maturation [3] . The second argument is that , if the computations underlying performance rest on adaptive , metastable brain dynamics [4] , there may be an age-related increase in trial-to-trial variability . Functional variability represents a greater repertoire of metastable brain states and the more facile state transitions [5] . In this work , we disambiguated between these competing hypotheses by relating trial-to-trial variability in behavior with brain electrical activity across a sample of children and adults . Neural systems can show a somewhat counterintuitive property , whereby optimal operations occur in the presence of a moderate amount of internal variability or noise [6] . For example , the phenomena of stochastic resonance describes how a simple nonlinear system can show an optimal signal-to-noise ratio with a moderate amount of noise , which enables the detection of weak periodic signals [7]–[9] . From cell channels to synapse to neural ensembles , noise seems to be an important parameter that shapes responsivity [10] , [11] . It has been suggested that there may need to be a degree of physiological variability for the brain to adapt effectively to an uncertain environment [12] . While there are sophisticated computational models demonstrating the beneficial effects of noise for network dynamics [13] , [14] , the direct relation between neural variability and the behavior variability of the organism has been largely unexplored [15]–[17] . We sought to characterize the relation of neurophysiological variability and behavioral variability in maturation . Critically , our focus was not on variability of the signal across individuals within a group ( interindividual ) , but rather the single-trial variability within an individual ( intraindividual ) [18] . Measures of single trial variability in electroencephalographic ( EEG ) signals within subjects were related to variability in response latency and accuracy . EEG signals were measured from children ( ages 8–15 yrs , n = 55 ) and young adults ( 20–33 yrs , n = 24 ) during the performance of a face recognition task [19] .
Behaviorally , all age groups showed high accuracy in the task , with adults near ceiling ( Figure 1a ) . Recognition accuracy for children , while lower than for adults ( one-way ANOVA , F ( 4 , 74 ) , = 9 . 07 , P<0 . 01 ) , was well above chance . Mean reaction time ( RT ) was much slower for children 8–11 years and similar for children 12–15 years and adults ( F ( 4 , 74 ) = 6 . 65 , P<0 . 01 ) . Importantly , the coefficient of variation in reaction time ( cvRT , standard deviation/mean RT within subject ) , which is an index of intra-individual variability , showed a gradual age-related decrease ( F ( 4 . 74 ) = 7 . 12 , P<0 . 01 ) . Though the standard deviation of RT ( sdRT ) showed the same trend as cvRT , because the standard deviation often scales with mean , we used cvRT which avoids this confound . Indeed the correlation between mean RT and sdRT was 0 . 63 and the correlation between mean RT and cvRT was 0 . 02 . The correlation between sdRT and cvRT was 0 . 78 . The average stimulus evoked electrical potentials across the age groups ( Figure 1b ) showed a characteristic maturational change with greater amplitude but longer latency for a positive deflection peaking at about 100 ms post-stimulus ( P100 ) [20] , [21] . A second observation was that the subsequent deflections ( N100 , P200 ) are easily observable in adults and gradually emerge in children [19] . The bias in children towards higher amplitude , but slower , electrophysiological signals is paralleled by a differential distribution in prestimulus , or baseline , spectral power . Across age , there was a gradual reduction in low frequency spectral power and a relative increase in power at higher frequencies ( Figure 1c , see supplementary material Text S1 for statistical analysis and Figures S1 ) . The relative change in spectral power density presumably underlies the reduction in the latency of the evoked responses and its multicomponent nature [22] , [23] , where the lower frequency bias in children would yield slow and broad evoked potentials . The emergence of higher frequencies with maturation would both decrease the evoked response latency and allow additional deflections to emerge ( i . e . , N100 , P200 , etc . ) . Two measures were used to evaluate brain signal variability . First , principal components analysis ( PCA ) was performed within each subject on their single trial EEG recordings . PCA identified the number of orthogonal dimensions , expressed as a proportion of the total possible , needed to express a certain amount of trial-to-trial variability ( 90% in the present case ) for each channel . In a deterministic system with highly stereotyped responses , only a few dimensions are needed to capture most of the variability . To the extent that trial-to-trial recordings differ from one another , total variability increases , and hence PCA dimensionality increases . PCA dimensionality estimates for the 200-ms intervals pre- and post-stimulus onset increased across the age groups ( Fig 2 , a , b ) . Statistical analysis of PCA dimensionality with partial least squares [PLS , 24] confirmed a significant linear increase across age groups in dimensionality that was expressed stably across most EEG channels . In other words , adults showed the most variability in the measured brain signals . A second aspect of signal variability is its temporal predictability , which can be measured using Multiscale Entropy [MSE 25] . MSE measures sample entropy [26] , [27] of the signal at successively downsampled time series , with a scale of 1 being the original time series and scale t indicating a time series created by averaging t adjacent points . MSE assigns low values to both highly deterministic and completely random signals , making it an explicit measure of signal complexity [28] . MSE estimation applied to single trial data for each channel showed that sample entropy measures were highest for adults across all temporal scales and lowest for the youngest children , with the intermediate age groups falling along an ordinal trend ( Fig 2b ) . Given that consistent age differences were observed at all time scales , the area under the MSE curve was computed for each subject to compare age-related differences . Multivariate statistical analysis with PLS confirmed a significant age-related increase in MSE that was expressed stably across most of the EEG channels ( Fig 2a ) . Taken together , the PCA and MSE measures indicate that , contrary to behavioral variability , brain variability increases with maturation . In some ways , this could be deduced from the group differences in the relative spectral density distribution , wherein one may expect signals that are dominated by low frequencies to show less variability than those with relatively stronger contributions from higher frequencies . For example , by increasing the relative magnitude of the Fourier coefficients for low frequencies in the adult EEG data , it is possible to get PCA dimensionality and MSE estimates similar to children ( Figure S2 ) . However , spectral density and variability are not completely interdependent , because jittering the phase of the Fourier coefficients , while maintaining their relative magnitude , has no impact on spectral density , but changes PCA dimensionality and MSE estimates ( Figure S3 ) . This is because PCA and MSE are sensitive to the dependencies within the signals that do not affect spectral density . Such sensitivities likely reflect transients in neural processing , and would be most evident in a system with enhanced capacity for signal processing and complexity . The final and most important part of this investigation was to relate behavioral variability , brain variability , and maturation . We addressed these issues using PLS to analyze the correlations between our measures of dynamical variability ( pre and post-PCA dimensionality and MSE ) and their phenotypic correlates ( RT-variability , accuracy , and age ) . We also include mean RT in the analysis to determine whether the correlation patterns we observed were specific to behavioral measures of variability , or to any metric showing a maturational change . Figure 3 shows the results of the analysis . Computed across all subjects , the correlation between behavioral variability ( cvRT ) and brain variability ( PCA dimensionality and MSE ) was negative and highly robust across most of the EEG channels ( Figure 3a ) . The correlation for accuracy was a mirror image of the pattern for cvRT , showing a positive correlation with PCA and MSE estimates ( Figure 3a ) . Mean RT , however , showed a much weaker , and statistically unreliable , correlation pattern with brain variability measures ( Figure 3a ) . Finally , the correlation of chronological age and brain variability was very strong and positive across most of the scalp . The impressions derived from the visual inspection of Figure 3a were confirmed by the PLS analysis ( Fig 3b ) . Measures of behavioral consistency ( cvRT and accuracy ) and chronological age showed stable correlations with the brain variability measured with PCA or MSE . Mean RT , however , did not . In other words , increased brain variability during maturation was associated with more stable and accurate behavior . The statement that the relation between behavioral variability and brain variability is mediated by maturation implies that if age differences were eliminated , the strong correlations seen in Figure 3 with cvRT and accuracy would be reduced . This turned out to be correct . We again used PLS to analyze the relation between the brain variability and behavior measures when chronological age was regressed out of both sets of measures . For cvRT and accuracy , the correlations for the full sample were reduced , and in the case of MSE , no longer statistically reliable ( Figure 4 ) . The effect was less dramatic when the analysis was performed on children only . Furthermore , the correlation between mean RT and brain variability remained nonsignificant after adjusting for chronological age for both the entire sample and the children . The age-adjustment process indicated that a large proportion of the relationship between behavioral consistency and brain variability was due to maturation , as measured by chronological age .
In contrast to behavioral variability , brain variability increases with maturation . With maturation comes differentiation and specialization of brain regions , but at the same time there is increased integration between distributed neuronal populations and establishment of new functional connections [29] . The change in balance between differentiation and specialization would produce more variability in on-going activity as the number of simultaneous processes possible at any given moment increases . Mature and integrated nervous systems generally have more prolonged and complicated neural transients [30] . Such transients are characteristic of a system with high neural complexity [31] . With the maturational increase in brain signal variability there is an increase in behavioral stability . Across the sample we studied , subjects with higher signal variability showed less variability in response latency ( measured with cvRT ) and greater performance accuracy . When the measures were adjusted for the chronological age , the relationship between brain and behavioral variability weakened , suggesting that a large part of the relationship represented a maturational effect . It is noteworthy that mean reaction time , which also showed a maturational change , did not significantly correlate with brain variability . It may be that other physiological factors are more important for the response speed change during maturation . By contrast , strong correlations with behavioral consistency indicates that cvRT and accuracy are likely tapping into aspects of the behavioral tuning which are more tightly related to the changes in brain complexity/variability . The present results may seem at odds with the intuitive notion of behavior and brain variability , where one would expect that they go hand in hand . However , the results do make sense when the nonlinear dynamics of the nervous system are considered . Internal variability may be vital to enable the brain to parse weak and ambiguous incoming signals [10] , [32] , [33] . Variability can facilitate the exchange signals between neurons [34] , [35] , transitions in metastable systems [7] , and the formation of functional networks [17] , [36] . As the nervous system matures , physiological variability increases , which is captured by increases in complexity [37] , [38] , and the system can better adapt to its environment . Maturational changes that have been reported in children's evoked potentials can also be related to increased brain signal complexity . Compared to adults , the average evoked responses in children tends to show higher amplitude and longer latency on early responses , and less well-defined later responses [20] , [21] , [39] . Spectral power distribution also changes with maturation , with a gradual reduction in low frequencies and an increase in higher frequencies [22] , [40] . The relative change in spectral power density presumably underlies the reduction in the latency of the evoked responses and its multicomponent nature , where the lower frequency bias in children would yield slow and broad evoked potentials . The emergence of higher frequencies with maturation would both decrease the evoked response latency and allow additional deflections to emerge ( i . e . , N100 , P200 , etc . ) . Both the spectral power and evoked response changes would be expected given the maturational increase in complexity . The emergence of higher frequencies would reflect the enhanced local processing ( segregation ) , whereas the multicomponent evoked response is thought to reflect reentrant interactions [41] , suggesting enhanced integration . In the age range of the children we studied the brain is in a state of structural and functional refinement [42] , [43] . Myelination and neural pruning increase differentiation of information flow in the brain , enabling a shift from a system that responds in a slow and stimulus-locked manner , to one that responds more rapidly and where the internal variability reflects the parallel exploration of the functional repertoire before converging to an optimal response [16] , [44] . In the case of normal development , the increased variability leads to a stabilization of behavior , increasing the cognitive repertoire of the system . One may postulate that internal variability would mature to some optimum level , based on both physiology and experience , but that further increases or decreases , coming from disease or damage would compromise behavioral stability . The suggestions derived from the present findings contribute to the growing evidence that internal dynamics are a key feature governing brain function [45]–[47] .
EEG recordings were collected from 24 adults and 55 children for a total of 79 subjects . Adults ( 18 females ) ranged from 20 to 33 years of age . Children were divided into four age groups as follows: 8–9 years ( n = 11 , 3 females ) , 10–11 years ( n = 16 , 8 females ) , 12–13 years ( n = 15 , 8 females ) , and 14–15 years ( n = 13 , 6 females ) . Adult subjects and children , along with their parents , signed informed written consent . All subjects were healthy with no known cognitive or neurological disorders and had normal or corrected-to-normal vision . All children successfully completed two sub-tests of the WISC III ( vocabulary and block design ) . The experimental procedure was approved by the French Comite Operationnel pour l'Ethique dans les Sciences de la Vie du CNRS . Continuous EEG was recorded ( NeuroScan 4 . 1 ) on an EasyCap ( 10/10 system ) containing 32 electrodes and Cz as reference , sampling rate of 500 Hz , a band-pass 0 . 1–100 Hz , and a gain of 500 ( SynAmps ) . Subjects performed a rapid face recognition task . Each trial started with a presentation of a novel or familiar face for 500ms and subjects responded by pressing either a target or a non-target button depending on whether they recognized the face . Detailed description of the stimuli and the task are given in Itier & Taylor [19] . Infraorbital electrodes for measuring eye movements were removed and an average reference was computed . The final number of electrodes was 31 . Continuous EEG recordings were lowpass filtered at 40 Hz . Data were epoched and baselined into [−200 1200] ms epochs with a [−200 0] ms pre-stimulus baseline . Preliminary artifact removal was performed using independent component analysis ( ICA ) as implemented in EEGLAB software [48] . Trials contaminated with excessive amplitudes were removed first , then ICA decomposition was performed on the remaining concatenated trials and components carrying ocular and muscle artifacts were subtracted . The number of kept trials per subject was between 236 and 761 , with an average of 529 . For the signal variability estimation , it was important to have equal amounts of artifact-free data across subjects . We therefore introduced an additional trial selection step based on the total global field power ( gfp ) , calculated as a sum of squared amplitudes across all electrodes and all time points for the trial duration . The 100 trials closest to the median for each subject were selected for further variability analysis . This selection criterion minimized potential presence of trials contaminated with high residual amplitude artifacts . This was particularly important for the convergence of multiscale entropy ( MSE ) algorithm ( see below ) . It is critical to note that , with the exception of MSE , the results were similar when all trials that passed initial screening were analyzed . For each subject we calculated two response time related measures: mean response time ( mean RT ) and coefficient of variation of the response time ( cvRT ) . The coefficient of variation of RT was calculated as the standard deviation divided by the mean RT within subject , and was taken as a measure of subject's behavioral variability . The scaling procedure in cvRT minimizes differences between groups that arise from differences in mean and standard deviations . Mean and cvRT were based on thresholded RTs ( <1200 ms ) and included both correct and incorrect responses . Exclusion of excessively long RTs as outliers enabled robust estimation of mean RT and cvRT . Because the trial selection focused on EEG signal regardless of the response , it is unlikely that such selection would have introduced any bias pertaining to the RT-related measures . Our results for mean RT are somewhat different than those presented in Itier & Taylor [19] , where subject's mean calculation included only correct-response trials and employed no outlier thresholding . As a third behavioural measure , we used subject accuracy ( percent correct responses ) , calculated from all recorded trials . Spectral power distribution of the baseline signal across single trials was calculated using Fast Fourier Transform ( FFT ) . Considering the known age-related differences in global signal power , the signal was first normalized ( mean = 0 , standard deviation = 1 ) in order to calculate relative contributions of different frequency bands to the total spectral power . With 500 Hz sampling rate and 200 ms baseline signal this gave us 100 time points and 5 Hz frequency resolution . For each subject , PCA was used as an estimate for the dimensionality of the single trial EEG space . Subject's data was divided into channel specific matrices of single trial data with trials as rows and time points as columns . The dimensionality of each matrix was determined as a minimum number of principal components capturing 90% of the variance across trials . This number was further expressed as a percent of the total number of trials and was taken as a measure of trial-to-trial variability for a given channel . Two time intervals were analyzed: [−200 0]ms pre-stimulus and [0 200]ms of the post-stimulus signal . In this way , we obtained dimensionality estimates of the trial space for pre- and post-stimulus signals across the scalp . For example , average pre-stimulus PCA dimensionality estimate for electrode O2 within the adult group was 10 . 8 . Given 100 trials , and hence 100 dimensions , this result means that for adults , 90% of the trial space signal variance occupies 10 . 8 dimensions . MSE was used to estimate entropy at different time scales . Full details of the MSE measure and its relevance for the analysis of signal complexity are given in Costa et al . , [49] and Costa et al . , [28] . We first calculated single trial MSE using the algorithm available at www . physionet . org/physiotools/mse/ with parameter values m = 2 , r = 0 . 5 . The algorithm calculates sample entropy as a measure of regularity ( predictability ) of the signal at different scales . It consists of two procedures: 1 ) coarse-graining of the time series and 2 ) calculating sample entropy for each coarse-grained time series . For scale t , the coarse-grained time series is constructed by averaging the data points within non-overlapping windows of length t . This procedure can be viewed as a smoother version of decimation . Sample entropy of each coarse-grained time series measures its regularity by evaluating the appearance of repetitive patterns . The length of single trial time series was 700 time points corresponding to [−200 1200]ms epoch at 500 Hz sampling rate . For each subject , a channel specific MSE estimate was obtained as a mean across single trial entropy measures for scales 1–14 . Entropy measures for scales >14 were not calculated because the corresponding coarse-grained ( downsampled ) time series were too short ( <50 time points ) for reliable sample entropy estimation . Statistical assessment of maturational trends in MSE and PCA was performed using partial least squares ( PLS ) for EEG data [24] . PLS was performed on data matrices consisting of subject and channel specific measures such that rows represented subjects within age groups . The columns of the data matrix were either the integrated measures for MSE or the PCA dimensionality estimation by channel . PLS data matrices were averaged within group and grand mean centered across all five age groups . The mean-centered matrices were then decomposed with singular value decomposition ( SVD ) to identify the strongest group differences and the corresponding scalp topography . For brain-behavior analyses , correlations were computed between each behavior measure and either the PCA or MSE measures across the entire sample . The four correlation “maps” ( one each for cvRT , accuracy , mean RT , and chronological age ) were then decomposed with SVD . The statistical significance of the effects was assessed using permutation tests for the overall relationship between either age group and brain variability or brain and behavior . The reliability of the topographies was determined with bootstrap estimation of confidence intervals , using 500 bootstrap samples . For scalp topographies , the singular vector weights for each channel were divided by the bootstrap estimated standard error , giving a bootstrap ratio . This is similar to a z-score if the distribution of singular vector weights is Gaussian . Details of these statistical tests are described in [24] , [50] , [51] . At the univariate level , correlations involving chronological age , cvRT , accuracy , and mean RT ( on the behavior side ) were assessed for statistical reliability . MSE and pre- and post-stimulus PCA ( on the brain side ) were also assessed . The stability of the correlations was estimated across subjects using a bootstrapping procedure . This allowed us to calculate confidence intervals ( CI ) around the correlation levels . For each brain measure we generated 1000 random samples of subjects with replacement and calculated the corresponding correlation with a behavioral measure . The lower and upper 95th percentiles across bootstrap samples were derived giving us the 95% confidence interval . Correlations were considered reliable if the CI did not include zero . For PCA trial space dimensionality estimation , the correlations across all 79 subjects were calculated for each channel . For MSE , we used area under the MSE curve ( sum of entropy values across all scales ) and calculated correlations with a behavioral measure for each channel . It should be emphasized that the inferential tests for the significance of these correlations were assessed with multivariate PLS . The channel-wise correlation analyses are an assessment of the reliability of the correlation patterns and are complementary to the multivariate PLS analysis .
|
Intuitive notions of brain–behavior relationships would suggest that because children show more variability in behavior , their brains should also be more variable . We demonstrate that this is not the case . In measuring brain signal variability with EEG and behavior in a simple face recognition task , we found that brain signal variability increases in children from 8–15 y and is even higher in young adults . Importantly , we show that this increased brain variability correlates with reduced behavioral variability and more accurate performance . A brain that has more variability also has greater complexity and a greater capacity for information processing . The implication of our findings is that variability in brain signals , or what some would call noise , is actually a critical feature of brain function . For the brain to operate at an optimal level , a certain amount of internal noise is necessary . In a certain way it could be stated that a noisy brain is a healthy brain .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"neuroscience/neurodevelopment",
"neuroscience/theoretical",
"neuroscience"
] |
2008
|
Increased Brain Signal Variability Accompanies Lower Behavioral Variability in Development
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.