Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
sequencelengths 1
91
| keywords
sequencelengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
In the current study , a comprehensive , data driven , mathematical model for cholera transmission in Haiti is presented . Along with the inclusion of short cycle human-to-human transmission and long cycle human-to-environment and environment-to-human transmission , this novel dynamic model incorporates both the reported cholera incidence and remote sensing data from the Ouest Department of Haiti between 2010 to 2014 . The model has separate compartments for infectious individuals that include different levels of infectivity to reflect the distribution of symptomatic and asymptomatic cases in the population . The environmental compartment , which serves as a source of exposure to toxigenic V . cholerae , is also modeled separately based on the biology of causative bacterium , the shedding of V . cholerae O1 by humans into the environment , as well as the effects of precipitation and water temperature on the concentration and survival of V . cholerae in aquatic reservoirs . Although the number of reported cholera cases has declined compared to the initial outbreak in 2010 , the increase in the number of susceptible population members and the presence of toxigenic V . cholerae in the environment estimated by the model indicate that without further improvements to drinking water and sanitation infrastructures , intermittent cholera outbreaks are likely to continue in Haiti .
After a massive earthquake struck the island nation of Haiti in 2010 , the introduction of an altered El Tor biotype of Vibrio cholerae O1 has led to one of the largest cholera outbreaks in recent history [1] [2] [3] . Almost four years after the identification of the first cholera cases , the transmission appears to have temporarily slowed , however the future of the cholera epidemic in Haiti remains uncertain [4] . After the initial isolation of toxigenic V . cholerae O1 from surface water monitoring sites in the Ouest Department of Haiti in 2012 and 2013 , there is evidence that the frequency of isolation from the environment has actually increased between 2013 and 2014 [5] [6] . In the absence of ongoing transmission , the presence of toxigenic V . cholerae O1 in the aquatic environment has left the international scientific community divided on the possibility that the causative bacterium has established environmental reservoirs in the surface waters of Haiti [7] [8] [9] . If this were to be the case , the goal of cholera elimination from the island of Hispaniola by 2022 would be more challenging , with the potential for cholera to become endemic in Haiti [10] . To assist in the planning and allocation of resources necessary to mitigate the outbreak , mathematical models have been developed to investigate the underlying dynamics of cholera transmission in Haiti . [11] . However , despite empirical evidence that V . cholerae O1 is increasingly present in the surface water as reported cases continue to decline , none of the previous models have considered the role of environmental reservoirs in cholera transmission [6] . Though the environmental compartment has been included in the models , it is assumed that V . cholerae O1 occupy a transient state where after being shed from the human host they will eventually become removed from the environment at a constant rate of decay [12] . However , in endemic countries , this assumption is often likely to be false; where V . cholerae O1 is able to persist and multiply in the environment in response to an influx of nutrients into surface waters after rainfall events or increases in water temperature leading to recurrent outbreaks after interepidemic periods where very few cases were reported [13] . Since both water temperature and rainfall have been associated with increased isolation frequency of toxigenic V . cholerae O1 in Haiti [6] , a dynamic cholera transmission model was created with the additional mechanism by which the environmental compartment responds to factors such as precipitation and surface water temperature that increase the concentration of the organism in the aquatic environment . Hopefully , these extra parameters will assist in the understanding of the underlying processes of cholera transmission in Haiti and allow for more accurate prediction of the potential for future outbreaks .
To reflect the basic differences in the modes of transmission , the model incorporates both the short cycle transmission from human-to-human and long cycle transmission from human-to-environment and environment-to-human . The short route relies on data suggesting that toxigenic V . cholerae assumes a short-lived hyperinfectious state immediately after passage from the human intestine [14] . This facilitates rapid transmission of V . cholerae from one person to another , often related to personal hygiene practices within the household . Alternatively , transmission may occur when V . cholerae is acquired from contaminated drinking water or by contact with the aquatic environment . The presence of toxigenic V . cholerae in the aquatic environment may reflect contamination of water sources by feces from an infected individual , and/or the existence of an aquatic reservoir in which the microorganism can persist for months to years [13] . Transmission through this aquatic route , while still having the potential for being relatively rapid , tends to involve more time than the short cycle transmission between humans . In the model , separate compartments for infectious symptomatic and infectious asymptomatic cases are used , even though it is not possible to estimate the size of the asymptomatic compartment . This is done to increase the model flexibility and to provide the option for sensitivity analysis . Dichotomization between symptomatic and asymptomatic cases also provides the option to address different infectivity levels for symptomatic and asymptomatic infections . The model has the following compartments: S ( t ) —number of susceptible people at time t . A ( t ) —number of asymptomatic people at time t . I ( t ) —number of symptomatic people at time t . R ( t ) —number of recovered people at time t . W ( t ) —bacteria concentration in the water at time t ( environmental compartment . ) The model diagram and the relationships between the model compartments and the observed data are summarized visually in the diagram provided in Fig 1 . In the model the movement of people between the compartments S , A , I , R is considered along with the growth and death of bacteria within the environmental compartment W . The system of ordinary differential equations ( ODE ) that defines our model has the form: d S ( t ) d t = μ R S R ( t ) - ( μ S A W + μ S I W ) S ( t ) f ( t ) - ( μ S A H + μ S I H ) S ( t ) ( A ( t ) + I ( t ) ) d A ( t ) d t = μ S A W S ( t ) f ( t ) + μ S A H S ( t ) ( A ( t ) + I ( t ) ) - μ A R A ( t ) d I ( t ) d t = μ S I W S ( t ) f ( t ) + μ S I H S ( t ) ( A ( t ) + I ( t ) ) - μ I R I ( t ) d R ( t ) d t = μ A R A ( t ) + μ I R I ( t ) - μ R S R ( t ) d W ( t ) d t = g ( t ) ( μ A W A ( t ) + μ I W I ( t ) ) + h ( t ) m ( t ) W ( t ) - γ W - ( t ) W ( t ) ( 1 ) In the model equations μ and γ indicate the transition rates with corresponding subscripts and superscripts that indicate the direction and the nature of the movement . The superscript H indicates the rates responsible for human-to-human transmission and superscript W indicates the rates responsible for environment-to-human transmission . To address the dynamic of the environmental compartment three main process that affect bacterial growth and survival in the environment were considered . The first process is the influx of bacteria via shedding by infected human hosts into the environment . Once shed into the environment the bacteria provide a source of exposure for susceptible humans . Those processes are modeled by the functions: f ( t ) = W ( t ) κ + W ( t ) and g ( t ) = ρ ( t ) δ + ρ ( t ) . The notations ρ ( t ) for the total weekly precipitation in mm and τ ( t ) for average weekly temperature in degrees Celsius at time t are used . Here κ and δ are the threshold parameters . The second process is the multiplication of the bacteria in the environment , which is affected by both temperature and precipitation . This process is modeled by functions h ( t ) and m ( t ) : h ( t ) = α exp [ - ( ρ ( t ) - ρ c ) 2 2 σ 2 ] + β τ ( t ) and m ( t ) = 1 - W ( t ) χ = χ - W ( t ) χ . Here α , ρc , σ and β are the parameters of interest and χ is the cap designed to constrain the excessive growth of bacteria in the environment . The functional form m ( t ) represents the logistic growth multiplier widely used in population dynamic models . This multiplier allows the growth to be proportional to the current bacterial concentration W ( t ) and limits the excessive growth when concentration approaches the limiting capacity using the cap parameter χ . The proposed multiplier h ( t ) has a novel structure . In the model it is assumed that bacterial growth is linearly related to the current temperature which is controlled by parameter β . Precipitation is assumed to have a maximum effect on the bacterial growth at the value ρc . It is assumed that for smaller amounts of precipitation than ρc there is not enough water to wash the bacteria into the environment which causes slower growth . For amounts of precipitation above ρc bacteria becomes diluted which diminishes the rate of bacterial growth in the environment . Graphically , the function h ( t ) has a bell-shaped curve where α and σ2 are the calibration parameters of inferential interest . The last process in the environmental reservoir is the natural decay ( death ) of bacteria in the environment which is modeled by the time-varying death rate γW− ( t ) . Please refer to the supplement S1 Text for more technical details on model formulation and assumptions . Overall , the model defined by Eq ( 1 ) is neither identifiable ( i . e . , there are too many unknown parameters ) nor estimable without extra assumptions [15] . Only precipitation , temperature and the symptomatic compartment I ( if underreporting is accounted for ) can be treated as observed . To summarize , a Susceptible-Infected-Recovered-Susceptible ( SIRS ) model has been implemented , where the V . cholerae concentration in the water is modeled via the environmental compartment W . In the model the SIRS piece is linked to the reported incidence via the symptomatic compartment I using the reporting probability pr . The reported incidence was adjusted before estimation by dividing it by the assumed reporting probability pr . To avoid identifiability issues , extra assumptions about the model parameters and the model itself are made . Since the period of time under consideration was very short , the population size was considered to be constant . Please refer to the supplement S2 Text for more technical details about the model parametrization . To account for the uncertainty in the deterministic model defined by the ordinary differential Eq ( 1 ) stochastic Gaussian terms were introduced into the model equations . The stochastic model was fitted to the reported incidence by using the least squares estimation ( LSE ) approach . Please refer to the supplement S3 Text for details on stochastic model fitting . Data were collected from multiple sources . The reported cholera incidence for the Ouest Department of Haiti , including the capital Port-au-Prince , was collected by the Haitian Ministry of Health ( Ministère de la Santé Publique et de la Population ( MSPP ) in French ) and compiled by the Pan American Health Organization ( PAHO ) [16] [4] . The weekly incidence of cholera cases was available from October 17 , 2010 until April 27 , 2014 . Daily precipitation ( in millimeters ) was obtained from the Tropical Rainfall Measuring Mission ( TRMM ) satellite data [17] , and daily temperatures ( in Celsius ) were obtained from the Port-au-Prince airport ( IATA: PAP ) monitoring station . The temperature readings were missing for 14 . 6% of the dates and the missing values were linearly interpolated . Precipitation data did not have any missingness . The environmental data were aggregated weekly so that it could be aligned with the incidence data . The average weekly temperature τ ( t ) and cumulative weekly precipitation ρ ( t ) were used as covariates . In the analysis , it was assumed that there was no time lag for temperature , whereas there was a 7-week lag for precipitation when we evaluated the environmental effects on the water compartment in the model . We did not observe any lag for the temperature from the data . Temperature had only a mild correlation with reported cholera incidence , and the empirical evidence suggested that the association between water temperature and isolations of toxigenic V . cholerae from the environment was the strongest with a time lag of 0 to 1 month [6] . At the same time a seven-week lag maximizes the sample correlation between total weekly precipitation and weekly reported cholera incidence . Moreover , there is empirical evidence that the bacteria concentration peaks in the environments three to four weeks after the rainfall , which is associated with an increase in the incidence approximately four weeks later [6] . Thus , a seven week time lag for precipitation was considered plausible . A visual presentation of aligned time series of incidence , temperature and seven-week-lagged precipitation is shown in Fig 2 . The transmissibility of a pathogen in a susceptible population is often measured using the basic reproductive number . Unfortunately , because of the complexity of the model , time-dependent covariates , and the multiple types of sources of infection ( humans and the aquatic environment ) , there was no straightforward epidemiological interpretation of R 0 for this model . Moreover , in this model R 0 was technically time dependent because of the time dependent environmental covariates and phage dependent bacterial death rate . The details on the computation of the the basic reproductive number are provided in the supplement S4 Text .
The obtained model fit provided a good understanding of the dynamics of the epidemic over time . The visual summary of the model fit together with the adjusted reported cholera incidence is shown in Fig 3 . First , the reported incidence was adjusted by rescaling to account for disease underreporting and plotted in orange in Fig 3 for better visual comparison with the model output . To produce the model realizations a different Gaussian white noise time series was generated for each set of 1000 parameter estimates obtained from the previous LSE fits . The corresponding model outputs are displayed in Fig 3 . The transparency was tuned to display the density of the curves in each part of the graph and to improve visualization . The symptomatic cases produced by the model are displayed in dark green in panel A of Fig 3 . The realizations of both the symptomatic ( dark green ) and total ( light green ) cases produced by the model are displayed for comparison in panel B of Fig 3 on a different scale . The total underlying realizations of the model that include both symptomatic and asymptomatic infections are much larger than the symptomatic realizations alone . The peak precipitation effect was estimated at ρ ^ c = 45 . 1 mm with 95% CI ( 43 . 0; 47 . 3 ) and the threshold parameter for the effect of shedding at δ ^ = 27 . 0 mm with 95% CI ( 6 . 7; 101 . 3 ) , which was estimated to be more variable than ρ ^ c . Those estimates did not change much from the starting points that were used for the iterative LSE minimization procedure , indicating the potential lack of information in the data about ρc and σ . The estimate for the effect of temperature had a median value much higher than the mean value , which indicated a heavy left tail . The median was used over 1000 realizations instead of the mean β to provide a more robust estimate . The estimate for β ^ was 0 . 014 with 95% CI ( −0 . 041; 0 . 027 ) , which led to the conclusion that temperature had a mild association with V . cholerae growth . The complete list of parameters is provided in Table A in S3 Text . If a single estimate for R 0 that summarizes the epidemic behavior is desired , a reasonable approach is to use the averaged values of the time dependent covariates and bacterial death rate to obtain the average estimate R ^ 0 = 1 . 6 with 95% CI ( 1 . 3 , 2 . 1 ) based on the mean of 1000 stochastic realizations . Alternatively , one may extend the definition of the basic reproductive number to allow for time-dependent covariates and denote it by R 0 ( t ) . Readers please refer to the supplement S4 Text for details . Another useful measure is the time-dependent effective reproductive number R ( t ) which is defined as a product of the basic reproductive number R 0 ( t ) and the proportion of susceptibles at a given time t . The change in the value of the estimated basic reproductive number R ^ 0 ( t ) ( using the extended definition ) and the estimated effective reproductive number R ^ ( t ) over time are shown in panel B of Fig 4 . Additional characteristics of the epidemic are illustrated in Fig 4 . In panel A the reported incidence adjusted for underreporting and the pointwise prediction bands for symptomatic infections are displayed . As shown the pointwise prediction bands were able to capture the majority the dynamics of the cholera epidemic . In panel C the concentration of V . cholerae in the environment and the corresponding prediction bands are shown over time . In panel D the changes in the proportion of the susceptible individual during the course of the epidemic produced by the model and corresponding prediction bands are shown . Based on the trend produced by the model and the observed incidence displayed in panel A , it was concluded that the epidemic of cholera likely stabilized in Ouest Department of Haiti after three years of transmission and became endemic . In the model output displayed in panel D the proportion of susceptible individuals at the end of the epidemic remains very high and is gradually increasing , which provides the necessary conditions to facilitate further cholera transmission . Furthermore , as displayed in panel C , since the concentration of toxigenic V . cholerae in the environment produced by the model remains sufficiently high at the end of the observation period , it is also likely that future cholera outbreaks will occur .
In this work , a dynamic model that incorporated the available environmental data was used to describe the transmission of cholera in Ouest Department of Haiti . The model output suggested the existence of a large environmental reservoir of toxigenic V . cholerae that reached a peak concentration early in 2012 , with a subsequent slow decline ( Fig 4 ) . The presence of such an environmental reservoir was consistent with environmental studies conducted in the Leogane flood basin of the Ouest Department , which identified V . cholerae O1 in multiple river and estuarine ecosystems [5] [6] . A similar trend was observed in the human susceptible compartment of the model , where the smallest number of susceptible population members was observed in early 2012 , with a slow but steady increase since that time ( Fig 4 ) . The model of the cholera epidemic in Haiti described by this study was novel in the way in which the environmental compartment was considered . As previously mentioned , most previous dynamic models of the cholera epidemic in Haiti postulated that toxigenic V . cholerae only occupy a transient state in the environment , where pathogenic bacteria shed into the surface water by humans decay at a constant rate and cannot increase without additional cases . This assumption precludes the ability for toxigenic V . cholerae to become more prevalent in the environment during periods of decreased cholera incidence and does not explain the resurgence of cholera cases after inter-epidemic periods; both of which have recently been observed in Haiti [4] [6] . As with any mathematical model of infectious disease transmission , this approach was not without limitations . One important theoretical concern was the assumption of homogenous mixing . The contact transition rates between compartments assume homogenous mixing and do not account for the local population density , presence of human mobility networks , and personal hygiene practices within households [18] . Likewise , the contact rates between humans and the environment are also dependent on the proportion of the population that consume contaminated surface water , which varies between urban and rural areas and by demographic factors [19] [20] . Besides the reliance of our model on previously published estimates of some parameters , there are also unobserved processes that occurred during the epidemic , such as increased consumption of bottled water in urban areas of up to 38% and the fluctuation in the number of cholera treatment centers ( CTC ) as the incidence began to decline [21] [22] . However , the demographic data as well as the number of interventions applied from the international network of aid organizations are also difficult to quantify , making their inclusion in the model speculative at best . Thus far , only a single serological study of cholera in Haiti was conducted in high-risk populations near the Artibonite River six months after the onset of the epidemic , which reported that 39% of the participants had antibody titers consistent with a recent cholera infection [23] . Our model , which used incidence data from the neighboring Ouest Department , where the onset of the epidemic occurred later , showed a projected population proportion of susceptibles that was somewhat higher at that time . Aside from the one study cited , there have been no further serologic studies reported in Haiti , so it is not possible to comment directly on the validity of the model’s projections . Nonetheless , a rising proportion of susceptibles is plausible , given the anticipated waning of immunity to El Tor cholera over time , and a birth rate that is over 40% higher than other developing countries in Latin America and the Caribbean [24] [25] . The combination of environmental reservoirs of toxigenic V . cholerae , lack of adequate sanitation and hygiene infrastructure , and a slowly rising proportion of susceptible population members suggests that seasonal epidemics are likely to be observed in the future . Furthermore , there remains the possibility of major cholera epidemics following hurricanes that generate severe flooding or other environmental disasters that could damage the existing sanitation and drinking water infrastructure . Given the potential for future cholera outbreaks and the demonstrated efficacy of the oral cholera vaccine in Haiti , it would be useful to have epidemic mitigation plans in place that include provisions for the use of the WHO mobile stockpile of cholera vaccine [26] [27] . | Based on the model-fitted trend and the observed incidence , there is evidence that after an initial period of intense transmission , the cholera epidemic in Haiti stabilized during the third year of the outbreak and became endemic . The model estimates indicate that the proportion of the population susceptible to infection is increasing and that the presence of toxigenic V . cholerae in the environment remains a potential source of new infections . Given the lack of adequate improvements to drinking water and sanitation infrastructure , these conditions could facilitate ongoing , seasonal cholera epidemics in Haiti . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Cholera Transmission in Ouest Department of Haiti: Dynamic Modeling and the Future of the Epidemic |
The Rhox cluster on the mouse X chromosome contains reproduction-related homeobox genes expressed in a sexually dimorphic manner . We report that two members of the Rhox cluster , Rhox6 and 9 , are regulated by de-methylation of histone H3 at lysine 27 by KDM6A , a histone demethylase with female-biased expression . Consistent with other homeobox genes , Rhox6 and 9 are in bivalent domains prior to embryonic stem cell differentiation and thus poised for activation . In female mouse ES cells , KDM6A is specifically recruited to Rhox6 and 9 for gene activation , a process inhibited by Kdm6a knockdown in a dose-dependent manner . In contrast , KDM6A occupancy at Rhox6 and 9 is low in male ES cells and knockdown has no effect on expression . In mouse ovary where Rhox6 and 9 remain highly expressed , KDM6A occupancy strongly correlates with expression . Our study implicates Kdm6a , a gene that escapes X inactivation , in the regulation of genes important in reproduction , suggesting that KDM6A may play a role in the etiology of developmental and reproduction-related effects of X chromosome anomalies .
Homeobox ( HOX ) genes are known for their ability to regulate embryogenesis and guide tissue differentiation . These genes encode transcription factors that specify cell identity and regulate many embryonic programs including axis formation , limb development , and organogenesis [1] . Control of HOX gene expression via epigenetic modifications that include DNA methylation and histone modifications is critical to this process . Notably , tri-methylation of lysine residue 27 of histone H3 ( H3K27me3 ) plays a major role in repression of HOX genes [2] . The histone demethylase KDM6A ( also known as UTX ) removes H3K27me3 from HOX genes to restore their activity [3] . KDM6A contains a tetratricopeptide motif predicted to mediate protein-protein interactions [4] , and is a member of a stable multi-protein complex that not only de-methylates H3K27me3 but also methylates lysine 4 at histone H3 to facilitate gene expression [5] , [6] . Different protein partners modulate KDM6A recruitment to specific chromatin regions since ectopic KDM6A expression does not result in significant reduction of genome-wide H3K27me3 levels but rather targets specific genes [3] , [7] , [8] , [9] . For example , KDM6A regulates muscle-specific genes during myogenesis and is necessary for proper cardiac cell differentiation [10] , [11] . KDM6A mutations have been discovered in patients with Kabuki syndrome , a rare syndrome associated with distinct facial features , intellectual disability , growth retardation , and skeletal anomalies [12] , [13] . Recent studies have also implicated KDM6A as a candidate tumor suppressor gene whereby ectopic expression leads to enhanced expression of the RB ( retinoblastoma ) and RBL2 ( retinoblastoma-like 2 ) genes [14] . KDM6A inactivating mutations have been discovered in acute promyelocytic leukemia and multiple other cancer types [15] , [16] , [17] . A large set of homeobox genes clustered on the X chromosome has been implicated in male and female reproduction . In mouse , this cluster called Rhox ( reproductive homeobox X-linked ) contains 33 adjacent genes organized into three sub-clusters: α , β , and γ [18] . The Rhox cluster evolved at a rapid pace in mammals: the rat cluster contains 11 genes , and the human cluster , only 3 genes . In mouse , members of each paralog family have nearly identical sequences and are thus considered to be functional , although few members have been studied in detail [18] . Rhox genes are selectively expressed in male and female reproductive tissues , including testis , ovary , and placenta [19] . Similar to other homeobox genes , Rhox genes are also expressed during early embryonic development [19] , [20] , [21] , [22] , [23] . Little is known about the biological significance of individual paralogs . Epigenetic regulation of the Rhox gene cluster has been mainly focused on DNA methylation and histone H1 control in placenta and during embryonic development [24] , [25] . It is unknown whether other histone modifications control Rhox expression and what histone modifiers might be responsible . An important contender is KDM6A , which is known to regulate the HOX cluster [3] . Interestingly , KDM6A is encoded by an X-linked gene that escapes X inactivation in somatic tissues of human and mouse [26] , [27] , [28] . Expression of most X-linked genes in somatic tissues is equalized between males ( XY ) and females ( XX ) by random silencing of one X chromosome in early development [29] . Genes that escape X inactivation represent exceptional genes with higher expression in females versus males , suggesting that they may be important for female-specific functions [29] , [30] , [31] . To explore the potential role of KDM6A in the sex-specific regulation of Rhox genes , chromatin analyses were done to follow KDM6A recruitment to the Rhox cluster in male and female ES cells . We focused on Rhox6 and 9 , two members of the Rhox cluster we discovered to be most affected by Kdm6a knockdown . KDM6A was specifically recruited to Rhox6 and 9 in female but not male ES cells , resulting in removal of the repressive histone mark H3K27me3 and in increased expression . KDM6A was also bound to Rhox6 and 9 in ovary where these genes are highly expressed . We conclude that KDM6A is important for removal of a repressive histone mark at the bivalent promoters of Rhox6 and 9 to facilitate their expression in female ES cells and in ovary .
Rhox6 and 9 expression levels were significantly higher in undifferentiated female versus male ES cells ( Figure 1A ) . Expression was measured using quantitative RT-PCR ( qRT-PCR ) in two female ( PGK12 . 1 and E8 ) and two male ( WD44 and E14 ) ES cell lines before and after differentiation . ES cell differentiation and embryoid body formation were induced by removal of LIF ( leukemia inhibitory factor ) . Rhox6 and 9 primers were verified to be gene-specific by cDNA sequencing ( Figure S1 ) . Sex-specific differences persisted to day 2 of ES cell differentiation ( Figure 1B ) . Analyses of sexed 8-cell pre-implantation embryos confirmed higher female than male expression of Rhox6 and 9 in early development in vivo ( Figure S2A ) . Furthermore , re-analyses of published microarray expression data [32] revealed a female bias in Rhox6 and 9 expression at later embryonic stages ( 11 . 5–13 . 5 dpc ) in both germ cells ( at all stages ) and somatic cells ( at 12 . 5–13 . 5 dpc ) ( Figure 1C ) . Note that expression was much higher in germ cells compared to somatic cells . A female bias in Rhox6 and 9 expression in ES cells was unexpected because these genes are solely expressed from the maternal allele due to paternal imprinting [25] . Thus , the significantly higher expression we observed in undifferentiated female versus male ES cells ( >6-fold for Rhox6 and >10-fold for Rhox9 , respectively ) must be due to another factor ( Figure 1A ) . Interestingly , levels of the histone demethylase KDM6A known to play a role in HOX gene regulation were approximately two-fold higher in female compared to male ES cells as measured by qRT-PCR and western blot analyses ( Figure 1D , 1E ) . This sex bias initially due to the presence of two active X chromosomes in undifferentiated female ES cells [33] , [34] persisted throughout differentiation and at later embryonic stages , as expected for a gene that escapes X inactivation ( Figure 1F , 1G ) . To determine whether KDM6A was involved in the sex-specific regulation of Rhox6 and 9 we measured occupancy using chromatin immunoprecipitation ( ChIP ) in male and female ES cells . KDM6A occupancy at the 5′ end of Rhox6 and 9 was greater in undifferentiated female than male ES cells as measured both by quantitative PCR ( ChIP-qPCR ) and by array analysis ( ChIP-chip ) ( Figure 2A and Figure S2B ) . At day 2 of differentiation the female bias in KDM6A occupancy persisted , but following differentiation ( day 15 ) KDM6A occupancy decreased ( Figure 2C and S2B ) . These results are in agreement with the observed timing of changes in Rhox6 and 9 expression ( Figure 1B ) . Furthermore , we observed corresponding changes in levels of H3K27me3 , the histone modification removed by KDM6A . By ChIP-qPCR H3K27me3 enrichment mirrored changes in KDM6A occupancy at Rhox6 and 9 in female PGK12 . 1 ES cells during differentiation ( Figure 2D ) . Quantitative analysis of H3K4me3 enrichment at Rhox6 and Rhox9 promoters revealed higher enrichment in female than male ES cells , as well as a decrease during differentiation correlating with expression changes ( Figure 2B , Figure 1A and 1B ) . During differentiation X inactivation initiates in female PGK12 . 1 ES cells , as confirmed by increased Xist expression and by the appearance of an Xist cloud detected by RNA-FISH in interphase nuclei of cells at day 15 ( Figure S3 ) [35] . Concomitantly , H3K27me3 enrichment at Rhox6 and 9 increased almost 2–3-fold between day 0–2 and day 15 ( Figure 2D ) . This increase was observed over the entire Rhox cluster , suggesting that the cluster is subject to silencing possibly by X inactivation ( Figure S4 ) ( see below ) [29] . In male ES cells , H3K27me3 levels were very low at Rhox6 and 9 at all time points ( Figure 2D ) . Taken together , our data indicate that KDM6A is specifically recruited to Rhox6 and 9 in undifferentiated female ES cells , which results in a 6–10-fold higher expression compared to male ES cells . To directly assess the role of KDM6A in regulation of the Rhox cluster we performed knockdowns in two female and two male ES cell lines by RNAi . Using a pool of siRNAs to target multiple regions of Kdm6a RNA , we achieved a 60–80% knockdown in ES cells as shown by qRT-PCR and expression array analyses ( Figure 3A ) . Immunoblots using two different antibodies confirmed a dramatic reduction ( 70–90% ) in the amount of KDM6A protein after 48 h of knockdown ( Figure 3A ) . Specificity of the siRNAs was confirmed using two individual siRNAs , each resulting in a ∼60% knockdown ( Figure S5A ) . Kdm6a knockdown caused a significant reduction in Rhox6 and 9 expression in the two female ( PGK12 . 1 and E8 ) but not in the male ( WD44 and E14 ) ES cell lines , indicating that the regulation of these genes by KDM6A is female-specific ( Figure 3B ) . Rhox6 and 9 expression levels measured by qRT-PCR and by expression array analyses were diminished by 30–50% after Kdm6a knockdown whereas the control gene β-actin did not change ( Figure 3A , 3B ) . By expression array analyses we found that among the Rhox genes , Rhox6 and 9 exhibited the highest expression decrease ( >1 . 25 fold ) ( Table S1 ) . The lesser decrease measured by expression arrays versus qRT-PCR can be attributed to the different methodologies; qRT-PCR was done using primers designed to be specific for either Rhox6 or Rhox9 ( Figure S1 ) , whereas expression changes measured by arrays may be dampened by cross-hybridization due to high sequence similarity between the genes . Importantly , Rhox6 and 9 expression depended on the amount of Kdm6a knockdown in a dose-sensitive manner , consistent with a sex-specific dosage effect ( Figure 3C ) . Note that Rhox5 also showed a significant decrease after Kdm6a knockdown but its analysis was not pursued at this time . As expected , KDM6A occupancy was reduced at the 5′ end and gene body of Rhox6 and 9 after knockdown in female ES cells ( Figure S5B ) . Conversely , H3K27me3 levels were significantly increased at the 5′ end , gene body , and 3′end of Rhox6 and 9 , while there was no significant change in levels of H3K4me3 ( Figure 3D and Figure S5C ) . Genes associated with pluripotency ( e . g . Cd9 , Nanog , Pou5f1 , Stat3 , Sox2 ) were not affected by Kdm6a knockdown in either female or male ES cells indicating no induction of differentiation ( Figure S5D ) . We conclude that KDM6A plays a critical and dose-dependent role in regulating Rhox6 and 9 expression in female but not male ES cells . Chromatin domains that contain both activating and inactivating histone marks in undifferentiated ES cells have been termed bivalent and are thought to be poised for activation during development [36] , [37] . Notably , bivalent genes include homeobox genes , such as HOX genes , suggesting that Rhox genes are also candidates for bivalency . ChIP-chip profiles in both female and male undifferentiated ES cells demonstrated that both Rhox6 and 9 in cluster β were enriched in H3K27me3 and H3K4me3 , indicating that these genes are bivalent and thus poised for activity during development ( Figure 4 and Figure S6 ) . Quantitative measurements showed higher H3K4me3 and H3K27me3 levels in female versus male ES cells ( Figure 2B , 2D ) . H3K4me3 levels decreased and H3K27 levels increased between day 0 and 15 of differentiation in female ES cells , consistent with a decrease in Rhox6 and 9 expression ( Figure 2B , 2D ) . In contrast , levels remained low in male ES cells . KDM6A binding being clearly female-biased would explain the female bias in gene expression at day 0–2 of differentiation , as described above ( Figure 1 and Figure 2 ) . Other Rhox genes within the α- and γ-clusters did not appear to be bivalent but rather were enriched with the repressive mark H3K27me3 , with no significant peaks of enrichment for the active mark H3K4me3 , suggesting that these genes are mostly contained in a silenced chromatin domain in both female and male undifferentiated ES cells ( Figure 4 and Figure S6 ) . Inspection of representative genes from each cluster , Rhox6 and 9 in cluster β , Rhox3e in cluster α , and Rhox12 in cluster γ , confirmed these findings ( Figure 4 ) . Note that while Rhox1 and 7 also appeared bivalent at day 0 , KDM6A was absent at their promoter , which may account for their low expression ( Table S1 ) . Re-analyses of published expression array data confirmed that Rhox6 and 9 and Kdm6a are expressed at a higher level in ovary than in testis ( Figure 5A ) [18] , [19] . This is consistent with measurements of expression in embryos , in which female germ cells have much higher expression than male germ cells ( Figure 1C ) . As expected for a gene that escapes X inactivation , Kdm6a expression was higher in all female tissues examined in comparison to male tissues , including brain and sexual organs , as well as somatic and germ cells from embryos ( Figure 5C and Figure 1G ) . To assess the in vivo binding of KDM6A to Rhox6 and 9 in reproductive tissues , chromatin extracted from adult mouse ovaries and testes was subjected to ChIP-qPCR . KDM6A occupancy was high at the promoters of Rhox6 and 9 in mouse ovary , consistent with high expression in this organ ( Figure 5A , 5B ) [38] , [39] . In mouse testis where Rhox6 and 9 expression is lower , KDM6A was still bound but to a lesser extent ( 38% of occupancy in ovary ) , reflecting lower expression ( Figure 5A , 5B ) . In mouse brain where the genes are not expressed [19] , KDM6A occupancy was almost undetectable in females and completely undetectable in males ( Figure 5B ) . Taken together , these data indicate that KDM6A occupancy is associated with Rhox6 and 9 expression in reproductive tissues , more significantly in females than in males . To determine the allele-specific expression of Rhox6 and 9 in ovary we employed F1 mice derived from crosses between C57BL/6J females with ( XistΔ ) or without ( XistΔ− ) an Xist mutation and Mus spretus males . In F1 animals that carry the mutant Xist ( XistΔ ) , X inactivation is completely skewed towards the M . spretus X chromosome . SNPs between the mouse species were used to distinguish alleles after RT-PCR and Sanger sequencing . In F1 mice with ( XistΔ ) or without ( XistΔ− ) the Xist mutation expression of Rhox6 and 9 was exclusively from the maternal C57BL/6J allele , with no evidence of the M . spretus allele , consistent with imprinting of the paternal allele ( Figure 5D ) . This is similar to what has been reported in mouse ES cells and placenta [25] . Control genomic DNA amplification confirmed the presence of the SNPs in the F1 mice ( Figure 5D ) . Our results suggest that imprinting has taken place in the germ cells from adult ovary , as we did not observe any evidence of paternal allele expression . By qRT-PCR Rhox6 and 9 expression was higher ( 1 . 7-fold and 3-fold , respectively ) in ovaries of F1 mice carrying the Xist mutation ( XistΔ ) in which the maternal allele is expressed in all cells ( due to skewing of X inactivation ) , compared to ovaries from non-mutant F1 mice ( XistΔ− ) in which the maternal allele is expressed in half of the cells ( due to random X inactivation ) ( Figure 5E ) . The X inactivation effect on Rhox6 and 9 expression would only be pertinent in somatic cells , but not in germ cells in which the inactive X chromosome is reactivated . This complicates interpretation of our data because expression was measured in whole ovary containing both germ cells with very high Rhox6 and 9 expression and supporting somatic cells with lower expression ( Figure 1C ) . Additional studies in germ cells and somatic cells of the ovary are needed to fully understand the developmental regulation of Rhox6 and 9 in this organ . Nonetheless , we conclude that female biased expression of Rhox6 and 9 in ovary is not due to bi-allelic expression in this tissue but rather to recruitment of KDM6A to activate Rhox6 and 9 on the active maternal X chromosome .
Rhox genes represent a set of X-linked homeobox genes specifically expressed in organs and cell types implicated in sexual development and reproduction [19] , [20] , [40] , [41] . Here , we provide functional evidence identifying KDM6A , an enzyme that removes methylation at lysine 27 of histone H3 , as an important regulator of a specific subset of Rhox genes , Rhox6 and 9 , in female ES cells and in ovary . Interestingly , KDM6A is encoded by an X-linked gene that escapes X inactivation and has higher expression in females , which may indirectly facilitate its sex-specific role in enhancing Rhox6 and 9 expression [26] , [27] , [28] . Our knockdown experiments clearly support an important role for KDM6A in regulating Rhox6 and 9 in female but not male ES cells . The 6–10 fold female bias in Rhox6 and 9 expression we measured in undifferentiated ES cells cannot be explained by the presence of two active X chromosomes in female ES cells prior to X inactivation since Rhox6 and 9 are paternally imprinted in these cells [25] . Rather , the female enhanced expression results from the specific recruitment of KDM6A at those genes to facilitate the transition from repressive to active histone modifications and to increase expression . A similar mechanism explains the female bias in Rhox6 and 9 expression in ovary where we demonstrate that the genes are imprinted as well . KDM6A is a member of a multi-protein complex that not only de-methylates H3K27me3 but also methylates lysine 4 at histone H3 to facilitate gene expression [5] , [6] . KDM6A counterbalances polycomb activity by regulating H3K27me3 levels [9] , which would help maintain Rhox6 and 9 expression in undifferentiated female ES cells and ovary . In differentiated female ES cells KDM6A occupancy decreases , which mirrors the accumulation of H3K27me3 at Rhox6 and 9 , consistent with low expression in most somatic cell types as well as with the heavy DNA methylation reported for these genes during development [24] . Our analyses of 8-cell embryos suggest that a female bias in Rhox6 and 9 expression is already present at this early stage , prior to gonadal development . Somatic and germ cells from developing embryos still show a female bias in Rhox6 and 9 expression at later stages ( 11 . 5–13 . 5dpc ) including those coincident with gonad differentiation , which confirms a previous study [20] . However , detailed analyses of sexed embryos at additional stages will be needed to fully follow developmental expression in multiple cell types . Strikingly , within the Rhox cluster only Rhox6 and 9 show a marked increase in KDM6A in female ES cells . Tellingly , these genes share a high degree of sequence similarity but differ from the other Rhox genes in other sub-clusters [19] , suggesting that they may contain a sequence motif to specifically recruit KDM6A . The question arises of which other histone demethylases would remove H3K27me3 at other Rhox genes to facilitate their expression in specific tissues . We determined that KDM6B , which also removes methylation at lysine 27 of histone H3 , has low expression in undifferentiated male and female ES cells ( data not shown ) . However , its expression increases after differentiation in both male and female ES cells , pointing towards a potential role for this enzyme in regulating expression of some of the other Rhox genes at later stages of development [42] . KDM6A binds to the promoter , gene body , and 3′end of Rhox6 and 9 in female ES cells , suggesting a mechanism of regulation at transcription initiation and elongation . Interestingly , recruitment of elongation factors to target genes has been demonstrated for KDM6B , in addition to its role in histone demethylation [43] . Epigenetic regulation of the Rhox cluster had been previously focused on DNA methylation [24] , [44] . Rhox5 whose expression peaks at day 9 after ES cell differentiation is repressed by DNA methylation at later stages , while it remains unmethylated and highly expressed in extra-embryonic tissues . Similarly , Rhox6 and 9 are repressed following the establishment of CpG methylation by DNA methyltransferases DNMT3b and DNMT1 at their promoter regions in the embryo proper but not in extra-embryonic tissues [24] . Rhox5 is the only gene together with Xist known to be expressed from the paternal X chromosome ( maternally imprinted ) at early embryonic stages ( until e6 . 5 ) ; surprisingly , it is expressed from the maternal X ( paternally imprinted ) in extra-embryonic tissues , like Rhox6 and 9 [18] , [24] , [25] , [45] . Our results are consistent with paternal imprinting of Rhox6 and 9 in mouse ovary , in agreement with other studies in placenta and ES cells [25] . We found that Rhox6 and 9 are bivalently marked in undifferentiated ES cells as they are occupied by nucleosomes containing histone H3 methylated at both lysine 27 and lysine 4 . Bivalent genes are usually silent while poised for expression [37] . However , Rhox6 and 9 are in fact expressed in undifferentiated ES cells , probably due to a specific recruitment of KDM6A in a portion of cells in which levels of H3K27me3 would be decreased . Bivalent modifications result from a dynamic equilibrium of negative and positive chromatin marks controlled by histone modifying enzymes such as KDM6A . Our findings of a H3K27me3 increase at Rhox6 and 9 after Kdm6a knockdown are in agreement with what has been reported for other bivalent genes and support a role for KDM6A in maintaining a balance between active and inactive marks at bivalent promoters [9] . Note that the extent of reduction in Rhox6 and 9 expression we measured in female ES cells is comparable to that reported for another HOX gene , HOXB1 , after KDM6A knockdown in human cells [8] . Many homeobox genes important for specification of cell types and organs contain bivalent domains , suggesting that bivalency is an important part of stem cell differentiation and development [46] , [47] . Except for Rhox6 and 9 , most other Rhox genes are not occupied by bivalent marks in undifferentiated ES cells , thus Rhox6 and 9 may be specifically activated to influence lineage commitment . It will be interesting to determine which pathways and specific lineages are stimulated by RHOX6 and RHOX9 proteins . Rhox6 has been implicated in the determination of the germ cell lineage [48] . So far , only Rhox5 and 9 have been studied in vivo . Whereas Rhox5-null male mice exhibit increased germ cell apoptosis and sperm motility defects leading to sub-fertility , Rhox9-null male or female mice do not have any apparent phenotypes [19] , [49] . It is possible that due to similarities in sequence , homeodomain , and expression patterns Rhox6 compensates for the loss of Rhox9 in these knockout mice [49] , [50] . Additional evidence based on knockouts in mouse and rat epididymis , suggests that Rhox5 may act as a master regulator of many of its paralogs [51] . Both Rhox6 and 9 are highly expressed in ovary and to a lesser extent , in testis ( this study ) and [19] . An intriguing finding from our study is that KDM6A occupancy is high at Rhox6 and 9 in ovary and thus may serve to keep these two Rhox genes active . KDM6A is also bound to Rhox6 and 9 in testis , although at a lower level ( 1 . 7-fold and 2 . 5-fold lower in testis than in ovary , respectively ) , suggesting a threshold effect and/or another level of control in testis . Our knockdown experiments do indicate that KDM6A affects Rhox6 and 9 expression in a dose-dependent manner . In embryonic gonads the majority of Rhox genes are already expressed in a sexually dimorphic manner from an early stage . Specifically , Rhox6 and 9 are predominantly expressed in female versus male primordial germ cells at 12 . 5–15 . 5dpc ( this study ) and [19] , [20] , [38] , [52] . In addition , Rhox6 and 9 are also expressed in somatic cumulus cells in ovary [53] , [54] . Our findings indicate that Rhox6 and 9 are both imprinted on the paternal allele and subject to X inactivation . This implies the existence of a population of somatic cells without any Rhox6 and 9 expression , suggesting that cumulus cells tolerate such mosaicism . In contrast , all germ cells would express Rhox6 and 9 following X re-activation and subsequent imprinting of the paternal X chromosome . This is similar to what has been reported for some of the Xlr genes , a family of mouse genes also implicated in reproduction , some of which are also imprinted and subject to X inactivation [55] . In addition to removal of H3K27me3 , KDM6A appears to have a demethylase-independent role in regulating chromatin structure [9] , [56] , [57] . Indeed , KDM6A and KDM6B regulate T-box family members through an interaction with SMARCA4-containing SWI/SNF complexes in T-cells [58] . Interestingly , Kdm6a knockout mice display a more severe phenotype at mid-gestation in female than male embryos [9] , [59] . Thus , the Y-linked paralog Uty compensates for Kdm6a deficiency allowing survival of male embryos by a demethylase independent mechanism , since UTY does not have demethylase activity [9] , [56] , [59] . However , while some KDM6A-deficient male mice survive , most do not or are runted throughout adulthood , indicating that H3K27 demethylation remains an important function of KDM6A for survival and growth [56] . Additionally , histone demethylation appears to be the predominant mechanism required for activation of genes important in differentiation since mouse and human cells lacking KDM6A but retaining UTY fail to reprogram [60] . Furthermore , male primordial germ cells lacking KDM6A do not develop , as H3K27me3 levels are retained when compared to wild type [60] . Our knockdown experiments are consistent with a role for KDM6A in controlling levels of H3K27me3 and expression of Rhox6 and 9 in female ES cells . However , we cannot rule out the contribution of a demethylase independent mechanism since we did not test for one in the context of Rhox expression control . It remains to be determined whether levels of KDM6A are critical for proper ovarian function . It is interesting that female mice with a single X chromosome , which would have a lower dose of KDM6A due to haploinsufficiency for Kdm6a , a gene that escapes X inactivation , have reduced fertility [26] , [61] , [62] . Furthermore , XO female mouse embryos are developmentally retarded when compared to XX littermates at early mid-gestation [63] . In human , the presence of a single X chromosome causes Turner syndrome associated with severe developmental defects and ovarian dysgenesis [62] . It will be important to determine whether any of the human RHOX genes are also regulated by KDM6A . Mutations of KDM6A in human cause Kabuki syndrome , associated with growth retardation , unique facial features , and severe intellectual disability . Both males with point mutations and females with complete heterozygous deletions have been reported [12] , [13] . The Kabuki phenotype , present in females who have one deleted KDM6A copy but absent in Turner syndrome females , may be due to partial silencing of the normal copy by X inactivation in Kabuki females , while Turner females would have one expressed copy in all cells . It would be interesting to examine ovaries in these patients . In summary , our study provides the first evidence that Rhox6 and 9 are regulated by the histone demethylase KDM6A in mouse ES cells and reproductive organs in a sex-specific manner . Our findings indicate that a gene that escapes X inactivation plays a sex-specific role in gene regulation in female ES cells and tissue . Higher female expression due to escape from X inactivation of Kdm6a may be favorable to Rhox6 and 9 expression in ovary .
Male ES cells WD44 ( from C . Ware , University of Washington , US ) and E14 ( BL6/Cast ) ( from J . Gribnau , Erasmus MC Rotterdam , NL ) , and female ES cells PGK12 . 1 [64] and E8 ( BL6/Cast ) ( from J . Gribnau , Erasmus MC Rotterdam , NL ) were grown in high glucose DMEM media supplemented with 15% fetal bovine serum ( FBS ) , 1% non-essential amino acids , 10 mg/ml APS , 0 . 1 mM 2-mercaptoethanol and 25 mM L-glutamine . ES cells were maintained in the presence of 1000 U/ml leukemia inhibitory factor ( LIF ) ( Millipore ) on a mono-layer of chemically inactivated mouse embryonic fibroblasts ( MEF ) and grown in a humidified incubator at 37°C and 5% CO2 . Plates were enriched for ES cells by incubation on 1% gelatin coated dishes for 30 min to allow MEFs to attach , followed by transfer to fresh gelatin coated plates for overnight culture . Differentiation was achieved by removing LIF and culturing on non-adherent dishes to facilitate the formation of embryoid bodies . After 7 days , embryoid bodies were transferred to cell culture dishes for 8 days . Cells were harvested on days 0 , 2 , 4 , and 15 after LIF removal . To follow X inactivation in PGK12 . 1 cells before and after differentiation , Xist expression was determined by RT-PCR and by RNA-FISH using standard protocols with a probe for Xist ( Vysis ) ( Figure S3 ) . Ovary , testis , and whole brain were collected from adult female and male C57BL/6J mice . Additionally , ovaries were collected from female F1 obtained by mating M . spretus males ( Jackson Labs ) with females that carry an Xist mutation ( XistΔ ) ( B6 . Cg-Xist<tm5Sado> ) ( from T . Sado , Kyushu University , available from RIKEN ) [65] . Female progeny were genotyped to verify inheritance of the mutant Xist allele . Female progeny with a XistΔ fail to silence the BL6 X chromosome and thus have complete skewing of inactivation of the spretus X chromosome . All procedures involving animals were reviewed and approved by the University Institutional Animal Care and Use Committee ( IACUC ) , and were performed in accordance with the Guiding Principles for the Care and Use of Laboratory Animals . Stealth Select siRNAs ( Invitrogen ) were selected to target three different locations of the Kdm6a mRNA . The three siRNAs were transfected together or individually and oligonucleotides with no target were used as negative controls . For transfection , 5 µl of Lipofectamine RNAi Max Reagent ( Invitrogen ) mixed with 250 µl of Opti-MEM I Reduced Serum Medium ( Invitrogen ) containing 100 pmol of siRNAs were incubated for 30 min prior to addition to 6-well plates seeded with 1 . 5×105 ES cells in 2 ml of DMEM supplemented as stated above . Cells were harvested after 48 h of RNAi treatment . Knockdown was confirmed by qRT-PCR , expression arrays , and Western blotting using standard procedures . Immunoblot analysis was done using a KDM6A/UTX antibody either from K . Ge ( NIDDK ) or from Bethyl Labs . Three siRNAs were pooled and protein levels were measured after 48 h of treatment . Western blot band densities were measured using ImageJ software ( http://rsbweb . nih . gov/ij/ ) . Tissues were homogenized using a glass homogenizer and ES cells were collected before , during , and after differentiation . Cells were incubated at room temperature for 15 min in 1% formaldehyde . Crosslinking was stopped by adding 50 µL glycine followed by a 5 min incubation at room temperature and cell lysis as described [66] . Chromatin was sonicated to yield fragments 300–1000 bp in length and was then pre-cleared with protein A agarose beads for 1 h at 4°C . An aliquot of 20 µL was kept to serve as the input fraction . Pre-cleared chromatin was incubated in immunoprecipitation buffer at 4°C overnight using the following antibodies: anti-KDM6A/UTX [7] , anti-UTX ( Bethyl Labs ) , anti-H3K27me3 ( Millipore ) , and anti-H3K4me3 ( Millipore ) . Samples were centrifuged at 1200 rpm for 1 min and a small portion of the suspension collected as the unbound fraction . Immunoprecipitated chromatin was collected and serially washed in increasingly stringent salt buffers . After elution , crosslinks were reversed in 5 M NaCl at 65°C overnight . DNA was purified using Qiaquick PCR purification kit ( Qiagen ) and subjected to PCR according to the following protocol: 95°C for 3 min followed by 35 cycles of 95°C for 30 sec , 56°C for 30 sec and 72°C 30 sec . Samples were incubated at 72°C for 10 min and analyzed by gel electrophoresis . Controls to assay the immunoprecipitation efficiency of KDM6A , H3K4me3 , and H3K27me3 antibodies included an active gene ( Kdm5c ) and an inactive gene ( Iqsec2 ) ( Table S2 ) . For qRT-PCR , total RNA was prepared using the Qiagen RNeasy kit with on-column DNaseI digestion . For cDNA synthesis , 500 ng-1 µg of mRNA was reverse transcribed using the SuperScript First Strand Synthesis system ( Invitrogen ) according to manufacturer's protocol . Table S2 lists the RT-PCR primers specific for Rhox6 , Rhox9 , and Kdm6a . Quantitative PCR was performed using a SYBR green master mix ( Roche ) and a standard curve for each primer pair . Data normalized to the 18s housekeeping gene were averaged for 2 to 3 separate reactions each assayed in duplicate . For chromatin analyses , ChIP DNA was subjected to real-time PCR using primers listed in Table S2 . Rhox6R1 and Rhox9R1 amplify regions upstream of the transcription start site of their respective gene . Rhox6/9R2 amplify regions in the 5′ gene body , and Rhox6/9R3 regions towards the 3′ end of both Rhox6 and Rhox9 . After normalization to the input fraction , relative enrichment was calculated based on two separate immunoprecipitation reactions each assayed in duplicate . Following PCR amplification , melting curves were used to ensure only a single product was amplified . Western blots were done to confirm sexually dimorphic KDM6A protein levels in female and male ES cells using standard procedures . Briefly , nuclear protein was captured using an anti-KDM6A antibody ( Bethyl Labs ) using 1∶5000 dilution . Anti-β-ACTIN was used at 1∶10 , 000 dilution ( Sigma ) as a loading control . KDM6A protein was detected using HRP conjugated donkey anti-rabbit IgG , and β-ACTIN was detected using HRP conjugated goat anti-mouse IgG . Allele-specific expression was determined by Sanger sequencing of RT-PCR products and control PCR of genomic DNA using primers listed in Table S2 . For Rhox6 , the SNP ( T>G ) that distinguishes between the maternal C57BL/6J ( Xa ) and paternal M . spretus ( Xi ) alleles is at nucleotide position 35180550 ( NCBI37/mm9 build ) . For Rhox9 , the SNP ( T>G ) that distinguishes between the maternal C57BL/6J ( Xa ) and paternal M . spretus ( Xi ) alleles is at nucleotide position 35254278 ( NCBI37/mm9 build ) . cDNA was hybridized to Affymetrix 1 . 0 ST and Affymetrix 430 2 . 0 mouse arrays . Array hybridizations were done at the Microarray Center or at the Center on Human Development and Disability ( University of Washington , Seattle WA ) . The raw data files from the 430 2 . 0 arrays were analyzed by the Affymetrix software ( GCOS 1 . 1 ) to produce the data in . CHP format ( Excel ) , while raw data files from the 1 . 0 ST arrays were analyzed with Affymetrix Expression Console Software ( http://www . affymetrix . com ) . Data was normalized with the RMA method as implemented in the Bioconductor Affymetrix package . Microarray quality control metrics were included according to the manufacturer's recommended guidelines . For analyses of published array data [32] , [38] , [39] , spots were normalized by dividing the signal intensity against the average fluorescent intensity of each array [67] . Expression array data have been deposited in NCBI's GEO database [68] and are accessible through series accession number GSE45034 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE45034 ) . ” Following ChIP , DNA was amplified by whole genome amplification using the GenomePlex Complete Whole Genome Amplification Kit ( Sigma ) with modifications previously described [69] . ChIP DNA was lyophilized and re-suspended in 10 µl of water . Library preparation buffer and stabilization buffer were added ( 2 µl and 1 µl , respectively ) , and samples incubated at 95°C for 2 min . After addition of library preparation enzyme , samples were incubated in a thermal cycler according to the following protocol: 16°C for 20 min , 24°C for 20 min , 37°C for 20 min , 75°C for 5 min . For amplification of the library , a master mix containing amplification master mix , water , and WGA DNA polymerase was added and samples subjected to 15 cycles of: 95°C for 3 min , 94°C for 15 sec , and 65°C for 5 min . Samples were purified using the Qiaquick PCR purification kit . ChIP and input fractions were labeled according to the standard Nimblegen sample labeling protocol prior to hybridization to HD2 Nimblegen tiling arrays for the entire mouse X chromosome ( Roche ) . Enrichment profiles were generated ( Genomics Resource Center , Fred Hutchinson Cancer Research Center , Seattle WA ) . Peak maps generated by the Nimblescan software consist of significant peaks ( FDR score <0 . 05 ) . Tiling array data have been deposited in NCBI's GEO database [68] and are accessible through series accession number GSE45390 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE45390 ) . ” All p-values shown represent paired two-tailed Student's t-tests . | Homeobox ( HOX ) genes are known to be under epigenetic control during development . Here , we report that two mouse X-linked homeobox genes implicated in reproduction , Rhox6 and 9 , are activated by the histone demethylase KDM6A that removes methylation at lysine 27 of histone H3 . Kdm6a is one in a small group of genes that escape X inactivation in mice and humans , and thus has female-biased expression . We found that knockdown of Kdm6a affects Rhox6 and 9 expression specifically in female ES cells . We also demonstrate that high expression of Rhox6 and 9 in mouse ovary is associated with recruitment of KDM6A to these genes , consistent with a role in a female-specific organ . Furthermore , we demonstrate paternal imprinting of Rhox6 and 9 in mouse ovary . The findings herein help to understand sex bias in the regulation of reproductive homeobox genes during early development and in ovary . Our findings provide clues into the sex-specific roles played by genes that escape from X inactivation , which may contribute to developmental defects and ovarian dysfunction in individuals with X chromosome abnormalities . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"gene",
"function",
"developmental",
"biology",
"x",
"chromosome",
"inactivation",
"rna",
"interference",
"stem",
"cells",
"chromosome",
"biology",
"embryonic",
"stem",
"cells",
"genomic",
"imprinting",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"epigenetics",
"molecular",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"chromatin",
"dna",
"transcription",
"histone",
"modification"
] | 2013 | Female Bias in Rhox6 and 9 Regulation by the Histone Demethylase KDM6A |
The infectious and diagnostic form of Entamoeba histolytica ( Eh ) , cause of amebic dysentery and liver abscess , is the quadranucleate cyst . The cyst wall of Entamoeba invadens ( Ei ) , a model for Eh , is composed of chitin fibrils and three sets of chitin-binding lectins that cross-link chitin fibrils ( multivalent Jacob lectins ) , self-aggregate ( Jessie lectins ) , and remodel chitin ( chitinase ) . The goal here was to determine how well the Ei model applies to Entamoeba cysts from humans . An Eh Jacob lectin ( EhJacob2 ) has three predicted chitin-binding domains surrounding a large , Ser-rich spacer . Recombinant EhJacob2 made in transfected Eh trophozoites binds to particulate chitin . Sequences of PCR products using primers flanking the highly polymorphic spacer of EhJacob2 may be used to distinguish Entamoeba isolates . Antibodies to the EhJacob2 , EhJessie3 , and chitinase each recognize cyst walls of clinical isolates of Entamoeba . While numerous sera from patients with amebic intestinal infections and liver abscess recognize recombinant EhJacob1 and EhJessie3 lectins , few of these sera recognize recombinant EhJacob2 . The EhJacob2 lectin binds chitin and is polymorphic , and Jacob2 , Jessie3 , and chitinase are present in cyst walls of clinical isolates of Entamoeba . These results suggest there are substantial similarities between cysts of the human pathogen ( Eh ) and the in vitro model ( Ei ) , even though there are quantitative and qualitative differences in their chitin-binding lectins .
The infectious and diagnostic stage of Entamoeba histolytica ( Eh ) , the cause of amebic dysentery and liver abscess , is a quadranucleate cyst [1] , [2] . Eh is morphologically indistinguishable from Entamoeba dispar ( Ed ) , a human commensal that does not cause disease [3] . Because Eh does not readily encyst in axenic culture , we have studied cyst walls formed in vitro by Entamoeba invadens ( Ei ) that infects reptiles [4] , [5] . The Ei cyst wall is composed of chitin ( a homopolymer of β-1 , 4-linked GlcNAc ) and three unique sets of chitin-binding lectins called Jacob , Jessie , and chitinase [6] , [7] . Ei Jacob lectins contain 3 to 6 tandemly arranged chitin-binding domains ( CBDs ) , each of which contains six Cys residues ( see Table S1 for a list of database accession numbers and a brief description of each protein ) . Spacer regions between CBDs of Ei Jacob lectins contain sites for cleavage by Cys proteases , as well as Ser residues that are modified by O-phosphodiester-linked glycans [7] . Ei Jessie lectins and chitinase each contain an N-terminal CBD , which contains eight Cys residues [8]–[10] . Ei Jessie3 lectins contain a self-aggregating domain that forms the mortar or daub between chitin fibrils [11] . As Eh cysts are difficult to obtain from patient stool in quantity , we have predicted components of the cyst wall from the whole genome sequence of Eh [2] , [12] . An Eh Jacob lectin ( EhJacob1 ) that has two CBDs binds chitin when expressed as a recombinant protein in transfected Eh trophozoites ( Table S1 ) [10] . Similarly , the N-terminal CBDs of Eh chitinase , Jessie2 , and Jessie3 each bind chitin [10] . The Eh chitinase , chitin synthase , and chitin deacetylases each have the expected activities when expressed as recombinant proteins in bacteria or yeast [8] , [13] , [14] . Messenger RNAs for chitinases , Jessie lectins , and Jacob lectins are expressed by Eh encysting in xenic culture [15] . A low complexity spacer region between the CBD and enzymatic domain of Eh and Ed chitinases contains a series of heptapeptide repeats that are polymorphic among clinical isolates [16] , [17] . Polymorphic tandem repeats have also been observed in the Ser-rich Eh protein ( SREHP or K2 antigen ) [16]–[19] . While abundant and immunogenic Eh trophozoite proteins such as the Gal/GalNAc lectin and SREHP are immunogenic and are therefore vaccine candidates [20]–[24] , little is known about the immunogenicity of Eh cyst wall proteins . In an effort here to test how well the Ei cyst model fits the human pathogen Eh , we characterized here a second Eh Jacob lectin ( EhJacob2: EHI_044500; see Table S1 ) that contains three predicted CBDs separated by a long , Ser-rich spacer similar to those present in EiJacob6 and EiJacob7 ) [7] . Questions asked included the following:
Culture and manipulation of Entamoeba , including production of cysts in vitro and handling of cysts from patient samples , has been has been approved by the Boston University Institutional Biosafety Committee ( BU IBC ) . Similarly , recombinant expression of Entamoeba proteins in bacteria has been approved by the BU IBC . Rabbit antibodies were made using approved protocols from the BU IACUC . An exemption has been received from the Boston University IRB for de-identified patient sera and for de-identified stool samples containing Entamoeba cysts . Patient sera , all of which bound to Gal/GalNAc lectin , came from five individuals with amebic liver abscess and five individuals with intestinal amebiasis . All of these serum samples , which were de-identified , were collected prior to the initiation of these studies . The Ethical Review Committee of the International Centre for Diarrhoeal Disease Research , Bangladesh ( ICDDR , B ) and the Human Investigation Committee of the University of Virginia reviewed and approved the design of the previous study under which these samples were obtained . Eh and Ed Jacob2 lectins were identified in BLASTP searches of the NR database at NCBI or at AmoebaDB using the EhJacob1 sequence ( see Table S1 for accession numbers ) [10] , [12] , [25] . N-terminal signals and transmembrane helices were predicted using Phobius [26] . Genomic DNA from axenic Eh strains ( HM-1:IMSS , HK-9 , 200:NIH , and SD157 ) was isolated using the Wizard Genomic DNA Purification Kit ( Promega ) . DNA from an axenic strain of Ed ( SAW760 ) was a generous gift from Graham Clark . DNAs from numerous clinical isolates of Eh were a generous gift from Egbert Tannich . PCR primers flanking the Ser-rich region between the second and third CBDs of Jacob 2 were designed from sequences that were identical in the Eh and Ed genomic sequences . The sense primer ( GCTGATGGATTCTACTGTGTT ) encoded the heptapeptide ( ADGFYCV ) . The anti-sense primer ( ACAGAAAAGACCATCTTGAGT ) was anti-sense to heptapeptide ( TQDGLFC ) . In the Eh genome project strain HM-1:IMSS the predicted product was 1260-nt long [12] . PCR was performed for 35 cycles of 30 sec at 94°C , 30 sec at 50°C , and 3 min at 72°C using the PCR Master Mix system ( Promega ) . Amplified products were analyzed using a 0 . 8% agarose gel in 1× Tris-acetate-EDTA ( TAE ) buffer . Selected PCR products were cloned into a TA-vector and sequenced from both ends . The entire coding region of the EhJacob2 gene ( 1722 nt encoding a 574-aa protein ) was PCR amplified from HM-1:IMSS strain gDNA using the Expand High Fidelity PCR system ( Roche ) . The sense primer ( GCGGTACCATGAAACAACTTATATTAGCA ) began at the start codon ( italic ) and included a KpnI site ( underline ) . The anti-sense primer ( GCGGATCCTTATAAATCTTCTTCTGAAATTAATTTTTGTTCCTTGTTTTCATTGTTATTATT ) included a BamHI site ( single underline ) and was anti-sense to the 3′ end of the coding region of EhJacob2 ( bold underline ) . This primer was anti-sense to a c-myc sequence ( bold ) and to a stop codon ( italic ) . This product was cloned into the pJST4 vector [27] between the 5′ and 3′ untranslated regions of the Eh actin gene , and this construct was used to transfect HM-1:IMSS trophozoites . Transfected Eh trophozoites were lysed by incubation in lysis/wash buffer ( 20 mM Tris-HCl , pH 8 . 0; 1 M NaCl , 0 . 1% Triton X-100 ) plus 250 µM E64 for 1 hr on ice . The lysate was centrifuged for 1 min at maximum speed in a microcentrifuge , and the supernatant was incubated with chitin beads ( New England Biolabs ) for 1 hr at room temperature . Unbound material was then removed , and the beads were washed 5 times in lysis/wash buffer . Bound material was removed by boiling the beads for 5 min in SDS buffer ( 50 mM Tris-HCl , pH 6 . 8; 2% SDS; 5% 2-mercaptoethanol , 5% glycerol ) . Protein samples were analyzed by SDS-PAGE on 4–20% Tris-glycine gels . After electrophoresis , proteins were stained with Coomassie Blue or blotted onto nitrocellulose . EhJacob2 was detected on the blots using an anti-c-myc antibody ( Invitrogen ) followed by a peroxidase-conjugated goat anti-mouse IgG antibody ( Jackson ImmunoResearch ) . Bound antibodies were detected with the LumiGLO chemiluminescent substrate ( KPL ) . The region of the EhJacob1 gene encoding a 53-aa C-terminal 6-Cys CBD , which begins with VNCTEVKE and ends with the stop codon , was PCR amplified from Eh DNA . The sense primer ( CGGGATCCGTCAATTGTACTGAAGTGAAAGAA ) had a BamHI site at the 5′-end ( underline ) . The anti-sense primer ( CCCAAGCTTTTAGTGGTGGTGGTGGTGGTGATAACATGGATTGTTATAAC ) , which had a 5′ HindIII site ( underlined ) , was anti-sense to a stop codon ( italics ) , a polyHis tail ( bold ) , and the 3′ end of the coding region of the EhJacob1 gene ( bold underline ) . The coding region of the EhJacob2 gene ( minus the first 48 nt that encode the N-terminal 16-aa-long signal peptide and minus 24-aa at the C-terminus ) was PCR amplified from Eh DNA . The sense primer ( GGGTACCTAATGGTATACCCAACTGGATGTAAGAAGAAA ) had a Kpn1 site at the 5′-end ( underline ) and encoded the peptide ( VYPTGCKKK ) that is C-terminal to the predicted cleavage site in EhJacob2 for the signal peptidase [26] . The anti-sense primer ( GGATCCTTAGTGGTGGTGGTGGTGGTGGTATTGGTAAGGACCTTCTTGT ) , which had a BamH1 site ( underline ) , was anti-sense to a stop codon ( italics ) , a polyHis tail ( bold ) , and the 3′ end of the coding region of the EhJacob2 gene ( bold underline ) . The EhJacob1 and EhJacob2 PCR products were cloned into pMAL-p2E ( New England Biolabs ) , using the same methods we used to clone Eh Jessie3 into this vector [11] . Maltose-binding protein ( MBP ) fusions containing the Eh cyst wall lectins were expressed in E . coli ( Bl21-DE3 strain ) using IPTG induction and an amylose resin ( NEB ) for purification . The purity of these recombinant proteins was checked on SDS-PAGE . The 363-aa long catalytic domain of Eichitinase1 , which begins with the peptide ( KVVSYYT ) was amplified from Ei DNA using a sense primer ( GGATCCATGAAGGTTGTCTCGTATTACACC ) that had a 5′ KpnI site ( underline ) . The anti-sense primer ( CTCGAGTTAGCAACCGATCAAGCTCTTTC ) had a XhoI site ( underline ) and was anti-sense to a stop codon ( italic ) and the C-terminal peptide ( KKELDQC ) ( bold underline ) . The Eichitinase1 PCR product was cloned into the pQE30 vector ( Qiagen ) and expressed in M15 strain of E . coli that contains a lac repressor-expressing plasmid ( pREP4 ) . Recombinant Eichitinase1 , which contains a C-terminal polyHis tag , was induced with IPTG and purified on Ni-NTA agarose beads . Mono-specific polyclonal rabbit antibodies to amylose resin-purified MBP-EhJacob1 and MBP-EhJacob2 were made at Strategic Biosolutions , using methods similar to that used previously to make a rabbit anti-EhJessie3 antibody [11] . Prior to their use in microscopy , rabbit antibodies were purified using MBP-EhJacob1 or MBP-EhJacob2 fusion-proteins chemically coupled to agarose . Similar methods were used to raise a mono-specific rabbit antibody to the catalytic domain of Eichitinase1 . This antibody cross-reacts with the catalytic domain of Ehchitinase1 . Approximately 2 to 3 grams of stool sample from patients infected with Entamoeba ( in Kharagpur , India ) were emulsified in 10 ml of chilled phosphate-buffered saline ( PBS ) and then passed through a mesh to remove the larger particles from the materials . Each sample was washed with ice cold PBS by centrifugation at 5000 rpm for 5 min thrice , and the pellet was resuspended in 1 ml PBS . The presence of Entamoeba cyst in sample was confirmed by iodine staining and light microscopy or by calcofluor white staining and epifluorescent microscopy . To localize the Jacob2 , chitinase and Jessie3 in the Eh cyst wall , we fixed stool samples fixed with 2% paraformaldehyde for 15 min at room temperature and washed three times with PBS . Fixed samples were incubated for two hrs with 1∶200 dilutions of rabbit anti-EhJacob2 , anti-Eh Jessie3 , and anti-chitinase ( catalytic domain ) antibodies ( described above ) . Samples were washed with PBS and then incubated with TRITC-conjugated goat anti-rabbit antibody ( 1∶500 dilution ) for 1 hr . Secondary antibody alone was used as negative control . Samples were again washed with PBS and examined with a FV1000 confocal microscope ( Olympus ) . Images were captured by FV10-ASW 1 . 6 viewer and processed with Adobe photoshop CS3 . Finally , xenic cysts of Eh , which were incubated with anti-Jacob2 and anti-Jessie3 antibodies , were a generous gift of Upinder Singh [15] . For Western blotting , ∼2 µg each of MBP , MBP-EhJacob1 , MBP-EhJacob2 and MBP-EhJessie3 were separated in 4–16% gradient SDS-PAGE ( Invitrogen , USA ) and transferred to a PVDF membrane by semi-dry method . Blotted membranes were incubated with each patient's sera ( 1∶10 dilution ) ( Dacca , Bangladesh ) on a rocker overnight at 4°C . Membranes were washed three times for 15 min with PBS-Tween 20 and then incubated with HRP-conjugated anti-human antibody ( Sigma ) ( 1∶2000 dilution ) for 1 hr . Bound antibody was detected using Super Signal West Pico Chemiluminescent kit ( Pierce ) , as per manufacturer's instruction . The strength of bound antibodies was qualitatively scored as no signal ( − ) , barely detectable signal ( +/− ) , weak signal ( + ) , stronger signal ( ++ ) , and strongest signal ( +++ ) .
Eh has only two predicted Jacob lectins [2] , [10] , [12] . EhJacob1 , which we previously characterized [10] , is present in three nearly identical copies in the genome ( see Table S1 ) . EhJacob1 , which contains two CBDs , is 151-amino acids long , has a predicted molecular weight of 17377 daltons , and has a predicted pI of 5 . 2 . In contrast , the EhJacob2 lectin , which contains three predicted CBDs , is 574-amino acids long , has a predicted molecular weight of 62862 daltons , and has a predicted pI of 4 . 65 ( Figs . 1A and 1B ) . The first two predicted CBDs of EhJacob2 are separated from the third CBD by a large spacer domain , which is 30% Ser . Large Ser-rich spacer domains are also present in minor components of the Ei cyst wall ( EiJacob6 and EiJacob7 ) and in chitin-binding proteins of insects ( peritrophins ) that are present in the wall surrounding the blood meal [7] , [28] . Large Ser-rich domains in EhJacob2 suggest the possibility of numerous O-phosphodiester-linked glycans , as demonstrated in Ei Jacob lectins [7] . In contrast , there are no sites for Asn-linked glycosylation in EhJacob2 [29] . Within the spacer domain of EhJacob2 are numerous short repeats that are polymorphic ( see next section ) . These repeats include sequences ( e . g . TTPSTGV ) that resemble sites for cleavage by Cys proteases in Ei Jacob lectins ( TTPVD ) [7] . The predicted Jacob2 from the commensal parasite Ed ( EDI_246160 ) is 743-amino acids long and contains three CBDs that closely resemble those of EhJacob2 ( Fig . S1 and Table S1 ) . In contrast , the Ser-rich domain of EdJacob2 contains numerous repeats ( marked in bold letters in Fig . S1 ) that are distinct from those of EhJacob2 . To determine if EhJacob2 is a chitin-binding lectin , EhJacob2 was expressed with a myc-tag at its C-terminus in transfected Eh trophozoites [10] , [27] . A total lysate from transfected Eh was incubated with chitin beads , and unbound proteins ( the vast majority ) were removed ( Fig . 2A ) . A single , ∼78-kDa protein binds to the chitin beads . A Western blot using an anti-myc antibody confirmed that this chitin-binding protein is the recombinant EhJacob2 protein ( Fig . 2B ) . In control non-transfected E . histolytica trophozoites , no proteins bind to chitin beads ( data not shown ) . We hypothesized that the Ser-rich region of EhJacob2 might be polymorphic because similar low complexity regions containing internal repeats in Entamoeba SREHP and chitinase are polymorphic [16]–[19] . EhJacob2 PCR products from DNA of axenized strains of Eh ( HM-1:IMSS , HK-9 , and 200:NIH ) , one clinical isolate ( SD157 ) , and axenized Ed strain ( SAW760 ) range in size from 1 . 1 kb to 2 . 3 kb ( Fig . 3A ) . EhJacob2 PCR products also range in size from clinical isolates of Eh ( Fig . 3B ) . Selected EhJacob2 PCR products were cloned and sequenced at both ends , and five groups of repetitive elements in the Ser-rich spacer were coded ( A to E in Fig . 4 ) , using methods to describe Entamoeba chitinase and Ser-rich protein repeats [16] . Nucleotide differences within groups included both silent and non-silent changes . While the repetitive elements differ among the four isolates examined , there are some similarities . For example , the repetitive regions all start with A1B3C2D2A1 and end with D1A1B1C4D1A1D3 . Blocks of ABCD are common , and HM-1:IMSS , HK-9 , and 200:NIH all have CB followed by variable numbers of E . In contrast , E repeats did not occur in the SD157 sequence . The idea here was to determine whether sera from patients with liver abscess or amebic intestinal infection , each of which recognizes the Gal/GalNAc lectin of Eh trophozoites [22] , [23] , also recognize recombinant Eh cyst wall proteins on Western blots . MBP alone was used as negative control . While 9 of 10 human anti-amebic sera recognized EhJessie3 , 6 of 10 sera recognized EhJacob1 ( Table 1 ) . In contrast , just 2 of 10 sera bound to EhJacob2 , suggesting EhJacob2 may be less antigenic than the other Eh cyst wall lectins . The results here and elsewhere generally support the idea that Ei is a good model for Eh cysts: Finally , it appears that EhJacob2 genes are at least as polymorphic as SREHP genes and are more polymorphic than chitinase genes [16]–[19] . These results support the general idea that polymorphisms in surface proteins that contain repetitive elements of Entamoeba , Cryptosporidium ( e . g . gp40/15 ) , and Plasmodium ( e . g . merozoite and circumsporozoite antigens ) may be used to distinguish clinical isolates [30]–[32] . The EhJacob2 polymorphisms may complement other methods such as tRNA gene-linked tandem repeats for finger-printing clinical isolates of Eh [33] , [34] . | For many years , we and others have used cysts of Entamoeba invadens ( Ei ) , a reptilian parasite , to model the infectious and diagnostic cysts of the human pathogen Entamoeba histolytica ( Eh ) . The Ei cyst wall is composed of chitin fibrils , as well as Jacob and Jessie lectins that have unique chitin-binding domains . Our recent results suggest a “wattle and daub” model of the Ei cyst wall , where the wattle or sticks ( chitin fibrils bound by multivalent Jacob lectins ) is constructed prior to the addition of the mortar or daub ( self-aggregating Jessie3 lectins ) . Here we “humanize” the Ei model of the cyst wall with four findings . First , a recombinant Eh Jacob2 lectin , which has three predicted chitin-binding domains surrounding a large spacer domain , binds chitin beads . Second , polymorphisms in the spacer domain of EhJacob2 discriminate clinical isolates of Entamoeba . Third , chitinase , Jacob2 lectin , and Jessie3 lectin are present in cyst walls of clinical isolates of Entamoeba . Finally , numerous sera from patients infected with Entamoeba recognize recombinant Eh Jacob1 and Jessie3 lectins . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"and",
"Discussion"
] | [
"cell",
"biology/morphogenesis",
"and",
"cell",
"biology",
"immunology/immune",
"response",
"biochemistry/macromolecular",
"assemblies",
"and",
"machines"
] | 2010 | The Jacob2 Lectin of the Entamoeba histolytica Cyst Wall Binds Chitin and Is Polymorphic |
French Guiana , a French overseas department located in South America between Brazil and Surinam , is the only European territory geographically located in the Amazonian forest complex and is considered endemic for yellow fever ( YF ) . In the context of the emergent threat of YF in Latin America , we conducted a large household cross-sectional survey from June to October 2017 to estimate vaccination coverage in the population and to determine associations with sociodemographic and geographical characteristics . In total , 1 , 415 households and 2 , 697 individuals were included from the 22 municipalities of French Guiana . YF vaccination coverage was estimated at 95 . 0% ( 95% CI: 93 . 4–96 . 2 ) in the entire territory but was spatially heterogeneous , with the lowest levels estimated in the western part of the territory along the Surinamese cross-border region , particularly in children under 16 years who were not enrolled in school , immigrant adults and disadvantaged populations with low socioeconomic indexes . Despite the good vaccination coverage against YF in the general population of French Guiana resulting from the compulsory nature of YF vaccination for residents and travelers , there is an urgent need to improve vaccination coverage in vulnerable populations living in the northwestern part of the territory to limit the risk of transmission in the context of the emerging YF threat in South America . Despite the relative rarity of YF and the significant number of infectious and tropical diseases in French Guiana , clinicians should adopt a high index of suspicion for YF , particularly in vulnerable and at-risk populations .
Yellow fever ( YF ) is the most severe arbovirus to circulate in the Americas , with symptoms ranging from mild non-specific illness to hemorrhagic fever , a systemic illness characterized by high viremia , hepatic , renal and myocardial injury , hemorrhage , and high lethality [1] . A single-dose vaccine has existed since the 1940s and has helped to substantially control and reduce YF transmission [2–4] . However , complete eradication is prevented by the sylvatic cycle of the virus within nonhuman primary hosts , and Aedes aegypti mosquitoes are responsible for occasional transmission to people [5] . Recent important outbreaks of YF in Africa and South America have confirmed the potential of arthropod-borne viruses to emerge or reemerge in risk areas [6] and have highlighted the urgent need to assess vaccination coverage efforts in the most exposed countries . Since November 2016 , after decades of silence , Brazilian authorities and scientists have reported an outbreak of YF associated with an exponential increase in the number of confirmed cases and deaths in humans [7–9] . The YF virus has spread into the coastal Atlantic forest zones and moved rapidly into the southeast and south of the country in less than one year , reaching several populous Brazilian states whose residents had not been included in the YF vaccination program [9 , 10] . The majority of reported cases have occurred in rural areas , clearly reflecting a typical sylvatic transmission cycle occurring between forest mosquitoes and forest-dwelling nonhuman primates , with humans serving only as accidental hosts . This important alert has prompted the Brazilian Ministry of Health to conduct massive vaccination campaigns among unvaccinated residents of affected areas [11] . French Guiana , a French overseas department located in South America between Brazil and Surinam , is the only European territory geographically located in the Amazonian forest complex and is considered endemic for YF [12] . Since 1967 , YF vaccination has been compulsory in French Guiana for all individuals older than 1 year of age ( with a booster dose every 10 years ) . The vaccination is free of charge and widely accessible in public vaccination centers and by accredited private practitioners . In February 2016 , according to the Strategic Advisory Group of Experts on Immunization and the modifications of the International Health Regulations [13] , French health authorities adopted the use of only a single dose of vaccine for most residents and travelers . Over the last decades , Ae . aegypti has been responsible for several major dengue fever outbreaks [14–18] and for the recent emergence of chikungunya in late 2013 [19–21] and Zika in 2016 [22 , 23] . Considering the large number of travelers moving through the Brazilian river border , the recent outbreak of YF has raised particular concern that an urban transmission could occur in French Guiana , specifically for nonvaccinated population subgroups , through the Ae . aegypti mosquitoes that are strongly represented in the territory . Although vaccination coverage reported by recent assessments was better than that observed 20 years ago , some geographical areas may present unsatisfactory levels of coverage , particularly in forest and exposed environments [24–27] . While a survey conducted in 2000 estimated YF vaccine coverage of 80–90% in children under 15 years old [24] , the overall vaccine coverage was estimated at 95 . 9% ( 95% CI 95·5–96·3 ) in 2009 in 9339 children from primary and secondary schools [25] . Lower coverage rates between 75% and 81% were observed in small municipalities located outside the urbanized and coastal areas . Moreover , French Guiana is experiencing continuous major waves of immigration facilitated by the natural and uncontrollable quality of the river borders , particularly from countries where vaccination against YF is not mandatory . In this context , two sporadic and fatal cases of YF were reported in French Guiana 1 year apart [28] , confirming that sylvatic YF circulation is active in this territory , particularly among nonvaccinated populations involved in important activities in the forest environment . The first case was confirmed in August 2017 in a 43-year-old Brazilian woman with unknown vaccination status . Epidemiological investigations reported a history of stay in the forest , suggesting that the patient could have been contaminated either in the Amapá state in Brazil or in French Guiana . In August 2018 , a second case was biologically confirmed in a non-vaccinated Swiss-citizen 47-year-old man who had entered through a river border and had lived in French Guiana for 4 months . Epidemiological investigations reported regular work activities on forest roads , suggesting autochthonous transmission in the forest environment . The last autochthonous case was identified in 1998 in the southeast of the territory[29] . These two recent case reports illustrate that despite the compulsory nature of YF vaccination in this French overseas region since 1967 , maintaining focus on the need for YF vaccination is important , especially in areas with favorable ecosystems for YF transmission . In the context of the emergent threat of YF in Latin America and , consequently , in French Guiana , we conducted a general population cross-sectional study to estimate vaccination coverage in the population and to determine associations with sociodemographic and geographical characteristics .
The study was approved by the “Sud-Ouest & Outre-Mer IV” Ethical Research Committee ( No . CPP17-007a/2017-A00514-49 ) and by the French Data Protection Authority ( No . DR-2017-324 ) , which is responsible for ethical issues and the protection of individual data collection . Publicity and information about the survey were provided through the media and contact with local and national authorities . Fieldworker teams were trained to visit all households , explain the project objectives , and , when allowed , collect participants’ signatures on a free and informed consent form and conduct the interviews . All household members of selected households who were 2–75 years of age were invited to take part in the study . For all participants under 18 years of age , one or two responsible adults signed the informed consent form . Data were collected through a standardized questionnaire installed on tablets to register demographic , socioeconomic and household characteristics . Vaccination cards or any other proof-of-vaccination documents were requested from all participants . Vaccination status was based on presented vaccination cards and verbal reports of vaccination when vaccination cards were not available . We employ the following notations to describe the study design: We considered that in each municipality i , the probability of selecting a particular subject was equal to the probability of selecting the subject’s household and was ( mi/Mi ) , corresponding to a statistical weight equal to ( 1/ mi/Mi ) = ( Mi/mi ) . This statistical weight indicates the number of people in the population represented by each subject in the sample . We applied a post-stratification adjustment to each of these weights to arrive at the final statistical weight for each subject . This adjustment helped us to weight the age-sex groups within each municipality to match the distribution in the total population of French Guiana . Ten age groups ( [2–5 years [ , [5–10[ , [10–15[ , [15–20[ , [20–25[ , [25–35[ , [35–45[ , [45–55[ , [55–65[ , and ≥65 years ) were defined within male and female groups . For each age-sex subgroup , we applied an adjustment factor cijk to obtain a final statistical weight: wijk = ( Mi/mi ) *cijk , where i indexes municipalities , j indexes sex groups and k indexes age groups . We constructed a household socioeconomic index combining a multiple correspondence analysis and a hierarchical cluster analysis based on household material possessions , socioprofessional category and household income . The weighted vaccination coverage estimation for one dose of YF was based on doses recorded on vaccination documents and/or reported by participants . Associated factors were identified using survey-weighted Poisson regression , and the strength of selected variables and vaccination coverage was estimated by raw and adjusted risk ratios ( RR ) and a 95% confidence interval ( CI ) . All RRs excluding 1 . 0 were considered significant . Analyses were conducted using the survey capabilities of Stata version 15 statistical software ( Stata Corp , College Station , TC , USA ) [31] . French Guiana’s layers were drawn using geodata from OpenStreetMaps ( http://www . openstreetmap . org ) , and mapping operations were performed using QGIS 2 . 18 software [32] .
In total , 1 , 415 households and 2 , 697 individuals were included from 22 municipalities ( Table 1 ) , representing 58% of eligible household members . The mean household size was 1 . 9 individuals [range: 1 to 11] . The mean age was 30 . 5 , ranging from 2 to 75 years old . Comparison of the sociodemographic characteristics of the study sample with census data demonstrated an overrepresentation of women ( 58 . 9% vs . 50 . 0% in the general population of French Guiana ) and adults over 25 years ( 64% vs . 53% in French Guiana ) . These differences were accounted for in the analyses of vaccination coverage and risk factors by allocating a poststratification weight to each participant . Vaccination coverage for YF ( at least one dose ) was estimated at 95 . 0% ( 95% CI: 93 . 2–96 . 3 ) throughout the entire region of French Guiana ( Table 1 ) . Eighty percent of the respondents presented a vaccination certificate or an equivalent document as evidence of vaccination , while 15 . 6% reported that they had received the vaccination but had no hard evidence . The number of booster doses received by vaccinated individuals ranged from 1 to 6 . The mean age at vaccination was 1 . 7 years among children under 10 born in French Guiana . The coverage was spatially heterogeneous and decreased from the central coastal area to the western part of the territory along the Maroni River , which forms the border with Suriname ( Fig 2 ) . The highest vaccination coverage levels were observed in small and remote villages or in municipalities with fewer than 3 , 000 inhabitants ( Antecume Pata , Twenke-Talhuen , Iracoubo , Roura , and Sinnamary ) , where all the respondents were vaccinated . While the majority of French Guiana had coverage levels higher than 90% , three municipalities located in the west border area ( Fig 2 ) had relatively low levels of vaccination coverage , including Grand-Santi ( 62 . 3% , 95% CI: 39 . 3%–80 . 9% ) , Saint Laurent ( 76 . 9% , 95%CI: 67 . 7%-84 . 1% ) and Papaïchton ( 78 . 3% , 95% CI: 60 . 2%-89 . 6% ) . More importantly , vaccination coverage was particularly low in children under 16 years in the municipalities of Saint-Laurent ( 40 . 1% , 95% CI: 21 . 6–61 . 9 ) , Papaïchton ( 51 . 9% , 95% CI: 26 . 0–76 . 8 ) and Grand-Santi ( 52 . 7% , 95% CI: 25 . 1–78 . 7 ) , while the coverage level was estimated at 97 . 1% , 95% CI: 94 . 5–98 . 5 in the other municipalities . More than 40% ( N = 23/51 ) of unvaccinated children aged 3–16 years living in these municipalities were not enrolled in school . Most of them ( 91% ) had parents born in Surinam or Guyana , and 72% of the children were born in French Guiana . Nearly 70% of the unvaccinated adults living in these municipalities were born outside of the country , and 30% had lived in French Guiana for less than 10 years . Disadvantaged groups benefiting from universal health coverage and state medical assistance schemes ( specifically conceived for undocumented migrants who become eligible after 3 months of residency in the French territory ) and those with low socioeconomic indexes were also associated with lower vaccination coverage in the entire territory ( Table 2 ) .
In 2017 , epizootic and sporadic human cases were observed in the northern part of the state of Pará , Brazil , near French Guiana . A recent case in neighboring Suriname was also identified in the Brokopondo Lake area , less than 100 km from the river border with French Guiana [28 , 33] . These situations suggest ongoing viral circulation and an emerging threat in the wider Guiana Shield region . In this context , it was incumbent upon us to estimate YF vaccination coverage throughout the entire region of French Guiana . Our results highlight good vaccination coverage against YF in the general population of French Guiana resulting from the compulsory nature of YF vaccination for residents and travelers [34] . However , vaccination rates appear to be insufficient in some western cross-border areas connected by river routes ( outside of vaccination control ) to countries potentially lacking sufficient vaccination coverage . This includes areas or countries where recent epidemics have occurred or where vaccination against YF is not mandatory [9 , 33] . While vaccination coverage estimates in French Guiana were the lowest in some western cross-border areas , including Grand Santi ( 62 . 3% ) , Papaïchton ( 78 . 3% ) , and Saint-Laurent ( 76 . 9% ) , the actual vaccination rate in these municipalities may be lower than the level recommended by the WHO to achieve and maintain a protective population-level immunity ( assumed to be approximately 60–80% ) [35] . This concern is based on the very low proportion of individuals who provided proof of vaccination in this part of French Guiana . Importantly , a large number of unvaccinated individuals in these areas were out-of-school children who were consequently not involved in vaccination monitoring and catch-up strategies conducted in formal educational settings . This situation poses an additional challenge for health authorities and prevention operators to reach and include these populations in vaccination catch-up strategies . While we estimated the population of children aged 3–16 years not attending school in the entire territory to be 5 . 4% , 75% of them lived in these western cross-border municipalities . Furthermore , it is possible that our study tends to overestimate vaccination coverage , particularly in specific areas associated with a low proportion of presentation of vaccination proof . Although a large majority of individuals who reported that they had received vaccination without hard evidence were able to provide the date and the service of vaccination , information on vaccination history without a card or hard evidence could be falsified . Another limitation of our study is that irregular immigrants without health coverage were underrepresented in our sample . Given that individuals without health coverage could not be enrolled in our survey because of restrictions from French legislation , this population was underrepresented in our study . Although this population was very small in most of the main municipalities of the coastal areas , some households were excluded in the western part of the territory , which is known for high levels of immigration , because the adults and referents of the selected household did not have health insurance status . Six individuals without health insurance status were included from households whose referents were eligible and enrolled in the survey . Only three of them had received single doses of vaccination , suggesting that recent immigrants , who are often in irregular situations and targeted by police operations , are at risk of being unvaccinated and are difficult for health professionals to reach . This situation may lead to YF cases or clusters in specific and unvaccinated population subgroups . In this context , it should be a priority to focus vaccination campaigns in the northwestern part of the territory where vaccination coverage rates are the lowest and most likely overestimated . Vaccination strategies and campaigns should be adapted to continuously improve vaccination coverage in children who are not enrolled in school , migrant populations who have recently arrived in French Guiana and other vulnerable populations , particularly if they are involved in essential activities in forest areas . Although the Social Security Fund and local health authorities provide yellow fever vaccination free of charge in the entire territory , vulnerable and unvaccinated populations without health insurance status may have poor access to health care providers and limited opportunities to be considered in the vaccination campaigns that are usually conducted in school and health centers . Alternative community initiatives based on places of religious worship , citizen’s centres or other places for social gathering should be used to reach these populations to increase vaccination coverage in target populations . Moreover , it is essential to maintain vaccine strategies and policies related to airport vaccination status controls and to raise awareness among health-care providers regarding the importance of verifying the immunization status of patients at each encounter regardless of patient origin to contribute to vaccination catch-up efforts . Despite the relative rarity of YF and the significant number of infectious and tropical diseases in French Guiana , clinicians should adopt a high index of suspicion for YF , particularly in unvaccinated travelers returning from affected regions . Daily air travel exchanges with metropolitan France could be the basis for the introduction of the YF virus in Europe . Despite the low epidemic risk in temperate countries , the local cycle of YF transmissions in regions where competent Aedes albopictus populations are established becomes a plausible scenario [36] . The latest YF case , confirmed in a Swiss man who supposedly arrived in French Guiana by land in April 2018 , illustrates this hypothesis . If the clinical symptoms of the patient had not developed while he was still in French Guiana , this could have led to an imported case in Europe and consequently to the occurrence of secondary cases , underlining the need for continued vigilance with respect to YF . | Yellow fever ( YF ) is the most severe arbovirus to circulate in the Americas . French Guiana , a French overseas department located in South America between Brazil and Surinam , is the only European territory geographically located in the Amazonian forest complex and is considered endemic for YF . We conducted a large general population survey from June to October 2017 to estimate vaccination coverage in the population and to identify target vulnerable populations for catch-up vaccination strategies . In total , 1 , 415 households and 2 , 697 individuals were included from the 22 municipalities of French Guiana . YF vaccination coverage was estimated at 95 . 0% ( 95% CI: 93 . 4–96 . 2 ) in the entire territory but was spatially heterogeneous , with the lowest levels estimated in the western part of the territory along the Surinamese cross-border region , particularly in children under 16 years who were not enrolled in school , immigrant adults and disadvantaged groups of populations with low socioeconomic indexes . Our findings showed that vaccination campaigns should be prioritized and adapted to improve vaccination coverage among vulnerable populations living in the northwestern part of the territory to limit the risk of transmission in the context of the emerging YF threat in South America . Despite the relative rarity of YF and the significant number of infectious and tropical diseases in French Guiana , clinicians should adopt a high index of suspicion for YF , particularly in vulnerable and at-risk populations . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"education",
"immunology",
"geographical",
"locations",
"sociology",
"social",
"sciences",
"ethnicities",
"preventive",
"medicine",
"vaccination",
"and",
"immunization",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"south",
"america",
"ecosystems",
"schools",
"european",
"people",
"brazil",
"french",
"people",
"yellow",
"fever",
"people",
"and",
"places",
"ecology",
"forests",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"viral",
"diseases",
"europe",
"terrestrial",
"environments"
] | 2019 | Vaccination coverage in the context of the emerging Yellow Fever threat in French Guiana |
We introduce a theory of sequential causal inference in which learners in a chain estimate a structural model from their upstream “teacher” and then pass samples from the model to their downstream “student” . It extends the population dynamics of genetic drift , recasting Kimura's selectively neutral theory as a special case of a generalized drift process using structured populations with memory . We examine the diffusion and fixation properties of several drift processes and propose applications to learning , inference , and evolution . We also demonstrate how the organization of drift process space controls fidelity , facilitates innovations , and leads to information loss in sequential learning with and without memory .
Genetic drift refers to the change over time in genotype frequencies in a population due to random sampling . It is a central and well studied phenomenon in population dynamics , genetics , and evolution . A population of genotypes evolves randomly due to drift , but typically changes are neither manifested as new phenotypes nor detected by selection—they are selectively neutral . Drift plays an important role in the spontaneous emergence of mutational robustness [6] , [7] , modern techniques for calibrating molecular evolutionary clocks [8] , and nonadaptive ( neutral ) evolution [9] , [10] , to mention only a few examples . Selectively neutral drift is typically modeled as a stochastic process: A random walk that tracks finite populations of individuals in terms of their possessing ( or not ) a variant of a gene . In the simplest models , the random walk occurs in a space that is a function of genotypes in the population . For example , a drift process can be considered to be a random walk of the fraction of individuals with a given variant . In the simplest cases there , the model reduces to the dynamics of repeated binomial sampling of a biased coin , in which the empirical estimate of bias becomes the bias in the next round of sampling . In the sense we will use the term , the sampling process is memoryless . The biased coin , as the population being sampled , has no memory: The past is independent of the future . The current state of the drift process is simply the bias , a number between zero and one that summarizes the state of the population . The theory of genetic drift predicts a number of measurable properties . For example , one can calculate the expected time until all or no members of a population possess a particular gene variant . These final states are referred to as fixation and deletion , respectively . Variation due to sampling vanishes once these states are reached and , for all practical purposes , drift stops . From then on , the population is homogeneous; further sampling can introduce no genotypic variation . These states are fixed points—in fact , absorbing states—of the drift stochastic process . The analytical predictions for the time to fixation and time to deletion were developed by Kimura and Ohta [5] , [11] in the 1960s and are based on the memoryless models and simplifying assumptions introduced by Wright [12] and Fisher [13] in the early 1930s . The theory has advanced substantially since then to handle more realistic models and to predict additional effects due to selection and mutation . These range from multi-allele drift models and -statistics [14] to pseudohitchhiking models of “genetic draft” [15] . The following explores what happens when we relax the memoryless restriction . The original random walk model of genetic drift forces the statistical structure at each sampling step to be an independent , identically distributed ( IID ) stochastic process . This precludes any memory in the sampling . Here , we extend the IID theory to use time-varying probabilistic state machines to describe memoryful population sampling . In the larger setting of sequential learning , we will show that memoryful sequential sampling exhibits structurally complex , drift-like behavior . We call the resulting phenomenon structural drift . Our extension presents a number of new questions regarding the organization of the space of drift processes and how they balance structure and randomness . To examine these questions , we require a more precise description of the original drift theory . We begin with the definition of an allele , which is one of several alternate forms of a gene . The textbook example is given by Mendel's early experiments on heredity [16] , in which he observed that the flowers of a pea plant were colored either white or violet , this being determined by the combination of alleles inherited from its parents . A new , mutant allele is introduced into a population by the mutation of a wild-type allele . A mutant allele can be passed on to an individual's offspring who , in turn , may pass it on to their offspring . Each inheritance occurs with some probability . Genetic drift , then , is the change of allele frequencies in a population over time . It is the process by which the number of individuals with an allele varies generation after generation . The Fisher-Wright theory [12] , [13] models drift as a stochastic evolutionary process with neither selection nor mutation . It assumes random mating between individuals and that the population is held at a finite , constant size . Moreover , successive populations do not overlap in time . Under these assumptions the Fisher-Wright theory reduces drift to a binomial or multinomial sampling process—a more complicated version of familiar random walks such as Gambler's Ruin or Prisoner's Escape [17] . Offspring receive either the wild-type allele or the mutant allele of a particular gene from a random parent in the previous generation with replacement . A population of diploid individuals will have total copies of these alleles . ( Though we first use diploid populations ( two alleles per individual and thus a sample length of ) for direct comparison to previous work , we later transition to haploid ( single allele per individual ) populations for notational simplicity . ) Given initial copies of in the population , an individual has either with probability or with probability . The probability that copies of exist in the offspring's generation given copies in the parent's generation is: ( 1 ) This specifies the transition dynamic of the drift stochastic process over the discrete state space This model of genetic drift is a discrete-time random walk , driven by samples of a biased coin , over the space of biases . The population is a set of coin flips , where the probability of HEADS or TAILS is determined by the coin's current bias . After each generation of flips , the coin's bias is updated to reflect the number of HEADS or TAILS realized in the new generation . The walk's absorbing states—all HEADS or all TAILS—capture the notion of fixation and deletion . Fixation occurs with respect to an allele when all individuals in the population carry that specific allele and none of its variants . Restated , a mutant allele reaches fixation when all alleles in the population are copies of and , consequently , has been deleted from the population . This halts the random fluctuations in the frequency of , assuming is not reintroduced . Let be a binomially distributed random variable with bias probability that represents the fraction of copies of in the population . The expected number of copies of is . That is , the expected number of copies of remains constant over time and depends only on its initial probability and the total number ( ) of alleles in the population . However , eventually reaches fixation or deletion due to the change in allele frequency introduced by random sampling and the presence of absorbing states . Prior to fixation , the mean and variance of the change in allele frequency are: ( 2 ) ( 3 ) respectively . On average there is no change in frequency . However , sampling variance causes the process to drift towards the absorbing states at and . The drift rate is determined by the current generation's allele frequency and the total number of alleles . For the neutrally selective case , the average number of generations until fixation ( ) or deletion ( ) is given by Kimura and Ohta [5]: ( 4 ) ( 5 ) where denotes effective population size . For simplicity we take , meaning all individuals in the population are candidates for reproduction . As , the boundary condition is given by: ( 6 ) That is , excluding cases of deletion , an initially rare mutant allele spreads to the entire population in generations . One important consequence of the theory is that when fixation ( ) or deletion ( ) are reached , variation in the population vanishes: . With no variation there is a homogeneous population , and sampling from this population produces the same homogeneous population . In other words , this establishes fixation and deletion as absorbing states of the stochastic sampling process . Once there , drift stops . Figure 1 illustrates this , showing both the simulated and theoretically predicted number of generations until fixation occurs for , as well as the predicted time to deletion for reference . Each simulation was performed for a different initial value of and averaged over 400 realizations . Using the same methodology as Kimura and Ohta [5] , we include only those realizations whose mutant allele reaches fixation . Populations are produced by repeated binomial sampling of uniform random numbers between and . An initial probability is assigned to allele and probability to allele . The count of in the initial population is incremented for each random number less than . This represents an individual acquiring the allele instead of . The maximum likelihood estimate of allele frequency in the initial sample is simply the number of alleles over the sample length: . This estimate of is then used to generate a new population of offspring , after which we re-estimate the value of . These steps are repeated each generation until fixation at or deletion at occurs . This is the Monte Carlo ( MC ) sampling method . Kimura's theory and simulations predict the time to fixation or deletion of a mutant allele in a finite population by the process of genetic drift . The Fisher-Wright model and Kimura's theory assume a memoryless population in which each offspring inherits allele or via an IID binomial sampling process . We now generalize this to memoryful stochastic processes , giving a new definition of fixation and exploring examples of structural drift behavior .
How can genetic drift be a memoryful stochastic process ? Consider a population of haploid organisms . Each generation consists of alleles and so is represented by a string of symbols , e . g . , where each symbol corresponds to an individual with a particular allele . In the original drift models , a generation of offspring is produced by a memoryless binomial sampling process , selecting an offspring's allele from a parent with replacement . In contrast , the structural drift model produces a generation of individuals in which the sample order is tracked . The population is now a string of alleles , giving the potential for memory and structure in sampling—spatial , temporal , or other interdependencies between individuals within a sample . At first , this appears as a major difference from the usual setting employed in population biology , where populations are treated as unordered collections of individuals and sampling is modeled as an independent , identically distributed stochastic process . That said , the structure we have in mind has several biological interpretations , such as inbreeding and subdivision [18] or the life histories of heterogeneous populations [19] . We later return to these alternative interpretations when considering applications . The model class we select to describe memoryful sampling is the -machine : the unique , minimal , and optimal representation of a stochastic process [4] . As we will show , these properties give an important advantage when analyzing structural drift , since they allow one to monitor the amount of structure innovated or lost during drift . We next give a brief overview of -machines and refer the reader to the previous reference for details . The -machine representations of the finite-memory discrete-valued stochastic processes we consider here form a class of ( deterministic ) probabilistic finite-state machine or unifilar hidden Markov model . An -machine consists of a set of causal states and a set of per-symbol transition matrices: ( 7 ) where is the set of alleles and where the transition probability gives the probability of transitioning from causal state to causal state and emitting allele . The causal state probability , , is determined as the left eigenvector of the state-to-state transition matrix . Maintaining our connection to ( haploid ) population dynamics , we think of an -machine as a generator of populations or length- strings: . As a model of a sampling process , an -machine gives the most compact representation of the distribution of strings produced by sampling . Consider a simple binary process that alternately generates s and s called the Alternating Process shown in Figure 2 . Its -machine generates either the string or depending on the start state . The per-symbol transition matrices are: ( 8 ) ( 9 ) Enforcing the alternating period-2 pattern requires two states , and , as well as two positive probability transitions and . Branching transitions are required for a process to structurally drift; the Alternating Process has none . Two simple -machines with branching structure are the smaller Fair Coin Process ( Figure 3 ) and more complex Golden Mean Process ( Figure 4 ) . Both are discussed in detail later . Beyond using -machines as generators of stochastic processes , as just described , several alternative reconstruction algorithms exist to infer -machines from data samples—tree-merging [2] , state-splitting [20] , and spectral [21] . These algorithms share a general approach: First , estimate the distribution of subsequences . ( If given data as a single string , for example , slide a window of length over the string and count subsequences of lengths . ) Second , compute the distinct probability distributions of future subsequences conditioned on past subsequences ( histories ) . Third , partition histories into equivalence classes ( causal states ) that give the same conditional future distributions . And , finally , calculate the transition dynamic between states . Properly reconstructed , the causal states form a minimal sufficient statistic for prediction in the sense of Kullback [22] . Here , we circumvent these methods' complications . Section Structural Innovation and Loss introduces an alternative that avoids them and is , at the same time , more computationally efficient . We are now ready to describe sequential learning , depicted in Figure 5 . We begin by selecting the -machine as an initial population generator . Following a path through , guided by its transition probabilities , produces a length- string that represents the first population of individuals possessing alleles . We then infer an -machine from the population . is then used to produce a new population , from which a new -machine is estimated . This new population has the same allele distribution as the previous , plus some amount of variance . The cycle of inference and re-inference is repeated while allele frequencies drift each generation until fixation or deletion is reached . At that point , the populations ( and so -machines ) cannot vary further . The net result is a stochastically varying time series of -machines ( ) that terminates when the populations stop changing . Thus , at each step a new representation or model is estimated from the previous step's sample . The inference step highlights that this is learning: a model of the generator is estimated from the given finite data . The repetition of this step creates a sequential communication chain . Sequential learning is thus closely related to genetic drift except that sample order is tracked , and this order is used in estimating the next generator . The procedure is analogous to flipping a biased coin a number of times , estimating the bias from the results , and re-flipping the newly biased coin . Eventually , the coin will be completely biased towards Heads or Tails . In our drift model the coin is replaced by an -machine , which removes the IID model constraint and allows for the sampling process to take on structure and memory . Not only do the transition probabilities change , but the structure of the generator itself—the number of states and the presence or absence of transitions—drifts over time to capture the statistics of the sample using as little information as possible . This is an essential and distinctive aspect of structural drift . Before we can explore this dynamic , we first need to examine how an -machine reaches fixation or deletion . Recall the Alternating Process from Figure 1 , producing the strings and depending on the start state . Regardless of the initial state , the original -machine is re-inferred from any sufficiently long string it produces . In the context of sequential learning , this means the population at each generation is the same . However , if we consider allele to be represented by symbol and by symbol , neither allele reaches fixation or deletion according to current definitions . Nonetheless , the Alternating Process prevents any variance between generations and so , despite the population not being all s or all s , the population does reach an equilibrium: half s and half s . For these reasons , one cannot use the original population-dynamics definitions of fixation and deletion . This leads us to introduce structural stasis to combine the notions of fixation , deletion , and the inability to vary caused by periodicity . Said more directly , structural stasis corresponds to a process becoming nonstochastic , since it ceases to introduce variance between generations and so prevents further drift . However , we need a method to detect the occurrence of structural stasis in a drift process . A state machine representing a periodic sampling process enforces the constraint of periodicity via its internal memory . One measure of this memory is the population diversity [23]: ( 10 ) ( 11 ) where the units are [bits] . ( For background on information theory as used here , the reader is referred to Ref . [24] . ) The population diversity of the Alternating Process is bit at any size . This single bit of information corresponds to the machine's current phase or state . Generally , though , the value diverges——for arbitrary sampling processes , and so population diversity is not suitable as a general test for stasis . Instead , the condition for stasis can be given as the vanishing of the growth rate of population diversity: ( 12 ) Equivalently , we can test the per-allele entropy of the sampling process . We call this allelic entropy: ( 13 ) where the units are [bits per allele] . Allelic entropy gives the average information per allele in bits , and structural stasis occurs when . While closer to a general test for stasis , this quantity is difficult to estimate from population samples since it relies on an asymptotic estimate of the population diversity . However , the allelic entropy can be calculated in closed-form from the -machine representation of the sampling process: ( 14 ) For example , the Alternating Process has , the Fair Coin Process , and the Golden Mean Process ; all in units of bits per symbol . When , the sampling process has become periodic and lost all randomness generated via its branching transitions . In this way , we replace the vanishing variance ( ) of a single bias parameter in the Kimura drift setting with a general measure of the sampling process's stochasticity . This new criterion subsumes the notions of fixation and deletion as well as periodicity . An -machine has zero allelic entropy if any of these conditions occur . More formally , we have the following statement . Definition Structural stasis occurs when the sampling process's allelic entropy vanishes: . Proposition Structural stasis is a fixed point of finite-memory structural drift . Proof Finite-memory means that the -machine representing the population sampling process has a finite number of states . Given this , if , then the -machine has no branching in its recurrent states: , where and are asymptotically recurrent states . This results in no variation in the inferred -machine when sampling sufficiently large populations . Lack of variation , in turn , means the transition probabilities can no longer change and so the drift process stops . If allelic entropy vanishes at time and no mutations are allowed , then it is zero for all . Thus , structural stasis is an absorbing state of the drift stochastic process .
The Biased Coin Process is represented by a single-state -machine with a self loop for both Heads and Tails symbols; see Figure 3 . It is an IID sampling process that generates populations with a binomial distribution of alleles . Unlike the Alternating Process , the coin's bias is free to drift during sequential inference . These properties make the Biased Coin Process an ideal candidate for exploring memoryless drift . Figure 6 shows structural drift , using two different measures , for a single realization of the Biased Coin Process with initial [Heads] = Pr [Tails] = 0 . 5 . Structural stasis ( ) is reached after generations . The initial Fair Coin -machine occurs at the left of Figure 6 and the final , completely biased -machine occurs at the right . Note that the drift of allelic entropy and [Tails] are inversely related , with allelic entropy converging quickly to zero as stasis is approached . This reflects the rapid drop in population diversity . After stasis occurs , all randomness has been eliminated from the transitions at state , resulting in a single transition that always produces TAILS . Anticipating later discussion , we note that during this run only Biased Coin Processes were observed . The time to stasis of the Biased Coin Process as a function of initial [Heads] was shown in Figure 7 . Also shown there was the previous Monte Carlo Kimura drift simulation modified to terminate when either fixation or deletion occurs . This experiment illustrates the definition of structural stasis and allows direct comparison of structural drift with genetic drift in the memoryless case . Not surprisingly , we can interpret genetic drift as a special case of the structural drift process for the Biased Coin . Both simulations follow Kimura's theoretically predicted curves , combining the lower half of the deletion curve with the upper half of the fixation curve to reflect the initial probability's proximity to the absorbing states . A high or low initial bias leads to a shorter time to stasis as the absorbing states are closer to the initial state . Similarly , a Fair Coin is the furthest from absorption and thus takes the longest average time to reach stasis . The Biased Coin Process represents an IID sampling process with no memory of previous flips , reaching stasis when Pr[Heads] = 1 . 0 or 0 . 0 and , correspondingly , when . We now introduce memory by starting drift with as the Golden Mean Process , which produces binary populations with no consecutive s . Its -machine was shown in Figure 4 . Note that one can initialize drift using any stochastic process; for example , see the -machine library of Ref . [25] . Like the Alternating Process , the Golden Mean Process has two causal states . However , the transitions from state have nonzero entropy , allowing their probabilities to drift as new -machines are inferred from generation to generation . If the transition probability ( Figure 4 ) becomes zero the transition is removed , and the Golden Mean Process reaches stasis by transforming into the Fixed Coin Process ( top right , Figure 6 ) . Instead , if the same transition drifts towards probability , the transition is removed . In this case , the Golden Mean Process reaches stasis by transforming into the Alternating Process ( Figure 2 ) . To compare structural drift behaviors , consider also the Even Process . Similar in form to the Golden Mean Process , the Even Process produces populations in which blocks of consecutive s must be even in length when bounded by s [24] . Figure 8 compares the drift of Pr[Heads] for a single realization of the Biased Coin , Golden Mean , and Even Processes . One observes that the Even and Biased Coin Processes reach stasis as the Fixed Coin Process , while the Golden Mean Process reaches stasis as the Alternating Process . For different realizations , the Even and Golden Mean Processes might instead reach different stasis points . It should be noted that the memoryful Golden Mean and Even Processes reach stasis markedly faster than the memoryless Biased Coin . While Figure 8 shows only a single realization of each sampling process type , the top panel of Figure 9 shows the large disparity in stasis times holds across all settings of each process's initial bias . This is one of our first general observations about memoryful processes: The structure of memoryful processes substantially impacts the average time to stasis by increasing variance between generations . In the cases shown , time to stasis is greatly shortened . To illustrate the richness of structural drift and to understand how it affects average time to stasis , we examine the complexity-entropy ( CE ) diagram [26] of the -machines produced over several realizations of an arbitrary sampling process . The CE diagram displays how the allelic entropy of an -machine varies with the allelic complexity of its causal states: ( 15 ) where the units are [bits] . The allelic complexity is the Shannon entropy over an -machine 's stationary state distribution . It measures the memory needed to maintain the internal state while producing stochastic outputs . -Machine minimality guarantees that is the smallest amount of memory required to do so . Since there is a one-to-one correspondence between processes and their -machines , a CE diagram is a projection of process space onto the two coordinates . Used in tandem , these two properties differentiate many types of sampling process , capturing both their intrinsic memory ( ) and the diversity ( ) of populations they generate .
Much of the previous discussion focused on structural drift as a kind of stochastic process , with examples and behaviors selected to emphasize the role of structure . Although there was a certain terminological bias toward neutral evolution theory since the latter provides an entree to analyzing how structural drift works , our presentation was intentionally general . Motivated by a variety of potential applications and extensions , we describe these now and close with several summary remarks on structural drift itself . The Fisher-Wright model of genetic drift can be viewed as a random walk of coin biases , a stochastic process that describes generational change in allele frequencies based on a strong statistical assumption: the sampling process is memoryless . Here , we developed a generalized structural drift model that adds memory to the process and examined the consequences of such population sampling memory . Memoryful sampling is a substantial departure from modeling evolutionary processes with unordered populations . Rather than view structural drift as a replacement for the well understood theory of genetic drift , and given that the latter is a special case of structurally drifting populations , we propose that it be seen as a new avenue for theoretical invention . Given its additional ties to language and cultural evolution , we believe it will provide a novel perspective on evolution in nonbiological domains , as well . The representation selected for the population sampling mechanism was the class of probabilistic finite-state hidden Markov models called -machines . We discussed how a sequential chain of -machines inferred and re-inferred from the finite data they generate parallels the drift of alleles in a finite population , using otherwise the same assumptions made by the Fisher-Wright model . The mathematical foundations developed for the latter and its related models provide a good deal quantitative , predictive power . Much of this has yet to be exploited . In concert with this , -machine minimality allowed us to monitor information processing , information storage , and causal architecture during the drift process . We introduced the information-theoretic notion of structural stasis to combine the concepts of deletion , fixation , and periodicity for drift processes . Generally , structural stasis occurs when the population's allelic entropy vanishes—a quantity one can calculate in closed form due to the -machine representation of the sampling process . We revisited Kimura and Ohta's early results measuring the time to fixation of drifting alleles and showed that the generalized structural drift process reproduces these well known results when staying within the memoryless sampling process subspace . Starting with structured populations outside of that subspace led the sampling process to exhibit memory effects including structural innovation and loss , complex transients , and greatly reduced stasis times . Simulations demonstrated how an -machine diffuses through isostructural process subspaces during sequential learning . The result was a very complex time-to-stasis dependence on the initial probability parameter—much more complicated than Kimura's theory describes . Nonetheless , we showed that a process' time to stasis can be decomposed into sums over these independent subspaces . Moreover , the time spent in an isostructural subspace depends on the value of the -machine probability parameters at the time of entry . This suggests an extension to Kimura's theory for predicting the time to stasis for each isostructural component independently . Much of the phenomenological analysis was facilitated by the global view of drift process space given by the complexity-entropy diagram . Drift processes with memory generally describe the evolution of structured populations without mutation or selection . Nonetheless , we showed that structure leads to substantially shorter stasis times . This was seen in drifts starting with the Biased Coin and Golden Mean Processes , where the Golden Mean jumps into the Biased Coin subspace close to an absorbing state . This suggests that even without selection , population structure and sampling memory matter in evolutionary dynamics . The temporal or spatial memory captured by the -machine can be interpreted as nonrandom mating , reducing the effective population size and , in doing so , increasing sampling variance . It also suggests that memoryless models restrict sequential learning and overestimate stasis times for structured populations . We demonstrated how structural drift—diffusion , structural innovation and loss—are controlled by the architecture of connected isostructural subspaces . Many questions remain about these subspaces . What is the degree of subspace-jump irreversibility ? Can we predict the likelihood of these jumps ? What does the phase portrait of a drift process look like ? Thus , to better understand structural drift , we need to analyze the high-level organization of generalized drift process space . Fortunately , -machines are in one-to-one correspondence with structured processes [25] . Thus , the preceding question reduces to understanding the space of -machines and how they can be connected by diffusion processes . Is the diffusion within each process subspace predicted by Kimura's theory or some simple variant ? We have given preliminary evidence that it does . And so , there are reasons to be optimistic that in face of the open-ended complexity of structural drift , a good deal can be predicted analytically . And this , in turn , will lead to quantitative applications . | Human knowledge is often transmitted orally within a group via a sequence of communications between individuals . The children's game of Telephone is a familiar , simplified version . A phrase is uttered , understood , and then transmitted to another . Genetic information is communicated in an analogous sequential communication chain via replication . We show that the evolutionary dynamics of both problems is a form of genetic drift which accounts for memory in the communication chain . Using this , one can predict the mechanisms that lead to variations in fidelity and to structural innovation . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"neutral",
"theory",
"population",
"genetics",
"markov",
"model",
"mathematics",
"effective",
"population",
"size",
"stochastic",
"processes",
"emergence",
"forms",
"of",
"evolution",
"theoretical",
"biology",
"evolutionary",
"modeling",
"biology",
"evolutionary",
"theory",
"nonlinear",
"dynamics",
"genetic",
"drift",
"probability",
"theory",
"natural",
"language",
"processing",
"adaptation",
"genetics",
"computational",
"biology",
"evolutionary",
"biology",
"evolutionary",
"processes",
"genetics",
"and",
"genomics"
] | 2012 | Structural Drift: The Population Dynamics of Sequential Learning |
Peripheral infection by Trypanosoma brucei , the protozoan responsible for sleeping sickness , activates lymphocytes , and , at later stages , causes meningoencephalitis . We have videoed the cortical meninges and superficial parenchyma of C56BL/6 reporter mice infected with T . b . brucei . By use of a two-photon microscope to image through the thinned skull , the integrity of the tissues was maintained . We observed a 47-fold increase in CD2+ T cells in the meninges by 12 days post infection ( dpi ) . CD11c+ dendritic cells also increased , and extravascular trypanosomes , made visible either by expression of a fluorescent protein , or by intravenous injection of furamidine , appeared . The likelihood that invasion will spread from the meninges to the parenchyma will depend strongly on whether the trypanosomes are below the arachnoid membrane , or above it , in the dura . Making use of optical signals from the skull bone , blood vessels and dural cells , we conclude that up to 40 dpi , the extravascular trypanosomes were essentially confined to the dura , as were the great majority of the T cells . Inhibition of T cell activation by intraperitoneal injection of abatacept reduced the numbers of meningeal T cells at 12 dpi and their mean speed fell from 11 . 64 ± 0 . 34 μm/min ( mean ± SEM ) to 5 . 2 ± 1 . 2 μm/min ( p = 0 . 007 ) . The T cells occasionally made contact lasting tens of minutes with dendritic cells , indicative of antigen presentation . The population and motility of the trypanosomes tended to decline after about 30 dpi . We suggest that the lymphocyte infiltration of the meninges may later contribute to encephalitis , but have no evidence that the dural trypanosomes invade the parenchyma .
Human African trypanosomiasis , or sleeping sickness , results from infection by sub-species of the protozoan Trypanosoma brucei and is normally fatal if untreated . At early stages , trypanosomes multiply in the blood and peripheral organs where they are readily killed by available drugs . In the absence of early treatment , trypanosomes can invade the brain parenchyma [1] and for this CNS stage of the disease current treatments are unsatisfactory in that they are logistically difficult to administer , often have severe side effects , and are confronted by the emergence of resistant strains [2–4] . The development of new drugs would be aided by a better understanding of the onset of CNS stage disease . Peripheral lymphocytes are activated in trypanosomiasis [5–8] , and the immune system and cytokines are central to the neuropathology [9–13] . An aspect of this immune response , infiltration of the meninges by leukocytes , was reported by Mott [14] and is frequently referred to in review articles ( e . g . , [10 , 11 , 13] ) , but descriptions in research articles seem to be scanty . Most experimental work on animal models of meningoencephalitis caused by trypanosomes ( or by other pathogens ) has depended on analysis of homogenized tissue ( e . g . [15] ) , histological sections ( e . g . , [16 , 17] ) or cerebrospinal fluid ( e . g . [16] ) . In the present work , we have profited from the development of two-photon microscopy , which allows in vivo imaging with micron resolution through the thinned skull of the mouse [18–20] . In this way , moving cells in the meninges and superficial cortex can be videoed while the integrity of the delicate tissue is preserved . T cells have previously been videoed in the exposed spinal meninges in experimental autoimmune encephalopathy [21 , 22] , and in the cortical meninges after occlusion of the middle cerebral artery [19] , but in the latter case the location within the meninges was not determined . We have briefly reported that in CD-1 mice infected with T . b . brucei GVR35 , a standard model of sleeping sickness , trypanosomes invade the meninges [20] . In order to image T cells and dendritic cells ( as well as trypanosomes ) we have now used genetically modified C57BL/6 mice , and followed the progression of meningitis up to 40 days post infection ( dpi ) . The question of where in the meninges the lymphocytes and trypanosomes are located would seem to be very relevant to the role , if any , of these actors in the development of CNS pathology . The meninges comprise two main compartments: the dura , and , below it , the leptomeninges ( subarachnoid space and pia mater ) , which are separated from the dura by the apparently impermeable arachnoid membrane [23–25] . The subarachnoid space contains CSF and is continuous with the perivascular spaces of vessels penetrating the neural brain [26] . Hence , lymphocytes and pathogens in the subarachnoid space should face little obstacle to movement into the neural brain [21] , while any in the dura would be relatively isolated from it . It is to be noted that in small mammals the subarachnoid space , except where it accommodates the surface pial vessels , is shallow or occluded over the cortical convexities [24 , 27 , 28] . The dural space [23] is vascularized by ramifications of the meningeal arteries , and richly innervated by the trigeminal system [29–31] . Studies on migraine have demonstrated that the dura is an interface between the nervous and immune systems [32 , 33] . It is known that in two other murine models of disease , experimental autoimmune encephalomyelitis and middle cerebral artery occlusion , there is an increase in T cells outside pial vessels [19 , 21] . In contrast , we find that , in our infection model , the great majority of the meningeal T cells , dendritic cells and trypanosomes are in the dura . We have analyzed their numbers , movements , and interactions . Since meningoencephalitis is a common outcome of infection by many pathogens , including bacteria , viruses , protozoa , fungi and cancer , and in autoimmune diseases such as multiple sclerosis , the observations may also be relevant to diseases other than trypanosomiasis .
All animal experiments were performed in accordance with the Animals ( Scientific Procedures ) Act 1986 and the University of Glasgow care and maintenance guidelines . All animal protocols and procedures were approved by The Home Office of the UK government and the University of Glasgow Ethics Committee . Specifically , the number of animals was kept to a minimum and all surgery and imaging were done under terminal anesthesia . The fluorescent protein genes , EGFP , mCherry , mKate2 ( Evrogen ) and tdTomato ( Clontech ) were amplified using primers to add HindIII and BamHI sites and cloned into pGEMT . The HindIII/BamHI digested genes were each cloned into pHD1034 ( from C . Clayton , [34] ) to generate pHD1034-EGFP ( pGL2179 ) , pHD1034-mCherry ( pGL2160 ) , pHD1034-mKate2 ( pGL2174 ) and pHD1034-tdTomato ( pGL2221 ) . C57BL/6 mice and C57BL/6 GM reporter mice were bred and maintained under specific pathogen-free conditions . Mice expressing EGFP or DsRed in T cells under control of the hCD2 promoter were bred from progenitors kindly given by D . Kioussis and A . Patel [35] . Mice expressing EYFP under control of the CD11c promoter ( in dendritic cells ) are described in [36] . We also used crosses: hCD2-DsRed x CD11c-EYFP . Adult mice ( 19–30 g body weight ) of either sex were infected with 3 x 104 T . b . brucei strain GVR35 ( WT or FP ) trypanosomes by intraperitoneal injection and monitored for parasitemia by counting trypanosomes in blood taken from the tail vein , using a haemocytometer with a detection limit of 2x104 parasites/mL . The very rare mice that showed motor abnormalities were culled . No differences in the parameters measured were observed between male and female C57BL/6 mice great enough to show up among the scatter of the data . Use of CD-1 mice ( for which we had no reporter strains ) and T . b . brucei Lister 427 ( which is normally lethal by 4 dpi ) is described in ref . [20] . Inhibition of T cell activation was studied by injecting abatacept i . p . at 10 mg/kg body weight on alternate days . The abatacept ( Orencia ) , was a gift from Bristol-Meyers Squibb . Blood plasma was labeled by injection in a tail vein of a fluorescent marker: 70 kD dextran , conjugated with either fluorescein isothiocyanate ( "dextran-FITC" ) or rhodamineB isothiocyanate ( "dextran-rhodamine" ) both from Sigma , was dissolved at 100 mg/mL PBS and 50–70 μL injected . Alternatively , 20–30 μL of quantum dot solution was used ( QTracker , Invitrogen , emission peak at 705 , 655 , or 625 nm ) . Furamidine ( [2 , 5-bis ( 4-amidinophenyl ) furan] , also known as DB75 [39] ) was a gift from David W . Boykin; it was routinely injected with the vascular marker . Furamidine , dissolved in DMSO at 1 mg per 40 μL and injected at a final concentration of 10 mg/kg body weight , labeled nuclei of trypanosomes and also certain host cells ( S4 Fig ) . Excitation light came from a Ti-sapphire femtosecond laser tunable from 700 to 1050 nm ( Chameleon Ultra II , Coherent , Santa Clara , USA ) . To obtain long wavelengths with higher power , the output of the Ti-S laser passed through an optical parametric oscillator ( OPO , Coherent ) : when pumped by the Ti-S laser at about 800 nm , outputs up to 1200 nm were obtained . It was possible to use part of the pump beam simultaneously with the OPO output . The intensity of the Ti-S beam bypassing the OPO was regulated by an acousto-optical modulator controlled by the imaging program ( Zen 2010 , Zeiss ) . The intensity of the OPO beam was varied manually by a polariser . The scan head ( Zeiss LSM7 MP ) had a maximum rate of 8 frames per sec . Almost all the imaging was done with a 20x water immersion objective , NA 1 . 0 . with high NIR transmission ( W Plan-Apochromat , Zeiss ) . Excitation and emitted light were separated in the microscope nose by a dichroic mirror with a cutoff at 740 nm . Five detectors of non-descanned fluorescence were available , three multialkali photodiodes , and two GaAsP detectors . Zen 2010 ( Zeiss ) , Volocity ( Perkin-Elmer ) , Imaris 7 . 4 . 1 . ( Bitplane ) , and ImageJ ( N . I . H ) were used . Care was taken to establish the anatomical orientation of the image files , and images were transformed as necessary so that rostral was up and left lateral to the left . In some cases , when signal from dendritic cells or blood vessels leaked into the T cell channel , the contamination was removed by subtracting the unwanted channels using Zen . On Volocity and Imaris , one or more channels of the image were sometimes smoothed , and usually the contrast was enhanced . Observation of fiducial features ( such as crossed collagen fibers ) showed that tissue drift in the X and Y directions was never more than 5 . 8 μm/h , and usually undetectable ( less than 1 μm/h ) . In the Z direction there was , on average , significant drift corresponding to increasing distance between the microscope objective and the dura at 4 . 9 μm/h , SD 4 . 9 μm/h , p < 0 . 026 . Student's t test was used for comparisons between mice ( log ( mean number of cells ) , mean speed of cells , etc . ) and the Mann-Whitney test for distributions that were not normal ( individual cell speeds , velocities , etc ) .
In uninfected reporter mice , small numbers of T cells expressing fluorescent protein ( FP ) under control of the CD2 promoter were visible at the level of horizontal meningeal vessels ( Fig 1C ) but were very rare in the parenchyma where the larger blood vessels are vertical and linked by characteristic sinuous capillaries ( Fig 1D ) . After intraperitoneal injection of trypanosomes , the number of T cells in the meninges appeared to increase by 7 dpi ( Fig 1E ) and increased further ( Fig 1F , 22 dpi ) . To count T cells in an image field , we acquired a Z-stack deep enough to include the upper and approximate lower limits of the population . Several random fields were imaged in each mouse and the mean number of T cells per unit area of meninges calculated . A significant increase was observed by 11 dpi ( p = 0 . 001 ) , the increase leveled off at about 20 dpi and was maintained to 40 dpi , the longest we maintained infected mice ( Fig 1G ) . T cells are dependent on signals from antigen-presenting cells , notably dendritic cells , for activation , and dendritic cells are present in the normal murine dura [28 , 41] . We imaged reporter mice expressing Enhanced Yellow Fluorescent Protein ( EYFP ) under control of the CD11c promoter so that myeloid dendritic cells and a small sub-population of macrophages were fluorescent [36] . In uninfected mice , small numbers of CD11c+ ( EYFP ) cells were present , located on meningeal vessels ( Fig 2A ) . Most had the irregular shape and constant extension and retraction of processes that identified them as dendritic cells ( Fig 2A and S1 Video ) . A few CD11c+ cells were spherical and usually not adjacent to blood vessels ( not shown ) . The numbers of CD11c+ cells increased markedly by 12 dpi ( Fig 2B ) . The great majority of them were irregularly shaped and many were velate . These morphologies made it difficult to count them with any accuracy , but we estimate that infection increased their numbers by a factor of at least ten ( Fig 2C ) . Nearly all were within 40 μm of the skull , apart from a minority close to pial veins ( Fig 2D ) . None was observed below the pia mater ( Fig 2E ) . In mice infected with T . b . brucei GVR35 expressing a fluorescent protein ( FP ) , such as EGFP or tdTomato , extravascular ( as well as intravascular ) trypanosomes were observed in the meninges , usually from about 11 dpi ( Fig 3A and S2 Video ) . In contrast , in the superficial parenchyma only intravascular ones were observed ( Fig 3B ) , in agreement with other reports that extravascular trypanosomes are rare in the sub-pial cortex of Murinae [17 , 20 , 42 , 43] . Rare extravascular trypanosomes were observed in ventral areas of brain slices ( S3 Video ) . Since the extravascular trypanosomes moved rapidly , they could not be counted without imaging rapidly , so we scanned in one nominal XY plane at the maximum rate of 8 f . p . s . ( S4 Video ) . However , in any one image field , trypanosomes were almost all confined within a depth extending about 5 μm thick above and below the plane of maximum density . Given the finite depth of focus , and the up-and-down motion of the trypanosomes ( so they crossed this plane ) , we could make an approximate count by examining videos of 100 or more frames . In exceptional cases , trypanosomes in the mouse lost expression of FP . To forestall this possibility , we routinely added the DNA-binding drug furamidine [39] to the solution of blood marker that was injected intravenously . Furamidine is taken up by trypanosomes in vitro and can be detected by its fluorescence [39 , 44] . Within about 15 min of intravenous injection in infected mice ( 10 mg/kg ) , trypanosomes in the blood and meninges were brightly labeled ( [20] Fig 3C and 3D ) . The labeling was variable: sometimes only the nucleus and kinetoplast were labeled , giving blue or red fluorescence , and sometimes , as in Fig 3C , the nucleus was red and the cytoplasm blue . In any case , trypanosomes could be identified unambiguously by their movement ( S5 Video ) , which was unaffected by furamidine over the imaging period of up to 3h [20] . By comparison of the signals from trypanosomes expressing FP and from furamidine , it was apparent that all meningeal trypanosomes were labeled within 15 min of the injection of furamidine . Extravascular trypanosomes were commonly observed in the meninges from about 12 dpi , and their numbers tended to increase with time of infection up to about 30 dpi ( Fig 3E ) . In general , the arrival of trypanosomes was later than the increase in T cells ( dashed lines in Fig 3E ) . After 30 dpi the number of meningeal trypanosomes appeared to fall and , in two mice , no trypanosomes were found 39–40 dpi despite parasitemia of 1 . 4–8 . 4 x 106 mL-1 ( Fig 3E ) . More generally , there was no correlation between the numbers of trypanosomes in the meninges and the numbers in the blood ( Fig 3F ) , suggesting that trypanosomes do not exchange freely between blood and the extravascular space . From 20 to 40 dpi the number of GVR35 trypanosomes in the parietal meninges of C57BL/6 mice ranged from 0 to 1107 mm-2 ( Fig 3G; 17 mice ) ; the median ( 89 mm-2 ) did not differ significantly from that in CD-1 mice ( Fig 3G , [20] ) . We considered the possibility that extravascular trypanosomes observed in the meninges ≈ 0 . 3–4h after we began to thin the skull had somehow arrived there as an artifactual result of the surgery . Previously , we had reported that , consistently , no extravascular trypanosomes were found in the meninges three days after infection with T . b . brucei Lister 427 , despite very high parasitemia [20] . This result is supported by the present experiments , in which , in a further nine mice with parasitemia , no extravascular trypanosomes were found in the meninges ( Fig 3F ) . When trypanosomes were present , there was no correlation between meningeal population and parasitemia ( Fig 3F ) . These results argue strongly against a rapid arrival of extravascular trypanosomes during the surgery or the first minutes after it . Nor did surgery initiate a significant arrival that was more gradual , since the number of meningeal trypanosomes did not noticeably increase during up to 3 hours of imaging . Recent papers have reported extravascular trypanosomes in the cortical pia mater [43] or the superficial cortical parenchyma [45] , while T cells in the spinal meninges have been reported in the leptomeninges ( arachnoid plus pia mater [21] ) . Knowing that the meninges were intact , and benefitting from signals from structural elements , we attempted to define the compartments containing the trypanosomes and T cells . The upper boundary of the compartment was clearly the skull , as trypanosomes were observed immediately below it . Often they were in the same plane as extracellular collagen , the main component of the dura mater ( Fig 4A and S6 Video ) . Also at about this depth were cell nuclei labeled blue by intravenous furamidine ( Fig 3D ) . These cells are likely to be the 'mesothelial lining cells' of the dura that are labeled by intravascular aminoacridines [24] . Defining the lower boundary of the compartment occupied by trypanosomes , like that of the dura itself [46 , 47] , was not so straightforward . Trypanosomes were observed close to small horizontal vessels , often with an irregular trajectory and close to collagen ( Fig 4B and S7 Video ) , but were not observed at the level of larger horizontal vessels just above the parenchyma ( Fig 4C ) . The latter were clearly in the subarachnoid space; the former were probably branches of the meningeal artery in the dura . In images acquired over many seconds , the trypanosomes traced out apparently isolated volumes within about 40 μm of the skull ( Fig 4D ) . To delineate the extent of the subarachnoid space we infused dye ( Texas Red ) in the cisterna magna . From this site , dye is known to flow along the perivascular spaces of surface arteries and enter the subarachnoid space over the cortex [26 , 29 , 48] . As shown in Fig 4E , Texas Red , infused in this way , labeled spaces adjacent to large pial blood vessels and a thin , patchy , space extending across the cortex , in general agreement with histological studies on Murinae [27 , 31 , 41 , 46 , 47] . The dye is excluded from a space beneath the skull that contains nuclei labeled by furamidine . It is this "dural space" , and not the subarachnoid space , that appears to be occupied by trypanosomes . Qualitative observation through the dissecting microscope , and vertical sections from sample Z-stacks , suggested that the distance from the skull to pial vessels increased in infected mice . Like trypanosomes , T cells were seen close under the skull at the level of extracellular collagen ( Fig 5A and S8 Video ) . At 12 dpi , they were mainly above the horizontal pial vessels , as seen in a 3D reconstruction ( Fig 5B ) and a slightly oblique single plane ( Fig 5C ) . Dendritic cells were at the same level ( Fig 5B and 5C ) . T cells ( at 11 dpi ) were seen to be present at the same level as trypanosomes ( Fig 5D and S9 Video ) . However , at 30 dpi , T cells were seen adjacent to large pial veins ( Fig 5E ) and at 39 dpi , occasional T cells were seen some 50 μm below the pia mater ( Fig 5F ) . In summary , the increase in T cells is greatest in the dura , but T cells can be present at deeper levels . The activation state of T cells and the nature of their environment are reflected in the speed and pattern of their movement [19 , 21 , 34 , 49–51] . To look for changes in T cell behavior in the cortical meninges during the progression of trypanosomiasis , we made videos , acquiring Z-stacks at 15–30s intervals over periods of up to 45 min . Fig 6A shows typical tracks in an uninfected mouse . Three T cells were almost stationary , one moved rapidly 28 μm along a blood vessel then careered off ( S10 Video ) . In videos from infected mice , a greater proportion of the cells were moving ( with speed >1 μm/min; Fig 6B and 6C and S11 Video ) . Particularly in infected mice , motile cells showed no evident preference for the vicinity of blood vessels . Nearly all the T cell movements were close to horizontal and within a space about 40 μm deep ( Fig 6D ) . Any net migration ( averaged movement ) in this plane was small ( from zero to four cell diameters per hour ) and apparently not in any particular direction , although no migration in the antero-lateral quadrant was observed ( S5 Fig ) . There were occasional vertical movements that extended below this layer; both downward and upward movements were tracked ( Fig 6D , S12 Video and S5 Fig ) , and , on the time scale of the videos ( up to 45 min ) we detected no net movement downwards towards the parenchyma . The mean speed of T cells in uninfected mice was 3 . 4 μm/min , S . D . = 1 . 4 μm/min , N = 3 mice . By 11–25 dpi this had increased to 9 . 8 ± 2 . 6 μm/min ( N = 6 , Fig 6E ) . The fastest brief spurt we noticed was at 30 . 1 μm/min over 40 sec . Measurement of T cell displacement showed that the T cells were not necessarily confined to small territories , as displacements as great as 93 μm were tracked . Median displacement rate was markedly greater in infected mice compared to uninfected mice , as illustrated by the scatter diagrams from individual mice in Fig 6F . To examine the role of antigen presentation to T cells in trypanosomal meningitis , we treated mice with abatacept ( CTLA4Ig , Orencia ) , which inhibits activation of naive T cells , in part by binding to CD80/86 on antigen-presenting cells , such as dendritic cells , and inhibiting their interaction with the co-stimulatory receptor CD28 on T cells [52] . To image both T cells and dendritic cells , we created hCD2 ( DsRed ) xCD11c ( EYFP ) mice . These were injected intraperitoneally with abatacept ( 10 mg/kg ) on alternate days from -1 dpi , and the meninges imaged at 11 or 12 dpi . This treatment markedly reduced the numbers of meningeal T cells and dendritic cells at 11–12 dpi , by factors of about ten ( Fig 7A , 7B and 7C ) . The remaining T cells had a significantly reduced displacement rate ( Fig 7D ) in part as a consequence of a reduction in mean speed from 11 . 64 ± 0 . 34 μm/min ( mean ± SEM , n = 3 mice ) to 5 . 2 ± 1 . 2 μm/min ( n = 3; p = 0 . 007 ) . However , treatment with abatacept had no evident effect on parasitemia ( Fig 7E ) . Nor was there a conclusive effect on the numbers of trypansomes in the dura: trypanosomes were detected in the meninges of 2 of 6 untreated mice , and in one of 6 treated mice ( Fig 7E ) , Thus abatacept greatly reduced and modified the immune response in the dura , but its effect , if any , on the early appearance of trypanosomes there is unresolved . Long-lasting antigen presentation by dendritic cells to T cells , which is inhibited by abatacept , can profoundly modify the phenotype of a developing immune response [53 , 54] . In infected hCD2 ( DsRed ) x CD11c ( EYFP ) mice , visual inspection of videos showed that nearly all contacts lasted less than 2 min , but some longer contacts were observed , as in Fig 8A–8C ( 6 min ) and Fig 8D and 8E ( > 20 min , arrows; S13 Video ) . At least 17 cases of contact lasting more than 15 min were observed ( in a total of 155 min of video record ) . T cells normally respond only to antigens presented on host MHC and in accordance with this rule they were not observed to make contact lasting more than a fraction of a second with trypanosomes ( S9 Video ) . To quantify the interactions of T cells in large populations , including those in mice without fluorescent dendritic cells , we used automated tracking to plot variations in speed against time for sample T cell tracks . Illustrative plots are shown in Fig 8F–8I . In uninfected mice , the plots confirmed that most cells were stationary and a small number moved ( Fig 8F ) . At early infection ( e . g . , Fig 8G ) , more cells were moving . By 27 dpi , some cells were seen to stop for some minutes , then move rapidly ( Fig 8H ) . Treatment with abatacept ( which reduced mean speed and displacement rate—Fig 7E ) also reduced major changes in speed ( Fig 8I ) . The distribution of instantaneous speeds of cells that briefly moved fast between longer periods of arrest will be positively skewed about the mean , so we calculated the skewness [40] for samples of tracks under various conditions . Skewness was positive , not obviously related to dpi , or to the numbers of dural trypanosomes , but correlated significantly with the number of T cells ( Fig 8J; r2 = 0 . 60 , slope different from zero with p = 0 . 008 ) . The results for three mice treated with abatacept lay on the same line ( purple symbols in Fig 8J ) . This graph quantifies the idea that when T cells are numerous they move rapidly between attachments , presumably to dendritic cells . Extravasation of leukocytes normally occurs by diapedesis , which takes several minutes and is preceded by a period of arrest on the vascular endothelium , which may last for hours [21 , 55 , 56] . Fluorescent T cells and dendritic cells were readily observed in meningeal blood vessels ( Fig 9A and 9E ) . T cells were observed to attach to the vascular endothelium and crawl slowly ( Fig 9A–9D ) . CD11c+ cells were not observed to crawl or halt , but , in rare cases , were observed to roll ( Fig 9E and 9F ) . These rolling cells did not have a dendritic shape , and may have been macrophages . For neither cell type did we observe an unambiguous extravasation , but the prolonged interaction sometimes seen between T cells and vascular endothelium suggests that at least some of the extravascular T cells had arrived by classical diapedesis . Previous authors have suggested that trypanosomes arrive in the brain parenchyma either by diapedesis [12 , 17] or by transport in CSF from the choroid plexus [57 , 58] , so we first looked for evidence of similar processes in the dura . Diapedesis of trypanosomes would require interaction lasting at least many seconds with the vascular endothelium , but in 24 C57BL/6 mice ( and a further 12 CD-1 mice included in [20] ) , we never observed a trypanosome slowed or arrested on vascular endothelium ( Fig 10A and S2 Video ) . It is therefore unlikely that trypanosomes extravasated from dural vessels by a diapedesis similar to that of leukocytes . Concerning transport by CSF , it is known that Trypanosoma brucei can appear in CSF , perhaps via the choroid plexus [43 , 59–63] . They might therefore be carried into subarachnoid space [43 , 57 , 64] . However , in the present work , trypanosomes were rare or absent in the subarachnoid space ( Fig 4C and 4D ) so the hypothesis would require that trypanosomes cross the arachnoid membrane [25] from a ( hypothetical ) population at a very low density to accumulate , at a higher density , in the dura , and this is unlikely . Nor did the movement of trypanosomes in the dura suggest that they arrived by migration from some source elsewhere in the cranium , in which case , they would migrate across the imaging field . Instead , the trypanosomes in the dura appeared to be mainly confined within localities less than 30 μm across and were never seen to move smoothly as if carried by a flow of fluid ( Fig 10B and S4 and S5 Videos ) . There were occasional abrupt displacements over longer distances ( arrow in Fig 10C ) , but these occurred in all directions . In a sample of 1901 tracks of trypanosomes recorded over up to 121 sec ( 1000 frames ) in XY scans there was no sign of net migration , the mean component of displacement rate in the rostral direction lying within 95% confidence limits of -0 . 104 and +0 . 305 μm/sec and in the medial direction of -0 . 169 and +0 . 232 μm/sec . In the inflamed dura , proteins can extravasate [65] , presumably by transient , localized , opening of the vascular endothelium [66–69] . Trypanosomiasis is known to increase vascular leakage in the meninges [70 , 71] . In the present work , most of our time-series imaging was done with excitation wavelengths chosen to maximize fluorescence from trypanosomes or lymphocytes rather than from the blood marker . However , plasma extravasation was observed at one vascular site ( with T . b . brucei Lister 427 in a CD-1 mouse ) : blood plasma labeled with fluorescent 70 kDa dextran was released over about 4 sec and a second release occurred 16 sec later . Remarkably , the first release of plasma was accompanied by a trypanosome ( Fig 10D , 10E and 10F and S14 Video ) . We did not observe another such extravasation , but we can estimate that the probability of observing one was small . Choosing , conservatively , a case of unusually rapid invasion , we counted 797 trypanosomes per mm2 in the dura at 11 dpi ( Fig 2B ) . If the invasion had occurred over , say , 48h , then that is one extravasation in 217 sec per mm2 . To count trypanosomes we typically imaged , at most , 8 fields of 0 . 02 mm2 for 12 sec each , in each mouse . We can therefore estimate that , on average , and if we used appropriate imaging conditions , we would capture one extravasation for every 113 mice . These arguments suggest that rapid extravasation might account for the arrival of trypanosomes in the dura , and explain why we observed it only once . The lack of correlation between the numbers in the dura and in the blood at any one time ( Fig 3F ) might be accounted for because the numbers in the dura were the cumulative result of extravasation at times with varying parasitemia . More than 4 , 025 extravascular trypanosomes in the meninges of more than 62 mice were videoed and counted ( including mice and trypanosomes of several strains ) , and more were observed but not videoed . No sign of cell division was observed: neither conjoined trypanosomes nor trypanosomes with more than two concentrations of DNA labeled with furamidine ( in the nucleus and the kinetoplast ) . The number of extravascular trypanosomes in the dura appeared to peak after 30 dpi and then fall to zero ( Fig 3D ) . This means that either trypanosomes moved out of the dura , or were destroyed within the dura , at a rate that could exceed the rate of entry . Automated tracking of trypanosomes suggested they displaced less at later infection times ( Fig 11A and 11B ) . To check this , without relying on automated tracking of the rapidly moving trypanosomes , we visually followed their positions on the videos and noted the co-ordinates of the two extreme positions visited by each trypanosome during the standard imaging period of 12 s . This analysis confirmed that the excursions were fewer and shorter at later stages of infection ( the mean values being 10 . 8 μm , SD = 7 . 7 μm at 36–39 dpi compared to 29 . 2 μm ± 13 . 9 μm at 11–14 dpi , p < 0 . 0001; Fig 11C ) . This progressive reduction in movement , and the decline in the dural population after 30 dpi ( Fig 3D ) , suggest that the dura becomes an unfavorable environment for trypanosomes .
In rats , experimental autoimmune encephalomyelitis ( a model of multiple sclerosis ) causes an increase in T cells outside leptomeningeal blood vessels in the exposed spinal cord [21] , and injection of adjuvant causes an increase in W3/25+ helper T cells in the leptomeningeal layers that adhere to the cortical parenchyma [72] . In a mouse stroke model , observation through the thinned skull revealed T cells outside horizontal , presumably pial , vessels [19] . In contrast , the major increases in the numbers of T cells and CD11c+ cells that we observed in trypanosomiasis were in the dura . Another parasite , the intestinal nematode Nippostrongylus braziliensis , is known to inflame the dura , as indicated by an increase in the mast cell population [32] . Since the dura is separated from the leptomeningeal spaces by the impermeable arachnoid membrane ( [23–25] , Fig 4E ) the distinction may have functional significance . The increase in lymphocyte numbers was greatly reduced by abatacept , which suggests that it required antigen presentation to T cells . This agrees with previous results showing that trypanosomiasis leads to activation of peripheral T cells [5–8] . It is also known that , in trypanosomiasis , lymphocytes can appear in the meninges of the basal brain [42] , in perivascular space [42] and in the CSF [59 , 73] . We did indeed observe small numbers of T cells in spaces filled with CSF: beside pial vessels ( Fig 5E ) or in the perivascular space of descending vessels ( Fig 5F ) , as well as the much greater numbers in the dura . Interactions lasting several minutes were observed between CD11c+ cells and T cells in the dura , suggesting that antigen presentation to T cells also took place in this site . This is supported by the reduction in the frequency of halts made by T cells in the dura after abatacept treatment ( Fig 8I ) . The data are compatible with antigen presentation in the dura occurring only after trypanosomes were present , but do not exclude the contrary . The dura is an interface between the nervous system and the immune system in that peptides released from efferent endings of the trigeminal nerve stimulate mast cells to release mediators , including substance P and histamine , which in turn activate afferent pain fibers [32 , 74] . Whether immune activity in the dura sends signals to the CSF appears to be unknown: conceivably , cytokines or chemokines might cross the arachnoid membrane , or arachnoid cells might be stimulated to release signaling molecules . Because we avoided making a craniotomy , the structure of the meninges was intact and we were able to locate the trypanosomes in the cortical dura . We were unable to confirm the observation of early invasion of the superficial parenchyma made by Frevert et al . [45] in craniotomized mice , except in one case , when the skull was accidentally penetrated . The blood vessels of the dura , like those of the choroid plexus , are relatively leaky compared to those of the subarachnoid space and most of the parenchyma [24 , 75] and it seems unsurprising that both should be sites of early extravasation of trypanosomes [42 , 43 , 76 , 77] . Drug treatments that kill trypanosomes in the periphery , but not in the brain parenchyma , kill trypanosomes in the meninges [20] , in the compartment we now identify as the dura . The number of trypanosomes in the dura of C57BL/6 mice did not continually increase with dpi , and appeared to fall between 30 and 40 dpi . Similarly , in CD-1 mice , low counts in the dura could be found at the later times [20] . Hence the balance of influx of trypanosomes into the dura and their removal , either by emigration or destruction , can shift towards removal . We did not find extravascular trypanosomes below the dura , so have no evidence of migration downwards towards the parenchyma . The arachnoid membrane , with its tight junctions , is expected to be a greater obstacle to trypanosome passage than , for example , the weak barriers between circumventricular organs and blood or CSF [78] . It therefore seems unlikely that trypanosome invasion spreads from the dura to the parenchyma . We did not see evident movement of trypanosomes along dural lymph ducts [30] , and we cannot comment on the possibility that trypanosomes moved back into the blood . However , the observations appear to be compatible with the hypothesis that trypanosomes in the dura are destroyed , this destruction occurring at a time when the dura is densely packed with T cells and CD11c+ cells and the extravascular trypanosomes are moving shorter distances ( Fig 11C ) . Likely contributors to this debilitation are cytokines such as interferon gamma [79] , prostaglandin D2 produced by trypanosomes [80] or neuropeptides from nerve endings or mast cells [81] . Another possibility is impaired energy metabolism . Trypanosomes and activated lymphocytes are crowded into the dural space and both produce ATP mainly by glycolysis , which requires rapid consumption of glucose [82–85] and , by producing pyruvic [86] or lactic acid , could acidify extracellular fluid [87–90] . The red fluorescence sometimes seen in the nucleus and kinetoplast after furamidine injection ( Fig 3C and 3D ) has not been reported in vitro , and might , perhaps , reflect an aberrant pH . A contrary possibility is that physical restriction of movement in the crowded environment of the dura is harmful to trypanosomes . In vitro , simply immobilizing trypanosomes can kill them [91] , and it has been argued that , in vivo , swimming protects trypanosomes against antibodies . Antibodies attached to a freely swimming trypanosome are swept from the forward-pointing flagellum into the flagellar pocket and endocytosed [92] and certain mutations that impair flagellar function [92 , 93] , but not all [94] , are lethal . Imaging through the calvaria was initially developed to study the neural brain ( e . g . [37] ) . Apart from the intrinsic interest of the cortical meninges in many diseases , it is technically a convenient place to study lymphocyte dynamics because the tissue is held rigidly by the skull , is approximately two-dimensional , and can be accessed optically without apparent damage .
In vivo imaging through the skull made it possible to localize and video accumulations of T cells , dendritic cells and trypanosomes in the cortical dura . Apart from using abatacept to show the requirement for T cell activation , we have confined ourselves in the present paper to describing the numbers and movements of these actors . Numerous questions are raised , including how ( or if ) infiltration of the dura by lymphocytes and trypanosomes contributes to neuropathology , and whether there is similar lymphocyte behavior in the dura in meningitis caused by other pathogens . | African trypanosomes are motile parasites that cause sleeping sickness . They multiply first in the blood then cause death mainly by effects on the brain: immune system cells , including T cells and dendritic cells , play major roles in this . Thinking we might see the attack on the brain , we infected mice with trypanosomes and used a two-photon microscope , which allowed us to image the superficial brain and the delicate tissue between the skull and the brain called the meninges without making a hole in the skull . The mice ( which were anesthetized ) had been genetically modified so that T cells and dendritic cells were fluorescent , as were the trypanosomes . We did not notice much happening in the brain itself , but in the meninges , in a compartment called the dura , huge numbers of T cells and dendritic cells appeared . Trypanosomes also moved from the blood into this compartment . Since T cells , dendritic cells and trypanosomes had not been videoed in the meninges before , we began by observing them carefully: their numbers , their movements and their interactions . The accumulation of lymphocytes is a sign of meningitis , a feature of infection by a wide range of pathogens and our results suggest interesting future work . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion",
"Conclusion"
] | [] | 2015 | Intravital Imaging of a Massive Lymphocyte Response in the Cortical Dura of Mice after Peripheral Infection by Trypanosomes |
In ecology , species can mitigate their extinction risks in uncertain environments by diversifying individual phenotypes . This observation is quantified by the theory of bet-hedging , which provides a reason for the degree of phenotypic diversity observed even in clonal populations . Bet-hedging in well-mixed populations is rather well understood . However , many species underwent range expansions during their evolutionary history , and the importance of phenotypic diversity in such scenarios still needs to be understood . In this paper , we develop a theory of bet-hedging for populations colonizing new , unknown environments that fluctuate either in space or time . In this case , we find that bet-hedging is a more favorable strategy than in well-mixed populations . For slow rates of variation , temporal and spatial fluctuations lead to different outcomes . In spatially fluctuating environments , bet-hedging is favored compared to temporally fluctuating environments . In the limit of frequent environmental variation , no opportunity for bet-hedging exists , regardless of the nature of the environmental fluctuations . For the same model , bet-hedging is never an advantageous strategy in the well-mixed case , supporting the view that range expansions strongly promote diversification . These conclusions are robust against stochasticity induced by finite population sizes . Our findings shed light on the importance of phenotypic heterogeneity in range expansions , paving the way to novel approaches to understand how biodiversity emerges and is maintained .
The dynamics and evolutionary history of many biological species , from bacteria to humans , are characterized by invasions and expansions into new territory . The effectiveness of such expansions is crucial in determining the ecological range and therefore the success of a species . A large body of observational [1 , 2] and experimental [3–6] literature indicates that evolution and selection of species undergoing range expansions can be dramatically different from that of other species resident in a fixed habitat . Theoretical studies of range expansions based on the Fisher-Kolmogorov equation [7 , 8] or variants [9–11] also support this idea . Adaptive dispersal strategies [2] and small population sizes at the edges of expanding fronts [12 , 13] are among the main reasons for such difference . Range expansions often occur in non-homogeneous and fluctuating environments . Under such conditions , it is possible to mathematically predict the expansion velocity of a community of phenotypically identical individuals [14–19] . However , diversity among individuals is expected to play an important positive role when populations expand in fluctuating environments . For instance , diverse behavioral strategies help animal populations to overcome different invasion stages and conditions [20–23] . Analyses of phenotypic diversity in motile cells suggest that it also may lead to a selective advantage at a population level [24–26] . Although several studies have tackled the problem of how individual variability affects population expansion [6 , 9 , 10 , 27–31] , systematic and predictive theory is still lacking [23] . Phenotypic diversification is often interpreted as a bet-hedging strategy , spreading the risk of uncertain environmental conditions across different phenotypes adapted to different environments [32–41] . Since its formalization in the context of information theory and portfolio diversification [42 , 43] , a large number of works have explored the applicability of bet-hedging in evolutionary game theory [44–47] and ecology [48–52] . Few studies have explored the benefits of bet-hedging in spatially structured ecosystems [53–55] . In this paper , we study how bet-hedging strategies can aid populations in invading new territories characterized by fluctuating environments . In particular , we analyze the effect of spatial expansion , different types of environmental heterogeneity , and demographic stochasticity on development of bet-hedging strategies for a population front evolving according to a Fisher wave . By employing mathematical as well as extensive computational analyses , we find that the advantage of bet-hedging in range expansions depends on the rate of environmental variation . In particular , bet-hedging is more convenient for infrequently varying environments , whereas its advantages vanish for frequent environmental variation . For the same model , bet-hedging is never an advantageous strategy in the well-mixed case , supporting the view that range expansions strongly promote diversification . We further find that spatial environmental variations provide more opportunities for bet-hedging than temporal fluctuations . Finally , we show that our conclusions still hold when considering stochastic effects on the front propagation induced by a finite population size . The paper is organized as follows . We introduce a general population model and an example with two available phenotypes and two environmental states . We present an extensive study of this example . We then generalize the main conclusions obtained for the example for a case with an arbitrary number of environmental states , and then with also an arbitrary number of strategies . We conclude with a discussion and future perspectives .
We consider a population consisting of individuals that can assume N alternative phenotypes . The population as a whole adopts a phenotypic strategy , that is identified by the fractions αi , i = 1…N of the population assuming each phenotype i with ∑i αi = 1 and 0 ≤ αi ≤ 1 ∀i ( Fig 1A ) . As customary in game theory , we say that a strategy is a “pure strategy” if αi = δik for some phenotype k , and a “mixed strategy” otherwise . We assume that the αi’s remain constant in time within the population . The environment can be found in one of M different states , which can randomly alternate either in time or in space . We call pi the probability of encountering environment i . We further define the growth rate sij ≥ 0 of phenotype j in environment i ( Fig 1A ) . When the population size is sufficiently large , so that demographic stochasticity can be neglected , the population-averaged growth rate given the state i = i ( x , t ) of the environment at position x and time t is σ i = ∑ j α j s i j . ( 1 ) Since Eq ( 1 ) is linear in the αj’s , the population-averaged growth rate in a given environment is always maximized by the pure strategy with the highest growth rate . However , in the presence of uncertainty about the environment , the population might choose other strategies . One possibility is to select a different pure strategy , less risky than the optimal one . This case is often termed “conservative bet-hedging” in the ecological literature [41] . Another option is to adopt a mixed strategy , with different phenotypes more adapted to different environments . This case is termed “diversifying bet-hedging” in the literature [41 , 56] . Since our interest is in diversification , the term “bet-hedging” will refer herein to diversifying bet-hedging . Before presenting our results in full generality , weconsider a simple , yet ecologically relevant instance of the model with only two phenotypes: “safe” and “risky” and two environmental states: “adverse” ( a ) and “favorable” ( b ) . The safe phenotype is characterized by a growth rate ss both in the adverse and favorable environments . The growth rate of the risky phenotype is sa in environment ( a ) and sb in environment ( b ) ( Fig 1B ) [57] . The two environments occur with the same probability , pa = pb = 1/2 . A fraction of individuals α adopts the risky phenotype and the complementary fraction ( 1 − α ) adopts the safe phenotype ( Fig 1B ) . For this model , the population-averaged growth rate reads σ ( x , t ) = { σ a = ( 1 - α ) s s + α s a , in env . a σ b = ( 1 - α ) s s + α s b . in env . b ( 2 ) Note that , with a slight abuse of notation , we use equivalently σi or σ ( x , t ) to denote the population-averaged growth rate in the environment i ( x , t ) . For pure strategies , α = 0 or α = 1 , the population-averaged growth rate σ reduces to the growth rate of the safe or risky phenotype , respectively .
We seek to understand those conditions under which bet-hedging is advantageous for the population . To this end , we shall compare three situations: i ) well-mixed populations , ii ) range expansions in environments that fluctuate temporally , but that are homogeneous in space ( Fig 1C ) , and iii ) range expansions in spatially fluctuating environments that are homogeneous in time ( Fig 1D ) . In this section , we generalize our results to a model with two strategies , but an arbitrary number i = 1…N of environmental states . Let us start with the temporally varying case . Following the usual logic , the mean velocity for k → 0 reads v m = 2 D ⟨ σ ( α ) ⟩ = 2 D ∑ i p i α s i 1 + ( 1 - α ) s i 2 ( 12 ) where si1 and si2 are the growth rates of the two strategies in environment i . The first derivative of the mean velocity respect to α reads ∂ v m ∂ α = D ∑ i p i s i 1 - s i 2 α s i 1 + ( 1 - α ) s i 2 ( 13 ) Since v ( α ) is a concave function , the condition for having a bet-hedging strategy , i . e . a maximum in the interior of the interval ( 0 , 1 ) is ∂ v m ∂ α | α = 0 = D ∑ i p i s i 1 - s i 2 s i 2 > 0 and ∂ v m ∂ α | α = 1 = D ∑ i p i s i 1 - s i 2 s i 1 < 0 . ( 14 ) These conditions reduce to the Eq ( 7 ) in the limiting case of the two-environment model . With a similar strategy we can compute the limits of the bet-hedging region also for the spatially varying case . In this case we have v M = 2 D ∑ i p i α s i 1 + ( 1 - α ) s i 2 ( 15 ) and therefore ∂ v M ∂ α = 2 D ( ∑ i p i α s i 1 + ( 1 - α ) s i 2 ) 2 ∑ j p j ( s j 1 - s j 2 ) ( α s j 1 + ( 1 - α ) s j 2 ) 2 . ( 16 ) To determine the bet-hedging region we follow the same logic as in the temporally varying case , yielding ∂ v M ∂ α | α = 0 = 2 D ( ∑ i p i s i 2 ) 2 ∑ j p j ( s j 1 - s j 2 ) s j 2 2 ∂ v M ∂ α | α = 1 = 2 D ( ∑ i p i s i 1 ) 2 ∑ j p j ( s j 1 - s j 2 ) s j 1 2 . ( 17 ) so that the condition in this case reads ∑ j p j ( s j 1 - s j 2 ) s j 2 2 > 0 and ∑ j p j ( s j 1 - s j 2 ) s j 1 2 < 0 . ( 18 ) Even in this case , the bet-hedging region is broader in the spatially-fluctuating than in the temporally-fluctuating case . This fact is proven in full generality in the next subsection . In this Section , we demonstrate that our main conclusions hold in full generality for arbitrary numbers of phenotypes N and environmental states M ( see Section Model ) . In particular , for a temporally fluctuating environment in the limit of very slow switching rates , the bet-hedging regime occupies a reduced region of parameter space compared to temporally constant environments fluctuating slowly in space . Also in this case , we find that for frequent environmental change , the propagation velocity tends to v M ≈ 2 D 〈 σ 〉 , regardless of whether the environmental fluctuations depend on time or space . Therefore , the optimal strategy maximizes the linear function of the αis 〈σ〉 and is therefore a pure strategy as discussed after Eq ( 1 ) . We consider a range expansion where the environment fluctuates in time and the stochastic switching rates among the M environmental states are small . Following the same line of thought of the two-strategy , two-environment model , the optimal strategy maximizes σ T = v M ( k → 0 ) 2 D = ∑ i p i σ i ( 19 ) where σi = ∑j sij αj . For spatially varying environments , the optimal strategy maximizes the harmonic mean σ S = v F ( k S → 0 ) 2 D = 1 ∑ i p i 1 σ i . ( 20 ) Both for Eqs ( 19 ) and ( 20 ) , maximization has to be performed with the constraint ∑j αj = 1 and 0 ≤ αj ≤ 1 ∀j . We recall that the bet-hedging regime is the region of parameter space where the optimal solution is a mixture of all phenotypes , αi > 0 ∀i . Here we show that if , for a given choice of the sij’s and pi’s , a population advancing in a temporally varying environment is in a bet-hedging regime , then the same holds for spatially varying environments . For the demonstration , we borrow a mathematical tool from evolutionary game theory [71] . We introduce the gradients F l T = ∂ σ T ∂ α l = ⟨ s l 2 σ ⟩ F l S = ∂ σ S ∂ α l = ( σ S ) 2 ⟨ s l 2 σ 3 / 2 ⟩ ( 21 ) where 〈x〉 = ∑i pi xi is the average over environments . We now associate replicator equations to Eqs ( 19 ) and ( 20 ) : d d t α l = α l ( F l T - F ¯ T ) = α l ⟨ s l - σ 2 σ ⟩ ( 22 ) d d t α l = α l ( F l S - F ¯ l S ) = α l ( σ S ) 2 ⟨ s l - σ 2 σ 3 / 2 ⟩ . ( 23 ) The system is in a bet-hedging regime when the replicator equations admit a stable fixed point in the interior of the unit simplex , 0 < αi < 1 . Instead of computing the fixed point explicitly , we check whether each phenotype l has a positive growth rate for αl ≪ 1 . Brouwer’s fixed point theorem ensures that , under this condition , there must be a fixed point in the interior ( see [71] , chapter 13 ) . For our aims , it is therefore sufficient to prove that , for small αl , if ( F l T - F ¯ T ) is positive , then ( F l S - F ¯ S ) must be positive as well . Note that for αl ≪ 1 , the average σ = ∑j sij αj does not depend on αl , and therefore , σ and sl are uncorrelated random variables respect to the average over the environment . Since σ > 0 , this means that the sign of ( F l T - F ¯ T ) is the same than the quantity 1 ⟨ σ ⟩ ⟨ s l ⟩ ⟨ 1 σ ⟩ - 1 . ( 24 ) Following the same logic , the sign of ( F l S - F ¯ S ) is the same than ⟨ s l ⟩ ⟨ 1 σ 3 / 2 ⟩ - ⟨ 1 σ ⟩ = ⟨ 1 σ ⟩ ( ⟨ s l ⟩ ⟨ 1 / σ 3 / 2⟩⟨ 1 / σ ⟩ - 1 ) . ( 25 ) This means that , in the general case , the bet-hedging region is defined by the conditions temporally varying case : 1 ⟨ σ ⟩ ⟨ s l ⟩ ⟨ 1 σ ⟩ - 1 > 0 ∀ l spatially varying case : ⟨ s l ⟩⟨ 1 / σ 3 / 2 ⟩⟨ 1 / σ ⟩ - 1 > 0 ∀ l . ( 26 ) We now turn to the demonstration that the bet-hedging region in the spatially varying case is always broader than in the temporally varying case . Since 〈sl〉 > 0 , we need to demonstrate that the following inequality always holds ⟨ 1 / σ 3 / 2 ⟩⟨ 1 / σ ⟩ ≥ ⟨ 1 σ ⟩ 1 ⟨ σ ⟩ . ( 27 ) This can be proven from the chain of inequalities ⟨ 1 / σ 3 / 2 ⟩⟨ 1 / σ ⟩ ≥ ⟨ 1 σ ⟩ ≥ ⟨ 1 σ ⟩ ⟨ 1 σ ⟩ ≥ ⟨ 1 σ ⟩ 1 ⟨ σ ⟩ . ( 28 ) In Eq ( 28 ) , the second and third inequalities are consequences of Jensen’s inequality , since both x2 and 1/x are convex functions . For the first inequality in Eq ( 28 ) , since s > 0 , we can use the result 〈xi〉 ≥ 〈xj〉i/j proved for i > j in [72] . Combining this result for ( i = 3 , j = 2 ) and ( i = 2 , j = 1 ) , we obtain 〈x3〉 ≥ 〈x2〉〈x〉 . Taking 〈 x 〉 = 〈 1 / σ 〉 we finally prove Eq ( 28 ) . Therefore , in the limit of small switching rates of the environment , the bet-hedging region is wider in the spatially varying case than in the temporally varying case . In the opposite limit of high rates of environmental switch , the function to be optimized is linear , and the optimal strategy is a pure strategy , i . e . the bet-hedging region shrinks to a set of measure zero . In this case , the particular phenotype l adopted by the whole population is that maximizing ∑i pi sil . This conclusion holds both for temporally and spatially varying environments .
Understanding the precise mechanisms of population expansions is of utmost importance , not only for understanding species diversity , but also to cope with invasive species in new habitats [20–23] , bacterial infections [24–26 , 73] , and cell migration , such as those occurring during tissue renewal or cancer metastasis [5] . Phenotypic diversity is a convenient strategy for the success of population expansions in a broad range of contexts [20–26] . Although precise experimental measures are not easy to obtain , a recent study shows that populations with increased variability in individual risk-taking can colonize wider ranges of territories [74] . In this work , we proposed a general mathematical and computational framework to analyze such scenarios . In particular , we introduced a population model with diverse phenotypes that perform differently depending on the type of environment . We focused on the “optimal” degree of diversity leading to the fastest average population expansion in an environment fluctuating either in space or in time . We found that , contrarily to the well-mixed case , bet-hedging can be convenient in expanding populations . This result complements the study in [53] for a fixed habitat and supports the view that diversification is of broad importance for spatially-structured populations . For environments varying slowly in time , the expansion is relatively slow , and diverse communities can be optimal depending on the parameters . On the contrary , for fast environmental changes , the optimal population always adopts a unique strategy . A remarkable outcome of our analysis is that spatial fluctuations create more opportunities for bet-hedging than temporal fluctuations , in that the region of parameter space where the optimal population is diverse , is always larger in the former case . One intuitive explanation is that in the case of spatial fluctuations , the population spends less time traversing favorable patches than adverse ones . This means that the beneficial effect of favorable patches is reduced with respect to the case of temporal fluctuations . Therefore , a pure risky strategy is less efficient in the case of spatial variability and can be more easily outcompeted by a diversified bet-hedging strategy . The framework presented here can be extended to accommodate other scenarios . We have assumed that the fraction of individuals adopting each phenotype is fixed by the phenotypic switching rates . To understand the evolution of bet-hedging , it could be interesting to study scenarios in which the phenotypic switching rates are slower , so that phenotypes can be selected , and/or are themselves subject to evolution and selection [57 , 70] . Another potentially relevant extension would be to consider two-dimensional habitats . Although the classic theory for Fisher waves [7 , 8] is unaffected in higher dimensions , in the presence of spatial heterogeneity the front shape can become anisotropic , potentially affecting the results . Similarly , it would be interesting to analyze the combined effect of spatial and temporal variability . We also limited ourselves to the case where the different environments affect individual growth rates , whereas in general , one could also expect them to have an effect on motility [14 , 15 , 75–77] , opening the way for different forms of bet-hedging . Finally , the present study was limited to pulled waves . It would be interesting to study the effect of bet-hedging on pushed waves , for example to describe population expansion in the presence of an Allee effect [78 , 79] . It would be also interesting to experimentally test our results . Experiments of expanding bacterial colonies in non-homogeneous environments have already been performed and shed light , for example , on the evolution of antibiotic resistance in spatially-structured populations [80] . To perform experiments within the limits of our theory , a challenge can be to maintain the environmental variability sufficiently low to avoid exposing the population to an excessive evolutionary pressure . Similar problems appear , for example , in studies of range expansion of mutualistic bacteria [81] . An extension of the theory including both phenotypic and genetic diversity could account for these scenarios . In summary , we have introduced a model to understand conditions favoring diversification of an expanding population . Our work provides a bridge between the theory of bet-hedging and that of ecological range expansion described by reaction-diffusion equations . The results of the model highlight the relation between population diversity and fluctuations of the environment encountered during range expansion . The flexibility and generality of our framework make it a useful starting point for applications to a wide range of ecological scenarios . | Ecological populations are often exposed to unpredictable and variable environmental conditions . A number of strategies have evolved to cope with such uncertainty . One of them is stochastic phenotypic switching , by which some individuals in the community are enabled to tackle adverse conditions , even at the price of reducing overall growth in the short term . In this paper , we study the effectiveness of these “bet-hedging” strategies for a population in the process of colonizing new territory . We show that bet-hedging is more advantageous when the environment varies spatially rather than temporally , and infrequently rather than frequently . | [
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
] | [
"bacteriology",
"organismal",
"evolution",
"species",
"colonization",
"ecology",
"and",
"environmental",
"sciences",
"invasive",
"species",
"applied",
"mathematics",
"population",
"genetics",
"microbiology",
"mathematics",
"microbial",
"evolution",
"population",
"biology",
"ecological",
"metrics",
"evolutionary",
"theory",
"species",
"diversity",
"game",
"theory",
"population",
"metrics",
"population",
"size",
"ecology",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology",
"numerical",
"integration",
"bacterial",
"evolution",
"numerical",
"analysis"
] | 2019 | Bet-hedging strategies in expanding populations |
Co-expression network analysis provides useful information for studying gene regulation in biological processes . Examining condition-specific patterns of co-expression can provide insights into the underlying cellular processes activated in a particular condition . One challenge in this type of analysis is that the sample sizes in each condition are usually small , making the statistical inference of co-expression patterns highly underpowered . A joint network construction that borrows information from related structures across conditions has the potential to improve the power of the analysis . One possible approach to constructing the co-expression network is to use the Gaussian graphical model . Though several methods are available for joint estimation of multiple graphical models , they do not fully account for the heterogeneity between samples and between co-expression patterns introduced by condition specificity . Here we develop the condition-adaptive fused graphical lasso ( CFGL ) , a data-driven approach to incorporate condition specificity in the estimation of co-expression networks . We show that this method improves the accuracy with which networks are learned . The application of this method on a rat multi-tissue dataset and The Cancer Genome Atlas ( TCGA ) breast cancer dataset provides interesting biological insights . In both analyses , we identify numerous modules enriched for Gene Ontology functions and observe that the modules that are upregulated in a particular condition are often involved in condition-specific activities . Interestingly , we observe that the genes strongly associated with survival time in the TCGA dataset are less likely to be network hubs , suggesting that genes associated with cancer progression are likely to govern specific functions or execute final biological functions in pathways , rather than regulating a large number of biological processes . Additionally , we observed that the tumor-specific hub genes tend to have few shared edges with normal tissue , revealing tumor-specific regulatory mechanism .
Gene co-expression network analysis is a useful tool for studying the complex regulatory machinery in organisms [1][2][3][4] . When the gene expression profiles under multiple conditions are available , comparing co-expression networks across conditions could reveal co-expression patterns that are common across conditions and those that are unique to a condition [5][6][7][8][9] , providing insights on how genes work together to regulate biological processes under different conditions . It has been demonstrated that complex diseases are likely to be regulated by condition-specific mechanisms while condition-specific hub genes are likely to be drug targets [10][11][12] . The Gaussian graphical model and its variants have been widely used for studying biological networks [13][14][15][16][17][18][19] . This method models the joint distribution of a set of variables and characterizes the conditional dependence between each pair of variables given all the other variables through the precision matrix ( a . k . a . inverse covariance matrix ) of the joint distribution [20] . Unlike co-expression models based on marginal correlation , e . g . WGCNA [21] , which do not distinguish the direct and indirect ( e . g . through intermediate genes ) relationship between genes , the direct relationship between a pair of genes can be inferred from the conditional independence estimated from the Gaussian graphical model . Many algorithms have been proposed to obtain a sparse estimate for the precision matrix , for example , graphical lasso [22] and neighborhood selection [23] . These algorithms make it possible to construct gene co-expression networks using graphical models . A graph generated from this estimate , where genes are represented as nodes and entries in the estimated precision matrix as edges , provides a useful tool for visualizing the relationships between genes and for generating biological hypotheses . In a multi-condition gene expression study , the co-expression profiles across conditions typically are related , for example , due to shared pathways in different tumor subtypes , or common regulatory mechanisms for housekeeping genes in different tissues . A joint analysis that borrows information across conditions potentially can reveal common structures and increase the power of statistical inference , which is especially useful when the sample sizes are small . Recently , several methods have been proposed to jointly analyze multiple graphical models . Meinshausen et al . [24] incorporated a non-convex hierarchical group lasso penalty into the graphical lasso to encourage common 0’s ( i . e . absence of edges ) in the precision matrix across conditions . Danaher et al . [6] proposed a joint graphical lasso model by adding an additional convex penalty to the graphical lasso objective function . They proposed two choices for the convex penalty: a group penalty that encourages a shared pattern of sparsity and a fused lasso penalty that encourages similarities in both network sparsity and edge weights . Despite their differences , these methods encourage similarities equally across all edges and all conditions . This inherently assumes that the similarity across conditions is similar for all edges and that the precision matrices in all conditions are equally similar to each other . For gene co-expression networks across different conditions , however , both assumptions are violated due to the heterogeneity across genes and across conditions . First , edges in the networks often have different levels of conservation across conditions . For example , in a network consisting of multiple pathways , the pathways involving basic cellular functions tend to be more conserved across tissues than those involving tissue-specific functions . Second , when there are multiple conditions , some conditions may be more similar to each other than others . For example , tissues with the same embryonic origin may have more similar pathways than those with different origins . More recently , several methods have been proposed to allow more structural heterogeneity in joint estimation . Zhu et al . [25] introduced a non-convex truncated l1 penalty on the pairwise differences between the precision matrices to encourage elementwise clustering of similar entries across conditions . To incorporate external information on shared subgraphs across conditions , Ma et al . [26] grouped edges shared across conditions based on external information and extended the neighborhood selection method to a joint analysis with the proposed penalty . To handle heterogeneity in similarities across conditions , Seagusa et al . [27] proposed a Laplacian shrinkage penalty to incorporate the pairwise distance between conditions , and proposed using hierarchical clustering to obtain the pairwise distance when it is unknown a priori . While these methods improve the flexibility in estimation , they do not completely address the issues in studying condition-specific co-expression networks . For example , though the approach in Zhu et al . [25] allows abrupt elementwise difference across conditions , it still implicitly assumes that the majority of edges are common across conditions and penalizes condition-specificity . The approach in Ma et al . [26] relies on the availability and the quality of external information , which is still limited for gene co-expression relationships . The approach in Seagusa et al . [27] uses external information or hierarchical clustering to define the weighted subpopulation network and only partially addresses the issue of condition specificity . In this work , we propose an adaptive approach to simultaneously addressing condition specificity and heterogeneity across conditions in the estimation of multiple co-expression networks . Our strategy is to incorporate a binary weight matrix that contains information on whether or not an edge is common between conditions in the fused graphical lasso framework . We propose a strategy to learn this matrix adaptively from the data based on a test for differential co-expression , though it can also be obtained from external sources . The incorporation of this matrix not only accounts for the difference between condition-common edges and condition-specific edges but also makes the estimation adaptive to the distance between different conditions . In this way , one can borrow information across conditions for common edges , while estimating differential edges in a condition-specific manner . We provide a computationally efficient implementation using the alternating direction method of multipliers ( ADMM ) algorithm . Our simulations show that this method generates more accurate results in both edge detection and edge weight estimation . We applied our method to a rat multi-tissue dataset and a TCGA breast cancer dataset ( TCGA BRCA ) and obtained interesting biological insights .
We first briefly describe the Graphical Lasso ( GL ) method [22] and the Fused Graphical Lasso ( FGL ) method [6] . Suppose the gene expression profiles are available across K conditions , where conditions are , for example , different tissues or disease statuses . Denote the gene expression levels Y ( k ) for the condition k , k = 1 , 2 , … , K , as a nk × p matrix , where p is the number of genes , which is common across all conditions , and nk is the number of observations , which can vary across conditions . Suppose that gene expression levels within each condition , y1 ( k ) , y2 ( k ) …ynk ( k ) ∈Rp , are identically drawn from N ( μk , Σk ) , where μk∈Rp and Σk is a positive definite p × p matrix . Then zero entries in the precision matrix Σk−1 correspond to the pairs of genes that are conditionally independent given all other genes in the dataset . Based on the precision matrix Θ ( k ) ≡Σk−1 , a gene co-expression network can be constructed by representing the genes as nodes and conditional dependencies as edges in a graph . The most direct way to analyze such data is to estimate K individual graphical models separately . We can use the graphical lasso method to compute a separate l1 penalized estimator of Σk−1 for each condition by solving , maximize{Θ ( k ) } ( log{det ( Θ ( k ) ) }−tr ( S ( k ) Θ ( k ) ) −λk‖Θ ( k ) ‖1 ) , ( 1 ) where S ( k ) = ( Y ( k ) ) TY ( k ) /nk is the empirical covariance matrix of Y ( k ) , λk‖Θ ( k ) ‖1 is a penalty term with non-negative tuning parameter λk and ‖Θ ( k ) ‖1 is the L1 norm of Θ ( k ) . However , when the conditions are related , separate estimation ignores the common structure shared across conditions and can also mask differences critical in understanding condition-specificity in the co-expression pattern . To address this issue , Danaher et al . [6] developed a fused graphical lasso model to jointly estimate multiple graphical models from related conditions . This model incorporates the generalized fused lasso penalty P ( {Θ} ) [28] to the log-likelihood , l ( {Θ} ) =∑k=1Knk[log{det ( Θ ( k ) ) }−tr ( S ( k ) Θ ( k ) ) ]−P ( {Θ} ) , ( 2 ) such that information can be borrowed across conditions . The penalty P ( {Θ} ) is a convex penalty with two terms , P ( {Θ} ) =λ1∑k=1K∑i≠j|θij ( k ) |+λ2∑k<k′K∑i≠j|θij ( k ) −θij ( k′ ) | , ( 3 ) where λ1 and λ2 are non-negative tuning parameters , and θij ( k ) is the ( i , j ) -th element of the matrix Θ ( k ) . The first term , which is the lasso penalty in GL [23][22] , is applied to the off-diagonal entries of the K precision matrices to encourage sparsity . The second term , which is the fused lasso penalty [28] , is applied to the differences between elements of each pair of precision matrices to encourage similarity between conditions . A large λ2 leads to similar edge patterns across conditions . It has been shown that FGL outperforms GL when conditions are related [6] . While borrowing strength across conditions is helpful for enlarging effective sample sizes , differences in co-expression patterns are present between different conditions . For example , if one were studying tumor-specific co-expression by analyzing two subtypes of tumor tissues and a normal tissue jointly ( Fig 1 ) , some edges in the co-expression networks may be common across all three conditions , while others may be specific to one condition or both tumor subtypes . A primary interest of the study would be to identify the tumor-specific or subtype-specific edges . If FGL is used to construct the co-expression network , it would encourage similarities among all edges across all conditions equally by imposing a constant penalty parameter . This has two drawbacks . First , it does not distinguish between shared edges and those unique to a condition , thus condition-specificity of edges is not preserved . Second , it imposes an equal amount of similarities across all pairs regardless of whether the pair consists of two tumor subtypes or a tumor tissue and a normal tissue . This is problematic as the two tumor subtypes are likely to be more similar to each other than to the normal tissue . To address these issues , we extend the fused graphical lasso method to incorporate condition-specificity in the integration of networks across conditions . Our strategy is to add a binary screening matrix W ( kk′ ) to the fused lasso penalty as follows , P ( {Θ} ) =λ1∑k=1K∑i≠j|θij ( k ) |+λ2∑k<k′K∑i≠jwij ( kk′ ) |θij ( k ) −θij ( k′ ) | , ( 4 ) where wij ( kk′ ) is the ( i , j ) -th element of W ( kk′ ) with wij ( kk′ ) ={1 , ifθij ( k}andθij ( k′ ) arenondifferentialbetweenconditionskandk′0 , ifθij ( k ) andθij ( k′ ) aredifferentialbetweenconditionskandk′ The matrix W ( kk′ ) controls whether similarity should or should not be encouraged between each pair of condition for each edge . It allows different edges to be penalized differently , and also allows the penalties for different pairs of conditions to vary according to the distance between the conditions . In doing so , one can borrow strength across conditions for estimating common edges , while allowing differential edges to be estimated in a condition-specific way . Therefore , we call our method condition-adaptive fused graphical lasso ( CFGL ) . Fig 2 illustrates the workflow of our method . The screening matrix can be obtained using prior knowledge , learning directly from the data , or a combination of both strategies . To determine the screening matrix using prior knowledge , one may extract information on co-expression regulation from public databases , such as the KEGG pathway database [29] , COXPRESdb [30] or MSigDB [31] . For example , if a pathway is known to be conserved across tissues [5][32] , one may specify the corresponding elements in W ( kk′ ) as 1 to reflect the conservation of co-expression regulation . However , it is difficult to construct the entire screening matrix solely based on prior information , because the gene relationships in the databases often are not provided in a condition-specific way ( e . g . not available for a specific disease type ) and the genes of interest may not be included . Therefore , we propose a data-driven strategy to estimate the screening matrix W ( kk′ ) from the data . As W ( kk′ ) reflects the status of differentiation between a pair of conditions , we determine W ( kk′ ) by identifying differential entries between the precision matrices of the two conditions , Σk−1 and Σk′−1 , through a hypothesis test . If the test determines that the entry ij is differential , we set wij ( kk′ ) =0 , otherwise we set wij ( kk′ ) =1 . As wij ( kk′ ) is binary , this approach is equivalent to using a l0 penalty to determine the support of the condition-specific edges . It is somewhat analogous to the Sure Independent Screening procedure for feature selection methods such as the lasso , Dantzig selector , and SCAD [33] , where an elementwise screening is first performed to reduce the dimension from ultra-high to moderate before variable selection . Here we test for differentiation using the test proposed by Xia et al . [34] . This method tests for a difference between a pair of precision matrices and reports differential entries in the precision matrices with proper false discovery rate ( FDR ) control . It directly estimates the difference between precision matrices , bypassing the estimation of the individual precision matrices . Other tests for differential entries are available[35][36] , but we selected Xia's test as it has been shown to provide more accurate estimates than the tests that require separate estimation of precision matrices due to leveraging information on the sparsity of the difference between precision matrices [37] . To avoid falsely imposing similarity for edges that are moderately differential , we use a relaxed FDR threshold in the test to encourage similarity only to the edges that are obviously non-differential across conditions . Similar to FGL and other penalty-based methods , this model can be estimated using the ADMM algorithm . We used BIC to guide the selection of penalty parameters . In the real data application , when the sample size is reasonably large to afford subsampling , we performed an additional stability selection [38] step . Instead of constructing networks using all the samples , the stability selection procedure constructs networks for a large set of subsamples generated from the original data and keeps only the edges that frequently occur across subsamples to obtain robust edges . Details on the stability selection procedure can be found in Methods . We used simulation studies to evaluate the performance of our method and compare it to FGL and GL . We first considered the two-condition scenario , evaluating the performance of these methods at different levels of differentiation between conditions . Then , we increased the complexity by introducing a third condition and allowing the level of differentiation to vary across all three conditions . In the first set of simulations , we generated the gene expression profiles from a co-expression network of 400 genes for 2 conditions . The network consists of 8 co-expression modules , each of 50 genes . To generate different levels of differentiation between conditions , we simulated four scenarios ( S1-S4 ) with a progressively increasing number of differential edges between conditions , where the networks in the two conditions are identical in S1 and are different at various levels in S2-S4 . While S1 is extremely rare in practice , it exactly follows the model assumptions of FGL and thus illustrates the methods performance under conditions ideal to FGL . Two samples size ( 50 , 100 ) are considered for each scenario . To simulate a network , we first simulated its constituent modules . To create different levels of differentiation , three types of modules were simulated: ( 1 ) identical network structure and identical edge weights between conditions ( II ) , ( 2 ) identical network structure but different edge weights between conditions ( ID ) , and ( 3 ) different network structures and different edge weights between conditions ( DD ) . We then combined these modules in various configurations to achieve the desired level of differentiation for the networks in different scenarios . In all scenarios , the 8 modules are evenly split into two groups , each of which consists of 4 modules of the same type . The configurations of modules in these scenarios are summarized in Table 1 . Detailed information on the data generating process are in Methods . For each simulation , we constructed the co-expression network using our method , FGL , and GL . To evaluate how the accuracy of the estimated screening matrix affects the performance of our method , we also included a version of CFGL with the true screening matrix , which is labeled as CFGL-oracle ( CFGLO ) ( see the Methods section ) . We compared the performance of these methods based on the estimation of network topology and edge weight across a grid of λ1 and λ2 . The accuracy of estimated network structure was evaluated according to the network topology , i . e . the presence or absence of edges . Specifically , we compared the estimated network topology with the true topology , then computed the sensitivity and specificity of the edge detection . If an edge is present in the true network but missed in the estimated one ( i . e . estimated edge weight = 0 ) , then it was counted as a false negative . If an edge is absent in the true network but identified in the estimated one ( i . e . estimated edge weight>0 ) , then it was counted as a false positive . The accuracy of edge weight estimation was assessed by computing the sum of squared error ( SSE ) between the estimated edge weight and the true precision matrix . We plotted the ROC curve for edge detection and the SSE for edge weight estimation at a varying level of λ1 with λ2 fixed at the value that achieves the minimal BIC value ( λ2 = 0 . 15 for n = 50 and λ2 = 0 . 10 for n = 100 ) . Because edges detected at a low false positive rate are of primary interest , we computed the partial area under the curve ( pAUC ) from the ROC curve for the range of FPR < 0 . 05 . Fig 3 shows the results at n = 50 . In all scenarios , our approach ( CFGL ) had a higher partial AUC ( pAUC: S1: 0 . 711 , S2: 0 . 671 , S3: 0 . 688 , S4: 0 . 620 ) than GL ( pAUC: S1: 0 . 583 , S2: 0 . 594 , S3: 0 . 588 , S4: 0 . 590 ) , and also a lower SSE . The gain is more apparent when the two conditions are relatively similar ( S1-S3 ) . This is because data integration improves the accuracy of edge detection , especially when networks are similar between conditions . The advantage is especially obvious when n = 50 ( S1 Table and S2 Fig for n = 100 ) , as this small sample size is likely not enough to support accurate estimation with GL based on the samples from a single condition . FGL performs well in these scenarios too ( pAUC: S1: 0 . 714 , S2: 0 . 609 , S3: 0 . 660 ) ; however , when the two conditions are fairly different ( S4 ) , FGL performs worse than GL ( pAUC: 0 . 578 vs . 0 . 590 ) . Compared with FGL , our method has a higher AUC and an apparently lower SSE in all scenarios with between-condition differences ( S2-S4 ) . Even in the scenario without between-condition differences ( S1 ) , i . e . the ideal setting for FGL , our method is still competitive: it has an almost identical ROC curve as FGL and a slightly higher SSE than FGL . In practice , it is much more common to encounter S2-S4 than S1 , as the networks of two different conditions are likely to be different . CFGLO has the best AUC ( pAUC: S1: 0 . 714 , S2: 0 . 717 , S3: 0 . 715 , S4: 0 . 723 ) and SSE among all the methods , suggesting that the performance of CFGL can be further improved by improving the estimation of screening matrix , for example , by incorporating external information . We also reported simulation results under several other λ2’s . The results are similar and can be found in S1 Table . Next , we allowed the level of differentiation between conditions to vary across conditions . Such a situation commonly arises when one performs co-expression network analysis for multiple conditions . Here , we simulated the gene expression profiles under three conditions for 450 genes comprised of 9 modules of 50 genes each . Similar to the 2-condition simulation , we included two groups of 4 modules of the same type . To better imitate real networks , we also included an additional type II module to mimic housekeeping co-expression across 3 conditions . In total , we considered 4 scenarios . Table 2 summarizes the configurations of these simulations . S1 and S2 represent the cases where pairwise similarities between conditions are constant across conditions , with a higher similarity in S1 than in S2 . S3 and S4 represent the case where pairwise similarities vary across conditions , with a higher similarity in S3 than S4 . In this simulation , we compared CFGL , FGL and GL . CFGLO was not included as its performance is similar to the previous case . In all scenarios , our approach has a higher AUC ( pAUC: S1: 0 . 649 , S2: 0 . 508 , S3: 0 . 650 , S4: 0 . 513 ) and a lower SSE than both GL ( pAUC: S1: 0 . 597 , S2: 0 . 474 , S3: 0 . 593 , S4: 0 . 472 ) and FGL ( pAUC: S1: 0 . 605 , S2: 0 . 494 , S3: 0 . 615 , S4: 0 . 499 ) ( Fig 4 , S2 Table ) . The gain over GL is most apparent when the differentiation between conditions is low ( S1 ) . This is again because data integration is most beneficial when networks are similar across conditions . FGL performs well in this case too . However , when the distance between conditions is different across conditions ( S3 ) , the advantage of FGL over GL diminishes; and when the differentiation between conditions is relatively high ( S2 and S4 ) , FGL performs worse than GL . This is expected , as imposing similarity across conditions as in FGL is improper for these scenarios . However , our method performs well in all scenarios . Taken together , we attribute the gain of our methods to its adaptive way of enforcing similarities . When networks are highly similar across conditions , enforcing similarities across all edges , as in FGL , is optimal . Our method adapts to this situation and produces similar results to FGL . In contrast , when networks are different across conditions , similarity should be encouraged only among the shared edges in data integration . Our method is again adaptive to the differential patterns across conditions , thus shows more gain when the difference between conditions is present . We applied our method to a microarray dataset collected from a recombinant inbred ( RI ) rat panel and compared with FGL , GL and WGCNA , which is a widely used network analysis method based on marginal correlation [21] . The gene expression profiles in the brain and heart tissues were measured for 19 rat strains using Affymetrix Rat Exon Array 1 . 0 ST . Details on data processing and normalization are provided in Methods . Because of the small sample size , we restricted the network construction to the 500 most differentially expressed ( DE ) genes between brain and heart ( see Methods ) . We used BIC to guide the selection of penalty parameters for CFGL , FGL and GL and used default parameters for WGCNA analysis . We applied our method to the breast cancer data from the TCGA project [44] . Breast cancer is the most common cancer among women [43] . According to the presence and absence of the estrogen receptor ( ER ) in cancer cells , breast cancer can be classified into two subtypes , ER+ and ER- . Approximately two-thirds of breast cancer are ER+ at the time of diagnosis , and the rest are ER- . The ER status provides important clinical implications for both mechanisms of carcinogenesis and therapeutic treatment [45] . The TCGA BRCA project [44] has collected gene expression RNA-seq data for 1100 breast cancer patients . Among them , 112 individuals have both tumor tissue and matched peripheral normal tissue . Our goal is to identify co-expression modules that are specific to ER+ or ER- subtype and those that are shared between the two tumor subtypes but are not present in normal tissue . To ensure the independence of the samples in our analysis , we used the normal tissue samples from these 112 individuals and tumor samples that are annotated as ER+ ( 187 samples ) or ER- ( 98 samples ) from different individuals . Due to the limited sample size , we restrict our analysis to a subset of 1000 genes that either show a significant association with survival time in a Cox model or have been reported to be related to breast cancer ( Details in Methods ) . To obtain robust co-expression networks , we applied the stability selection procedure in [38] in conjunction with CFGL , FGL and GL . In addition , we also performed WGCNA on the same dataset as a comparison . Details of the data processing steps and the procedure of the stability selection can be found in Methods . In this paper , we present a method , called condition-adaptive fused graphical lasso ( CFGL ) , to construct gene co-expression networks for multiple conditions simultaneously . By incorporating a data-driven penalty that reflects the condition-specific co-expression pattern in the FGL framework , this method takes condition specificity into account while borrowing information across conditions in the network construction . Our results have shown that it effectively accounts for heterogeneity between samples and between co-expression patterns introduced by condition specificity . It outperforms GL and FGL methods in both edge detection and estimation of edge weights across a range of scenarios in simulation studies . Our analysis on a rat multi-tissue dataset and TCGA breast cancer data reveals interesting biological insights . In both datasets , the modules in the condition-specific subnetwork identified by our method consistently show biologically relevant functions , demonstrating the suitability of our method for studying tissue-specific or disease-specific co-expression networks . The analysis on TCGA breast cancer data also reveals several interesting findings related to the mechanism of ER+ and ER- tumor subtypes . We found that the genes most significantly associated with survival time are less likely to be hubs . This suggests that most genes associated with cancer progression may govern specific functions or locate downstream in pathways to execute the final biological functions , rather than regulating a large number of biological processes . Similarly , we also observed that the hub genes in the tumor-specific subnetworks tend to not harbor edges shared with normal tissue . Several previously known cancer-related genes , including PTPN22 , BRIP1 , and CEACAM6 , were found as hubs in the tumor-related subnetworks . Together , these results confirm the biological relevance of the results from our method . Interestingly , we noticed that the methods that construct networks separately for each condition ( GL and WGCNA ) consistently produce very few condition-common edges ( < 5% ) , far fewer than the joint analysis methods ( FGL and CFGL ) . This is even the case when some common biology is expected to be shared between conditions , for example , ER+ and ER- breast cancer subtypes . The results reported by these analysis methods , and their suitability for studying condition specificity , are therefore questionable . Another interesting observation is that the graphical lasso based methods ( GL , FGL and CFGL ) generally report smaller modules than WGCNA . This is partly because the former uses sparse estimation and partly because the former evaluates conditional independence between genes , rather than marginal independence as in the latter . It is generally believed that very large gene sets may encompass multiple cellular processes and make GO enrichment results less specific [64][65][66] , thus smaller modules may have the benefit of improving the interpretability of results . Though our method was motivated by co-expression networks , it is suitable for other data applications with multiple conditions but shared network structures , such as learning condition-specific binary networks with sparse Ising models [67][68] . The R package for our method CFGL is available on GitHub , https://github . com/Yafei611/CFGL .
CFGL estimates the precision matrices {Θ} by solving maximize{Θ} ( ∑k=1Knk[log{det ( Θ ( k ) ) }−tr ( S ( k ) Θ ( k ) ) ]−P ( {Θ} ) ) . The penalty term P ( {Θ} ) is P ( {Θ} ) =λ1∑k=1K∑i≠j|θij ( k ) |+λ2∑k<k′K∑i≠jwij ( kk′ ) |θij ( k ) −θij ( k′ ) | , where nk is the sample size of kth condition and S ( k ) = ( Y ( k ) ) TY ( k ) /nk is the empirical covariance matrix for the kth expression data set . We implemented the ADMM algorithm [69] to solve the above problem . The detailed optimization procedure can be found in the supplementary materials ( S1 Text ) . We determine the tuning parameters λ1 and λ2 ( λ2 only for CFGL and FGL ) according to the Bayesian information criterion ( BIC ) [70] . BIC ( λ1 , λ2 ) =∑k=1K[nktr ( S ( k ) Θ^ ( λ1 , λ2 ) ( k ) ) −nklog{det ( Θ^ ( λ1 , λ2 ) ( k ) ) }+pklog ( nk ) ] , where Θ^ ( λ1 , λ2 ) ( k ) is the estimated precision matrix for the kth condition obtained at ( λ1 , λ2 ) , and pk is the number of non-zero elements in Θ^ ( λ1 , λ2 ) ( k ) . We ran the analysis on a series of combinations of λ1 and λ2 , then chose the tuning parameters that achieve the minimal BIC value . In the simulation , we used BIC to select tuning parameters . For the rat data analysis , because the sample size is very small , it is difficult to obtain meaningful estimates from subsamples in the stability selection . Instead , we first identified the model that achieves the minimum BIC and the models with similar BIC values , then selected the model that had the fewest edges to obtain biologically interpretable results . In the TCGA data analysis , we applied a stability selection procedure to identify reliable edges ( see stability selection section ) . We determined the screening matrix for CFGL by testing the differences between two precision matrices using the method proposed by Xia et al . [34] . This method tests whether the difference ( Δ=Σk−1−Σk′−1 ) between two precision matrices is 0 , i . e . H0: Δ = 0 vs H1: Δ ≠ 0 . To avoid falsely imposing similarity for edges that are moderately differential , we used a relaxed FDR threshold ( FDR = 0 . 4 ) to determine differential entries , such that only the edges that are obviously non-differential across conditions ( i . e . FDR>0 . 4 ) were encouraged to be similar ( wij ( kk′ ) =1 ) . We implemented this method in R and included it in the CFGL package . In the simulation study , we generate the gene expression data for multiple network configurations . Suppose each condition contains M disjoint modules and each module consists of p genes . For each module , the gene constitution is constant across conditions , but the connectivity and the edge weight may vary across conditions . To generate conditions with a specified level of similarity , we first generate the network for condition 1 , and then generate the network for other conditions based on their similarities to condition 1 . To assess the performance of CFGL-oracle , we obtained the true screening matrix as Wtrue ( kk′ ) ( i , j ) ={1Σk−1 ( i , j ) −Σk′−1 ( i , j ) <0 . 010otherwise where Σk−1 is the simulated precision matrix for the kth condition . The accuracy of edge identification is assessed by checking if the presence of edges is correct in the estimated matrix Σk−1^ . We define true positive as Σk−1 ( i , j ) ≠0 and Σk−1^ ( i , j ) ≠0 and false positive as Σk−1 ( i , j ) =0 and Σk−1^ ( i , j ) ≠0 ( i>j ) . The accuracy of edge weight estimation is assessed by the sum of square error ( SSE ) between the true and estimated edge weights SSE=∑k=1K∑i=2P∑j=1i−1 ( Σk−1 ( i , j ) −Σk−1^ ( i , j ) ) 2 , where K is number of conditions and P is number of nodes ( genes ) . For each λ2 , we generate an ROC curve by computing the true positive rate and false positive rate over a grid of λ1 . Similarly , SSE is computed over a grid of λ1 . To compare the performance of different methods , we calculated partial AUC ( pAUC ) , which is the area under the ROC curve over a restricted range of false positive rate ( FPR ) . Because the primary interests are edges detected at a low false positive rate , we compute pAUC on the FPR range of ( 0 , 0 . 05 ) in the simulation study . We performed WGCNA analysis for rat and TCGA BRAC expression data . Because WGCNA does not allow joint analysis for more than one condition , we performed WGCNA for each condition separately . We used the WGCNA R package ( version 1 . 63 ) and chose the tuning parameters according to its manual [21] . For the rat expression data , the soft threshold was set to 8 for brain tissue and 7 for heart tissue , and the module size was set to 20 to accommodate the relatively small total number of genes . For TCGA data , the soft threshold was set to 6 for normal tissue , 5 for the ER+ tumor tissue and 4 for the ER- tumor tissue , and the module size was set to 50 . Default settings were used for all the other parameters . To study hub genes and edges using WGCNA , we obtained edge weights from the topological overlap matrix ( TOM ) calculated from WGCNA . The TOM is a quantity computed by WGCNA for measuring the topological similarity between genes . Each entry can be viewed as an edge weight between a pair of genes . Unlike graphical lasso based methods , which provide sparse networks , the TOM matrix is dense . To ensure the comparison with graphical lasso based methods is on the same basis , we only kept the edges with high values of TOM and removed other edges . The number of nonzero edges is chosen according to the number of edges identified by CFGL in the same dataset . Heart and brain RNA expression levels were measured in a recombinant inbred ( RI ) rat panel , HXB/BXH , using the Affymetrix Rat Exon 1 . 0 ST Array ( Affymetrix , Santa Clara , CA ) . This rat panel was originally generated using gender reciprocal crossing between the congenic Brown Norway strain with the polydactyly-luxate syndrome ( BN-Lx/Cub ) and the spontaneous hypertensive rat strain ( SHR/OlaIpcv ) , with sixty generations of brother/sister mating after the F2 generation [73] . The CEL files for the heart and brain RNA expression data from 3 to 4 male rats per strain ( 19 strains ) are publicly available through the PhenoGen website ( http://phenogen . ucdenver . edu ) [74] along with a probe mask for the ‘core’ ( Affymetrix defined ) transcript clusters that eliminates probes that do not align uniquely to the RN6 version of the rat genome or align to a region of the genome that harbors a single nucleotide polymorphism between either of the parental strains ( SHR and BN-Lx ) and the reference genome . Further detail about this type of probe mask are available in Saba et al 2015 [75] . Transcript cluster estimates on the log2 scale were estimated using the rma-sketch pipeline for normalization and aggregation using Affymetrix Power Tools ( Irizarry et al 2003; Lockstone 2011 ) [76][77] . Individual rat estimates were summarized as strain mean values for each transcript cluster and strain combination . Given the small sample size , we restricted the network construction to the 500 most differentially expressed genes between the two tissues . The differential expression was determined using the R package LIMMA with the default parameter settings . We ran CFGL and FGL for a grid of λ1 and λ2 . They both achieved the lowest BIC at λ1 = 0 . 001 and λ2 = 0 . 0008 . To investigate the effect of λ2 , we report the results at λ1 = 0 . 001 and λ2 = 0 . 0008 , 0 . 0010 , 0 . 0012 . We set λ = 0 . 0009 for GL since it gives similar sparsity ( edge number ) to the other two methods . For WGCNA analysis , we kept the edges with the highest TOM values ( 1000 for brain and 500 for heart ) from the WGCNA results , such that the number of edges is consistent with that of the estimated network from CFGL/FGL . TCGA has collected gene expression RNA-seq data for 1092 breast cancer patients [44] . We used the normal tissue samples from the individuals ( n = 112 ) who have both tumor tissue and matched peripheral normal tissue , and all tumor samples from different individuals that were annotated with ER+ ( n = 187 ) and ER- ( n = 98 ) [44] . We obtained the gene expression level by downloading the RNA-seq V2 data , which are reads counts normalized by RSEM , from the TCGA website ( https://cancergenome . nih . gov/ ) . We then took log transformation for the expression level ( with 0 . 5 added to the counts of each gene to avoid 0 ) and standardized the transformed expression level to mean 0 and standard deviation 1 . Prior to network construction , we first removed genes with very low counts ( less or equal than 5 ) in more than 10% ( 40 ) samples . After this step , the log summed read counts over all samples approximately follow a normal distribution . Due to the limitation of sample size , we restricted our analysis to a subset of 1000 genes . To select 1000 genes in the analysis , we first included 39 genes that were previously reported as breast cancer-related genes ( S10 Table ) [23][44] . Then we selected other 961 genes that are most strongly associated with the survival time based on a univariate Cox regression: h ( t ) =h0 ( t ) ×exp ( βxg ) where t is the survival time and xg is the expression level of the gth gene . We constructed the co-expression network using CFGL/FGL/GL in conjunction with the stability selection procedure ( See next section ) . For the WGCNA-based network , we kept the edges with the highest TOM values ( 920 edges for normal tissue , 1154 edges for ER+ tumor tissue and 840 edge for ER- tumor tissue ) . In order to obtain reliable co-expression networks , we applied the stability selection procedure in [78] to CFGL , FGL , and GL . This procedure first generates a large set of subsamples from the original data and then builds networks based on the subsamples . The edges that frequently occur in subsamples are kept . This method provides an upper bound for the FDR control and has been shown to outperform the standard GL when being applied to GL [78][38] . In our analysis , we created 100 subsamples , each of which contains half of the original samples . To reduce the computational load , we first determined the optimal choice of λ1 based on the original dataset , and then used this value for all subsamples . For all methods , λ1 = 0 . 2 achieves both reasonable sparsity and low BIC across a series λ1 ( 0 . 01–0 . 50 ) on the original dataset , thus we fixed λ1 = 0 . 2 in all subsamples . For FGL and CFGL , we performed the analysis on a series of λ2 for each subsample and then used the tuning parameters that achieve the minimal BIC value to select edges . The minimal BIC for all subsamples was found in the range of λ2 = 0 . 002–0 . 02 . We keep the edges that appear in more than 90% subsamples . According to the false discovery rate ( FDR ) calculation in [38] , this threshold guarantees that the number of wrong edges is less than 800 among the 499500 possible edges in the graph . The GO enrichment analyses were conducted using ToppFun[79] , which is publicly available at https://toppgene . cchmc . org/enrichment . jsp . All parameters are used at their default setting . | Gene co-expression networks provide insights into the mechanism of cellular activity and gene regulation . Condition-specific mechanisms may be identified by constructing and comparing co-expression networks of multiple conditions . We propose a novel statistical method to jointly construct co-expression networks for gene expression profiles from multiple conditions . By using a data-driven approach to capture condition-specific co-expression patterns , this method is effective in identifying both co-expression patterns that are specific to a condition and that are common across conditions . The application of this method to real datasets reveals interesting biological insights . | [
"Abstract",
"Introduction",
"Results",
"Method"
] | [
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"breast",
"tumors",
"cardiovascular",
"anatomy",
"gene",
"regulation",
"cancers",
"and",
"neoplasms",
"random",
"variables",
"covariance",
"simulation",
"and",
"modeling",
"oncology",
"mathematics",
"network",
"analysis",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"gene",
"expression",
"breast",
"cancer",
"probability",
"theory",
"anatomy",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"heart"
] | 2018 | Condition-adaptive fused graphical lasso (CFGL): An adaptive procedure for inferring condition-specific gene co-expression network |
Effector responses induced by polarized CD4+ T helper 2 ( Th2 ) cells drive nonhealing responses in BALB/c mice infected with Leishmania major . Th2 cytokines IL-4 and IL-13 are known susceptibility factors for L . major infection in BALB/c mice and induce their biological functions through a common receptor , the IL-4 receptor α chain ( IL-4Rα ) . IL-4Rα–deficient BALB/c mice , however , remain susceptible to L . major infection , indicating that IL-4/IL-13 may induce protective responses . Therefore , the roles of polarized Th2 CD4+ T cells and IL-4/IL-13 responsiveness of non-CD4+ T cells in inducing nonhealer or healer responses have yet to be elucidated . CD4+ T cell–specific IL-4Rα ( LckcreIL-4Rα−/lox ) deficient BALB/c mice were generated and characterized to elucidate the importance of IL-4Rα signaling during cutaneous leishmaniasis in the absence of IL-4–responsive CD4+ T cells . Efficient deletion was confirmed by loss of IL-4Rα expression on CD4+ T cells and impaired IL-4–induced CD4+ T cell proliferation and Th2 differentiation . CD8+ , γδ+ , and NK–T cells expressed residual IL-4Rα , and representative non–T cell populations maintained IL-4/IL-13 responsiveness . In contrast to IL-4Rα−/lox BALB/c mice , which developed ulcerating lesions following infection with L . major , LckcreIL-4Rα−/lox mice were resistant and showed protection to rechallenge , similar to healer C57BL/6 mice . Resistance to L . major in LckcreIL-4Rα−/lox mice correlated with reduced numbers of IL-10–secreting cells and early IL-12p35 mRNA induction , leading to increased delayed type hypersensitivity responses , interferon-γ production , and elevated ratios of inducible nitric oxide synthase mRNA/parasite , similar to C57BL/6 mice . These data demonstrate that abrogation of IL-4 signaling in CD4+ T cells is required to transform nonhealer BALB/c mice to a healer phenotype . Furthermore , a beneficial role for IL-4Rα signaling in L . major infection is revealed in which IL-4/IL-13–responsive non-CD4+ T cells induce protective responses .
Experimental Leishmania major infection is widely used to explore the control of T helper 1 ( Th1 ) /Th2 differentiation and elucidate mechanisms underlying susceptibility/resistance to intracellular microbial infection [1 , 2] . Typically , susceptible BALB/c mice infected subcutaneously with L . major develop severe pathology , manifested by progressive lesion development , necrosis , and death , while resistant C57BL/6 mice are able to control and heal dermal lesions [3] . Nonhealing disease in BALB/c mice is associated with a Th2 response characterized by secretion of mainly IL-4 , IL-5 , IL-9 , and IL-13 [2 , 4–7] , high anti-Leishmania antibody titres , arginase-1 production by macrophages [8 , 9] and visceral dissemination of parasites [10] . In contrast , resistance to L . major infection is mediated by development of a protective Th1 response , in which sustained IL-12 production , interferon-γ ( IFN-γ ) release and macrophage killing via effector nitric oxide ( NO ) production catalyzed by inducible NO synthase ( iNOS ) underlie protective responses [9 , 11–14] . CD4 T cell–derived cytokines drive L . major responses , and , as such , events that control T cell differentiation in response to L . major appear to be critical for disease outcome [15] . Disruption of Th1 differentiation by neutralization of IL-12 renders resistant C57BL/6 mice susceptible , whereas susceptible BALB/c mice treated with IL-12 become resistant to L . major infection [12] . IL-12 production must be sustained to control infection [13] . While both resistant and susceptible mice produce IL-4 early after infection [16 , 17] , production of this cytokine is sustained in susceptible mice and transient in resistant mice [16–18] . Neutralization of IL-4 allowed control of L . major infection in BALB/c mice [19] . Subsequent studies in knockout mice proved that IL-4 was indeed important but not the sole mediator of susceptibility in BALB/c mice . L . major infection was controlled in BALB/c IL-4−/− mice , but parasite burdens remained greater than those of resistant animals [6 , 20] . These observations remain controversial , with some laboratory strains developing IL-4–independent susceptibility and indicating that further factors are involved [21] . IL-13 has been implicated as a susceptibility factor in L . major infection [4] . Susceptible IL-13 transgenic C57BL/6 mice develop impaired IL-12 and IFN-γ production during acute leishmaniasis , while IL-13−/− BALB/c mice remain comparatively resistant [4 , 22] . IL-13 can influence Th1 differentiation by modulating macrophage function and suppressing secretion of NO , IL-12 , and/or IL-18 [22 , 23] , partially attributed to IL-4/IL-13 activated alternative macrophages ( aaMphs ) , recently demonstrated in mice deficient for this activation status [9 , 24] . IL-4 and IL-13 share a common signaling pathway through the IL-4 receptor α ( IL-4Rα ) chain . A functional IL-4R ( type I ) requires assembly of IL-4Rα with a γc chain , while interaction of IL-4Rα with an IL-13Rα1 subunit leads to formation of a functional IL-13 receptor ( type II ) [25] . IL-4Rα–deficient mice therefore lack responsiveness to IL-4 and IL-13 . Careful analysis of footpad swelling and lesion development showed that initial control of L . major infection is equivalent in IL-4−/− and IL-4Rα−/− BALB/c mice . However , in contrast to IL-4−/− mice , IL-4Rα−/− mice develop progressive chronic disease . These data clearly indicate a protective role for IL-13 signaling in protection against chronic L . major infection , at least in the absence of IL-4 responsiveness [20] . Expression of IL-4Rα reflects the pleiotropic nature of IL-4 biology , as this receptor subunit is expressed upon a wide range of cells [26] . Given the central role of T cells in controlling L . major infection [15] and of IL-4 in driving Th2 responses [27] , CD4+ T cell–specific IL-4Rα knockout mice were generated to elucidate the role of IL-4Rα–mediated signaling in CD4+ T cells independently of non-CD4+ T cell populations . Our results demonstrate a successful generation of transgene-bearing hemizygous LckcreIL-4Rα−/lox BALB/c mice that have effective deletion of IL-4Rα on CD4+ T cells , an incomplete deletion on CD8+ T cells and other T cell subpopulations , and normal expression on non–T cells . LckcreIL-4Rα−/lox mice infected with L . major developed a healing disease phenotype and clinical immunity similar to genetically resistant C57BL/6 mice . Consequently , our studies demonstrate that impairment of IL-4Rα–dependent Th2 polarized CD4+ T cells in the presence of IL-4/IL-13–responsive non-CD4+ T cells is required to transform nonhealer BALB/c mice to a healer phenotype .
Recently established IL-4Rαlox/lox BALB/c mice [24] were intercrossed with BALB/c mice expressing Cre-recombinase under control of the T cell–specific promoter Lck [28] and IL-4Rα−/− BALB/c mice [20] to generate LckcreIL-4Rα−/lox mice ( Figure 1A ) . IL-4Rα hemizygosity ( −/lox ) increases probability of Cre-mediated deletion of the “floxed” allele [24] . LckcreIL-4Rα−/lox mice were identified by PCR genotyping ( Figure 1B ) . Fluorescence-activated cell sorter ( FACS ) analysis of IL-4Rα surface expression confirmed efficient deletion on CD3+CD4+ T cells from LckcreIL-4Rα−/lox mice when compared with IL-4Rα−/− and IL-4Rα−/lox BALB/c ( WT ) controls ( geometric mean channel florescence [geo . mean]: WT = 18 . 11 , IL-4Rα−/− = 8 . 5 , LckcreIL-4Rα−/lox = 9 . 48 ) , but incomplete and variable deletion efficiency was observed on CD8+ T cells ( Figure 1C and Figure S1 ) ( geo . mean: WT = 18 . 69 , IL-4Rα−/− = 9 . 06 , LckcreIL-4Rα−/lox = 13 . 96 ) and γδ+ ( geo . mean: WT = 7 . 6 , IL-4Rα−/− = 3 . 15 , LckcreIL-4Rα−/lox = 6 . 72 ) and NK–T cells ( geo . mean: WT = 9 . 03 , IL-4Rα−/− = 5 . 25 , LckcreIL-4Rα−/lox = 7 . 28; Figure 1C ) . The cellular specificity of IL-4Rα deletion was confirmed because B cells ( CD19+ ) , macrophages , and dendritic cells ( DCs; Figure 1C ) of LckcreIL-4Rα−/lox mice maintained expression of IL-4Rα . Efficiency of deletion of IL-4Rα in CD4+ T cells was analyzed at the genomic level by quantitative real-time PCR . The number of exon 5 alleles ( both present in all cells ) relative to exon 8 alleles ( deleted in −/− , one copy in −/lox mice ) of IL-4Rα was determined in CD4+ T cells sorted to high purity . As expected , exon 8 was efficiently deleted in CD4+ T cells and B cells from IL-4Rα−/− mice ( Figure 1D ) . Confirming FACS analysis , efficient deletion of lox-p–flanked IL-4Rα exon 8 was observed in CD4+ T cells from LckcreIL-4Rα−/lox mice . Analysis revealed 0 . 114 copies of exon 8 were retained relative to exon 5 , equating to 95 . 48% ± 5 . 8% deletion efficiency of exon 8 within the CD4+ T cell population . In agreement , no CD4+ T cell exon 8 product was visible following 75 PCR cycles ( Figure 1D ) . An equivalent ratio of exon 8 versus exon 5 was maintained in CD19+ B cells in LckcreIL-4Rα−/lox mice compared with WT controls . These data provide evidence of efficient deletion of IL-4Rα in CD4+ T cells from LckcreIL-4Rα−/lox BALB/c mice . IL-4 promotes proliferation of naive CD4+ T cells in vitro [29] . In order to assess functional impairment of IL-4Rα on CD4+ T cells from LckcreIL-4Rα−/lox mice , naive CD4+ T cells were stimulated with IL-4 , and proliferation was measured by [3H] thymidine incorporation ( Figure 2A ) . CD4+ T cells isolated from naive LckcreIL-4Rα−/lox BALB/c mice were unable to proliferate in response to IL-4 , as were those from IL-4Rα−/− mice . In contrast , WT CD4+ T cells showed dose-responsive proliferative responses to IL-4 . Impairment of IL-4 signaling was IL-4Rα specific , as proliferative responses to IL-2 , which shares a γc-chain with the type I IL-4R , were unaffected in all three strains ( Figure 2A ) . Impairment of CD4+ T cell IL-4 responsiveness was further verified using the Th cell differentiation assay . Th1 versus Th2 differentiation of noncommitted CD4+ T cells may be achieved in vitro by treatment with either IL-12/anti–IL-4 or IL-4/anti–IFN-γ , respectively [29] . Naive CD4+ T cells stimulated with anti-CD3/CD28 and polarized with cytokine/neutralizing mAb treatment demonstrate that Th2 polarization , indicated by IL-4 production , was impaired in LckCreIL-4Rα−/lox and IL-4Rα−/− but not WT mice ( Figure 2B ) . As expected , Th1 polarization was achieved in all three strains . Functional macrophage IL-4Rα data from LckcreIL-4Rα−/lox mice were demonstrated in Figure 2C . NO production was suppressed by IL-4 and IL-13 in macrophages from LckcreIL-4Rα−/lox and WT mice ( Figure 2C ) , but not IL-4Rα−/− macrophages , showing IL-4Rα specificity . As a positive control , IL-10 suppressed NO production in all three strains . Production of IgE antibodies is strictly dependent on IL-4 signaling [30] . IL-4Rα responsiveness of B cells in LckcreIL-4Rα−/lox mice was demonstrated in Figure 2D . Antigen-induced IgE antibody was present at slightly reduced levels in OVA-immunized LckcreIL-4Rα−/lox mice when compared with those of WT mice , while IgE production was completely abrogated in IL-4Rα−/− mice ( Figure 2D ) . Together , these data provide evidence for effective impairment of IL-4Rα–mediated functions in LckcreIL-4Ra−/lox CD4+ T cells , but not in other lymphocyte subpopulations such as B cells and macrophages . Controversy remains as to whether IL-4 [6 , 20 , 21] and/or IL-4Rα signaling [20 , 31] are key components of susceptibility to L . major infection . Polarized Th2 cells certainly play a significant role in contributing to susceptibility [32] . To investigate the consequence of CD4+ T cell–specific IL-4Rα unresponsiveness in leishmaniasis , mice were infected subcutaneously with 2 × 106 stationary phase metacyclic promastigotes of L . major LV39 ( MRHO/SV/59/P; Figure 3A ) . As expected , WT mice developed rapidly growing nonhealing lesions ( Figure 3A ) within 12 wks of infection and were unable to control parasite burden with high parasite numbers in the footpads ( Figure 3B ) and LNs ( Figure 3C ) . IL-4Rα−/− mice initially controlled infection with intermediate parasite load in the draining lymph nodes ( LNs ) and footpad . However , as previously described [20] , global IL-4Rα deficiency does not confer resistance to L . major infection , as the mice progressed to develop necrotic lesions in the chronic phase ( Figure 3A ) . In contrast , LckcreIL-4Rα−/lox mice were able to resolve infection with lesion growth comparable with resistant C57BL/6 mice ( Figure 3A ) . LckcreIL-4Rα−/lox mice carried low parasite burdens in the footpad , with approximately 2 , 000-fold less parasites in the footpad compared with that of WT 6 wk after infection ( Figure 3B ) , and maintained an intermediate parasite burden in the draining LNs when compared with C57BL/6 and WT mice ( Figure 3C ) . Resistance to L . major infection in CD4+ T cell–specific IL-4Rα–deficient mice was profound , as parasite load in the footpad was equivalent to that observed in C57BL/6 mice at 36 wk after infection using PCR to detect kinetoplast DNA at the lesion site ( Figure 3D ) . LckcreIL-4Rα−/lox mice were also shown to be resistant to reinfection . At 6 wk after L . major infection , mice were reinfected in the contralateral footpad . LckcreIL-4Rα−/lox mice were again comparable with genetically resistant C57BL/6 mice in lesion development , while L . major reinfection in WT mice progressed to necrosis in acute phase ( Figure 3E ) . LckcreIL-4Rα−/lox mice were also resistant to the more virulent L . major ( MHOM/IL/81/FEBNI ) strain ( Figure 3F ) , again with lesion kinetics comparable with that of C57BL/6 mice . IL-10 is a highly immunosuppressive cytokine , profoundly reducing NO production by macrophages ( Figure 2C ) [33] , and is a susceptibility factor in L . major infection [31] . Intracellular cytokine staining revealed increased numbers of antigen-specific CD4+ IL-10–secreting T cells in the draining LNs of WT mice compared with C57BL/6 and LckcreIL-4Rα−/lox mice ( Figure 4A and 4B ) . In order to examine an in vivo correlate demonstrating IL-10 inhibition of protective parasite specific responses , IL-12/IFN-γ–driven delayed type hypersensitivity ( DTH ) responses were investigated in L . major–infected mice . C57BL/6 develop sustained footpad swelling when challenged with soluble L . major antigen ( SLA; Figure 4C ) , and LckcreIL-4Rα−/lox mice showed intermediate sustained swelling , whereas minimal DTH responses were observed in WT mice ( Figure 4C ) . As expected , addition of IL-10 to SLA diminished DTH responses in all mice ( Figure 4D ) . Neutralization of IL-10 function by blockade of IL-10R lifted suppression of the DTH in the low-responder WT mice on a par with DTH responses observed in the resistant strains ( Figure 4E ) . Confirming that increased DTH responses observed in LckcreIL-4Rα−/lox mice resulted from increased Th1 responses , significant levels of IL-12p70 ( Figure 4F ) and IFN-γ ( Figure 4G ) were detected in footpad lysates taken from resistant mice , while little or no IL-12p70 or IFN-γ were induced in susceptible WT mice ( Figure 4F and 4G ) . IL-12 is a key protective cytokine involved in inducing protective responses following L . major infection [34] . We therefore examined IL-12 expression in LckcreIL-4Rα−/lox mice . Although IL-12p35 mRNA production was equivalent at 1 wk after infection ( unpublished data ) , levels of IL-12p35 mRNA were increased in draining LNs of LckcreIL-4Rα−/lox and C57BL/6 mice at 3 wk after infection when compared with those of WT mice ( Figure 5A ) . Levels of IL-12p35 mRNA increased from 1 wk to 3 wk after infection in resistant mice while remaining low in susceptible mice ( Figure 5B ) . IFN-γ–driven iNOS production by macrophages is a key control mechanism in L . major infection [35] . CD4 T cell antigen–specific IFN-γ cytokine production was therefore examined . CD4 T cells from LckcreIL-4Rα−/lox mice induced 2 . 5- , 1 . 6- , and 2-fold more IFN-γ when compared with those from IL-4Rα−/− and WT or IL-4Rα−/lox mice at 10 , 6 , and 12 wk after infection ( Figure 5C ) , respectively . Furthermore , greater IFN-γ levels were detected in footpad homogenates from infected LckcreIL-4Rα−/lox compared with WT mice at 10 wk after infection ( Figure 5D ) . Importantly , IL-4Rα–independent IL-4 production was observed in LckcreIL-4Rα−/lox mice with similar levels of IL-4 production being observed in WT and LckcreIL-4Rα−/lox mice in antigen-specific CD4+ T cell restimulation ( Figure 5E ) and footpad lysates ( Figure 5F ) . Consistently increased IFN-γ production had an influence on downstream macrophage effector functions . This was shown at 6 wk after infection , when more copies of iNOS mRNA/parasite were observed in resistant strains of mice ( Figure 5G ) . Together , these data demonstrate that resistance to acute leishmaniasis in LckcreIL-4Rα−/lox mice is associated with an early induction of increased protective type 1 immunity and reduced suppression of responses by IL-10–secreting CD4+ T cells .
IL-4 and IL-13 share a common signaling pathway through the IL-4Rα chain [26] , and as such the combined role of both cytokines can be studied in vivo in IL-4Rα−/− mice . While IL-4 mediates multiple effects on T cells , murine T and B cells do not respond to IL-13 [7] . Generation of CD4+ T cell–specific IL-4Rα–deficient ( LckcreIL-4Rα−/lox ) mice therefore allows investigation into the role of IL-4 signaling specifically on CD4+ T cells while maintaining IL-4/IL-13–mediated functions on non-CD4+ T cells . CD4+ T cell–specific IL-4Rα–deficient BALB/c mice were generated using the Cre/LoxP recombination system in BALB/c embryonic stem cells . Previous studies have shown efficiency of cell-specific Cre-mediated gene disruption may vary between 38%–85% depending on recombinase efficiency and promoter activity [36] . Efficiency of CD4+ T cell–specific IL-4Rα disruption ( 95 . 48% ) was increased by using hemizygous WT mice instead of IL-4Rαlox/lox as mating partners for transgenic LckCre mice , thereby reducing the LoxP substrate for Cre-recombinase by 50% . FACS analysis showed efficient disruption of IL-4Rα gene expression in CD4+ T cells and incomplete deletion in CD8+ and NK–T cells with variable deletion efficiency . γδ T cells and non–T cells retained unaltered receptor expression in LckcreIL-4Rα−/lox mice . The data suggest that while the Lck promoter is functional and mediates deletion of loxP-flanked DNA sequences in CD4+ , CD8+ , and NK–T cell subsets , deletion is more efficient in CD4+ T cells using this promoter construct . Functional analysis further demonstrated effective and specific impairment of the IL-4 responsiveness of CD4 T cells , while B cells and macrophages retained IL-4– and IL-13–mediated functions . Thus , LckcreIL-4Rα−/lox mice are CD4+ T cell–specific IL-4Rα knockout mice , whereas all other cell types remain responsive to IL-4/IL-13 . LckcreIL-4Rα−/lox mice infected with L . major developed similar kinetics of lesion development and resolution as those observed in C57BL/6 mice genetically resistant to two strains of L . major . In contrast , control IL-4Rα−/lox ( WT ) and IL-4Rα−/− BALB/c mice developed progressive lesion swelling leading to necrosis during the acute and chronic phases of disease as expected . LckcreIL-4Rα−/lox BALB/c and C57BL/6 mice also resisted secondary parasite challenge , unlike WT mice , which showed no signs of footpad pathology . A similar resistant phenotype to L . major infection was also noted in an independent line of mice in which IL-4Rα is efficiently deleted from CD4 , CD8 , NK–T , and γδ T cells ( unpublished data ) , indicating that IL-4–responsive CD4+ T cells control susceptibility to L . major infection , and that the resistant phenotype is not associated with Cre activity in T cells or hypothetical mutations introduced by the transgene . Together , our study demonstrates that clinical immunity can be achieved in mice on a susceptible BALB/c background by abrogating IL-4Rα responsiveness on CD4+ T cells while retaining IL-4/IL-13–mediated function on non-CD4+ T cells . IL-10 is a potent suppressor of macrophage activation [37] , can abolish IFN-γ/LPS–induced killing of L . major by macrophages [38 , 39] , and can suppress development of DTH responses [40] . In agreement , L . major–infected C57BL/6 and LckcreIL-4Rα−/lox mice developed DTH responses to SLA , inhibited by coadministration of IL-10 . In contrast , DTH responses in WT mice were absent . Neutralization of IL-10 signaling allowed WT mice to mount a significant response to SLA . Together , DTH data demonstrated that IL-10 produced in response to SLA in susceptible mice was able to suppress protective cell-mediated immune responses . IL-10 production is increased in BALB/c mice compared with resistant mice [41] , can regulate parasite survival in resistant C57BL/6 mice [1 , 42] , and is a susceptibility factor for L . major infection [31 , 39] . In agreement , the draining LNs of infected resistant LckcreIL-4Rα−/lox and C57BL/6 mice contained reduced numbers of CD4+ IL-10–secreting cells ( 4- and 9-fold less , respectively ) compared with WT mice . Variable amounts of IL-10 staining were observed in the non-CD4+ T cell population; however , this was found to be nonspecific ( Figure 4A ) . Increased IL-10 secretion was also observed in anti-CD3–stimulated CD4+ T cells derived from WT mice compared with T cells derived from LckcreIL-4Rα−/lox and C57BL/6 mice ( not shown ) . IL-10 production by macrophages [43] and CD4+ T cells [31] has been linked to susceptibility to L . major infection . Using our assay system , IL-10–secreting cells were identified as CD4+ T cells . IL-10–producing CD4+ T cells have been implicated in controlling L . major parasite survival/infection in genetically resistant C57BL/6 mice . CD4+CD25+FoxP3+ IL-10–producing natural T regulatory cells ( Tregs ) have been elegantly shown to control parasite survival [44 , 45] . More recently , a novel disease controlling FoxP3− IL-10/IFN-γ–coproducing Th1 cell population has been identified [46] . The role for Tregs in control of L . major is unclear in BALB/c mice and potentially obscured by the predominant polarized Th2 response . The moderately specific method of Treg depletion using anti-CD25 antibody has produced contradictory results either enhancing [47] or reducing [48] susceptibility to L . major infection . Certainly , IL-4 has the ability to enhance the proliferation and function of CD4+CD25+ T cells in BALB/c mice [49 , 50] . However , the generation of CD4+FoxP3+ T cells was unaffected by IL-4Rα deficiency ( unpublished data ) . Therefore , while not excluding a role for macrophage IL-10 production [43] , our data suggest that IL-10 is predominantly produced by activated/effector T cells or Tregs , and further characterization of the CD4+IL-10+ T cells is ongoing . The absence of IL-4Rα specifically on CD4+ T cells resulted in consistently higher levels of IFN-γ secretion by CD4+ T cells compared with WT mice . However , as previously shown , induction of increased IFN-γ responses alone does not guarantee control of L . major infection . Substantially increased L . major–specific CD4+ T cell IFN-γ production was observed in macrophage/neutrophil-specific IL-4Rα–deficient mice when compared with WT controls . However , infection also induced a potent polarized Th2 response , and lesion development was delayed but uncontrolled [9] . In contrast , in the absence of a polarized Th2 response , increased IFN-γ production correlated with protection against infection in LckcreIL-4Rα−/lox and C57BL/6 mice . Significant DTH responses upon injection of SLA into the footpad were observed as early as 3 wk after infection in LckcreIL-4Rα−/lox and C57BL/6 mice , but not in WT mice ( unpublished data ) . Sustained tuberculin-like DTH responses are driven by IL-12–induced IFN-γ–producing Th1 cells [34 , 51] , resulting in macrophage recruitment and activation , and are indicative of protective cell-mediated immune responses against intracellular pathogens . This was confirmed by increased IL-12 protein detected in tissue lysate of footpads of resistant mice compared with WT mice 24 h after DTH induction . Furthermore , increased levels of IFN-γ secretion were associated with increased expression of iNOS mRNA/parasite in infected footpads . Together , these results demonstrate that in the absence of IL-4Rα signaling on CD4 T cells , a polarized Th2 response , and IL-10 production , protective Th1 immune responses during cutaneous leishmaniasis result in effective macrophage activation and intracellular parasite elimination . IL-4Rα−/− mice are susceptible to L . major infection in the acute [31] or the chronic [20] phase . Despite the absence of Th1 downregulatory signals through the IL-4Rα , IL-4Rα−/− mice do not produce increased amounts of IFN-γ following L . major infection when compared with WT controls [7] . Resistance to L . major in LckcreIL-4Rα−/lox mice has therefore revealed the protective role of IL-4/IL-13–responsive non-CD4+ T cells in control of infection in BALB/c mice . Crucial to resistance in LckcreIL-4Rα−/lox mice is CD4+ T cell IL-4Rα–independent IL-4 production . Not only induced following L . major infection [7 , 31] in IL-4Rα−/− mice , IL-4Rα–independent IL-4 production has been observed in response to Nippostrongylus brasiliensis [52] and Schistosoma mansoni [53] infections and following immunization with protein precipitated in alum [54] . As our study suggests , IL-4Rα–independent IL-4 production in LckcreIL-4Rα−/lox mice drives the induction of protective responses by non-CD4+ T cells . Both IL-4 and IL-13 are able to indirectly promote protective Th1 responses . Elegant experiments have demonstrated that IL-4 is able to instruct DCs to produce IL-12 and subsequent protection from L . major infection in BALB/c mice [55] . Furthermore , IL-4 is required for protective type 1 responses to Candida [56] . IL-13 can prime monocytes for IL-12 production [57] and drive protective cell-mediated immune responses during listeriosis [58] . Indeed , levels of IL-12p35 mRNA were increased in draining LNs of LckcreIL-4Rα−/lox and C57BL/6 mice by 3 wk after infection ( Figure 5A ) , coincident with increased DTH responses ( unpublished data ) . As macrophage IL-12 production is actively downregulated by L . major [18] , it is likely that increased IL-12p35 mRNA levels in the LNs at 3 wk after infection were produced by DCs . In agreement , infected DCs appear in draining LNs in two waves; the first transient wave peaks at 24 h , and the second commences 15–21 d after L . major infection [59] . Therefore , IL-4Rα–independent IL-4 production and subsequent IL-12 production by DCs in the absence of Th2 polarization may explain the protection of LckcreIL-4Rα−/lox from L . major infection . Furthermore , the protective effect of IL-4 signaling in non-CD4+ T cells may also explain the requirement for IL-4 in effective treatments against visceral leishmaniasis [60 , 61] . In summary , in the absence of a polarized Th2 response where non-CD4+ T cells retain IL-4/IL-13 responsiveness , increased protective immune responses are induced by 3 wk in LckcreIL-4Rα−/lox mice . As IL-12 may also negate Treg cell action on activated T cells [62] , this regulation is likely to enhance beneficial Th1 responses and immunity following L . major infection in LckcreIL-4Rα−/lox mice , possibly reflecting a similar scenario in the healer C57Bl/6 . In contrast , IL-4Rα expression on CD4+ T cells allows Th2 polarization and induction of IL-10 production in the nonhealer BALB/c strain . As a consequence , Th1 responses and protective macrophage effector functions are downregulated , IL-10 is upregulated , and subsequently , BALB/c mice succumb to L . major infection in the acute phase . In conclusion , where CD4+ T cells are unable to respond to IL-4 , IL-4/IL-13 signaling in non-CD4+ T cells is beneficial in BALB/c mice following infection with L . major .
Transgenic Lckcre mice [28] back-crossed to BALB/c for nine generations were intercrossed with IL-4Rα−/− and IL-4Rαlox/lox mice to generate LckcreIL-4Rα−/lox BALB/c mice . WT littermates were used as controls in all experiments . Mice were genotyped as described previously [24] . All mice were housed in specific pathogen–free barrier conditions at the University of Cape Town , South Africa , and used in accordance with University ethical committee guidelines . All experimental mice were age and sex matched and used between 8–12 wk of age . DNA was prepared from CD3+CD4+ and CD19+ sorted LN cells from LckcreIL-4Rα−/lox , WT , or IL-4Rα−/− mice using a FACsvantage flow cytometer ( BD , http://www . bd . com ) to >99% purity . A standard curve was prepared from serial 10-fold DNA dilutions of cloned IL-4Rα exon 5 and exon 8 DNA . Primers: exon 5 forward 5′ AACCTGGGAAGTTGTG 3′ , exon 5 reverse 5′ CACAGTTCCATCTGGTAT 3′; exon 8 forward 5′ GTACAGCGCACATTGTTTTT 3′ , exon 8 reverse 5′ CTCGGCGCACTGACCCATCT 3′ . DNA was prepared from homogenized tissues samples . A DNA standard curve was prepared from serial 10-fold parasite DNA dilutions in PBS . L . major kinetoplast primers used: forward 5′ CGCCTCCGAGCCCAAAAATG 3′ and reverse 5′ GATTATGGGTGGGCGTTCTG 3′ . Real-time PCR amplification and data analysis performed using the “Fit Points” and “Standard Curve” methods as described previously [63] . IL-4Rα was detected by anti-IL-4Rα–PE ( M-1; BD ) , and leukocyte subpopulations were identified using anti-CD19 ( 1D3 ) , anti–δ-TCR ( GL3 ) , anti-CD11c ( HL3 ) , anti-F4/80 , anti–I-Ad ( AMS-32 . 1 ) , anti-CD11b ( M1/70 ) ( all from BD ) , anti-CD3 ( 145–2C11 ) , anti-CD4 ( GK1 . 5 ) , and anti-CD8 ( 53 . 6 . 72 ) mAbs , which were purified from hybridoma supernatants by protein G sepharose ( Amersham Biosciences , http://www . amersham . com ) and labeled with FITC or biotin . Biotin-labeled antibodies were detected by streptavidin–allophycocyanin ( BD ) . Dead cells were stained by 7-AAD and excluded from analysis ( Sigma , http://www . sigmaaldrich . com ) . Acquisition was performed using FACSCalibur , and data were analyzed by Cellquest ( BD ) . CD4+ T cells , positively selected by anti-CD4 Dynabeads ( Invitrogen , http://www . invitrogen . com ) to a purity of >85% as described [7] , were stimulated with serial dilutions of IL-4 , IL-13 , or IL-2 ( BD ) in complete IMDM containing 10% FCS , penicillin , and streptomycin , 1 mM sodium pyruvate , NEAA ( Invitrogen ) , 10 mM HEPES , and 50 μM β2-ME ( Sigma ) . After 48 h of incubation at 37 °C and 5% CO2 , cells were pulsed with 1 μCi ( 0 . 037 MBq ) [3H] thymidine ( Amersham Biosciences ) for a further 18 h . [3H] incorporation was measured in a liquid scintillation counter . In vitro Th1/Th2 differentiation of purified CD4+ T cells was induced as described previously [7] . Suppression assay was performed as described [20] . Briefly , adherent macrophages derived from peritoneal exudate cells elicited with 3% Brewers thioglycollate ( Difco Laboratories , http://www . bd . com/ds ) were incubated for 16 h with medium or with IL-4 , IL-13 , or IL-10 at 1 , 000 U/ml ( R&D Systems , http://www . rndsystems . com ) . Cells were subsequently stimulated with LPS ( 15 ng/ml; Sigma ) and IFN-γ ( 100 U/ml; BD ) and NO was measured by Griess reaction after 48 h . Mice were immunized subcutaneously with 10 μg of OVA in CFA ( Sigma ) and boosted at 7 and 14 d with OVA intraperitoneally . IgE production was detected as described previously [20] . L . major LV39 ( MRHO/SV/59/P ) and MHOM/IL/81/FEBNI strains were maintained by continuous passage in BALB/c mice and cultured in vitro as described previously [20] . Mice were inoculated subcutaneously with 2 × 106 stationary phase metacyclic promastigotes into the left hind footpad in a volume of 50 μl HBSS ( Invitrogen ) . Swelling was monitored every week up to a maximum of 40 wk using a Mitutoyo pocket thickness gauge ( http://www . mitutoyo . com ) . For reinfection studies , 6 wk after primary infection , mice were injected subcutaneously with 2 × 106 stationary phase metacyclic promastigotes into the contralateral footpad . Footpad swelling was monitored for 18 wk . Infected organ cell suspensions were cultured in Schneider's culture medium ( Sigma ) . Parasite burden was estimated according to a previously described limiting dilution method [20] . Total RNA from footpad or LN was purified using mini-elute columns ( Qiagen , http://www . qiagen . com ) and cDNA was generated using the Inprom-II re-verse transcription system ( Promega , http://www . promega . com ) . Primers pairs used to detect IL-12p35 message: forward 5′-GATGACATGGTGAAGACGGCC-3′ , and reverse 5′-GGAGGTTTCTGGCGCAGAGT-3′ . iNOS message forward 5′-AGCTCCTCCCAGGACCACAC-3′ , and reverse 5′-ACGCTGAGTAC CTCATTGGC-3′ . Data analysis was performed using the “Fit Points” and “Standard Curve” methods using beta-2-microglobulin as a housekeeping gene . Mice were inoculated subcutaneously with 10 μg SLA into the right hind footpad alone or with 0 . 5 μg mouse rIL-10 or 1 . 5 μg anti–IL-10Rα ( R&D Systems ) . Footpad swelling was measured every 24 h . Footpads were homogenized , and lysates were taken 24 h after induction of DTH . CD4+ T cells were positively selected using anti-CD4 Macs beads ( Miltenyi Biotec , http://www . miltenyibiotec . com ) to a purity of >90% according to the manufacturer's instructions . Thy1 . 2-labeled splenocytes were T cell depleted by complement-mediated lysis ( Cedarlane , http://www . cedarlanelabs . com ) to produce antigen-presenting cells ( APCs ) . APCs fixed with mitomycin C ( 50 μg/ml , 20 min at 37 °C ) and washed extensively in complete IMDM . A total of 2 × 105 purified CD4+ T cells and 1 × 105 APCs were cultured with SLA at 50 μg/ml , supernatants were collected after 48 h , and cytokines were analyzed as previously described [20] . IFN-γ and IL-4 were detected in footpad tissues using the method previously described [24] . L . major–infected mice; popliteal LN cells at 2 × 105 cells/well were stimulated with SLA ( 5 μg/ml ) for 24 h . Cultures were supplemented with monensin ( 2 μM ) for the final 4 h of culture . Cells were stained with anti-CD4 FITC ( mAb , GK1 . 5 ) , fixed , permeabilized , and stained with anti–IL-10 APCs ( BD ) . Values are given as mean ± SD and significant differences were determined using Student's t test ( Prism software , http://www . prism-software . com ) . | Leishmaniasis is a disease induced by a protozoan parasite and transmitted by the sandfly . Several forms of infection are identified , and the different diseases have wide-ranging symptoms from localized cutaneous sores to visceral disease affecting many internal organs . Animal models of human cutaneous leishmaniasis have been established in which disease is induced by infecting mice subcutaneously with Leishmania major . Different strains of inbred mice have been found to be susceptible or resistant to L . major infection . “Healer” C57BL/6 mice control infection with transient lesion development . The protective response to infection in this strain is dominated by type 1 cytokines inducing parasite killing by nitric oxide . Conversely , “nonhealer” BALB/c mice are unable to control infection and develop nonhealing lesions associated with a dominant type 2 immune response driven by cytokines IL-4 and IL-13 . However , mice deficient in IL-4/IL-13 signaling are not protected against development of cutaneous leishmaniasis . Here we describe a BALB/c mouse where the ability to polarize to a dominant type 2 response is removed by cell-specific deletion of the receptor for IL-4/IL-13 on CD4+ T cells . These mice are resistant to L . major infection similar to C57BL/6 mice , which highlights the role of T helper 2 cells in driving susceptibility and the protective role of IL-4/IL-13 signaling in non-CD4+ T cells in BALB/c mice . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mus",
"(mouse)",
"none",
"infectious",
"diseases",
"immunology"
] | 2007 | Deletion of IL-4Rα on CD4 T Cells Renders BALB/c Mice Resistant to Leishmania major Infection |
Safe treatment of Plasmodium vivax requires diagnosis of both the infection and status of erythrocytic glucose-6-phosphate dehydrogenase ( G6PD ) activity because hypnozoitocidal therapy against relapse requires primaquine , which causes a mild to severe acute hemolytic anemia in G6PD deficient patients . Many national malaria control programs recommend primaquine therapy without G6PD screening but with monitoring due to a broad lack of G6PD deficiency screening capacity . The degree of risk in doing so hinges upon the level of residual G6PD activity among the variants present in any given area . We conducted studies on Sumba Island in eastern Indonesia in order to assess the potential threat posed by primaquine therapy without G6PD screening . We sampled 2 , 033 residents of three separate districts in western Sumba for quantitative G6PD activity and 104 ( 5 . 1% ) were phenotypically deficient ( <4 . 6U/gHb; median normal 10U/gHb ) . The villages were in two distinct ecosystems , coastal and inland . A positive correlation occurred between the prevalence of malaria and G6PD deficiency: 5 . 9% coastal versus inland 0 . 2% for malaria ( P<0 . 001 ) , and 6 . 7% and 3 . 1% for G6PD deficiency ( P<0 . 001 ) at coastal and inland sites , respectively . The dominant genotypes of G6PD deficiency were Vanua Lava , Viangchan , and Chatham , accounting for 98 . 5% of the 70 samples genotyped . Subjects expressing the dominant genotypes all had less than 10% of normal enzyme activities and were thus considered severe variants . Blind administration of anti-relapse primaquine therapy at Sumba would likely impose risk of serious harm .
The majority of people suffering acute malaria caused by Plasmodium vivax cannot safely receive primaquine therapy to prevent multiple recurrent attacks called relapses . This extraordinary gap likely explains most of the heavy burdens of morbidity and mortality imposed by this long neglected species of parasite . Misunderstood as relatively harmless over past decades , recent work affirms an often pernicious and sometime fatal course with a diagnosis of P . vivax malaria [1][2][3][4][5] . This realization awakened concern regarding the long neglect of therapy against relapse due to the problem of primaquine toxicity in patients with an inborn deficiency of glucose-6-phosphate dehydrogenase ( G6PD ) . Glucose-6-phosphate dehydrogenase deficiency ( G6PDd ) affects more than 400 million people , or about 8% of the general population of malaria endemic nations [6][7] . This problem hampers the malaria elimination aims employing primaquine as anti relaps therapy . G6PD is an inherited , X-linked recessive trait expressed by any one of many dozens of known single nucleotide polymorphisms ( SNPs ) impairing the function of G6PD enzyme to an extent ranging from very slightly to almost completely [8] . G6PDd remains silent in almost everyone until exposure to certain infections , foods , chemicals or drugs provokes an acute hemolytic anemia ( AHA ) ranging from mild and self-limiting to severe and life threatening [9] . The severity of AHA appears to be directly correlated with the extent to which G6PD activity is impaired , and such is the basis of the classification of the many known G6PDd variants put forth by the World Health Organization [10] . The most important clinical and public health problem with G6PDd stems from the hemolytic toxicity of primaquine . This drug stands alone as the only available therapy against both onward transmission of the infection via sexual blood stages ( called gametocytes ) and in preventing relapses caused by latent forms in the liver ( called hypnozoites ) . Therapeutic doses of primaquine cause a mild to severe AHA , depending upon dose delivered and the G6PDd variant involved , and there is great variation in each in the clinical setting . Gametocytocidal therapy has been a single adult dose of 45mg primaquine base , whereas hypnozoitocidal therapy has been a daily dose of 15 or 30mg daily for 14 days ( 210 or 420mg total dose ) [11][12] . Even the single 45mg dose provoked a substantial AHA in otherwise healthy adult volunteers with the relatively severe Mediterranean variant of G6PDd [13] , but not among those with a mild A- variant common in Africa and African-Americans [14] . That toxicity , along with recognition of good gametocytocidal efficacy at a single 15mg adult dose , prompted WHO in 2012 to recommend the lower dose [15] . Hypnozoitocidal therapy applies much greater amounts of primaquine , and important differences also occur with regard to variant-specific sensitivity to the drug . When otherwise healthy G6PDd A- adult volunteers were exposed to daily doses of 15mg or 30mg primaquine , hemolysis typically did not commence until after the third or fourth day , and the nadir of hematocrit occurred on about day 7 or 8 of dosing [16] . Thereafter hemolysis seemed to cease and subjects recovered normal hematocrit despite continuous daily dosing of 30mg for many weeks [17] . Only older red blood cells ( RBC ) were destroyed and the younger RBC replacing them could manage continued primaquine dosing . AHA in those subjects was thus considered mild and self-limiting . In stark contrast , similar experiments in adult Mediterranean G6PDd variants showed exquisite sensitivity to daily dosing without induction of even the slightest tolerance of primaquine [17] . Even reticulocytes of those subjects were destroyed by primaquine challenge . Continued dosing would cause severe and threatening AHA . The danger of primaquine therapy thus hinges upon single ( gametocytocidal ) versus daily ( hypnozoitocidal ) dosing , and the ability to identify those most at risk of a relatively severe AHA , i . e . , patients having severe G6PDd variants , like Mediterranean . The majority of malaria patients live where G6PDd cannot be even crudely assessed using commercially available qualitative kits [18][19][20] . These kits are expensive , require specialized equipment and laboratory skills , and a cold chain of supply and storage [21] . Consequently , most malaria patients are not offered primaquine therapy due to the danger it poses to a minority of patients . The opportunity to prevent multiple relapses in the coming weeks and months is thus lost . Multiple preventable attacks , typically at least 3 and sometime 10 or more , occur with attendant deepening risk of morbidity , mortality , and onward transmission [22] . The benefit of withholding primaquine therapy for fear of causing harm among unscreened G6PDd patients must be weighed against the risks borne of repeated clinical attacks and new opportunities for onward transmission . Risk versus benefit deliberation of primaquine therapy should be informed by the prevalence and severity of G6PDd in any given population burdened with endemic malaria vivax transmission . The proportion of people vulnerable to primaquine therapy and the extent of their sensitivity to primaquine underpin that weighing . In the current study , we characterized the epidemiology of G6PDd at three sites on Sumba Island in the malaria-endemic eastern Indonesian archipelago . Those sites represented relatively low to high risk of malaria . We quantitatively assessed G6PD activity in blood from nearly 2000 residents of most ages and both sexes and genotyped those exhibiting impaired G6PD activity . The effort represents a primary step in beginning to grasp the likely clinical consequences of primaquine therapy without G6PDd screening by delineating quantitative G6PD activity and genotype in a population at risk . The study also lays the foundation for quantitative definition of G6PDd diagnostic device performance and pitfalls in anticipation of practical point-of-care screening later becoming available in such settings .
The study has been approved by the Eijkman Institute Research Ethics Commission ( Project Number 46 , July 29th , 2011 ) and the study has been conducted according to the principles expressed in the Declaration of Helsinki . Written informed consent was obtained from all subjects whose 8 ml of blood were taken . Parents or guardians signed the informed consent for minors . Screening of 2033 residents from two different ecosystems ( inland and coastal ) in western Sumba in the Lesser Sundas Archipelago of eastern Indonesia ( Fig . 1 ) occurred during January-February 2012 . The cross-sectional surveys were conducted at three districts in western Sumba comprising inland and coastal ecosystems . The inland ecosystem included Lendiwacu and Wairasa in Central Sumba and Palla health centres in Southwest Sumba whereas the coastal ecosystems included Kabukarodi and Lahihuruk health centres in West Sumba and Bondo Kodi in Southwest Sumba . The sample represented 1–3% of the total population in the areas assessed . The villages sampled represented those in proximity to the health centres from which the research team operated . Most villages were beyond practical reach of sampling that included quantitative G6PD assessments in the field . Residents were invited for screening and the sample thus represents all of those willing to do so . All volunteers were assessed for malaria , hemoglobin level , and quantitative G6PD activity in an improvised field laboratory utilizing a finger-stick sample of blood in EDTA micro-tubes . Subjects were classified as G6PD normal or deficient by employing 4 . 6 U/gHb G6PD as cut off point as per recommendation of the quantitative assay manufacturer’s instructions . All G6PDd subjects and 31 G6PD normal subjects who were available on the day of blood collection , having at least 10g/dL hemoglobin ( Hb ) and being more than 6 years of age were invited to submit to venipuncture ( 8mL whole blood ) for assessment of purified G6PD enzyme kinetics and genotyping at a laboratory in Jakarta . Among the 104 G6PDd residents identified , only 80 volunteered to submit to venipuncture and provided written informed consents , the balance of 24 people were either absent or declined . Finger stick blood was first placed on a clean glass slide and prepared for thin and thick blood film staining by Giemsa in the standard manner . Certified malaria microscopists read at least 100 visual fields of oil immersion ( 1000X ) magnification of the stained thick film prior to considering a slide negative . Positive slides were reported according to species observed in the microscopic examination from thin blood film . Blood from finger stick of volunteers was placed in EDTA micro-tubes ( Becton-Dickinson Microtainer ) , placed in a cool dark container , and within 8hr was quantitatively assayed for G6PD activity using a commercially available kit ( Trinity Biotech , Ireland; Cat . No . 345-B ) with deficient ( Cat . No . G5888 ) , intermediate ( Cat . No . G5029 ) and normal ( Cat . No . G6888 ) G6PD controls . The reaction was read at 340nm wavelength using a Biochrom ( UK ) WPA Biowave II UV/Vis spectrophotometer . G6PD activity was calculated from that optical density reading as instructed by the kit manufacturer . Ten microliters of EDTA blood from the microtube was put into a micro-cuvette supplied by the manufacturer of the HemoCue system ( HemoCue AB , Sweden ) and immediately read in the instrument ( Hb201+ ) of that system for hemoglobin measurement prior to the G6PD quantitative assay . A qualified phlebotomist collected 8mL venous blood from the arm into 8 . 5mL vacuum capped tubes containing acid citrate dextrose ( ACD; Becton Dickinson yellow top vacutainer ) . These tubes were immediately stored at 4°C for no more than 5 days in Sumba prior to transport by air to the laboratory in Jakarta . The tubes were also stored at 4°C in the laboratory until the day of processing . All blood was analyzed within 30 days of collection . Venous blood was processed for genotyping and G6PD purification and enzyme kinetics assays . After being spun 1500xg for 15min , the plasma was decanted by pipette , followed by transfer of the buffy coat into a 1 . 5mL Eppendorf tube held at −25°C for later genotyping . The red blood cell ( RBC ) pellet was then washed with isotonic solution ( 0 . 9% NaCl with 1mM EDTA ) and spun at 500xg for 15min 3 times before adding lysis solution ( 2 . 7 mM EDTA with 0 . 7 mM β-mercaptoethanol ) in 1:5 ratios . The hemolysate was then centrifuged at 20 , 000xg for 30 min and the supernatant poured slowly into the DEAE column ( GE Healthcare , USA ) equilibrated with Buffer 1 ( 5 mM Na-phosphate buffer , pH 6 . 4 ) . After lysate was loaded to this column , Buffer 2 ( 5 mM Na-phosphate buffer , pH 6 . 4 containing: 1mM EDTA , 20 μM NADP , 1 mM β-mercaptoethanol ) was added . Partially purified G6PD was eluted from the column using Buffer 3 ( 0 . 1 M Na-phosphate Buffer pH 5 . 8 containing: 0 . 5 M NaCl , 1 mM EDTA , 20 μM NADP , 1 mM β-mercaptoethanol ) . G6PD activity of this crude eluate ( DEAE fraction ) was measured prior to being put through a 10 KD Milipore size-exclusion chromatography spin column to concentrate and to exchange Buffer 3 with ADP Wash Buffer 1 ( 0 . 1M K-acetate + 0 . 1M K-phosphate pH 6 ) ready for next purification step . That eluate ( Millipore fraction ) was then put into 2’ , 5’ ADP-Sepharose affinity chromatography column ( BioRad , USA ) and the column was washed in ADP Wash Buffer 2 ( 0 . 1M K-acetate + 0 . 1M K-phosphate pH 7 . 85 ) , ADP Wash Buffer 3 ( 0 . 1M KCl + 0 . 1M K-phosphate pH 7 . 85 ) , and finally with Equilibration Buffer ( 50nM K-phosphate buffer containing 1 mM EDTA pH 7 . 5 ) all with flow rate of 50ml/hr . The elution buffer ( 80mM K-phosphate , 80mM KCl , 1 mM EDTA pH 7 . 85 and 0 . 4mM NADP+ ) was added and fractions 4 and 5 ( 1 ml each ) , containing purified G6PD from 2’5’ ADP-Sepharose affinity chromatography , were stored on ice no more than 4hr prior to biochemical characterization . This enzyme purification method was adapted from protocols describe elsewhere [23][24] . In each tube , 0 . 2mM NADP+ , 0 . 1M Tris HCl pH 8 . 0 , 0 . 01M MgCl2 and purified enzyme from fractions 4 and 5 of 2’5’ ADP-Sepharose affinity chromatography ( diluted 1:5 with lysis solution ) was incubated for 10 min at 30°C . Water ( blank ) or substrate ( 0 . 6mM glucose-6-phosphate , G6P ) was then added to the mixture and absorbance read at 340nm using the 25 Lambda™ UV/Vis spectrophotometer ( Perkin Elmer , USA ) for 5 min with 1 min interval for optical density ( OD ) readings . The calculation of activity was as follows: Activity ( U ) = ΔOD/min× 10 5 6 . 22×enzyme volume ( enzyme volume = μl of enzyme/ml reaction mixture ) where one unit of G6PD consists of the amount of enzyme which reduces one μmol of NADP+ per minute . One μmol/min of reduced NADP ( NADPH ) has an absorbance of 6 . 22 in a light path of 1 cm . Five different concentrations of NADP+ ( 0 . 01mM , 0 . 02mM , 0 . 03mM , 0 . 04mM and 0 . 05 mM ) and constant G6P concentration or 5 different concentrations of G6P ( 0 . 03mM , 0 . 06mM , 0 . 12mM , 0 . 18mM and 0 . 16mM ) with constant NADP+ and all reactions used constant amount of enzyme pooled from fraction 4 and 5 of 2’ , 5’ ADP-Sepharose affinity chromatography column [10] . These activities were then measured at 340 nm wavelength as described above and used to calculate enzyme activity . The frozen buffy coat from venous blood samples was thawed and its DNA was extracted using Qiagen Flexigene DNA extraction kit according to the manufacturer’s instruction . PCR of the G6PD gene was done essentially as described by Saunders et al [25] . The gene was divided into 3 segments in which internal primers were designed to amplify the entire gene starting from exon 3 to 13 . Primers ( ordered from 1stBASE ) for sequencing were designed internally from each of the 3 parts and employed a primer walking sequencing strategy [25] . Purifications of the PCR products with High Pure PCR Product Purification Kit from Roche were done prior to cycle sequencing . Sequencing reaction was performed by using the ABI Prism BigDye Terminator cycle sequencing ready kit version 31 and run on 3130 XL genetic analyzer ( Applied Biosystems , France ) . Electropherograms were visualized and analyzed with FinchTV . Nucleotide sequences were compared to the sequence of G6PD in GenBank ( accession no . NG_009015 . 1 ) for mutation identification . Subjects were classified as normal or deficient on the basis of Trinity Biotech’s quantitative G6PD activity above 4 . 6 U/gHb as normal G6PD . For those with G6PD deficiency , their enzyme activities ( phenotype ) and G6PD mutations ( genotype ) were measured . The primary outcome was the prevalence of people in the sample with G6PD deficiency . The result was stratified by study site and gender . Statistical significance of G6PDd prevalence by site and gender was evaluated by Chi-square test . Odd ratios and Fisher’s exact 95% confidence intervals were used to determine the relationship between haemoglobin , gender and parasitemia on deficient phenotype . Mean , median , standard deviation and range of G6PD enzymatic activities were calculated to determine reference values in normal and deficient subjects . An extended Wilcoxon rank-sum test was used to evaluate the trend of G6PD activities across anemia and normal Hb level . Data were analyzed using Stata 9 .
In total we covered the villages surrounding the six health centres from these three districts , with three health centres per ecosystem . Among the screened 2 , 033 subjects , 58% were female at both habitat types . The median age at the inland sites was 29 and 33 years of age ( males and females which ranged from 5–76 years and 4–80 years , respectively ) , older than at the coastal sites ( median 16 and 26 for males and females which ranged from 3–80 years old and 3–78 years old respectively ) as summarized in Table 1 . Hemoglobin levels were essentially similar between inland and coastal sites , with females consistently showing higher rates of anemia . The prevalence of G6PDd phenotype among males at the coastal sites was significantly higher than among those at inland sites ( 10 . 8% vs . 3 . 6%; OR = 3 . 2; 95%CI = 1 . 7–6 . 4; P<0 . 0001 ) . Among females , no such difference occurred ( 4 . 2% vs . 3 . 0%; OR = 1 . 4; 95%CI = 0 . 7–2 . 8; P = 0 . 282 ) . Among the 70 G6PDd subjects successfully genotyped , 32 were Vanua Lava ( 17303T→C ) , 22 Viangchan ( 19451G→A ) , 15 Chatham ( 19583G→A ) , and 1 Kaiping ( 20316G→A ) . The prevalence of malaria differed sharply between the two habitat types as shown in Table 1 . At the inland sites only 0 . 2% of the sample was parasitemic compared to 5 . 9% among coastal residents ( P<0 . 0001 ) . Likewise , the prevalence of of G6PDd among males also differed; 3 . 8% ( 16/417 ) and 10 . 5% ( 45/427 ) , respectively ( OR = 0 . 33 , 95%CI = 0 . 18–0 . 59 ) . The risk of parasitemia among the G6PD normal was slightly lower than among G6PDd , but not significantly different; 3 . 0% vs . 4 . 8% ( OR = 0 . 60 , 95%CI = 0 . 23–1 . 54; P = 0 . 29 ) . Fig . 2 shows the distributions of G6PD activity values at each of the 2 different ecosystems . Visually , the manufacturer’s recommended cut off for deficient versus normal G6PD activity appears valid . The tail of the normal distribution of G6PD activity approaches a frequency of zero at about that point at both locations . Below that level , the frequency of subjects rises , in most instances to the highest frequency at or near the lowest increment of enzyme activity ( 0–0 . 5 U/gHb ) . Put another way , G6PD activity below about 45% of normal represented the classification of deficient , and the majority of these measurements in men were <10% of normal . Fig . 2B showed that the 45% of normal cut off was even more distinct in inland compared to coastal ( Fig . 2A ) populations . Table 2 summarizes the essential statistics of G6PD activity among normal and deficient subpopulations with Hb ≥ 8g/dL , respectively , and for male and female subpopulations within each . Normal male G6PD activity ranged from 4 . 1–47 . 7 U/gHb and female G6PD activity ranged from 4 . 6 to 129 . 9 U/gHb compared to deficient male and female G6PD activities ranged from 0–2 . 83 U/gHb and 0 . 46–8 . 12 U/gHb respectively . The sample with 8 . 12 U/g Hb had been selected as a normal control for female , however gene sequencing revealed this person was heterozygous for Vanua Lava mutation . The value of 4 . 1 U/gHb in one normal male was later classified as falsely deficient in the field quantitative test since this individual showed no mutation by sequencing analysis . In this subject both genotyping and enzyme kinetics were wild type and normal ( 11 . 61 U/gHb ) , respectively . Fig . 3 illustrates the statistical summary of these measurements according to genotype and zygosity in scatterplot . The normal median of 10 . 6 U/gHb came with a great deal of variance skewed to higher G6PD levels . Relative to the mostly balanced variance among all classes of G6PDd summarized , the G6PD normal class is relatively very wide and imbalanced . This may be attributed to the outliers seen to the right of the normal distributions of Fig . 2 . Variance in G6PD activity among the three dominant genotypes observed ( Vanua Lava , Viangchan , and Chatham; VL , VC , & CT , respectively ) was very slight among the hemizygous males but was comparatively prominent among heterozygous females especially for VL ( Fig . 3 ) . The median G6PD activity among hemizygotes for VL , VC , and CT was 0 . 13 , 0 . 27 , and 0 . 12 U/gHb , respectively . These are approximately 1% to 2% of normal G6PD activity . The G6PD activities for heterozygous females of these genotypes were 5 . 3 , 1 . 9 , and 2 . 4 U/gHb for VL , VC , and CT , respectively . These medians ranged from 18% to 52% of normal G6PD activity . Note that the 95% confidence intervals for these heterozygous females all exceed 50% of normal G6PD activity at the upper end , and all reach below 8% at the lower end . Hb level as a covariate of G6PD activity is illustrated in Fig . 4 . We considered subjects having Hb levels less than 10g/dL as at least moderately anemic . The cluster of normal Hb between 10 and 15g/dL also falls along the G6PD activity norm of about 10 U/gHb . The minor cluster below , representing the G6PDd residents , also scatters evenly between 10 and 15g/dL Hb . The G6PD activity trend line for all of the Hb values above 10g/dL is relatively flat , i . e . , apparently not impacting or biasing G6PD activity measurements . In stark contrast , below 10g/dL the trend line rises sharply with diminishing Hb levels , i . e . , increasingly severe anemia . Although represented by relatively few data points , the trend was statistically significant ( P<0 . 001 ) . Age may bear upon risk of anemia , and could possibly account for the trend observed , but when G6PD activity was plotted as a function of age ( Fig . 5 ) , no trends emerged . Anemia alone profoundly impacted observed G6PD activity measurements . Table 3 summarizes the enzymatic analyses conducted on purified G6PD from study subjects . The first column from the left lists normal Km values for G6P substrate and NADP+ cofactor , along with activity of the purified enzyme . These results , though based on relatively few samples ( listed ) , essentially agreed with the quantitative results derived from hemolysate already detailed . In other words , hemizygous males ranged from 1% to 3% of normal G6PD activity , whereas heterozygous females ranged from 13% to 35% of normal . Low Km for NADP in all variants except CT indicated resistance to inhibition by NADPH and conversely high Km for NADP indicated strong competitive inhibition by NADPH which rendered the variant enzyme to be scarcely functional [26] . High Km for G6P normally resulted in little residual activity in the red cells but does not necessarily mean chronic hemolysis . However , Km values for G6P in hemizygous males for all 3 variants are well below that of normal in comparison to heterozygous females showing close to normal Km for the substrate . All the kinetics parameters ( Km for G6P and NADP ) as well as the hemolysate enzymes activities showed that the 3 dominant variants found in Sumba have very low activity in the red cell which was in line with reported Km values for each variant . These Km values generally reflect on the characteristic of the enzyme variant in vivo . Similarly , pH optimum for each hemizygous variant was found to be in line with the reported values for each ( VL is between pH 8 . 5–9 . 5 , VC is between pH 8 . 0–9 . 5 and CT is between pH 9 . 0–10 . 0 ) . However , to further characterize the enzyme according to the genotype , further biochemical experiments such as electrophoretic mobility and thermostability were needed , which were not feasible in this study due to time limitation and limited amount of extracted enzyme
The findings reported here corroborate smaller G6PDd surveys done at various locations on Sumba [27][28] . The diversity of G6PDd variants surprises in light of the ethnic homogeneity of people native to Sumba and presumed island founder effect [29][30][31] . It thus seems likely that they acquired that diversity prior to settling Sumba about 30 000–10 000 years ago [32] . In contrast , the Khmer people of Cambodia typically have higher prevalence of G6PDd and over 95% of it represented by Viangchan variant alone [33][34] . Our findings cannot be directly extrapolated to any of Indonesia’s many hundreds of other ethnic groups , but the limited data now available also suggest relatively diverse G6PDd variant representation among them [27][35][36][37][38][39][28] . The apparent diversity of G6PDd in Indonesia imposes difficulties for those evaluating risk of primaquine therapeutic policy and practice in the nation . Understanding this may be illustrated by exploring the implications to such made evident by the observations from Sumba . Each of the three dominant variants among hemizygous males expressed relatively very low residual G6PD enzyme activity—as low or lower than typically reported for the exquisitely primaquine-sensitive Mediterranean variant [40] . We would conclude that all G6PDd males resident in western Sumba would also be exquisitely sensitive to primaquine anti-relapse therapy ( not necessarily gametocytocidal therapy ) in terms of risk of AHA . Indeed , one male G6PDd subject enrolled in a study executed at western Sumba experienced a steep hemolytic crisis after being dosed for 5 days with 30mg primaquine daily following his misclassification as G6PD normal ( Syafruddin D , personal communication ) . A baseline Hb of 12 . 4 g/dL registered at 7 . 2g/dL at clinical assessment , and then 5 . 6 g/dL at admission to hospital a few hours later . After six days in hospital and blood transfusion therapy , he completely recovered . Close clinical monitoring by the research team had averted deeper harm to that research subject . The blind administration of primaquine anti-relapse therapy on Sumba , in light of our findings and that event and the reality of impractical and improbable clinical monitoring in routine practice , would appear to incur significant risk of serious harm . Screening for G6PDd prior to primaquine therapy would likely be required to protect patients diagnosed with vivax malaria on Sumba . The heterozygous females impose greater complexity to consideration of primaquine therapy and its safety even with G6PDd screening . The random inactivation of one or the other X chromosome during embryonic development ( Lyonization ) results in individual females having populations of RBC expressing G6PDd in fixed proportions ranging between 0% to 100% . This may be visualized most clearly among the Vanua Lava heterozygotes ( Fig . 3 ) . Although the mean and 95% confidence intervals for G6PD activity are well below normal , the range of values extends above the mean for normal G6PD . These values appear to represent a normal distribution between 0 and 100% of normal , as random Lyonization would yield . About half of these females may have more than 50% of their RBC populations as fully deficient as seen among Vanua Lava male hemizygotes . They also would perhaps prove vulnerable to a threatening AHA with primaquine anti-relapse therapy . The females heterozygous for Viangchan exhibited a significantly narrower and lower range of G6PD activity . The reasons for this difference are unclear , but the finding bears important clinical , public health and genetic implications worthy of follow up investigation . Non-random Lyonization somehow favoring the mutant X chromosome would perhaps explain the observations [41] . Although Lyonization is random as a whole organism , non-random or skewed lyonization does occur in individual cells or tissues [41][42] . Regardless of mechanism , the findings from Sumba suggest that Viangchan females may be intrinsically more vulnerable to primaquine therapy than females heterozygous for Vanua Lava . Chatham appears intermediate between the two ( Fig . 3 ) . Another important phenomenon observed in the survey at Sumba is also evident in Fig . 3 . The mean value of those expressing normal G6PD activity was sharply skewed upwards . Fig . 4 illustrates the basis of this skewing . Below a hemoglobin level of 10g/dL , G6PD activity rose sharply in linear fashion with decreasing hemoglobin levels . In other words , anemia appeared to directly impact G6PD activity in a quantitative manner . There are at least two possible explanations for this observation: 1 ) reticulocytemia among the anemic; or 2 ) mathematical treatment of G6PD activity estimates based on hemoglobin levels systematically biasing the estimate . Reticulocytes express the highest levels of G6PD activity , and that decays naturally as RBC age [43][44] . That subpopulation being overrepresented among the anemic would push G6PD values upwards . Alternatively , this effect may simply be due to the hemoglobin concentration being the denominator of the observed enzymatic activity for the per g/dL hemoglobin estimation . In other words , actual enzyme activity may be artificially inflated when the concentration of Hb is naturally low . This phenomenon may obscure truly deficient G6PD by falsely reporting normal activity . Kinetic parameters of G6PD activity may provide deeper insights into relative vulnerability to AHA by primaquine . NADPH competitively inhibits G6PD in the phosphate pentose shunt pathway . Further , G6PD is very strictly regulated in the cytosol where inhibition by NADPH becomes stronger with lower concentration of substrate , glucose-6-phosphate . Therefore , a high Km for NADP or low Ki for NADPH would more strongly inhibit G6PD , rendering the enzyme far less functional , whereas low Km for NADP ( high Ki for NADPH ) would result in diminished NADPH inhibition and more active G6PD [26] . These parameters obviously bear upon realized G6PD activity under conditions of oxidative stress and threat of RBC destruction by it . Our findings with G6PD kinetics hint at possibly important distinctions among the variants . Whereas Vanua Lava and Viangchan hemi/homozygotes had markedly low Km for NADP+ compared to normal ( Table 3 ) , Chatham variants were higher than normal . How this may bear upon primaquine sensitivity bears further investigation . The kinetics studies were hampered by relatively low numbers of successfully analyzed samples , and no firm conclusions may be drawn from them . The kinetics studies were severely limited by the relatively low volume of blood collected and the laborious and expensive methods of analysis required within a short span of time . Pioneering G6PD scientist Ernest Beutler expressed that low residual enzyme in a deficient sample may be a product of transportation of samples , extraction steps , length of sample storage or estimations during the biochemical parameters themselves [45] . We accept that such may have occured in our experiments , but the kinetics trends among variants and accordance with the quantitative lysate measurements offers reassurance . The sample for kinetics was not controlled for reticulocyte counts or for the presence of other inherited blood disorder common in Sumba , like Southeast Ovalocytosis , hemoglobin E , and other thalassemias . The possibility of an important sampling bias in the kinetics analyses may not be ruled out . We considered these analyses exploratory in nature rather than conclusive in findings . In summary , our large survey of G6PDd in western Sumba revealed the disorder to be prevalent , diverse and severe , informing the assessment of risk versus benefit with primaquine therapy , both with and without G6PDd screening . We observed possibly important differences in G6PD expression among heterozygous variants , and revealed anemia as the basis of G6PD measurements skewed far above normal . This epidemiological evaluation of G6PDd at Sumba highlights the complexity of this disorder in light of primaquine safety with therapy against relapse of vivax malaria . | G6PD deficiency affects over 400 million people worldwide . This enormously diverse disorder causes acute hemolytic anemia upon exposure to oxidizing chemicals , e . g . , naphthalene , some sulfa drugs , and certain antimalarials , including primaquine . The primary public health concern with G6PD deficiency involves that latter drug , the only one available for the radical cure of vivax and ovale malarias . Absent primaquine therapy , patients will suffer multiple recurrent attacks called relapses in the two years following the primary attack . Primaquine in G6PD-deficient patients triggers a mild to severe acute hemolytic anemia , depending upon dose administered and the specific variant involved . Relatively high therapeutic doses in severely deficient variants will threaten life . Malaria therapeutic policy and practice regarding primaquine may hinge upon the prevalence and severity of G6PD deficiency weighed against the therapeutic benefit of averting risk of relapse and attendant morbidity , mortality and onward transmission . In the current study we aimed to inform that weighing by characterizing the frequency and type of G6PD deficiency occurring in populations enduring endemic vivax malaria transmission on a single island in eastern Indonesia . The findings infer risk of serious harm caused by primaquine administered to residents of unknown G6PD status . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | G6PD Deficiency at Sumba in Eastern Indonesia Is Prevalent, Diverse and Severe: Implications for Primaquine Therapy against Relapsing Vivax Malaria |
The study of Onchocerca volvulus has been limited by its host range , with only humans and non-human primates shown to be susceptible to the full life cycle infection . Small animal models that support the development of adult parasites have not been identified . We hypothesized that highly immunodeficient NSG mice would support the survival and maturation of O . volvulus and alteration of the host microenvironment through the addition of various human cells and tissues would further enhance the level of parasite maturation . NSG mice were humanized with: ( 1 ) umbilical cord derived CD34+ stem cells , ( 2 ) fetal derived liver , thymus and CD34+ stem cells or ( 3 ) primary human skeletal muscle cells . NSG and humanized NSG mice were infected with 100 O . volvulus infective larvae ( L3 ) for 4 to 12 weeks . When necropsies of infected animals were performed , it was observed that parasites survived and developed throughout the infection time course . In each of the different humanized mouse models , worms matured from L3 to advanced fourth stage larvae , with both male and female organ development . In addition , worms increased in length by up to 4-fold . Serum and urine , collected from humanized mice for identification of potential biomarkers of infection , allowed for the identification of 10 O . volvulus-derived proteins found specifically in either the urine or the serum of the humanized O . volvulus-infected NSG mice . The newly identified mouse models for onchocerciasis will enable the development of O . volvulus specific biomarkers , screening for new therapeutic approaches and potentially studying the human immune response to infection with O . volvulus .
Onchocerciasis , caused by the parasitic filarial nematode Onchocerca volvulus , remains a significant source of morbidity throughout sub-Saharan Africa [1] . O . volvulus infection is commonly diagnosed through the presence of microfilariae in skin snips , but skin samples can be further analyzed by qPCR for enhanced sensitivity [2] . Antibody tests ( e . g . Ov16 [3 , 4] ) are available but do not have the ability to differentiate between past and present infections which is problematic in areas where the infection is endemic [5] . Recently , a limited number of biomarkers have been identified in the urine that can distinguish between O . volvulus-infected and non-infected individuals [6–8] . A significant obstacle for studying the biology of O . volvulus and for the development of new therapeutics and diagnostics has been the absence of small animal models . The only susceptible animal hosts for O . volvulus are chimpanzees [9 , 10] and mangabey monkeys [11] . Chimpanzees infected with O . volvulus had patent infections that lasted between 6 to 9 years with adult-worm bundles located in deep tissues with microfilariae in skin snips being detected 12–18 months post infection [12 , 13] . While immunologically intact mice are resistant to infection with the infective larvae ( L3 ) of O . volvulus [14] , adult worms within nodules have been successfully transplanted into SCID mice ( NOD . CB17-Prkdcscid/J ) with the worms surviving greater than 20 weeks [15] . This observation was confirmed by the successful transplantation of adult Onchocerca ochengi into SCID mice [16] . As an alternative approach , O . volvulus L3 were implanted in primates and rodents within diffusion chambers that consist of a Lucite ring enclosed with permeable membranes , allowing for migration of cells and other humoral factors into the diffusion chamber with simultaneous containment of the parasites . O . volvulus in diffusion chambers implanted in multiple species of rodents and primates showed limited growth of larvae in all host species [17] . Several strategies have been employed to overcome murine resistance to infection with various human pathogens . The Collaborative Cross ( CC ) is a large group of inbred mouse strains that were developed to address many of the different shortcomings found within the existing experimental mouse populations , including small numbers of homozygous strains , limited genetic diversity , and non-ideal population structures . Based on the hypothesis that there is a genetic basis for mouse susceptibility and resistance to infection , novel strains of CC mice have been identified that are susceptible to specific bacteria , viruses , and parasites of humans [18–23] . As an alternative approach to overcome murine resistance to infection , mice have been developed with greatly diminished immune responses . SCID mice were shown to be susceptible to infection with Brugia malayi while immunocompetent mice were resistant to the infection [24] . NOD . Cg-PrkdcscidIl2rgtm1Wjl/SzJ ( NSG ) mice are a highly immune-compromised strain of mice that have profound defects in the adaptive and innate immune responses [25] . The most notable defects include those in: 1 ) macrophages , dendritic cells and in the complement cascade [26–28]; 2 ) maturation of T and B cells [29]; 3 ) NK cells; 4 ) signaling of 6 different cytokines [30] and 5 ) the presence of eosinophils in the peripheral circulation and in tissue [31] . NOD-Rag1tm1Mom IL2rgtm1Wjl mice ( NRG ) are phenotypically similar to NSG mice with disruption in the B and T-Cell production [32] . Interestingly , NSG mice can support the complete lifecycle of the human nematode Strongyloides stercoralis , whereas immunologically intact mice cannot [31] . A third strategy to enhance pathogen survival in mice has been to add human-derived cells required for survival and growth [33] . NSG mice have the unique ability to support several different xenografts including human hematopoietic stem cells ( CD34+ stem cells ) that allow for the development of an immature partially-functional human immune system [25 , 27 , 30 , 34] . NOD . Cg-Prkdcscid Il2rgtm1Wjl Tg ( CMV-IL3 , CSF2 , KITLG ) 1Eav/MloySzJ ( SGM ) mice have an additional three human genes IL3 , CSF2 , and KITLG under the CMV promoter to enhance the overall microenvironment for the development of human xenografts . This results in increased numbers of CD33+ myeloid cells , B-cells , T-cells , and hematopoietic stem cells [35] . Humanized BLT mice are NSG mice that received a xenograft of human CD34+ stem cells and a transplant of human fetal thymus and liver implanted under the kidney capsule , which results in the formation of a “human immune organ” [36] . Control and potential elimination of onchocerciasis has been significantly impeded by the limited number of available drugs and the development of resistance to those therapies [37 , 38] . In addition , biomarkers to assess the infection status of treated individuals or macrofilaricidal activity are sorely lacking [6–8] . One of the critical barriers blocking drug and biomarker development has been the absence of suitable small animal hosts for experimentation . Hence , this study was focused on the development of a small animal model that would support the growth and maturation of O . volvulus and could serve to identify parasite specific biomarkers . To this end we tested multiple genetically defined mouse strains and xenografted ( with human cells ) immunodeficient mice to identify those microenvironments suitable for O . volvulus development . In so doing , we were able to develop convenient and tractable murine models that support the development of O . volvulus L3 into advanced larval stages . Moreover , we were able to use these O . volvulus-infected animals to identify parasite-derived biomarkers measurable in both urine and serum .
The parasite material was collected during the years 1994–1999 in the research facility at the Tropical Medicine Research Station , Kumba , Cameroon . The procedures used for the production of O . volvulus forest strain third-stage-larvae ( L3 ) were approved by an NIH accredited Institutional Review Board of the Medical Research Council Kumba , Cameroon ( Protocol 001 ) . The protocol was reviewed and approved annually . L3 were collected from black flies ( Simulium damnosum ) that were fed on consenting infected donors . After seven days the flies were dissected and the developed L3 were collected , cleaned and cryopreserved . The cryopreserved L3 were shipped to the New York Blood Center in liquid nitrogen and upon arrival in New York were stored in liquid nitrogen . All protocols using the L3 cryopreserved samples in this study were approved by the New York Blood Center's IRB ( Protocol 321 and Protocol 603–09 ) . All L3 samples were anonymized . All experimental procedures in mice were performed in compliance with the ethical and regulatory standards set by the NIH for animal experimentation . The animal use protocol ( 01469 ) was approved by the Thomas Jefferson University Institutional Animal Care and Use Committee . The animal care and use protocol adhered to the “Guide for the Care and Use of Laboratory Animals” published by the National Research Council , USA . Cryopreserved L3 were prepared as previously described [39–41] . Briefly , black flies ( Simulium damnosum ) were fed on consenting donors infected with O . volvulus , and after 7 days the developed L3 were collected from dissected flies , cleaned , and cryopreserved in dimethyl sulfoxide and sucrose using Biocool II computerized freezing equipment ( FTS Systems Inc . , Stone Ridge , NY ) [42] . Cryopreserved L3 were removed from liquid nitrogen storage and placed on dry ice for 15 minutes followed immediately by a 37° water bath . The L3 were then washed 5 times in a 1:1 mixture of NCTC-135 and Iscove’s modified Dulbecco’s medium ( Sigma , St . Louis MO ) supplemented with 100 U penicillin , 100 μg streptomycin ( Corning , Tewksbury MA ) 100 μg gentamicin and 30 μg chloramphenicol per ml ( Sigma ) . L3 isolated from different collection days were tested first for viability in diffusion chambers implanted in BALB/ByJ for 21 days , as previously described [17] . Batches of L3 with viabilities greater than 50% at 21 days post implantation were used in these studies . One hundred worms ( except where noted ) were then counted and loaded into 1 ml tuberculin syringes with 21 g needle for subcutaneous injection of the larvae into the nape of the neck . All mice were housed in micro-isolator boxes in a pathogen-free room at the Laboratory Animal Science Facility at Thomas Jefferson University ( Philadelphia , PA ) . Collaborative Cross ( CC ) mouse strains , Cast/EIJ , IL16680 ( CC055/TauUnc ) , AU8052 ( CC052/GeniUnc ) , AU8049 ( CC038/GeniUnc ) and OR13067 ( CC003/Unc ) , were purchased and imported from the Systems Genetics Core Facility of University of North Carolina ( UNC-Chapel Hill ) . Information about the CC strains can be found on the UNC Systems Genetics website at http://csbio . unc . edu/CCstatus/index . py . The mice were kept under temperature , humidity and light cycle-controlled conditions and fed autoclavable rodent chow and given water ad libitum . NOD-scid IL2Rgnull ( NSG ) , NOD-Rag1null IL2rgnull ( NRG ) , and NOD-scid IL2Rgnull-3/GM/SF ( SGM ) mice were obtained from The Jackson Laboratories ( Bar Harbor , ME ) . An NSG , NRG , and SGM mouse breeding colony was maintained in the Laboratory Animal Science Facility , at Thomas Jefferson University with breeding trios given acidified water and low fat 5K52 animal chow ( LabDiet , St . Louis , MO ) . The following human cell types were individually transferred into NSG mice: ( 1 ) human keratinocytes ( HaCat ) ( ATCC , Manassas , VA ) , ( 2 ) bovine embryo skeletal muscle cells ( BESM ) [43] , ( 3 ) lymphatic endothelial cells ( LEC ) ( PromoCell , Heidelberg , Germany ) , and ( 4 ) human skeletal smooth muscle cells ( HuSkMc ) ( Cell Application , San Diego , CA ) . HaCat and BESM were maintained in Dulbecco’s Modification of Eagle’s Medium ( DMEM ) ( Corning , Manassas , VA ) supplemented with 100 U penicillin , 100 μg streptomycin ( Corning ) , 200 nM L-glutamine ( Corning ) and 10% fetal bovine serum ( FBS ) ( Gemini BioProducts , West Scaramento , CA ) . LEC were maintained in EGM-2MV media ( Lonza , Walkersville , MD ) , and HuSkMc were maintained in Skeletal Muscle Growth Medium ( Cell Applications Inc , San Diego , CA ) following manufacturer’s recommendations . In the initial experiments , mice were injected with 5×106 BESM , HaCaT , LEC , or HuSkMc cells subcutaneously weekly throughout the experiment . The frequency of injection was subsequently determined by in vivo imaging experiments . Genes encoding green fluorescent protein ( GFP ) and luciferase were inserted into HuSkMc cells using lentiviral vectors following the manufacturer’s recommendations ( Cell Biolabs , Inc , San Diego , CA ) . Cells expressing GFP were isolated using the GFP marker by fluorescence-activated cell sorting using a BD FACS Aria ( BD Biosciences , Franklin Lakes , NJ ) . The isolated cells were grown in HuSkMc media as described above and 5×106 cells were injected subcutaneously into NSG mice . Mice were injected with VivoGlo ( Promega , Madison , WI ) and imaged following manufacturer’s recommendations on an IVIS Lumina XR ( Promega ) . Two approaches were used to create humanized mice containing multiple human-cell types [25 , 36] . Human umbilical cord blood was obtained through collaboration with Thomas Jefferson University Hospital Department of Obstetrics and Gynecology from full term natural deliveries . CD34+ stem cells were isolated from cord blood using magnetic assisted cell sorting ( MACS ) and cryopreserved until use . SGM , NRG and NSG mice were humanized with CD34+ umbilical cord derived stem cells by intrahepatic injection of 5x105 CD34+ stem cells into 48-hour old pups that were irradiated with 1 . 5 gray . Six weeks following injection of the stem cells , peripheral blood from the mice was screened for the presence of human cells and mice with counts greater than 600 human CD45+ hematopoietic cells per μl of whole blood were used for experimentation , following previously published protocols [31] . BLT mice were purchased from The Jackson Laboratories or were prepared following previously established protocols [27] . Briefly , NSG mice ( 4- to 6-week old ) were implanted with 1 mm3 sections of fetal thymus and liver ( Advanced Biomedical Resources , Alameda , CA ) under the kidney capsule . Two weeks post-implantation the mice were treated with busulfan ( Sigma ) ( 20 mg/kg IP ) and were injected retro-orbitally with 5x105 CD34+ stem cells , isolated from the donor fetal liver using MACS , ( Miltenyi Biotec Inc . Auburn , CA ) . Eight weeks following the stem cell xenograft the peripheral blood from the BLT mice was screened for the presence of human cells . BLT mice were screened and selected using the same protocol described above for the CD34+ cord blood mice . PCR screening for O . volvulus DNA was performed on all the infected mouse tissues to identify the presence of current or past O . volvulus larvae in that location . Mice were anesthetized and exsanguinated , and the internal organs were removed , and the skin was removed from the muscle . The muscle and skin were then divided into 100 different sections and individually frozen in 1 . 7 ml Eppendorf tubes . DNA was extracted from the tissue sections using the Promega genomic DNA kit A1125 following the manufacturer’s directions . Realtime PCR was performed using custom Taqman probes ( Integrated DNA Technologies , Coralville , IA ) against the Ov-150 [44–46] repeats and an ABI OneStep-Plus ( ThermoFisher ) . Mice were necropsied following previously established protocols for the isolation of filarial worms from tissues [47 , 48] . Briefly , mice were anesthetized using isoflurane gas and exsanguinated . The head was removed from the body of the mouse and discarded . The remaining internal organs and skin were removed from the muscles , and the muscle was divided into upper and lower sections at the bottom of the rib cage . All portions of the mouse ( muscle , skin , and all internal organs with the exception of the head ) were soaked overnight in RPMI containing 10% FBS and with 100 U penicillin , 100 μg streptomycin ( Corning ) , and emerging parasites were then collected and enumerated . Infected mice were evaluated using two criteria: 1 ) percent established , measured the proportion of mice in a group of infected animals from which live parasites were recovered; and 2 ) the geometric mean number of live worms recovered per mouse within the group . Recovered worms were placed in boiling fixative consisting of 95% ethanol ( Deacon Labs , King of Prussia , PA ) and 5% glycerol ( Fisher , Fair Lawn , NJ ) . After allowing the alcohol to evaporate , glycerol was added , and the worms were transferred to glycerin jelly ( gelatin 10 g , ddH2O 60 . 0 ml , glycerine 70 . 0 ml , Phenol 1 . 0 ml ) . Fixed worms were measured using an Olympus SZX16 dissecting scope connected to a DP26 camera ( Olympus , Center Valley , PA ) . CellSens Dimensions software ( Olympus ) was used to measure the length of the recovered worms . Serum and urine were collected and frozen as terminal procedures during necropsy . Serum was thawed on ice and 25 μL removed for processing from each mouse . Sera from 4 mice from each group/strain were pooled for maximizing the protein identifications . Abundant proteins were depleted using an affinity chromatography ( MARS-Ms-3 , Agilent ) according to the manufacturer’s directions . Urine was thawed on ice , centrifuged and then filtered through a 0 . 22 μM filter ( Corning ) . Serum and urine samples were prepared for mass spectrometry by digestion using the filter-assisted sample preparation ( FASP ) method [49] . Briefly , the samples brought to 1% sodium deoxycholate ( SDC ) , 50 mM Tris-HCl , pH 7 . 6 , 3 mM dithiothreitol , sonicated briefly , and incubated in a Thermo-Mixer at 90o C , 1 , 000 RPM for 20 min . Samples were centrifuged to clarify and the supernatant was transferred to a passivated 30 kD MWCO device ( Millipore , Merck KGaA , Darmstadt , Germany ) and centrifuged at 13 , 000g for 30 min . The remaining sample was buffer exchanged with 1% SDC , 100 mM Tris-HCl , pH 7 . 6 , then alkylated with 15 mM iodoacetamide . The SDC concentration was reduced to 0 . 1% . Samples were digested using trypsin at an enzyme to substrate ratio of 1:100 , overnight , at 37o C in a thermo-mixer at 1 , 000 RPM . Digested peptides were collected by centrifugation and the filter washed with 0 . 5 NaCl to elute electrostatically bound peptides . Digested peptides were desalted using reversed phase stop-and-go extraction tips [50] . Peptides were eluted with 80% acetonitrile , 0 . 5% formic acid and lyophilized in a SpeedVac ( Thermo Savant , Holbrook , NY ) to near dryness , approximately 1 h . Each digestion mixture was analyzed by ultra-high performance liquid chromatography tandem mass spectrometry ( UHPLC-MS/MS ) . LC was performed using an Easy-nLC 1000 UHPLC system ( Thermo Fisher Scientific , Waltham , MA ) . Mobile phase A was 97 . 5% MilliQ water , 2% acetonitrile , 0 . 5% formic acid . Mobile phase B was 99 . 5% acetonitrile , 0 . 5% formic acid . The 240 min LC gradient ran from 0% B to 35% B over 210 min , then to 80% B for the remaining 30 min . Samples were loaded directly to the column . The column was 50 cm x 75 um I . D . and packed with 2 μm C18 media ( Thermo Easy Spray PepMap ) . The LC was interfaced to a quadrupole-Orbitrap mass spectrometer ( Q-Exactive , Thermo Fisher Scientific , Waltham , MA ) via nano-electrospray ionization using a source with an integrated column heater ( Thermo Easy Spray , Thermo Fisher Scientific , Waltham , MA ) . The column was heated to 50°C . An electrospray voltage of 2 . 2 kV was applied . The mass spectrometer was programmed to acquire , by data-dependent acquisition , tandem mass spectra from the top 10 ions in the full scan from 400–1200 m/z . Dynamic exclusion was set to 15 s , singly-charged ions were excluded , isolation width was set to 1 . 6 Da , full MS resolution to 70 , 000 and MS/MS resolution to 17 , 500 . Normalized collision energy was set to 25 , automatic gain control to 2e5 , max fill MS to 20 ms , max fill MS/MS to 60 ms and the underfill ratio to 0 . 1% . Mass spectrometer RAW data files were converted to mzML format using msconvert [51] . MGF files were generated from mzML using the Peak Picker HiRes tool , part of the OpenMS framework [52] . All searches were performed on Amazon Web Services-based cluster compute instances using the Proteome Cluster interface . Detailed search parameters are printed in the search output XML files . Briefly , all searches required 10 ppm precursor mass tolerance , 0 . 02 Da fragment mass tolerance , strict tryptic cleavage , up to 2 missed cleavages , fixed modification of cysteine alkylation , variable modification of methionine oxidation and protein-level expectation value scores of 0 . 0001 or lower . Proteome Cluster builds species- and genus-specific protein sequence libraries from the most current UniProtKB distribution [53] . MGF files were searched using the most recent protein sequence libraries available from UniProtKB using X ! Tandem [54] and OMSSA [55] . XML output files were parsed and non-redundant protein sets determined using Proteome Cluster based on previously published rules [56] . MS1-based isotopoic features were detected and peptide peak areas were calculated using the FeatureFinderCentroid tool , part of the OpenMS framework [52] . Proteins were required to have 1 or more unique peptides across the analyzed samples with E-value scores of 0 . 0001 or less . Geometric means ( GM ) were used as measures of central tendency . Data were analyzed for larval growth by multifactorial analysis of variance ANOVA with post-hoc Fisher’s Least Significant Difference ( LSD ) testing in Systat v . 11 ( Systat Inc . , Evanstown , IL , USA ) . Probability values less than 0 . 05 were considered statistically significant . All experiments were performed a minimum of 2 times .
To determine if there was an underlying genetic basis for the resistance of mice to infection with O . volvulus [14] , 5 mouse strains having a wide range of genetic diversity from the CC ( Cast/EIJ , IL16680 , AU8052 , AU8049 and OR13067 ) [57] were screened for their susceptibility to O . volvulus . Five mice from each of the different strains were infected with 100 O . volvulus L3 and necropsied at 4-weeks post infection . None of the 5 strains of CC mice tested was susceptible to infection with O . volvulus L3 . To assess the role of the mouse immune system in mediating resistance to O . volvulus , immunodeficient mouse strains were assessed for their susceptibility to infection with O . volvulus . In a preliminary set of studies , 250 O . volvulus L3 were injected into 2 NSG mice . After 4-weeks the skin and muscle from the mice was divided into 100 anatomically distinct sections and qPCR for O-150 was performed on extracted DNA from each section . Twenty-three of the sections were positive from , spatially widespread regions of the body . It was concluded that parasites survived in NSG mice and had the ability to migrate extensively . This indicated that all regions of the mice had to be inspected for the presence of parasites following infection . NSG mice were infected with 100 O . volvulus L3 and necropsied at 4- and 8-weeks following infection . Infected mice had an established infection rate of 63% with a GM worm recovery of 2 . 0 ( range 1 to 4 ) worms recovered at 4-weeks following infection . At 8-weeks , 75% of mice had an established infection , with a GM recovery of 1 . 4 ( range 1 to 3 ) worms per mouse ( Fig 1A ) . The recovered worms were measured and found to be significantly increased in size ( p<0 . 0001; Min: 677 μm , Max 843 μm , GM: 717 μm ) at 4-weeks compared to L3 . They also significantly increased in size between 4 and 8 weeks ( p<0 . 0001 ) reaching a maximum of 1 , 085 μm ( Min: 700 μm , GM: 1034 μm ) ( Fig 2 ) . Because of the species-specific infectivity of O . volvulus ( humans and primates ) , we next examined the ability of nutrients or growth factors from human cells to promote the growth and maturation of the parasite . Human cells were selected for study based on the anatomical niche they typically inhabit that would parallel the niches of O . volvulus in humans . Luciferin labeled HuSkMc cells injected into NSG mice had a life span of approximately one week based on in vivo imaging . Thus , NSG mice received weekly injections of BESM , LEC , HaCaT , or HuSkMc cells . Mice were injected with 100 O . volvulus L3 immediately after the cell inoculations . At 4-weeks post-inoculation of the L3 , 60% of BESM cell engrafted mice established O . volvulus infection with a GM recovery of 1 worm per mouse with a maximum of 1 worm/mouse recovered ( S1 Table ) . Mice engrafted with LEC had 20% established infection and GM of 4 worms recovered per mouse with a range of 1 to 4 worms per mouse ( S1 Table ) . HaCaT cell engrafted mice had a 60% established infection and GM recovery 1 . 7 worms recovered per mouse with between 1 and 2 worms per mouse ( S1 Table ) . HuSkMc cell engrafted mice had an 81% established O . volvulus infection rate at 4-weeks post-infection and GM recovery 2 . 4 worms recovered per mouse with between 1 and 8 per mouse . Based on the enhanced survival of O . volvulus in mice engrafted with HuSkMc cells , extended infections were evaluated . At 8-weeks post-infection , mice with HuSkMc cells had a 58% established infection rate with a GM parasite recovery of 1 . 4 with a range of 1 to 3 per mouse . At 12-weeks post-infection , 80% of the mice had established infections at recovery with a GM parasite recovery of 1 . 6 with 1 to 3 per mouse ( Fig 1B ) . At each time point tested the parasites in HuSkMc-engrafted mice demonstrated continued parasite growth . At 4-weeks recovered parasites had GM lengths of 713 μm ( Min: 592 μm , Max: 788 μm ) , at 8-weeks lengths of 751 μm ( Min: 642 μm , Max: 1 , 081 μm ) , and at 12-weeks the GM length was 1 , 121 μm with a maximum length of 2 , 086 μm observed ( Min: 748 μm ) , representing a 4-fold increase in length over L3 ( Fig 2 ) . These growth rates represented significant changes between L3 and 4-week worms ( p = 0 . 0010 ) and between 8- and 12-week worms ( p<0 . 0001 ) . SGM , NRG and NSG ( HuNSG ) mice humanized with CD34+ umbilical cord derived stem cells were infected with 100 O . volvulus L3 . Each of these humanized mice had human hematopoietic lineage cells at a concentration greater than 600 cells per μl of blood . While a complete picture of the cell populations in these specific mice was not determined , based upon flow analysis during the screening process all humanized mice had both human B and T-cells present in their blood . At 4-weeks post-infection humanized SGM mice contained a GM of 2 . 4 ( range 1–7 ) worms/mouse , humanized NRG mice had a GM of 1 . 7 ( range 1 to 4 ) worms/mouse ( S1 Table ) . HuNSG at 4-weeks post-infection had a 69% established infection rate with a GM of 3 . 4 ( range 1–14 ) worms/mouse , and at 8-weeks post infection HuNSG mice had a 46% established infection rate and a GM of 2 . 6 ( range 1–18 ) worms/mouse ( Fig 1C ) . Larval growth in HuNSG mice was comparable to that seen in NSG and in NSG mice engrafted with HuSkMc ( Fig 2 ) . A significant increase in length was seen between the L3 and 4-week recovery ( p<0 . 0001 , Min: 395 μm , Max: 886 μm , GM: 627 μm ) , but no significant changes were seen between 4- and 8-weeks ( Min: 440 μm , Max: 886 μm , GM: 678 μm ) following infection . BLT mice humanized with fetal thymus , liver and CD34+ stem cells , which display an enhanced repertoire of the developing human cells [58] , were infected with 100 O . volvulus L3 and then necropsied at 4- and 8-weeks post infection . At 4-weeks post infection BLT mice had an established infection rate of 77% with a GM of 4 . 8 ( range 2–10 ) worms/mouse . At 8-weeks post infection BLT mice had an established infection rate of 60% with a GM of 2 . 3 ( range 2–3 ) worms/mouse ( Fig 1D ) . Growth of the parasites was equivalent to that seen in the NSG mice and the other humanized models with the maximum length reaching 1 , 080 μm at 4-weeks ( Min: 391 μm , GM: 623 μm ) and 1 , 448 μm at 8-weeks ( Min: 730 μm , GM: 949 μm ) ( Fig 2 ) . The overall growth was significant between the L3 and worms 4-weeks post infection ( p<0 . 0001 ) and between worms recovered at the 4- and 8-week time points ( p<0 . 0001 ) . During necropsy , mice were sectioned into 4 groupings: upper muscles , lower muscles , skin , and the complete set of internal organs . No nodules were found in any of these tissues upon necropsy . Worms were recovered from all four of the different tissue groupings with no apparent preference for any region of the animal in all of the mice tested . Detailed morphological analyses focused on worms recovered from BLT mice infected for 8-weeks and HuSkMc humanized mice infected for 12-weeks . Although there were clear differences in lengths of these worms , their morphological characteristics were similar . No differences in the ratio of males and female worms recovered from the mice was observed , however it was noted that most of the longer worms were females . Both the anterior and posterior ends in both sexes were bluntly rounded and only slightly tapered ( Fig 3A , 3B and 3F ) . Other than growth in length , the major change from L3 was development of the reproductive systems . In the L3 , both the female and male systems are rudimentary genital primordia consisting of only several cells . In the 8-12-week old worms , the female ovejector had formed and had attached to the body wall . The ovejector was ovoid in shape , relatively large and filled the body cavity , and had a distinct lumen ( Fig 3C and 3D ) . Rudimentary cellular growth of the reproductive tubes was also evident ( Fig 3D ) . In males , the testis , located at approximately mid-body , had become elongate in shape and had looped posteriorly to form a classic shepherd’s crook ( Fig 3E ) . In addition , the spicule pads were well developed and demarcated ( Fig 3F ) but were still oval in shape and had not yet started to take on the shape of the spicules nor was there any evidence of cuticularization . These observations are consistent with parasite development into advanced fourth stage larvae ( L4 ) . Global proteomic analyses were performed with serum and urine collected from BLT mice infected for 8-weeks and HuSkMc humanized mice infected for 12-weeks . A total of 7 , 430 proteins were identified based on the spectral matching to a combined protein database of human , mouse , O . volvulus and its Wolbachia ( wOv ) endosymbiont . Because of the ambiguity in distinguishing certain spectral matches for proteins commonly found in both humans and mice , these proteins were grouped as non-O . volvulus proteins ( 4 , 743 in serum , 2 , 836 in urine , S1 Fig ) . The present study , however , focused on only O . volvulus derived proteins as potential parasite-encoded biomarkers of O . volvulus infection . Of all the proteins identified , 155 O . volvulus proteins ( 111 in serum and 44 in the urine ) were detected in the infected mice and not in the control mice . While there were no proteins that were commonly detected in the serum and urine of the infected mice , the BLT and HuSkMc mice had 5 proteins ( OVOC11556 , OVOC835 , OVOC10244 , OVOC4009 and OVOC9087 ) in common in the serum and 5 ( OVOC7220 , OVOC4139 , OVOC224 , OVOC8249 and OVOC9267 ) in the urine ( Fig 4B ) . Almost all the proteins identified have been shown through RNAseq to be transcribed by various stages of the O . volvulus parasite ( Fig 4B ) .
The objective of this project was to identify small animal models that would support the development of black-fly derived O . volvulus L3 into advanced mammalian-adapted stages of the parasite . These small animal models are critically needed for identifying biomarkers released from the early stages of the infection and for screening potential new anthelmintics . Published findings on the susceptibility of mice to infection with the L3 of O . volvulus suggest that mice are resistant to infection when the larvae were injected subcutaneously [14] . However , O . volvulus L3 have been shown to survive and develop in mice when implanted within diffusion chambers at rates comparable to those seen in susceptible primates [17] . Genetic traits can play a significant role in the susceptibility of animals to infection as has been clearly demonstrated by the diversity of susceptibility of different mouse strains to infection with Litomosoides sigmodontis [59] . The CC mouse project was developed to produce an extremely diverse set of inbred animals that could be used for mapping different genetic traits [57] . Five different CC mouse strains , selected based on their diverse genetic backgrounds , were tested for susceptibility to O . volvulus and were completely resistant to the infection . This observation suggests that either mice are missing some integral factors required for parasite growth or the immune responses in mice are effective at eliminating the infection . The question remains as to why parasites are recovered live from diffusion chambers implanted in mice but not when injected into the tissues . There are several possible explanations including: ( 1 ) the diffusion chamber acts as a barrier from the immune response creating an immune privileged site , ( 2 ) the larvae within the diffusion chamber are blocked from migrating through the tissue releasing excretory and secretory products and thereby eliciting an immune reaction , or ( 3 ) the diffusion chamber attracts host components to the parasite microenvironment that are beneficial for parasite development . To test the hypothesis that mouse-intrinsic immune responses control O . volvulus infections , NSG mice that lack both functional innate and adaptive immune systems were infected with O . volvulus L3 . Advanced stages of the parasites were consistently recovered from the infected NSG mice , and parasites survived and developed over the 8-week time course into advanced L4 . These findings demonstrate that the mouse immune response was capable of controlling infection with O . volvulus , with elements of the mouse immune response eliminating the infection in immunologically intact mice . Many mechanisms have been described for innate immune control of nematode infections in mice [39 , 60–67] all or some of which may be effective against O . volvulus . Interestingly , tissues and cells in NSG mice provide required factors for parasite development and based on PCR analyses , the larvae actively migrated far from the infection site . The present study did not identify the point at which O . volvulus ceased surviving and developing in NSG mice . Studies with the human parasite S . stercoralis have demonstrated that the entire parasite life cycle will develop in NSG mice within 4-weeks [31] . Given the significant difference in time that it takes for O . volvulus adults to develop ( 12–15 months in chimpanzees [9 , 10] ) and their size as mature adults ( females are 50 cm [68] ) it is unlikely that the entire O . volvulus life cycle , including mating and development of microfilariae , will occur in NSG mice . In vitro studies on the development of filarial worms including O . volvulus [69 , 70] and Brugia malayi [71] have demonstrated that host cells are needed in the culture wells to optimize parasite growth and development . Furthermore , optimal development and survival of larval T . spiralis in mice requires the presence of mouse eosinophils [33] . It was thus hypothesized that adding human cells to the NSG mice , from the tissues that the parasites are normally found juxtaposed in humans , might provide additional required nutritional or developmental elements found in humans required for parasite development and survival . Four different single cell xenografts were screened: HuSkMc , LEC , HaCaT and BESM . Of these four cell lines , HuSkMc was found to support the highest average percent survival of the implanted worms and consistent infection rates over a 12-week time period ( Figs 1 and 2 ) As an alternative to adding single cell populations to the NSG mice , multipotential umbilical cord stem cells were transferred to the immunodeficient mice . NSG , NSG-SGM , and NRG mice , all of which lack functional immune responses , were humanized with CD34+ umbilical cord stem cells . The humanization of the various immunocompromised mouse strains resulted in the development of an immature human immune system . The developing immune system in these mice displays a T-independent response , limited antigen-specific IgM responses , and the presence of multiple innate immune cells has been noted [25] . When humanized NSG , NSG-SGM , and NRG mice were infected with O . volvulus L3 , higher GM parasite recoveries were observed when compared to NSG mice without any human cells but these enhancements were not significant . Although NSG-SGM did have consistent infection rates with O . volvulus we had significant difficulties establishing reliable engraftment of human cells in this strain of mice . Extending this concept further , BLT mice which contain CD34+ stem cells in addition to fetal tissue were used . BLT mice are known to develop a more mature version of a human immune system . Limited T-dependent recognition is seen within these mice and better overall functionality of both the B and T-cells has been documented [36] . BLT mice supported the highest average parasite recoveries of any mouse model tested at 4-weeks , although the average fell to be in line with the HuNSG and HuSkMc models by 8-weeks post infection . The mean lengths of the parasites recovered from 4- and 8-week time points from the NSG mice were comparable to those recovered from BLT mice . This suggests a link between the enhanced survival , may be related to the presence of the human immune cells , but the growth of the parasites that survive may not be related to the human cell byproducts . Although cellular engraftments levels in BLT and HuNSG mice were not rechecked at the end of the experiment previously published data have shown that these animals reliably hold their engraftments for extended periods of time [72 , 73] . The mouse models developed in this study offer a number of advantages over the models that currently exist . L3 implanted within diffusion chambers in both non-human primates and mice successfully molt into L4 but the lengths of the recovered parasites were significantly shorter than those recovered at 8 and 12 weeks post infection from the mice tested in this study [17] . While non-human primates can support the development of the parasites from L3 to microfilariae producing adults , cost and ethical considerations limit the utility of primates for drug discovery and antigen identification [10 , 14] . Nodules containing adult worms recovered from humans have been implanted into SCID mice and the female worms survived and released microfilariae [15] . This method allows for identification of adult antigens and biomarkers , but lacks the intermediate L3 and L4 stages of development . Finally , an alternative model using adult male O . ochengi implanted into SCID mice has been developed for the identifying filaricides [16] , which may also be effective against O . volvulus . The NSG models developed in this study have the critical advantage of working with O . volvulus , thereby allowing species-specific screening of filaricides and identification of biomarkers . It was of interest to note that the immune response found in normal mice was highly efficient at eliminating the L3 of O . volvulus , but the human immune cells in NSG mice were supportive rather than destructive of the parasites . It is possible that the mouse immune cells were evolutionally adapted to eliminate the parasites , whereas the human cells were evolutionally adapted to support the parasites as observed in nature . A comparison between mouse and human immune reactions to the worms might yield important new insights into the etiology of resistance and susceptibility to infection with O . volvulus . The source of the L3 used in these studies was from parasites cryopreserved in liquid nitrogen . After defrosting , 100 individual parasites were selected , attempting to identify only the viable/undamaged L3 . It is reasonable to predict that a percentage of larvae exiting from the mouth parts of a black fly have the potential to resume development in the human host and that the cryopreservation process damaged some of those worms . Approximately 50 percent of the larvae recovered after cryopreservation survived in diffusion chambers implanted in mice , which may explain the overall number of worms recovered from the different mouse models in this study . While minor differences in the overall larval recovery levels were observed between individual batches of cryopreserved larvae , multiple batches were combined before implantation to help ensure a consistent viability level going into the animal hosts . Even with the combination of multiple batches of larvae it cannot be ruled out that the overall viability of the injected larvae played a role in the observed recovery rates . The recovery rates of O . volvulus larvae after developing in NSG mice with or without human cells was in the same order of magnitude as that reported for the recovery of adult filarial worms , where infections were initiated by larvae recovered directly from the insect vector . These infections included Brugia pahangi in cats [74] , Brugia malayi in leaf-monkey [75] , and Onchocerca ochengi in cattle [76] . In the final analysis , optimal parasite recovery was observed in mice humanized with HuSkMc ( maximum of 10 worms ) , HuNSG ( maximum of 18 worms ) and BLT mice ( maximum of 10 worms ) . It was clear that worms increased in size during the infection period with individual worms achieving up to 4 times the size of the original L3 . Growth of worms within a single mouse was not consistent and suggests that there is significant variability within the infecting larval population . It does verify , however , that humanized mice have the potential to support extended development of O . volvulus . Both male and female worms grew in length and resumed their sexual development . Although no cast cuticles were observed , it was evident from organ development that the parasites had molted into fourth-stage larvae . Urine and serum was collected from humanized mice infected for 8-12-weeks with O . volvulus for the identification of biomarkers . Infected HuSkMc mice or BLT mice were selected for this analysis so the biomarkers identified would develop in the presence of human cells thereby potentially enhancing their specificity . Several O . volvulus-specific peptides were identified in the serum and urine of BLT and HuSkMc mice ( S3 Table ) , however , no O . volvulus-specific proteins were found in both urine and serum from either mouse source . The most likely useful biomarkers were the proteins listed in Fig 4B . Though most of the proteins were identified by one or more unique peptide ( s ) , among the proteins with unknown function ( OVOC9087 , OVOC835 , OVOC224 ) , OVOC9087 does not have orthologues in other filarial species and hence would likely be able to distinguish O . volvulus from other filarial infections . Because of the expected low number and abundance of O . volvulus-specific protein identification in the serum and urine from any given mouse in the current system , mass spectrometry was carried out with pooled serum and urine samples from each group of mice . In conclusion , novel small-animal hosts , NSG mice , have been identified that support the survival and development of O . volvulus L3 into advanced L4 mammalian stages . Humanized mice have also been shown to be effective at identifying biomarkers for early O . volvulus infections . It is anticipated that these small-animal hosts for O . volvulus will also be useful as part of the effort to identify new anthelminthic drugs . Finally , the fact that O . volvulus survives and develops in NSG mice humanized with human immune cells may provide the opportunity to study the human immune response to early infection with O . volvulus in a small animal model . | Onchocerciasis , caused by the filarial parasitic nematode Onchocerca volvulus , remains a significant source of morbidity throughout sub-Saharan Africa and is a primary cause of infectious blindness . Research on this disease has been hindered by an absence of suitable small animal hosts . Here we describe the development of humanized mouse models that are susceptible to O . volvulus infection supporting both growth and maturation of the parasite . These novel mouse models have enabled the identification of a number of new O . volvulus biomarkers with potential to play a role in the development of specific and sensitive diagnostic tests for presence of viable parasites . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"onchocerca",
"volvulus",
"helminths",
"immunology",
"biomarkers",
"parasitic",
"diseases",
"animals",
"onchocerca",
"animal",
"models",
"urine",
"developmental",
"biology",
"model",
"organisms",
"experimental",
"organism",
"systems",
"research",
"and",
"analysis",
"methods",
"animal",
"studies",
"proteins",
"life",
"cycles",
"mouse",
"models",
"immune",
"response",
"biochemistry",
"eukaryota",
"anatomy",
"physiology",
"nematoda",
"biology",
"and",
"life",
"sciences",
"larvae",
"serum",
"proteins",
"organisms"
] | 2018 | Development of Onchocerca volvulus in humanized NSG mice and detection of parasite biomarkers in urine and serum |
The ultimate goal of metabolic engineering is to produce desired compounds on an industrial scale in a cost effective manner . To address challenges in metabolic engineering , computational strain optimization algorithms based on genome-scale metabolic models have increasingly been used to aid in overproducing products of interest . However , most of these strain optimization algorithms utilize a metabolic network alone , with few approaches providing strategies that also include transcriptional regulation . Moreover previous integrated approaches generally require a pre-existing regulatory network . In this study , we developed a novel strain design algorithm , named OptRAM ( Optimization of Regulatory And Metabolic Networks ) , which can identify combinatorial optimization strategies including overexpression , knockdown or knockout of both metabolic genes and transcription factors . OptRAM is based on our previous IDREAM integrated network framework , which makes it able to deduce a regulatory network from data . OptRAM uses simulated annealing with a novel objective function , which can ensure a favorable coupling between desired chemical and cell growth . The other advance we propose is a systematic evaluation metric of multiple solutions , by considering the essential genes , flux variation , and engineering manipulation cost . We applied OptRAM to generate strain designs for succinate , 2 , 3-butanediol , and ethanol overproduction in yeast , which predicted high minimum predicted target production rate compared with other methods and previous literature values . Moreover , most of the genes and TFs proposed to be altered by OptRAM in these scenarios have been validated by modification of the exact genes or the target genes regulated by the TFs , for overproduction of these desired compounds by in vivo experiments cataloged in the LASER database . Particularly , we successfully validated the predicted strain optimization strategy for ethanol production by fermentation experiment . In conclusion , OptRAM can provide a useful approach that leverages an integrated transcriptional regulatory network and metabolic network to guide metabolic engineering applications .
Microbial-based cell factories can be used to advance environmentally friendly and economically viable industrial bioprocesses . Various strategies have been suggested to modify industrial strains to improve desired product yields . Traditional methods of strain screening mainly rely on mating , hybridization and mutagenesis techniques [1 , 2] , which are time consuming and costly , and have struggled to keep up with current industrial needs . In 1991 , Jay Bailey proposed the term "metabolic engineering" to show how using recombinant DNA and other techniques could improve specific metabolic activity in cells by manipulating enzymes , transporters , and regulation to make cells meet human-specified goals [3] . Rational strain design methods suggest particular genes or enzymes to alter in order to achieve desired strain characteristics for metabolic engineering [4] . Systems biology is a powerful approach to uncover genotype-phenotype relationships , which can guide rational design-build-test iterations on strains to improve phenotypic properties in metabolic engineering . Next-Generation Sequencing ( NGS ) [5] and semi-automatic annotation techniques [6] have produced an increasing number of well annotated microbial genomes , enabling the collection of reasonably comprehensive information about which metabolic enzymes are encoded . This information has greatly contributed to the reconstruction of the genome-scale metabolic models of various organisms [7] . GEnome-scale metabolic Models ( GEMs ) are mathematical representations of the complete network of known biochemical reactions that can occur in a particular cell , assembled as a collection of metabolites , reaction stoichiometries , compartmentalizations , and gene-protein-reaction associations [8 , 9] . One of the main analysis approaches of GEMs is the well-known Flux Balance Analysis ( FBA ) [10] , which can predict phenotypes for cells under different genetic and environmental conditions based on the stoichiometric matrix without requiring kinetic parameters [11 , 12] . It has been demonstrated that computational simulation on GEMs can predict effective engineering strategies for strain design [13 , 14] . Since the first strain design method OptKnock [15] was proposed in 2003 , several computational methods for efficient automated identification of genetic strain modifications have been developed , such as RobustKnock [16] , OptGene [17] , OptORF [18] , GDLS [19] , and FSEOF ( Flux Scanning based on Enforced Objective Flux ) [20] . These algorithms have already yielded successful strain design applications . In an early example , Fong et al . designed E . coli strains for lactate production with a maximum 73% increase by using OptKnock [21] . Researchers from Tianjin University utilized a GEM of B . subtilis and elementary mode analysis to design an engineering strain for isobutanol production , and experimentally verified a 2 . 3-fold increase compared to wild type strain [22] . Recently , Otero et al . designed a strain using OptGene to overproduce succinate in S . cerevisiae , and experimentally validated a 43-fold improvement in succinate yield on biomass after directed evolutions [23] . However , a metabolic model alone has a significant limitation in revealing condition-specific metabolic activity [24 , 25] because gene regulation plays an important role in constraining the particular metabolism available under any given condition . Also , the complex crosstalking mechanisms between gene regulation and metabolism are not captured by a metabolic model alone . To overcome the limitation , methods that systematically integrate a transcriptional regulatory network and a metabolic network have been developed [26] , including regulatory Flux Balance Analysis ( rFBA ) [27] , steady-state rFBA ( SR-FBA ) [28] , Probabilistic Regulation of Metabolism ( PROM ) [29] , and Integrated Deduced REgulation And Metabolism ( IDREAM ) , developed by our group [30] . Modification of gene regulatory circuits is an important strategy for transforming engineering strains [14] . In fact , modifications of regulatory factors ( e . g . upregulation of biosynthetic pathway activators ) contribute to more than half of the genetic operations in E . coli and S . cerevisiae engineering strains , but most of these interventions are based on human intuition [31] . Therefore , some strain design methods have utilized transcriptional regulation information to propose more effective metabolic engineering strategies . OptORF [18] was the first approach using integrated regulatory-metabolic models , which followed the framework of two-layer optimization as did OptKnock . In 2011 , the heuristic strain design method OptGene also updated a version which introduced integrated regulatory-metabolic models [32] . In 2012 , a series of approaches based on minimal cut sets ( MCSs ) was further developed to include a new tool ( rcMCSs ) , that incorporates regulatory constraints [33] . However , since the above algorithms used manually curated integrated regulatory-metabolic models , where the regulatory network is a Boolean network , there are some limitations to application . Firstly , only some well-studied microorganisms may have existing integrated networks , such as E . coli [34] , M . tuberculosis[35] , and yeast [36] . Reconstructing such models requires extensive manual adjustment and additional information for generating Boolean logic rules in the regulatory network [37] , which hinders the ability of these algorithms to be broadly applicable across many organisms . Secondly , these algorithms have to assume that the target gene is completely active or inactive , which ignores the range of possible regulatory intensities between regulatory factors and the target genes . In addition , Boolean networks can only suggest the manipulation of transcription factor by knockouts ( ON to OFF ) and cannot provide guidance for more quantitative adjustment of transcriptional regulation . Recently , a method named Beneficial Regulator Targeting ( BeReTa ) , used gene expression to infer the interaction between regulatory factors and target genes , combined with FSEOF [20] for identifying transcription regulators to enhance desired production . According to the correlations between the different transcription factor expression levels and target reaction flux rates , beneficial scores are calculated to judge whether the transcription factor can enhance or inhibit the target reaction . The algorithm was applied to E . coli , as well as S . coelicolor , which does not currently have an integrated metabolic-regulatory model . BeReTa represents a significant advance , but it cannot predict an expected product rate or yield of the mutant , or make predictions about the combined manipulations of multiple sites . Herein , we report a new strain design algorithm , named OptRAM ( Optimization of Regulatory And Metabolic Networks ) , which can identify combinatorial optimization strategies including overexpression , knockdown or knockout of both metabolic genes and transcription factors , based on our previous IDREAM integrated network [30] . OptRAM also aims to achieve optimal coupling between biomass and target production , and can be used for strain design of bacteria , archaea or eukaryotes . The other advantage of OptRAM compared with previous heuristic approaches is that we systematically evaluated the implementation cost of different solutions and selected strain designs which are more likely to be achieved in experiments .
Saccharomyces cerevisiae S288c has been studied and simulated extensively through a series of models [38] reconstructed based on the genome sequence and literature annotations [39] . We used the latest metabolic reconstruction , Yeast 7 . 6 , which includes 3493 metabolic reactions , 2220 metabolites and 909 metabolic genes . Integration of a gene regulatory network with a metabolic network at the genome-scale poses significant challenges , in part because they are distinct network types requiring very different modeling frameworks . PROM uses probabilities to represent gene states and TF–gene interactions from abundant gene expression data , and then uses these probabilities to constrain the fluxes through the reactions controlled by the target genes [29] . A limitation of PROM is that a pre-built transcriptional regulatory network is required as an input . In our previous work , we developed a framework called Integrated Deduced REgulation And Metabolism ( IDREAM ) , which uses bootstrapping-EGRIN-inferred [40 , 41] transcription factor ( TF ) regulation of enzyme-encoding genes , and then applies a PROM-like approach to apply regulatory constraints to the metabolic network . In Yeast , we collected 2929 microarray datasets with 5939 yeast genes and evaluated 392 of those genes as possible regulators . For each of the 5939 target genes , we constructed separate models from 200 randomly selected subsets of the 2929 experiments , as well as a 201st model constructed using the entire data set . This resulted in 201 generated gene regulatory models for each of the 5939 yeast genes , for a total of 1 , 193 , 739 models . For each gene , we estimated a false discovery rate ( FDR ) for each factor by tallying the fraction of models from random subsets that identified that factor as a regulator . Thus , if factor X was predicted to regulate gene Y in 191 of 200 models , then X would have an FDR = 1–191/200 = 0 . 045 . If X is predicted to activate Y with an FDR of 0 . 045 , only 4 . 5% of Y’s activity would be predicted to remain if X was deleted . If X is predicted to deactivate Y , then we use the much larger 1-FDR ( e . g . , 95 . 5% of activity ) to represent that Y is somehow disturbed without a significant reduction in activity . We included only those interactions that passed an FDR cutoff of 0 . 05 . Then we predicted whether a factor was an activator or repressor by testing if its mRNA expression was correlated or anti-correlated with the expression of its target using the model from the entire expression dataset . Finally , we retrieved an integrated regulatory-metabolic network including 2626 inferred influences consisting of 91 TFs transcriptionally regulating 803 genes encoding enzymes of the metabolic network . It should be noted it is impossible to cleanly differentiate between the false discovery rate and the strength of the regulatory role for multi-cell microarrays or RNA-Seq because from bulk measurements we can’t differentiate a strong regulation occurring in a small portion of cells from a weak regulation happening in a large fraction of the cell population . In other words , because we are using a ground up mixture of cells , then we might reasonably expect a few cells with high expression related to a rarely used promoter to appear similarly to a low level , frequently used promoter . Perhaps single cell RNA-Seq will help disentangle this problem in the future . OptRAM is a meta-heuristic strain optimization method based on an integrated model of an inferred regulatory network and a constraint-based metabolic network . It aims to identify the modifications of TFs and metabolic genes , including overexpression , knockdown , and knockout , to achieve the maximal production of desired chemical . OptRAM will simulate a series of mutations to get the optimized strategy for target overproduction by simulated annealing . We adopt 11 kinds of mutations on gene expression of TFs or metabolic genes , represented as FC ( TF ) and FC ( G ) , with overexpression and knockdown fold change of 2 , 4 , 8 , 16 , 32 respectively , and 1 knockout , as shown in Table 1 . The expression level of these genes will be translated to corresponding metabolic reactions by the integrative network . First , expression levels of metabolic genes are calculated according to expression of corresponding transcription factors . In the EGRIN algorithm , a linear equation of the target gene and the TFs is generated: target=coeff1TF1+coeff2TF2+⋯+coeffnTFn ( 1 ) Where the variable target is the expression level of a target gene regulated by n TFs , TFi are the expression level of these TFs , and Coeffi are the corresponding coefficients of each TF . In OptRAM , for a target gene regulated by one TF , tfExpr is the relative expression level of the mutated TF , then the relative expression level of the target gene is calculated as: targExpr=2coeff×log2tfExpr ( 2 ) When a target gene is affected by more than one TF , the expression level of the target gene is calculated as: targExpr=2∑incoeffi×log2tfExpri ( 3 ) Having the relative expression level of all metabolic genes , the change of relevant reactions , represented as FC ( R ) , is calculated according to the gene-reaction rules in the metabolic model . For reactions with ‘AND’ rules of different genes , we selected the minimum value of relative gene expression level , since ‘AND’ indicates that the combination of multiple enzymes is required , so the enzyme with lower expression determines the upper bound of the reaction rate . For reactions with ‘OR’ rules of different genes , the mean value of relative gene expression level is calculated , because ‘OR’ indicates that multiple enzymes have the same catalytic function and can be substituted for each other . Therefore , the average expression level of enzymes in the set can better reflect upper bound of the reaction rate . While the average was used herein , it would also be reasonable to use the max of the enzymes being expressed in this scenario . In order to simulate the flux change of reactions induced by the gene expression mutation , we first need a reference flux value for each reaction , which is obtained by pFBA ( Parsimonious enzyme usage FBA ) method [42] . pFBA is an algorithm based on FBA . For a metabolic network with M metabolites and N reactions , the FBA formulation is shown below: MaximizevobjectiveSubjectto∑j=1NSijvj=0 , i=1 , … , Mlbj≤vj≤ubj , j=1 , … , N ( 4 ) Where Sij stands for the stoichiometric coefficient of metabolite i in reaction j , and vj stands for the flux of reaction j , lbj and ubj are the constraints for reaction j . The most commonly used objective function ( vobjective ) is biomass synthesis [43] . Here the simulation condition for yeast metabolic flux was set corresponding to the YPD medium , with glucose 20 g/L and blocking other carbon source . The pFBA algorithm is divided into three steps . First , the max biomass rate is obtained by FBA with the original model . Secondly , the constraint of biomass is set equal to the max biomass value . Finally , a new objective function is set as the minimization of total flux values carried by all reactions , to generate the flux distribution . According to the reference flux values from pFBA and the level of expression change for mutated reactions , we set the new constraints of reactions as shown in Table 2 , where FC ( R ) is the change of reaction , v is the reference flux value , lb and ub are the original lower bound and upper bound for the reaction . In the previous meta-heuristic strain optimization methods , such as OptGene , BPCY ( biomass-product coupled yield ) is used as the objective function [17]: BPCY=Product×GrowthSubstrate ( 5 ) Where Product represents the flux of the reaction synthesizing the desired product , Growth represents the flux of biomass , and Substrate represents the uptake rate of the nutrient substrate . The ultimate goal of the optimization algorithm is to identify the mutated solution with the largest BPCY value , which ensures a considerable growth when improving the target product . A limitation of the simulation using pFBA is that this framework does not guarantee that the target reaction flux will be coupled to biomass . That is , even if the BPCY score of a mutated solution is high , the flux value of the target reaction is unstable with the max biomass . Because the flux variability of target reaction is a wide range and the minimum flux may even be zero , there is no guarantee that the target product can have a certain output under natural growth . Moreover , since the objective function of pFBA is biomass , there is often no flux through the desired target reaction , although the flux range of that reaction may be 0 to a positive value . In this situation , BPCY remains 0 and the algorithm reports no feasible solution . Therefore , we defined a new objective function ( Eq 6 ) in OptRAM to couple maximizing biomass production and the target reaction of interest . Where Target=Vmax+Vmin2 , Range=Vmax−Vmin2 , Vmax is the maximum flux value of target reaction and Vmin is the minimum flux value by FVA ( flux variability analysis ) [44] . Target means the average flux value of target product . Range is set to half of the interval between min and max target flux value . When Vmin is 0 , RangeTarget=1 , the coefficient ( 1−logRangeTarget ) is 1 . And when Vmin is greater than 0 , RangeTarget>1 the coefficient will be greater than 1 , which is a reward coefficient for BPCY . Compared to BPCY , this objective function will induce solutions to have a higher and narrower flux range of target product , which reduces the uncertainty caused by alternative solutions in constraint-based modeling . Hence , by using the refined objective function , OptRAM can provide solutions with better biomass-product coupled . Fig 1 illustrates the strategy of the OptRAM approach , and the detailed pipeline can be downloaded from supplemental files ( S1 Script and S1 Code ) . OptRAM requires a transcriptional regulatory network ( or a gene expression data set ) and a genome-scale metabolic model as input . Then IDREAM method will be run to get an integrated model . For organisms with no existing TRN , users can input a set of expression data from which the IDREAM method will automatically infer the TRN and generate an integrated model . Then the core strain design process within OptRAM will simulate a series of mutations to get the optimized strategy for target overproduction . The output from OptRAM includes the maximized objective score , flux of the target reaction , and the corresponding mutated solution with suggested modification of TFs and/or metabolic genes . We used simulated annealing for the core strain design part , which is able to accept a worse solution in the early stage ( avoiding getting stuck in local maximal ) , to pursue finding the global optimal screening for mutated models . The simulated annealing algorithm is derived from the simulation of the solid annealing process , an idea first proposed by Metropolis in 1953 [45] . In 1983 , Kirkpatrick et al . introduced the idea of simulated annealing algorithm into the field of optimization problems , making the algorithm practical in engineering [46] . The simulated annealing algorithm introduces Metropolis criterion , which help escaping local optimum , to accept new solutions , including not only better solutions but also worse ones , according to the probability . By simulating the drop of temperature , the algorithm controls parameters during the process and gives an approximate optimal solution in polynomial time . In our algorithm , we replace the internal energy of the annealing process with the refined objective function Obj ( Eq 6 ) , which is the prospective score for each mutated strain . The following steps show the implementation of the simulated annealing algorithm in our specific optimization problem: 1 . Initialize the simulated annealing parameters , including the initial temperature T0 of the control parameter T , the attenuation factor ( α<1 ) , and the maximum number of iterations L at each temperature . Then generate the initial solution , Ind0 . 2 . When T = T ( k ) , search L times according to the following process: ( 1 ) For the current solution Indk , randomly mutate the expression of TFs and metabolic genes , translate to effects on reaction flux , and get a new Obj score and a new mutation solution Ind'k . ( 2 ) Calculate ΔObj = Obj ( Ind'k ) —Obj ( Indk ) , where Obj ( Ind ) is the value of objective function for each solution . ( 3 ) If ΔObj > 0 then Ind'k is received as the new solution , let Indk = Ind'k; otherwise generate a random number R on the even distribution in ( 0 , 1 ) , calculate the probability P according to Metropolis criterion: P=eΔObjT ( k ) ( 7 ) If R<P , then accept the new solution , let Indk = Ind'k; otherwise keep the current solution; ( 4 ) If the number of iterations is less than L at this temperature , repeat step 2; otherwise , proceed to step 3 . 3 . If the convergence condition is satisfied , the algorithm ends and the current solution is the optimal solution , represented by a two-dimensional array of the mutated strain . The first array contains all the mutated gene IDs . The second array stores the expected fold change of expression level of the mutated gene over the wild type expression . The convergence condition is that the value of objective function has not improved for a number of continuous temperatures ( here we set 100 temperatures ) ; otherwise , proceed to step 4 . 4 . Decrease the control parameter T , let T ( k + 1 ) = T ( k ) ×α , simultaneously reset the number of iterations and go to step 2 .
We used flux variability analysis on the Yeast7 . 6 model to analyze the range of possible flux values for all reactions and excluded the reactions that could not have a non-zero value since these would not have any effect on the strain optimization calculations . The processed metabolic model has fewer reactions , metabolites and genes compared to the original Yeast7 . 6 model ( Table 3 ) . By using the IDREAM approach , we integrated the deduced regulatory network from 2929 microarray datasets with the Yeast7 . 6 metabolic network to generate an integrative gene regulatory-metabolic network ( Table 3 ) . We performed 10 parallel simulated annealing runs , which set succinate as the target product , and obtained 10 solutions ( Supplemental S3 Table ) . We first filtered the solutions with in silico knockdown or knockout of known essential genes . Then we selected solutions with the maximal fluxes of the target reaction . For solutions with the same predicted yield in succinate , we chose the one with a smaller implementation cost and a larger Cosine function representing smaller variation of genetic manipulation . Finally , we selected one optimized strain design from ten runs , with the mutation sites and expression modifications shown in Table 4 . The upper two rows show mutation sites with overexpression manipulation , while the rows below show mutation sites with knockdown manipulation . Corresponding numbers are perturbation fold change of expression as defined in Table 1 . PDR1 and PHO4 are TFs , and others are metabolic genes . For this complicated combination of TFs and metabolic genes , we analyzed the critical sites corresponding to metabolic reactions . A ‘critical reaction’ here means that the removal of the reaction can cause more than a 10% reduction in the objective function ( See Methods ) . There are 5 critical reactions ( marked in Fig 3 ) involved in this solution: up-regulation in cystathionine beta-synthase , down-regulation in ferrocytochrome-c: oxygen oxidoreductase , pyruvate decarboxylase , citrate transport , and oxoglutarate/malate exchange . There are two main processes in the production of succinate from glucose ( Fig 3 ) . First , glucose is used to make pyruvate in cytoplasm . Next , pyruvate converts to oxaloacetate , then to malate , then to fumarate , and finally to succinate . Comparison of the flux values in the two models revealed that the succinate exchange flux increases about 66-fold theoretically in the strain-optimized model . All critical reactions we found are closer to the path from pyruvate to succinate . Up-regulation in cystathionine beta-synthase will promote more L-cystathionine to pyruvate . Decreasing the constraints of ferrocytochrome-c:oxygen oxidoreductase and pyruvate decarboxylase directly affects the pyruvate flux from lactate and acetaldehyde . So pyruvate mostly comes from phosphoenolpyruvate in the mutated strain . Meanwhile , citrate transport and oxoglutarate/malate exchange are limited to prevent malate in cytoplasm to transport to mitochondrion , which enable more malate flux to succinate synthesis . We compared the best solution in Table 4 with the previous literature in which the succinate was successfully improved in yeast assisted by strain design method [23] . This study used OptGene to get the optimized strategy including deletion of SDH3 , SER3 and SER33 . SDH3 ( Succinate dehydrogenase ) catalyzes the reaction from succinate to fumaric acid , its deletion will obviously promote flux to succinate . Deletion of SER3 and SER33 can facilitate TCA Cycle , and improve succinate . We also identified SER1 as engineering site by OptRAM . To compare with other strain design methods , we chose the widely used OptFlux [50] to find overexpression , knockdown , and knockout of metabolic genes . For the mutant models generated from literature , OptFlux , and OptRAM , we used FVA to investigate the range of target reaction under constraints of being able to achieve 99% and 50% biomass respectively . The 99% biomass constraint means that the lower bound of biomass reaction is set as the 99% of maximum biomass in each mutant model respectively , similarly for 50% biomass . The 99% biomass is a relatively strict constraint on biomass and the 50% biomass is a loose constraint . A strict constraint on biomass will make the flux range of target reaction much more narrow . We compared the resulting flux range of these mutant models with the wild type model ( Table 5 ) . When biomass is constrained to 99% of the maximal theoretical value , our strategy makes an obvious improvement for predicted succinate production , even the minimal succinate exchange flux is higher than the max value in wild type and the strategy in literature [23] . Although the max biomass flux is relatively low , it is still acceptable for strain growth . When biomass is constrained to half of the theoretical max value , the model comparison suggests that our strategy makes succinate production strongly coupled with growth . The performance of solution from OptFlux seems good , but the overall interval in OptRAM solution is much larger than other strategies . We also compared all the mutation sites from 10 solutions with the experimental design in LASER database , as shown in S4 Table . There are 9 genes presenting in more than 2 solutions matched with the experimental modifications in LASER , such as MDH2 , SDH3 , and several mutated TFs , including CAT8 , HAP2 , have targets modification improving succinate validated in LASER . Overexpression of CAT8 can promote succinate production by increasing flux of glycolysis and TCA Cycle , DIC1 and ICL1 are regulated by CAT8 , which have significant effects on succinate reported by Agren et al [51] and Otero et al [23] . HAP2 is a global regulator responsible for respiration gene expression , one of its targets IDP1 has been validated to improve succinate [52] . We performed 10 parallel simulated annealing runs with 2 , 3-butanediol as target product ( Supplemental S3 Table ) . According to the filtering process similar with succinate case , we selected one optimized solution from the results of the ten runs . Table 6 shows the mutation sites suggested by OptRAM , among which GLN3 and RTG3 are TFs . There are 4 critical metabolic reactions ( marked in Fig 4 ) involved in this mutated model , respectively catalyzed by mitochondrial alcohol dehydrogenase , ferrocytochrome-c:oxygen oxidoreductase , malate dehydrogenase , and cytoplasmic alcohol dehydrogenase ( acetaldehyde to ethanol ) , all of them are predicted to be knocked down . There are two main processes in the production from glucose to 2 , 3-butanediol ( Fig 4 ) , just as for succinate . Pyruvate comes from glucose and is converted to acetoin and then to 2 , 3-butanediol . By comparing the flux values in the two models ( mutated and wild type ) , the flux of 2 , 3-butanediol exchange reaction increases about 61-fold theoretically in the mutated model . Reduction of ferrocytochrome-c:oxygen oxidoreductase will affect pyruvate in cytoplasm coming from lactate . Repressing alcohol dehydrogenase and mitochondrial alcohol dehydrogenase , both of which convert acetaldehyde to ethanol , will prevent acetaldehyde to ethanol . Downregulation of malate dehydrogenase changes the direction of the reversible reaction from malate to more oxoglutarate , and promote the flux of pyruvate to final 2 , 3-butanediol . We also compared the best solution in Table 6 with the previous literature in which Ng et al succeeded in improving 2 , 3-butanediol in yeast [53] . They used OptKnock to explore optimization design including the deletion of ADH1 , ADH3 and ADH5 under an anaerobic condition . We used OptRAM to identify knockdown of ADH3 , and the TF STE12 regulating ADH1 and ADH3 to improve 2 , 3-butanediol . Also , we ran OptFlux to get strain design solution for comparison . For the mutant models generated from literature , OptFlux , and OptRAM , we used FVA to obtain the range of 2 , 3-butanediol target under constraints of being able to achieve 99% and 50% biomass , respectively . The comparison of mutant models with wild type model is shown in Table 7 . Similarly , when biomass is constrained to 99% of the max theoretical value , even the minimal flux value of 2 , 3-butanediol exchange reaction by OptRAM is higher than other strategies . While the minimum predicted target flux in OptFlux drops to 0 with the constraints of 99% max biomass , it means the good performance of OptFlux for succinate production coupled with biomass might be a case-specific outcome , whereas OptRAM can ensure the coupling for overproduction of different targets , because of the improved objective function . We also compared all the mutation sites from 10 solutions with the experimental design in LASER database ( Supplemental S4 Table ) . There are 6 genes presenting in more than 2 solutions matched with the experimental modifications in LASER , such as ADH3 , GPD2 . Knockdown of ADH3 ( alcohol dehydrogenase ) can keep more flux to acetaldehyde and impress the flux to ethanol . Deletion of GPD1/2 ( glycerol-3-phosphate dehydrogenase ) can improving ethanol by effectively decreasing flux to glycerol [54] . We also found several mutated TFs , such as STE12 and OAF1 , have targets modification improving 2 , 3-butanediol validated in LASER . STE12 is an important global regulator for yeast growth , whose target genes including ADH1/2 , ALD6 , BDH1 , GPD1/2 , have been reported as effective modification for improving 2 , 3-butanediol [49] . OAF1 also regulates ADH1/2 . It demonstrated that by our strain design method based on integrated model , the global TF could be identified for modification to accomplish the roles of several metabolic genes . We performed 10 parallel simulated annealing runs with ethanol as target product ( Supplemental S3 Table ) . According to the similar filtering process , we selected one optimized solution from the results of the ten runs . Table 8 shows the mutation sites suggested by OptRAM . There are 3 critical metabolic reactions ( marked in Fig 5 ) involved in this mutated model , respectively catalyzed by ( R , R ) -butanediol dehydrogenase , ferrocytochrome-c:oxygen oxidoreductase and malate dehydrogenase in cytoplasm , all of them are predicted to be knocked down . We experimentally implemented the OptRAM-based design by modifying yeast strains using CRISPR-Cas9 and measuring the effects on ethanol production by fermentation . Three genes ( BDH1 , MDH2 and COX4 ) in totally different pathways were firstly considered simultaneously to enhance the ethanol pathway , which were deleted separately as well as in combination . The growth of the modified strains was slightly reduced compared to wildtype with all of the strains showing an increase in ethanol production ( Fig 6 ) . The performance of these genes depends on their molecular function in yeast cells . BDH1 encoding 2 , 3-butanediol dehydrogenase had the smallest effect on growth and ethanol production . MDH2 and COX4 encode cytoplasmic malate dehydrogenase in the TCA pathway and subunit IV of cytochrome c oxidase in mitochondrial inner membrane electron transport chain , respectively . These two pathways are primary competitors to the fermentation pathway for carbon flux . Thus the significant ramping up of ethanol , as well as growth decline , was observed on the yeast cells without MDH2 or COX4 . Since the metabolic burden and metabolic optimization are two sides of coin , the fermentation results of double deletion become much more complex . The yeast without genes BDH1 and COX4 produced the highest ethanol , but the yeast without BDH1 and MDH2 even could not keep consistent ethanol titer . Taken together , the prediction of OptRAM for three key genes to improve ethanol fermentation has been partially validated . However , the metabolic burden must be taken into consideration especially when carrying out multiple gene manipulation on microbes . We also compared the best solution with the previous literature in improving ethanol in yeast [55] . The best strategy in this study included the deletion of GDH1 and overexpression of GLT1 and GLN1 . Also , we ran the optimization algorithm OptFlux to get its strain design solution . For the mutant models generated from literature , OptFlux , and OptRAM , we used FVA to obtain the range of ethanol target under constraints of being able to achieve 99% and 50% biomass , respectively ( Table 9 ) . When biomass is constrained to 99% of the max theoretical value , even the minimal flux value of ethanol exchange reaction in OptRAM is higher than other strategies . When biomass is constrained to half of the theoretical max value , we can still see our strategy make ethanol production coupled with growth . In addition , we compared all the mutation sites from 10 OptRAM solutions with the experimental design in the LASER database ( Supplemental S4 Table ) . MDH2 ( malate dehydrogenase ) is the top recurrent mutation found by all 10 solutions , whose deletion represses the TCA Cycle , increases flux of anaerobic respiration , and improves ethanol production [56] . BDH2 is another recurrent mutation found by 8 solutions , whose knockdown can increase flux from acetaldehyde to ethanol . There are also some TFs having targets whose modification improves ethanol validated in LASER . HCM1 is found as a TF to modulate using OptRAM and it regulates ACC1 , GLN1 , and TKL1 , all of which have been reported to improve ethanol production as reported in LASER [57–59] . For example , knockdown of HCM1 can repress ACC1 ( acetyl-CoA carboxylase ) , and make more flux from pyruvate to ethanol . We systematically compared our OptRAM method with existing representative constraint-based strain optimization methods . OptKnock is the first method in this field , OptGene is the first trial of meta-heuristic method , OptORF is the first strain design method utilizing regulatory information , and BeReTa is the latest method utilizing regulatory information , which can suggest manipulations of transcription factors to be knocked down or overexpressed . Table 10 shows the different properties of these methods . We can see that only OptORF , BeReTa and OptRAM integrate a regulatory network . OptORF utilizes the Boolean network which has more limitations in practice , while BeReTa cannot give a combination of multiple engineering sites . OptRAM can simultaneously identify transcription factors and metabolic genes to be targeted for overexpression , knockdown , and knockout . Solutions from OptRAM ensure that the target product is better coupled with cell growth , and further systematical evaluation can help biologists to choose a relatively reliable solution for experiment validation . When performing OptRAM on MATLAB ( 2017a ) with GUROBI version of 7 . 5 , the average time of one SA process is 3 . 7 hours . Overall , the computation time is comparable to the time running optimization algorithm once on the OptFlux platform . The processor of PC is i7-6700 CPU with 3 . 40GHz frequency , and the RAM is 16 . 0GB . The performance of OptRAM has been compared with OptFlux by three strain design cases in yeast , including production of succinate , 2 , 3-butanediol , and ethanol . To make a more suitable comparison with a previous strain optimization method using an integrated regulatory-metabolic model , we compared OptRAM with OptORF , the first strain design method utilizing regulatory information , for ethanol overproduction . We used the integrated E . coli model iMC1010 by OptORF to simulate the modification strategy for ethanol production with OptRAM . We found the best solution of OptRAM includes knockdown of Pta , TpiA , AldB , ZntA , YbiV , Fre , OmpL , AceA , and GpmB , as well as overexpression of Acs . The ethanol production improved 1 . 8 fold with 7% reduction in biomass compared to wild-type . In comparison , OptORF suggested deletion of ArcA , Pta , TpiA , EutD , and PtsH , and overexpressing gene Edd , which was predicted to improve ethanol production by 2 . 2 fold with 54% reduction in biomass . However , ArcA has positive regulation on AckA , which is one of the essential genes for E . coli ( from DEG ) . Hence , this solution from OptORF may cause death of E . coli . While there are benefits to both approaches , in this case it may be that OptRAM identified a more biologically feasible modification strategy with similar improvement on target production compared to OptORF .
With the development of industrial biotechnology , there is an increasing need to design high producing strains in an economic and efficient manner . Computational strain optimization algorithms have been developed for this purpose as an important application of metabolic network reconstructions and constraint-based modeling . These methods can automatically search for sites of genetic modification for increasing any desired product . However , most strain optimization algorithms can only utilize a metabolic network alone and cannot provide strategies also involving transcriptional regulation . Although some methods can utilize gene regulatory information now , they have some limitations since they are based on integrating a boolean regulatory network , which is not suitable for TF overexpression ( e . g . OptORF ) . Reconstructing such models requires extensive manual adjustment and additional information for generating boolean logic rules in the regulatory network [37] , which hinders the ability of these algorithms to be broadly applicable across many organisms . In this study , we developed OptRAM to identify the manipulations of both TF and metabolic genes including overexpression , knockdown and knockout . OptRAM uses the framework of simulated annealing and is based on the integration of an inferred regulatory network with a metabolic network from our previous work ( IDREAM ) [30] . Through the in silico strain design case studies for producing succinate , 2 , 3-butanediol , and ethanol in yeast , we demonstrated that OptRAM can identify solutions containing both TF and metabolic gene manipulations that are predicted to increase production beyond what is seen currently , or found as potential designs using alternative methods . OptRAM used simulated annealing with a novel objective function , which can ensure a favorable coupling between desired chemical production and cell growth . We applied OptRAM in succinate , 2 , 3-butanediol , and ethanol overproduction in yeast . By setting a loose constraint ( at least 50% ) and a strict one ( 99% ) to biomass respectively , we compared the flux ranges of target reaction in different mutant models by different methods . In both cases , strategies from OptRAM led to higher minimum fluxes for target production under the strict constraints ( Tables 5 , 7 and 9 ) than alternative approaches , which indicated that the target chemical is strongly coupled with growth from the OptRAM designs . However , the minimum predicted output fluxes of target in other models are lower and may even be close or equal to zero , which means the coupling relationships are likely weak . We also found most of the genes predicted to be altered could be matched with the reported gene modifications , and some altered TFs having their target genes validated to improve the desired chemical in LASER database . In particular , we conducted fermentation experiment to validate the predicted deletion of MDH2 , BDH1 , and COX4 has significant improvement on ethanol production . Therefore , OptRAM provides in silico predictions of improved strains over other methods tested . Meta-heuristic algorithms commonly provide several optimized solutions with close objective scores , and it is difficult to select a best one for practical operation , such as OptFlux . Thus , we try herein to give a systematic evaluation for these solutions . First of all , essential genes cannot be knocked out without making growth of the cell impossible , so strategies are filtered according to essential genes with experimental validation in DEG and SGD databases . Then , we estimate the implementation cost by setting a score according to connection distance of shortest paths from critical reactions to the main path . Another factor that is weighed is minimizing the adjustment needed to make from the wild type in order to achieve the optimal designed performance . The latter two quantitative indicators are used to assist our selection , along with the flux value of producing the target compound . For succinate overproduction in yeast , there were three solutions excluded since some essential genes for growth were predicted to be knocked out or knocked down . After eliminating these three solutions , we then selected the best remaining solution with maximum target production , which also has lower score of implementation cost and global flux adjustment ( S3 Table ) . For the case of 2 , 3-butanediol overproduction in yeast , also three solutions with essential gene knockouts were excluded . Of the other solutions that have the same maximum target production ( S3 Table ) , we selected the solution with the lowest summation of path score and flux variation , which provided more suitable design modifications for real experiment design . For the case of improving ethanol production in E . coli , OptORF predicted deletion of ArcA as the modification site , but ArcA positively regulated AckA , which is one of the essential genes for E . coli . While OptRAM will avoid knockout or knockdown of such essential genes to keep better growth and/or viability of the organism . Despite the above highlights of our new algorithm , OptRAM can be further strengthened in various ways . First , the performance of integrative regulatory-metabolic modeling for phenotype simulation can be improved by introducing more information such as kinetic parameters to set more precise ranges for critical reactions in a kinetic model . Also , by introducing more potential sites , the solution space for exploration increases sharply . It is also favorable to seek new ideas for optimization framework other than focusing on the branches from previously proposed computational strain design methods [60] . On the other hand , the in-silico strain design can be enhanced by the inclusion of other forms of regulation , such as allosteric regulation . A constraint-based method ( arFBA ) for modeling the contribution of allosteric regulation for flux control in the central carbon metabolism of E . coli has been reported [61] . Most importantly , the method needs to be validated experimentally in many different situations and cases and these data compiled , so that this method and future methods can be iteratively enhanced . In conclusion , in current situation , OptRAM provides a good solution to assist the biologists to identify strain design strategies for particular applications . | Computational strain design algorithms based on genome-scale metabolic models have increasingly been used to guide rational strain design for metabolic engineering . However , most strain optimization algorithms only utilize a metabolic network alone and cannot provide strategies that also involve transcriptional regulation . In this paper , we developed a novel strain design algorithm , named OptRAM ( Optimization of Regulatory And Metabolic Network ) , which can identify combinatorial optimization strategies including overexpression , knockdown or knockout of both transcription factors and metabolic genes , based on our previous IDREAM integrated network framework . OptRAM uses simulated annealing with a novel objective function , which can ensure a favorable coupling between the production of a desired chemical and cell growth . This strategy can be deployed for strain design of bacteria , archaea or eukaryotes . The other advantage of OptRAM compared with previous heuristic approaches is that we systematically evaluated the implementation cost of different solutions and selected strain designs which are more likely to be achievable in experiments . Through the in-silico strain design case studies for producing succinate , 2 , 3-butanediol , and ethanol in yeast , we demonstrated that OptRAM can identify strategies that increase production beyond what is seen currently , or found as potential designs using alternative methods . We also validated the modified genes chosen by OptRAM in example cases against previous in vivo experiments in the LASER database . Additionally , we experimentally validated the ethanol strain design by evaluating its performance in fermentation . OptRAM provides a robust approach to strain design across gene regulatory network modification and metabolic engineering . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"chemical",
"compounds",
"gene",
"regulation",
"applied",
"mathematics",
"metabolic",
"networks",
"organic",
"compounds",
"simulation",
"and",
"modeling",
"algorithms",
"fungi",
"model",
"organisms",
"mathematics",
"simulated",
"annealing",
"network",
"analysis",
"experimental",
"organism",
"systems",
"alcohols",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"computer",
"and",
"information",
"sciences",
"transcriptional",
"control",
"animal",
"studies",
"gene",
"expression",
"chemistry",
"ethanol",
"yeast",
"eukaryota",
"organic",
"chemistry",
"gene",
"regulatory",
"networks",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"physical",
"sciences",
"computational",
"biology",
"saccharomyces",
"cerevisiae",
"organisms"
] | 2019 | OptRAM: In-silico strain design via integrative regulatory-metabolic network modeling |
Antigenic variation to evade host immunity has long been assumed to be a driving force of diversifying selection in pathogens . Colonization by Streptococcus pneumoniae , which is central to the organism's transmission and therefore evolution , is limited by two arms of the immune system: antibody- and T cell- mediated immunity . In particular , the effector activity of CD4+ TH17 cell mediated immunity has been shown to act in trans , clearing co-colonizing pneumococci that do not bear the relevant antigen . It is thus unclear whether TH17 cell immunity allows benefit of antigenic variation and contributes to diversifying selection . Here we show that antigen-specific CD4+ TH17 cell immunity almost equally reduces colonization by both an antigen-positive strain and a co-colonized , antigen-negative strain in a mouse model of pneumococcal carriage , thus potentially minimizing the advantage of escape from this type of immunity . Using a proteomic screening approach , we identified a list of candidate human CD4+ TH17 cell antigens . Using this list and a previously published list of pneumococcal Antibody antigens , we bioinformatically assessed the signals of diversifying selection among the identified antigens compared to non-antigens . We found that Antibody antigen genes were significantly more likely to be under diversifying selection than the TH17 cell antigen genes , which were indistinguishable from non-antigens . Within the Antibody antigens , epitopes recognized by human antibodies showed stronger evidence of diversifying selection . Taken together , the data suggest that TH17 cell-mediated immunity , one form of T cell immunity that is important to limit carriage of antigen-positive pneumococcus , favors little diversifying selection in the targeted antigen . The results could provide new insight into pneumococcal vaccine design .
Diversifying selection on genes encoding pathogen antigens is a well known effect of host immunity [1] , [2] . Diversifying selection can maintain multiple alleles of a gene at appreciable frequencies in a population [3] . Acquired immune responses provide a fitness advantage for antigenic variants that evade immune recognition , reducing the probability that the allele encoding the targeted antigen will fix with a single allele . In viruses such as HIV [4] , [5] , [6] and influenza [7] , [8] , neutralizing antibody and cytotoxic T-lymphocytes ( CTLs ) drive antigenic diversification . Strong diversifying selection was also identified in major antigen genes in the malaria parasite Plasmodium falciparum [9] , [10] . In bacteria , diversity of surface structures ( such as capsular polysaccharides ) that are targeted by host antibodies is thought to result from such diversifying selection [1] . However , a few exceptions exist . Measles virus antigens show little variation , partially because exposure to the virus would generate polyclonal antibodies that efficiently neutralize a broad range of antigenic variants [11] . Human T cell epitopes of Mycobacterium tuberculosis show a substantially lower level of sequence variation than seen in other genomic regions , suggesting T cell immune responses might limit diversification in the antigen genes [12] . Therefore , we hypothesized that the effect of host immunity on diversifying selection depends on the specific mechanism involved . Recent studies have indicated that acquired immunity elicited by natural exposure to Streptococcus pneumoniae includes three distinct arms [13]: ( 1 ) type-specific , antibody-mediated immunity to the highly variable polysaccharide capsule [14] , [15] , [16] , [17] , ( 2 ) antibody-mediated immunity to pneumococcal proteins , some of which are variable and some of which are more conserved [15] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , and ( 3 ) CD4+ TH17 cell- mediated , antibody independent immunity to pneumococcal proteins and to the cell-wall polysaccharide [15] , [25] , [26] , [27] , [28] . The first two forms of immunity are thought to operate by the standard mechanisms of antibody binding to surface antigens , leading to opsonophagocytosis , reduced attachment and/or other mechanisms of reduced colonization [22] , [29] . In the last form of immunity , antigen-specific CD4+ TH17 cells secrete interleukin ( IL ) -17A , leading to the activation and recruitment of effector cells ( neutrophils and macrophages ) that then kill pneumococci [25] , [30] , [31] , [32] . TH17 cell-mediated immunity primarily accelerates the clearance of pneumococcus rather than preventing initiation of carriage [31] . Even in combination , these forms of immunity to S . pneumoniae are imperfect . Humans can be repeatedly colonized despite the immune responses from multiple arms . While antibody binding is by definition specific to bacteria bearing the target antigen , we have previously shown that the CD4+ TH17-based effector activity may extend beyond antigen-expressing bacteria , accelerating the clearance of co-colonized pneumococci that even do not bear the relevant antigen [23] . It is unclear whether CD4+ TH17-mediated immunity would still create a fitness advantage for antigenic variants and thus promote diversifying selection on the genes encoding the targets of such immunity in S . pneumoniae . Here we report the assessment of two hypotheses: first , a competition assay was performed to examine whether an antigen-negative strain shows a colonization advantage over the antigen-positive strain in mice with antigen-specific TH17 immunity . Second , pneumococcal genes that show signs of being under diversifying selection were systematically identified and their association with either Antibody antigens or TH17 antigens was examined . The results indicate little evidence of diversifying selection in the targets of CD4+ TH17 cell immunity , unlike the targets of antibody immunity .
Immunization with a pneumococcal whole cell vaccine displaying a peptide from ovalbumin ( OVA323–339 ) delivered with cholera toxin ( CT ) adjuvant results in CD4+ TH17 cell-mediated and antibody-independent protection against subsequent pneumococcal colonization [23] . To examine whether the TH17 cell immunity against S . pneumoniae , given its in trans clearance effect [23] , allows a competitive advantage for a non-recognizable ( antigen-negative ) strain , twenty BALB/c mice were immunized by either ovalbumin with adjuvant ( OVA+CT ) or adjuvant alone ( CT ) . The mice were challenged with a 1∶1 mix of an antigen-negative strain ( AVO ) and an antigen-positive strain ( OVA ) . The two strains were isogenic except that only the OVA strain displays OVA323–339 peptides that can be recognized by the ovalbumin-induced , TH17 immunity in mice [23] . The AVO strain can be viewed as an antigenic variant of the OVA strain and the AVO/OVA ratio would increase if there were a competitive advantage for the antigen-negative strain . The mixture of pneumococci colonized the ovalbumin-immunized and control mice equally well on day 1 . No significant difference in colonization density was observed ( Figure 1A , p = 0 . 87 , Mann-Whitney test ) . By day 4 , the median colonization density in ovalbumin-immunized mice was about 7-fold lower than that in the control mice , although the difference was not statistically significant ( Figure 1A , p = 0 . 48 , Mann-Whitney test ) . By day 8 , the median colonization density in the immunized mice was about 40-fold lower than that in the control mice and the difference was statistically significant ( Figure 1A , p = 0 . 02 , Mann-Whitney test ) . The effect was consistent with an accelerated clearance of colonization mediated by TH17 immunity [31] . The AVO/OVA ratio remained approximately 1∶1 in the control mice during the course of the experiment ( Figure 1B ) . The medians of log10 ( AVO/OVA ) were 0 . 185 ( n = 10 ) , −0 . 028 ( n = 11 ) , and 0 . 011 ( n = 16 ) on days 1 , 4 and 8 , respectively ( Table 1 ) , indicating that the AVO strain was competitively neutral in the absence of antigen-specific immunity . In the ovalbumin-immunized mice , the medians of log10 ( AVO/OVA ) were 0 . 334 ( n = 8 ) , 0 . 042 ( n = 10 ) and 0 . 730 ( n = 13 ) on days 1 , 4 and 8 , respectively ( Table 1 ) . The median log10 ( AVO/OVA ) was not significantly different between the control and the immunized group on days 1 , 4 or 8 ( Figure 1B , p = 0 . 067 , p = 0 . 50 , and p = 0 . 12 , respectively , Mann-Whitney test ) , although there was a trend toward an increase in AVO/OVA ratio in the immunized mice . To better quantify the potential competitive advantage for the antigen-negative strain , we constructed nonparametric confidence intervals for the median of the difference in log10 ( AVO/OVA ) between the immunized group and the control group ( Table 1 ) . A median greater than 0 would indicate a competitive advantage for the AVO strain in the immunized group . The 95% confidence intervals for median difference in log10 ( AVO/OVA ) were ( −0 . 006 , 0 . 563 ) , ( −1 . 437 , 0 . 456 ) , and ( −0 . 2319 , 1 . 015 ) on days 1 , 4 , and 8 , respectively ( Table 1 ) . Thus , the loss of an antigen was unlikely to provide a more than 10 . 4-fold ( 1 . 015 log10 ) median increase in competitive advantage for the AVO strain by day 8 . We also note that the increased frequency of AVO strains was almost entirely found in mice who have nearly cleared colonization ( Figure 1C ) . In absolute CFU numbers , therefore , the relative advantage is unlikely to be associated with much overall superiority . In mice that remain colonized on days 4 and 8 , a negative correlation between the AVO/OVA ratio and total CFU recovered was observed in the immunized group ( Figure 1C ) but not in the control group ( Figure 1D ) . These results suggested that the antigen-negative strain gains a relative advantage only for the period where bacterial numbers are rather low . To determine whether CD4+ TH17 cell-mediated immunity to S . pneumoniae affects antigenic variation in the context of human colonization and disease , S . pneumoniae antigens recognized by human TH17 cells were identified . CD4+ TH 17 cells were enriched from peripheral blood cells and IL-17A secretion in response to pneumococcal protein pools was measured by ELISA ( see Materials and Methods , Figure S1 , and Figure S2 ) . To identify the common antigens in the sample population of 36 healthy adults , a Mann-Whitney test was used to compare normalized values for each pool to the normalized values for E . coli expressing GFP . Each protein was then ranked by its antigenicity score , which was calculated by multiplying together the p-values resulting from the Mann-Whitney test for both pools containing the protein , lower antigenicity scores indicating more commonly recognized antigens . An N-terminal fragment of PtrA ( SP0641 . 1 ) was the most strongly recognized antigen in the screen with an antigenicity score of 1 . 58×10−17 ( Figure 2B ) . Clones with a score less than 0 . 05 were defined as the common antigens ( Table 2 ) . To evaluate genetic diversity and the underlying selection pressure on pneumococcal proteins , we systematically examined protein-encoding regions from the genome sequence data of 39 publicly-available pneumococcus strains for evidence of diversifying selection . Based on information accompanying the genome sequence data , the collection of strains covered 14 common serotypes ( Table S1 in Text S1 ) . Although the strains used in our study are not a random sample of any population and may overrepresent clinical ( invasive ) isolates , the distribution of serotype frequency in this study was reasonably consistent with distribution reported in human carriage [33] ( Figure S3 ) . A flowchart of the analysis is shown in Figure 3A . Open reading frames ( ORFs ) that were inferred to represent the same gene in different strains were grouped together to form an orthologous group . A total of 2773 unique unambiguous groups were generated by the Proteinortho4 software [34] . Sequence alignment of genes within an orthologous group was performed using the PRANK software [35] . Extensive sequence variation was observed for many pneumococcal protein-encoding genes . The nucleotide diversity for a gene ranged from 0 to 0 . 23 with a median of 0 . 0091 ( Figure 3B ) . To identify pneumococcal genes that show signs of being under diversifying selection , we analyzed the non-synonymous to synonymous substitution ( dN/dS ) ratio for codon sites in each gene using the PAML package as described by Yang [36] . Signs of being under diversifying selection were detected by a likelihood ratio test in which a null model ( dN/dS < = 1 for all codons ) was compared with an alternative model ( dN/dS>1for at least one codon ) , as described in the Materials and Methods . We concluded signs of diversifying selection for a gene if the null model was rejected at the significance level of 0 . 05 . By this criterion , 658 genes ( 23 . 7% ) showed signs of being under diversifying selection . The subsequent Bayes Empirical Bayes ( BEB ) analysis [37] identified 1410 codon sites , or 0 . 178% of total codon sites , to be under diversifying selection ( Figure 3C ) . Codon sites under diversifying selection were enriched in cell envelope genes ( Table S2 in Text S1 ) , consistent with that interaction with antibodies might be a source of selection pressure on the pneumococcal protein antigens . We hypothesized that if human immunity had promoted diversifying selection in pneumococcal antigens , the antigen genes would exhibit higher sequence diversity than non-antigen genes . Genes encoding CD4+ TH 17 antigens were identified as described above . Genes encoding Antibody antigens were obtained from the list published by Giefing et al [24] . TIGR4 genes belonging to an orthologous group of two or more genes were analyzed , including 1648 non-antigens , 48 TH17 antigens and 80 Antibody antigens . In addition , the regions of Antibody antigens genes that included epitopes were also noted by Giefing et al . , facilitating our comparisons of non-antigens , Antibody antigen-encoding genes , and the epitope-containing and non-epitope-containing regions of these antigen-encoding genes . The average non-synonymous substitution rate ( dN ) of Antibody antigens was significantly higher than that of non-antigens ( Figure 4A; median 0 . 0032 vs . 0 . 0025; p = 0 . 022 , Mann-Whitney test ) . However , there was no significant difference in dN between TH17 antigens and non-antigens . ( Figure 4A; median 0 . 0026 vs . 0 . 0025; p = 0 . 65 , Mann-Whitney test ) . Genes encoding Antibody antigens also showed a significantly higher proportion of genes with signs of being under diversifying selection ( Figure 4B , OR = 1 . 95 , p = 0 . 006 , Fisher's Exact test ) . In contrast , TH17 antigen genes showed no evidence of being under diversifying selection ( Figure 4B , OR = 0 . 77; p = 0 . 52; Fisher's Exact test ) . Not all codon sites within a gene need be under the same selective force . To understand the contribution of host immunity to diversifying selection , we were particularly interested in whether the codon sites that did show an estimated dN/dS ratio greater than 1 were equally distributed among antigen categories . We found that 0 . 183% of the codon sites located in the non-antigen genes showed dN/dS ratio greater than 1 ( Figure 4C ) . For codon sites in the CD4+ TH 17 antigen genes , a higher fraction ( 0 . 33% , Figure 4C ) showed a dN/dS ratio greater than 1 . An even higher fraction ( 0 . 46% , Figure 4C ) of the Antibody antigen codon sites showed a dN/dS ratio greater than 1 . Furthermore , within the Antibody antigens , the regions in antibody epitopes showed a higher density of codon sites with dN/dS greater than 1 than the non-epitope regions ( 0 . 62% vs . 0 . 42% , Figure 4C ) . Thus , the genomic regions that interact with antibody-mediated immunity appeared to be more enriched for codon sites with signs of being under diversifying selection , with a weaker signal of diversifying selection in the CD4+ TH17 antigens . To account for correlations between different codon sites within a gene and for differences in gene length that would make longer genes more likely , by chance alone , to have sites with elevated dN/dS ratios , we employed a generalized-estimating-equation ( GEE ) model to examine the “population-averaged” effect of being recognized by human immunity on the probability that a gene is under diversifying selection [38] . Essentially , we treated the status of each individual codon in a gene ( whether or not the codon showed sign of being under diversifying selection ) as the outcome of a repeated measurement for the status of the gene ( whether or not the gene showed sign of being under diversifying selection ) . During model fitting , the covariance structure across codon sites within a gene was treated as a nuisance parameter . The output of the model fitting showed that being an Antibody antigen is a highly significant predictor for being under diversifying selection ( Figure 4D; OR = 2 . 23 , p = 0 . 0016 ) and being a TH17 antigen is a weaker , and not statistically significant predictor ( Fig . 4D , OR = 1 . 57 , p = 0 . 17 ) . Taken together , these results indicated that antibody immunity made a greater contribution than CD4+ TH 17 cell immunity to diversifying selection on antigen genes in S . pneumoniae . To examine the robustness of our results , we carried out the analysis of diversifying selection using a different alignment algorithm [39] , as well as another evolution model proposed by Wilson et al . , which allows estimation of the dN/dS ratio in the presence of recombination [40] . All analyses yielded qualitatively similar results ( Table S3 and Table S4 in Text S1 ) .
In this study , we investigated the contribution of host immunity to the diversifying selection in S . pneumoniae . We found that CD4+ TH17 cell-mediated immunity , elicited by exposure to pneumococci bearing a targeted antigen , cleared pneumococci that do not bear this antigen in trans almost as efficiently as it cleared the antigen-bearing cells . Thus , TH17 cell immunity limited the competitive benefit of antigenic variation within a colonized host , potentially reducing a driving force of diversifying selection . Consistent with this notion , we found a weak , and not statistically significant association between diversifying selection and recognition by human TH17 cell immunity . We hypothesize that this lack of selection is due to in trans killing of antigen-negative bacteria by innate cells recruited through TH17 cells recognition of antigen-expressing bacteria . However , the promiscuity of CD4+ T cell epitope recognition [41] could also play a role as it may be more difficult for bacteria to mutate the recognized antigens to avoid T cell recognition . In contrast to TH17 antigens , there was a significant association between recognition by human antibody and diversifying selection on the antigen . These data suggest that these two mechanisms of acquired immunity exert distinct selection forces on their respective antigens in S . pneumoniae . We observed that an antigen-negative ( AVO ) /antigen-positive ( OVA ) ratio higher than 1 was associated with lower CFU in the ovalbumin-immunized mice but not in the control mice . This supported the antigen-specificity of the immunity recalled by the OVA strain . In principle , there are three stages of the pneumococcal life cycle in which escape from immunity might be beneficial: ( 1 ) an advantage for an escape variant by mutation or deletion of an antigen that is the target of an immune response during infection; ( 2 ) an advantage for a variant in colonizing a host already responding to a “wild-type” strain that is resident and targeted by the host's response; ( 3 ) an advantage for a variant in colonizing a host that is currently uncolonized with any pneumococcal strain , but has immunity to wild-type alleles of the antigen from previous exposure . Cis-acting immune effectors , such as antibodies , would be expected to provide an advantage for a variant at all three of these stages . Our animal experiments suggest that for CD4+ TH17 cells , the advantage of an immune-escape variant would be small at stages 1 and 2 , because of in-trans killing; the first stage is particularly important because this is where a variant would likely first arise . Still , one would expect some advantage for CD4+ TH17 cell escape variants at the third stage – colonization of an uncolonized but partially immune host; this possibly may account for the weaker , less statistically convincing evidence of enrichment for diversifying selection in CD4+ TH17 antigens . Escape from CD4+ TH 17 cell immunity in our in vivo model should be more favored than in natural settings , for two reasons . First , we constructed a model in which the TH17 epitope was completely deleted ( and replaced with the reverse amino acid sequence ) , rather than creating a point mutation; given the promiscuity of T cell responses , many point mutations might make little or no difference to T cell recognition . Second , natural exposure to pneumococci would induce immunity to multiple T cell ( and antibody ) antigens , so that escape from a single response would not necessarily create a major advantage . The fact that we saw a modest benefit of losing the sole CD4+ T cell epitope against which the mice had been immunized argues that the benefit would be even weaker under natural conditions . The high throughput screen was designed to pick up the antigens with the strongest TH17 responses in the studied sample . This strength includes both frequency of response in the studied population and the strength of the response within individuals . The Mann-Whitney analysis does not allow us to define whether an antigen was positive in any given subject . However , if we use a different analysis method of taking antigens that induce a response greater than 1 . 2 MAD above the median , we find that the most common antigen was recognized by 47% of the subjects , with most antigens present in 10–20% of subjects ( data not shown ) , indicating a reasonably broad TH17 response . We acknowledge that there are weaker responses in these individuals that may have not been detected , but we posit that any selective pressure on TH17 antigens should be more robust in the strongly recognized antigens . Since no association between signs of diversifying selection and the human TH17 antigens we identified was found , the observation supports our hypothesis that CD4+ TH17 cell immunity in humans allows minimal competitive benefit for antigenic variation in S . pneumoniae . It is also important to note that only antigens recognized by IL-17A secreting T cells were identified . If the antigens recognized by different T cell lineages are distinct [42] , [43] , other T cells lineages may exert stronger selective pressure depending on their mechanism of action . We found that genomic regions that showed signs of being under diversifying selection were enriched in the antibody antigen genes and further enriched in the epitopes targeted by antibodies . This finding was consistent with the conventional understanding that avoidance of antibody-recognition can provide a substantial competitive benefit . The magnitude of the enrichment was consistently modest among all analyses . It is possible that multiple ways to avoid antibody-recognition exist , reducing the dependence on non-synonymous substitutions in the antigens . For example , antigens can be temporarily down regulated at the expression level to escape from host immunity , as was seen in the malaria parasite Plasmodium falciparum [44] and suggested for meningococci under vaccine pressure [45] . Antigens are also proteins carrying out physiological functions for the pathogen at the same time . They might be subjected to diversifying , purifying or other selective forces in addition to those imposed by acquired immunity . However , the significant association between antibody-recognition and diversifying selection despite these putative competing mechanisms suggested that antibodies impose a strong fitness cost on the antigen-bearing pneumococcus . In addition , it would be interesting to understand whether the diversifying selection differs in selected genes according to the invasive potential and transformability of the strain . Appropriate comparison would require much larger samples , which we hope to investigate in future studies . CD4+ T subsets other than the TH17 cells , such as the IFN-γ producing TH1 cells , have been proposed to play important roles in the control of pneumococcal invasive disease [42] , [43] but not , to our knowledge , colonization . In fact , in our colonization model , the IFN-γ mediated mechanism appeared to be dispensable [31] . Our screen would not have picked up antigens that elicited CD4+ T responses unless they also stimulated IL-17A production . Further work might address the contribution of other forms of T cell mediated immunity to diversifying selection . This study suggests that CD4+ TH17 cell immunity creates little selective pressure for antigenic variation while efficiently protecting against pneumococcal colonization , and suggest that the reason for this lack of selection may be due to efficient in trans killing of antigenic variants arising within a host . It is conceivable that a vaccine designed to induce TH17 cell immunity might limit the immune escape of antigenic variants and result in broader and longer protection . To this end , further research is ongoing to characterize the major TH17 cell antigens in pneumococcus and identify methods for eliciting this type of immunity through vaccination [27] , [46] .
All human subjects enrolled in this study provided written informed consent . The protocols for this study were IRB-approved by Quorum Review , Inc . All animal work has been conducted in compliance with the Animal Welfare Act and the guidelines of the U . S . Public Health Service Policy on Humane Care and Use of Laboratory Animals , and specifically approved by the Institutional Animal Care and Use Committee ( IACUC ) of Harvard Medical School . ( Animal Welfare Assurance of Compliance A3431-01 and AAALAC Accreditation #000009 , 6/19/09 ) The antigen-positive S . pneumoniae stain ( OVA ) was a serotype 6B strain 603 derivative that expressed the OVA323–339 peptide ( ISQAVHAAHAEINEAGR ) on the bacterial surface as fusion proteins with both pneumococcal surface protein A ( PspA ) and pneumolysin ( Ply ) [23] . To construct the antigen-negative S . pneumoniae ( AVO ) , the OVA coding sequence in the pspA and ply loci of the OVA strain was replaced by a nucleotide sequence encoding the OVA323–339 peptide in reversed sequence ( RGAENIEAHAAHVAQSI ) by using a Janus-cassette mediated transformation protocol [47] . Wild-type , female BALB/c mice were obtained from the Jackson ImmunoResearch Laboratories , Barr Harbor , ME . All mice were 5 to 6 weeks old at the start of experiments and kept in a BL2 facility . Ovalbumin ( Sigma-Aldrich , St . Louis , MO ) and cholera toxin ( CT ) mucosal adjuvant ( List Biological Laboratories , Compel , CA ) were purchased and stored according to the manufacturer's protocols . Mice were intranasally immunized twice , one week apart , with10 µL of PBS containing 10 µg Ovalbumin plus 1 µg CT ( OVA+CT ) or 1 µg CT alone ( CT ) . Four weeks after the second immunization , mice were inoculated intranasally with a mix of the OVA and the AVO strains in 10 µl of PBS containing approximately 5×106 CFU of each strain . On days 1 and 4 after challenge , samples from live animals were collected by applying 10 µl of ice cold PBS to either nostril of a mouse and collecting droplets discharged by the animal . On day 8 after challenge , upper respiratory tract samples were collected post mortem from retrotracheal washes of sacrificed mice . Aliquots of sample were titered to determine the colonization density . The remaining samples were cultured on gentamicin plates overnight and the resulting colonies were harvested for genomic DNA extraction . Genomic DNA was purified from cultures of samples collected from animals using DNeasy Blood and Tissue kit ( QIAGEN , Valencia , CA ) . The OVA strain- and the AVO strain-specific primer sets were designed based on the nucleotide sequence difference in the pspA locus between the two strains . The quantity of strain-specific genomic DNA in a sample was determined by absolute quantification protocol . A standard curve was built for each qPCR plate and was based on two replicates . All samples were measured based on averaged value of qPCR duplicate . The CFU ratio between the two strains was calculated by using the absolute amount of OVA DNA and AVO DNA in the same sample . The detection limit of AVO/OVA ratio was set as from ( 1×total CFU ) −1 to ( 1×total CFU ) . The qPCR-derived ratios outside this range were rounded to the nearest detection limit . Approval for blood collection was obtained from the Institutional Review Boards of each institution . IL-17A-secreting CD4+ T cells were first enriched from peripheral blood cells using negative magnetic selection of CD4+ T cells and a previously published IL-17A cytokine capture protocol [48] . S . pneumoniae-specific TH17 cells were further enriched by culturing the cells with autologous monocyte-derived dendritic cells ( MoDCs ) pulsed with inactivated S . pneumoniae . IL-17A secretion from the cells was measured after three days of co-culture with MoDCs pulsed with E . coli expressing a previously validated 2 , 547 clone ORFeome library of the S . pneumoniae TIGR4 genome [49] arrayed in pools of four clones . Enriched cells from 36 peripheral blood samples were screened with the pooled library ( see SI for methods detail ) . The results of the IL-17A ELISA were first normalized by plate by averaging the duplicates for each well , subtracting the plate median from each average and then dividing the result by the median absolute deviation of the plate , yielding the MAD score for each well in the screen . The most common antigens recognized by the population were identified by comparing the population response to each pool in the library to the measured responses to the all the wells that received E . coli expressing GFP using a one-tailed Mann-Whitney test . Each individual antigen was then scored by multiplying the p-values from the Mann-Whitney test of the two wells in which it was present . Genome sequence data of 39 pneumococcal strains were retrieved from the NCBI FTP site , ftp://ftp . ncbi . nih . gov/genomes . The collection included 14 annotated genomes and 25 draft genomes . Accession numbers of genome sequence were listed in Table S1 in Text S1 . For the annotated genomes , the annotation and nucleotide sequence of each gene were downloaded from the NCBI FTP site . For the draft genomes , putative protein-encoding genes were identified by using the Glimmer3 software [50] . Orthology analysis of pneumococcal proteins was carried out by using the Proteinortho4 software [34] , which assigned orthologous proteins from different strains into a same orthologous group based on the reciprocal best alignment heuristic . Cellular roles of TIGR4 genes were categorized according to the JCVI Annotation Gene Attributes ( http://cmr . jcvi . org ) . The gene sequences of each orthologous group were aligned based on the amino acid sequences they encode ( codon alignment ) and a gene tree was constructed using either the ClustalW software or the PRANK software [35] , [39] . A likelihood ratio test was applied to compare a null model with an alternative model of the distribution of the dN/dS ratio parameter , ω , among codon sites , as described in [37] . In the null model ( nearly-neutral model ) , each codon site within a gene is assumed to be either under purifying selection ( ω0<1 ) or under neutral evolution ( ω1 = 1 ) . In the alternative model ( positive selection model ) , a codon site can be under purifying selection ( ω0<1 ) , under neutral evolution ( ω1 = 1 ) or under diversifying selection ( ω2>1 ) . For each model , the log likelihood value was calculated by the CodeML program from the package PAML [36] . If the null model was rejected by the likelihood ratio test at a significance level of 0 . 05 , the gene represented by the orthologous group would be considered as being under diversifying selection . For such genes , a Bayes Empirical Bayes ( BEB ) analysis implemented in the CodeML program [36] was used to determine the particular codon sites that were under diversifying selection . The output file of the CodeML program included non-synonymous substitution rate ( dN ) derived from pair wise sequence comparison . The average dN for each orthologous group was estimated by averaging over all pair wise dNs . The dN/dS ratio for codon sites was also estimated by a method developed by Wilson et al . , which applied a population genetics approximation to the coalescent to accommodate recombination events [40] . The codon alignment of each orthologous group was analyzed by Omegamap software with a prior exponential distribution of ω and a prior ω mean of 1 . Each codon site was assumed to have independent ω and the posterior distributions of ω were obtains by 500 , 000 iterations . A codon site was defined to show evidence of being under diversifying selection if 95% of its posterior distribution of ω was above 1 . A gene was considered to show evidence of being under diversifying selection if any codon site within the gene showed sign of being under diversifying selection . The analyses took 3–4 weeks on a Linux cluster comprised of 4708 processor cores . Statistical analysis was performed by using the R package ( http://www . r-project . org ) . Graphs were created in Graphpad Prism and in Microsoft Excel . List of NCBI-Gene ID numbers for genes and proteins mentioned in the text: 929896 ( PspA ) , 931915 ( Pneumolysin ) , 930590 ( PtrA ) . | Streptococcus pneumoniae , or pneumococcus , is a leading cause of morbidity and mortality in young children and elderly persons worldwide . Current pneumococcus vaccines target a limited number of clinically important serotypes , while strains with serotypes not targeted by current vaccines are increasing in importance in both carriage and invasive disease . As a result , there has been a substantial interest to develop novel , cost-effective vaccines based on protein antigens from pneumococcus . To this end , it is critical to understand how the human immune system exerts selection pressures on the targeted antigens . Two immune mechanisms targeting pneumococcal protein antigens have been documented , mediated by antibody and T cells , respectively . In this study , we screened for pneumococcal antigens that are commonly recognized by human CD4+ TH17 cells . Using a mouse model of pneumococcal colonization , we demonstrate that TH17 cell-based immunity almost equally reduces colonization by both an antigen-positive strain and a co-colonizing , antigen-negative strain . Furthermore , we demonstrate that the DNA sequences of TH17 cell antigens demonstrate no detectable signs of being under selective pressure , unlike pneumococcal antigens known to be strong antibody targets . Thus , one form of the T cell-mediated immunity that is important to limit carriage of antigen-positive pneumococcus favors little diversifying selection in the targeted antigen . These results suggest evolution of escape from TH17 -based vaccines may be slower than from antibody-based vaccines . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"of",
"the",
"immune",
"system",
"genomic",
"evolution",
"immunity",
"evolutionary",
"immunology",
"biology",
"evolutionary",
"biology",
"comparative",
"genomics",
"immunology",
"evolutionary",
"processes",
"immune",
"response"
] | 2012 | Distinct Effects on Diversifying Selection by Two Mechanisms of Immunity against Streptococcus pneumoniae |
Innate lymphoid cells ( ILCs ) are severely depleted during chronic HIV-1 infection by unclear mechanisms . We report here that human ILC1s comprising of CD4+ and CD4- subpopulations were present in various human lymphoid organs but with different transcription programs and functions . Importantly , CD4+ ILC1s expressed HIV-1 co-receptors and were productively infected by HIV-1 in vitro and in vivo . Furthermore , chronic HIV-1 infection activated and depleted both CD4+ and CD4- ILC1s , and impaired their cytokine production activity . Highly active antiretroviral ( HAART ) therapy in HIV-1 patients efficiently rescued the ILC1 numbers and reduced their activation , but failed to restore their functionality . We also found that blocking type-I interferon ( IFN-I ) signaling during HIV-1 infection in vivo in humanized mice prevented HIV-1 induced depletion or apoptosis of ILC1 cells . Therefore , we have identified the CD4+ ILC1 cells as a new target population for HIV-1 infection , and revealed that IFN-I contributes to the depletion of ILC1s during HIV-1 infection .
Innate lymphoid cells ( ILCs ) represent a novel family of cellular subsets that produce large amounts of T cell-associated cytokines in response to innate stimulation in the absence of antigens [1 , 2] . Based on the expression of specific transcription factors , cell surface markers and signature cytokines [1 , 3 , 4] , ILCs can be divided into three groups . Group 1 ILCs ( ILC1s ) have been defined as lineage-CD127+CD117- cells and can produce interferon ( IFN ) -γ and depend on T-bet for their functions [5] . Group 2 ILCs ( ILC2s ) are a population of lineage-CD127+CRTH2+ cells that preferentially produce IL-5 and IL-13 and require GATA3 for differentiation [6] . Group 3 ILCs ( ILC3s ) are lineage-CD127+CD117+cells that have the potential to produce IL-17 and/or IL-22 , and are dependent on RORγt [3 , 7] . An increasing number of studies have indicated that ILCs represent a heterogeneous family of cells [8–10] . ILC1s were recently divided into CD4+CD8- , CD4-CD8+ and CD4-CD8- cell populations , and ILC3s comprise CD62L+ naïve cells and HLA-DR+ ILC3 subsets [8] . These novel ILC subsets still need to be explored with regard to their functionality and clinical significance in humans . ILCs have emerged as central players in homeostatic and inflammatory conditions . In particular , changes in the number of ILCs have been found to be associated with the pathogenesis and progression of a number of human diseases including chronic infections and inflammatory diseases [1 , 3 , 11–13] . For example , IFN-γ production by intraepithelial ILC1s promotes inflammation in mouse models of colitis , and blocking of IFN-γ reduces disease severity [12] . In addition , ILC1s may also contribute to human inflammatory bowel diseases , as their numbers have been found to be higher than normal in patients with Crohn’s disease [5 , 12] . Changes in the number and function of ILCs have also been documented during HIV-1 or SIV infection . Further , it has been reported that SIV infection results in persistent loss of IL-17-producing ILCs , especially in the jejunum [14] . NKp44+ ILC3s are also rapidly depleted in the intestinal mucosa during acute SIV infection [15] . In HIV-1-infected patients , too , ILCs are found to be severely depleted [16–18] . We have previously demonstrated that in HIV-1/SIV infection , ILC3s are depleted through plasmacytoid dendritic cell ( pDC ) activation and CD95-mediated apoptosis [17] or TLR signaling [19] . However , it is not clear whether HIV-1 influences ILCs through infection and how ILCs are depleted , especially ILC1s , during HIV-1 infection . In this study , we first showed that tissue ILC1s , as reported previously in the case of human peripheral blood mononuclear cells ( PBMCs ) [20] , consist of CD4+ , CD8+ and CD4-CD8- cells , three populations that widely exist in various lymph organs in human . In addition , we found that CD4+ ILC1s exhibit significant differences from CD4- ILC1s with regard to their phenotype , cytokine production and expression profile of transcriptional factors . Thus , we have identified a previously unknown CD4+ ILC1 population that serves as a target for HIV-1 productive infection . We showed that HIV-1 can infect , activate and preferentially deplete these CD4+ ILC1s . Our data was also indicative of the pathogenic effect of sustained type I interferon ( IFN-I ) signaling during HIV-1 infection , including depletion of ILC1s .
It was recently reported that ILC1s in human peripheral blood contain CD4+ , CD8+ and CD4-CD8- subpopulations [20]; however , it is unclear whether these cell populations are present in human lymphoid organs . Here , we investigated the distribution of each ILC1 subpopulation in various human lymph organs . By gating on live human CD45+ cells that were negative for lineage-specific surface markers of B cells ( CD19 and CD20 ) , T cells ( CD3 ) , conventional natural killer ( NK ) cells ( CD16 ) , monocytes and dendritic cells ( CD14 , CD11c and CD123 ) , and surface markers of hematopoietic precursors ( CD34 ) , ILC2 cells ( CRTH2 ) as well as ILC3 cells ( CD117 ) , we identified ILC1s as hCD45+Lin-CD117-CRTH2-CD127+CD56- cells ( S1A Fig ) . Similar to the results of a previous study [20] , we found that ILC1s comprise of CD4+CD8- , CD4-CD8+ and CD4-CD8- subpopulations ( S1A Fig ) . All the ILC1 subsets don’t express the T cell marker TCRαβ , TCRγδ and NK cell marker CD94 which excludes T cell and NK cell contamination; while they express CD5 ( S1B Fig ) . More importantly , we found that the all the three ILC1 subsets , including CD4+ ILC1s , were all present in various human lymphoid organs including the spleen , bone marrow , large intestine , small intestine and liver perfusion ( Fig 1A ) . Further analysis indicated that CD45+ cells constituted 0 . 019%–0 . 818% of the total ILC1 cells ( Fig 1B ) and CD4+ ILC1s constituted 2 . 35%–39 . 2% of the total ILC1s in different organs ( Fig 1C ) . We further investigated the expression of transcriptional factors such as T-bet and eomesodermin ( Eomes ) in the three ILC1 subsets in human peripheral blood ( Fig 1D ) . We found that CD4+ and CD4-CD8- ILC1s expressed lower levels of T-bet than CD8+ ILC1s ( Fig 1E , left ) . In addition , CD4+ ILC1s also expressed lower levels of Eomes than CD4- ILC1 subsets in the blood ( Fig 1E , right ) . We also examined the expression of T-bet and Eomes in ILC1 subsets from various human lymphoid organs by flow cytometry ( S2A Fig ) . We found that in most tissues that we examined , CD4+ ILC1s expressed lower levels of T-bet than CD8+ or CD4-CD8- ILC1s . Notably , the expression levels of T-bet were significantly lower in all ILC1 subsets from the small intestine than in the corresponding subsets from the other organs . This indicates that ILC1s present in the small intestine may have a unique function or activity ( S2B Fig ) . CD4+ ILC1s also expressed lower levels of Eomes than CD4- ILC1 subsets in the blood , spleen , bone marrow and liver perfusion . However , the opposite phenomenon was observed in the large and small intestine , where CD4+ ILC1s expressed higher levels of Eomes than CD4- ILC1s ( S2C Fig ) . These data suggest that the transcriptional factor profiles of ILC1s differ according to subsets and tissue types . In particular , CD4+ ILC1s are characterized by lower expression of T-bet and Eomes transcriptional factors in human peripheral blood . With regards to phenotypic characteristics , CD4+ ILC1s in peripheral blood expressed CD45RA , the NK cell-related molecule CD161 , the chemokine receptors CCR6 and CXCR3 , death receptor CD95 and adhesive molecule CD11a , which indicate an immature phenotype . However , peripheral blood CD4+ ILC1s did not express the integrin CD103; activation markers CD69 , CD38 and HLA-DR; proliferation marker Ki67; and death molecules DR5 , caspase 1 and caspase 3 . Moreover , they expressed low levels of the ILC progenitor marker IL-1R1 ( S3 Fig ) . However , there was no significant difference in the expression of most of the molecules between CD4+ and CD4- ILC1 subsets from peripheral blood . As for functionality , we evaluated cytokine production by peripheral blood ILC1 subsets after PMA/ionomycin or IL-12/IL-18 stimulation ( Fig 1F and 1G ) . We found that CD4+ ILC1s produce more TNF-α and lower levels of IFN-γ than CD8+ and CD4-CD8- ILC1s under PMA/ionomycin stimulation ( Fig 1F ) . Under IL-12/IL-18 stimulation , CD4+ ILC1s also produced lower levels of IFN-γ but similar levels of TNF-α than CD8+ and CD4-CD8- ILC1s ( Fig 1G ) . These ILC1 subsets produced no detectable IFN-γ and TNF-α without stimulation . These data indicate that peripheral blood ILC1 subsets are characterized by functional heterogeneity , and that CD4+ ILC1s preferentially produce TNF-α in response to stimulation , as opposed to CD4- ILC1 subsets , which produce lower levels of TNF-α . Taken together , these comprehensive analyses indicate that CD4+ and CD4- ILC1s exist in various human lymphoid tissues , and their relative numbers , transcription and functionality depend on the subsets and the tissues . In particular , CD4+ ILC1s display relatively unique expression of transcriptional factors , immune phenotypes , and cytokine production in relation to CD4- ILC1s . Since a significant proportion of ILC1s express CD4 , the receptor for HIV-1 infection , we investigated whether HIV-1 can infect CD4+ ILC1s . First , we examined the expression of the HIV-1 co-receptors CCR5 and CXCR4 on ILC1s by flow cytometry . Both CCR5 and CXCR4 were expressed on CD4+ ILC1s from human PBMCs and the spleen of humanized mice ( Fig 2A ) . CD4- ILC1s also expressed comparable levels of CCR5 and CXCR4 ( Fig 2A ) . Further analyses indicated that 12% of human CD4+ ILC1s express CCR5 , while 60% express CXCR4 ( Fig 2B ) . The expression of CCR5 and CXCR4 was also detected on CD4+ ILC1s in lymphoid organs , including the spleen , peripheral lymph node and bone marrow , and peripheral blood from humanized mice , but the expression level was slightly lower than that in human PBMCs ( Fig 2B ) . We then examined whether HIV-1 can infect human CD4+ ILC1s . We infected resting human PBMCs with the CXCR4 tropic virus NL4-3 and the CXCR4 and CCR5 dual-tropic virus R3A in vitro . Productive infection by HIV-1 was detected by staining of the HIV-1 protein p24 in ILC1s ( Fig 2C ) and in CD3+ T cells ( control cells ) ( S4A Fig ) . We found that HIV-1 p24 protein was detected in 2 . 2% of ILC1s after R3A infection and in 3% of ILC1s after NL4-3 infection ( Fig 2D ) , which was comparable to the p24 levels in CD3+ T cells ( S4A Fig ) . A neutralizing monoclonal antibody ( Clone CH31 ) specific to the CD4 binding site [21] blocked both R3A and NL4-3 infection by 90% ( Fig 2C and 2D and S4 Fig ) . We also found that HIV-1 infection down-regulated CD4 expression in ILC1s ( Fig 2C ) , as observed in T cells ( S4 Fig ) . Interestingly , when PBMCs were activated by PHA ( Fig 2E and 2F ) , both ILC1s and T cells were infected at higher levels by HIV-1 in vitro ( Fig 2E and 2F and S5A and S5B Fig ) . These results indicate that HIV-1 can productively infect ILC1s via the CD4 receptor . We also examined whether HIV-1 also infected ILC1s in vivo in human patients and in humanized mice . We purified CD4+ ILC1s from HIV-1-infected patients and determined the cell-associated HIV-1 DNA level by real-time PCR . On average , we detected 800 copies of cell-associated HIV-1 DNA in one million CD4+ ILC1s ( Fig 2G ) . As controls , 3200 copies of HIV-1 DNA were detected in one million CD4+ T cells , while no HIV-1 DNA was detected in CD8+ T cells ( Fig 2G ) . HIV-1 can effectively infect and replicate in vivo in humanized NOD-Rag2-/-γc-/- ( NRG-hu ) mice transplanted with human CD34+ hematopoietic stem cells [22] , a highly relevant model for studying HIV-1 induced pathology in vivo [23] . We therefore investigated whether CD4+ ILC1s could be directly infected by HIV-1 in vivo in humanized mice . At 3 weeks after R3A infection , 4 . 9% of ILC1s expressed p24 , while 2 . 7% of CD3 T cells were positive for p24 ( Fig 2H , left ) . To exclude the possibility that the p24 protein detected here was from virions taken into cells by endocytosis , we infected humanized mice with an engineered R3A reporter virus which expresses the mouse CD24 gene in the Vpr ORF [24] . We found that 8% of ILC1s expressed the mouse CD24 protein ( Fig 2H , right ) . Taken together , our results show that HIV-1 can productively infect CD4+ ILC1s both in vitro and in vivo . We next investigated whether HIV-1 infection also activates ILC1s in patients . We analyzed the expression of CD38 and Ki-67 in ILC1s ( Fig 3A and S6 Fig ) . Both CD4+ and CD4- ILC1s expressed higher levels of CD38 and Ki67 in HIV-1-infected patients than in the healthy control ( HC ) subjects , while highly active antiretroviral therapy ( HAART ) reduced the activation and proliferation of both CD4+ and CD4- ILC1s ( Fig 3B ) . As expected , HIV-1 also activated CD8 T cells in HIV-infected patients , and that the activation level was significantly decreased after HAART ( Fig 3C and 3D ) . Further , the percentage of Ki67-expressing CD4+ ILC1s , but not CD4- ILC1s , was found to positively correlate with the plasma HIV-1 viral load ( Fig 3E and 3F ) . In contrast , Ki-67 expression in CD8 T cells was not correlated with plasma HIV-1 load in these patients ( Fig 3G ) . These data indicate that HIV-1 infection activated both CD4+ and CD4- ILC1s . In particular , the activation of CD4+ ILC1s , the HIV-1 target population , was positively correlated with the HIV-1 viral load . We next investigated whether HIV-1 infection depletes ILC1s in vivo . Compared to the HCs , ILC1s in CD45+ cells were significantly reduced in the peripheral blood of patients with chronic HIV-1 infection ( Fig 4A and 4B ) , and HAART partially reversed the reduction of total ILC1s ( Fig 4A and 4B ) . Further analysis indicated that the percentage of both CD4+ and CD4- ILC1s in total CD45+ cells was lower in patients with HIV-1 infection than in the HC subjects , while only the CD4+ ILC1s but not CD4- ILC1s were significantly rescued by HAART ( Fig 4C ) . The absolute cell counts of total ILC1s and CD4+ and CD4- ILC1s were found to be largely reduced in patients with chronic HIV-1 infection as compared to those of HC subjects; and HAART successfully recovered the absolute cell counts of total ILC1s and CD4+ ILC1s but not CD4- ILC1s ( Fig 4D ) . Correlation analysis indicated that the percentage of peripheral CD4+ ILC1s was negatively correlated with the plasma HIV-1 viral load ( Fig 4E ) and positively correlated with the CD4/CD8 ratio in the HIV-1-infected subjects ( S7 Fig ) . We further examined whether ILC1s in the gut were also depleted by HIV-1 infection in humans , which is the key lymphoid organ in HIV-1-associated pathogenesis . As shown in Fig 4F , CD4+ ILC1s were significantly depleted in the large intestine in HIV-1-infected patients as compared to the HC donors . The summarized data also showed that the percentage of total ILC1s within CD45+ cells was significantly decreased in the large intestine in patients with HIV-1 infection ( Fig 4G ) . Further analysis indicated that the percentage of both CD4+ and CD4- ILC1s was reduced during chronic HIV-1 infection ( Fig 4H ) . Importantly , when gated on ILC1 populations , the percentage of CD4+ ILC1s was largely decreased and the percentage of CD4- ILC1s was increased accordingly , which indicates that the CD4+ ILC1s were preferentially depleted ( Fig 4I ) . These data indicate that CD4+ ILC1s from both peripheral blood and large intestine are preferentially depleted during chronic HIV-1 infection . ILC1s can produce large amounts of Th1-associated cytokines in response to innate stimulation . We next analyzed whether persistent HIV-1 infection affected the cytokine production ability of ILC1s . As shown in Fig 5A , IFN-γ and TNF-α production by both CD4+ and CD4- ILC1 subsets induced by PMA/ionomycin stimulation were significantly lower in HIV-1-infected patients than in HCs . Similar phenomena were also observed when the ILC1s were stimulated by IL-12 and IL-18 ( Fig 5B ) . HAART failed to rescue the function of ILC1 subsets , with the exception that IFN-γ production was rescued by HAART after IL-12 and IL-18 stimulation ( Fig 5A and 5B ) . We thus conclude that chronic HIV-1 infection impaired the ability of the remaining ILC1s , including CD4+ ILC1s , to produce cytokines . We next examined how HIV-1 infection leads to ILC1 depletion . We discovered that chronic HIV-1 infection significantly up-regulated active caspase-3 expression in both CD4+ and CD4- ILC1s ( Fig 6A and 6B ) . In contrast , caspase1 was not significantly up-regulated in CD4+ ILC1s ( and only slightly increased in CD4- ILC1s ) of patients with HIV-1 infection as compared to the HC subjects ( S8A and S8B Fig ) . HAART could significantly decrease the expression of active caspase-3 in both CD4+ and CD4- ILC1s ( Fig 6A and 6B ) , correlated with rescued ILC1s . These findings indicate that HIV-1 infection leads to depletion of ILC1 subsets via apoptosis-dependent mechanisms . We further investigated whether the Fas/FasL pathway was involved in the apoptosis of ILC1s ( up-regulation of active caspase-3 ) , as reported in ILC3s in our previous study [17] . We found that expression of CD95 was significantly up-regulated in both CD4+ and CD4- ILC1s from patients with chronic HIV-1 infection compared with HC subjects ( Fig 6C and 6D ) . HAART decreased the expression of CD95 in CD4+ but not CD4- ILC1s ( Fig 6C and 6D ) . In contrast , the expression of death receptor 5 ( DR5 ) was not up-regulated in ILC1 subsets in patients with chronic HIV-1 infection ( S8C and S8D Fig ) . Notably , the expression of caspase-3 and CD95 was also up-regulated in CD8+ T cells in HIV-1-infected patients as compared to HC subjects ( S9A and S9B Fig ) . We then investigated whether the Fas/FasL pathway is involved in the apoptosis of ILC1 subsets . After in vitro stimulation with the anti-CD95 agonist antibody , both CD4+ and CD4- ILC1s from HIV-1-infected patients displayed higher levels of active caspase-3 expression than those from HCs ( Fig 6E ) . Accordingly , the number of live CD4+ and CD4- ILC1s was significantly reduced after treatment with the anti-CD95 agonist antibody as compared to the IgG control in HIV-1-infected patients but not in the HC subjects ( Fig 6F ) . Thus , the number of live CD4+ and CD4- ILC1s in HIV-1-infected patients was markedly less than that in HC subjects in response to in vitro stimulation with the same anti-CD95 agonist antibody ( Fig 6F ) . This indicates that ILC1 subsets from HIV-1-infected patients are more sensitive to Fas/FasL signaling than those from HC subjects . We conclude that the Fas/FasL pathway is actively involved in the apoptosis of ILC1 subsets in patients with chronic HIV-1 infection . Sustained IFN-I signaling has been reported to be correlated with and contribute to SIV and HIV-1-induced immune pathogenesis [25–27] . We have proved that depletion of pDCs or blocking IFN-I signaling prevents HIV-1-induced T cell and ILC3 depletion in vivo [17 , 26 , 28] . We thus investigated whether IFN-I signaling also contributes to HIV-1-induced ILC1 depletion in vivo . We treated HIV-1-infected humanized mice with the anti-IFNAR1 mAb [26] from week 6 through week 10 after infection . At 10–12 weeks after infection , we terminated the mice and measured ILC1 number and phenotype in each group . We found that blockade of IFN-I signaling with the anti-IFNAR1 mAb rescued both CD4+ and CD4- ILC1s cells in percentages ( Fig 7A–7C ) and in numbers ( Fig 7D ) as compared to the isotype IgG control group . In addition , we found that blocking the IFN-I pathway significantly decreased CD95 expression on CD4+ ILC1s in humanized mice with persistent HIV-1 infection ( Fig 7E and 7F ) . We further cultured PBMCs from HIV-1-infected patient ex vivo in the absence or presence of pDC-depleting 15B mAbs conjugated with the SAP toxin ( immune toxin 15B-sap ) or the anti-IFNα/β receptor blocking antibody . We observed significant downregulation of both CD95 and active caspase-3 expression in CD4+ ILC1s from HIV-1-infected patients cultured in vitro in the presence of the immune toxin 15B-sap or anti-IFN-α/β receptor antibodies as compared to the IgG control ( Fig 7G ) . Therefore , depletion of pDCs or blockade of IFNAR1 both prevents HIV-1 induced ILC-1 depletion in vitro ( Fig 7H ) . These data indicate that IFN-I signaling contributes to ILC1 depletion during chronic HIV-1 infection . Since HAART fails to restore ILC function in HIV-1-infected patients , we therefore investigated whether blocking IFN-I signaling combined with combined antiretroviral therapy ( cART ) in vivo can rescue the function of ILC1s in HIV-1 infected humanized mice . We treated HIV-1 infected mice with cART at 4 weeks post infection ( wpi ) . As reported [26] , 3 week after cART , the infected humanized mice received α-IFNAR1 mAb treatment for 3 weeks from 7 to 10wpi . The function of ILC1 was analyzed at 12wpi . Interestingly , we found that cART alone restored IFN-γ and TNF-α production by splenic ILC1s under PMA/Ionomycin stimulation ( S10B Fig ) . IFN-I blockade in combination with cART did not further increase IFN-γ and TNF-α production by splenic ILC1s ( S10B Fig ) . The result is different from human studies which indicated that HAART cannot rescue ILC1 function ( Fig 5 ) . One possible reason for the differences is that we started cART treatment in humanized mice at early infection phase ( 4 weeks post HIV-1 infection ) , while in HIV-1 infected patients HAART is usually initialized years after infection at chronic phase of the infection included in our study . Further studies are needed to unveil the effect of IFN-I signaling on ILC1 function .
Our study investigates the heterogeneity of ILC1s in human lymphoid organs , and provides the first piece of evidence to show that HIV-1 can directly infect CD4+ ILC1s and lead to their activation , depletion and functional impairment in vivo in humans and in humanized mice . Successful HAART rescued the number of CD4+ ILC1s but not cytokine production activity via the inhibition of Fas/FasL-mediated apoptosis of ILC1s . This study , therefore , is the first to identify CD4+ ILC1s as important HIV-1 target cells , and may serve as a novel target of HIV-1 therapies aimed at human immune reconstitution . Through a comprehensive analysis of lymphocytes from human spleen , bone marrow , large intestine , small intestine and liver , we found that human ILC1s consist of CD4+ , CD8+ and CD4-CD8- populations and that all these populations are widely present in all lymphoid organs , which has not been described in previous studies [8 , 9 , 20] . We also observed that CD4+ ILC1s expressed immature phenotypes and lower levels of Th1-associated transcriptional factor T-bet and Eomes than CD4- ILC1s and higher level of TNF-α in response to stimulation . It Is not clear where these CD4+ ILC1s are developed and how the immature ILC1 subsets traffic to various lymphoid tissues . A recent study based on mass cytometry- and t-SNE-based analysis showed ILC1s were undetectable across different human tissues [10] . However , in a re-analysis on CyTOF dataset , ILC1s are clearly clustered in lymphoid tissues [29] . Another report also suggested that ILC1s reported in previous studies may be attributable to CD5+ T cell contamination [30] . However , CD5 is also expressed and functions independently of T cells [31] . Indeed , ILC1s express high levels of CD5 in our study and previous report [32] . Therefore , the use of CD5 with CD4 or CD8 in ILCs without confirming surface CD3 or TCR expression does not definitively identify CD4+ and CD8+ T cells [20] . Furthermore , human patients with RAG1 deficiency , who lack T cells , are characterized by the presence of circulating ILC1s at frequencies comparable to those of ILC2s and ILC3s [33] . ILC1s have also been cloned under T-cell-promoting conditions , and have been detected in inflamed intestinal tissues of patients suffering from Crohn’s diseases [5 , 34] . Taken together , our data provide a comprehensive description of the heterogeneity of CD4+ and CD4- ILC1s ( not T cells ) across various lymphoid tissues in humans . The identification of CD4 expression on ILC1s led to the question of whether this population can be infected by HIV-1 . Our results clearly showed that CD4+ ILC1s also express CCR5 and CXCR4 and can be productively infected by HIV-1 both in vitro and in vivo . The relative infection and replication of HIV-1 in CD4+ ILC1s is comparable to that in CD4 T cells . Interestingly , PHA activation of PBMC enhanced HIV infection in both CD4+ ILC1 and T cells . These results indicate that CD4+ ILC1s are HIV-1 target cells and possibly support HIV-1 persistence in patients with chronic HIV-1 infection . Therefore , we identified CD4+ ILC1s as a new target for HIV-1 infection . Further studies to identify whether CD4+ ILC1s serve as an HIV-1 reservoir in HIV patients during HAART will be important for developing strategies for HIV-1 treatment . It has been reported that HIV-1 infection leads to depletion of all ILC subsets , including ILC1s , in circulation [16 , 17 , 19] and lymphoid organs [18] . We discovered here that HIV-1 infection also depleted ILC1s in the large intestine of patients . Unlike the results of a previous study [16] , we found that HAART can rescue the number of peripheral ILC1s in HIV-1-infected patients . This discrepancy could be explained by the difference in the cohorts enrolled in the two studies . Differences in the time of HAART onset may lead to differences in immune reconstitution [35 , 36] , which may then affect the restoration of the number of ILC1s . Our results indicate that HIV-1 infection depletes ILC1s both in circulation and in lymphoid organs . Of particular note , we found that CD4+ ILC1s were preferentially depleted within the total ILC1 population , which indicates that they are more sensitive to HIV-1-induced apoptosis . The mechanism underlying HIV-1-induced depletion of ILC1s is poorly defined . We have reported previously that HIV-1 infection induces depletion of ILC3s via Fas/FasL signaling in a pDC/IFN-I-dependent manner [17] . In the present study , we found that the depletion of ILC1s was also associated with cell apoptosis mediated by the Fas/FasL pathway during HIV-1 infection . We therefore tested the pDC/IFN-I axis in humanized mice with HIV-1 infection . Our data clearly showed that blocking IFN-I signaling with an antibody against IFNAR1 prevented HIV-1-induced depletion of ILC1s in vivo in humanized mice . Furthermore , blocking IFN-I signaling or depletion of pDCs during in vitro culture of PBMCs from HIV-1 infected patients also significantly reduced ILC1 apoptosis and rescued their number . We thus have demonstrated that pDC and IFN-I signaling plays a critical role in ILC1 depletion during chronic HIV-1 infection . Our data demonstrate that HIV-1 infection not only depletes ILC1s but also leads to their activation and functional impairment , as indicated by the significant decrease observed in their production of cytokines , including IFN-γ and TNF-α . Interestingly , HAART rescues ILC1s in number but fails to recover their function of cytokine production in HIV-1-infected patients . In HIV-1-infected humanized mice , however , we found that HAART starting during early phase of infection ( 4wpi ) rescued both ILC1 number and functions in IFN-γ and TNF-α production . This differential effect of HAART on ILC1 function may be due to different treatment time in patients and in humanized mice . For patients in the study , HAART was usually initialized years after HIV-1 infection at chronic infection phase; while HAART was given at early phase of HIV-1 infection ( 4 weeks ) for humanized mice in the study . Indeed , our findings are similar to a previous report in which antiretroviral therapy initialized during acute infection could preserve ILCs in patients [16] . These data also indicate that depletion of ILC1s and their functional impairment may be mediated by various mechanisms during short acute and long chronic infection . We have recently found that pDC depletion or blockade of IFN-I signaling could significantly reduce residual immune activation and restore anti-HIV immunity in HIV-1-infected humanized mice without or with cART [26] . Future studies should focus on the differential mechanisms underlying cell depletion and functional impairment of ILC1 subsets , and determine whether HAART combined with IFN-I blockade can restore ILC1 function in chronic HIV-1 infection in human patients . In summary , we identified subset- and tissue-dependent heterogeneity of ILC1s and provided evidence to show that CD4+ ILC1s are a novel target for HIV-1 infection . Further , we demonstrated that IFN-I signaling contributes to the depletion of ILC1s , at least partly through the Fas/FasL pathway during HIV-1 infection . These new findings , therefore , extend our earlier findings which show that sustained pDC activation and IFN-I production contributes to HIV-1 pathogenesis . Therefore , blockade of the pDC/IFN-I axis will be a novel therapeutic stratagem to reverse HIV-1-induced pathogenesis , including ILC1 depletion and impairment .
Approval for animal work was obtained from the University of North Carolina Institutional Animal Care and Use Committee ( IACUC ID: 14–100 ) . The study protocol on human samples was approved by the Institutional Review Board and the Ethics Committee of Beijing 302 Hospital in China . The written informed consent was obtained from each subject . All samples were anonymized in the study . Human tissue samples , including the spleen , small intestine , large intestine , bone marrow and liver perfusion , used in this study were obtained from adult donors who had undergone liver transplantation as healthy controls . Gut mucosa from HIV-1-infected patients were obtained for pathological diagnosis . Written informed consent was obtained from each donor . Complete RPMI media were used for all cell isolation experiments . Human fetal livers and thymuses ( gestational age 16 to 20 weeks ) were obtained from medically indicated or elective termination of pregnancies through a non-profit intermediary working with outpatient clinics ( Advanced Bioscience Resources , Alameda , CA ) . Written informed consent from the maternal donor was obtained in all cases under regulations governing the clinic . All animal studies were conducted following NIH guidelines for housing and care of laboratory animals . The project was reviewed by the University’s Office of Human Research Ethics , which determined that this submission does not constitute human subjects research as defined under federal regulations [45 CFR 46 . 102 ( d or f ) and 21 CFR 56 . 102 ( c ) ( e ) ( l ) ] . Thirty HIV-1-infected HAART-naïve individuals and 12 HIV-1-infected patients who underwent successful HAART were enrolled in our study ( S1 Table ) . The majority of these individuals had been infected with HIV-1 via sexual transmission , while a few subjects were paid blood donors . Twenty-six uninfected subjects were employed as healthy controls ( HCs ) . The study protocol was approved by the Ethics Committee of Beijing 302 Hospital , and written informed consent was obtained from each subject . Immune cells from human samples were isolated according to previously reported protocols . In brief , peripheral blood mononuclear cells ( PBMCs ) and bone marrow cells were isolated by Ficoll-Hypaque density gradient centrifugation of heparinized blood of enrolled subjects . The spleen was first ground on ice , after which the cells were collected and filtered . The liver perfusion was directly filtered and concentrated by centrifugation ( 750 g , 15 min , 20°C ) , and was layered onto the Ficoll gradient . The small intestine and large intestine were first finely minced using scalpels , and were then incubated with 0 . 8 mg/mL collagenase type IV ( Worthington-Biochemical ) and DNase I ( Roche ) for 1 h before they were filtered through a 70-mm strainer . The filtered cells were collected and isolated in a similar manner to PBMCs . Upon isolation , all the cells were cryopreserved in 90% fetal calf serum plus 10% DMSO for subsequent assay . We constructed NRG-hu mice using a previously reported method [22] . Briefly , human CD34+ cells were isolated from 16- to 20-week-old fetal liver tissues ( Advanced Bioscience Resources , Alameda , CA ) . The tissues were digested with liver digest medium ( Invitrogen , Frederick , MD ) . The suspension was filtered through a 70-μm cell strainer ( BD Falcon , Lincoln Park , NJ ) and centrifuged for 5 min to isolate mononuclear cells by Ficoll gradient centrifugation . After selection with the CD34+ magnetic-activated cell sorting ( MACS ) kit , CD34+ hematopoietic stem cells were injected into the liver of each irradiated ( 300 rad ) 2- to 6-day-old NRG mouse ( 0 . 5 × 106/mouse ) . More than 95% of the humanized mice were stably reconstituted with human leukocytes in the blood ( 60%–90% at 12–14 weeks ) . The level of engraftment was similar in each cohort . All the mice were housed at the University of North Carolina at Chapel Hill . Total leukocytes were isolated from the spleen of humanized mice as previously described [22] . Lymphoid tissues , including red blood cells , were lysed with the ACK buffer , and the leukocytes were stained and fixed with 1% formaldehyde before FACS analysis . The total cell number was quantified by Guava Easycytes with the Guava Express software . An R5-tropic strain of HIV-1 , JR-CSF ( NIH AIDS reagents program , Cat# 2708 ) , was used for inducing persistent HIV-1 infection . Viruses were generated by transfection of 293T cells ( SIGMA-ALORICH , Cat# 12022001-1VL ) . R3A-HSA was constructed by replacing the vpr gene with mouse heat stable antigen ( HSA; CD24 ) as reported previously . Humanized mice with stable human leukocyte reconstitution were infected with JR-CSF or R3A-HSA at a dose of 10 ng p24/mouse , through an intra-orbital injection . Humanized mice infected with mock-transfected 293T cell culture supernatant were used as control groups . For acute HIV-1 infection , viral genomic RNA present in the plasma was measured by real-time PCR ( ABI Applied Biosystem ) . An X4 and R5 dual-tropic strain of HIV-1 , R3B/Av1v2 , was used for the in vitro experiment . Fresh PBMCs were incubated with the infectious HIV-R3A stock , NL4-3 stock or mock stock with or without the neutralizing monoclonal antibody ( Clone CH31 ) for 2 h at 37°C . Then , the cells were incubated in complete RPMI 1640 medium at a density of 2 × 106 cells/ml in the presence of IL-2 ( 50 IU/ml ) and IL-7 ( 20ng/ml ) for an additional 3 days . Alternatively , fresh PBMCs were activated with phytohemagglutinin ( PHA , 5 μg/ml ) or medium in the presence of IL-2 ( 50 IU/ml ) and IL-7 ( 20 ng/ml ) for 24 hours . Then the cells were incubated with the infectious NL4-3 stock or mock stock for an additional 4 days . Intracellular p24 expression on ILC1 subsets or CD3+ T cells was determined by flow cytometry as described above . An anti-IFNAR1 blocking antibody was developed as per our recent report [26] . Briefly , the human IFNAR1 expression cell line 293T was first incubated with the supernatant of the hybridoma and then incubated with the PE-labeled goat anti-mouse IgG secondary antibody . Then , an IFN-I reporter cell line 293T stably transfected with a mouse A2 promoter-driven EGFP was used to screen antibody clones that could block human IFNAR1 signaling . Humanized mice with HIV-1 infection were treated intraperitoneally with anti-IFNAR1 blocking antibodies from 7 to 10 weeks post-infection twice a week at a dose of 400 μg/mouse at the first treatment and 200 μg/mouse for the following treatments . The same dose of mouse isotype IgG2a control was used in all the experiments . Alternatively , the HIV-1-infected mice were treated with combination antiretroviral therapy ( cART ) as reported [26] . HIV-1 infected , cART treated mice were treated i . p . with IFNAR1 blocking antibodies from 7 to 10 wpi twice a week with 400 μg/mouse at the first injection and 200 μg/mouse for the following treatments . A same dose of mouse isotype IgG2a control was use in all experiments . Flurochrome-conjugated antibodies or regents obtained from Biolegend , BD Bioscience , eBioscience and R&D Systems were used in the study . Live/dead fixable violet dead cell dye ( LD7 ) was purchased from Molecular Probes ( Eugene , OR ) . For humanized mice , live human leukocytes ( Y7-mCD45-hCD45+ ) were analyzed for ILC1 subsets and other cell subsets or phenotypes with CyAn FACS ( Dako , Beckman Coulter , Denmark ) . The data were analyzed with the Summit Software . For human PBMCs and various tissue-derived lymphocytes , dead cells were excluded using the fixable viability dye eFluor 450 ( eBioscience ) . The remaining live CD45+ cells were analyzed for phenotypic expression with FACS CANTO II , and the data obtained were further analyzed with the FlowJo software ( TreeStar , San Carlos , CA ) . Cytokines , including IL-2 , IL-12 and IL-18 , were purchased from PeproTech ( Rocky Hill , NJ ) . For surface marker staining , leukocytes were incubated with antibodies on ice for 30 min and then washed and fixed for further analysis . For staining of HIV-1 gag p24 , transcriptional factors , Ki67 and the apoptotic marker active caspase-3 , the cells were stained with the surface marker first , and then permeabilized using a Cytofix/Cytoperm kit ( BD Bioscience ) and stained for intracellular protein . Alternatively , fresh cells were mixed with caspase-1 for 2 h for caspase-1 staining and were then subjected to surface staining . For intracellular cytokine detection , freshly isolated cells were stimulated for 6 h by culturing with PMA ( 50 ng/ml , Sigma ) and ionomycin ( 1 μM , Merck ) in the presence of BFA ( 1 μM ) . Alternatively , the cells were incubated with IL-12 ( 20 ng/ml ) plus IL-18 ( 20 ng/ml ) for 12 h , followed by Golgi-stop for an additional 6 h . The cells were then collected for surface marker staining; this was followed by cell permeabilization and intracellular cytokine staining . For CD107a staining , the cells were incubated with anti-CD107a antibodies from the onset of stimulation . Then , the cells were further incubated with BFA for an additional 6 h . Freshly isolated PBMCs from HC and HIV-1-infected patients were enriched for ILCs by depletion of CD3+ T cells , CD14+ monocytes and CD19+ B cells using microbeads ( Miltenyi Biotech , Germany ) . Then , the enriched cells were sorted on a FACSAria II ( BD Biosciences ) . CD4+ ILC1s were isolated by sorting on live cells , singlets , scatter , and lineage-CD56-CD127+CD4+ cells ( lineage including CD3 , CD14 , CD16 , CD19 , CD34 , CD11c , CD123 , CD117 and CRTH2 ) . CD4+ and CD8+ T cells were directly sorted from PBMCs . Then , nucleic acid was extracted by sorting CD4+ ILC1s , CD4+ T cells and CD8+ T cells using the DNAeasy minikit ( Qiagen ) to measure total cell-associated HIV-1 DNA . HIV-1 DNA was quantified by real-time PCR according to our previous protocol . DNA from serial dilutions of ACH2 cells , which contain 1 copy of the HIV-1 genome per cell , was used to generate a standard curve . Frozen PBMCs from HCs and HIV-1-infected patients were thawed and cultured in complete RPMI ( RPMI 1640 containing 10% heat-inactivated fetal bovine serum , 2 mM l-glutamine , 100 U/ml penicillin and 100 mg/ml streptomycin sulfate ) ( Cellgro , Manassas , VA ) with IL-12 ( 10 ng/ml ) , IL-18 ( 10 ng/ml ) and IL-2 ( 50 IU/ml ) for 12 h . Then , the cells were collected to perform in vitro assays . The cells were incubated in the presence of plate-bound anti-CD95 monoclonal antibody or isotype control antibody ( 5 μg/ml , clone CH11 , Millipore ) for an additional 24 h . Alternatively , the cells were incubated with 15B mAb conjugated with the toxin sap ( 15B-sap , 8 ng/ml ) to deplete pDCs or with anti-IFN-α/β receptor antibodies ( 10 μg/ml , Millipore ) to block IFN-I signaling for an additional 72 h . Then , the cells were harvested , and the number of live cells was counted and stained for active caspase-3 and/or CD95 expression by ILC1 subsets . Data were analyzed using GraphPad Prism software version 5 . 0 ( GraphPad software; San Diego , CA , USA ) . The data represent the mean ± s . e . m values . One-way ANOVA was used for primary comparisons between different groups , and the result was represented by the overall p value . Secondary comparisons between any two different cohorts of mice or patients were performed using a two-tailed unpaired Student’s t-test . Correlations between variables were evaluated using the Spearman rank-correlation test . Results were considered significant at p values <0 . 05 . | Innate lymphoid cells ( ILCs ) , including ILC1 , ILC2 and ILC3 populations , represent a novel cellular family of the immune system and have potentials to produce large amounts of T cell-associated cytokines in response to innate stimulation in the absence of specific antigen stimulation . ILCs have emerged as central players in homeostatic and inflammatory conditions , and correlated with the pathogenesis and progression of multiple human diseases . It is reported that ILCs are depleted in HIV-1 infected patients . However , it is not clear whether HIV-1 can infect ILCs and how ILCs are depleted during HIV-1 infection . Here , we find that ILC1s consist CD4+ and CD4- subsets and both are present in various human lymphoid organs . We show that HIV-1 can directly infect CD4+ ILC1s . HIV-1 infection leads to activation , depletion and functional impairment of ILC1s in humans and in humanized mice in vivo . Blocking IFN-I signaling prevents HIV-1-induced apoptosis of ILC1s both in vitro and in humanized mice in vivo . Our study reveals the CD4+ ILC1 population as a new target for HIV-1 infection and identifies an IFN-I mediated mechanism of ILC1 depletion during chronic HIV-1 infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"highly-active",
"antiretroviral",
"therapy",
"immune",
"physiology",
"cytokines",
"antiviral",
"therapy",
"pathogens",
"immunology",
"microbiology",
"lymphoid",
"organs",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"preventive",
"medicine",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"rna",
"viruses",
"antiretroviral",
"therapy",
"lymphatic",
"system",
"experimental",
"organism",
"systems",
"molecular",
"development",
"cytotoxic",
"t",
"cells",
"vaccination",
"and",
"immunization",
"research",
"and",
"analysis",
"methods",
"public",
"and",
"occupational",
"health",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"t",
"cells",
"mouse",
"models",
"hiv-1",
"immune",
"system",
"blood",
"cell",
"biology",
"anatomy",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"lentivirus",
"organisms"
] | 2018 | Infection and depletion of CD4+ group-1 innate lymphoid cells by HIV-1 via type-I interferon pathway |
Infection with human BK polyomavirus , a small double-stranded DNA virus , potentially results in severe complications in immunocompromised patients . Here , we describe the in vivo variability and evolution of the BK polyomavirus by deep sequencing . Our data reveal the highest genomic evolutionary rate described in double-stranded DNA viruses , i . e . , 10−3–10−5 substitutions per nucleotide site per year . High mutation rates in viruses allow their escape from immune surveillance and adaptation to new hosts . By combining mutational landscapes across viral genomes with in silico prediction of viral peptides , we demonstrate the presence of significantly more coding substitutions within predicted cognate HLA-C-bound viral peptides than outside . This finding suggests a role for HLA-C in antiviral immunity , perhaps through the action of killer cell immunoglobulin-like receptors . The present study provides a comprehensive view of viral evolution and immune escape in a DNA virus .
Viral evolutionary rates can vary strongly depending on the method used to estimate them [1 , 2] . Among Baltimore groups , the fastest evolving entities are single-stranded ( ss ) RNA and reverse-transcribing ( RT ) viruses , with rates ranging between 10−2 and 10−5 substitutions per site per year ( s/s/y ) . The rates of double-stranded ( ds ) RNA and ssDNA viruses range between 10−3 and 10−6 s/s/y , whereas dsDNA viruses evolve more slowly ( 10−3 and 10−8 s/s/y ) [3 , 4] . It is important to note that only few estimates on dsDNA viruses are published . In fact , higher estimates are based on specific genes , as estimated for human papillomavirus 16 ( E6 and E7 ) , human adenovirus ( hexon ) , or JC virus ( VP1 ) , which are in the order of 10−3 s/s/y [4 , 5] . Regarding estimates based on dsDNA complete genomes , all of them range between 10−5 and 10−7 s/s/y [3 , 5] . This finding confirms that viruses are fast evolving entities whereas humans have much lower evolutionary rates ( 10−8–10−9 s/s/y ) . However , the well-established co-divergence of viral populations with their hosts suggests the possibility of low evolutionary rates in viruses as well . For example , polyomaviruses were historically considered to be examples of human-virus co-divergence , and have been used as markers for human migration patterns , with proposed estimates ranging from 1 . 41 × 10−7 to 4 × 10−8 s/s/y [6 , 7] . Detailed studies are needed to better understand dsDNA virus evolution in vivo , especially in viruses that can be considered as potential pathogens . In vertebrates , the major driving force in anti-viral immunity is the high level of polymorphism in human leukocyte antigen ( HLA ) genes . Despite a few recent reports [8 , 9] , limited information is presently available on the extent of viral variability in vivo , especially at the whole viral genome level , and only a few studies have tackled this variability in conjunction with the HLA genotype of infected individuals . Consequently , viral escape mutants—i . e . , viruses that produce mutated peptides that are no longer able to bind to cognate HLA molecules—have been mainly studied for limited model epitopes in in vitro systems and in highly relevant RNA viruses such as HIV , HCV , influenza or dengue ( see the following historical references [10 , 11]; for a recent review and full bibliography on the subject see [12] ) . It is not surprising that RNA viruses can adapt to circumvent the immune responses [4] , but little is known about viral escape in DNA viruses . A better understanding of the epitopes involved in viral escape from the immune system could be useful for the development of vaccines and specific treatments . Here , we initiate a dual approach using the BK virus ( BKV ) as a model . BKV , which was detected for the first time in 1971 , is a 5 . 1 kb dsDNA virus of the Polyomaviridae family that harbors six genes ( Agnogene , VP1 to VP3 , large T antigen “LTA” and small t antigen “stA” ) [13] . The primary infection occurs essentially in childhood and the virus infects up to 90% of the human population . The virus remains persistent throughout life , primarily in the urinary tract [14] . High-level replication mainly occurs in immunocompromised hosts and , more specifically in those receiving modern immunosuppressive regimens , notably post-kidney transplantation . BKV-associated diseases , especially BKV-associated nephropathy , affect 1–10% of transplant recipients [15 , 16] and may lead to loss of the allograft and even death [17] . There are no specific prophylactic or curative treatments , and early diagnosis , as well as quick restoration of immunity ( through dampening of immunosuppression ) , remain the most effective strategies to control the disease .
Access to the virus in the bloodstream and/or urine within a transplant setting , where HLA alleles of both donors and recipients are known , provides a unique opportunity to study viral evolution in vivo in the context of the individual’s ( both recipient and donor ) HLA class I genotype . A retrospective cohort of 96 patients—225 samples—that underwent solid organ ( N = 83 ) or hematopoietic cell transplantations ( N = 13 ) , harboring a minimum of 104 viral copies/mL in whole blood or urine , was selected . Quantitative real-time PCR showed that the viral titers in blood ( 8 . 98 × 104 ± 2 . 47 × 104 copies/mL ) were significantly lower than those in urine ( 2 . 16 ×109 ± 3 . 94 × 108 copies/mL ) ( Mann-Whitney U = 315 . 0 , two-tailed , P < 0 . 0001 ) . After complete deep genome sequencing of all 225 samples and alignment to the BKV Dunlop reference strain ( GenBank accession number NC001538 ) , an average of 110 ± 3 polymorphisms per sample was observed with an average median coverage of 3043 ± 78 reads/position ( S1 Table , GenBank accession numbers KT896230-KT896454; see Methods ) . In total , 37 . 88% of all amino acid positions in the Agnoprotein , 12 . 43% in VP1 , 9 . 97% in VP2 , 11 . 21% in VP3 , 8 . 20% in LTA and 8 . 72% in stA , were found to be polymorphic ( S2 Table ) . Agnogene is the only gene that is not under apparent selective constraints ( Nei-Gojobori test , P = 0 . 8663 ) , while the others are under purifying selection ( Nei-Gojobori test , P < 0 . 0001 , Fig 1 , see Methods ) . Only a few single nucleotide insertions or deletions were detected in the viral genes ( S3 Table ) . Due to methodological limitations ( short reads ) the non-coding control region was not included in the analyses . The occurrence of mutations is the main process generating genetic variability , but other processes , such as genetic drift , gene flow , selection and recombination , are responsible for shaping the genetic structure and variation of viral populations . Here , we present evidence that BKV is under strong purifying selection even in the immunocompromised host . Several specific features of the Polyomaviridae ( e . g . , limited size of the genome , small number of genes and overlapping transcription units ) likely account for this outcome . In addition , the prevalence of purifying selection in essential genes is anticipated in all viruses as there is a requirement to complete the viral cycle , even in immunocompromised hosts . Most mutations in coding regions must be deleterious , and a high substitution rate implies the accumulation of mutations with deleterious effects [18] . This phenomenon is well known in RNA viruses , which have high mutation rates and short replication times . Similar results have been shown comparing mutational fitness effects and evolution in ssRNA and ssDNA viruses [19 , 20] . Our study supports the hypothesis , in concordance with other recent findings [21] , that the evolutionary rate gap between small dsDNA and RNA viruses might not be as wide as previously thought . A recent study in lentiviruses has revealed that the combined effects of sequence saturation and purifying selection can explain the time-dependent pattern of rate variation . Purifying selection acts on the genetic diversity over long timeframes by removing a large number of transient deleterious mutations that are still present within short timeframes [4] . Phylogenetic analysis with all BKV complete genomes available from GenBank ( Fig 2A ) suggested the existence of three large groups or genotypes represented by serotypes I , II/III , and IV , with subtypes within genotypes . Limited differences ( short branch lengths ) between the previously designated genotypes II and III suggested the existence of only one genotype II/III with two subtypes ( in contrast to more pronounced differences between serotypes II and III ) . A similar phylogenetic classification was observed by analyzing only the VP1 gene ( Fig 2B ) . Incidentally , this finding indicated that the current BKV classification should be revised due to inconsistencies between serotyping and genotyping . Next , to establish the genotype of our samples , one reference strain of each genotype and subtype was used for the phylogenetic analysis ( Fig 2C ) . Most of our samples ( 80 . 88% ) belonged to genotype I , whereas genotypes IV and II/III were less represented ( 13 . 78% and 5 . 3% respectively ) . The clustering was patient-dependent but independent of the sample origin ( urine or blood ) and suggested that some samples likely contained a mixture of genotypes . This mixture might be due to multiple lifelong infections or the replication of viruses from the recipient and/or the donor . Intra- and inter-patient evolutionary rates were estimated . BKV sequences from samples with possible recombination or a mixture of genotypes according to the RDP output [22] were removed from the analysis ( see Methods ) . We estimated an intra-patient substitution rate for BKV in transplanted patients in the range of 4 . 90 × 10−4–1 . 22 × 10−3 substitutions per nucleotide site per year ( s/s/y ) . No differences between substitution rates in solid organ and hematopoietic cell transplant recipients were found ( t-test , P = 0 . 2581 ) . To estimate the inter-patient evolutionary rate , the best substitution ( molecular clock ) and demographic model according to marginal likelihood analyses was the relaxed log-normal uncorrelated clock with Bayesian skyline demographic prior . The estimated inter-patient evolutionary rate ranged from 1 . 00 × 10−5–2 . 15 × 10−4 ( 95% HDI ) for a maximum sampling interval of 568 days . The estimate was quite robust to different demographic and molecular clock models ( S4 Table ) . The evolutionary rates based on the maximum likelihood and least-squares methods implemented in treedater were similar when applied to the whole data set ( 4 . 30 × 10−3 s/s/y ) but with large parametric bootstrap confidence intervals ( in the 10−20 to 1014 range ) , thus preventing their consideration as reasonable estimates . However , when the dataset was reduced to the sequences of genotype I ( n = 56 ) the average evolutionary rate was estimated at 1 . 33 × 10−4 ( 95% CI = 3 . 13 ×10−6–5 . 59 × 10−3 ) . These values were close to those obtained with the Bayesian approach described previously . It is usually assumed that RNA viruses evolve at a rate of 10−4 s/s/y , while dsDNA can be close to 10−8 s/s/y [23] . ssDNA viruses with small genomes evolving at rates similar to those of RNA viruses have been reported previously [24 , 25] , as illustrated by the canine parvovirus , with a substitution rate of 1 . 7 × 10−4 s/s/y [26] . In the case of dsDNA , many evolutionary rates have been calculated under the assumption of co-divergence between viral and human populations , as observed for polyomaviruses . Recently , the substitution rate for JC polyomavirus was evaluated at 1 . 7 × 10−5 s/s/y [27] . Based on this result , Bayesian analyses suggested the substitution rate of BKV to be on the order of 10−5 s/s/y [5 , 28] , while another study found only minor nucleotide substitutions in the genes encoding late proteins [29] . Here we estimated a substitution rate for BKV on the order of 10−3–10−5 s/s/y ( Fig 3 ) . Our experimental results show , for the first time using whole-genome sequencing of in vivo viral populations ( in a large monocentric cohort ) , that the genomic evolutionary rate of a dsDNA virus can be as high as that of RNA viruses . It is important to note that the sampling window of sequences may affect the estimates of evolutionary rates , because very short timescales can inflate them . A recent study has shown that estimates of evolutionary rates were lower for broader sampling levels and longer timeframes for both , DNA and RNA viruses , suggesting that the time dependence of substitution rates is ubiquitous among all viruses [4] . For example , lentivirus evolutionary rates from serial samples over a few years within a single patient or host are in the order of 10−3 s/s/y [30] , reflecting those observed in this study in a small dsDNA virus . In addition , a previous study comparing the evolution of ssRNA and ssDNA viruses has shown that small genomes ( < 5 kb ) can evolve rapidly [24] regardless of their encoding material , and that the well-known correlation between genome size and mutation rate [70] can also hold for evolutionary rates . Here , we show that small dsDNA genomes can also evolve as fast as single-stranded ones . Although BKV uses the host DNA polymerase for its replication , the virally-encoded Agnoprotein inhibits dsDNA break repair activity , thereby potentially increasing the error rate during BKV DNA replication [71] . Interestingly , cell tropism of RNA viruses was recently suggested as a key factor in their capacity to evolve , since viruses replicating in epithelial cells ( as BKV ) are characterized by rapid replication and higher substitution rates [72] . To investigate the relationship between the evolutionary rate of the virus and the immunosuppressive drug regimen—hence the strength of the immune system—we analyzed such information in our kidney transplant recipient cohort ( the largest subgroup in our cohort ) . Kidney transplant patients were given either anti-thymocyte globulin ( ATG ) ( immunological high-risk patients ) or anti-Interleukin-2 receptor ( anti-IL-2R ) ( immunological low-risk patients ) as induction treatments , and tacrolimus ( immunological high-risk patients ) or cyclosporine ( immunological low-risk patients ) as maintenance therapy . Mycophenolate mofetil and steroids were also part of both drug regimens ( for high- and low-risk patients ) . Evolutionary analysis of the different subgroups showed no significant differences in the mutational load ( full negative binomial mixed model regression with random effect intercept to account for repeated measures ) nor in inter-patient substitution rates where ranges overlapped between treatments ( ATG 6 . 12 × 10−4–1 . 03 × 10−5 s/s/y , Anti-IL-2R 8 . 60 × 10−4–1 . 36 × 10−5 s/s/y , tacrolimus 4 . 64 × 10−4–9 . 31 × 10−6 s/s/y , and cyclosporine 1 . 72 × 10−3–1 . 11 × 10−5 s/s/y ) . To investigate the genetic immune escape mechanism of BKV , potential T-cell epitopes presented by HLA class I were predicted using both donor and recipient HLA alleles , combined with the viral substitutions found herein ( S1 Fig , S2 , S5 and S6 Tables , see Methods ) . In this way , we determined the putative HLA ligandome of the virus as linked to the individual’s cognate HLA genotype . Interestingly , the two codons in VP2 that appeared to be under positive selection corresponded to codons within predicted epitopes . The VP2 103 codon , the one with the highest level of significant difference , was found in three predicted HLA-C epitopes ( KFFDDWDHKVSTV , FFDDWDHKV and FFDDWDHKVSTV ) , and codon 340 was located within two HLA-A predicted epitopes ( TTNKRRSR and TTNKRRSRSSR ) . We also found a higher fraction of observed amino acid substitutions within HLA-C epitopes compared with the fraction of amino acid substitutions outside of HLA-C epitopes ( one-sided Wilcoxon signed test , P = 3 . 71 × 10−10 ) . The opposite behavior was observed for HLA-A and -B presented epitopes ( one-sided Wilcoxon signed , HLA-A: P = 4 . 17 × 10−29; HLA-B: P = 1 . 35 × 10−26 ) ( Fig 4 ) . This difference in contribution of HLA loci was independent of the transplantation type ( solid organ or hematopoietic ) or the origin of the HLA loci ( whether from the donor or the recipient ) as assessed by a three-way ANOVA ( P = 0 . 7947 ) . Therefore , our results suggest that HLA-C might be specifically involved in the immune response against BKV through its peptide selection capacity for viral peptides . A possible mechanistic explanation for this finding stems from the amply documented interaction of HLA-C with natural killer ( NK ) and T cells expressing the killer cell immunoglobulin-like receptors ( KIR ) . Notably , the relevance of KIR and HLA-C interactions has been described for viral infections [73 , 74] , and the involvement of NK cells in the immune response against BKV has also been reported [75 , 76] , although further investigations should be done to confirm this hypothesis . High evolutionary rates in RNA viruses allow them to escape immune pressures . Interactions between HLA epitopes and viruses have been described for a variety of RNA viruses , such as HIV , HCV , influenza or dengue , while little is known about immune escape in DNA viruses . A few studies in HPV-16 or herpes simplex virus have been done to improve vaccine design and drug development , but those studies have only examined a fraction of the proteins and not at whole-genome sequencing data [77–79] . This work , to our knowledge , is the first in which predicted epitopes from whole genome sequencing have been studied in an in vivo cohort , in conjunction with cognate HLA alleles , to understand the mechanism involved in immune escape in a DNA virus . Our results of viral escape combined with the high evolutionary rate described herein suggest that a combination of drugs should be used as potential treatment against BKV , as commonly used in highly variable viruses such as HIV and HCV , due to the variable viral populations present in a single patient as observed in our study . The present work describes an unusually fast evolutionary rate for BKV in vivo and charts its interaction with the immune system—through the analysis of cognate HLA alleles—whilst considering the whole viral genome and not only candidate epitopes . It further offers a blueprint for similar analyses in other viruses and helps to better rationalize anti-viral therapy and candidate vaccine development . Our results suggest that small dsDNA viruses should be treated as RNA viruses due to their similarities in evolution and immune escape . Thus , a combination of drugs might be necessary for the treatment of BKV , as used for fast evolving RNA viruses . It is important to note that new analytic methods for the study of the evolutionary rates are needed to better understand the effect of time spans and improve the comparison between estimates .
Ninety-six transplanted patients between 2012 and 2013 from the Strasbourg University Hospitals ( France ) with high levels of post-transplant BKV viruria—as detected by routine BKV testing at the hospital’s clinical virology laboratory—were enrolled in this study . Sixty-eight patients underwent kidney transplantation , 12 were lung recipients , 3 received double ( kidney-heart; heart-lung or kidney-pancreas ) transplants and 13 hematopoietic stem cell transplantation . A total of 225 samples , including 197 urine ( from 94 patients ) and 28 whole blood ( from 13 patients ) were included . Urine samples were collected longitudinally for 36 patients . All patients were enrolled in the study following the Helsinki guidelines . Written informed consent for genetic testing was obtained from all patients and the study was approved by the Strasbourg University Hospitals institutional review board ( RNI DC-2013-1990 ) . Urine and whole blood samples were collected , and DNA was purified using the QIAxtractor instrument ( Qiagen , Hilden , Germany ) , following the DX protocol . Extracted DNA was stored at -80°C until analysis . Blood and urine specimens were assessed using the BK virus R-gene quantification kit ( Biomérieux , Lyon , France ) following the manufacturer’s recommendations . DNA was amplified by Phusion Polymerase ( New England Biolabs , MA , USA ) using specific overlapping primers . Nested PCR was performed for samples with a low BKV DNA load ( usually blood samples ) . PCR products were purified using the GeneJET DNA purification Kit ( ThermoFisher Scientific , Waltham , MA , USA ) and quantified with Qubit ( ThermoFisher Scientific , Waltham , MA , USA ) . Twenty-one urine-blood paired samples were used for sequencing by the Sanger method using an ABI Prism 3130 Genetic Analyzer ( ThermoFisher Scientific , Waltham , MA , USA ) . Bi-directional sequencing was performed with the Big Dye Terminator v3 . 1 kit ( ThermoFisher Scientific , Waltham , MA , USA ) following the manufacturer’s recommendations . Chromatograms were analyzed with the Staden package ( 24 ) to obtain the consensus sequence for each sample . These consensuses were obtained to compare with the results after the next-generation sequencing assembly to validate our pipeline . All 225 urine and blood samples were sequenced by NGS . PCR products from the same samples were pooled in equimolar amounts and library construction with barcodes was performed according to the Fragment Library Preparation protocol using the AB Library Builder System ( ThermoFisher Scientific , Waltham , MA , USA ) . Libraries were quantified by Qubit ( ThermoFisher Scientific , Waltham , MA , USA ) and then pooled in equimolar amounts for Template beads preparation using the SOLiD EZ beads System ( ThermoFisher Scientific , Waltham , MA , USA ) . Template beads were subjected to sequencing using SOLiD 5500 ( ThermoFisher Scientific , Waltham , MA , USA ) with the paired-end 75 bp / 35 bp workflow . Sequences were assembled against the Dunlop reference strain ( GenBank accession number NC001538 ) using LifeScope software ( ThermoFisher Scientific , Waltham , MA , USA ) . Comparison with Sanger sequencing was performed to ascertain the correct assemblies . To quantify the variability per sample , mutations were analyzed with SeqMan software ( DNASTAR , Madison , Wisconsin , USA ) . For each sample , we obtained a list of variants with their genomic location , coverage , and quality metrics , among others . To establish a cutoff for variant calling , we introduced internal controls including ( a ) a clone from the Dunlop reference strain , pBK ( BKV34-2 ) plasmid ( ATCC 45025 ) prepared by minipreparation ( ThermoFisher Scientific , Waltham , MA , USA ) ; ( b ) PCR amplicons from the same clone; and ( c ) PCR amplicons in duplicate from three of the samples . These controls were processed using the same sequencing methodology to establish the rate of sequencing and PCR errors . The final list of variants was selected by means of a Fisher's exact one-sided test comparing evidence obtained from the data for every potential polymorphism to the estimated error rate using our internal controls . Based on this analysis , BKV sequence variants found in less than 0 . 5% of reads were removed from the analysis . Sequences were aligned and assembled against the Dunlop strain by Muscle implemented in MEGA version 6 [80] with default parameters in order to compare and determine point mutations , insertions , deletions , and other sequence variations . For better analysis of coding regions , individual datasets per gene were obtained . Further analysis of synonymous and non-synonymous substitutions and the Nei-Gojobori test of neutrality were performed with MEGA version 6 [80] . Phylogenetic analyses of the whole genome consensus sequences obtained from all samples , and for each gene separately , were performed using MEGA version 6 [80] . Maximum likelihood phylogenetic trees were constructed with the general time reversible model ( GTR ) of nucleotide substitution with gamma distribution to account for rate heterogeneity among sites , as this model achieved the lowest AIC score . Similar analyses were performed for 309 BKV complete genome sequences collected from GenBank ( all items found by searching the NCBI nucleotide database for “BK polyomavirus complete genome” ) . To genotype the populations in the different samples , two approaches were performed . First , phylogenetic trees with all our samples and one of the reference strains for each genotype and subtype were obtained following the methodology explained previously . We determined the genotype as the shortest branch distance to one reference . The second approach was based on the methodology proposed by Luo and colleagues , in which point mutations specifically reported in particular genotypes are described [81] . To estimate the evolutionary rates of BKV , intra- and inter-patient analyses were performed . Upon multiple alignment , consensus sequences were tested using RDP software [22] for potential recombination , and those with positive results using at least two different methods implemented in the RDP package were removed from the ensuing analyses . Samples showing mixtures of genotypes were also excluded since they could interfere with the calculation of the substitution rate . To estimate the intra-patient substitution rate , we used urine samples from twenty-five patients collected at different times ( the first positive samples and after 6 months ) . To calculate substitutions per site per year , we considered all the different genomic positions between two different times that were fixed in the populations . All the substitutions that reverted to the reference base were not included since the possibility of them already being present in the ancestral population at a low frequency could not be ruled out . Thereby only substitutions appearing de novo and exhibiting a high proportion in the population ( fixed substitutions , more than 80% of the reads ) were included in this approach . With this methodology , we obtained conservative estimates . To estimate the inter-patient substitution rate , the consensus sequence for the first available urine sample of each patient with a known date of sampling was selected . After being tested by RDP , a dataset of 79 BKV sequences was used to estimate the inter-patient evolutionary rate ( sequences from 15 patients were potential recombinants ) . A maximum likelihood phylogenetic tree was obtained using Phyml [82] with the GTR model with gamma distribution and invariant sites to account for heterogeneity among sites . This model was determined to be the most appropriate for this dataset with jModeltest [83] . TempEst analysis was conducted to detect a correlation between genetic divergence and sampling time , and it assured a temporal signal in our inter-patient dataset ( S2 Fig ) [84] . We used Bayesian estimates of the evolutionary rate with dated tips as implemented in BEAST [85] . Based on previous results by Firth et al . [5] , we considered three molecular clock models ( strict , relaxed log-normal uncorrelated , and relaxed exponential uncorrelated ) and two demographic models ( constant population size and Bayesian skyline ) . The GTR model with a gamma distribution and invariant sites was used as the nucleotide substitution model in all combinations . Model selection was performed through computation of the marginal likelihood using path sampling and stepping stone sampling analyses [86] . A lognormalPrior with a mean of 1 × 10−6 and a standard deviation of 1 . 0 was used for the substitution rate . Two independent runs of 30 million steps with 10% burn-in were used to obtain the median and 95% high probability density intervals for the relevant parameters in each model . In all cases , the effective sample size was > 200 , as checked with Tracer v . 1 . 5 ( available from http://beast . bio . ed . ac . uk ) . In addition , we used the recently developed method of Volz and Frost which uses maximum likelihood and least squares to estimate evolutionary rates and dates based on relaxed molecular clocks . The method is implemented in the R package treedater [87] . To predict BKV-encoded T-cell epitopes that can be presented by HLA alleles , HLA high-resolution typing ( 2 fields ) was done at the Etablissement Français du Sang Grand Est ( Strasbourg ) using a sequence-specific oligonucleotide technology . High-resolution typing data of HLA-A , -B and -C of 75 available donor / recipient pairs were used in each analysis , using the recipient’s viral populations in each case ( S5 Table ) . NetMHC 3 . 4 [88] was used to predict the peptide binding affinities of potential HLA class I epitopes occurring in BKV Dunlop reference proteins to HLA class I alleles of the patients and donors . Peptides eliciting a predicted IC50 of less than 50 nM were considered epitopes . IC50 values represent the concentration of the peptide that will displace 50% of a standard peptide from the HLA molecule . The lower the IC50 value , the stronger is the affinity of the peptide for the tested HLA molecule . According to the NetMHC parameters , peptides with IC50 < 50 nM were considered high-affinity binders . IC50 values of 5 nM and 500 nM were also tested , but a cutoff of 50 nM was chosen as the best indicator ( at a 5 nM threshold not enough peptides were predicted to bind; at 500 nM all possible peptides within a given proteins were predicted to bind ) . Furthermore , all predicted epitopes were tested with NetChop 3 . 1 [89] to predict whether the epitopes could have been produced by the human proteasome using default parameters . All strong binding peptides with a high likelihood of being correctly cleaved ( score prediction higher than the default threshold of 0 . 5 ) were included in further analyses . To calculate the fraction of substituted amino acids within and outside of HLA epitopes , the substitutions detected in the specific viral populations of each patient were mapped onto viral reference proteins , and the number of substitutions that occurred within and outside of the predicted epitopes were calculated for each protein and HLA allele of each patient and donor respectively . The counts were normalized to the number of potentially mutable amino acids per category ( i . e . , within or outside of epitopes ) , to make them comparable across proteins of varying length . Statistical comparison of the internal and external fractions was performed with a one-sided Wilcoxon signed test for each HLA allele to identify the direction of the difference . The P-values were Bonferroni corrected to account for multiple testing . | Little is known about the mechanisms of evolution and viral immune escape in double-stranded DNA ( dsDNA ) viruses . Here , we study the evolution of BK polyomavirus and observe the highest genomic evolutionary rate described so far for a dsDNA virus , in the range of RNA viruses , which usually evolve rapidly . Furthermore , the prediction of viral peptides to determine immune escape suggests a specific role of HLA-C in antiviral immunity . These findings are helpful for future advances in antiviral therapies and provide a step forward in our understanding of in vivo viral evolution in humans . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] | [
"taxonomy",
"organismal",
"evolution",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"genome",
"evolution",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"pathogens",
"microbiology",
"organic",
"compounds",
"viruses",
"urine",
"phylogenetics",
"data",
"management",
"rna",
"viruses",
"amino",
"acid",
"substitution",
"phylogenetic",
"analysis",
"dna",
"viruses",
"microbial",
"evolution",
"amino",
"acids",
"computer",
"and",
"information",
"sciences",
"evolutionary",
"rate",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"evolution",
"chemistry",
"evolutionary",
"systematics",
"polyomaviruses",
"viral",
"evolution",
"biochemistry",
"anatomy",
"organic",
"chemistry",
"evolutionary",
"processes",
"virology",
"viral",
"pathogens",
"genetics",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"genomics",
"evolutionary",
"biology",
"computational",
"biology",
"organisms"
] | 2018 | An unusually high substitution rate in transplant-associated BK polyomavirus in vivo is further concentrated in HLA-C-bound viral peptides |
KSHV is the causative agent of Kaposi sarcoma ( KS ) , a spindle-shaped endothelial cell neoplasm accompanied by an inflammatory infiltrate . To evaluate the role of KSHV vFLIP in the pathogenesis of KS , we constructed mice with inducible expression of vFLIP in endothelial cells . Abnormal cells with endothelial marker expression and fusiform appearance were observed in several tissues reminiscent of the spindle cells found in KS . Serum cytokines displayed a profound perturbation similar to that described in KSHV inflammatory cytokine syndrome ( KICS ) , a recently described clinical condition characterized by elevated IL6 and IL10 . An increased myeloid component with suppressive immune phenotype was found , which may contribute to functional changes in the microenvironment and cellular heterogeneity as observed in KS . These mice represent the first in vivo demonstration that vFLIP is capable of inducing vascular abnormalities and changes in host microenvironment with important implications for understanding the pathogenesis and treating KSHV-associated diseases .
Kaposi sarcoma herpesvirus ( KSHV ) , also called human herpersvirus 8 ( HHV-8 ) , one of the most recently discovered human oncoviruses [1] , displays tropism for different cell types and a dual oncogenic role , both in lymphomagenesis and vascular oncogenesis . KSHV is specifically associated with Kaposi sarcoma ( KS ) and two B-cell lymphoproliferative diseases , namely primary effusion lymphoma ( PEL ) and a large subset of cases of multicentric Castleman’s disease ( MCD ) [1–3] . KSHV is also associated with KSHV inflammatory cytokine syndrome ( KICS ) , a newly described clinical condition characterized by systemic illness , poor prognosis , elevated KSHV titers , increased levels of viral IL6 and IL10 comparable to those seen in KSHV–MCD but lacking the characteristic lymphadenopathy of KSHV–MCD [4 , 5] , and KSHV-associated hemophagocytic syndrome ( VAHS ) , an extremely rare syndrome reported in immunocompromised patients with MCD and markedly elevated levels of serum human IL6 [6] . KSHV has been found associated also with POEMS syndrome , a rare multisystemic nosological entity characterized by polyneuropathy , organomegaly ( particularly cardiomyopathy ) , endocrinopathy , monoclonal gammopathy and skin lesions [7]; however , a role for KSHV in this disease is controversial , and POEMS may be part of the spectrum of the inflammatory abnormalities seen in MCD , whether KSHV-associated or not . Similarly to other related herpesviruses , there is dependency on latency for transformation , although this dogma encountered exceptions and has been subjected to debate [8–11] . KSHV genes regulating viral genomic persistence and capable of inducing cellular transformation are transcribed during latency ( i . e . , LANA , v-cyclin , vFLIP ) , and the KSHV mode of infection is predominantly latent in KSHV-induced tumors [12] . Experimental data indicate a role for the viral FLICE-inhibitory protein ( vFLIP ) in KSHV pathogenesis , as it is a latent gene capable of activating NF-κB [13 , 14] , a hallmark cellular pathway constitutively active in PEL and indispensable for the maintenance of lymphoma cell survival [15–17] . FLIP proteins are a group of cellular and viral proteins identified as inhibitors of death-receptor ( DR ) -induced apoptosis [18 , 19] . They contain two death effector domains ( DED ) capable of inhibiting DED-DED interactions between FAS-associated protein with death domain ( FADD ) and pro-caspase 8 and 10 within the death-inducing signaling complex ( DISC ) responsible for DR-induced apoptosis [20] . Based on the homology of KSHV vFLIP with cFLIP proteins , it has been thought that vFLIP becomes part of the DISC , preventing the recruitment and processing of procaspase 8 and , thereby , FAS-induced apoptosis [19] , although there is little experimental evidence supporting this direct role in apoptosis inhibition . Nonetheless , it is clear is that vFLIP directly binds to IκB kinase ( IKK ) γ , inducing IKKα/β phosphorylation , IκBα degradation and p100 cleavage , resulting in the activation of both the classical and alternative NF-κB pathways [13 , 14 , 21] . Another established function of vFLIP is inhibition of cell death by blocking autophagy [22] . Several groups have developed mice expressing vFLIP in B-cells [23–25] . Among these , our group used a Cre-Lox recombination approach to express vFLIP in all B-cells and specifically in germinal center B-cells , confirming its role in lymphomagenesis and defining the in vivo immunological functions of vFLIP as an abrogator of germinal center formation and immunoglobulin ( Ig ) maturation [23] . Tumors occurring in mice expressing vFLIP in B cells retain major features of PEL , namely B-cell origin , as formally demonstrated by the presence of monoclonal Ig gene rearrangements , and remodeling of BCR with downregulation of B-cell markers , including CD19 and lambda . However , these tumors were also characterized by expression of histiocytic/dendritic cell ( DC ) antigens , consistent with transdifferentiation from B-cells into the myeloid lineage , without excluding a coexisting paracrine effect on the surrounding myeloid cells [23] . Notably , KS lesions are characterized by the presence of inflammatory cells , including numerous histiocytes [26] . Thus , induction of myeloid cell proliferation by vFLIP could be part of the cellular events and microenvironment alterations that occur during KS pathogenesis . The role of vFLIP in vascular oncogenesis is suggested by the in vitro observations that vFLIP induces spindle cell morphology and expression of inflammatory cytokines in endothelial cells and phosphorylation of STAT1 and STAT2|[27–29] . Both spindling and a proinflammatory microenvironment are key features of KS , defined as a chronic inflammation-associated malignancy due to the presence of spindle-shaped endothelial cells , slit-like neovascular structures , and abnormal vascular spaces with extravasation of red blood cells , as well as variable quantities of infiltrating inflammatory cells and secretion of angiogenic and inflammatory cytokines such as VEGF , PDGF , bFGF , TGFβ , IL1β , IL6 and INFγ [30] . However , the role of vFLIP in the initiation of KSHV-related vascular pathogenesis , if any , is largely unknown . A substantial number of studies have indicated that the cell of origin of KS spindle cells is of endothelial origin as these cells express both blood ( e . g . , CD34 ) and lymphatic ( e . g . , VEGFR3 , podoplasmin , LYVE-1 , Prox1 ) endothelial cell markers ( BEC , LEC ) [31–34] and display a gene signature that falls in between the two cell types , albeit closer to LEC [35] . KSHV can infect both BECs and LECs and is capable of reprogramming their transcriptomes to make BECs more alike LECs and viceversa [35–37] . Therefore , to address the role of vFLIP in vascular oncogenesis , we generated mice that express vFLIP under the control of VE-Cadherin promoter , which has been reported to be active in both BECs and LECs [38] . These transgenic ( TG ) mice showed systemic endothelial alterations with increased spindle-like cells and changes in serum cytokines , reminiscent of certain features of KS and KICS . We also observed remodeling of myeloid differentiation toward cell types known to have implications in host microenvironment , tumor immune evasion , angiogenesis and vascular lesion development .
We generated mice expressing vFLIP in endothelial cells by using a recombinant inducible system . Previously generated conditional mice for vFLIP ( ROSA26 . vFLIP knock-in mice ) [23] were bred with mice expressing cre recombinase in the form of a fusion protein with the estrogen receptor under the transcriptional control of VE-Cadherin promoter ( Cdh5 ( PAC ) . creERT2 mice ) [39] , thus resulting in vFLIP expression in endothelial cells upon tamoxifen treatment ( Fig . 1A ) . Before generating ROSA26 . vFLIP;Cdh5 ( PAC ) . creERT2 TG mice , we tried to constitutively express vFLIP in endothelial cells by crossing ROSA26 . vFLIP mice with mice expressing cre recombinase under the control of Tie2 promoter . However , embryonic lethality was observed , suggesting that constitutive expression of vFLIP is detrimental for embryogenesis and incompatible with life . Instead , the inducible ROSA26 . vFLIP;Cdh5 ( PAC ) . creERT2 TG mice ( carrying both cre and vFLIP ) were born at the expected Mendelian frequency and were indistinguishable from their wild-type ( WT ) littermate controls ( carrying only vFLIP ) in terms of fertility and developmental features . Expression of vFLIP was evaluated in 2–3 month-old mice , approximately one month after intra-peritoneal ( i . p . ) injection of tamoxifen in both TG and littermate control mice . vFLIP expression was detected at the RNA and protein level in lung , spleen , liver and heart ( Fig . 1B ) . The level of vFLIP expression was assessed by quantitative real-time RT-PCR in lung , spleen , liver and heart derived from both TG and controls mice , as well as in BC3 PEL cell line and primary KS with lymph node involvement ( Fig . 1C ) . As expected , the highest level of expression was observed in BC3 , where all the cells harbor KSHV multiple copies of the viral genome . The splenic fraction derived from B-cell-specific TG mice show also high level of vFLIP expression , comparable with BC3 , and this reflects with the high percentage of B-cells in the spleen and the fact that vFLIP expression is controlled by a strong promoter ( i . e . , CD19 ) . Instead , the endothelial-specific TG mice express lower levels of vFLIP , although comparable with vFLIP expression seen in primary KS . This is consistent with the percentage of endothelial cells in the organs analyzed , which is lower than the percentage of splenic B-cells . Since antibodies to vFLIP are not adequate for immunohistochemistry or flow cytometry , we monitored transgene expression using antibodies to EGFP , which is expressed in a common transcript with vFLIP due to the insertion of an IRES between the two gene sequences ( Fig . 1A ) . EGFP was detected by immunohistochemistry in cells lining vascular spaces and with the morphologic appearance of endothelial cells in different organs , including intestine ( S1A Fig . ) ; these cells were also positive for the endothelial marker CD34 . While B-cell-specific TG mice expressed vFLIP in the splenic B-zone as expected , the endothelial-specific TG mice expressed vFLIP in the vascularized interfollicular area ( S1B Fig . ) . The endothelial identity of transgenic cells and the endothelial specificity of vFLIP expression was confirmed by flow cytometry performed in the heart , where EGFP was expressed in the vast majority ( 70 . 8% ± 1 . 4% ) , of endothelial cells defined as CD45−CD11b−CD31+ ( Fig . 1D , middle panel ) , but not in splenic B-cells ( Fig . 1E , middle panel ) . Conversely , B-cell specific vFLIP TG mice expressed EGFP in splenic B-cells ( Fig . 1E , right panel ) , but not in endothelial cells ( Fig . 1D , right panel ) . Taken together , these data showed that vFLIP had the expected pattern of expression restricted to endothelial cells . Virtually all organs and tissues were affected by pathological changes ultimately related to endothelial dysregulation . Numerous elongated cells frequently lining poorly formed vascular spaces was diffusely found throughout several organs , but most notably in the myocardial parenchyma of TG mice . In the heart , these endothelial cells lined the capillaries surrounding individual myofibers , but also proliferated into the parenchyma , expressed vFLIP and many retained endothelial markers ( CD34 and/or CD31 ) and expressed Ki67 ( Fig . 2A ) . Similar findings with the presence of spindle cells , and plump endothelial cells lining vascular spaces , were found in several organs including skeletal muscles ( Fig . 2B ) , brown fat ( Fig . 2C ) and brain ( S2 Fig . ) . These proliferating endothelial cells do not express the lymphatic marker PROX1 , in spite of successful staining of lymphatic endothelial cells in sites where these normally occur including skin , intestines and splenic red pulp ( S3A-S3C Fig . ) . An abnormal perineurial proliferation of endothelial-like cells was found in several tissues in the TG mice , including perirenal capsule , diaphragm muscle , salivary gland , pancreas , but not in the controls ( Fig . 3 ) . The abnormalities observed in the pancreas prompted us to check for signs of endocrinopathy ( e . g . , diabetes ) ; serum glucose levels were slightly increased , but the difference was not significant ( Fig . 3 ) . On the abdominal side of diaphragm and in the peripancreatic region , few nerve bundles were also surrounded by hyperplastic perineurial cells , mixed inflammatory cells , lymphocytes and plasma cells and inflammation extended to the adjacent connective tissue . Chronic inflammation , documented with the presence of mixed cell infiltrate of neutrophils , lymphocytes , plasma cells and histiocytes , was found in several tissues , including the peritoneum ( Fig . 4A ) , meninges ( Fig . 4A ) , kidney and skeletal muscle . Both kidneys showed subcapsular areas with numerous spindle cells ( Fig . 3 ) , and the perirenal fat was infiltrated by neutrophils , lymphocytes , plasma cells and histiocytes . Extra-medullary hematopoiesis , with both erythroid and myeloid hyperplasia , was present in the spleen and liver . Peripheral blood analysis showed that TG mice have left-shift ( i . e . , high metamyelocytes and bands with normal neutrophil count ) , suggesting a demand for neutrophils that exceeded their production and release , a scenario usually seen in case of chronic inflammation at different anatomic sites as observed in our TG mice . The mice were viable after tamoxifen administration , but starting as early as few weeks after induction they developed the pathological abnormalities here described and by the age of 3–4 months more than 60% of mice had died ( Fig . 4B ) as result of a systemic illness that comprised myocardial , meningeal , skeletal muscular , peritoneal and perineurial pathological changes . Although i ) the pattern of cytokine perturbation indicates the existence of M2-type polarization , which eventually favors immune suppression and tumor immune evasion rather than autoimmunity , [ii ) vFLIP does not appear to be a particularly immunogenic protein and iii ) KSHV , in general , has developed a wide array of strategies to evade the host immune responses , the mice were not exposed to the transgene during their embryonic development and , thus , they could have theoretically developed immune response toward vFLIP , resulting in a pseudo-autoimmunity that could partially account for the pathological findings and the poor mouse overall survival . Thus , we assessed the presence of a humoral immune response against vFLIP by immunoblotting , but no cross-reactivity was found between a pool of mouse sera isolated from seven TG mice and whole cell lysates derived from lung , spleen , liver and heart of both TG and control mice ( Fig . 4C ) . In vitro ectopic expression of vFLIP in either endothelial or B-cells has been shown to confer a myeloid-prone gene expression profile with production of cytokines that have potential tropism for myeloid cells [28 , 40] . To assess whether in vivo expression of vFLIP is capable of exerting similar effects , a panel of fourteen cytokines and growth factors ( IL10 , IL6 , INFγ , IL1β , IL12p70 , TNF , IL4 , IL2 , IL13 , GM-CSF , Phospho Stat1 , RANTES , IL12/IL23p40 , MCP1 ) was tested in serum samples collected from mice one month after vFLIP induction by tamoxifen ( Fig . 5 ) . We used a flow cytometry bead-based assay , which provides quantitative data and is linear within a large range of concentration ( from 30 fg/ml to 200000 fg/ml ) ( S4 Fig . ) . Compared to control mice , vFLIP TG mice showed increase of IL10 , IL6 , IL2 , IL13 , INFγ , TNF , MCP1 and RANTES . These findings are in line with in vitro data on gene expression profiling obtained in PEL and endothelial cells that ascribed to vFLIP the ability to activate the expression of several cytokines and growth factors potentially implicated in remodeling of the tumor microenvironment by myeloid cells [28 , 40] . Noteworthy , the systemic illness with poor prognosis and the profound changes in cytokines profile , particularly with increased IL6 and IL10 , are aspects similar to those described for MCD and KICS . To gain insights into the mechanism and consequences of the cytokine storm and further assess the effect of transgene expression in vivo , myeloid differentiation was analyzed by flow cytometry with particular emphasis at the cell subsets that could be influenced by or responsible for the observed cytokine perturbation . A large increase in number of CD45+CD11b+Gr1+/− cells was found in lung , spleen , liver and heart , both in endothelial and B-cell specific vFLIP TG mice ( Fig . 6A ) . A more detailed analysis revealed that the myeloid subpopulation preferentially expanded was Ly6G+Ly6Cint ( Fig . 6B ) . These cells were large ( FSChigh ) , have high granularity ( SSChigh ) and expressed high levels of Gr1 , therefore they likely represent granulocytic myeloid derived suppressor cells ( MDSCs ) ( also called polymorphonuclear-MDSCs , PMN-MDSCs ) , as opposed to monocytic-MDSCs ( Ly6GintLy6C+ ) that lack granularity and express lower level of Gr1 [41–43] . Moreover , Ly6G−Ly6C− cell population , which was expanded only in lung and heart , represented cells that did not express Gr1 , were smaller , had no granularity and potentially represent tumor associated macrophages ( TAM ) or DCs based on immunophenotype , although a functional characterization would be necessary to confirm this . Considering that the endothelial cells are a component of the hematopoietic niches , we compared bone marrow from control and TG mice , but no abnormalities in myeloid or lymphoid hematopoiesis were found that could be ascribed to the expression of vFLIP in endothelial cells . The tumor microenvironment has been shown to be deeply affected by myeloid cells , including CD11b+Gr1+ cells , which are able to produce soluble factors , such as Bv8 , that influence angiogenesis , extracellular matrix remodeling , anti-VEGF resistance and mobilization of additional myeloid cells toward premetastatic sites [44] . Therefore , we checked whether the expanded myeloid cell subpopulations were differentially expressing any these factors . No differences were observed in TG versus WT mice in the levels of expression of Bv8 , VEGF and MMP9 , indicating that these cell subsets exert their function in vFLIP-mediated pathogenesis through different mechanisms .
In this study , we have investigated the effect of inducible recombinant vFLIP expression in endothelial cells to model KSHV-associated vascular pathogenesis as observed in KS . Mice developed pathological abnormalities with systemic changes and appearance of elongated spindle-like endothelial cells , mimicking aspects of KS and other KSHV-associated diseases . Mice developed a profound proinflammatory phenotype with perturbation of serum cytokines , similarly to KICS , as well as expansion of myeloid cells , which unveiled a key role of vFLIP in initiating a cascade of events that lead to changes in host microenvironment , ultimately favoring tumor immune evasion , angiogenesis and tumor progression during KSHV pathogenesis . Given the evidence that KSHV can infect both BECs and LECs [35–37] , and vFLIP induces spindling of endothelial cells in vitro [27] , we tested the hypothesis that in vivo expression of vFLIP in endothelial cells would lead to the development of KS-like disease . Mice developed vascular abnormalities with the presence of spindle cells expressing endothelial antigens in virtually all organs , but , unexpectedly , not in the skin , which is the most common location for KS in humans . While the reasons for this finding are unclear , other viral genes are likely to contribute to the many aspects of KSHV pathogenesis in humans in the context of natural infection , including specific organ involvement [45] . Nevertheless , the endothelial-specific vFLIP TG mice generated showed a proliferation of spindle cells , and a proinflammatory phenotype , indicating that this characteristic of KS can be induced by vFLIP alone . In this setting , vFLIP induces expression of cytokines including those that can result in formation of autocrine loops . For example , there is increased production of IL2 , and the IL2 receptor alpha chain is upregulated by NF-κB [46] , which is turn is activated by vFLIP . Similarly , there is an increase of TNF production in the vFLIP TG mice , and the TNF receptor ( CD120B ) molecule is induced by NF-κB [47] , which in turn can further activate the NF-κB pathway creating a positive regulatory loop . However , we did not obtain complete KS phenotype , so it is likely that cooperation with other KSHV proteins ( e . g . , vGPCR , LANA , vCyclin , vIL-6 , K1 ) and/or noncoding transcripts ( e . g . , miR 17–92 , miR K12-7 ) , which are co-expressed in KS and relevant for vascular tumorigenesis , are required for full pathogenesis [48–52] . In this regard , previously reported TG mice for vGPCR and vCyclin also failed to fully recapitulate KSHV-associated vascular diseases although tumorigenic properties of the viral products were otherwise demonstrated [48 , 49 , 53–58] . Expression of multiple viral products has been achieved in B-cells using the latency locus under the control of the native viral promoter , but specific expression in endothelial cells has not been assessed [24] . Our mouse model contrasts with previous TG models of KSHV-encoded genes in the extent of a proinflammatory phenotype . The severity and systemic nature of the endothelial changes were reminiscent of certain features of the POEMS syndrome [7] . Although the etiopathogenesis of this syndrome is still largely unknown , a role for KSHV has been suggested by few studies . First , there is frequent association with KSHV-associated MCD and angioma formation . Second , in POEMS syndrome there is overproduction of proinflammatory cytokines , including TNFα , IL1β , IL10 , IL6 , VEGF [59] , similarly to what is observed in MCD and KICS , suggesting that these three clinical entities partially overlap . Moreover , POEMS is characterized by the presence of monoclonal Ig , usually IgG or IgA with lambda light chain , and KSHV encodes for viral IL6 that is functionally active on human myeloma cells [60] . KSHV was found in the lymphoid cells of MCD , as well as in the microvenular hemangioma , the pathognomonic endothelial lesion , positive for CD34 , CD31 , LYVE-7 and Prox-1 , that characterizes this syndrome [61–64] . However , other studies failed KSHV detection in this syndrome [65 , 66] . Similarities of POEMS with ROSA26 . vFLIP;Cdh5 ( PAC ) . creERT2 TG mice included: i ) neuropathic symptoms , which in mice are likely related to hyperplasia of the perineurium around nerve bundles in spinal nerve roots , ganglion and skeletal muscle , ii ) systemic presence of elongated endothelial cells , particularly in the heart , which is also increased in size , reminiscent of organomegaly seen in POEMS iii ) proneness to develop endocrinopathy ( e . g . , diabetes ) , as suggested by increased glycemic levels observed in TG mice , and iv ) overproduction of proinflammatory cytokines , including TNFα , IL10 , IL6 . While the association between POEMS and KSHV remains controversial , a role for KSHV in KICS is well-established and the cytokine storm observed in the vFLIP TG mice is very reminiscent of that seen in this syndrome [4 , 5] . We also observed remodeling of myeloid differentiation with expansion of CD11b+Gr1+Ly6G+Ly6C+/− cells , phenotypically corresponding to granulocytic myeloid derived suppressor cells ( MDSCs ) . Under physiological conditions , immature myeloid cells from the bone marrow differentiate into granulocytes , macrophages or dendritic cells ( Fig . 7 ) . Tumors are capable of secreting several factors in the tumor microenvironment responsible for changes in myeloid differentiation that ultimately can favor tumor immune evasion , angiogenesis and tumor progression . M1 toward M2 polarization is favored by increase in IL10 and reduction in IL12 , which lead to reduced Th1 activity and tumor immune evasion , along with angiogenesis and tumor promotion . The main myeloid subpopulations responsible for these effects in tumors are TAM , MDSC , and suppressive DC ( Fig . 7 ) . Aberrant CD11b+Gr1+ myeloid cells have also been found in the mouse placenta , where most likely exert immune suppressive and angiogenetic functions to promote immune tolerance and growth of the developing embryo [67] . Mouse MDSCs consist of two major subsets: granulocytic CD11+Ly6G+Ly6Clow cells and monocytic CD11b+Ly6G+/−Ly6Chigh cells ( M-MDSCs ) , which differ in their immunosuppressive mechanisms [43 , 68] . MDSCs derive from the bone marrow hematopoietic precursors due to the altering of myelopoiesis by chronic inflammatory mediators [69] , such as STAT1 and NF-κB , signaling pathways known to be vFLIP targets ( 56 , 59 ) . MDSCs exert their immunosuppressive functions primarily by inhibiting antitumor T-cell function . Moreover , MDSCs are able to secrete angiogenic factors , matrix metalloproteinases and cytokines promoting neoangiogenesis and tumor growth and skewing immune responses towards protumoral Th2-type with activation of Tregs . Thus , MDSCs play a central role in the development of immunosuppressive tumor microenvironment [43] , as also emphasized by the fact that functionally active tumor-specific CD8+ T-cells can develop anergy or undergo apoptosis when adoptively transferred into a microenvironment containing MDSCs; moreover , depletion of MDSCs restore CD8+ T cell function , thus confirming their role in induction and maintenance of host immunosuppression [41] . The cooperation between chronic inflammation and myeloid cell expansion is particularly relevant . In our vFLIP TG mice there is evidence of chronic inflammation at different anatomic sites , sustained also by left-shift in myeloid differentiation . Moreover , vFLIP transcriptome , as defined by in vitro gene expression profiling of both vFLIP-expressing endothelial cells and PEL [28 , 40] , highlights the fact that vFLIP activates several proinflammatory cytokines directly implicated in tumor microenvironment and remodeling of myeloid cells , particularly IL4 , IL10 , IL6 , IL13 , TGFβ , CCL5/RANTES , IL2 , IL1β , G-CSF , similar to those seen in our in vivo data . The myeloid phenotype observed in our vFLIP TG mice , with expansion of phenotypically bona fide granulocytic-MDSCs , is the first demonstration that vFLIP exerts in vivo induction and remodeling of myeloid differentiation with changes in critical components of the microenvironment toward a proinflammatory , angiogenic and immunosuppressive effect . The aberrant myeloid differentiation seems to be a consequence of vFLIP-mediated perturbation of cytokine profiles; once the microenvironment is polarized toward M2 , development of MDSCs rather than Th1 activity is favored . In turn , MDSCs , through the upregulation of molecules such as VEGF , Bv8 and MMP9 , can favor angiogenesis , tumor progression and tumor immune evasion ( Fig . 7 ) . Additional studies are necessary to dissect whether this cytokine storm is produced by myeloid cells or , alternatively , by the endothelial cells with the myeloid cells being a target of this cytokine overproduction . However , the myeloid phenotype with expansion of CD11b+Gr1+cells was observed in both endothelial and B-cell specific vFLIP TG mice , therefore it likely represents myeloid cells chemotactically recruited by the ectopic expression of vFLIP in either cell type , which , in turn , precedes the expression of cytokines known to have tropism for myeloid cells . In addition to vFLIP’s ability to impair GC formation and Ig maturation , this change in cytokine profile with remodeling of myeloid differentiation might represent a novel mechanism developed by KSHV to achieve immune evasion by altering the microenvironment to prevent immune recognition of KSHV-infected cells . Considering that Th1-type responses promote cellular immunity against intracellular pathogens and tumors , particularly meaningful is the evidence that KSHV as oncovirus has developed mechanisms to induce Th2 polarization and sabotage host immunity through manipulation of the microenvironment . Interestingly , also KSHV miR-K12-7 induces the expression of IL6 and IL10 [70] , which by inhibiting DC maturation protect PEL from host immune recognition [71] and simultaneously act as independent growth factors for these cells [72 , 73] . It is likely that myeloid differentiation is also perturbed in KSHV-infected individuals , with M2 polarization and impairment of Th1 activity . Although there is need for prospective studies on myeloid cells in KSHV-infected patients , quantitative and functional defects of peripheral blood DC and monocytes with reduced IL12 and increased IL10 were reported as becoming even more pronounced in advanced stages of KS [74] . Moreover , KSHV-specific CTLs are very rare in patients who progress to KS , supporting the role of Th1 immune responses in controlling KSHV replication and transformation [75] . Finally , the cytokine profile from patients with KSHV-associated disease further sustains the hypothesis based on our in vivo finding that vFLIP-induced M2 polarization of the microenvironment ( with increased IL10 , IL13 , IL4 , INFγ and reduction in IL12 ) is critical for KSHV pathogenesis . KSHV is associated with KS in which tumor identity has been made extremely puzzling by the presence of a rich myeloid component , as well as KICS and MCD , both associated with inflammatory cytokines . Our findings suggest this phenomenon is a result of vFLIP-driven remodeling of the microenvironment through a paracrine effect due to the secretion of myeloid-stimulating factors from vFLIP-expressing endothelial or B-cells ( Fig . 7 ) . Most macrophages in KS lesions do not contain KSHV , largely favoring a paracrine effect , although rare cells co-express LANA and histiocytic antigens [76] . In conclusion , we have revealed a previously unknown function for vFLIP in inducing in vivo expansion of the myeloid compartment with the emergence of a cellular component of immunosuppressive phenotype . This has important implications for the pathogenesis of KSHV-associated malignancies that invariably display a rich myeloid inflammatory infiltrate , which remains poorly characterized . The profound myeloid phenotype induced by vFLIP supports the key role vFLIP has in contributing to host immune dysfunction with development of tumor immune evasion during KSHV pathogenesis . The high-level coordination between cellular and soluble components seen in these mice provide a model to test inhibitors of vFLIP or other immunotherapeutic approaches targeting the microenvironment as potential anticancer agents for KSHV-associated diseases .
To generate mice expressing the transgene in an endothelial-cell specific manner , homozygous ROSA26 . vFLIP TG mice [23] were crossed with heterozygous Cdh5 ( PAC ) . creERT2 knock-in mice [39] of C57BL/6 genetic background; therefore , all experimental mice were on 129/Sv-C57BL/6 genetic background and age-matched littermates were used as controls . Genotyping was performed by PCR analysis on mouse tail DNA . All mice were housed , bred and studied according to the guidelines of Institutional Animal Care and Use Committee at Cornell University . Mice were monitored for pathological changes weekly and sacrificed when visibly ill , according to approved protocols . Statistical analysis of event-free survival was performed by GraphPad Prism v . 5 ( San Diego , CA , USA ) using Kaplan-Meier cumulative survival curve and the log-rank test to evaluate statistical significance . Lung , spleen , liver and heart were isolated during autopsy and promptly processed to obtain a single cell suspension using collagenase A and DNaseI treatment . RNA extraction , RT-PCR and quantitative RT-PCR were performed using standard protocols as detailed in Supporting Methods ( S1 Methods ) . Total protein extracts were prepared from lung , spleen , liver and heart using RIPA buffer , gel electrophoresed on 12% SDS-PAGE gel , transferred to a polyvinylindene difluoride membrane ( Millipore ) and immunostained according to standard methods using anti-FLAG ( M2; Sigma ) and anti-β-actin ( Sigma ) antibodies . Single-cell suspensions prepared from lung , spleen , liver and heart were stained using standard procedures with a panel of fluorescent-labeled antibodies ( see S1 Methods ) . 7AAD was used for the exclusion of dead cells . Data were acquired on LSRII or Aria flow cytometer ( Becton Dickinson ) and analyzed using FlowJo software ( Tree Star ) . Four μm thick formalin-fixed , paraffin-embedded sections were stained for H&E or immunostained with the following antibodies: anti-EGFP ( Abcam ) and anti-CD34 ( MEC14 . 7; Abcam ) . Mice 8–12 weeks of age were subjected to i . p . injection with 0 . 2 ml of tamoxifen ( 150 mg ) ( Sigma ) , dissolved in a mixture of 90% corn oil ( Sigma ) and 10% ethanol ( Sigma ) , and analyzed after 30–45 days . Transgene expression was assessed as early as 1 week after induction and remained constitutive over time . To determine the concentration of a panel of fourteen serum cytokines , a flow cytometry bead-based assay was used , which exploits particle with discrete fluorescence intensities to detect soluble analytes at very low concentrations . GM-CSF , Phospho Stat1 , RANTES , IL12/IL23p40 and MCP1 were quantified using BD Cytometric Bead Array ( CBA ) Mouse/Rat Soluble Protein Master Buffer Kit , while for the detection of IL10 , IL6 , INFγ , IL1b , IL12p70 , TNF , IL4 , IL2 , IL13 BD CBA Mouse Enhanced Sensitivity Master Buffer Kit was used . Each capture bead has a distinct fluorescence and is coated with a capture antibody specific for a soluble protein . The bead populations are resolved in two fluorescence channels of a flow cytometry , and each bead population is given an alphanumeric position indicating its position relative to other beads . Beads with different position can be combined to create multiplex assay and analyze multiple proteins from a single sample . After incubation of the capture beads with analytes and detection reagent , the PE mean fluorescence intensity ( MFI ) of the complex was measured and readings within the assay linear range were used to calculate the serum cytokines concentrations against cytokines standard curve for each analyte ( Becton Dickinson ) . Statistical significance , defined as P<0 . 05 , was assessed by two-tailed unpaired Student’s t-test . | Kaposi’s sarcoma ( KS ) is the most common cancer in men infected with HIV , and also among the most frequent malignancies in Sub-Equatorial Africa . KS is a tumor of endothelial cell origin that is caused by infection with a gamma-herpesvirus , called KS herpesvirus ( KSHV ) or human herpesvirus 8 ( HHV-8 ) . KSHV vFLIP is a viral oncoprotein expressed during latent infection . We report here the generation and characterization of mice expressing KSHV vFLIP in an inducible manner in endothelial cells . Transgenic mice showed: 1 ) systemic endothelial abnormalities , with the presence of fusiform cells reminiscent of the spindle cells found in KS , 2 ) development of a profound perturbation in serum cytokines , reminiscent of the cytokine storm characteristic of KSHV-associated cytokine syndrome ( KICS ) , and 3 ) remodeling of myeloid differentiation with expansion of myeloid cells displaying a suppressive immunophenotype that potentially favors host immune evasion , angiogenesis and tumor progression . This is the first example of significant changes in myeloid differentiation , vascular abnormalities and cytokine perturbation entirely initiated by ectopic expression of a single viral gene , making this mouse model a useful system to dissect the mechanisms viruses use to manipulate the host microenvironment culminating in sabotage of immunity and development of vascular lesions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Systemic Expression of Kaposi Sarcoma Herpesvirus (KSHV) Vflip in Endothelial Cells Leads to a Profound Proinflammatory Phenotype and Myeloid Lineage Remodeling In Vivo |
Crossovers ( COs ) are at the origin of genetic variability , occurring across successive generations , and they are also essential for the correct segregation of chromosomes during meiosis . Their number and position are precisely controlled , however the mechanisms underlying these controls are poorly understood . Neddylation/rubylation is a regulatory pathway of posttranslational protein modification that is required for numerous cellular processes in eukaryotes , but has not yet been linked to homologous recombination . In a screen for meiotic recombination-defective mutants , we identified several axr1 alleles , disrupting the gene encoding the E1 enzyme of the neddylation complex in Arabidopsis . Using genetic and cytological approaches we found that axr1 mutants are characterised by a shortage in bivalent formation correlated with strong synapsis defects . We determined that the bivalent shortage in axr1 is not due to a general decrease in CO formation but rather due to a mislocalisation of class I COs . In axr1 , as in wild type , COs are still under the control of the ZMM group of proteins . However , in contrast to wild type , they tend to cluster together and no longer follow the obligatory CO rule . Lastly , we showed that this deregulation of CO localisation is likely to be mediated by the activity of a cullin 4 RING ligase , known to be involved in DNA damage sensing during somatic DNA repair and mouse spermatogenesis . In conclusion , we provide evidence that the neddylation/rubylation pathway of protein modification is a key regulator of meiotic recombination . We propose that rather than regulating the number of recombination events , this pathway regulates their localisation , through the activation of cullin 4 RING ligase complexes . Possible targets for these ligases are discussed .
Meiosis is a modified cell cycle where two rounds of chromosome segregation follow a single S phase , resulting in the production of haploid gametes . Recombination is a key step in meiosis I , as it results in genetic crossover ( CO ) formation , which establishes physical links between the homologues cytologically visible as chiasmata [1] , [2] . In most species , each chromosome pair has at least one CO ( referred to as the obligatory CO ) , which is required to hold the homologues together during the first meiotic division , ensuring their correct segregation . In most organisms , homologues that lack a CO often segregate improperly , leading to the formation of aneuploid gametes [3] . Meiotic recombination can also lead to gene conversion not associated with COs ( NCOs ) [4] . Meiotic recombination is initiated by the induction of DNA double-strand breaks ( DSBs ) catalysed by SPO11 [5] . DSBs are then resected by exonucleases to generate 3′ single-stranded DNA molecules ( ssDNA ) . In the subsequent step , RecA homologues RAD51 and DMC1 assemble on the ssDNA to form nucleoprotein filaments . These filaments search for homologous sequences and trigger single-strand invasions [6] to generate displacement loop ( D-loop ) recombination intermediates [7] . Depending on the way these D-loop intermediates are processed , different recombination products can be formed . For example , capture of the second DSB end leads to the formation of a double Holliday junction that can be resolved to generate either a non-CO ( NCO ) or a CO [8]–[10] . Alternatively , NCOs can also be formed when a single strand end is displaced after priming a limited amount of DNA synthesis , annealing with the other DSB end in a process called synthesis-dependent strand annealing ( SDSA ) [11] . In most organisms , when multiple COs occur on the same chromosome , they are distributed nonrandomly: One CO prevents other COs from occurring close by , in a distance-dependent manner . This phenomenon results in COs being more evenly spaced along chromosomes than would be expected if they occurred randomly . The term used to describe this phenomenon is CO interference [12] , [13] . In budding yeast , two kinds of COs are known to coexist: class I COs , which are interference-sensitive COs and whose formation depends on the ZMM proteins ( Zip1 , Zip2 , Zip3 , Zip4 , Msh4 , Msh5 and Mer3 ) in addition to Mlh1 and Mlh3 , and class II COs , which are not subject to interference and depend on Mus81 and Eme1/Mms4 [10] . Arabidopsis thaliana , like yeast and mammals , has two recombination pathways: one that exhibits CO interference and another one that does not [14]–[20] . In A . thaliana , disruption of genes acting in the interference-sensitive pathway causes a loss of approximately 85% of COs [21] . In addition , there is evidence that the MUS81 gene accounts for some , but not all , of the 15% MSH4-independent COs , suggesting that MUS81 is involved in a secondary subset of meiotic COs that are interference insensitive [14] , [22] . Very little information is available on the mechanisms controlling interference and the number and distribution of COs during meiosis in general [23] , [24] . Eukaryotes possess a highly conserved mechanism to control protein degradation mediated by the action of the ubiquitin ( Ub ) proteasome system ( UPS ) [25] . In this system , E3 Ub ligases are required to ubiquitylate specific protein targets . Cullin RING ligases ( CRLs ) are the largest class of E3 ligases . Several mechanisms control CRL activity: It can be activated by covalent attachment of the Ub-like protein NEDD8/RUB ( a process called neddylation or rubylation ) [26] , [27] or inhibited by the COP9 signalosome-directed deneddylation [28] . Neddylation/rubylation has been shown to play a crucial role in processes such as morphogenesis in mice [29] , cell division in budding yeast [30] , embryogenesis in C . elegans [31] , meiosis to mitosis transition in C . elegans [32] , and response to various plant hormones [33] , [34] including auxin [35]–[38] . However , neddylation/rubylation had not been connected to homologous recombination ( HR ) . Cullin RING Ligase 4 ( CRL4 ) is associated with DNA repair in plants and humans; the DDB1-CUL4ADDB2 E3 ligase initiates nucleotide excision repair ( NER ) by recognizing damaged chromatin with concomitant ubiquitylation of core histones at the lesion site [39]–[41] . Additionally , CUL4A plays a role in meiotic recombination and spermatogenesis in mice [42] , [43] . Inactivation of cul4a affected male fertility , with increased death of pachytene/diplotene cells and defects in MLH1 dissociation from the SCs . Here we show that the E1 enzyme of the neddylation complex , AXR1 , is a major regulator of meiotic recombination in Arabidopsis . In axr1 mutants , the average number of meiotic COs is unchanged; they are still under the control of the ZMM proteins , but they tend to cluster together and no longer follow the obligatory CO rule . We were able to show that this recombination defect is correlated with strong synapsis defects . In addition , we found that this deregulation of CO localisation is likely mediated by a CRL4 .
In the process of screening A . thaliana T-DNA ( Agrobacterium tumefaciens transferred DNA ) insertional lines for meiotic defects , we isolated three mutants [EGS344 , EIC174 , and EVM8 ( Ws-4 strain ) ; Figure 1 and Figure S1] allelic for disruption in At1g05180 , the AXR1 gene , previously shown to encode the E1 enzyme of the Arabidopsis neddylation complex [44] . Another insertion line in At1g05180 available in the public collection ( http://signal . salk . edu/ ) Sail_904E06 ( N877898 , Col-0 strain ) and the historical axr1 allele ( axr1-12/N3076 , Col-0 ecotype [44] , with a single nucleotide substitution in exon 11 of At1g05180 ) were also included in this study ( Figure 1 ) . The mutant plants all show the same vegetative phenotypes as previously described for axr1 mutants: They are dwarfed , excessively branched , with small rosettes and crinkled leaves ( shown for N877898 in Figure 2A–B and in Figure S2 for the other alleles ) [45] , [46] . They also have small flowers and short fruits , indicating fertility defects ( Figure S2 ) . We examined the reproductive development of these mutants and found that all alleles showed a high level of male and female gametophyte abortion [shown for N8779898 male gametophytes ( pollen grains ) in Figure 2D] . In plants , male gametogenesis occurs in the anthers where groups of meiocytes undergo meiosis synchronously , each producing four haploid cells ( called microspores ) . The four products of each meiosis remain temporally encased in a common callose wall , forming tetrads of microspores that can be visualised after tissue clearing ( Figure 2E ) . Each microspore is then released from its tetrad and continues to develop into a mature pollen grain ( the male gametophyte ) containing the male gametes . Study of the early stages of pollen development in axr1 revealed the presence of abnormal meiotic products . Instead of the regular tetrahedral structure observed in the wild-type , asymmetric tetrads ( containing four daughter cells of unequal size ) or “polyads” ( containing more than four products ) were observed ( Figure 2F ) , suggesting that the meiotic program is disrupted in these mutants . To confirm that the reduced fertility was caused by a defect in meiosis , we investigated male meiosis via chromosome spreading and DAPI ( 4′ , 6-diamidino-2-phenylindole ) staining ( Figure 3 ) . During wild-type meiotic prophase I ( Figure 3A–D ) , DNA fibres of each sister chromatid are organised as chromatin loops connected to a common protein axis ( the axial element [AE] ) [47] . When chromosomes start to condense at leptotene , they become visible as threads ( Figure 3A ) . At this stage , meiotic recombination is initiated by the formation of a large number of DNA DSBs ( not shown ) . HR repairs these breaks concomitantly with the progression of synapsis , the close association of the homologous chromosome axes through the polymerisation of the central element ( CE ) of the synaptonemal complex ( SC ) . Synapsis begins at zygotene ( not shown ) and is complete by pachytene , when complete alignment of homologous pairs can be detected in DAPI-stained chromosomes ( Figure 3B ) . DNA repair and recombination are thought to be achieved during pachytene , yielding at least one CO per homologous chromosome pair . At diplotene ( Figure 3C ) , when the CE of the SC is depolymerised , the homologous chromosomes are therefore connected to each other by COs in which chromatids from homologous chromosomes have been exchanged . These connections between homologous chromosomes become apparent only at diakinesis ( Figure 3D , arrows ) , when chromosomes are sufficiently condensed . At this stage in Arabidopsis , chiasmata ( the cytological manifestations of COs ) cannot be scored precisely , but chiasma-carrying chromosome arms can sometimes be identified based on bivalent appearance ( see Figure 3D , arrows ) . Next , condensation proceeds and , at metaphase I , the five Arabidopsis bivalents are easily distinguishable , aligned on the metaphase plate ( Figure 3E ) . During anaphase I , sister chromatid cohesion is released from chromosome arms , allowing homologous chromosomes to segregate to the two opposite cellular poles ( Figure 3F ) . The second meiotic division then separates the sister chromatids , generating four pools of five chromosomes ( Figure 3G and 3H ) , which gives rise to the tetrads of four spores ( Figure 2E ) . In A . thaliana axr1 mutants , the leptotene and zygotene stages appeared similar to those in the wild type . However , no pachytene cells were identified in the 457 meiocytes analysed , in contrast to wild type , where this stage is present in approximately 35% of the cells ( n = 334 ) . Instead , we observed pachytene-like stages , with only partial chromosome alignment ( Figure 3J ) . This suggests that axr1 is defective in synapsis . Diplotene cells were indistinguishable from those in the wild type ( Figure 3K ) . Then , chromosome condensation could be followed until metaphase I , although diakinesis stages were rarely observed ( 1% of all stage cells , n = 457 for N877898 , 12% in wt , n = 334 ) ( Figure 3L ) . At wild-type metaphase I , the five typical Arabidopsis bivalents could be observed aligned on the metaphase plate ( Figure 3E ) . Each bivalent was composed of two homologous chromosomes connected by chiasmata either on one chromosome arm ( rod bivalent , Figure 3E# ) or on both pairs of chromosome arms ( ring bivalent , Figure 3E* ) . Chiasma numbers could therefore be estimated based on the bivalent structure . However , because multiple COs on a single arm cannot be cytologically differentiated from single COs , these estimates only correspond to a minimum chiasma number ( MCN; Figure 4 , Table S1 ) . In axr1 mutants , we observed reduced bivalent formation , and instead of five bivalents , a mixture of bivalents and univalents could be identified ( Figure 3M ) . The reduction in bivalent formation resulted in chromosome mis-segregation during subsequent anaphase I ( Figure 3N ) , whereas the second meiotic division separated sister chromatids ( Figure 3O ) , giving rise to a variable number of daughter cells containing aberrant numbers of chromosomes ( Figure 3P ) . We quantified the decrease in bivalent formation as well as the MCN at metaphase I from all axr1 mutants and their respective wild-type accessions ( Figure 4 , Table S1 ) . On average , axr1 mutants had 78% of the wild-type number of bivalents for the Col-0 background and 52% for the Ws background . In terms of the chiasma number , axr1 mutants displayed a residual level of 56% and 41% of the wild-type levels for Col-0 and Ws strains , respectively ( Figure 4 ) . Within a single ecotype ( Col-0 or Ws ) , all alleles were statistically different from the wild type but not different from each other . Finally , when the partitioning of the residual chiasmata in axr1 was analysed , we observed that a large proportion of metaphase I cells showed both ring bivalents ( at least two chiasmata ) together with univalents ( no chiasma ) ( 42% of the N877898 cells , n = 47 ) , showing that in axr1 , the obligatory CO is lost . To further analyse the bivalent shortage observed in axr1 , we used fluorescence in situ hybridization ( FISH ) analyses on PMCs . Metaphase I chromosomes were labelled with probes for the 45S and 5S rDNA repeats , allowing specific identification of chromosomes 1 , 2 , and 4 ( Figure 5 ) . Chromosomes 3 and 5 could not be discriminated from each other with these probes and were pooled . First , we observed that in axr1 as in wild type , bivalents were always formed between homologous chromosomes ( n = 147 bivalents for axr1 , n = 165 for wt ) . Then , we considered each bivalent individually and determined which pair of chromosomes was involved in its formation . As shown in Figure 5D , in axr1 , as in the wild type , each pair of chromosomes was equally involved in bivalent formation , showing that the decrease in bivalent formation observed in axr1 affected all chromosomes in the same way . In wild-type Arabidopsis , the majority of COs ( 85%–90% , depending on the genetic background Col-0 versus Ws-4 ) depend on the ZMM proteins ( MSH4 , MSH5 , MER3 , ZIP4 , SHOC1/ZIP2 , HEI10 , and PTD ) as well as on MLH1 and MLH3 [21] , [48] , whereas MUS81 is responsible for 10%–15% of the remaining COs [14] , [22] . We measured bivalent formation frequencies and the chiasma frequencies in various genetic combinations compared to the single axr1 mutant ( Figure 4 , Table S1 ) . For all the zmmaxr1 double mutants ( except mer3axr1 ) and regardless of strain ( Col-0 versus Ws-4 ) , the level of bivalent formation was reduced by more than 95% with hardly any bivalents observed ( from 0 . 13 to 0 . 18 bivalent per cell; Table S1 ) , showing that almost all the COs in axr1 are ZMM-dependent . We also analysed the bivalent frequency in the axr1mus81 double mutant , which was the same as for the axr1 single mutant ( 3 . 77±1 . 03 against 3 . 75±1 . 12; p = 0 . 9 ) ( Figure 4 ) . We then quantified bivalent frequency in the axr1msh5mus81 triple mutant and observed , as expected , a dramatic decrease in bivalent formation compared to axr1mus81 ( Figure 4 ) . No difference could be detected between the axr1msh5mus81 triple mutant and the axr1msh5 double mutant ( p = 0 . 2 ) . These results show that CO formation in axr1 mutants is almost exclusively dependent on ZMM proteins , whereas the MUS81 pathway plays only a limited role , if any . To further analyse recombination events in axr1 , we immunolabelled chromosomes with antibodies directed against HEI10 and MLH1 , two markers of class I COs in Arabidopsis [48] , [49] . MLH1 foci can be seen from late pachytene to diakinesis [49] , whereas HEI10 is first loaded early during prophase on a large number of sites forming foci of different sizes on chromosomes . A limited number of these foci then remain ( Figure 6A and B ) at sites that correspond to class I COs where they co-localise with MLH1 until the end of prophase [48] . We therefore counted HEI10 and MLH1 foci in late pachytene and diplotene cells in wild type and axr1 . Surprisingly , the average foci number per cell was not different between wild type and axr1 , for either HEI10 ( 8 . 30±0 . 29 , n = 54 and 7 . 49±0 . 40 , n = 84 , p = 0 . 15 ) or MLH1 ( 8 . 61±0 . 29 , n = 33 and 7 . 58±0 . 54 , n = 91 , respectively , p = 0 . 263 ) . In addition , we confirmed that these foci localise to chiasma-containing arms at diakinesis ( Figure 6E and F and Figure S3 ) , showing that they are likely to mark CO sites in axr1 as in wild type [49] . We also observed that there was higher variability in the numbers of HEI10 and MLH1 foci in axr1 than in wild type ( Figure 6G ) , with the coefficient of variation ( standard deviation divided by the mean ) varying from 26% ( HEI10 , wt ) to 50% ( HEI10 , axr1 ) or from 19% ( MLH1 , wt ) to 68% ( MLH1 , axr1 ) . Another striking feature of axr1 was the frequent occurrence at the pachytene-like and diplotene stages of portions of paired chromosome axes where adjacent HEI10 and MLH1 foci could be seen ( Figure 6C , D , arrows and Figure 7A , arrows ) . Forty-seven percent ( HEI10 , n = 60 ) or 53% ( MLH1 , n = 66 ) of the cells had at least two foci localised on the same portion of a chromosome axis , whereas in wild type , this scenario occurred only in 7% ( HEI10 , n = 57 ) or 3% of the cells ( MLH1 , n = 39 ) ( Figure 7B ) . In addition , although we never observed more than two adjacent foci in wild type , we observed 22% ( HEI10 ) and 13% ( MLH1 ) of the cells with more than two adjacent foci , with a maximum of five adjacent HEI10 foci observed in axr1 ( Figure 7B ) . Therefore , although the average level of class I COs is the same in axr1 and in wild type ( Figure 6G ) , these class I COs tend to cluster together in at least 50% of the axr1 cells . We then estimated the scale at which this clustering arises . The distance between clustered foci was measured and compared to the total length of chromosome axis . The distance between two adjacent foci was on average 1/400 of the total axis length of a cell , ranging from 1/1600 of the genome to a maximum of 1/90 of the genome ( Figure S4A ) . Extrapolated in DNA distance , with the additional assumption that genome condensation is homogeneous , the distance between two adjacent foci in a cluster is therefore expected to vary from 150 kb to 3 , 000 kb , with an average of 625 kb . We also observed that the distance between two adjacent foci does not vary significantly in clusters with exactly two foci compared with clusters with more than two foci . As a consequence , cluster size increases proportionally with the number of foci present in the cluster ( Figure S4B ) . The size of the clusters was on average 1/200 of the genome for HEI10 foci ( n = 14 , 1 , 200 kb ) and 1/300 for MLH1 ( n = 21 , 800 kb ) . Finally , we examined whether the clustered foci displayed interference , as might be expected for class I COs . We thus considered the hypothesis H0 that the foci in clusters are not subject to interference . The test was based on the distribution of distances between adjacent foci , specifically using the coefficient of variation for the statistical test and comparing to 105 simulations under H0 ( see Materials and Methods ) . For the clusters of three or more foci ( Table S2 ) , we rejected the H0 hypothesis of no interference for MLH1 foci ( p = 0 . 0024 based on seven clusters ) , for HEI10 foci ( p = 0 . 0028 based on six clusters ) , and when pooling the MLH1 and HEI10 data ( p = 2 . 4×10−5 based on 13 clusters ) . Specifically , inside clusters , MLH1 and HEI10 foci are more evenly distributed than at random , showing that COs within clusters still interfere . Taken together , these results show that the shortage in bivalent formation observed in axr1 mutants is not due to a general decrease in CO formation but rather to a mislocalisation of class I COs that tend to cluster together . The level of genetic recombination on several chromosomal intervals was measured using the Fluorescent-Tagged Lines ( FTL ) tool developed by Copenhaver et al . [50] . The FTL system is a visual assay based on segregation of genetically linked fluorescent proteins expressed in the pollen grains of the quartet mutant ( qrt1 ) , in which the pollen grains remain attached as tetrads . With these lines , a large number of meiotic products can be visually scored and then a subset of multiple CO events can be identified ( two- , three- , and four-strand double COs in adjacent intervals and four-strand double COs within a single interval ) ( [50] and Table S3B ) . Six different intervals were used , either on chromosome 3 ( I3b and I3c ) or 5 ( I5a , I5b , I5c , and I5d ) , with sizes ranging from 1 , 200 to 4 , 900 kb ( Table S3A ) . We first measured recombination rates for each interval using the standard Perkins genetic mapping equation [51] . As shown in Table 1 , recombination rates in axr1 vary differently depending on the interval tested , from 70% to 180% of the wild-type level . On average , axr1 shows an increase in recombination , but these data should be taken with caution , as recombination measurements rely only on a subset of tetrads ( the viable tetrads ) . Out of the six intervals considered , intervals located close to the telomeres ( I3b and I5b ) showed the most significant increase in recombination , whereas proximal intervals appeared less affected . This could indicate that the level of recombination is affected differently according to the location on the chromosomes , although additional data will be required to determine if telomere proximity increases CO frequency in the mutant . We then used the FTL data to estimate interference between COs occurring in adjacent intervals ( Table 2 and Table S4 ) . We calculated the Interference Ratio ( IR ) as defined by Malkova et al . [18] , which compares the genetic length of one interval with and without the presence of a simultaneous event in the neighbouring interval . When the occurrence of a CO in one interval reduces the probability of a CO occurring in the adjacent interval , the IR is less than 1 , indicating ( positive ) CO interference . When COs in the two adjacent intervals are independent of each other , the IR is 1 , and if the presence of one CO in an interval increases the probability of an additional CO in the adjacent interval , the IR is greater than 1 , indicating negative interference . As shown in Table 2 , all wild-type IRs were less than 1 , in agreement with the presence of CO interference . For axr1 , however , all IRs increased dramatically and were statistically significantly different to wild type ( p<0 . 0001 , Table 2 and Table S4 ) . In addition , all axr1 IR values were greater than 1 , although only one pair of intervals tested was significantly different from 1 ( I5a I5b , first data set , IR = 1 . 63 , p = 4×10−3 ) . Therefore , in axr1 , adjacent COs appear to occur more frequently than in wild type , which is in agreement with the previously observed clustering of class I COs scored cytologically ( Figure 7 ) . The cytologically observed clustering is occurring at a very small scale , namely a few hundred kb ( on average 1 , 200 kb for HEI10 foci and 800 kb for MLH1 foci , see above ) , whereas in FTLs pairs of intervals correspond to more than 3 , 000 kb ( I5cd , I3bc ) and up to 7 , 500 kb ( I5ab ) . Consequently , most of the clusters are expected to be present within a single interval and to only occasionally affect two adjacent intervals , which could explain why only one pair of intervals showed significant negative interference . Double CO events within a single interval can be detected using the FTLs if the two COs involve four different chromatids ( Table S3B ) because they will generate nonparental ditype ( NPD ) tetrads [50] . Interference within single intervals can be estimated by comparing the observed number of double COs ( NPD frequency ) to the expected number of double COs under the hypothesis of no interference [52] . The ratio between these two numbers ( NPDr ) gives the strength of interference within the considered interval , even if an important proportion of multiple COs will be silent . We calculated NPDr for all intervals considered for wild type and axr1 ( Table 1 and Table S5 ) . In wild type , the NPDr indicated strong interference ( NPDr close to 0 . 3 ) within all the intervals ( except for I3c , which is too small for statistically meaningful data , Tables S3A and S5 ) . In axr1 , however , the NPDr increased systematically ( between 0 . 7 and 1 . 47 ) and was mostly greater than 1 . For two intervals ( I5a and I5b ) , the NPDr values of 2 . 69 and 1 . 63 were statistically significant ( p<0 . 01 ) , showing negative interference ( Table 2 ) . Thus , genetic analyses allowed us to measure negative interference in several of the intervals tested , confirming the CO clustering observed in cytology . To verify whether the recombination defect in axr1 could be linked to a defect in recombination initiation , we used two methods to investigate DSB formation . We first introgressed the axr1 mutation into a rad51 mutant , defective for meiotic DSB repair . In this mutant , DSBs are formed but are then repaired abnormally , leading to significant chromosomal defects ( such as chromosome bridges and chromosome fragmentation ) during anaphase I ( Figure S5A ) . These chromosomal defects persisted in axr1rad51 , showing that DSBs are present in the axr1 mutant ( Figure S5B ) . Second , we analysed the nuclear distribution of the DMC1 protein , a meiosis-specific recombinase that forms foci at recombination sites . The dynamics and number of AtDMC1 foci in axr1 ( 237±40 , n = 7 ) were indistinguishable from wild type ( 234±89 , n = 28 ) ( t , p = 0 . 9 ) ( Figure S5 ) . Thus , the meiotic defects observed in axr1 are not correlated with a decrease in the amount of recombination initiation events . During meiotic prophase , chromosomes are structured in the context of a protein axis ( the AE ) , which is crucial for most meiotic events , including meiotic recombination and synapsis [53] , [54] . The meiotic chromosome axis is composed of specific AE proteins , such as ASY1 and cohesion proteins ( REC8 and SCC3 , [55] , [56] ) . In wild-type meiotic cells , cohesins are loaded as early as premeiotic G1 , whereas ASY1 appears at leptotene first as foci , then as a linear signal throughout the entire chromosome length ( Figure S6A ) , in a pattern similar to that of cohesins ( Figure S6C , [56] ) . As shown in Figure S6 , the signal observed in axr1 mutants cannot be differentiated from wild type , showing that no major alteration of the axis can be detected in axr1 mutants . We then analysed the progression of synapsis by immunolocalisation of ZYP1 , the A . thaliana CE component [57] . In wild type , ZYP1 appeared on chromosomes as foci that quickly elongated to yield a mixture of foci and short stretches of ZYP1 ( Figure 8A , B , red signal and Figure S7 ) . Synapsis then progressed until complete synapsis was reached , defining the pachytene stage ( Figure 8C , D and Figure S7 ) . In axr1 , the early stages of synapsis could not be distinguished from wild type , showing a mix of foci and short ZYP1 stretches ( Figure 8E , I and Figure S7 ) . As meiosis progressed , ZYP1 elongation could be detected ( Figure 8F–L and Figure S7 ) , but full synapsis was never achieved ( n = 66 ) , confirming the synapsis defect detected after DAPI staining of meiocyte spreads ( Figure 3 ) . In addition , in approximately half of the cells , ZYP1 signals appeared strongly perturbed , uneven in thickness and forming dotted lines rather than a homogeneous continuous signal ( Figure 8J or G and Figure S7 ) . In some cases , only short and thick ZYP1 stretches were detected . These could correspond to ZYP1 poly-complexes rather than to CE polymerisation ( Figure 8L and Figure S7 ) . To follow the progression of meiotic recombination events , we co-immunolocalised ZYP1 and HEI10 , using a lipsol spreading protocol that has the advantage of allowing the simultaneous detection of these two proteins [58] but also the disadvantage of preventing examination of prophase after pachytene [59] . As mentioned above , HEI10 is detected as foci on meiotic chromosomes from leptotene to diakinesis , and its dynamics reflect the progression from early recombination intermediates to mature class I COs [48] . During leptotene and early zygotene , HEI10 forms numerous foci of variable size on chromatin ( Figure 8A and Figure S8 ) . Then , during synapsis initiation , bigger and brighter HEI10 foci appear , often co-localising with synapsed regions ( Figure 8B and Figure S8 ) . At this stage and later on , a combination of large and small foci are observed , forming “strings of HEI10 pearls” on ZYP1 stretches ( Figure 8B , C and Figure S8B , C , arrows ) . At late pachytene , only a few bright HEI10 foci , corresponding to mature class I COs , are retained ( Figure 8D and Figure S8 ) . Nevertheless , during most of the pachytene stage , bright HEI10 foci are present , together with faint HEI10 signal marking the CE ( Figure S8C , D ) . In axr1 , the dynamics of HEI10 progression were the same as in wild type with HEI10 detected as multiple foci during early prophase stages ( Figure 8E , I and Figure S8 ) . Brighter foci then appeared as synapsis progressed , also forming a string of pearls on ZYP1 stretches ( Figure 8J and Figure S8 , arrows ) . A subset of very bright foci was retained at the later stages ( Figure 8G , H , K , L and Figure S8 ) . We noticed that at these late stages ( based on the HEI10 pattern ) , the level of synapsis varied considerably from one cell to another . In addition , although these late HEI10 foci were always observed on ZYP1 stretches , the reverse was not true and ZYP1 stretches without late HEI10 signals were observed ( see , for example , Figure 8H , where four late HEI10 foci are clustered on a single ZYP1 stretch , whereas many ZYP1 stretches are deprived of HEI10 foci ) . Therefore , it appears that class I CO clustering in axr1 is correlated with strong synapsis defects , but cannot be explained by the limited extension of the SC . Because neddylation is known to regulate the activity of CRLs , we investigated whether axr1 meiotic defects are dependent on a specific CRL . In A . thaliana only four cullins are neddylated: cullin 1 , cullin 3A , cullin 3B , and cullin 4 [33] . To identify possible AXR1 downstream players , we scored cullin-deficient lines for meiotic defects . Complete suppression of any of cullin functions ( null cul1 or cul4 or the double cul3a cul3b mutants ) is lethal , but various genetic backgrounds deficient in cullin activities are available We first investigated meiosis of the auxin response defective cul1 mutant alleles—cul1–6 [60] , axr6-2/N3818 [61] , and axr6-3/eta1 [62]—and observed perfectly normal meiosis ( not shown ) . Next , considering cullin 3 activity , we analysed the CUL3a/3b hypomorphic mutant [cul3w ( cul3a3cul3b1 ) ] described for its defects in various aspects of the ethylene biosynthesis pathway and root development [63] . cul3w plants also showed normal meiotic development of male meiocytes ( not shown ) . Finally , we analysed the cul4-1 mutant in which a T-DNA is inserted occurred in the 12th exon of the gene , leading to aberrant CUL4 mRNA expression , which varies depending on the developmental stage [64] . We observed significant male and female gametophyte abortion in cul4-1 ( shown for the male , compare Figure 9A to Figure 2C ) . Although in wild type only balanced tetrads of microspores were observed , asymmetric tetrads and polyads were seen in cul4-1 mutants ( compare Figure 9B to Figure 2E ) . Male meiosis was then investigated . The first stages of meiosis proceeded normally in cul4-1 mutants , however we observed metaphase I phenotypes reminiscent of the axr1 defects , with a large proportion of cells showing a clear reduction in bivalent formation ( Figure 9C ) . The MCN per meiotic cell in cul4-1 ( 6±3 . 2 , n = 71 ) was significantly different from wild type ( 8 . 9±0 . 9 , n = 51 , p<0 . 0001 ) , and slightly different from axr1 ( 5 . 1±1 . 5 , n = 74 , p = 0 . 02 ) . Nevertheless , the number of MCN per cell in cul4-1 was far more variable than in axr1 ( Figure 9E ) , due to an overrepresentation of cells with wild-type levels of chiasmata ( Figure 9D , E ) . We then introgressed the axr1 mutation ( N877898 ) into cul4-1 and found that the double mutant cannot be distinguished from the single axr1 in terms of meiotic phenotype ( not shown ) , the average level of MCN per cell ( 4 . 9±1 . 8 , n = 98 , p = 0 . 412 ) , and in terms of variability of the values ( Figure 9E ) , showing that axr1 is epistatic to cul4-1 . Overall , our results suggest that AXR1 acts during meiotic recombination through the activation of a CRL4 complex .
We observed that in axr1 mutants , meiotic nondisjunction is correlated with defects in bivalent formation . However , our results indicate that the general level of meiotic recombination in axr1 is close to that of wild type , as we showed that COs are mostly under the control of the ZMM pathway and that their average number , revealed by MLH1 and HEI10 foci , is unchanged . Furthermore , these CO events show a completely aberrant distribution in axr1 . First , cytogenetic data showed that clustered MLH1 or HEI10 foci are observed in approximately 50% of the meiocytes ( Figure 7B ) . Second , genetic data showed that adjacent COs in most tested intervals no longer display genetic interference , showing that CO distribution is abnormal in axr1 . More strikingly , in several intervals , strong significant negative interference was detected , a genetic demonstration of CO clustering . We can therefore conclude that COs in axr1 tend to cluster together and that the observed shortage in bivalent formation is not due to a global decrease in meiotic recombination but rather mislocalisation of these events , resulting in a loss of the obligatory CO . Very little information is available on the mechanisms that control CO distribution during meiosis . Nevertheless , it has been known for a long time that COs are not randomly distributed among chromosomes , as in most organisms , adjacent COs display interference and therefore tend to be evenly spaced within chromosomes [13] . In addition , the phenomenon of the “obligatory CO” ( or CO assurance ) ensures the formation of at least one CO per bivalent , whatever the total number of CO precursors per cell . The relationship between these two phenomena is still under debate [23] , but recent modelling analyses suggested that the obligatory CO is a direct consequence of interference [65] . Numerous mutants with altered interference were described , but they nearly always also change CO rates , either because of increased MUS81-dependant COs [66]–[68] or because they are defective in the ZMM CO pathway ( see , for example , [48] , [69] ) . One possible exception is the Saccharomyces cerevisiae pch2 mutant , for which two independent studies showed that CO interference is alleviated without changes to meiotic recombination rates , at least on the smallest yeast chromosome ( III ) [70] , [71] . Nevertheless , the generalisation of this observation to the whole genome seems unlikely [71] . To our knowledge , axr1 is therefore the first mutant that specifically modifies the localisation of class I COs , changing interference among them and resulting in the loss of the obligatory CO , but without changing the global average number of CO events . Thus , AXR1 is a key regulator of meiotic recombination outcomes . From our data , we can exclude that the meiotic defects observed in axr1 are due to a major decrease in DSB formation or to a drastic mislocalisation of these events ( Figure S5 ) . We have also shown that axr1 meiotic defects are not associated with major chromosome axis defects ( Figure S6 ) , but instead with major perturbations in the polymerisation of the SC CE ( Figure 8 ) . The relationship between CO control and SC polymerisation is a long-standing question in the field of meiosis [72] . In yeast , SC polymerisation is not necessary for CO interference , as it occurs after CO patterns have been imposed [73] , [74] . In Caenorhabditis elegans , however , it was recently shown that the SC central region limits the formation of COs and imposes total interference [75] . This could also be the case in rice , where zep1 mutants ( ZEP1 being the rice CE ZIP1 homologue ) show an increase in chiasma formation at diakinesis [76] . This suggests that , in plants as in C . elegans , SC polymerisation could be necessary to limit CO formation . However , in Arabidopsis , ZYP1 appears to be required to prevent non-HR rather than acting on homologous CO formation [57] . To further complicate our understanding of the relationship between polymerisation of the SC CE and CO controls , in yeast , SC polymerisation requires the stabilisation of recombination intermediates by the ZMM proteins [77] , [78] . In Arabidopsis , SC polymerisation is also dependent on the formation of HR intermediates , as no synapsis is observed either in spo11 , dmc1 , or rad51 mutants where recombination is either not initiated or is blocked at the invasion step . However , Arabidopsis zmm mutants all display normal synapsis [20] , [48] , [69] , showing that SC polymerisation in these species depends on the formation of recombination intermediates , but not on their stabilisation by the ZMMs . The limited synapsis progression observed in axr1 mutants therefore suggests that recombination only proceeds far enough in a limited fraction of the genome where SC can polymerise and COs are formed . This appears to explain what is seen in Figure 8K , where mature HEI10 foci are concentrated on a few ZYP1 stretches . In that sense , aberrant SC formation illustrates that either recombination is blocked in a portion of the genome or that only a limited portion of the genome is competent to support recombination maturation , resulting in the loss of the obligatory CO . Nevertheless , the observation of nuclei where mature HEI10 foci are clustered on a single ZYP1 stretch , whereas the level of synapsis is high ( as illustrated on Figure 8H ) shows that the amount of ZYP1 polymerisation can be uncoupled from clustering of mature HEI10 foci . This suggests that CO clustering in axr1 is not only a consequence of limited synapsis progression . It is interesting to note that , within clusters of more than two class I CO foci , the distance among foci is not random , showing that they still display interference . During wild-type meiosis , no more than two adjacent foci could be scored , showing that these events are either less frequent or much more distant than in axr1 . In this latter case , our results suggest that interference strength is considerably modified in axr1 , resulting in CO clustering , at least in some areas of the genome . In Arabidopsis , interference strength is not uniform within chromosomes and increases toward the chromosome extremities [79] . This suggests that regional modification of interference parameters could be affected in axr1 . In the future , it will therefore be crucial to determine whether CO clustering in axr1 is region-specific or not . Regardless , the average number of final CO events is unchanged in axr1 , suggesting that ( i ) the total number of class I COs is precisely controlled , ( ii ) this control is still active in axr1 , and ( iii ) the mechanism underlying this CO homeostasis is independent of the obligatory CO mechanism . Neddylation stimulates several subclasses of cullin RING Ub ligases . We provide evidence that during meiotic recombination neddylation acts through cullin 4 activation to regulate the localisation of class I COs . Cullin 4 is a widely conserved cullin , involved in a large range of cellular and developmental controls , many of which are associated with genome integrity maintenance [80] . CRL4 complexes are composed of a CUL4 scaffold , a small RING domain containing RBX1 protein , a WD40-like repeat-containing adaptor DDB1 ( DNA-damage binding 1 ) , and a substrate receptor subunit called DWD ( DDB1-binding WD40 protein ) or DCAF ( DDB1- and CUL4-associated factor ) [80] . Evidence for CRL4 functions in genome integrity control come from multiple sources and concern mostly cell responses to UV damage and replication controls by regulating the accumulation of the replication licensing factor CDT1 . For example , DDB1- and/or CUL4A-depleted human cells accumulate DSBs and have an activated ATM-ATR cell cycle checkpoint [81] . The budding yeast cul8 mutants ( cullin 8 is thought to be the functional homologue of cullin 4 in S . cerevisae ) also accumulate DNA damage [82] . In fission yeast , mutation in Ddb1 increases the spontaneous mutation rate by more than 20-fold and prevents premeiotic S phase entry [83] . CRL4 activity is also required for the NER pathway by controlling the detection and processing of DNA lesions induced by UV in plants [41] , [84]–[86] , but also in mammals , as loss of CUL4A in mice leads to an increase in susceptibility to UV skin cancer [87] . Evidence for the role of CRL4 complexes in DSB repair was also provided in Drosophila , where DDB1 depletion promotes loss of heterozygosity in somatic cells [88] . In addition , CRL4s complexes may also be involved in HR regulation , as in fission yeast , ddb1 mutants are defective in HR probably by regulating the pool of available dNTPs [89] . Interestingly , we observed that hardly any COs are retained in double axr1zmm mutants , suggesting that the MUS81 recombination pathway may be shut down in axr1 . Because this pathway accounts for only a small proportion of all COs in Arabidopsis [14] , [22] , this disruption would not have a strong impact on meiosis . However , it could have a dramatic effect on somatic DNA repair , as the MUS81 pathway is one of the major pathways of somatic HR in eukaryotes [90] . It would therefore be interesting to study the involvement of AXR1 in somatic DNA recombination , above all considering that Dohmann and collaborators observed DSB accumulation in axr1 ( axr1-3 and axr1-12 ) somatic cells [91] . Considering the crucial role of CRL4s in genome maintenance and the activation of DNA repair pathways including HR , it is hardly surprising to find that it is involved in the regulation of meiotic HR in Arabidopsis . Considering the conservation of CRL4 functions across kingdoms , it is likely that the regulation of meiotic recombination by one ( or several ) CRL4 complex ( es ) will be also observed in other eukaryotes . Indeed , two converging studies in mice recently showed that cullin 4 is also required for meiosis also in mammals , as depletion of Cul4a ( one of the two mammalian Cul4 genes ) led to male infertility [42] , [43] . Whether this infertility is associated with early recombination [42] or later CO resolution defects [43] is still under debate . Nevertheless , the observation that MLH1 foci number is unchanged in cul4a but that a fraction of meiotic cells show pachytene bivalents without any MLH1 foci [43] is reminiscent of our data on axr1 . Therefore , we propose that neddylation is acting on one or several CRL4 complex ( es ) to regulate the localisation of class I COs not only in Arabidopsis but also in mammals . In A . thaliana , there are more than 85 substrate receptor DWD domain proteins that can assemble with DDB1A or DDB1B or directly with CUL4-RBX1 to form CRL4 complexes [92] , [93] . Further studies will be necessary to identify which of these is acting during meiosis .
Ws-4 lines ( including EGS344 , EIC174 , and EVM8 ) were obtained from the Versailles collection of Arabidopsis T-DNA transformants available at http://www-ijpb . versailles . inra . fr/en/sgap/equipes/variabilite/crg/[94] . Col-0 lines [including N877898 ( Sail_904E06 ) and N3076 = axr1-12] were obtained from the collection of T-DNA mutants from the Salk Institute Genomic Analysis Laboratory ( Columbia accession ) ( SIGnAL , http://signal . salk . edu/cgi-bin/tdnaexpress ) [95] and provided by NASC ( http://nasc . nott . ac . uk/ ) . Other mutant alleles used in this study are as follows: msh4Ws ( EXY25 ) [48] , msh5Col ( SALK_026553 ) [96]; hei10Ws ( EQO124 ) [48] , zip4Col ( SALK_068052 ) [69]; mer3Col ( mer3-2 , SALK_091560 ) [20] , mlh1Col ( SK_25975 ) [48] , mus81Col ( SALK_107515 ) [14] , rad51Col ( Gabi_134A01 ) [97] , mre11Col ( mre11-4 , Salk_067823 ) , cul1-6 Col [60] , axr6-2 Col ( N3818 ) [61] , axr6-3 Col ( eta1 ) [62] , cul3w Col [63] , and cul4-1 Col [64] . Plants were grown in a greenhouse ( photoperiod 16 h/d and 8 h/night; temperature 20°C day and night; humidity 70% ) . Screening for A . thaliana T-DNA ( A . tumefaciens transferred DNA ) insertions that provoke meiotic defects , we isolated three mutant lines: EGS344 , EIC174 , and EVM8 . They all segregated 3∶1 for reduced fertility , meiotic defects , and a bushy vegetative phenotype . Linkage analysis ( as described by Grelon et al . [98] ) showed that none of the mutations were linked with a T-DNA insertion . We therefore undertook a rough positional cloning of the three mutations as described by De Muyt et al . [99] . The most closely linked marker was chr1_02991901 for all three mutants ( based on 31 F2 mutant plants for EVM8 , 31 for EGS344 , and 31 for EIC174 ) . Fine gene mapping was then carried out as described by De Muyt et al . [99] using chromosome 1 microsatellite markers located between 1 , 243 , 352 and 1 , 573 , 000 bp . Among the predicted genes by TAIR10 SeqViewer server ( http://www . arabidopsis . org/ ) , we retained AXR1 ( At1G05180 ) as the best candidate , as axr1 mutants were previously shown to display the same vegetative developmental defects as EGS344 , EIC174 , and EVM8 [44] , [46] . Sequencing of At1g05180 in the three mutant lines showed that all three are disrupted in this open reading frame ( see below ) . We further analysed the axr1 reference allele ( axr1-12 ) and another insertion line ( Sail_904E06 ) available in the public databases ( http://signal . salk . edu/ ) . They all displayed the same meiotic phenotype as the previously isolated lines . Sequencing of At1g05180 in the EIC174 mutant line revealed a single nucleotide insertion in exon 6 ( position 1364 of the genomic sequence , corresponding to nt 688 in the cDNA ) , leading to a premature stop codon ( a 222 aa protein is produced instead of 540 aa in wild type ) . In the EGS344 mutant , a deletion of 898 bp ( from nucleotide 91 of the genomic sequence ) together with an insertion of Agrobacterium plasmid Ti DNA disrupts At1g05180 ( Figure S1 ) . In the EVM8 line , an in-frame deletion of 312 bp occurred between exons 3 and 4 , generating a 20 aa deleted protein . Details are shown in Figure S1 . In axr1-12 , corresponding to the N3076 line , a single C-T nucleotide substitution in position 1295 of the cDNA occurred , leading to a premature stop codon ( 415 aa instead of 540 ) , as described by Leyser et al . [44] . In N877898 , corresponding to the Sail_904E06 line , a T-DNA insertion occurred in intron 11 . Sequence references are as follows: Tair Accession 4010763662 for the genomic sequence , and Tair Accession 4010730885 for the cDNA sequence . For EGS344 and EVM8 , wild-type alleles were amplified with primers 05180-P1 ( ACCCTGATTGAAGAAAAGTCT ) and 05180-P2 ( CGGAGGTCGTCAAGAAAA ) ( 60°C , 30 PCR cycles , 1 , 200 bp ) . The EGS344 mutant allele was amplified with primers 05180-P1 and 05180-AgroP1 ( ACATCACAGCACCTCGATCCTGG ) ( 60°C , 30 PCR cycles , 300 bp ) . The EVM8 mutant allele was amplified with 05180-P1 and 05180-P2 ( 60°C , 30 PCR cycles , 980 bp ) For N877898 , the wild-type allele was amplified with primers N877898U and N877898L ( 60°C , 30 PCR cycles , 957 bp ) . The mutant allele was amplified with primers N877898L and Lb3SAIL ( TAGCATCTGAATTTCATAACCAATCTCGATACAC ) ( 60°C , 30 PCR cycles , 500 bp ) . For all other genotypes , the primer list and PCR amplification conditions are shown in Table S6 . The anti-ASY1 polyclonal antibody was described by Armstrong et al . [101] . It was used at a dilution of 1∶500 . The anti-ZYP1 polyclonal antibody was described by Higgins et al . [57] . It was used at a dilution of 1∶500 . The anti-DMC1 , anti-MLH1 , and anti-HEI10 antibodies were described by Chelysheva et al . in [69] , [49] , and [48] , respectively . These were used at a dilution of 1∶20 , 1∶200 , and 1∶200 , respectively . The anti-REC8 polyclonal antibody was described by Cromer et al . [102] and the anti-SCC3 by Chelysheva et al . [56] . These were used at a dilution of 1∶250 and 1∶500 , respectively . Comparison of the early stages of microsporogenesis and the development of PMCs was carried out as described in Grelon et al . [98] . Preparation of prophase stage spreads for immunocytology was performed using Carnoy's fixative and acetic acid chromosome spreads [59] , except for DMC1 detection and double HEI10/ZYP1 immunolabelling where lipsol spreading and paraformaldehyde fixation were used [58] . Chiasma numbers were assessed by analysing metaphase I spread PMC chromosomes stained with DAPI , as described by Sanchez-Moran et al . [103] . In brief , a rod bivalent stands for a single chiasma , whereas a ring bivalent as two ( one on each arm ) . Observations were made as described by Chelysheva et al . [48] . | During meiosis , two successive chromosomal divisions follow a single S phase , resulting in the formation of four haploid cells , each with half of the parental genetic material . This reduction in chromosome number occurs during the first meiotic division , when homologous chromosomes ( paternal and maternal ) are separated from each other . For this to happen , homologous chromosomes associate in structures called bivalents , where each chromosome is linked to its homologue by a point of contact known as chiasmata . These chiasmata reflect the formation of crossovers ( COs ) , one of the manifestations of the exchange of genetic material occurring during homologous recombination . CO number varies little at around two per chromosome pair , and they tend to be evenly spaced on chromosomes . Thus , CO number and distribution are very tightly controlled . However , the mechanisms underlying these controls are very poorly understood . In this study , we identified a regulatory pathway of meiotic recombination . We show that this pathway does not regulate the amount of recombination events per se , but instead controls their localisation , as when it is defective , CO events cluster together in a few regions of the genome , leading to bivalent shortage and progeny aneuploidy with incorrect numbers of chromosomes . This regulatory pathway is a posttranslational protein modification system called neddylation ( or rubylation in plants ) , known to be required for numerous cellular processes in eukaryotes . We identify an enzyme of the neddylation complex as a major regulator of meiotic recombination in Arabidopsis and show that this process may be also conserved in mammals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"meiosis",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"brassica",
"gene",
"function",
"plant",
"science",
"model",
"organisms",
"molecular",
"genetics",
"plants",
"arabidopsis",
"thaliana",
"research",
"and",
"analysis",
"methods",
"chromosome",
"biology",
"plant",
"genetics",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"cell",
"biology",
"organisms"
] | 2014 | Crossover Localisation Is Regulated by the Neddylation Posttranslational Regulatory Pathway |
In eutherian mammals , embryonic growth and survival is dependent on the formation of the placenta , an organ that facilitates the efficient exchange of oxygen , nutrients , and metabolic waste between the maternal and fetal blood supplies . Key to the placenta's function is the formation of its vascular labyrinth , a series of finely branched vessels whose molecular ontogeny remains largely undefined . In this report , we demonstrate that HOXA13 plays an essential role in labyrinth vessel formation . In the absence of HOXA13 function , placental endothelial cell morphology is altered , causing a loss in vessel wall integrity , edema of the embryonic blood vessels , and mid-gestational lethality . Microarray analysis of wild-type and mutant placentas revealed significant changes in endothelial gene expression profiles . Notably , pro-vascular genes , including Tie2 and Foxf1 , exhibited reduced expression in the mutant endothelia , which also exhibited elevated expression of genes normally expressed in lymphatic or sinusoidal endothelia . ChIP analysis of HOXA13–DNA complexes in the placenta confirmed that HOXA13 binds the Tie2 and Foxf1 promoters in vivo . In vitro , HOXA13 binds sequences present in the Tie2 and Foxf1 promoters with high affinity ( Kd = 27–42 nM ) and HOXA13 can use these bound promoter regions to direct gene expression . Taken together , these findings demonstrate that HOXA13 directly regulates Tie2 and Foxf1 in the placental labyrinth endothelia , providing a functional explanation for the mid-gestational lethality exhibited by Hoxa13 mutant embryos as well as a novel transcriptional program necessary for the specification of the labyrinth vascular endothelia .
For placental mammals , fetal development is contained in an intrauterine environment where the efficient exchange of oxygen , nutrients , and metabolic waste between the maternal and fetal blood supplies is facilitated by the placenta . Central to the placenta's function is its vascular labyrinth , a juxtaposed series of finely-branched blood vessels and trophoblasts that regulate nutrient and waste exchange while maintaining the separation of the maternal and fetal blood supplies [1] . After implantation , labyrinth vascularization proceeds from the allantois , where angiogenic and vasculogenic processes promote the formation of a dense , highly arborized vascular bed [2]–[7] . The formation of the labyrinth vascular bed requires many of the same signals controlling embryonic vascular development including: VEGF and its associated receptors FLT1 , FLK1 , and NEUROPILIN-1 , as well as ANG-1 and ANG-2 and its receptor TIE-2 [8]–[16] . Interestingly , while loss of function studies clearly demonstrate that transcription factors such as: TBX4 , CDX2 , CDX4 , HAND1 , DLX3 , FOXF1 , and CITED2 are required for placental development , the target genes regulated by these proteins in the developing placenta are largely undefined [3] , [17]–[26] . In this report , we describe a novel role for HOXA13 in the developing placenta and identify both direct and indirect targets of HOXA13 functioning in the placental labyrinth endothelia . In the absence of HOXA13 function , labyrinth endothelial cell morphology , vessel branching , and vessel integrity are compromised , a consequence we attribute to a loss in the regulation of several essential pro-vascular genes . Chromatin immunoprecipitation of the HOXA13-DNA complexes confirmed that HOXA13 directly associates with the Tie2 and Foxf1 promoters in vivo in the developing placenta . Quantitation of HOXA13's affinity for these promoter regions confirmed that HOXA13 binds these regions with high affinity and can utilize these bound DNA sequences to facilitate gene expression in vitro . Together these findings reveal a novel temporal and spatial domain for HOXA13 function in the developing embryo and identify a key transcriptional hierarchy necessary for the development of the placental vascular labyrinth .
Among the 39 murine Hox genes , only mutations in Hoxa13 cause mid-gestational lethality from embryonic day ( E ) 11–15 . 5 [27]–[29] . Initially , this phenotype was attributed to premature stenosis of the umbilical arteries [29] . Extensive analysis of the umbilical artery ( UA ) defect in Hoxa13 homozygous mutants revealed that only one of the two UAs exhibited complete stenosis from E11 . 5–15 . 5 ( Figure 1A and 1B ) . This finding prompted the hypothesis that an additional defect must be contributing to the mid-gestational lethality . Because malformations of the heart and placenta are the most commonly cited reasons for mid-gestational lethality , we first examined whether Hoxa13 is expressed in the developing heart or placenta [30]–[34] . No HOXA13 expression was detected in the cardiac crescent; however , the earliest component of the placenta , the allantoic bud mesoderm , strongly expressed HOXA13 from E7 . 75 and maintained expression in the maturing allantois and developing labyrinth microvessels at E8 . 5 and E9 . 5 respectively ( Figures 1C–1G and 2 ) . At E10 . 5 , HOXA13 expression was readily detected in the developing placental labyrinth ( Figures 1H and 2 ) whereas the chorionic ectoderm exhibited little or no HOXA13 expression from E9 . 5 to E10 . 5 ( Figures 1E–1H , 2 ) . To note , the timing and occurrence of chorioallantoic fusion was unaffected by the loss of HOXA13 function ( data not shown ) . To determine the identity of the cells expressing HOXA13 in developing labyrinth , we examined whether these cells co-express the endothelial marker PECAM-1 ( Figure 3 ) [35]–[37] . Characterization of HOXA13 and PECAM-1 expression confirmed that only the cells expressing PECAM-1 ( cell surface ) also express HOXA13 ( nucleus ) , suggesting that HOXA13 is functioning in the labyrinth vascular endothelial cells ( EC ) ( Figure 3A–3F ) . HOXA13 expression was not detected in the placental trophoblasts ( data not shown ) . Interestingly , the elongated morphology normally attributed to the labyrinth vascular endothelia was also affected in homozygous mutants , which appeared rounded compared to controls in the E12 . 5 labyrinths ( Figure 3C and 3D ) . To determine the onset of the EC phenotype , we characterized the labyrinth vessels at E10 . 5 ( Figure 3E and 3F ) . At E10 . 5 , the EC in both the heterozygous control and homozygous mutant labyrinths exhibited only the rounded EC morphology ( Figure 3E and 3F ) . This result suggests that the vascular specification of the EC , as indicated by their elongated morphology , occurs between E10 . 5 and E12 . 5 and denotes when the loss of HOXA13 function phenotype first manifests in the developing labyrinth EC ( Figure 3C–3F ) . Close examination of the affected EC using transmission electron microscopy confirmed the timing of the onset of this phenotype as wild type EC exhibited lengthening of the cell body as early at E11 . 5 whereas homozygous mutant EC exhibited shortened cell bodies that lacked uniform contact with the underlying vessel walls ( Figure 4A–4D ) . Finally , consistent with the reduction in the EC cell body was the loss of vessel wall integrity in the E11 . 5 and E13 . 5 mutant vessels , resulting in extracellular edema between mutant labyrinth vessels and the underlying syncytiotrophoblasts , while edema was not detected in the wild type labyrinths ( Figure 4A–4D ) . No gross morphological defects were observed in the basement membranes between the EC and the syncytiotrophoblast ( Figure 4E and 4F ) . Quantitation of the affected labyrinth EC in the homozygous mutants revealed that nearly 50 percent exhibited an intermediate or grossly affected morphology compared to only 12 percent for age-matched controls ( Figure 4G ) . Finally , the affected EC did not appear apoptotic as the nuclei lacked a pyknotic phenotype as well as TUNEL positive staining ( Figure 4 and unpublished data ) . As the umbilical vessels cross the chorionic plate , they exhibit non-sprouting angiogenesis parallel to the chorionic plate to produce the chorionic plate vessels , followed by additional EC migration and branching angiogenesis into the labyrinth to create a complex vascular tree ( see Figure 2 ) . In the murine placental labyrinth , these vascular branches are interconnected , signifying that intussusceptive angiogenesis and vessel fusion also play a role in their angiogenic remodeling [38] , [39] . Recognizing that EC mediate many of the angiogenic processes necessary for vessel remodeling and branching , we hypothesized that defects in the mutant labyrinth EC would affect vessel branching , which would compromise the capacity of this structure to sustain embryonic survival ( Figures 2–4 ) [40]–[47] . To best visualize the branched vasculature within the labyrinth , we utilized whole tissue immunohistochemistry using the PECAM-1 antibody and hemisected placentas . Indeed , while the initial vascular invasion of the chorionic plate at E10 . 5 appears normal in Hoxa13 homozygous mutants , there is a qualitative reduction in the level of PECAM-1 staining of the labyrinth vasculature as early as E11 . 5 ( Figure S1 ) . Quantitation of the vascular branches in the mutant and control placental labyrinths at E13 . 5 confirmed that the number of branches is reduced in the Hoxa13 homozygous mutants which exhibited 33 ( ±6 . 3 ) branches per 400 mm2 grid analyzed , whereas wild type and heterozygous mutant controls contained 56 ( ±12 . 7 ) and 55 ( ±9 . 7 ) vessel branches respectively ( Figure 5 ) . Qualitatively , the reduction in labyrinth vascularity in homozygous mutants was persistent throughout labyrinth development which is complete by E14 . 5 , suggesting that decreased vessel branching is phenotypic of the loss of HOXA13 function rather than a delay in the labyrinth maturation ( Figures 3 , 5 , and S1 ) [48] . Next , because changes in vessel branching can also affect the overall size of the labyrinth , we examined whether the labyrinth region was smaller in Hoxa13 mutant placentas . Analysis of labyrinth sections taken from E13 . 5 wild type and Hoxa13 homozygous mutants confirmed that the mutant labyrinths were nearly half the thickness as their age-matched controls ( 800±200 µm vs . 1600±250 µm ) ( Figure 6 ) . Since defects in endothelial cell migration can also contribute to perturbations in vessel branching , we examined whether Hoxa13 mutant endothelia were competent to migrate and participate in de novo angiogenesis ( Figure 7 ) . A comparison of mutant and heterozygous control cultured arterial sections revealed comparable levels of neo-vessel production after five days of growth ( Figure 7A and 7B and 7D and 7E ) . Characterization of the cells contributing to the neo-vessels revealed strong HOXA13 expression in the PECAM-1 positive endothelia in both heterozygous and homozygous mutant ( Figure 7C and 7F ) . This result suggests that endothelial cell migration is not affected by the loss of HOXA13 function in the in vitro angiogenesis assay . Taken together , these results suggest that placental insufficiency caused by decreased labyrinth vascularity and size may be causing the mid-gestational lethality associated with the loss of HOXA13 function . To test this hypothesis we examined the developing heart , an organ that does not express Hoxa13 but is severely affected by placental insufficiency [28] , [29] , [49]–[53] . Analysis of E14 . 5 hearts from heterozygous control and homozygous mutants revealed a substantial thinning of the right and left ventricular walls in homozygous mutants ( Figure 8 ) . Quantitation of the left ventricular wall thicknesses confirmed nearly a 43% reduction in wall thicknesses in homozygous mutants which contained an average of 3 . 6 ( ±1 . 1 ) cells per wall section measurement versus 6 . 3 ( ±0 . 8 ) cells in the comparable sections of age-matched heterozygous controls ( Figure 8C and 4D ) . To identify candidate HOXA13 target genes which are functioning in the placental labyrinth , we performed microarray analysis on E13 . 5 labyrinth tissues . Twelve microarray hybridizations ( 6 mutant; 6 wild type ) were performed using independent isolates of placental labyrinth total RNA . Statistical analysis of the gene expression signals detected by the microarray probe sets identified significant reductions in pro-vascular gene expression ( FDR≤0 . 05 ) including: CD36 , necdin , Enpp2 , Adrb1 , Tie2 , Foxf1 , Neuropilin-1 , Magel-2 , and Caveolin-1 ( Table 1 ) . Interestingly , the homozygous mutant placentas also exhibited significant over-expression of several genes normally expressed in sinusoidal and lymphatic endothelia including: Lyve-1 , Igfbp3 , Selenbp1 , Bmp1 , and Ednrb ( Table 2 ) [54]–[58] . Finally , genes expressed in specific trophoblast lineages such as: gcm1 ( chorionic trophoblast ) , esx1 ( labyrinth trophoblasts ) , tpbp ( spongiotrophoblast ) , Id1 ( chorionic trophoblast ) , and Id2 ( chorionic trophoblast ) exhibited no significant changes in expression between control and Hoxa13 mutant placentas ( FDR≤0 . 05 ) ( Figure 6 and Table 3 ) [59]–[63] . Next , to validate that the genes detected by the microarray analysis were mis-expressed in the affected EC , we performed quantitative real-time PCR ( qRT-PCR ) using total RNA derived from affinity purified vascular labyrinth EC . In all cases , the mis-expression trend determined by the microarray analysis ( increased or decreased in mutant placentas ) was conserved ( Table 1 ) . Moreover , the enrichment of the EC also caused an increase in the detected fold-change differences between wild type and homozygous mutant EC , a finding consistent with an endothelial-specific expression pattern or function for the affected genes . Immunohistochemical analysis of TIE2 , LYVE-1 , NEUROPILIN-1 , and ENPP2 confirmed the altered EC-specific expression levels detected by microarray and qRTPCR ( Figures 9 and S1 , Table 1 , and unpublished data ) . In particular , the pro-vascular receptor tyrosine kinase , TIE2 , was noticeably reduced in the mutant labyrinth EC which also express HOXA13 at E10 . 5–13 . 5 ( Figures 9A–9D and S1 ) . Because Tie2 and Foxf1 are strongly expressed in the placental labyrinth and mice lacking TIE2 or FOXF1 exhibit vascular defects most similar to those present in Hoxa13 mutant labyrinths , we hypothesized that HOXA13 directly regulates Tie2 or Foxf1 to facilitate labyrinth vascular development [8] , [39] , [64]–[70] . Testing this hypothesis , we first examined whether HOXA13 binds the promoters for Tie2 or Foxf1 in vivo using a HOXA13 antibody to immunoprecipitate HOXA13-DNA complexes present in labyrinth chromatin ( Figure 10A ) . Previous characterization of the HOXA13 antibody confirmed that it can bind both wild type and mutant HOXA13 proteins and facilitate chromatin immunoprecipitation ( ChIP ) of HOXA13-bound gene regulatory elements [71]–[73] . PCR analysis of the first 3000 base-pairs ( bp ) ( −3000 to +1 ) of the Tie2 and Foxf1 promoters revealed a single region in each locus bound by HOXA13 in wild type ( not shown ) and heterozygous mutant placental labyrinths ( Figure 10A and 10B ) . In contrast , the homozygous mutant HOXA13 protein , which lacks its DNA binding domain , failed to associate with the same promoter regions ( Figure 10A ) . Parallel ChIP assays did not detect HOXA13 association with the promoters of other pro-vascular genes mis-expressed in the mutant labyrinth including CD36 , Caveolin-1 , and Neuropilin-1 ( data not shown ) . Sequence analysis of the immunoprecipitated Tie2 and Foxf1 promoter regions confirmed the presence of several of the recently identified HOXA13 binding sites ( Figure 10B ) [71] . Next , using an electrophoretic mobility shift assay ( EMSA ) , the HOXA13 DNA binding domain was confirmed to bind the promoter regions detected by the ChIP assay in a concentration-dependent manner ( Figure 10C ) . Quantitation of HOXA13's affinity for the Tie2 and Foxf1 ChIP-positive regions using fluorescence polarization anisotropy revealed high affinity for the binding sites present in Tie2 ( Kd = 27±1 . 4 nM and 22 nM±1 . 6 nM ) and Foxf1 ( Kd = 48±4 nM ) compared to a control sequence lacking the HOXA13 binding site ( Kd = 250±22 nM ) ( Figure 10D and 10E ) . Next , the capacity of HOXA13 to regulate gene expression through the 140 base-pair Tie2 and 121 base-pair Foxf1 ChIP-positive DNA fragments was examined ( Figure 10F ) . The pGL3-Basic vector was selected for this analysis based on previous studies that confirm its capacity to assess promoter/enhancer activity in vitro , including previous characterizations of HOXA13's capacity to regulate transcription from minimal promoter elements [72] , [74]–[78] . In the absence of the Tie2 or Foxf1 DNA elements , the empty pGL3-basic luciferase plasmid exhibited only a minor increase in luciferase expression when co-transfected with a Hoxa13 expression plasmid ( Figure 10F ) . Similarly , the same luciferase vector containing either the Tie2 or the Foxf1 ChIP-positive regions also exhibited minimal luciferase expression in the absence of HOXA13 ( Figure 10F ) . Co-transfection with a Hoxa13 expression vector stimulated luciferase expression from these minimal promoter elements resulting in low but significant increases in normalized luciferase expression: 3 . 7 fold for Tie2 and 3 . 2 fold for Foxf1 ( Figure 10F ) .
While the timing of Hoxa13 expression at E7 . 75 was surprising for a 5′ Hox gene , its localization to the allantoic bud is consistent with the later functions of HOXA13 in the gut , where the contributing posterior allantoic bud mesoderm receives patterning instructions from the 5′ HOX proteins [5] , [28] , [29] , [79]–[83] . The vascular-specific expression of Hoxa13 appears to be restricted to the placental labyrinth , genital tubercle , and umbilical arteries; structures whose ontogeny can be linked to tissues within the posterior embryo [28] , [29] , [73] , [84] . Here , the mesenchymal functions of HOXA13 during genitourinary development appear co-opted to facilitate vascular specification of the labyrinth endothelia [28] , [29] , [73] , [80] , [84] . HOXC13 also exhibits co-opted function in the hair follicle , suggesting that the development of specialized structures may utilize the co-opted functions of the group 13 HOX proteins [85] . While other Hox proteins such as HOXD3 , HOXB3 , HOXA5 , and HOXD10 have been shown to regulate angiogenesis , our analysis of HOXA13's function in the vascular endothelia represents , to our knowledge , the only group 13 HOX protein functioning in this capacity [86]–[89] . In the absence of HOXA13 function , labyrinth endothelial cell integrity and vessel branching are compromised , resulting in a placental labyrinth incapable of sustaining fetal growth beyond mid-gestation [27]–[29] . Placental insufficiency also causes secondary cardiac defects [49]–[53] . The absence of Hoxa13 expression in the early cardiac field as well as its absence in the affected heart tissues supports the conclusion that thinning of the ventricular walls in Hoxa13 mutants is a secondary defect caused by placental insufficiency . The over-expression of non-capillary endothelial genes in Hoxa13 homozygous mutant labyrinths raises an intriguing possibility that the vascular endothelia are trans-differentiating towards a sinusoidal or lymphatic phenotype . Indeed , the genes exhibiting the highest degree of over-expression in Hoxa13 mutant placentas are either elevated during lymphatic re-specification or have discrete functions in sinusoidal and/or lymphatic endothelia including: Lyve-1 , Igfbp3 , Selenbp1 , and Ednrb ( Table 2 ) [54] , [56]–[58] , [90]–[92] . Moreover , the rounded phenotype of Hoxa13 mutant EC is also more similar to sinusoidal EC , although our analysis of the mutant endothelia by TEM did not reveal substantial fenestrations usually present in sinusoidal endothelia [54] . Perturbations in the expression of the transcription factor , PROX1 , also promote Lyve-1 over-expression in vascular endothelia , suggesting some degree of plasticity in endothelial specification [91] , [92] . Transformations of sinusoidal endothelia towards a capillary phenotype have also been induced by selenium , thus the over-expression of the selenium binding protein in the mutant endothelia ( Selenbp1 ) , could alter endothelial specification by limiting selenium bioavailability[54] . Alternatively , it is possible that the over-expression of LYVE1 in Hoxa13 homozygous mutant EC could affect the uptake and degradation of matrix hyaluronan , impacting endothelial cell adhesion [90] , [93]–[97] . While mice lacking LYVE1 do not exhibit defects in the placental labyrinth , its over-expression in Hoxa13 mutant EC may influence how these cells interact with the underlying basement membrane resulting in the rounded appearance and perturbations in EC migration [92] . Arguing against this possibility is the result that Hoxa13 homozygous mutant EC are competent to migrate during angiogenesis , suggesting that LYVE1 over-expression in the mutant EC is a consequence of the loss of endothelial specification . Finally , the over-expression of the metalloprotease , Bmp1 , could also contribute to the loss in labyrinth vascularization as BMP1 cleaves placental PROLACTIN to produce a potent angiogenic inhibitor [55] . Previous studies of labyrinth development suggest that signals from the trophoblast lineages participate in labyrinth formation and vascularization [5] , [13] , [22] , [50] , [51] , [53] , [62] , [63] , [98]–[102] . The normal expression of trophoblast genes such as Gcm1 , Id1 , Id2 ( chorionic ) ; Tpbp ( spongiotrophoblast ) ; and Esx1 ( labyrinth trophoblast ) in Hoxa13 mutant placentas suggests that trophoblast tissues are not affected by the loss of HOXA13 function . Instead , HOXA13 appears to be functioning in the vascular endothelia where it regulates pro-vascular genes necessary for labyrinth vascular development . Similar labyrinth defects are also seen in mice lacking WNT2 and HGF which appear to be predominantly expressed in the allantoic mesoderm and its derivatives [103] , [104] . Finally , the normal expression of trophoblast-specific genes also suggests that tetraploid chimeras consisting of a wild type tetraploid embryo and Hoxa13 homozygous mutant embryonic stem cells would not be effective in rescuing the labyrinth defects [1] , [105] . In Drosophila , HOX proteins such as UBX regulate the formation of specific structures by controlling genes at multiple levels of a developmental cascade [106]–[108] . Our analysis of HOXA13 function during limb , genitourinary , and placental development suggests a similar mode of gene regulation where the combinatorial functions of direct and indirect target genes are coordinated to facilitate the formation of specific tissues and structures [28] , [71]–[73] , [80] . Evidence for this coordination is seen in the labyrinth endothelia where genes necessary for cell adhesion and vascular branching are concomitantly affected by the loss of HOXA13 function including: Neuropilin-1 , Enpp2 , Lyve1 , Caveolin-1 , Foxf1 , and Tie2 , resulting in hypomorphic levels of the provascular factors necessary for labyrinth vascular development . The binding of HOXA13 to the Tie2 and Foxf1 gene regulatory elements in vivo and the reduction of Tie2 and Foxf1 expression in Hoxa13 mutant labyrinths suggest that these pro-vascular genes are direct targets of HOXA13 in the labyrinth endothelia . Indeed , the loss of function phenotypes associated with Tie2 and Foxf1 are consistent with the endothelial defects in Hoxa13 mutant labyrinths . Mice lacking TIE2 exhibit lethality by E10 . 5 due to a severe lack of angiogenic branching and remodeling of the primary vascular network [39] , [70] , [109] . Foxf1 is also essential for placental development [21] , [110] . Mice lacking Foxf1 exhibit mid-gestational lethality resulting from defects in the patterning and vascularization of extra-embryonic tissues [20] . Interestingly , Foxf1 haploinsufficiency also affects vascular integrity , causing hemorrhaging in the lung and foregut , suggesting that reduced Foxf1 in the Hoxa13 mutant labyrinth may be sufficient to affect EC integrity and vessel branching [66] . Similarly , mice lacking ENPP2 also exhibit defects in allantois , yolk sac , and embryonic vessel formation , suggesting that this previously identified HOXA13 target gene may function to mediate labyrinth vascularization . Because the Neuropilin-1 , Caveolin-1 , and CD36 promoter regions were not detected by the HOXA13 ChIP assay , we are presently classifying these genes as indirect targets of HOXA13 . Developmentally , perturbations in Neuropilin-1 , Caveolin-1 , and CD36 expression are also consistent with the Hoxa13 mutant labyrinth defects . NEUROPILIN-1 , a receptor for VEGF , PlGF-2 , and VEGF-B , is essential for endothelial migration and proliferation , and mice lacking this protein die at mid-gestation from vascular and heart defects [14] , [15] , [111]–[113] . The caveolae-associated molecules CD36 and CAVEOLIN-1 also modulate cell mobility and permeability , angiogenesis , and intracellular trafficking , and caveolin-1 knockout mice exhibit decreased vascular tone and decreased angiogenic responses to exogenous stimuli [114]–[117] . Factors regulating sprouting and non-sprouting angiogenesis are required for labyrinth development . Initially , non-sprouting angiogenesis regulated by factors such as CYR61 are required to produce the network of chorionic plate vessels ( Figure 2 ) [1] , [118] . While Hoxa13 and CYR61 homozygous mutants both exhibit reductions in labyrinth vascularization , the vascularization defects are more severe in CYR61 mutants , a consequence of the earlier developmental function of CYR61 which regulates the formation of chorionic plate vessels whose subsequent branches form the primary vessels of the vascular labyrinth ( Figure 2 ) [118] . Similarly , defects in labyrinth vascularization have also been attributed to perturbations in NOTCH/DELTA signaling [41] , [119]–[121] . In particular the loss of Notch1 function or its target genes , Hey1 and Hey2 , cause a complete loss of angiogenic sprouting necessary to form the vascular labyrinth [41] , [119]–[121] . Similar to HOXA13 , the NOTCH ligand , DLL4 , appears to be expressed in the umbilical arteries , as well as the developing chorionic plate vessels and invading labyrinth branches [119] . In these tissues , haploinsufficiency of DLL4 was sufficient to cause gestational lethality , a consequence of the regression of the chorionic plate vessels which integrate the developing umbilical vessels to the placental labyrinth [119] . While this phenotype is substantially different from the labyrinth defects in Hoxa13 homozygous mutants the similarities in the expression domains of Hoxa13 , Notch1 , and Dll4 raise the possibility that HOXA13 and activated Notch receptors function through common factors to promote the angiogenic processes necessary for labyrinth development . Factors such as NEUROPILIN-1 may be a common link between HOXA13 and NOTCH/DELTA signaling as this essential pro-angiogenic molecule is down-regulated in Hoxa13 ( this work ) , Notch1 , and Hey1/Hey2 mutants [41] . The formation of a functional placenta is one of the most critical steps in human and mouse intrauterine development . In this study we have identified a role for HOXA13 in the formation of this vital organ . These findings suggest that HOXA13 regulates a series of genes in the vascular endothelia that are necessary for adhesion and vessel branching , providing a functional explanation for the mid-gestational lethality exhibited by Hoxa13 mutant mice . More importantly , these findings provide a novel genetic pathway to consider when characterizing pathologies of the placenta or placental evolutionary ontogeny .
All animal care and handling was done following an approved institutional animal protocol . Mice used in this study were from the Hoxa13-GFP line , previously described [28] . Hoxa13 mutant embryos were derived from heterozygous intercrosses as described [28] , [80] . Timed matings were used to establish embryonic gestational age and are depicted in embryonic days ( E ) where E0 . 5 represents the first day of vaginal plug detection . The Hoxa13 mutant allele encodes a fusion protein where the last 34 amino acids of HOXA13 , which encodes the DNA binding domain , have been replaced with an EGFP reporter as described [28] , [73] . The nuclear localization , protein turnover , and tissue-specific expression of the HOXA13-GFP protein were similar to the wild type protein [28] . The HOXA13-GFP fusion protein produces a robust fluorescent signal . Analyses examining endogenous HOXA13-GFP localization are labeled as HOXA13-GFP . Because E11 . 5–13 . 5 placentas exhibit high autofluorescence in the GFP emission spectra it was necessary to use a GFP antibody ( AB3080 , Chemicon ) to visualize the co-localization of the HOXA13-GFP with candidate target gene proteins . Characterizations of HOXA13-GFP expression using the GFP antibody are denoted as α-GFP in the figures . A Texas-red labeled secondary antibody ( Jackson Immunological ) was used to detect the localization of the GFP antibody . Placentas were dissected in cold 1× PBS and fixed 3 hours to overnight in 4% paraformaldehyde/PBS at 4°C rocking . For frozen sections , placentas were treated with a 10–30% sucrose/PBS gradient , embedded in OCT ( Tissue-Tek , Inc ) , and stored at −80°C . Frozen OCT-embedded placentas were sectioned at 17–20 µm using a Leitz Kryostat 1740 and mounted on Superfrost plus slides ( Fisher ) . Immunohistochemistry ( IHC ) and in situ hybridization ( ISH ) experiments were carried out as previously described [73] , [80] . The Tpbp riboprobe plasmid was kindly provided by Dr . James Cross ( University of Calgary ) . For whole placenta IHC , placentas were bisected with a double-edged razor blade and fixed in 4% paraformaldehyde overnight at 4°C . After fixation the bisected placentas were dehydrated with a 4∶1∶1 methanol-DMSO-peroxide solution and incubated in primary and secondary antibodies overnight followed by extensive PBST washes . The following antibodies and dilutions were used: PECAM-1 for section ( #550274 , BD Pharmingen , 1∶200 ) , PECAM-1 for whole placenta IHC ( MEC13 . 3 , #553369 BD Pharmingen , 1∶200 ) , GFP ( AB3080 , Chemicon , 1∶100 ) , Lyve1 ( ab14917 , Abcam , 1∶200 ) , Tie2 ( #MAB1148 , Chemicon , 1∶200 ) , ENPP2 ( Cosmo Bio , 1∶100 ) . For co-localization studies , CY5- or Texas Red-labeled secondary antibodies were used as described [73] , [80] . To visualize the co-localization of PECAM-1 and LYVE1 in the vascular endothelia , the red CY5 signal detecting the distribution of the PECAM-1 antibody was pseudo-colored green using the Laser Sharp 2000 software ( BioRad ) . E11 . 5 placentas were bisected and immediately immersed in 1 . 5% glutaraldehyde/1 . 5% paraformaldehyde with 0 . 05% tannic acid and 5 . 0% sucrose in DMEM media for 2 hours on ice , rinsed in several changes of DMEM , then post-fixed in 1 . 0% OsO4 in DMEM for an additional 90 minutes on ice . Fixed tissues were rinsed in several changes of DMEM over 15 minutes , then dehydrated in a graded ethanol series to 100% , rinsed in propylene oxide , and finally infiltrated and embedded in Spurr's epoxy . Ultrathin sections were cut at 80 nm , contrasted with uranyl acetate and lead citrate , and then examined using a Philips 410 TEM operated at 80 KV . E13 . 5 placentas were fixed by perfusion using 1 . 5% glutaraldehyde/1 . 5% paraformaldehyde with 0 . 05% tannic acid in DMEM , then bisected and immersed in the same fixative for an additional 60 minutes . Post-fixation , dehydration and embedding was identical to the E11 . 5 embryos described above . E13 . 5 placentas were sectioned at 1 µm , mounted on slides , and counterstained with toluidine blue with basic fushin and the sections containing the labyrinth regions were photographed . For each placental sample , seven to ten photographs representing nearly the complete vascular labyrinth region were taken . Within each photo , fetal labyrinth lumens were identified , and all ECs with a visible nucleus were counted and their morphology scored as either normal ( smooth elongated EC layer and lumen ) , intermediate ( some cell rounding or irregular shape ) , or abnormal ( cell detachment , edema , diminished cell body ) . Three wild type placentas ( 25 photos; 1510 endothelial cells ) , five Hoxa13 heterozygous placentas ( 38 photos; 1938 ECs ) , and five Hoxa13 homozygous mutant placentas ( 39 photos; 1301 ECs ) were analyzed . The average percentage of each cell morphology was calculated and plotted with their standard deviations using Excel ( Microsoft ) . Placental labyrinth thickness was measured in 8 independent E12 . 5 Hoxa13 heterozygous and homozygous mutant placentas . The placentas were fixed , OCT embedded , and 20 µm sections were hybridized with the Tpbp or Hoxa13 ( not shown ) riboprobes to identify the labyrinth and overlying spongiotrophoblast regions . Digital photographs were taken from two separate placental sections that represent middle regions of each placenta , using a Leica DML light microscope with a 10× objective and a Q-Imaging camera . Using the NIH Image J software , labyrinth thicknesses ( in micrometers ) were measured from the labyrinth-spongiotrophoblast border to the chorionic plate at five independent points along the placental section . For each point , three to six independent measurements were taken . The average thickness and standard deviation was calculated using Excel ( Microsoft ) and plotted using Sigmaplot 9 . 0 ( Systat ) . Labyrinth vessel branching was quantitated using a modified grid analysis of the labeled labyrinth vessels as described [122] . Hemisected E13 . 5 placentas were fixed in 4 percent paraformaldehyde and processed for whole mount immunohistochemistry using the PECAM-1 antibody as described earlier . The hemisected placentas were photographed using a Leica MZ-12 . 5 stereoscope fitted with a Nikon Coolpix 990 digital camera at 4× magnification . Placenta photographs were resized to 4×2 . 5 inches using Adobe Photoshop CS3 . Three hemisected placentas were characterized for each genotype . A 400 mm2 ( 20 mm×20 mm ) grid was placed at six independent locations in the labyrinth region of the placental photographs and the PECAM-1 labeled vessels branches present in the grid were counted . The average number of vessel branches and their standard deviations were calculated using Excel ( Microsoft ) and plotted using Sigmaplot 9 . 0 ( Systat ) . Primary umbilical arteries entering the placenta ( E13 . 5 embryos ) were dissected free using sterilized tungsten needles . The dissected vessels were cut into 1–3 mm sections using sterilized microvasculature scissors ( Fine Science Tools ) and embedded into 1 percent type I collagen ( BD Biosciences ) containing 2 . 3 mg/ml sodium bicarbonate , and 2× enrichment of EGM-2 MV base media ( Lonza , Walkersville , MD ) as described [123] . Four-well chambered microslides ( NUNC ) were used for the embedding and culture of the vessel sections . After 10 minutes of incubation at 37 degrees Celsius , the chambered slides were filled with complete 1× EGM-2MV media supplemented with 5 percent fetal bovine serum , and the Single-Quot® growth factor/antibiotic cocktail containing hydrocortisone , hFGF-B , VEGF , IGF , ascorbic acid , EGF , gentamicin , and amphotericin-B as described by the manufacturer ( Lonza , Walkersville , MD ) . After five days , the cultured umbilical arteries were fixed overnight in 4 percent paraformaldehyde/PBS at 4 degrees Celsius . After fixation the cultured vessels were rinsed with 1× PBS containing 1% Triton X 100 , and characterized for HOXA13-GFP and PECAM-1 expression as described [73] , [80] . Three separate E13 . 5 Hoxa13 heterozygous control and homozygous mutant embryos were fixed , paraffin-embedded , sectioned , and stained with hematoxylin and eosin using standard histological techniques as described [73] . The heart regions from the sectioned embryos were photographed using a Leica DML compound microscope and a Q-Imaging digital camera . The cellular thickness of each ventricular wall was determined by counting the number of hematoxylin-stained nuclei present in 6–10 perpendicular lines drawn from outer edge of the left ventricular wall to the trabeculae as shown in Figure 8 . The average cell number and standard deviation was calculated using Excel ( Microsoft ) . E13 . 5 placentas were dissected and the isolated umbilical vessels , embryonic labyrinth , and spongiotrophoblasts tissues were used as the source for the embryonic placental RNA . Tissues were dissected in RNAlater ( Ambion ) , flash-frozen in liquid nitrogen , and stored at −80°C . The RNA STAT-60 ( CS-110 , Tel-Test , Inc . ) and RNeasy Micro Kit ( QIAGEN ) systems were used for RNA extraction , following the manufacturer's protocol . RNA quality was assessed using an Agilent Bioanalzyer at the OHSU Affymetrix Microarray Core facility ( AMC ) , UV spectroscopy , and agarose gel electrophoresis . Three RNA samples of like genotype were pooled for each microarray analysis . Twelve independent microarray hybridizations ( 6 Hoxa13 homozygous mutant , 6 wild type ) were performed using the MOE430A and B microarrays ( Affymetrix ) . MAS 5 . 0 software ( Affymetrix ) was used by the AMC to collect and normalize the array data . Statistical analysis of the microarray data sets was performed by the Biostatistics and Bioinformatics Shared Resources Core Facility at OHSU . A two-factor analysis of variance ( ANOVA ) was used to determine the false discovery rate and to compare transcript signal intensities between wild type and homozygous mutant placental tissues . E13 . 5 placental EC were isolated by dissecting fresh placentas as in the microarray method . Two to three placentas of the same Hoxa13 genotype were combined . Samples were treated with 0 . 2% Collagenase Type IV ( #17104-019 , Gibco ) at 37°C for 30 minutes , with occasional shaking . Tissue was then transferred to a Netwells Dish ( Costar ) and treated with digestion media ( 0 . 1% Trypsin/EDTA , 0 . 2% Collagenase IV , in PBS ) for 30 minutes at 37°C using gentle pipetting to dissociate the tissue . Magnetic beads ( Dynabeads M450 , DYNAL Inc ) were coated with PECAM-1 antibody ( MEC 13 . 3 , #553369 BD Pharmingen ) as described by the manufacturer . 2 . 5×106 cells were dispensed into 0 . 7 ml micro-tubes with the Dynabeads-PECAM1 antibody ( 3× more beads than cells ) , and incubated for 1 hour at 4°C on rotating platform . The bead-antibody-cell complexes were isolated using a magnet and the remaining cell supernatant was collected as a PECAM ( - ) control . The cell complexes were gently washed with 0 . 1% BSA/PBS and collected for RNA extraction . RNA was extracted from both the PECAM+ and PECAM− cell samples using the RNA Stat-60 ( CS-110-Tel-Test , Inc . ) and RNeasy Micro Kit systems ( QIAGEN ) and treated with DNase I ( Promega ) to remove genomic DNA . RNA quality was analyzed by UV spectroscopy and agarose gel electrophoresis . One microgram of RNA was used for cDNA synthesis using the Superscript First-Strand Synthesis system ( Invitrogen ) . The SYBR-green based system of gene expression detection was utilized to quantify the relative levels of control and placenta-specific genes in each cDNA sample . To distinguish amplicons generated from genomic DNA , PCR primers were designed to span an intron-exon boundary . Melting curves were empirically determined for each primer set to establish precise temperature for data collection . Reaction amplifications were performed using a 1× SYBR Green PCR Master Mix as described by the manufacturer ( Applied Biosciences ) . QRT-PCR reactions were performed in triplicate , and each gene primer set was tested against a minimum of four cDNA samples derived from independent RNA isolates . All qRT-PCR reactions were performed using an Applied Biosystems Model 7700 Sequence Detector . The primer sequences for the control genes ( gapdh , actin , and Pecam-1 ) and candidate genes are as follows: MageL-2 , For-5′-GAACCCACGACCAGAACC-3′ , Rev-5′-CTTAGTGTTGGCACGGTTGA-3′ ( 132 bp , Ex1 ) ; Necdin , For-5′-AACGCTTTGGTGCAGTTTCT-3′ , Rev-5′- AACACTCTGGCGAGGATGAC-3′ ( 134 bp , Ex1 ) ; Adrb1 , For-5′-TGGTACGTGTTGGTGAAGGA-3′ , Rev-5′- AAGTCCAGAGCTCGCAGAAG-3′ ( 100 bp , Ex1 ) ; Enpp2 , For-5′-CCGACCTGACAATGATGAGA-3′ , Rev-5′- AAATCCAAACCGGTGAGATG-3′ ( 120 bp , Ex24–25 ) ; Tie2 , For-5′-TGAGGACGCTTCCACATTC-3′ , Rev-5′- CAACAGCACGGTATGCAAGT-3′ ( 104 bp , Ex13–14 ) ; Foxf1 , For-5′-GCAGCCATACCTTCACCAA-3′ , Rev-5′-GCCATGGCATTGAAAGAGA-3′ ( 126 bp , Ex1–2 ) ; Neuropilin-1 , For-5′-TGTCCTGGCCACAGAGAAG-3′ , Rev-5′- CCAGTGGCAGAATGTCTTGT-3′ ( 115 bp , Ex12–13 ) ; Caveolin-1 , For-5′-GGGAACAGGGCAACATCTAC-3′ , Rev-5′-ACCACGTCGTCGTTGAGAT-3′ ( 136 bp , Ex1 ) ; CD36 , For-5′-GAGTTGGCGAGAAAACCAGT-3′ , Rev-5′-GTCTCCGACTGGCATGAGA-3′ ( 143 bp , Ex3 ) ; Lyve-1 , For-5′-AGCCAACGAGGCCTGTAA-3′ , Rev-5′-CACCTGGGGTTTGAGAAAAT-3′ ( 150 bp , Ex 2–3 ) ; Pecam-1 , For-5′-CCAGTGCAGAGCGGATAAT-3′ , Rev-5′-GCACCGAAGTACCATTTCAC-3′ ( 148 bp , Ex7–8 ) ; Actin , For-5′-CCTGCCATGTATGTGGCTAT-3′ , Rev-5′-CTCATAGATGGGCACGTTGT-3′ ( 114 bp , Ex3–4 ) ; Gapdh For-5′-CACTGCCACCCAGAAGACTGT-3′ , Rev-5′-GGAAGGCCATGCCAGTGA-3′ ( 147 bp ) . Whole placentas were cut into 2 mm sections and fixed for 60 seconds in a 1% formaldehyde solution followed by treatment with 1M glycine/PBS/protease inhibitors . The fixed tissues were incubated 10 minutes in SDS lysis buffer ( Upstate Biotech ) and homogenized using disposable plastic dowels . The released chromatin was sheared at 4°C using a Biorupter sonicator ( CosmoBio , Japan ) . Proper sonication was confirmed by agarose gel electrophoresis . A ChIP assay kit ( Upstate Biotech ) was used to isolate HOXA13-DNA fragments from sheared placental chromatin . Prior to ChIP , the presence of the HOXA13 protein in the sheared chromatin was confirmed by western-immunoblot using the HOXA13 antibody previously described [72] . Chromatin produced by the HOXA13-ChIP assay was examined for the Foxf1 and Tie2 promoter sequences containing the previously described HOXA13 binding sites [71] using PCR and 30 cycles of 94 degrees Celsius for 30 seconds , 54 degrees Celsius for 30 seconds , and 72 degrees Celsius for 30 seconds . Primers flanking the potential HOXA13 binding sites were selected using the genomic sequences listed for Foxf1 ( Ensembl:MUSG00000042812 ) and Tie2 ( Ensembl:MUSG00000006386 ) . The PCR primers were: Foxf1#1 ( −2368 to −2231 , 138 bp ) , For-5′-TGAGGTACAGCCCAGAGTCC-3′ , Rev-5′-CACACCCCCAAGTTTTCTTC-3′; Foxf1#2 ( −1284 to −1163 , 122 bp ) , For-5′-CGCGGGCTTCTCTACTCTTA-3′ , Rev-5′-CCTTTTACAAGCGCAGGTTC-3′; Tie2#1 ( −2413 to −2273 , 141 bp ) , For-5′-GGGAAGGGGAGTGGATAACA-3′ , Rev-5′-CTAATCCCAGCCCTGCTGTA-3′; Tie2#2 ( −859 to −702 , 158 bp ) , For-5′-CTTCCTGTGCCAAGTTCTCC-3′ , Rev-5′- GACCAGATTCCACAGCCATT-3′ . Each ChIP assay was performed at least two times using unique placenta samples in order to confirm the ChIP results . The HOXA13 DNA-binding domain peptide was purified and gel shifts performed as previously described [72] . The Foxf1 and Tie2 primers described in the ChIP assay were used to amplify the genomic regions for the EMSA assay . The amplified PCR products were isolated using the QIAquick Gel Extraction Kit ( Qiagen ) , quantified by UV spectroscopy , and sequenced to confirm the correct gene promoter sequence . PCR amplicons were radiolabeled using γ-32P dATP ( 3000 Ci/mmol; 1Ci-37 GBq ) and T4 polynucleotide kinase as described by the manufacturer ( Promega ) . EMSA assays were performed using the Gel Shift Binding System following manufacturer's protocol ( Promega ) . The 121 bp Foxf1 and 140 bp Tie2 DNA elements identified by ChIP to be bound by HOXA13 in the labyrinth vascular endothelia were evaluated in vitro for their capacity to facilitate gene expression in the presence of HOXA13 . Based on the small size of these ChIP-positive DNA sequences as well as our previous characterization of HOXA13's relatively low in vitro transcriptional activity , a promoter- and enhancer-less luciferase reporter plasmid , pGL3-Basic ( Promega ) , was selected for this analysis [65] , [66] . The pGL3-Basic vector has been previously shown to measure minimal promoter activity with high reproducibility [71] , [72] , [74]–[77] , [124] . Luciferase assays and cell culture were performed as previously described [71] , [72] . The NG108-15 cell line ( ATCC#HB-12317 ) was selected for the luciferase assay based on the absence of endogenous HOXA13 expression [124] . Transfections were performed in 12 well plates ( Costar ) using 2 µg of the Tie2 or Foxf1 luciferase vectors , 0 . 25 µg pRL-CMV Renilla , and 0 . 5 µg pCAGGS-Hoxa13 or empty pCAGGS control plasmid per well as described [71] . Cell lysates were processed to detect luciferase activity using the Dual-Glo Luciferase Assay System ( Promega ) in OptiPlate-96F black plates ( Packard ) as described ( Promega ) [72] . Luciferase activity was detected using a Packard Fusion Microplate Analyzer ( Perkin Elmer ) , wells were read 3 times for 1 sec each and averaged . Three replicates of each transfection were performed and each transfection assay was repeated 3 times . Results were normalized for Renilla expression and averaged . The averaged data and standard errors were plotted using SigmaPlot 9 . 0 ( Systat ) . Quantification of HOXA13 affinity for the binding sites present in Foxf1 and Tie2 was determined by fluorescence polarization ( FP ) anisotropy as described [71] . Fluorescein-labeled self-annealing hairpin oligonucleotides specific for the binding sites in the ChIP-positive regions of Tie2 and Foxf1 were synthesized by Integrated DNA Technologies ( Coralville , IA ) : Tie2#1: 5′-ctgtaattaaataccccccgtatttaattacag-3′; Tie2#2: 5′-catttaataaaaaccccccgtttttattaaatg-3′; Foxf1: 5′-cttattattaaaggccccccctttaataataag-3′; TGAC control: 5′-tgactgactgactgccccccagtcagtcagtca-3′ . 1 nM of the annealed fluorescein-labeled oligonucleotides was incubated with 0–200 nM HOXA13-DBD peptide at 15 degrees Celsius and the FP values were measured using the Beacon 2000 Fluorescence Polarization System ( Invitrogen ) . The detected millipolarization values ( mP ) were plotted using a non-linear least squares fit iteration as described [71] . Each data point represents an average ( ±standard error ) of 3 or more independently derived mP values for each concentration of HOXA13-DBD peptide . | Defects in placental development are a common cause of mid-gestational lethality . Key to the placenta's function is its vascular labyrinth , a series of finely branched vessels that facilitate the efficient exchange of gases , nutrients , and metabolic waste between the maternal and fetal blood supplies . In this study , we identify a novel role for the transcription factor HOXA13 in formation of the placental vascular labyrinth . In the absence of HOXA13 function , labyrinth vessel branching and endothelial specification is compromised , causing mid-gestational lethality due to placental insufficiency . Analysis of the genes affected by the loss of HOXA13 function revealed significant reductions in the expression of several pro-vascular genes , including Tie2 and Foxf1 . Analysis of the Tie2 and Foxf1 promoters confirmed that HOXA13 binds sites present in each promoter with high affinity in the placenta , and in vitro , HOXA13 can use these bound sequences to regulate gene expression . These results suggest that Tie2 and Foxf1 are direct transcriptional targets of HOXA13 in the developing placental labyrinth , providing a novel transcriptional pathway to consider when examining pathologies of the placenta and placental insufficiency , as well as the evolutionary mechanisms required for the emergence of the vascular placenta in eutherian mammals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/cell",
"differentiation",
"developmental",
"biology/molecular",
"development",
"developmental",
"biology/developmental",
"evolution",
"developmental",
"biology/developmental",
"molecular",
"mechanisms"
] | 2008 | HOXA13 Is Essential for Placental Vascular Patterning and Labyrinth Endothelial Specification |
The initial response to Leishmania parasites is essential in determining disease development or resistance . In vitro , a divergent response to Leishmania , characterized by high or low IFN-γ production has been described as a potential tool to predict both vaccine response and disease susceptibility in vivo . We identified uninfected and healthy individuals that were shown to be either high- or low IFN-γ producers ( HPs and LPs , respectively ) following stimulation of peripheral blood cells with Leishmania braziliensis . Following stimulation , RNA was processed for gene expression analysis using immune gene arrays . Both HPs and LPs were shown to upregulate the expression of CXCL10 , IFI27 , IL6 and LTA . Genes expressed in HPs only ( CCL7 , IL8 , IFI44L and IL1B ) were associated with pathways related to IL17 and TREM 1 signaling . In LPs , uniquely expressed genes ( for example IL9 , IFI44 , IFIT1 and IL2RA ) were associated with pathways related to pattern recognition receptors and interferon signaling . We then investigated whether the unique gene expression profiles described here could be recapitulated in vivo , in individuals with active Cutaneous Leishmaniasis or with subclinical infection . Indeed , using a set of six genes ( TLR2 , JAK2 , IFI27 , IFIT1 , IRF1 and IL6 ) modulated in HPs and LPs , we could successfully discriminate these two clinical groups . Finally , we demonstrate that these six genes are significantly overexpressed in CL lesions . Upon interrogation of the peripheral response of naive individuals with diverging IFN-γ production to L . braziliensis , we identified differences in the innate response to the parasite that are recapitulated in vivo and that discriminate CL patients from individuals presenting a subclinical infection .
Cutaneous Leishmaniasis ( CL ) caused by Leishmania braziliensis is characterized by a broad spectrum of clinical manifestations , ranging from localized CL to Mucosal Leishmaniasis ( rev . in [1] ) . A hallmark feature of the immunological response in localized CL is a strong Th1 type immune response to soluble Leishmania antigen ( SLA ) , demonstrated by a positive delayed type hypersensitivity ( DTH ) reaction to the Leishmania skin test , as well as lymphocyte proliferation and high levels of IFN-γ and TNF-α [2–4] . Since T-cell-mediated immunity plays a central role in the host’s response to intracellular parasites , in vitro experimental settings have been used to address the initial lymphocyte response to Leishmania: PBMCs from naive volunteers stimulated with L . major produce mainly IFN-γ and this effect is regulated by IL-10 and IL-12 [5] . Using an in vitro priming system with Leishmania amazonensis antigen , Pompeu et al . showed that cells from naïve individuals produce either high or low amounts of IFN-γ [6] . These two patterns of in vitro anti-Leishmania response correlated with the in vivo post-vaccination response: low in vitro IFN-γ producers exhibit a delayed response to vaccination with SLA , whereas an accelerated immune reaction vaccine is observed in those who were high IFN-γ producers [6] . Upon stimulation with L . amazonensis , high IFN-γ producers also secrete more TNF [6] , more IL-12 and less IL-13 [7] . These results indicate that a low IFN-γ response in vitro accompanies a slower IFN-γ production in vivo and authors suggested that in vitro responses could be used to predict , for example , the pace of post vaccination responses . IFN-γ , produced primarily by T cells and natural killer cells , is an important mediator of macrophage activation and intracellular pathogen killing , including Leishmania . We previously demonstrated that PBMCs from healthy uninfected individuals respond differently to Leishmania stimulation ( secreting either high or low amounts of IFN-γ ) . In this study , we aimed at characterizing the immune gene signature that parallels these two responses . Further , we investigated whether such in vitro responses had in vivo equivalents by probing the gene expression of CL patients and that of individuals presenting a subclinical infection which is associated with absence of lesions , a positive Leishmania skin test ( LST ) [8] , and lower levels of both IFN-γ and TNF [9] . We expand the current knowledge in the field by identifying genes that are expressed in association with the capacity to produce IFN-γ upon stimulation with Leishmania braziliensis . The immune signature associated with IFN-γ production also discriminates CL patients from individuals with subclinical infection .
Peripheral Blood Mononuclear Cells ( PBMCs ) were obtained from healthy uninfected individuals ( n = 9 ) recruited in the city of Salvador ( Bahia state , Brazil ) , where L . braziliensis transmission in not endemic ( S1 Table ) . These individuals had negative serology results for leishmaniasis , negative serology for Chagas’ disease , hepatitis and human immunodeficiency virus . CL patients and individuals presenting a subclinical ( SC ) infection were recruited from the area of Jiquiriça ( Bahia state , Brazil ) , where L . braziliensis transmission is endemic ( S2 Table ) . Patients with active CL ( n = 5 ) were diagnosed based on the presence of a typical clinical leishmaniasis lesion , a positive Leishmania skin test ( LST ) and documentation of parasites in culture or by histopathology . SC individuals ( n = 8 ) were identified in the same endemic area , following a medical interview . These individuals had no history of past CL ( absence of scars consistent with CL or Mucosal Leishmaniasis in the skin , nose and soft palate ) and a positive LST to Leishmania . This research was conducted with the approval of the ethical committee of Centro de Pesquisas Gonçalo Moniz ( CPqGM ) , Fundação Oswaldo Cruz ( FIOCRUZ ) ( Salvador , Bahia , Brazil; 177/2008 ) and Comissão Nacional de Ética em Pesquisa ( Brazilian National Ethics Committee , Brazil ) , and written informed consent was obtained from each participant . No minors participated in the study . L . braziliensis promastigotes ( strain MHOM/BR/01/BA788 ) were grown in Schneider medium ( Sigma ) , supplemented with 100 U/ml of penicillin , 100 ug/ml of streptomycin and 10% heat-inactivated fetal calf serum ( all from Invitrogen ) . PBMCs from healthy individuals ( n = 9 ) were obtained from heparinized venous blood layered over a Ficoll-Hypaque gradient ( GE Healthcare ) . Cells were washed and resuspended in RPMI1640 supplemented with 10% human AB serum , 100 IU/ml of penicillin and 100μg/ml of streptomycin ( all from Invitrogen ) . PBMCs ( 3x106/ml ) were placed in the wells of a 24-well plate at 500 μl per well . L . braziliensis stationary phase promastigotes were added to the cultures at a parasite/cell ratio 1:1 . Control cultures were maintained in medium only . Cultures were performed in triplicate and maintained at 37°C/5% CO2 . After 72h , IFN-γ levels in culture supernatants were determined by ELISA ( R&D Systems ) , following manufacturer´s instructions . PBMCs obtained from previously defined HPs ( n = 3 ) and LPs ( n = 3 ) were stimulated with L . braziliensis promastigotes for 72h , as described above . After stimulation , total RNA was obtained using Trizol ( Invitrogen ) , according to manufacturer's instructions . RNA ( 500 ng ) was suspended in 50μl DEPC-treated water and stored at –70°C until use . cDNA was synthesized from DNAse-treated RNA by reverse transcription using RT2 First Strand kit ( Qiagen ) , following manufacturer´s instructions . cDNA obtained from cultures stimulated with L . braziliensis or from control cultures ( maintained in the absence of stimulus ) was then employed in PCR array analysis using RT2 Real-Time SYBR Green PCR Master Mix ( Qiagen ) and the following human RT2 Profiler PCR arrays: Th1 & Th2 responses , Toll-Like Receptor Signaling Pathway , Interferon & receptors and Chemokines & Receptors ( Qiagen ) , following manufacturer´s instructions . Reactions were performed on ABI 7500 Sequence Analyzer ( Applied Biosystems ) . Fold changes in gene expression between L . braziliensis-stimulated and control cultures were calculated using the RT2 Profiler PCR array data analysis tool , based on the ΔΔCt method , after normalization to housekeeping genes , determined by the manufacturer . A gene was considered differentially expressed when fold change was above or below 2 , compared to control cultures , and p<0 . 05 when comparing the different groups [HPs ( n = 3 ) and LPs ( n = 3 ) ( such genes are indicated in Supplemental Data Set 1 ) ] . For confirmation of results obtained in the RT2 Profiler PCR arrays , PBMCs from HPs ( n = 3 ) and from LPs ( n = 3 ) were stimulated with L . braziliensis or were cultured in the absence of stimulus ( control ) . RNA was employed in individual quantitative Real Time PCR ( qRT-PCR ) reactions using primers for IFNG , CXCL10 , IFI27 , IL6 and IRF1 designed using Primer Express Software ( ThermoFisher Scientific ) , reactions were performed as described elsewhere [10] . qRT- PCRs reactions were run in triplicates for each gene of interest and compared with a housekeeping gene ( GAPDH ) , also using the ΔΔCt method [11] . We compared the stability of B2M , HPRT1 , ACTB and GAPDH ( housekeeping genes ) in our PBMC samples ( stimulated or not with L . braziliensis ) . Only ACTB and GAPDH displayed normal distribution ( by both Shapiro-Wilk and Kolmogorov-Smirnov tests ) and low Coefficient of Variation ( B2M 94 . 39% , HPRT1 277 . 36% , GAPDH 60 . 61% and ACTB 50 . 75% ) . However , GAPDH displayed lowest skewness ( 0 . 03 vs . 0 . 65 for B2M , 3 . 97 for HPRT1 and 0 . 28 for ACTB ) , i . e . was very close to Gaussian distribution across all samples . Also in this experiment , IFNG transcripts were strongly correlated to B2M ( r = 065 , p = 0 . 004 ) , HPRT1 ( r = -0 . 44 , p = 0 . 078 ) and ACTB ( r = -0 . 44 , p = 0 . 080 ) transcripts , in agreement with their annotation as IFN-regulated genes . In contrast , GAPDH transcripts were not significantly correlated to IFNG transcripts ( r = -0 . 13 , p = 0 . 62 ) . To eliminate this small , but possible bias , we choose GAPDH as the housekeeping gene for the validation experiments . As a positive control , PBMCs from HPs and LPs were stimulated with Phytohaemagglutinin ( SIGMA ) ( 10μg/ml ) ; RNA was extracted and submitted to qRT-PCR for IFNG expression as described above . PBMCs from CL patients ( n = 5 ) and from SC individuals ( n = 8 ) were also stimulated with L . braziliensis; RNA was then submitted to qRT-PCR against IFI27 , IFIT1 , TLR2 , IRF1 , JAK2 and IL6 using custom designed primers . In these experiments , PBMCs from CL patients and SC individuals were obtained before placement of LST . IFN-γ levels in culture supernatants and gene expression levels after PBMC stimulation were compared by the Mann-Whitney test using Prism ( GraphPad , V . 6 . 0 ) . Functional analyses were generated using Ingenuity Pathway Analysis ( IPA , QIAGEN , V . 01–06 ) , using as the Reference Set ( Population of genes considered for p-value calculation ) , a User Data Set instead of the Ingenuity Knowledge Base . The User Data Set consisted of the 269 genes evaluated in the arrays . The genes differentially modulated in HPs ( 32 ) and LPs ( 29 ) were tested against this panel of 269 genes collectively present in the arrays . Hierarchical cluster analysis using the Ward’s method with bootstrap and principal component analysis ( PCA ) were performed using and JMP Statistical Discovery ( V 12 ) . The number of transcripts for ( IFI27 , IFIT1 , TLR2 , IRF1 , JAK2 and IL6 ) was quantified in data sets GSM1341365 [12] and GSM1560512 [13] containing transcriptomic data from CL lesions and from healthy ( control skin ) . Data were normalized using RMA ( Robust Multichip Average ) for each dataset and the number of transcripts were compared by unpaired t-test using Prism ( GraphPad Software , V 6 . 0 ) . For all comparisons , a p-value ≤ 0 . 05 was considered significant .
We have previously shown that PBMCs from naive volunteers exposed to Leishmania secrete IFN-γ in high or low amounts , allowing the classification of such individuals as either high- or low- producers ( HPs and LPs , respectively ) [6] . Herein , we aimed at further dissecting these differential responses to Leishmania , particularly in relation to the expression of immune-related genes that parallels IFN-γ production following exposure to L . braziliensis . Initially , PBMCs from healthy individuals were exposed to Leishmania promastigotes and supernatants collected after 72h were assayed for the presence of IFN-γ . The IFN-γ concentrations detected in culture supernatants evidenced individuals who are high IFN-γ producers ( HPs ) and individuals who are low IFN-γ producers ( LPs ) ( Fig 1 ) . Among all volunteers ( n = 9 ) , the median IFN-γ level was 233 pg/ml , this value was further considered as the cut off for defining HPs and LPs . Therefore , in HPs , IFN-γ levels in culture supernatants were >300pg/ml whereas LPs were defined as presenting IFN-γ levels <300pg/ml ( Fig 1 ) . As a positive control , PBMCs from HPs and LPs were stimulated with a mitogen and IFNG expression , as accessed by qRT-PCR , did not differ significantly between the two groups ( median IFNG expression was 11 . 24 for HPs and 7 . 66 for LPs ) . The difference in IFN-γ levels observed in HPs and LPs did not result from distinct responses to L . braziliensis infection since both HPs and LPs displayed similar percentages of infected macrophages and of amastigotes per infected macrophages ( S1 Fig ) . Also , the significant difference in IFN-γ production comparing HPs and LPs was replicated in a subsequent experiment ( S2 Fig ) . To investigate the expression profile of immune genes paralleling the two patterns of L . braziliensis-induced IFN-γ responsiveness ( HPs vs . LPs ) , total RNA obtained from L . braziliensis-stimulated cultures was employed in PCR arrays covering Th1-Th2-Th3 responses , IFN and receptors , chemokines and TLRs . Among the 269 genes evaluated , we identified 49 genes differentially expressed in L . braziliensis-stimulated cultures relative to control cultures , considering both HPs and LPs ( fold change above or below 2 , compared to control cultures , and p<0 . 05 ) . These genes ( indicated in Supplemental Data Set 1 ) were identified based on differential expression following stimulation with L . braziliensis , as described in Materials and Methods . Twenty genes were uniquely modulated in HPs whereas 17 genes were uniquely modulated in LPs ( Fig 2A ) . Twelve genes were commonly modulated in both HPs and LPs and , within these , IFNG was the top up-regulated gene ( Fig 2B ) . IFNG expression was ~8-fold higher in HPs compared to LPs , a finding that recapitulated the higher IFN-γ levels detected in culture supernatants ( Fig 1 ) . In addition to IFNG , eleven other genes were also modulated in both HPs and LPs ( Fig 2B ) including CXCL10 and IL6 ( for which the expression level was ~4-fold higher in HPs compared to LPs ) . HPs also upregulated genes associated with a type I interferon ( IFN ) response such as OAS1 , MX1 and IRF1 . IFI17 , upregulated in response to stimulation by interferon [14] , was highly expressed in HPs . On the converse , expression of CD180 , LY86 and TLR5 was suppressed at similar levels in both HPs and LPs ( Fig 2B ) . These results demonstrate that HPs and LPs modulated , in general , a similar number of genes , twelve of which were common between both categories . We then selected the top four genes modulated in PCR arrays in both HPs and LPs ( IFNG , CXCL10 , IFI27 and IL6 ) plus IRF1 and validated their expression by qRT-PCR , using custom designed primers ( Fig 3 ) . Reactions performed with RNA from the same HPs ( n = 3 ) and LPs ( n = 3 ) confirmed that HPs express higher levels of IFNG , CXCL10 , IFI27 , IL6 and IRF1 compared to LPs . Therefore , higher production of IFN-γ in response to L . braziliensis stimulation is accompanied by upregulation of a series of known IFN-stimulated genes such as CXCL10 , IFI27 and IRF1 . Within the genes differentially expressed only in HPs ( n = 20; indicated with p< 0 . 05 in ( Supplemental Data Set 1 ) , CCL7 , IL8 , IFI44L , IL1B and CSF2 were expressed >20-fold in L . braziliensis-stimulated cultures compared to control cultures ( Fig 4A ) . Genes coding for chemokines ( CXCL1 ) , cytokines ( IL1A , TNF , IL7 ) , transcription factors ( STAT1 , ELK1 ) , protein kinases ( JAK2 , RIPK2 ) , F3 ( coagulation factor III ) , IFIT2 , IL31RA and JUN were also upregulated in HPs whereas LY96 and CMKLR1 were downregulated ( Fig 4A ) . Following the identification of the immune genes differentially expressed in HPs [n = 32 , 12 common genes ( Fig 2 ) plus 20 unique genes , indicated with p< 0 . 05 in Supplemental Data Set 1] , we inquired the immune pathways associated with that expression profile . Among the pathways significantly enriched were IL17 and TREM1 ( triggering receptor expressed on myelois cells ) signaling ( Fig 4B ) . In LPs , fold expression was in general lower than in HPs and the uniquely upregulated genes ( n = 17; genes are indicated with p< 0 . 05 in Supplemental Data Set 1 ) were cytokines ( IL9 ) , pattern recognition receptors ( TLR2 ) , cytokine receptors ( IL2RA , TNFSRF9 , IL3RA ) , IFN-related molecules ( IFI44 , IFIT1 , IFITM2 ) and LAG3 , which belongs to the Ig superfamily ( Fig 5A ) . Moreover , in LPs , MAF , IL5RA , TLR4 , MAPK8 , IL10 , STAT6 and CD14 expression was suppressed upon stimulation with L . braziliensis . As before , upon identification of the differentially expressed genes in LPs [n = 28 , 12 common genes ( Fig 2 ) plus 17 unique genes , indicated with p< 0 . 05 in Supplemental Data Set 1] , pathways involved in the role of pattern recognition receptors , IL-12 and in interferon signaling were enriched ( Fig 5B ) . Altogether , our findings show that HPs , but no LPs , modulate the expression of genes associated with the inflammatory response ( CCL7 , IL8 , IL1B , IL1A and TNF ) and the overall immune signature is associated with IL-17-related pathways . Following the observation that HPs modulate genes associated with an inflammatory response , we hypothesized that these features could have an in vivo equivalent in CL patients . In CL , the immune response is characterized by strong production of IFN-γ and TNF , cytokines that are important for controlling infection but that are also associated with pathogenesis ( rev . in [15] ) . The in vivo counterpart to LPs would be individuals with subclinical infection ( SC ) . SC individuals do not present a clinical lesion , display a positive LST [8] and lower IFN-γ and TNF production upon PBMC stimulation [9 , 16] . We therefore selected genes modulated in both HPs and LPs ( IL6 , IFI27 and IRF1 ) , modulated in HPs only ( JAK2 ) or in LPs only ( IFIT1 and TLR2 ) . Additionally , these genes have been implicated in the control or progression of Leishmania infection: type I IFN positively regulates SOD1 levels , decreasing superoxide and increasing L . amazonensis and L . braziliensis burden in vitro [17 , 18]; TLR2 has been associated with reduced pathology in vaccination studies [19]; JAK2 is modulated in Leishmania-infected cells [20] and IL6 has been associated with CL/mucosal leishmaniasis susceptibility [21] . PBMCs obtained from CL patients and from SC individuals were stimulated with L . braziliensis and expression of these genes was measured by qRT-PCR . Expression of IFI27 , IFIT1 and TLR2 did not differ significantly comparing CL patients and SC individuals ( Fig 6A ) but expression of IRF1 , JAK2 and IL6 was significantly higher in CL individuals . Hierarchical clustering showed that the pattern of expression of IFIT1 , TLR2 , IFI27 , IRF1 , JAK2 and IL6 successfully discriminated CL patients from SC individuals ( Fig 6B ) and this result was further confirmed by principal component analysis ( Fig 6C ) showing that immune markers that accompany High- or Low IFN-γ production in naïve individuals , following exposure to L . braziliensis , are recapitulated in vivo . To address whether these genes are also expressed in CL lesions , we performed in silico analysis of microarray transcriptomic data generated from human CL lesions caused by L . braziliensis [12 , 13] . Expression of IRF1 , JAK2 , IL6 and IFI27 was significantly higher in CL lesions , corroborating our findings tiwht PBMCs ( Fig 7A and 7B ) . Expression of TLR2 and IFIT1 was also significantly higher in CL lesions , suggesting that these molecules maybe differentially modulated at the infection site , compared to PBMCs . These findings indicate that differential expression we observed in PBMCs from both HPs and CL patients is also observed in vivo .
The initial encounter among Leishmania parasites and cells from the host´s immune system is fundamental in determining disease development or resistance . In naïve volunteers , this event is reflected in either high or low IFN-γ production [6] . Herein , we investigated the expression of a set of immune response-related genes that accompanies these polarized responses in PBMCs from naïve volunteers exposed to L . braziliensis . We initially confirmed earlier findings by Pompeu et al . [6] regarding two patterns of IFN-γ production in the peripheral response to Leishmania , suggesting that this dichotomy may be a common feature following exposure of naïve cells to these parasites . Importantly , this dichomoty is stable as LPs remained low IFN-γ producers 40 days after the initial investigation [6] . In parallel to IFNG , genes such as CXCL10 , IFI27 , IL6 and LTA were upregulated in both HPs and LPs , though to different extents . IFN-γ enhances the production of CXCL10 [22] and CXCL10 activates NK cells , further inducing the secretion of IFN-γ [23] . Schnorr et al . suggested that NK cells maybe a source of IFN-γ in the response to Leishmania and since we also detected IFNG mRNA and IFN-γ protein after 72h of L . braziliensis stimulation , we suggest that NK cells may play a role . Also , IFI27 is highly induced by type I IFN and IFN-α stimulates IL-12 secretion , promoting IFN-γ production [24] . We speculate that IFI27 , highly expressed in HPs , could have contributed towards the greater IFN-γ level observed in HPs . On the other hand , IL6 drives the differentiation of CD4+ Th2 cells by inducing early production of IL4 [25] and by interfering with SOCS1 phosphorylation[26] . In experimental Visceral Leishmaniasis , IL-6 deficient mice showed enhanced control of L . donovani infection [27] , suggesting a possible deleterious role for IL6 , which was expressed >4-fold in HPs , compared to LPs . Apart from the commonly upregulated genes , HPs and LPs equally suppressed the expression of CD180 , LY86 and TLR5: CD180 belongs to the TLR family of pathogen receptor and it is associated with MD-1 ( LY86 ) . MD-1 cooperates with CD180 and TLR4 in the recognition of LPS [28] whereas TLR5 is a receptor for bacterial flagellin . Downmodulation of such pathogen recognition receptors in both HPs and LPs , we speculate , may limit the initial inflammatory reaction , enabling the establishment of infection . Upon examination of genes uniquely upregulated in HPs , the pro-inflammatory signature suggested earlier is further supported by expression of CCL7 ( a chemoattractant for macrophages during inflammation ) , IL8 , IL1A and IL1B , all of which have been detected in CL lesions [29] and TNF , a cytokine extensively associated with the pathogenesis of CL [4 , 30–32] also present in high IFN-γ producers , as described by Pompeu et al . [6] . PTGS2 , the inducible Prostaglandin-endoperoxide synthase/cyclooxygenase , is also expressed in CL lesions caused by L . braziliensis [33] and , again , was upregulated uniquely in HPs . The top immune pathways enriched in HPs were IL-17 related . PBMCs from CL patients produce elevated levels of IL-17 compared to healthy controls [34] and Th-17-related cytokines are overexpressed in lesions from mucosal leishmaniasis patients [35] , indicating an association between IL-17 and pathogenesis in CL . TREM1 is selectively expressed on neutrophils , monocytes and macrophages and engagement of this receptor leads to a pro-inflammatory immune response [36] . This pathway triggers expression of IL1B and TNF [37] , all of which were upregulated in HPs . The possible engagement of TREM preferentially in HPs seems to suggest that a stronger anti-Leishmania response in humans maybe implicated in tissue destruction , leading to lesion development . Infection of Trem1-deficient mice with L . major induced a milder inflammatory infiltrate and smaller lesions but the absence of TREM1 signaling did not impair parasite control [38] suggesting that the TREM1 pathway is associated with excessive inflammation rather than the capacity to control experimental infection . In contrast to HPs , LPs expressed mainly IL9 , interferon-related genes ( IFI44 , IFIT1 and IFITM2 ) and receptors IL2RA , IL3RA and TLR2 . IL-9 has been associated with the Th2 phenotype: susceptible BALB/c mice infected with L . major expressed IL9 [39] and crossing of IL-9 transgenic mice to Th2 cytokine-deficient mice promoted Th2 cytokine production [40] . IL3RA ( CD123 ) is expressed by plasmacytoid DCs ( pDCs ) which can activate NK cells ( rev . in [41] ) again suggesting the participation of these cells in the early response to L . braziliensis [16] . Several transcripts of the IL2 pathway are present in CL lesions caused by L . braziliensis and certain IL2RA gene polymorphisms were associated with a poor IFN-γ response and lower activation of regulatory Foxp3+ cells [13]; IL2RA was also upregulated in LPs though herein we did not probe for any of these polymorphisms . Among the top two pathways identified in the LP response were "Role of Pattern Recognition Receptors in Recognition of Bacteria and Viruses" and " Interferon Signaling" , corroborating the upregulation in IFN-related genes such as IFI44 , IFIT1 and IFITM2 . Based on the premise that HPs displayed a more inflammatory signature , compared to LPs , we hypothesized that features of the HP response , identified in PBMCs in vitro , had an in vivo counterpart , in CL patients . On the contrary , the in vivo equivalent of LPs would be SC individuals , characterized by the presence of lower IFN-γ and TNF production upon PBMC stimulation and absence of CL lesions [9] . A group of six genes differentially expressed in HPs and LPs allowed the discrimination of CL patients from SC individuals , suggesting that features of the early in vitro response to L . braziliensis are at play in vivo . As shown elsewhere , IFN-γ and CXCL10 levels are significantly higher in CL patients compared to SC individuals [16] which agrees with our data regarding HPs and LPs , respectively . We also observed upregulated expression of JAK2 , IRF1 , IL6 and IFI27 in CL lesions , strengthening our findings regarding the expression of these molecules in PBMCs from HPs and from CL patients . In the present study we identified immune markers associated with High- and Low-IFN-γ producers , upon investigation of the peripheral response of naïve individuals to L . braziliensis . Certain markers were also expressed in vivo , in PBMCs from CL patients and SC individuals , respectively , and were further observed in situ , in CL lesions . Limitations of the study include the small number of individuals ( n = 6 ) employed in the gene expression profiles and the limited number of genes selected for evaluation in CL patients and SC individuals . Despite these important limitations , our findings highlight the importance of addressing the initial response to Leishmania since , as shown here , such approaches can potentially lead to the identification of markers of CL development . | Control and development of Cutaneous Leishmaniasis ( CL ) are dependent on the host immunological response . One of the key molecules in determining elimination of Leishmania parasites from the infected host cell is the cytokine interferon gamma ( IFN-γ ) . The aim of this study was to investigate which immune response genes are associated with the production of IFN-γ in the context of Leishmania infection . We identified individuals that are high- or low IFN-γ producers upon stimulation of their peripheral blood cells with Leishmania parasites . We then determined the immune gene expression profile of these individuals and we identified a set of genes that are differentially expressed comparing high- and low IFN-γ producers . The expression of these genes was also evaluated in patients with CL and in individuals with a subclinical Leishmania infection ( SC ) . In this setting , the overall pattern of expression of this particular gene combination discriminated the CL patients x from SC individuals . Understanding the initial response to Leishmania may lead to the identification of markers that are associated with development of CL . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoans",
"signs",
"and",
"symptoms",
"leishmania",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"white",
"blood",
"cells",
"zoonoses",
"animal",
"cells",
"proteins",
"gene",
"expression",
"lesions",
"protozoan",
"infections",
"immune",
"response",
"biochemistry",
"diagnostic",
"medicine",
"cell",
"biology",
"leishmaniasis",
"genetics",
"interferons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
] | 2016 | Gene Expression Profile of High IFN-γ Producers Stimulated with Leishmania braziliensis Identifies Genes Associated with Cutaneous Leishmaniasis |
Human TRIM5α potently restricts particular strains of murine leukemia viruses ( the so-called N-tropic strains ) but not others ( the B- or NB-tropic strains ) during early stages of infection . We show that overexpression of SUMO-1 in human 293T cells , but not in mouse MDTF cells , profoundly blocks N-MLV infection . This block is dependent on the tropism of the incoming virus , as neither B- , NB- , nor the mutant R110E of N-MLV CA ( a B-tropic switch ) are affected by SUMO-1 overexpression . The block occurred prior to reverse transcription and could be abrogated by large amounts of restricted virus . Knockdown of TRIM5α in 293T SUMO-1-overexpressing cells resulted in ablation of the SUMO-1 antiviral effects , and this loss of restriction could be restored by expression of a human TRIM5α shRNA-resistant plasmid . Amino acid sequence analysis of human TRIM5α revealed a consensus SUMO conjugation site at the N-terminus and three putative SUMO interacting motifs ( SIMs ) in the B30 . 2 domain . Mutations of the TRIM5α consensus SUMO conjugation site did not affect the antiviral activity of TRIM5α in any of the cell types tested . Mutation of the SIM consensus sequences , however , abolished TRIM5α antiviral activity against N-MLV . Mutation of lysines at a potential site of SUMOylation in the CA region of the Gag gene reduced the SUMO-1 block and the TRIM5α restriction of N-MLV . Our data suggest a novel aspect of TRIM5α-mediated restriction , in which the presence of intact SIMs in TRIM5α , and also the SUMO conjugation of CA , are required for restriction . We propose that at least a portion of the antiviral activity of TRIM5α is mediated through the binding of its SIMs to SUMO-conjugated CA .
Cells have developed many mechanisms to restrict viral infection . The adaptative immune response provides major protection against viral pathogens , but recently dominant-acting inhibitory gene products , called restriction factors , have been discovered that also play an important role in limiting host susceptibility to viral infections . One class of such restriction factors blocks retroviral infection by targeting the incoming capsid protein ( CA ) ( for review see [1] ) . Early studies identified the Friend virus susceptibility factor 1 ( Fv1 ) locus as a host gene determining the susceptibility of mice to infection by various strains of murine leukemia viruses ( MLV ) [2] . Two major alleles of Fv1 were identified: Fv1n and Fv1b . Fv1n renders NIH/Swiss mice resistant to B-tropic MLV ( B-MLV ) infection and Fv1b renders BALB/c mice resistant to N-tropic virus [3] , [4] . The critical difference between the N- and B-tropic MLVs was traced to specific residues of the viral capsid ( CA ) protein [5]–[7] . Some strains of MLV , including Moloney MLV , termed NB-tropic , are insensitive to Fv1 restriction [4] . The restriction by Fv1 occurs early in infection , after reverse transcription but before viral DNA integration , and is saturable by large amount of virus [8]–[10] . The mechanism by which Fv1 restricts MLV infection is unknown , but it is generally presumed that Fv1 somehow recognizes the incoming CA protein structure and prevents normal infection . More recently , rhesus monkey TRIM5α and human TRIM5α were identified as intracellular restriction factors capable of blocking infection by human immunodeficiency virus type-1 ( HIV-1 ) and N-MLV , respectively [11]–[15] . TRIM5α blocks retroviral replication early in the life cycle , after viral entry but before reverse transcription [16]–[21] . The same residues of MLV capsid determine sensitivity to the Fv1 and human TRIM5α-mediated restriction [14] , [21] . Some human cell lines are able to potently block N-MLV infection ( HeLa , TE671 ) while others ( 293T cells ) do not block N-MLV or do so only weakly [20] . The mechanism by which TRIM5α restricts infection is unclear . TRIM5α is a member of the tripartite motif family of proteins , characterized as having three domains: a RING domain , either one or two B-boxes , and a coiled-coil domain [22] . The C-terminus of TRIM5α , unlike that of most TRIMs , consists of a B30 . 2 domain . This domain binds to CA molecules of incoming retroviruses , and its sequence determines which retroviruses a specific TRIM5α will restrict [12] , [14] , [21] , [23]–[27] . The RING domain is a cysteine-rich zinc binding domain commonly found in E3 ubiquitin ligases , and there is some evidence suggesting that TRIM5α could be a ubiquitin ligase [28] . The B-box domains are thought to act as protein-protein interaction domains and thereby determine RING box ubiquitin ligase substrate specificity [29] . The coiled-coil domain has been shown to be involved in homo- and hetero-multimerization of the TRIM proteins , and deletion of this domain in TRIM5α completely abrogates HIV-1 and N-MLV restriction [30]–[32] . SUMO proteins are small ubiquitin-related proteins that become conjugated to cellular substrates and regulate diverse cellular processes , including intracellular trafficking , cell cycle progression , transcription , DNA repair and nuclear receptor activities ( for review see [33] , [34] ) . SUMO conjugation , like ubiquitination , requires an E1 activating enzyme ( in the case of SUMO , these are AOS1-UBA2 ) , an E2 conjugation enzyme ( UBC9 ) [35] , [36] and often an E3 ligase ( RanBP2 and PIAS 1 , -3 and -4/y ) which recognize the substrate and determine target specificity [37]–[39] . In mammals , three SUMO paralogues are commonly expressed: SUMO-1 , SUMO-2 and SUMO-3 . SUMO-1 is distinct from SUMO-2 and SUMO-3; SUMO-2 and SUMO-3 are 97% identical to each other but only 47% identical to SUMO-1 . SUMO-1 and SUMO-2/3 are conjugated to different target proteins in vivo [40] , [41] , [42] and likely serve distinct functions . SUMO proteins are usually transferred to lysines of a UBC9 binding site motif of consensus sequence ΨKxE ( where Ψ is a hydrophobic residue and x any amino acid ) [43] , [44] , though lysines in other contexts can be modified . In addition to targeting different substrate proteins , the functional properties of SUMO isoforms might also reflect their ability to mediate distinct protein-protein interactions in vivo . Recent work has identified specific motifs that mediate non-covalent interactions with SUMO modified proteins [43] , [45] , [46] . The best characterized of the SUMO-interacting motifs ( SIMs ) have the consensus sequence V/I/L-x-V/I/L-V/I/L or V/I/L-V/I/L-x-V/I/L ( where x is any amino acid ) [46] , [47] . In the past few years , there have been many reports demonstrating the involvement of SUMO conjugation in virus replication . In some instances , conjugation of SUMO to either viral proteins or host proteins can impair viral infection , and hence , viruses have found ways to interfere with the pathway [48]–[50] . In other instances , SUMO conjugation of viral proteins can be essential for viral replication [41] , [48] , [51] . It has been reported that Gag proteins from Mazon-Pfizer monkey virus , Moloney murine leukemia virus , and HIV-1 interact with the SUMO conjugation pathway [52]–[54] . Our laboratory has previously reported that the E2 and E3 SUMO-conjugating enzymes , UBC9 and PIAS4/y , interact with the capsid ( CA ) protein of Moloney murine leukemia virus ( MoMLV or NB-tropic MLV ) . The UBC9 and PIAS4/y binding sites within CA were identified , and it was also demonstrated that co-expression of CA and tagged-SUMO-1 proteins resulted in SUMO conjugation of CA in vivo . Mutation of lysine residues to arginine near the UBC9 binding site and ablation of the PIAS4/y binding site reduced or eliminated CA SUMO conjugation , and impaired virus replication . This block occurred in the early stages of viral infection , after reverse transcription and before nuclear entry and viral DNA integration . The findings suggest a role for the SUMO machinery in the early stages of viral infection [54] . In an effort to further elucidate the relationship between the SUMO conjugation pathway and early events of the MLV life cycle , we tested the effects of manipulating the components of the SUMO transfer machinery in several cell lines . Surprisingly , we found that the normally weak N-MLV restriction by TRIM5α in human cells is profoundly enhanced by overexpression of SUMO-1 . The presence of two SIMs in TRIM5α is required for the enhanced N-MLV restriction . Mutation of lysines preventing CA SUMOylation , abolish TRIM5α-mediated restriction of N-MLV . Our data suggest a novel aspect of the TRIM5α-mediated restriction of N-tropic MLV , in which the presences of intact SIMs in TRIM5α , and SUMO conjugation of CA , are required for N-MLV restriction . We propose that TRIM5α recognition of CA is augmented by binding SUMO-modified CA via its SUMO-interacting motif .
To explore the significance of SUMO conjugation in the early stages of viral infection , we transduced 293T cells with an empty retroviral vector or a vector encoding HA-tagged versions of human SUMO-1 , SUMO-2 or SUMO-3 . Pools of cell lines stably expressing these proteins were selected . The presence of the HA-tagged SUMO proteins was detected by Western blot ( Figure 1A ) . The empty vector control , HA-SUMO-1 , HA-SUMO-2 , and HA-SUMO-3 overexpressing cell lines were then assayed in single round infection experiments . The different cell lines were infected in parallel with increasing amounts of VSV-G pseudotyped B- ( Figure 1B ) , N- ( Figure 1C ) , or NB-tropic MLV ( Figure 1D ) virus particles delivering a firefly luciferase reporter ( B-MLV luc , N-MLV luc and NB-MLV luc respectively ) . The overexpression of HA-SUMO-1 in 293T cells reduced the infectivity of N-MLV by an average of more than 8-fold as compared to the empty vector cell line ( Figure 1C ) . In contrast , SUMO-1 overexpression had no significant effect on susceptibility to infection by B-MLV ( Figure 1B ) or NB-MLV ( Figure 1D ) . Overexpression of HA-SUMO-2 did not affect susceptibility to infection by any of the viruses . Overexpression of HA-SUMO-3 reduced the susceptibility to infection by N-MLV by only 2 fold , again with no effect on infection by B- or NB-MLV . These data show that SUMO-1 overexpression induces or enhances a block to N-MLV infection in 293T cells . This data raised the possibility that SUMO-1 overexpression could also block N-MLV infection in murine cells that are non-restrictive for N-MLV , such as the Fv1-null Mus dunni tail fibroblasts ( MDTF ) . We transduced MDTF cells with the retroviral vectors encoding HA-SUMO-1 used above , and detected the presence of HA-SUMO-1 by Western blot ( Figure 2A ) . Overexpression of SUMO-1 in MDTF did not affect N-MLV infectivity when compared to the empty vector cell line ( Figure 2B ) . This result indicates that the inhibition of N-MLV infection by overexpression of SUMO-1 is cell-type dependent , and could require a baseline level of restriction . 293T cells have been described as non-restrictive or very weakly restrictive for N-MLV infection when compared to other human cell lines such as TE671 or HeLa [20] . To test if overexpression of SUMO-1 in other more restrictive human cells could enhance the restriction to N-MLV , we transduced TE671 cells with retroviral vectors encoding HA-SUMO-1 used above , and confirmed expression by Western blot ( Figure 2A ) . Overexpression of SUMO-1 in TE671 cells reduced infection of N-MLV by an average of 3 fold as compared to the empty vector cell line ( Figure 2C ) . This result suggests that overexpression of SUMO-1 can enhances a restriction activity present in multiple human cell lines . Since overexpression of SUMO-1 is enhancing restriction in human cells , we wondered if the knock down of endogenous SUMO-1 would be able to release restriction . We knocked down the endogenous SUMO-1 in 293T , HeLa and TE671 , then we infected the cells with N-MLV or B-MLV luc , virus infection was not significantly increased ( data not shown ) . These results are very difficult to interpret; SUMO-1 is required for early events during infection [54] and any effect observed with the different shRNA used was equivalent for both N-MLV and B-MLV luc infection , although the restriction to N-MLV infection in HeLa and TE671 cells is still present . Our data show that SUMO-1 specifically blocks N-tropic MLV and not B- or NB-tropic MLV . This suggests that the viral capsid , and its associated tropism , is a critical determinant of the antiviral effects of SUMO-1 on N-MLV . Capsid ( CA ) protein amino acid 110 appears to be the most important determinant of Fv-1 N ( specified by R110 ) and B ( specified by E110 ) tropisms [7] . To determine if CA is also the target for the block observed in SUMO-1-overexpressing 293T cells , we generated a mutant version of N-MLV luc in which amino acid 110 in CA was mutated from arginine to glutamic acid ( N-MLV luc CA R110E ) , a change known to convert Fv-1 sensitivity from N to B-tropism . N-MLV luc CA R110E infection was not blocked by HA-SUMO-1 overexpression in 293T when compared with the empty vector cell line ( Figure 3A ) . Therefore the antiviral effects of SUMO-1 on N-MLV are dependent on CA amino acid 110 . The observation that the antiviral effects of SUMO-1 are CA-dependent and cell-type specific suggests that SUMO-1 overexpression enhances the antiviral activity of an intrinsic restriction factor . Human cells contain the restriction factor TRIM5α , which confers resistance to N-MLV but not B- or NB-MLV infection [13] , [20] , [32] , [55] . TRIM5α antiviral activity was found to be dependent on CA amino acid 110 and could be abrogated with high multiplicities of infection [20] , [56] . To determine whether the SUMO-1 enhanced block could also be abrogated , 293T cells transduced with the empty vector control or HA-SUMO-1 were pretreated with increasing amounts of N-MLV containing a green fluorescent protein ( GFP ) reporter gene ( N-MLV GFP ) . Four hours later , the cells were superinfected with a fixed amount of N-MLV luc . Those 293T cells expressing the empty vector did not significantly restrict N-MLV luc infection ( Figure 1C ) . Pretreatment of those cells with N-MLV GFP had little effect on the luciferase activity after infection with N-MLV luc ( Figure 3B ) . Pretreatment of 293T cells overexpressing HA-SUMO-1 with N-MLV GFP at high doses , however , resulted in a dose-dependent loss of the SUMO-1-mediated block to N-MLV luc infection . Thus , as observed for the TRIM5α-mediated restriction , the SUMO-1 block of N-MLV infection can be abrogated . Under normal circumstances , TRIM5α blocks retroviral replication early in the life cycle , after viral entry but before reverse transcription [19] , [57] . To determine whether the block is occurring at a similar time in the SUMO-1 overexpressing cells , the course of viral DNA synthesis was examined by qPCR after acute infection . 293T cells stably transduced with the empty vector or HA-SUMO-1 were infected with N-MLV luc at two different dilutions . The levels of amplified viral DNA products correlated well with the levels of input virus in the empty vector control cell line . Examination of an early step in reverse transcription , using primers that detect the minus strand strong stop ( MSS ) DNA ( the first detectable product of viral DNA synthesis ) , revealed a significant reduction in MSS DNA products in the SUMO-1-overexpressing cell line as compared to the empty vector cell line ( Figure 3C , left panel ) . In correlation with this , very low levels of luciferase DNA sequences , which correspond to linear fully reverse transcribed viral DNA from our luciferase reporter ( Figure 3C , middle panel ) , and nuclear viral DNA forms , analyzed by detection of the 2-LTR junction ( Figure 3C , right panel ) were detected in the SUMO-1-overexpressing cell line as compared to the empty vector cell line . These data shows that SUMO-1 overexpression resulted in an early block in the viral life cycle , at a step prior to reverse transcription , and that the block to early forms continues to affect later DNA forms in the course of infection . TRIM5α-mediated restriction of N-MLV infection is dependent on CA amino acid 110 , occurs prior to reverse transcription , and can be saturated with high multiplicities of infection [14] , [20] , [56] . The SUMO-1 block of N-MLV infection in 293T has the same characteristics , suggesting that TRIM5α could be mediating the block of N-MLV by SUMO-1 overexpression . To determine whether TRIM5α mediates SUMO-1 restriction of N-MLV , we knocked down the expression of TRIM5α in the HA-SUMO-1 overexpressing cell line using shRNAs specific for the coding sequence or the 3′ UTR region of human TRIM5α mRNA ( Figure 4A ) . An shRNA containing a scrambled , nonsilencing ( scr ) sequence was used as a control . TRIM5α mRNA was efficiently reduced by all of the targeted shRNAs tested , as determined by qPCR ( Figure 4B ) . Expression of all five of the shRNAs directed specifically to the TRIM5α transcript abolished the HA-SUMO-1 block to N-MLV luc infection ( Figure 4C ) , while expression of the control scr shRNA had no effect . This result strongly suggests that human TRIM5α is required for the SUMO-1 block of N-MLV infection . To confirm that elimination of the SUMO-1 block to N-MLV infection by TRIM5α shRNA was due to reduced levels of TRIM5α and not the result of an off-target effect , we transiently transfected a plasmid encoding a FLAG-tagged human TRIM5α ORF , but lacking any TRIM5α UTR sequences into the HA-SUMO-1/shRNA4 cell line , in which the shRNA is directed to the 3′UTR region of the TRIM5α transcript . The SUMO-1 block to N-MLV was restored when we reintroduced the shRNA-resistant TRIM5α in the HA-SUMO-1/shRNA4 cell line ( Figure 4D ) . We confirmed the restored TRIM5α expression by Western blotting of the same extracts used for the luciferase assay ( Figure 4D bottom ) . These results demonstrate that the block of N-MLV infection upon SUMO-1 overexpression requires human TRIM5α . There is a complex relationship between the SUMO and ubiquitin pathways , and it is possible that the ubiquitin ligase activity of the RING domain of TRIM5α is required for N-MLV restriction upon SUMO-1 overexpression . To test this , we generated a mutant version of TRIM5α in which cystein 15 and 18 in the first zing finger were mutated to alanine ( C15A/C18A ) , in the context of the FLAG-TRIM5α vector . The SUMO-1 block to N-MLV was restored when we reintroduced the C15A/C18A TRIM5α in the HA-SUMO-1/shRNA4 cell line ( Figure 5C ) . This result indicated that the RING domain E3 ubiquitin ligase activity of TRIM5α is not required for the block of N-MLV infection upon SUMO-1 overexpression . As our data suggested an unanticipated relationship between SUMO-1 and TRIM5α , we asked if TRIM5α contains any amino acid motifs indicative of an interaction with the SUMO conjugation pathway . Analysis of the TRIM5α protein sequence revealed an N-terminal consensus site for SUMO conjugation and three potential SUMO-interacting motifs ( SIM ) in the B30 . 2 domain ( Figure 5A ) , motifs often identified in proteins involved in non-covalent SUMO binding . The first potential SIM ( SIM1 ) consists of an ILGV hydrophobic core that is followed by a cysteine residue . The second potential SIM ( SIM2 ) consists of a VIGL hydrophobic core that is juxtaposed with two acidic amino acids ( EE ) . The third potential SIM ( SIM3 ) consists of an IVPL hydrophobic core that is followed by a Ser residue . To determine the importance of these sites to the antiviral activity of TRIM5α , we independently mutated the consensus SUMO conjugation site ( K10R ) , the first SIM ( SIM1mut ) , the second SIM ( SIM2mut ) , and the third SIM ( SIM3mut ) in the context of the FLAG-TRIM5α vector ( Figure 5B ) . We transiently transfected the HA-SUMO-1/shRNA4 cell line with constructs encoding these human TRIM5α variants and , twenty-four hours after transfection , infected the cells with N-MLV luc . Luciferase activity was measured forty-eight hours post-infection , and TRIM5α protein expression was confirmed by Western blotting ( Figure 5C , bottom ) . Both wild-type TRIM5α and K10R mutant proteins were able to restore the block of N-MLV infection in the HA-SUMO-1/shRNA4 cell line ( Figure 5C ) . This suggests that SUMO conjugation of TRIM5α at K10R is not required for restriction of N-MLV . Although SUMOylation of TRIM5α may occur , in spite of many efforts , we have not been able to demonstrate biochemically that human TRIM5α undergoes SUMO conjugation ( data not shown ) . In contrast , the mutations in the first and second SIM sequences of TRIM5α ( SIM1mut , SIM2mut ) relieves the restriction activity ( Figure 5C ) . The SIM3mut protein restricted N-MLV infection to levels similar to wild type and K10R TRIM5α . These results suggest that SIM1 and SIM2 motifs present in the B30 . 2 domain of TRIM5α are important for restriction activity of TRIM5α in the context of the 293T HA-SUMO-1 cell line . We wondered if SUMO-1 overexpression impacted on TRIM5α expression levels . To answer this we transiently express FLAG-TRIM5α wild type or the mutant versions in the empty vector control cell line or the cells that express HA-SUMO1 , and compared by Western blot the levels of TRIM5α expressed in the 2 cell lines . When SUMO-1 is overexpressed we found that there was only 1 . 2-fold more wild-type and SIM3mut TRIM5α than in the empty vector control cells , and that no change was observed for K10R , SIM1mut or SIM2mut TRIM5α ( Figure S1A and B ) . Therefore , SUMO-1 overexpression has no impact on TRIM5α expression levels . It has been documented that exogenous expression of human TRIM5α in permissive cells , such as Crandall feline kidney ( CRFK ) fibroblast or Mus dunni ( MDTF ) cells , imparts a block against N-MLV as potent as the one observed in human cells [12] , [13] , [58] . To determine the ability of TRIM5α mutants to inhibit infection of N-MLV in cells that do not overexpress SUMO-1 and are permissive for N-MLV infection , we generated MDTF and CRFK cells stably expressing either wild-type or mutant Myc-tagged versions of TRIM5α . The expression levels of the mutants were somewhat variable ( Figure 5D ) , though TRIM5α protein levels have not been found to correlate with the strength of restriction in cells where TRIM5α is overexpressed [25] , [58] . To measure the retroviral restriction activities of the TRIM5α wild-type , K10R , and SIM mutants , populations of MDTF and CRFK cells expressing these various TRIM5α proteins were infected with increasing doses of N-MLV luc , and the cultures were analyzed for luciferase activity . In MDTF cells , wild-type and K10R TRIM5α restricted N-MLV . Importantly , the SIM1mut and SIM2mut proteins were completely unable to mediate restriction of N-MLV ( Figure 5E ) , and the SIM3mut showed modestly reduced restriction as compared to the wild-type TRIM5α . As in the wild type TRIM5α overexpressing cells , both K10R and SIM3mut , restricted N-MLV infection before reverse transcription ( Figure S2 ) . Similar phenotypes of restriction for the different TRIM5α mutant proteins were observed in CRFK cells ( Figure 5F ) . Thus , SIM1 and SIM2 of human TRIM5α are crucial for N-MLV restriction in general , not only in the context of SUMO-1 overexpression . The SIMs present in human TRIM5α are conserved in several primates orthologs ( Figure S3 ) . To determine if the SIM1 and SIM2 are also required for restriction of N-MLV by TRIM5α of other species , we generated the same mutations used above in the rhesus monkey ( Macaca mulatta ) TRIM5α , which has been reported to restrict N-MLV infection [15] . We generated CRFK cell lines stably expressing either wild-type or mutant FLAG-tagged versions of rhesus TRIM5α , and the presence of the different rhesus TRIM5α proteins was detected by Western blot ( Figure 6A ) . To measure the retroviral restriction activities of the rhesus TRIM5α wild-type , K10R , and SIM mutants , populations of CRFK cells expressing these various proteins were infected with increasing doses of N-MLV luc , and the cultures were analyzed for luciferase activity . Wild-type and SIM3mut TRIM5α restricted N-MLV . Mutant K10R showed modestly reduced restriction as compared to the wild-type TRIM5α . Most importantly , SIM1mut and SIM2mut rhesus TRIM5α were completely unable to mediate restriction of N-MLV ( Figure 6B ) . These results suggest that SIM1 and SIM2 are important for restriction activity of different TRIM5α orthologs . When expressed from transgenes , TRIM5α has been reported to form large cytoplasmic bodies [11] , [59] or to exhibit a diffuse reticular pattern in the cytoplasm [32] . Several groups have demonstrated full retroviral restriction activity of TRIM5α in the absence of detectable cytoplasmic bodies [24] , [32] , [59] , while others have shown the bodies to be highly dynamic structures that are key intermediates in the restriction process [60] , [61] . We wondered if the mutations we introduced in TRIM5α could be altering its subcellular localization and if this possible change was the cause of the defective restriction observed . The CRFK cell lines expressing the TRIM5α wild-type or the different mutants were examined by immunofluorescence using a confocal microscope . In all cases , a punctate cytoplasmic staining pattern was observed with no difference between wild-type and the different mutant proteins ( Figure S4 ) . Therefore , the mutations introduced on TRIM5α are not altering its subcellular localization . The above results show that SIM1 and SIM2 in TRIM5α are required for restriction activity in the 293T SUMO-1-overexpressing and MDTF and CRFK TRIM5α overexpressing cell lines . We wondered if TRIM5α SIMs are required for binding SUMO-modified CA or another SUMO-modified cellular factor . First we tested if TRIM5α is able to bind SUMO-1 . We transiently overexpressed FLAG-TRIM5α wild-type in 293T cells , and cellular lysates were used for a GST-pull down assay with GST or GST-SUMO-1 fusion proteins produced and purified from bacteria . We found that although weakly , TRIM5α is in fact , able to interact with SUMO-1 ( Figure 7A ) . Our laboratory has previously described that Mo-MLV CA interacts with the SUMO E2 ligase UBC9 , and undergoes SUMO-1 conjugation . We have identified the domain of CA that interacts with UBC9 and PIAS4/y , and those lysine residues present in this domain and nearby that were required for CA SUMO conjugation [54] . To test whether these features of CA were involved in SUMO-1 restriction , we generated a series of mutant versions of N-MLV in which CA lysine residues in and near the UBC9 interaction were mutated to arginine ( Figure 7B ) , analogous to mutations previously tested in the context of Mo-MLV . Although all the N-MLV luc mutant viruses showed reduced infectivity , as was reported for the equivalent mutants of Mo-MLV [54] , they retained sufficient activity to allow us to infect cells , score for luciferase activity and observe the difference between restrictive and non restrictive cells ( Figure S5 ) . We infected both the empty vector control and HA-SUMO-1-overexpressing 293T cell lines with various mutant N-MLV luc viruses at a range of multiplicities and calculated the fold restriction as the ratio between the luciferase activities of these cultures ( Figure 7C and S5A ) . As previously observed , wild type N-MLV was restricted an average of 8-fold by SUMO-1 overexpression . The K193R and K201 , 203R N-MLV mutants were still restricted , but the restriction was reduced to 5- and 4-fold respectively . Strikingly the K218R and K201 , 203 , 218R mutant viruses were not significantly restricted ( 1 . 5 and 1 . 8 fold restriction respectively ) by SUMO-1 overexpression . This result indicates that CA mutations altering the UBC9 interaction site , and lysines required for SUMO conjugation reduced the SUMO-1 block to N-MLV infection . Therefore , SUMO conjugation of CA is required for the SUMO-1-enhanced TRIM5α restriction activity . To determine the effect of CA mutations on TRIM5α restriction when SUMO-1 is not overexpressed , we infected the MDTF empty vector and TRIM5α-overexpressing cell lines with either wild-type or CA mutant versions of N-MLV luc and measured the luciferase activity in culture lysates . The fold restriction was calculated as the ratio between the luciferase activities observed upon infection of the empty vector and TRIM5α cell lines . MTDF cells expressing TRIM5α restricted wild-type N-MLV by an average of 166 fold over the control cells . The K193R and K201 , 203R mutants were also profoundly restricted and had no significant reduction of restriction ( p>0 . 1 ) when compared to wild-type virus . In contrast , the restriction of the K218R and K201 , 203 , 218R mutants was significantly , though only modestly , reduced ( p<0 . 01 ) when compared to wild-type virus ( 71 and 50 fold respectively ) ( Figure 7D ) . Thus , mutation of the major UBC9 interaction site ( K218 ) and the lysines required for SUMO conjugation of CA ( K218 , K201 and K203 ) reduced TRIM5α antiviral activity . These results , together with the evidence that the TRIM5α SIM1 and SIM2 are required for antiviral activity against N-MLV suggests that at least a portion of the antiviral activity of TRIM5α is mediated through the binding of its SIMs to SUMO-conjugated CA .
In this study , we have identified the involvement of the SUMO conjugation machinery in the TRIM5α mediated restriction of N-MLV . Our findings indicate that overexpression of SUMO-1 in 293T and TE671 cells enhances an intrinsic block to N-MLV infection of human cell lines ( Figure 1C and 2C ) . We show that this enhanced block is dependent on virus tropism ( Figure 3A ) , occurs before reverse transcription ( Figure 3C ) , and can be abrogated by pre-infection with restricted virus ( Figure 3B ) . These characteristics are shared by TRIM5α restriction of N-MLV in human cells , and by RNAi-mediated knockdown of TRIM5α in the SUMO-1 overexpressing cells , we confirmed that the enhanced block of N-MLV infection is due to TRIM5α activity ( Figure 4C and D ) . TRIM5α orthologues of primate and non-primate species have restrictive activity against a variety of retroviruses [62]–[65] . The four domains ( RING , B-box , coiled-coil and B30 . 2 ) of TRIM5α have been extensively studied in efforts to elucidate their functions in retroviral restriction [21] , [25] , [27] , [32] , [66]–[68] . These studies have revealed the importance of both the RING and B30 . 2 domains of rhesus TRIM5α in the inhibition of HIV-1 infection [11] and the requirement of the RING , B-box and B30 . 2 domains of human TRIM5α in the inhibition of N-tropic MLV [32] . TRIM5α has been found to bind to the retroviral capsid via its B30 . 2 domain [25] , [26] , [58] . Several groups have proposed that TRIM5α leads to an acceleration of the viral uncoating process , which results in deleterious disassembly of the capsid structure and exposure of the viral RNA to destructive cellular factors [26] , [64] , [69] . There is also evidence that suggests a role for ubiquitination and proteasome-mediated degradation of CA in TRIM5α-mediated restriction [70] , [71] . There has been no report indicating a relationship between the SUMO conjugation pathway and TRIM5α restriction activity . We consider that TRIM5α SUMO-1 conjugation could somehow enhance TRIM5α restriction activity . However , we have not been able to demonstrate SUMO-1 modification of TRIM5α , and in addition our data indicates that mutation of a potential SUMO conjugation site does not affect TRIM5α antiviral activity either in the 293T HA-SUMO-1 cell line ( figure 5C ) or in the TRIM5α-expressing MDTF ( figure 5E ) or CRFK cells ( figure 5F ) , indicating that this is not the likely mechanism of enhanced antiviral activity of TRIM5α upon SUMO-1 overexpression . We also considered the possibility that SUMO-1 overexpression impacts TRIM5α expression levels . Although we cannot eliminate this possibility for the endogenous protein , an overexpressed FLAG-tagged TRIM5α had very similar levels of expression in control and HA-SUMO-1 overexpressing cells ( Figure S1A ) , which argues against an enhancement of TRIM5α protein levels as the mechanism of increased activity upon SUMO-1 overexpression . The ubiquitin and SUMO pathways have been described as having an antagonistic relationship , but further studies have revealed a more complex interplay between these pathways . There are multiple reports that SUMO conjugation can act as a signal to recruit E3 ubiquitin ligases , leading to proteasome-mediated degradation of the modified protein . To date , the so-called SUMO-targeted Ubiquitin ligase ( STUbL ) family has been described in yeast , Dictyoselium , Drosophila , and mammals . Notably , all known STUbL proteins have a RING finger domain and an active SIM , which mediates the noncovalent binding to SUMO [72] , . Analysis of the TRIM5α protein sequence revealed three potential SIMs in the B30 . 2 domain . We found that mutation of the first two hydrophobic residues of SIM1 and SIM2 to lysine had dramatic effects on TRIM5α restriction of N-MLV . SIM1mut and SIM2mut versions of TRIM5α were not able to fully restore TRIM5α antiviral activity in the HA-SUMO-1/shRNA4 overexpressing cell line ( Figure 5C ) and had absolutely no restriction activity in MDTF and CRFK cells ( Figure 5E and 5F respectively ) . The same was observed in the rhesus TRIM5α overexpressing cells ( Figure 6B ) . Mutation of SIM3 had a more moderate effect on TRIM5α restriction; cells expressing SIM3mut were still able to restrict N-MLV , but not as effectively as the wild-type protein . These results are consistent with a model in which TRIM5α activity is partially mediated by SIMs binding to SUMO-modified CA . If TRIM5α is a novel STUbL , it would explain our observations of the SUMO-1 antiviral effects in 293T cells and the current model of a proteasome-dependent TRIM5α-mediated restriction . SUMO-1 can be conjugated to Mo-CA in vivo [54] , and the amino acid sequence of the SUMO conjugation site in N-CA is identical to that of Mo-CA . We can speculate that SUMO conjugation to N-CA facilitates or stabilize TRIM5α binding via the SIMs . With multiple SIMs , TRIM5α could either bind multiple SUMO-1 conjugated viral CA proteins present in the incoming virus , or one CA molecule with multiple SUMO modifications . Once TRIM5α binds to the SUMO-CA , the RING domain of TRIM5α could activate the proteasome-mediated degradation of CA or other viral proteins associated with it , leaving the viral RNA unprotected and vulnerable to cellular factors . It is also possible that the mere binding of TRIM5α to the SUMO-modified CA is sufficient to interfere with the viral life cycle , which could explain why in the SUMO-1 overexpresing cell line , a RING domain mutant is able to complement as efficiently as the wild-type protein the restriction observed upon SUMO-1 overexpression . TRIM5α belong to a large family of TRIM proteins that were originally observed to oligomerize into high-order structures localizing to specific compartments in the cytoplasm and nucleus [75] . TRIM19 , also known as PML , forms the so-called PML nuclear bodies , which are important host antiviral defense against DNA virus ( for review see [76] ) . PML was shown to be SUMOylated on 3 lysine residues and also contain a SIM . Accordingly intramolecular interactions between the PML SUMO and SIM were proposed to underlie PML nuclear bodies formation and recruitment of partners [77] , [78] . Although an appealing model , specific PML isoforms that do not contain the SIM are able to form normal bodies [79] and still have antiviral activity [80] . When expressed from transgenes , TRIM5α has been reported to form large cytoplasmic bodies [11] , [59] . We observe that TRIM5α mutants that lack the SIM are still able to form cytosolic bodies , which is similar to the case of the PML protein that do not contain a SIM , but in our case this TRIM5α mutant proteins lack the antiviral activity . TRIM5α bodies have been described as highly dynamic structures that interact with cytoplasmic viral complexes [60] , [81] , and it could be possible that the mutations we have introduced are affecting the dynamics of these structures , making them incapable to get to the viral complexes . Also , if TRIM5α cytosolic bodies are highly dynamic structures , the presence of the SIMs may stabilize the binding to the incoming viral cores , which are primarily recognized through the N tropism determinants ( arginine 110 ) on CA . It could be interesting in further studies analyze the localization of TRIM5α SIM mutants and viral complexes upon infection . Our data showing that viruses with CA mutations in the SUMO conjugation site and nearby lysines are not fully restricted support the theory that TRIM5α SIMs bind SUMO-CA . More importantly , mutant viruses containing the K218R mutation , which removes the lysine in the major UBC9 interaction site , are poorly restricted relative to wild-type virus in some settings ( Figure 7C and 7D ) . While these mutants are still restricted in the MDTF-TRIM5α cells ( Figure 7D ) , this could be due to other lysines acting as acceptor sites for SUMO conjugation , and to the high levels of TRIM5α expression . Portions of TRIM5α in the B30 . 2 domain have been described as important for binding CA [26] , [58] , and TRIM5α may be able to bind CA molecules with reduced efficiency in the absence of SUMOylation . The SUMO-conjugation sites of CA are located in the C-terminal domain of the protein , and upon maturation the C-terminal domain of CA migrates to fill the interstitial spaces between hexamers in the N-terminal domain layer , in the current model of MLV CA structure ( Burns and Goff unpublished data and [82] ) . This rearrangement of CA could allow the SUMOylated lysines of CA to be exposed in the surface of the incoming virus . It is possible that the TRIM5α protein can recognize the arginine residue 110 of N-CA and the SUMOylated lysines at the same time or that different TRIM5α molecules can independently recognize each feature . Unfortunately the exact location of the SUMO-conjugation sites of CA in the mature lattice is still not available . Earlier work has suggested that the SUMO modification of CA is important for successful infection: MLV mutant viruses with alterations in the UBC9 interaction site or mutations that abolish SUMO conjugation are blocked in an early step of infection [54] . If SUMO conjugation of CA is an important step during viral infection , is likely that innate immunity might take advantage of this process to restrict viral infection . The presence of SIMs in TRIM5α could be an adaptative advantage of this protein for its antiviral activity . SIM1 and SIM2 , which do have a role in human and rhesus TRIM5α restriction of N-MLV , are located outside of the variable regions , while SIM3 which does not have a role on restriction activity is inside of the variable region 3 . The location of SIM1 and SIM2 suggest that they are important for TRIM5α function and in fact several TRIM5α orthologs contain the SIMs present in the human protein ( Figure S3 ) . In conclusion , our data show that the SIMs of TRIM5α are important for restriction of N-MLV , and that their mutation abolishes antiviral activity . The presence of an intact UBC9 interaction site and lysines that potentially can be SUMO modified in N-CA are required for full restriction by human TRIM5α . We propose that TRIM5α SIMs increase affinity of recognition of SUMO-modified CA . Further analyses are required to determine if TRIM5α is indeed restricting virus as a STUbL .
Human embryonic fibroblast ( 293T ) , Mus dunni tail fibroblast ( MDTF ) , human medulloblastoma cell line TE671 and Crandall feline kidney ( CRFK ) fibroblast were maintained in Dulbeccos's modified Eagle medium supplemented with 10% fetal bovine serum , 100 UI/ml penicillin and 100 mg/ml streptomycin . All cells were cultured at 37°C in 5% CO2 . pCIG3-N and pCIG3-B express gag and pol from N-MLV and B-MLV respectively [83] . pCMVI expresses gag and pol from NB-MLV . p8 . 91 encodes gag and pol of HIV-1 . pMD . G expresses the vesicular stomatitis virus envelope glycoprotein [21] . pFBLuc ( Stratagene ) is a reporter plasmid containing the firefly luciferase coding sequence flanked by MLV-based LTRs . pCNCG is a CMV-driven reporter plasmid containing the green fluorescent protein ( GFP ) coding sequence flanked by MLV-based LTRs . pQCXIH Retroviral vector ( Clontech ) is a bicistronic expression vector that expresses an inserted gene along with the hygromicin selection marker . pQCXIP Retroviral vector ( Clontech ) is a bicistronic expression vector that expresses an inserted gene along with the puromycin selection marker . pcDNA3xFLAG is a CMV-driven expression vector that allows expression of N-terminal FLAG epitope proteins [84] . SUMO-1 was subcloned from pSG5 His-SUMO-1 ( gift from Dr . Anne Dejean of the Institut Pasteur ) . SUMO-2 and SUMO-3 were subcloned from pCDNA4 HisMaxC-SUMO-2 or HisMaxC-SUMO-3 ( gifts from Dr . Yoshiaki Azuma , The University of Kansas ) into pQCXIH , such that the His tag was replaced with the HA epitope . Human and rhesus TRIM5α were subcloned from pMIP-TRIM5α ( gift from Dr . Jeremy Luban ) [85] into pcDNA3xFLAG . Human TRIM5α K10R and SIMs substitution mutants were generated by two-step overlapping PCR and cloned into pcDNA3xFLAG . Human TRIM5α wild type , K10R and SIMs mutants were subcloned from pcDNA3xFLAG-TRIM5α vectors into pQCXIP , such that the N-terminal FLAG epitope was replaced with a C-terminal Myc epitope . FLAG rhesus TRIM5α wild type , K10R and SIMs mutants were subcloned from pcDNA3xFLAG-TRIM5α vectors into pQCXIP . The pCIG3-N mutant versions CA R110E , CA K193R , CA K201 , 203R , CA K218R and CA K201 , 203 , 218R were generated by site-directed mutagenesis using Quick Change Lightning kit ( Stratagene ) . Note: All primer sequences are available upon request . Retroviruses for transduction were produced by transfection of 293T cells with 1 µg pMD . G , 1 µg pCMVI and 1 . 5 µg of either pQCXIH , pQCXIH-HA-SUMO-1 , pQCXIP , pQCXIP-TRIM5α wild-type or mutant versions , using FUGENE ( Roche ) . Viruses were harvested 48 h after transfection , filtered ( 0 . 45 µm ) and used to infect 5×105 cells in 100 mm dishes in the presence of 8 µg/ml polybrene . 293T , MDTF and TE671 cells infected with vectors delivering the hygror gene were selected in 200 µg/ml hygromycin . Cells infected with vectors containing the puror gene were selected either in 5 µg/ml puromycin ( MDTF cells ) or in 7 . 5 µg/ml puromycin ( CRFK cells ) . Lentiviruses for transduction were produced by transfection of 293T cells with 1 µg pMD . G , 1 µg p8 . 91 and 1 . 5 µg of pGIPz ( Open Biosystems ) or pGIPzTRIM5α DNAs containing shRNAs #1 to #5 ( Open Biosystems ) . Viruses were harvested 48 h after infection , filtered ( 0 . 45 µm ) and used to infect 5×104 293T HA-SUMO-1 cells in 35 mm dishes in the presence of 8 µg/ml polybrene . Cells were selected in 200 µg/ml hygromycin and 1 . 5 µg/ml puromycin . Cells were lysed in 20 mM Tris-HCl ( pH 8 . 0 ) , 137 mM KCl , 10% glycerol , 1% NP-40 and Complete protease inhibitor ( Roche ) or Reporter lysis buffer ( Promega ) . Samples were then boiled in 5× sodium dodecyl sulphate ( SDS ) loading buffer , and the proteins were resolved by SDS-polyacrylamide gel electrophoresis ( PAGE ) . After transfer to nitrocellulose membranes , the blots were probed with mouse anti-β actin ( Sigma ) , mouse anti-HA ( Covance ) , mouse anti-FLAG ( Sigma ) or mouse anti-c-Myc ( Santa Cruz Biotechnology ) . B- , NB- and N-tropic luciferase reporter viruses were produced by transfection of 293T cells with 1 µg pCIG3-B or pCMVI or pCIG3-N ( the wild type or mutant versions ) , 1 µg pMD . G and 1 . 5 µg pFBluc or pCNCG ( per 100 mm plate ) using FUGENE ( Roche ) . Reporter virus stocks were harvested 48 h after transfection , filtered ( 0 . 45 µm ) and stored at −80°C . 293T ( 3×104 per well ) , MDTF ( 2 . 5×104 per well ) , TE671 ( 2 . 5×104 per well ) and CRFK ( 3×104 per well ) cells were seeded in 24-well plates and infected with MLV luc reporter viruses . For reactions involving transient transfections , cells were transfected with 100 ng of pcDNA3xFLAG or pcDNA3xFLAG-TRIM5α wild type or mutant versions . Twenty-four hours post-transfection , cells were infected with N-MLV luc virus . Forty-eight hours post-infection cells were collected and assayed for firefly luciferase activity ( Promega ) in a luminometer . Cells 293T ( 1×105 ) plated in 35 mm dishes were infected with N-MLV luc for six hours . Twenty-four hours post-infection , cells were trypsinized , pelleted and total DNA was collected using DNeasy Qiagen kit ( Qiagen ) . Quantitative PCR ( qPCR ) analysis was performed using primers to amplify the minus-strand strong stop ( MSS ) , 2-LTR circles ( LTR-LTR junction ) and luciferase DNA as previously described [86] . Cells were harvested and total RNA was extracted using TRIZOL reagent ( Invitrogen ) . 2 µg of total RNA per cell line was used for reverse transcription reactions to produce cDNA using random hexamers and SuperScript III kit ( Invitrogen ) . 2 µl of each cDNA was used for qPCR analysis of TRIM5α and GAPDH transcript levels . Fold change was calculated using the relative standard curve method . GST and the fusion protein GST-SUMO1 were produce in Escherichia coli BL-21 ( DE3 ) as previously described [87] . 293T cells were transfected with 3 µg of pcDNA3xFLAG-humanTRIM5α or the empty vector . Forty-eight hours after transfection the cells were lysed on 20 mM Tris-HCl ( pH 8 . 0 ) , 137 mM KCl , 10% glycerol , 1% NP-40 and Complete protease inhibitor ( Roche ) 20 minutes at 4°C , the lysate was clarified by centrifugation at 13000×g for 10 minutes at 4°C . GST-pulldown assays were performed using 100 µl of cell lysate and 2 µg of purified GST or GST-SUMO1 and 20 µl of Glutathione Sepharose 4B ( Amersham biosciences ) in binding buffer ( 20 mM Tris-HCl ( pH 8 . 0 ) , 100 mM KCl , 10 mM EDTA , 0 . 5 mM DTT , Complete protease inhibitor ) for 2 hours , the beads were washed 4 times with wash buffer ( 20 mM Tris-HCl ( pH 8 . 0 ) , 100 mM KCl , 0 . 2% NP-40 10 mM EDTA , 0 . 5 mM DTT , Complete protease inhibitor ) , the beads were resuspended with 40 µl of GLB2x , and the proteins were resolved by SDS-PAGE . After transfer to nitrocellulose membranes , the blots were probed with mouse anti-FLAG ( Sigma ) or mouse anti-GST ( Covance ) . SUMO-1: AACC5096 . 1 SUMO-2: P61956 . 2 SUMO-3: NP_008867 . 2 MoMLV: AF033811 N-tropic MLV gag-pol region: K01203 . 1 B-tropic MLV gag-pol region: K01204 . 1 TRIM5α: H . sapiens ABB90543 , M . mulatta NP_0010228082 | TRIM5α is an intrinsic immunity protein that provides a post-entry block of retroviral infection , which depends on its specific ability to recognize retroviral capsid ( CA ) . Human TRIM5α is able to recognize and block infection by N-tropic murine leukemia virus ( N-MLV ) as well as other viruses . The exact mechanism by which TRIM5α exerts its action is still controversial . In this study , we have identified a new aspect of TRIM5α-mediated restriction of N-MLV: the involvement of the SUMO conjugation machinery . SUMO conjugation of MLV CA is an important step during viral infection , and our data suggest that innate immunity takes advantage of this process to restrict viral infection . We show that human TRIM5α protein contains two SUMO interacting motifs ( SIMs ) that are required for its antiviral activity against N-MLV . We propose that at least a portion of the antiviral activity of TRIM5α is mediated through the binding of its SIMs to SUMO-conjugated CA . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"molecular",
"cell",
"biology",
"virology",
"genetics",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"genetics",
"and",
"genomics"
] | 2011 | SUMO-Interacting Motifs of Human TRIM5α are Important for
Antiviral Activity |
Metachondromatosis ( MC ) is a rare , autosomal dominant , incompletely penetrant combined exostosis and enchondromatosis tumor syndrome . MC is clinically distinct from other multiple exostosis or multiple enchondromatosis syndromes and is unlinked to EXT1 and EXT2 , the genes responsible for autosomal dominant multiple osteochondromas ( MO ) . To identify a gene for MC , we performed linkage analysis with high-density SNP arrays in a single family , used a targeted array to capture exons and promoter sequences from the linked interval in 16 participants from 11 MC families , and sequenced the captured DNA using high-throughput parallel sequencing technologies . DNA capture and parallel sequencing identified heterozygous putative loss-of-function mutations in PTPN11 in 4 of the 11 families . Sanger sequence analysis of PTPN11 coding regions in a total of 17 MC families identified mutations in 10 of them ( 5 frameshift , 2 nonsense , and 3 splice-site mutations ) . Copy number analysis of sequencing reads from a second targeted capture that included the entire PTPN11 gene identified an additional family with a 15 kb deletion spanning exon 7 of PTPN11 . Microdissected MC lesions from two patients with PTPN11 mutations demonstrated loss-of-heterozygosity for the wild-type allele . We next sequenced PTPN11 in DNA samples from 54 patients with the multiple enchondromatosis disorders Ollier disease or Maffucci syndrome , but found no coding sequence PTPN11 mutations . We conclude that heterozygous loss-of-function mutations in PTPN11 are a frequent cause of MC , that lesions in patients with MC appear to arise following a “second hit , ” that MC may be locus heterogeneous since 1 familial and 5 sporadically occurring cases lacked obvious disease-causing PTPN11 mutations , and that PTPN11 mutations are not a common cause of Ollier disease or Maffucci syndrome .
Cartilage tumor syndromes are characterized by multiple cartilaginous bone tumors that develop in childhood , often causing significant morbidity and predisposing to chondrosarcoma . Tumors can form as exostoses ( on the surface of bone ) , as in the autosomal dominant , multiple osteochondroma ( hereditary multiple exostoses ) syndromes ( MO; MIM 133700 and 133701 ) , or as endosteal tumors ( within bone ) , as in the sporadically occurring multiple enchondromatosis disorders ( MIM 166000 ) Ollier disease and Maffucci syndrome . In MO , mutations in EXT1 or EXT2 , which encode heparan sulfate glycosyltransferases , affect chondrocyte orientation in the growth plate [1] . A small percentage of patients with Ollier syndrome have mutations in PTH1R , which encodes the receptor for parathyroid hormone and parathyroid hormone-related protein , causing altered chondrocyte differentiation in the growth plate [2] . The cause of Maffucci syndrome is unknown [3] . Patients with MO do not develop endosteal tumors , and patients with Ollier disease or Maffucci syndrome do not develop exostotic tumors [1] , [3] , [4] . Patients with metachondromatosis ( MC; MIM 156250 ) form exostotic and endosteal tumors ( Figure 1 ) . Fewer than 50 cases of MC have been published since Maroteaux's initial description in 1971 [5] . Exostotic lesions in MC occur frequently in the digits , involve metaphyses and epiphyses , and tend to grow toward the joint; in contrast , exostotic lesions in MO occur frequently in the long bones , involve only the metaphyses , and tend to grow away from the joint [6]–[11] . MC exostotic lesions can also spontaneously decrease in size and completely regress [6] , [7] , [9] , [12] . Endosteal lesions in MC are common in the metaphyses of long bones and in the pelvis [7]–[11] . Avascular necrosis of the femoral head , due to endosteal tumors , has been a frequent complication in patients with MC [7] , [8] , [13]–[15] . Hand deformity due to endosteal tumors is uncommon in patients with MC , whereas it is often a significant problem for patients with Ollier disease and Maffucci syndrome [3] . Finally , malignant transformation has only been reported in one patient with MC , whereas it has been more frequently reported in patients with MO , Ollier disease , and Maffucci syndrome [3] , [4] , [16] . The distinct distribution and clinical behavior of lesions in patients with MC , suggest that MC is pathophysiologically distinct from these other cartilage tumor syndromes . We therefore sought to better characterize MC and to determine its genetic basis .
We diagnosed participants as having MC based upon the presence of both multiple exostotic and endosteal cartilaginous lesions as previously described [5]–[10] , [15] . We excluded from analysis participants with solitary lesions , contiguous endosteal lesions suggestive of Ollier disease , soft tissue lesions suggestive of Maffucci syndrome , or radiographs suggestive of MO . We included participants who had clinical and radiographic features of MC , even if they lacked a positive family history . For each patient , the clinical history and radiographs were reviewed by at least 3 authors . MC patients from 17 unrelated families from 9 countries were identified ( Supplementary Table 1 ) . All participants gave their informed consent following the guidelines of each referring institution . In 10 families disease segregation is consistent with autosomal dominant inheritance . In 7 families , the disease is suspected to have arisen de novo . Immediate family members of patients with sporadically occurring MC were interviewed and examined , although detailed imaging was not performed . For 8 familial cases , blood or DNA was available from additional family members . Figure 1 depicts features seen in affected participants with MC . No phenotypic differences could be found between sporadic or familial cases of MC . Radiographs identify exostotic and endosteal lesions of the digits ( Figure 1A–1C ) and long bones ( Figure 1E–1H ) , along with degenerative hip disease secondary to endosteal lesions in the femoral neck ( Figure 1I ) . Spontaneous regression of exostotic lesions is seen in radiographs obtained 10 years apart in the same patient ( Figure 1F , 1G ) . Also depicted in Figure 1 are histopathologic features that distinguish exostoses in patients with MO from those in patients with MC , based upon a comparison of 30 exostoses excised from children with MO and 15 exostotic lesions excised from 3 affected individuals with MC . Exostoses in children with MO have cartilage caps with endochondral bone growth immediately beneath the cap ( Figure 1J ) . In contrast , exostoses in children with MC have a predominantly fibrous cap and a core of disorganized cartilage surrounded by trabecular bone ( Figure 1K , 1L ) . In all MC cases , the lesions were bilateral and not obviously confined to a single body segment as in Ollier or Maffucci patients . We performed linkage analysis in the largest family ( Family A , Figure 2A ) to identify a genetic locus for MC . Raw genotype data were generated using Affymetrix 6 . 0 SNP arrays and multipoint parametric linkage analysis of the autosomal genome was performed using MERLIN [17] . Because non-penetrance and non-ascertainment are potential confounding factors in the diagnosis of MC , we analyzed only founders and affected individuals ( Figure 2A ) . Although this limited the maximum attainable LOD score to 2 . 7 , which is lower than the genome-wide significance threshold of 3 . 3 , the dense marker set ensured a reasonable probability that only one large interval that achieved the maximum LOD score would be observed , with the remainder of the genome being excluded . We identified a single interval on chromosome 12 , from 121 . 5 to 132 . 3 cM , that attained the maximum LOD score of 2 . 7 ( Figure 2B ) . No other autosomal interval >1 cM yielded a peak LOD score >−1 . 9 . Several intervals <1 cM attained LOD scores >0 ( Figure S1 ) , but we considered these unlikely to be candidate intervals and instead assumed they represented either unfiltered genotyping errors or short ancestral population haplotypes , rather than familial haplotypes inherited from a common ancestor . We next performed array-based capture , followed by Illumina GAII sequencing , using bar-coded DNA libraries created from 16 individuals in 11 families . We sequenced 1 affected individual in 8 families , 2 affected individuals in 1 family , and 2 affected and 1 unaffected individual in 2 families ( Table S1 ) . We prepared bar-coded genomic DNA libraries , having an average insert size of 150 bp , using sheared DNA from each of the 16 individuals ( Figure S2 ) . We performed array-based targeted capture by pooling each DNA library and hybridizing the pooled sample to an Agilent Technologies 1M SureSelect DNA capture array containing 973 , 952 probes targeting 844 , 339 bp within the 8 . 6 Mb candidate interval , including 88 . 4% and 98 . 6% of UCSC exons and CCDS coding sequence , respectively . After hybridization and elution , the captured DNA was PCR amplified , purified to remove primer dimers , and sequenced on two lanes of an Illumina GA II . We obtained 50 million , 80 bp , single-end reads . Novobarcode software was used to sort the reads according to their 3 bp barcode , and Novoalign was used to align the reads to the reference genome ( hg19 ) . We obtained between 1 and 6 million reads for each individual . Among individuals , an average of 61% ( ±2% ) of the aligned reads mapped to regions targeted by the capture array ( Figure S3A ) . Of the bases targeted by the capture array , 75% ( ±7% ) had a read depth of at least 5× , which diminished to 55% ( ±14% ) after the removal of PCR duplicates ( Figure S3B ) . Twenty percent of targeted bases were not captured . In the 3 families for which pairs of affected family members were sequenced , total filtered variants in the candidate interval ( 388–1499 per individual ) were analyzed to find variants shared by both affected individuals from the same family ( Table 1 ) . In all 3 families , frameshift mutations in exon 4 of PTPN11 were the only novel coding variants present in both affected family members and , for Families A and B , absent in the unaffected individual . Family A had a 5 bp deletion , Family B had a more complex deletion/insertion , and Family C had a 2 bp deletion ( Table S1 ) . In the remaining 8 families for which only 1 affected individual per family was sequenced , there were 18 novel coding variants present in ≥3 reads , one of which was a nonsense mutation in exon 13 of PTPN11 ( p . Q506X ) ( Figure S4 ) . We used Sanger sequence analysis of PCR amplimers to demonstrate that affected family members from these 4 families had PTPN11 mutations , and that unaffected family members lacked PTPN11 mutations . Sanger sequence analysis of the 15 coding exons of PTPN11 in the 7 families for whom we had not found mutations by array capture and Illumina-sequencing detected a 1 bp deletion in exon 11 in 1 of the 7 families ( Family D ) . This deletion was within a 98 bp segment that had been targeted but not captured in any of the DNA samples . Another family ( F ) had a splice-acceptor site mutation ( AG>CG ) in intron 5 in 2 affected siblings , but not in either parent . The siblings' mother was clinically affected with MC , although less severely than her children . The mother was the first in the family to have MC and was the only member of the family who was included in the Illumina sequencing . The site of the splice-site mutation identified in her children was covered 25× in her DNA sequence and was always wild-type , as were her Sanger sequence results . These data suggest the mother is mosaic for a PTPN11 mutation and that the family's mutation would have been found by Illumina sequencing had we initially sequenced her children's DNA . We subsequently collected DNA from an additional 6 MC families . Sanger sequence analysis revealed a nonsense mutation involving exon 3 ( p . K99X ) in one family ( I ) , and a splice site mutation in intron 9 ( c . 1093-1G>T ) in another family ( G ) . In total , we found PTPN11 mutations in 10 of 17 families . Five mutations were frameshift , 2 were nonsense , and 3 disrupted a splice-acceptor site ( Figure 3A , Table S1 ) . Each family had a different mutation and mutations were scattered across the gene ( Figure 3A ) . In two families without mutations we had performed aCGH and did not find evidence of PTPN11 intragenic deletions or duplications ( Table S1 ) . We did not have other family members' DNA samples from the one familial MC patient who lacked a PTPN11 mutation to be able to test for locus heterogeneity by linkage analysis . To test whether the patients without PTPN11 coding mutations had noncoding mutations in PTPN11 or had mutations in other genes , we designed a second Agilent 1M capture array . Firstly , we included probes to target the entire PTPN11 gene , excluding Alu repeats . Secondly , we included probes targeting the exons of 74 genes that function in the same pathways as PTPN11 , including the Ras/MAPK and PI3K/Akt pathways ( Table S2 ) . Thirdly , we included probes targeting the exons of the MO genes EXT1 and EXT2 , to determine whether any of our patients lacking mutations and classic radiographic features of MC had been misdiagnosed . Barcoded genomic libraries for an individual from each of the 7 MC families without PTPN11 coding mutations and from 2 families originally referred with MC , but whose radiographic features were more consistent with MO , were pooled and hybridized to the capture array . The captured DNA was then sequenced using two lanes of Illumina GAII 42 bp single end sequencing . For each barcoded sample , 4 . 3 ( ±1 . 2 ) million reads were obtained , of which 37% ( ±7% ) mapped to regions targeted by the array ( Figure S3A ) . Of the bases targeted by the array , 85% ( ±6% ) were covered by a read depth of at least 10× , which dropped to 83% ( ±8% ) after the removal of PCR duplicate reads ( Figure S3B ) . Identified variants with a quality score >20 were filtered to remove SNPs listed in the SNP database ( version 132 ) and the 1000 genomes project ( Nov . 2010 release ) . No PTPN11 coding mutations were found . We then analyzed the noncoding regions and identified one 3′ UTR mutation , and 7 intronic mutations , of which 6 were in LINE elements or other repetitive regions ( Table S3 ) . All intronic mutations were at least 300 bp from the nearest exon and , using an online splice prediction tool ( http://genes . mit . edu/GENSCAN . html ) , were not predicted to alter splicing . We then analyzed variants identified in the exons of the 76 other genes included in the capture array ( Table S2 ) . No nonsense , frameshift or splice site mutations were identified . Of the missense mutations that were present in more than 4 independent sequencing reads , 4 were nonsynonomous ( ERBB2 p . S1050L , MTOR p . P1408S , MVP p . R49S , SOS2 p . D952N ) and 4 were synonomous ( PIK3C2B p . D478D , RAF1 p . L351L , MAP2K2 p . D140D , MVP p . T199T ) . Further experiments will be needed to determine if any of the novel noncoding PTPN11 mutations or novel variants in the other genes are disease causing . We next analyzed the sequencing read depth across the PTPN11 locus to detect deletions or duplications . In one individual ( Patient S ) , we identified an ∼15 kb region spanning exon 7 that contained half as many reads as would be expected based upon the read depths of the other patients included in the capture array ( Figure 3B ) . As expected , PCR primers that flank this 15 kb region failed to produce amplimers when wild-type genomic DNA was used as template . However , PCR amplification using genomic DNA from Patient S yielded an ∼700 bp PCR product and Sanger sequence analysis of this product indicated that 14 , 629 bp of genomic sequence ( chr12:112 , 897 , 487–112 , 912 , 115 ) had been replaced with a single CA dinucleotide ( Figure 3B ) . In addition , PCR amplification and sequencing of PTPN11 in peripheral blood cDNA from this patient , using a forward primer in exon 6 and a reverse primer in exon 8 , detected a mutant cDNA that lacked exon 7 ( data not shown ) . The loss of exon 7 results in a frameshift with introduction of a premature stop codon ( T253LfsX54 ) . The read depth of the remaining 76 genes targeted by the array was also analyzed to detect deletions or duplications . Two patients initially included in the study , but on radiographic review were felt more likely to have MO than MC , were found to have deletions involving EXT1 ( Figure S5 ) . In one patient ( Q ) , the first exon of EXT1 contained half as many reads over its 1 . 8 kb as expected ( Figure S5A ) . In a second patient ( N ) , all exons of EXT1 had half as many reads as expected ( Figure S5A ) . In additional to skeletal lesions , this patient has developmental delay , microcephaly and mild dysmorphism , suggesting a possible contiguous gene deletion syndrome . EXT1 deletions were confirmed in both patients by multiplex ligation-dependent probe amplification ( MLPA ) ( Figure S5B , S5C ) . Our finding of nonsense , frameshift , and splice-site mutations in multiple exons , as well as a large deletion , suggests that MC-causing PTPN11 alleles are loss-of-function . We tested this hypothesis by performing Western blots on whole protein extracts from white blood cells and from an excised exostotic lesion in a patient ( B-IV-7 ) with a PTPN11 frameshift mutation in exon 4 . An anti-SHP2 antibody that recognizes an epitope amino-terminal of the polypeptide encoded by the frameshifted exon detected only full-length , wild-type SHP2 protein ( Figure S6 ) . We next determined whether MC exostoses arise from a “second hit , ” similar to what has been observed in autosomal dominant MO [18] . We looked for a second hit in cells of the cartilage core of an MC lesion ( e . g . , Figure 1K ) by performing microdissection , PCR amplifying the mutation containing exon , and Sanger sequencing the amplimers . In tumors from two different patients ( A-IV-5 , A-IV-8 ) , with a 5 bp frameshift mutation in exon 4 , we observed a clear excess of mutant sequence versus wild-type sequence in the tumors' cartilage cores , as compared to the patients' peripheral blood and bone/marrow from the lesion ( Figure 4A ) . We quantified the amount of mutant versus wild-type sequence , by extracting DNA from the cartilage core of patient A-IV-8 , PCR amplifying exon 4 , and subcloning amplimers to determine the percent that contained the mutant allele . Forty-four of 52 individual subclones contained the mutant allele , which is significantly higher ( p<0 . 001 ) than expected for a heterozygous mutation . In contrast , 58% of subclones ( 34/59 ) from adjacent unaffected bone/bone marrow contained the mutant allele , which is not significantly different from the expected value of 50% ( p = 0 . 24 ) . These data are consistent with an MC exostosis arising from a second hit ( loss of the wild-type allele ) within a cell that ultimately contributes to the lesion's cartilage core . Because the mutant allele is 5 bp shorter than wild-type PTPN11 in these two patients , we tested for loss of heterozygosity at a second polymorphic site in PTPN11 to control for potential PCR bias in amplifying the exon with the deletion . In their peripheral blood DNA , patient A-IV-5 and her unaffected mother are heterozygous for a benign polymorphism in intron 11 of PTPN11 ( rs41279092 ) . The abundance of this SNP , which is in the wild-type PTPN11 allele , was markedly reduced in the lesion's cartilage core , again consistent with LOH occurring in the cell that drives formation of the cartilage core ( Figure 4B ) . We finally asked whether mutations in PTPN11 are associated with other cartilaginous tumor syndromes . We sequenced the coding exons of PTPN11 in 38 lesions excised from patients with Ollier disease , 2 peripheral blood samples from patients with Ollier disease , 15 lesions excised from patients with Maffucci syndrome , 4 solitary enchondromas , 9 chondrosarcomas ( 1 polyostotic ) , and 3 osteochondromas without EXT1 or EXT2 mutations . We did not find PTPN11 coding sequence mutations in any patient sample . In 24 percent of the samples we observed heterozygosity for noncoding SNPs that are known common variants , suggesting that large PTPN11 gene deletions and other causes of LOH are not frequently associated with these other cartilaginous tumors .
We identified 17 unrelated families with MC . Clinical features were similar to previously published cases [5]–[10] , [15] . The exostoses of MC had been assumed to be identical to the osteochondromas of MO; however , we demonstrate that they are histologically unique lesions with a large cartilaginous core ( Figure 1J–1L ) . We combined linkage analysis in a single MC family with DNA capture and parallel sequencing of bar-coded DNAs from several MC families to identify mutations in PTPN11 as a cause of MC . In MC patients without PTPN11 coding sequence and splice site mutations , we generated and pooled barcoded DNAs , and performed a second targeted capture that included the entire PTPN11 gene and exons from 76 other genes . This led us to detect an ∼15 kb deletion in a patient by analyzing the depth of sequencing reads ( Figure 3B ) . In total , we found likely disease-causing PTPN11 mutations in 11 of 17 families . Concurrent with our studies of MC , Sobreira et al . ( 2010 ) reported PTPN11 mutations in 2 MC families [11] . They performed whole-genome sequencing ( WGS ) in a single affected individual who was a member of family in which MC was segregating . This approach also required these investigators to include linkage data to reduce the number of novel potentially disease-causing heterozygous changes that are identified by WGS [19] , [20] . The investigators next identified an independently arising PTPN11 mutation in an unrelated patient to strengthen the evidence for causality . However , having only studied genomic DNA and finding frameshift mutations in the same exon ( exon 4 ) in their two unrelated patients , Sobreira et al . ( 2010 ) could not definitively determine the mechanism by which PTPN11 mutations cause MC . Missense mutations in PTPN11 have previously been identified in patients with Noonan , Noonan-like , and LEOPARD syndromes , as well as in juvenile myelomonocytic leukemia [21] . In these disorders , the mutations are gain-of-function and/or dominant negative for SHP2 , which is the PTPN11 protein product [22] , [23] . SHP2 is a protein tyrosine phosphatase and an important intracellular signaling molecule linking several growth factor receptors to the Ras/MAPK and other signaling pathways ( Reviewed in [24] ) . Therefore , frameshift mutations in exon 4 might also create an abnormal protein product by altering PTPN11 mRNA splicing . Alternatively , the frameshift mutations might result in loss-of-function because of nonsense mediated mRNA decay or rapid degradation of a truncated SHP2 polypeptide . Our finding of nonsense , frameshift , and splice-site mutations in multiple exons , as well as a whole-exon deletion , suggests that MC-causing PTPN11 alleles are loss-of-function . We tested this hypothesis by performing Western blots on whole protein extracts from white blood cells and from an excised lesion containing affected and unaffected tissue , and detected only full-length , wild-type SHP2 protein ( Figure S6 ) , confirming that the mutant alleles are loss-of-function . Exostoses in MO originate from the “second hit” mutations [18] . Mice with conditional alleles at the EXT1 locus demonstrate that only a few cells having two mutant alleles are sufficient to cause other cells to become misdirected and form an exostosis [25] . By performing microdissection , we found evidence for loss of the wild-type PTPN11 alleles in the majority of cells within the cartilage cores of exostoses from two MC patients ( Figure 4 ) , consistent with a “second hit . ” Recently , Bauler et al . used a ubiquitously expressed Ert2-Cre driver in 6–8 week-old Ptpn11 floxed mice to generate mice that were Ptpn11-null in multiple tissues . Among the consequences of completely deleting SHP2 was the appearance of ectopic cartilage islands in the animals' metaphyseal trabecular bone and growth plates [26] . These findings are consistent with the distribution of endosteal tumors and exostoses seen in patients with MC . The findings in mice with homozygous deletion of Ptpn11 contrast with the absence of skeletal lesions in mice that have heterozygous loss-of-function mutations [27] . We suspect that mice with heterozygous mutations have a much lower incidence of noticeable “second hits” compared to humans because they have fewer skeletal cells and shorter lifespans . Homozygous inactivation of Ptpn11 solely in mouse chondrocytes may be required to enable a detailed understanding of how SHP2 deficiency leads to tumorigenesis . We did not detect PTPN11 mutations in 6 of 17 patients with MC phenotypes , including 1 patient with a family history of MC and 5 patients who are the first affected members in their families . DNA is not available from other affected family members of the familial case to determine whether MC exhibits locus heterogeneity . Two patients with de novo disease did have DNA variants found in the 3′ UTR and/or in introns . None of these variants are in likely regulatory regions or in regions important for mRNA splicing; however , we cannot conclude they are benign . Furthermore , we cannot exclude the possibility that patients with de novo disease are somatic cell mosaics for PTPN11 mutations that are not present in white blood cell DNA , similar to the mildly affected mother in Family F who had two affected children . Despite these caveats , MC could be locus heterogeneous , similar to Noonan syndrome , which can be caused by mutations in PTPN11 or in other components of the Ras/MAPK pathway [28]; however , our targeted capture and sequencing of 74 genes that included most of the Ras/MAPK and PI3K/Akt signaling pathways did not find an obvious mutation in another gene in any of the 6 PTPN11 mutation-negative MC patients . We found no evidence of PTPN11 coding mutations in other cartilage tumor syndromes , including Ollier disease and Mafucci syndrome . Although sequencing was performed on lesional tissue rather than whole blood , it is possible that we may have missed causative mutations that are present in only a subset of cells within the lesion . We may also have missed mutations in the 5′ and 3′ untranslated regions of PTPN11 contained within exons 1 , 15 , and 16 , that we did not sequence in these patients . Based on our finding heterozygosity for noncoding SNPs in many of these samples , it is unlikely that large PTPN11 gene deletions or other causes of LOH are common in these syndromes . Despite the aforementioned limitations of our mutation detection method , our data are consistent with the separation of MC from the other cartilage tumor syndromes based on clinical and pathologic features . In conclusion , we combined linkage analysis in a single family with DNA capture and parallel sequencing of bar-coded DNAs from several families to identify mutations in PTPN11 as a cause of MC . The advantages of this approach are its ability to identify a region of interest , then simultaneously sequence affected individuals from multiple unrelated families , and then focus on genes for which novel SNPs or other mutations are seen in more than one family , all at reasonable cost ( ∼$10 , 000 in consumables ) . In patients with MC and PTPN11 mutations , we conclude that the mutations are loss-of-function since the mutant protein is not expressed , and that the loss of the remaining wild-type allele via a “second hit” is responsible for the formation of the exostoses . Since we did not detect PTPN11 mutations in all MC families , MC may be locus heterogeneous , although we have not found evidence after sequencing more than 70 genes that function in related pathways . Finally , precisely how mutations in PTPN11 give rise to the exostoses and endosteal tumors in patients with MC is not yet known . However , this question can now be addressed since mice with alleles of Ptpn11 that can be conditionally inactivated in temporal and site-specific manner are available [26] , [29] .
Informed consent was obtained through a Children's Hospital Boston IRB approved protocol . Specimens and/or DNA received from external institutions were collected under IRB approved protocols at host institutions and received coded without identifying information . Raw genotype data were generated for multiple members of family A using Affymetrix 6 . 0 SNP arrays , and genotypes were called using Affymetrix Genotyping Console with the Birdseed v2 algorithm and a confidence threshold of 0 . 02 . SNPs with <100% sample call rate or pedigree minor allele frequency of 0 were removed , then multipoint parametric linkage analysis of the remaining 421 , 922 autosomal and 17 , 169 X-linked SNPs was performed using MERLIN and its derivative MINX , respectively , with Affymetrix Caucasian allele frequencies and deCODE Genetics genetic map positions . The disease allele frequency was estimated at 1E-7 , and phenocopies and non-penetrance were not permitted ( affectation probability 0/1/1 ) . Because non-penetrance and non-ascertainment are potential confounding factors in MC , we analyzed only founders and affected individuals . To generate genomic libraries for each individual , 2 µg of genomic DNA were first sheared to ∼100 bp–200 bp using Adaptive Focused Acoustics following the manufacturer's protocol ( Covaris , Inc ) . Blunt-ended fragments were generated using an End-it DNA End-Repair kit ( Epicenter ) , purified using Agencourt AMPure XP magnetic beads ( Beckman Coulter ) , and eluted in 10 mM Tris Acetate , pH 8 . 0 . The fragments were A-tailed using the Klenow fragment ( NEB ) , purified , eluted in 1× Quick Ligase Buffer ( Quick Ligation kit , NEB ) , and incubated with Quick T4 DNA ligase and 100 µM barcoded-adapters ( Table S4 ) to create a library of adapter-ligated fragments . A different barcoded adapter was used for each genomic DNA library . Each library was again purified using Agencourt AMPure XP magnetic beads and eluted in 40 µl of 10 mM Tris Acetate , pH 8 . 0 . Libraries used for hybridization to the first capture array were amplified according to two strategies: 3 µl amplified for 18 cycles in four 50 µl PCR reactions ( Phusion High-Fidelity DNA polymerase , Finnzymes ) , or 2 µl amplified for 11 cycles in one 50 µl PCR reaction that was then purified and amplified for 17 cycles in ten 50 µl PCR reactions ( FastStart Taq DNA polymerase , Roche ) . For the libraries used for hybridization to the second capture array , 13 µl was amplified for 15 cycles in ten 50 µl PCR reactions ( Phusion High-Fidelity DNA polymerase , Finnzymes ) . Primers are provided in Table S5 . Sizes of amplified libraries were confirmed to be between 200–300 bp necessary for Illumina GA II sequencing prior to hybridization ( Figure S2 ) . To enrich regions of interest in the linked interval for sequencing , we used an Agilent Technologies 1M SureSelect DNA capture array . Target regions were defined using the UCSC Genome Browser and included: the union of exons from multiple GRCh37/hg19 gene , mRNA , and Alt Events tracks; 30 bp of proximal and distal intronic flanking sequence; and 1000 bp of upstream promoter sequence . Targets were padded with 60 bp of additional proximal and distal flanking sequence to promote uniform capture coverage , for a total size of 1 , 187 , 477 bp . Probes were designed against NCBI36/hg18 using the Agilent eArray software ( https://earray . chem . agilent . com/earray/ ) and translated coordinates , with 60-nt length , 3-nt spacing , and repetitive elements masked . The resulting 243 , 488 probes spanned 844 , 339 bp ( GRCh37/hg19 ) , and included 71 . 1% , 72 . 2% , 88 . 4% , and 98 . 6% of the padded target , unpadded target , UCSC exons , and CCDS coding sequence , respectively . The probes and their reverse complements were each applied in duplicate to the capture array for a total of 973 , 952 probes . For the second 1M SureSelect DNA capture array , Biomart ( http://uswest . ensembl . org/biomart/ ) was used to obtain the Ensembl NCBI37/hg19 coordinates for the exons of 76 genes ( Table S2 ) . Exons were padded with 90 bp to define a 718 , 566 bp target region . eArray was used to design 568 , 634 probes to target the repeat masked sequences of this region ( 91% ) . The target region for PTPN11 was defined as 93 , 180 bp spanning 1 kb upstream to 2 kb downstream of the gene . Repeat masker ( http://www . repeatmasker . org/ ) was used to mask only Alu repeats ( 37% of the region ) resulting in a target region of 61 , 365 bp , for which 55 , 205 probes were designed using eArray . All probes were 60-nt in length and spaced every 1-nt . The capture array was designed to include all probes ( 623 , 839 total ) , as well as the reverse complement of every 2nd probe and every 17th probe , for a total of 972 , 455 probes . For the first capture array , 1 . 4 µg of each of the 16 amplified libraries was pooled and hybridized to the array following Agilent's SureSelect DNA Capture Array protocol version 1 . 0 . Different blocking oligonucleotides ( Table S5 ) were added to the hybridization . After elution from the array , half of the captured library was amplified in five 50 µl PCR reactions for 18 cycles using Phusion High-Fidelity DNA polymerase ( Finnzymes ) and post-capture primer pair ( Table S5 ) , purified using an E-Gel CloneWell ( Invitrogen ) to remove primer dimers , and re-amplified using fifteen PCR cycles with the same primer pair . The amplified library was again purified to eliminate primer dimers using E-Gel . For the second capture array , 2 µg of each of the 12 amplified libraries was pooled and hybridized to the array . After elution , half of the captured library was amplified in five 50 µl PCR reactions for 15 cycles and purified using Agencourt AMPure XP magnetic beads . Further details of the array design and methods for sequence analysis are provided in the supporting information . Additional methods for Illumina data analysis , copy number analysis , Sanger sequence analysis of PTPN11 ( Table S6 ) , DNA extraction from lesional tissue , PCR product subcloning experiments , aCGH analysis , MLPA ( Table S7 ) , and immunodetection of SHP2 are also provided in the supporting information ( Text S1 ) . | Children with cartilage tumor syndromes form multiple tumors of cartilage next to joints . These tumors can occur inside the bones , as with Ollier disease and Maffuci syndrome , or on the surface of bones , as in the Multiple Osteochondroma syndrome ( MO ) . In a hybrid syndrome , called metachondromatosis ( MC ) , patients develop tumors both on and within bones . Only the genes causing MO are known . Since MC is inherited , we studied genetic markers in an affected family and found a region of the genome , encompassing 100 genes , always passed on to affected members . Using a recently developed method , we captured and sequenced all 100 genes in multiple families and found mutations in one gene , PTPN11 , in 11 of 17 families . Patients with MC have one mutant copy of PTPN11 from their affected parent and one normal copy from their unaffected parent in all cells . We found that the normal copy is additionally lost in cartilage cells that form tumors , giving rise to cells without PTPN11 . Mutations in PTPN11 were not found in other cartilage tumor syndromes , including Ollier disease and Maffucci syndrome . We are currently working to understand how loss of PTPN11 in cartilage cells causes tumors to form . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"cancer",
"genetics",
"pediatric",
"oncology",
"pediatrics",
"molecular",
"genetics",
"pediatric",
"orthopedics",
"pediatrics",
"and",
"child",
"health",
"biology",
"autosomal",
"dominant",
"clinical",
"genetics",
"gene",
"identification",
"and",
"analysis",
"genetics",
"human",
"genetics",
"genetics",
"and",
"genomics"
] | 2011 | Loss-of-Function Mutations in PTPN11 Cause Metachondromatosis, but Not Ollier Disease or Maffucci Syndrome |
Deleterious mutations appearing in a population increase in frequency until stopped by natural selection . The ensuing equilibrium creates a stable frequency of deleterious mutations or the mutational load . Here I develop the comparable concept of a damage load , which is caused by harmful non-heritable changes to the phenotype . A damage load also ensues when the increase of damage is opposed by selection . The presence of a damage load favors the evolution of asymmetrical transmission of damage by a mother to her daughters . The asymmetry is beneficial because it increases fitness variance , but it also leads to aging or senescence . A mathematical model based on microbes reveals that a cell lineage dividing symmetrically is immortal if lifetime damage rates do not exceed a threshold . The evolution of asymmetry allows the lineage to persist above the threshold , but the lineage becomes mortal . In microbes with low genomic mutation rates , it is likely that the damage load is much greater than the mutational load . In metazoans with higher genomic mutation rates , the damage and the mutational load could be of the same magnitude . A fit of the model to experimental data shows that Escherichia coli cells experience a damage rate that is below the threshold and are immortal under the conditions examined . The model estimates the asymmetry level of E . coli to be low but sufficient for persisting at higher damage rates . The model also predicts that increasing asymmetry results in diminishing fitness returns , which may explain why the bacterium has not evolved higher asymmetry .
Evolution by natural selection generally produces phenotypes that maximize fitness , but many factors can interfere . Genetic constraints can lead to a suboptimal phenotype , but an optimal phenotype may not be achieved even when an optimal genotype is possible . If the rate of deleterious mutations is sufficiently high , selection is unable to eliminate all mutations and a mutation-selection equilibrium at a lower fitness ensues . An asexual population has at equilibrium a mean fitness of ( 1 ) where U is the deleterious genomic mutation rate [1] , [2] . The mutational load equals ( references [3]–[5] ) . However , an optimal phenotype can also be prevented by the direct action of non-heritable damage . Bones can be broken , muscles torn , and macromolecules oxidized . All these lower fitness despite the perfection of the genotype . Although the study of deleterious mutations is long standing in evolution [1]–[7] , interest in damage is recent . The transmission of deleterious mutations across generations follows the rules of genetics . While damage is not heritable , it can still be transmitted from mother to daughter and its transmission rules are just being explored as an evolutionary phenomenon [8]–[11] . Here I develop the concept of a damage load and analyze its evolutionary consequences . Mutational and damage loads may appear on the surface similar , but key and fundamental differences are revealed by a comparison . Because recent experimental work has stimulated an interest in the effects of damage in microbes [12]–[14] , the analysis focuses on a single-celled organism reproducing by binary fission . A model for damage load can be developed by allowing the generation of damage , the operation of selection , and the attainment of the ensuing damage-selection equilibrium . Recent models have in fact used such an approach to examine the evolution of transmission rules for damage , i . e . how a mother cell distributes her damage to her two daughter cells [8]–[11] . However , with the exception of the most recent model by Erjavec et al . [11] all of these models were limited because a key difference between damage and mutations is the timing of their effects . In metazoans , the consequences of damage are immediate because the soma is affected . Mutational damage to the germline is delayed to the next generation . Because somatic mutations are not inherited through the germline , they are effectively non-heritable in most organisms and equivalent to damage for the present analysis . For single-celled organisms damage such as oxidized proteins has immediate effects , while genetic damage is again delayed in expression . If damage acts immediately , an early event during the lifetime of an organism has more impact than a later one . More importantly , an early damage can extend generation time and expose the organism to even more damage . To incorporate the timing of damage , a new model was developed . The model shared some similarities with the one by Erjavec et al . , but it assumed instead that the effect of damage was linear . Ackermann et al . [8] examined a range of damage effects , including a linear relationship , but their effects were mapped directly to fitness and thus did not incorporate the timing of damage . Given that evidence supporting either a linear or higher order effect of damage is lacking , a linear assumption is parsimoniously reasonable and provides more statistical power by reducing the number of parameters . Linearity additionally allows for simpler but explicit solutions and facilitates fitting data to estimate key parameters in the model .
Let k0 be the amount of damage a mother cell receives at birth . She immediately acquires new damage and , if λ is the intrinsic rate of damage , her damage at any time t is To divide into two daughter cells , the mother cell is assumed to build up an intracellular product P to a checkpoint Π . Assuming that damage hinders function linearly , P accumulates at a rate ( 2a ) ( 2b ) by integration and letting t = T0 and P ( T0 ) = Π when the mother cell divides and T0 is her doubling time . The integration constant P ( 0 ) is set to zero because a new pool of the product P is assumed to be built de novo for every cell division . If k0 = λ = 0 , T0 = Π in Equation 2b . Thus , Π represents both the checkpoint and the shortest possible doubling time achieved by the fittest and damage-free bacteria . The dual meaning of Π results because dP/dt is scaled to have a maximum value of 1 in the absence of damage ( k0 = λ = 0 in Equation 2a ) . The scaling assumes that P increases linearly with time in the absence of damage and also renders time into units of P . Although the true regulator of bacterial division is not known [15] , [16] , a model requiring the build-up of a product to a checkpoint is reasonable [11] . Various cellular ( volume , mass and length ) and biochemical attributes have been postulated to serve as checkpoints , but distinguishing between primary ( causative ) and secondary ( downstream ) regulators has been difficult . Regardless , the constancy of bacterial cell size shows that some accounting mechanism and a checkpoint must exist . Upon dividing , the mother cell partitions her damage D0 to two daughters andTo allow for variation in the partitioning , let a and ( 1−a ) be the proportion of D0 given to the daughters , which are subscripted 1 and 2 and 0≤a≤½ . Thus , daughter 1 always receives less damage if a<½ and the damage given to the daughters is ( 3a ) ( 3b ) Because each daughter in turn becomes a mother , Equation 2b can be subscripted to describe the daughters or ( 4 ) ( 5 ) by the quadratic formula and i = 1 or 2 . Thus , given T0 for a mother cell , Ti of her two daughters can be determined . k0 in Equations 3a and 3b is obtained by rearranging Equation 2b as ( 6 ) The ability to predict T1 and T2 given T0 allows projecting forward in time the doubling time , and hence fitness , of every individual in a population . T1 and T2 serve as T0 for the next generation and Equations 3 , 5 and 6 only need to be reiterated . Equations 3 , 5 and 6 are hereafter referred to as the model .
To determine if a lineage of dividing cells converged to a determinable level of damage over successive generations , the model was examined for equilibria . Daughter cells reach an equilibrium when and . Substituting these conditions into Equations 3a and 3b , yields ( 7a ) ( 7b ) where α = a/ ( 1−a ) . Substituting Equations 7a and 7b into Equation 4 ( 8a ) ( 8b ) Equilibrium values and are possible if roots to the quadratic solutions of Equations 8a and 8b are real or ( 9a ) ( 9b ) Thus , equilibria are possible , depending on the level of asymmetry and the product of the two parameters Π and λ . The linking of Π and λ into a single product or fundamental parameter facilitates the analysis of the model by reducing the effective number of parameters in the model from three to two . Because λ is the intrinsic damage rate and Π is the doubling or life time of a damage-free individual ( Equation 2b ) , all damage in such a cell is acquired over its lifetime and Πλ represents its total or lifetime damage rate . The partitioning of damage from the mother cell to her daughter cells is symmetrical if a = ½ , in which case the daughters are identical , k1 = k2 , and T1 = T2 . Letting i = 1 represent both daughters , the equilibrium conditions are provided by Equation 8a with α = 1 or ( 10 ) The roots to Equation 10 are real if ( 11 ) The equilibrium can be locally stable and the stability can be assessed graphically ( Figure 1A ) . The stability results because the doubling times in a lineage descending from a mother cell with a doubling time less than increase until equaling . On the other hand , if the doubling time of a mother cell is greater than , the doubling time of her lineage decreases to . If Πλ>1/6 , the doubling time of a lineage never attains an equilibrium and it increases over generations until it is infinitely long ( Figure 1B ) . When doubling time is infinitely long , a mother cell is alive but unable to divide because its damage content is too high and P cannot be built up to Π . At the threshold of Πλ = 1/6 , P is built up to Π and , by Equation 10 , , which corresponds to the equilibrium doubling time of the least fit symmetrical cell . To obtain an estimate of the damage load as with symmetrical transmission , the equilibrium mean fitness was estimated for a population of cells with doubling time of relative to a population with the highest fitness or the shortest possible doubling time of Π ( Equation 2b ) . A population with doubling time of Π increases by definition by a factor of 2 during a time period Π . A population with a doubling time of increases during the same period by a factor of . The mean relative fitness at equilibrium resulting from damage is therefore the ratio of , or ( 12 ) ( 13 ) by letting equal Equation 10 . A summary of the results , including new ones to follow , and the definitions of all parameters and variables for the model are presented in Table 1 . The evolution of the asymmetrical transmission of damage can be examined by letting 0≤a<½ . Unlike a = ½ ( Figure 1A ) , a separate equilibrium is now possible for each of the daughters ( Equations 8a , 8b ) . Because a<½ , daughter 1 gets less damage . Inspection of Equations 9a and 9b reveals that as Πλ increases from zero , and go through conditions in which both , one or none attain real equilibrium values ( Table 1 ) . If Πλ<1/6 and a is sufficiently large , Equations 9a and 9b are satisfied and both and attain equilibria ( Figure 2A ) . If a is not sufficiently large , only Equation 9a is satisfied and has an equilibrium while does not . The same outcome ensues if Πλ≥1/6 and a is sufficiently small ( Figure 2B ) . Thus , the threshold of Πλ = 1/6 still plays an important role ( Table 1 ) . If 0≤a<½ , unlike when a = ½ , knowing the values and does not allow an estimate of . and represent extreme values to which the doubling time of daughters converge as they replicate , e . g . , as it would be illustrated if a graphical projection ( see Figure 1 ) were applied to Figure 2 . As new daughter 1 and 2 cells are generated , lineages descending from each type converge to and , respectively , if the equilibria exist . If a daughter 1 is at the equilibrium , it still generates daughter 2 cells , which create new lineages that now converge onto . The presence of different lineages generates a population with mixed doubling times . The distribution of doubling times in the population is in turn shaped by natural selection and can be estimated only after the distribution reaches a selection-damage equilibrium . In contrast , if a = ½ , offers a direct estimate of ( Equation 10 ) because a population with mixed doubling times is not possible at equilibrium . Once a lineage converges to all descending daughters have a doubling time of . was therefore estimated by using the model to simulate a population of cells under selection until a fitness equilibrium was reached . Selection was imposed by allowing cells with shortest doubling times to divide before other cells ( Figure 3; Table 1 ) . Πλ behaves again as a single parameter because the model uses Equation 5 , which when combined with Equation 12 collapses Π and λ into a product . For values of 0<Πλ<1/6 , is highest with a = 0 and maximum asymmetry is favored . However , the advantage shrinks as Πλ decreases to zero . The difference between for a = ½ and a = 0 decreased from 8 . 6% , 0 . 57% to 0 . 0025% when Πλ goes from 0 . 165 , 0 . 1 to 0 . 01 . For Πλ>1/6 , asymmetry is favored more strongly but relationship between and Πλ is non-monotonic . When Πλ = 0 . 17 , is maximized at a = 0 . 1 and an intermediate asymmetry is favored . Such non-monotonicity is also present , but less apparent , for other values of Πλ , including when Πλ<1/6 . Inspection of the relationship between and a for extreme values reveals why it is not monotonic for Πλ>1/6 . Simulations showed that for Πλ>0 . 25 populations were unable to grow , regardless of a . When Πλ>0 . 25 and a = 0 , Equation 9a is satisfied and daughter 1 achieves its equilibrium of . However , daughter 2 from a cell at receives too much damage and is unable to divide . Thus , the mother cell just replaces herself with daughter 1 and there is no net reproduction . A similar effect explains the left side of the hump when Πλ = 0 . 17 ( Figure 3 ) . However , in this case declines not because daughter 2 is unable to divide , but because the grand-daughter 2 is unable ( analysis not presented ) . The right side of the hump for Πλ = 0 . 17 results because with low asymmetry Equation 9a is not satisfied and daughter 1 is now unable to divide . The combination of all these effects explains also many of the inflections seen in Figure 3 . Recent E . coli data [14] reporting growth rates of mother and daughter cells allow determining where in the parameter space of the model a biological organism resides . By converting the reported growth rates to doubling times , observed values of T0 , T1 and T2 were obtained for the bacterium ( Figure 4 ) . Each observed T0 was then inputted into the model over the parameter space to derive expected T1 and T2 values . The parameters of the model were determined as those that minimized the difference between the observed and expected T1 and T2 by maximum likelihood via a conjugate gradient method implemented in the software package HyPhy [17] . The parameter a was estimated to have a mean value of 0 . 4836 and a 95% Bayesian Confidence Interval ( BCI ) of [0 . 4716–0 . 4905] . Estimated mean value of Π was 18 . 95 min ( 95% BCI [16 . 61–21 . 71] ) and of λ was 0 . 007737 min−1 ( 95% BCI [0 . 005347–0 . 009717] ) . Applying these mean values of Π , λ , and a to Equations 8a and 8b estimated of and min , which show that the doubling times of the two daughter lineages attained separate equilibria . The presence of these two equilibria in E . coli was also revealed by a phase plot overlaying the observed and expected values ( Figure 4 ) . Applying the estimates of Πλ = ( 18 . 95 ) ( 0 . 007737 ) = 0 . 1467 and a = 0 . 4836 to the model also showed that was higher , though only by a small amount , for these E . coli relative to a symmetrical ( a = ½ ) bacterium with the same value of Πλ ( Figure 5A ) . However , the advantage became greater if Πλ were increased . While bacteria with a = ½ could not reproduce once Πλ>1/6 ( Equation 11 ) , these E . coli were able up to Πλ = 0 . 173 ( Figure 5A ) . If Πλ>0 . 173 , neither bacteria could reproduce , but damage accumulated and shut down division more slowly in cells with a = 0 . 4836 than a = ½ ( Figure 5B ) .
Just as deleterious mutations and selection create at equilibrium a mutational load , non-heritable damage and selection generate a damage load . Both type of loads contribute to the phenotypic load [18] . Analysis of a model for damage load in organisms dividing by fission revealed that a single fundamental parameter equal to Πλ determines equilibrium mean fitness , where the damage load is 1− ( see Table 1 for summary ) . Πλ is the product of two separate parameters in the model , λ the intrinsic damage rate and Π the doubling or life time of a damage-free individual . Thus , Πλ represents the total or lifetime damage rate of damage-free cell . A damage load selects for mother cells that partition damage asymmetrically between daughters . If division were symmetrical , the daughters are identical and the equilibrium doubling time of descending lineages attains a real value when lifetime damage rate is less than 1/6 ( Figure 1A ) . If the rate is greater 1/6 , becomes infinitely long because the mother cells acquire too much damage and are unable to build cellular products to the amount needed for fission . The cell is alive but the lineage dies because doubling time becomes infinitely long ( Figure 1B ) . Asymmetry allows cells to survive by division up to a rate less than 0 . 25 ( see Transmission Rules; Figure 2; Table 1 ) . Moreover , within the range of 0≤Πλ<0 . 25 , equilibrium mean fitness is generally maximized as asymmetry decreases to the extreme when one daughter receives all of the damage harbored by a mother cell . Such extreme asymmetry is represented in the model with an asymmetry coefficient with a value of a = 0 . However , for some intermediate lifetime damage rates an optimal and intermediate value of asymmetry is favored ( Figure 3; Πλ = 0 . 17 ) . The evolution of asymmetry due to a damage load is comparable to the evolution of sex from a mutational load . The evolution of sex requires that the distribution of deleterious mutations is underdispersed in a population , i . e . , that the variance is less than the mean [2] . Because sex shuffles genetic variation , its net effect is to redistribute mutations by a Poisson process , in which case the variance converges to the mean . If the variance and mean are equal , sex is not advantageous because the variance cannot be changed . If sexual reproduction were to overdisperse deleterious mutations an advantage can result , but that is prevented by the rules of genetic transmission . Asymmetry likewise increases variance in a population , but selection for asymmetry is much stronger because transmitting all or none of the damage to the daughters overdisperses the variance . Asymmetrical transmission impacts the life history of a lineage by creating the two types of daughters . While daughter 1 is rejuvenated at birth , daughter 2 is loaded with damage . The deterioration of daughter 2 , her daughter 2 in turn , and so forth constitutes currently one of the main hypothesis for the evolution of aging or senescence in microbes [11]–[14] . Fission in microbes results in the creation of a new and old pole . Because the new pole harbors less damage , it tags daughter 1 . However , the long term consequences of asymmetry are debated . With symmetry and low damage , single-celled organisms are immortal ( Figure 1A ) . Do high damage and asymmetry make them mortal [14] , [19]–[22] ? The model shows that it depends on the level of asymmetry and the rate of lifetime damage . If lifetime damage rate is less than 1/6 and asymmetry is sufficiently large , both the equilibrium doubling times and of daughters 1 and 2 attain real values and the microbe is immortal ( Figure 2A ) . If the asymmetry is not sufficiently large , has a real equilibrium value , but becomes infinitely long and the daughter 2 lineage is mortal . The same outcome ensues if lifetime damage rates are greater than or equal to 1/6 and asymmetry is sufficiently small ( Figure 2B ) . The mortality of daughter 2 renders all lineages in microbe mortal because all new poles eventually become old and reside in a daughter 2 . Thus , although asymmetry matters , a lifetime damage rate of 1/6 is a key threshold . If the rate is less than 1/6 , immortality is possible . If the rate is greater than or equal to 1/6 , a microbe is mortal . A recent study recording the division of E . coli cells over generations [14] provided an estimate of parameters for lifetime damage rate and asymmetry ( see Estimating Model Parameters ) . The estimates placed E . coli in an area of the parameter space where the bacterium was immortal; both and attained real equilibrium values ( Figure 4 ) . However , the estimate of the asymmetry coefficient a at a value of 0 . 4836 was at first surprising . Given that fitness is generally maximized with extreme asymmetry ( a = 0; Figure 3 ) , a lower a could have been expected . What is the level of advantage provided by such a small degree of asymmetry ? Could the level of asymmetry just be noise rather than an adaptation [22] , [23] ? If asymmetry is adaptive , why has it not evolved to be much higher ? Resolution of these issues requires more information , but the current model can be used to provide guidance at this point . On the basis of the parameter values estimated , the model predicts that the equilibrium mean fitness for an asymmetrical E . coli with a = 0 . 4836 is higher by a difference of 3×10−5 when compared to that of a hypothetical and symmetrical E . coli with a = ½ ( Figure 5A ) . Although small , such a difference is more than sufficient for evolving asymmetry in large microbial populations . However , the difference may be on the low end of the range experienced by E . coli because the model was based on parameter values estimated in a benign laboratory environment . The parameter lifetime damage rate was estimated to be 0 . 1467 , but it could be much higher for E . coli in the wild . If the rate were increased by 14% to the threshold of 1/6 ( Equation 11 ) , the advantage of a small degree of asymmetry is magnified . At this new rate , a symmetrical E . coli can no longer persist by reproduction and its doubling time becomes infinitely long ( Figure 1B ) . With an asymmetry of just 0 . 4836 , asymmetrical E . coli can persist up to rates as high as 0 . 173 ( Figure 5A ) . Moreover , the advantage continues to increase if rates were further elevated . If they were greater than 0 . 173 , E . coli with both a = 0 . 4836 and symmetry cannot persist , but damage accumulates and cell division shuts down more slowly in the asymmetrical bacterium . For example , if the rate were 0 . 175 , the frequency of dividing cells drops to 0% in slightly over 500 min with symmetry , but only after more than 1300 min with a = 0 . 4836 ( Figure 5B ) . Retaining a few dividing cells for several extra hundred minutes could be invaluable to an organism capable of rapid growth . Thus , the fitness advantage of a small degree of asymmetry , such as a = 0 . 4836 , could be high . However , why has asymmetry level in E . coli not evolved to be higher than 0 . 4836 ? An obvious answer is there may be a cost . Although the model assumes no costs , it reveals the fitness constraints . Assuming that E . coli experiences a higher lifetime damage rate of 0 . 17 , inspection of Figure 3 shows that the equilibrium mean fitness has the greatest curvature when asymmetry equals 0 . 475 . Equilibrium mean fitness equals 0 . 68 when asymmetry is 0 . 475 and it increases only by 3% to 0 . 70 when asymmetry is 0 . 450 . If in the simplest scenario the cost of reducing a from ½ to 0 . 475 equals from 0 . 475 to 0 . 450 , fitness gain for the second reduction could be too small to override the costs . The fitness gain for the first reduction is large because symmetrical bacteria cannot reproduce when the lifetime damage rate is 0 . 17 ( Figure 3; Table 1 ) . Unless the cost is extremely small or the second reduction is less costly , asymmetry should evolve to reside where the curvature is greatest [24] . The level of asymmetry can be shifted by different cost functions , but the curvature constrains its evolution to the neighborhood of 0 . 475 . Estimating the costs will be needed for a full resolution , but the low asymmetry estimated for E . coli may well be anticipated by the model . A higher lifetime damage rate may be reasonable for E . coli and other microbes . Defenses and weapons by microbial competitors and hosts routinely employ mechanisms that inflict non-genetic damage often through oxidation [25]–[29] . Microbes face damage even in apparently benign environments . E . coli grown under standard laboratory conditions do not experience much oxidative damage . However , 48 hr after reaching stasis , oxidative damage to proteins increases six fold [30] . The damaged population can be separated into two fractions by centrifugation . One fraction , which accounts for 40% of the cells , contains bacteria capable of dividing and forming colonies on agar plates . The bacteria in the second fraction are not , although they remain intact and metabolically active . Most importantly , almost 90% of the detectable oxidative damage is in the second fraction , which demonstrates well outcomes comparable to Figure 2A and Figure 5B . Because a damage load is created by non-heritable variation , it has characteristics that are attributable to the soma . The asymmetrical transmission of non-heritable damage in microbes , and the subsequent division of labor [31] , has led to suggestions that these organisms have the equivalent of soma and germline [21] , [32] . From this perspective , the evolution of germline , soma , and senescence in metazoans [32] is just the extension of microbial asymmetry and the damage load could be called the somatic load . Although the present model was formulated for microbes , it could be generalized to include metazoans . It may also be useful for describing a population of cells within a metazoan . Do stem cells partition damage asymmetrically ? Many aspects of the present model are not novel . Previous models have shown that the asymmetrical transmission of damage can be favored during binary fission by both directional and stabilizing selection [8]–[10] . Erjavec et al . [11] demonstrated qualitatively with simulations a threshold for cells dividing symmetrically . However , the present model offers some new perspectives . First , the derivation of a damage load allows a comparison to a mutational load . In metazoans with large genomes , the mutational load [33] could be as large as some of the damage load estimates ( Figure 3 ) . On the other hand , because the mutational load is smaller in microbes [33] , the damage load could be a stronger evolutionary force . Second , the present model shows that the two parameters Π and λ combine to form a single fundamental parameter as the product Πλ or the lifetime damage rate . Moreover , the model was also able to predict key thresholds for Πλ at 1/6 and 0 . 25 without any empirical/data calibration . The threshold of 1/6 delineates the boundary for when cell dividing by fission is mortal or immortal . This outcome stands in contrast to the absence of any theoretical framework for whether the genomic mutation rate U , a key parameter for determining the mutational load ( Equation 1 ) has an upper limit , despite the fact that metazoans and RNA viruses have independently evolved maximum rates of 1 ( reference [33] ) . Third , a fit of the model to experimental data from E . coli provided estimates for all of the key parameters in the model . The parameter values showed that E . coli was immortal under the conditions examined . The determination of where a real organism resides in parameter space offers a powerful predictive tool for studying evolution . | Almost all living organisms deteriorate with time through the process of aging or senescence . Because most studies on senescence examined organisms possessing a juvenile state , it was thought that bacteria , which reproduce by producing two apparently identical daughter cells , were immortal and not senescent . Recent studies have demonstrated that bacteria senesce because one daughter is allocated a larger share of the mother's load of non-genetic damage . Nonetheless , it is still equivocal whether bacterial senescence renders them mortal . I have developed a model that demonstrates that bacteria can be immortal if they experience damage below a threshold rate . A fit of the model to data shows that bacteria grown under standard laboratory conditions are immortal because they encounter a rate below the threshold . Because bacteria often experience higher damage rates in nature , it is likely that bacteria are generally mortal . The allocation of more damage to one daughter and the resulting mortality is the price bacteria pay to survive higher damage rates . These results suggest that senescence originated with the evolution of the first single-celled organisms and that it is ancestral in all multicellular organisms . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] | [
"computational",
"biology/population",
"genetics",
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"evolutionary",
"biology/evolutionary",
"ecology",
"microbiology/microbial",
"evolution",
"and",
"genomics",
"computational",
"biology/evolutionary",
"modeling",
"evolutionary",
"biology",
"genetics",
"and",
"genomics/population",
"genetics"
] | 2010 | A Model for Damage Load and Its Implications for the Evolution of Bacterial Aging |
The proteasome inhibitor MG132 had been shown to prevent galactose induction of the S . cerevisiae GAL1 gene , demonstrating that ubiquitin proteasome-dependent degradation of transcription factors plays an important role in the regulation of gene expression . The deletion of the gene encoding the F-box protein Mdm30 had been reported to stabilize the transcriptional activator Gal4 under inducing conditions and to lead to defects in galactose utilization , suggesting that recycling of Gal4 is required for its function . Subsequently , however , it was argued that Gal4 remains stably bound to the enhancer under inducing conditions , suggesting that proteolytic turnover of Gal4 might not be required for its function . We have performed an alanine-scanning mutagenesis of ubiquitin and isolated a galactose utilization-defective ubiquitin mutant . We have used it for an unbiased suppressor screen and identified the inhibitor Gal80 as a suppressor of the transcriptional defects of the ubiquitin mutant , indicating that the protein degradation of the inhibitor Gal80 , and not of the activator Gal4 , is required for galactose induction of the GAL genes . We also show that in the absence of Gal80 , Mdm30 is not required for Gal4 function , strongly supporting this hypothesis . Furthermore , we have found that Mediator controls the galactose-induced protein degradation of Gal80 , which places Mediator genetically upstream of the activator Gal4 . Mediator had originally been isolated by its ability to respond to transcriptional activators , and here we have discovered a leading role for Mediator in the process of transcription . The protein kinase Snf1 senses the inducing conditions and transduces the signal to Mediator , which initiates the degradation of the inhibitor Gal80 with the help of the E3 ubiquitin ligase SCFMdm30 . The ability of Mediator to control the protein degradation of transcriptional inhibitors indicates that Mediator is actually able to direct its own recruitment to gene promoters .
Cells regulate the expression of their genes according to requirement [1] . Activators recruit chromatin-remodeling or chromatin-modifying complexes that change the structure of chromatin to promote transcription [2] , [3] , while repressors recruit chromatin-modifying complexes that change the structure of chromatin to prevent transcription [4] , [5] . Repressors also bind directly to activators and prevent the recruitment of the transcription machinery [6] . According to the reverse recruitment hypothesis [7] , the transcription factors do not move to the highly transcribed genes , but the highly transcribed genes move to the gene expression machines ( GEMs ) , which are protein complexes with fixed locations in the nuclear periphery . GEMs , which host all transcription factors that are required for gene expression from RNA Polymerase to RNA capping , splicing , poly-adenylation , and export factors [8] , are associated with the nuclear pores , and the mature mRNAs , once produced at the GEM , are immediately exported out of the nucleus to be translated at the ribosomes of the rough endoplasmic reticulum [7] . The Saccharomyces cerevisiae GAL genes are a paradigm for transcriptional regulation in eukaryotes [9] . In cells grown with glucose , Gal80 binds to Gal4 and blocks its activation function [10] , while Mig1 binds to an upstream silencer and recruits the general repressor Tup1 to prevent gene expression [11] . Upon the switch to galactose media , Snf1 phosphorylates Mig1 , causing its translocation from the nucleus to the cytoplasm [12] , while Gal80 dissociates from Gal4 [13] and is sequestered in the cytoplasm by Gal3 [14] , leaving Gal4 free to activate the GAL genes , which are required for galactose utilization [7] . Proteolytic stability of transcription factors offers an intriguing possibility for the eukaryotic cell to control gene expression [15] . Ubiquitin proteasome-dependent degradation ( UPD ) of activators and repressors plays an important role in gene regulation [16] , and treatment of S . cerevisiae cells with the proteasome inhibitor MG132 abolished galactose induction of the GAL1 gene [17] . Ubiquitin is a small protein of 76 amino acids that is transferred by E3 ubiquitin ligases to proteins to be targeted for degradation by the 26S proteasome [18] . F-box proteins confer substrate specificity to SCF ( Skip1-Cullin-F-box protein ) E3 ubiquitin ligases [19] . When cells are grown with galactose , an SCF E3 ubiquitin ligase containing the F-box protein Mdm30 , SCFMdm30 , ubiquitinates Gal4 [20] . The deletion of MDM30 stabilizes Gal4 under inducing conditions and leads to defects in galactose utilization , suggesting that recycling of Gal4 is required for its transcriptional activator function [20] . Subsequently , however , it was argued that Gal4 remains stably bound to the enhancer under inducing conditions , suggesting that proteolytic turnover of Gal4 might not be required for its function [21]–[23] . Previously , it had been shown that mono-ubiquitination protected Gal4 from the promoter-stripping activity of proteasomal ATPases [24]–[26] , suggesting a role for ubiquitin in transcriptional activation other than protein degradation . Recently , it has been reported that the proteolytic stability of Mediator subunits is inversely correlated with their ability to activate transcription when fused to a DNA-binding domain [27] . Mediator is a complex of more than 20 proteins that is conserved from yeast to man [28] . It was discovered by its ability to respond to transcriptional activators in vivo and in vitro [29] . Genome-wide gene expression studies with temperature-sensitive alleles have shown that Mediator is required for the transcription of nearly all RNA Polymerase II–dependent genes in yeast [30] . Mediator interacts directly with activators , General Transcription Factors , and RNA Polymerase II [31] . In higher eukaryotes , Mediator facilitates a DNA loop between enhancer and basal promoter via its interaction with cohesin [32] . In addition , Mediator affects steps that are downstream of the recruitment of RNA Polymerase II to the core promoter , as Med26-containing metazoan Mediator switches RNA Polymerase into the productive transcription elongation mode by an interaction of Med26 with TBP ( TATA-binding protein ) and the CTD ( C-terminal domain of RNA Polymerase II ) kinase P-TEFb [33] . Mediator also modifies chromatin via its own CDK8 subunit , which phosphorylates histone H3S10 , and by its interaction with histone acetyl- and methyltransferases [34] , [35] . Metazoan Mediator plays important roles in neurogenesis , cancer formation , and stem cell proliferation [31] . All of these reported functions of Mediator are genetically downstream of transcriptional activators . Here , we have found that Mediator additionally is able to act upstream of the transcriptional activator Gal4 by controlling the ubiquitin-mediated protein degradation of the inhibitor Gal80 . In the absence of Gal80 , Gal4 is free to recruit Mediator to the promoter of the GAL genes . Therefore , Mediator actually orchestrates its own recruitment to the GAL promoters upon galactose induction .
The role of ubiquitin proteasome-dependent protein degradation in the transcriptional regulation of the GAL genes has been controversial [19]–[22] . We performed an alanine-scanning mutagenesis of ubiquitin in order to isolate galactose-utilization defective ( gal− ) mutant strains and use these for unbiased multi-copy suppressor screens . However , no ubiquitin single point mutant displaying the gal− phenotype was isolated ( Figure S1 , even lanes; Figure S2 ) . The addition of an N-terminal tag can sometimes enhance the phenotype of point mutants , and so we fused a stretch of 10 N-terminal histidines to all ubiquitin mutant proteins . S . cerevisiae cells expressing H10UbF4A , H10UbK6A , H10UbI13A , H10UbR42A , H10UbF45A , H10UbD58A , and H10UbT66A in the place of endogenous ubiquitin displayed growth defects on galactose plates containing the respiration inhibitor Antimycin A ( AA; Figure S1 , lanes 5 , 11 , 23 , 85 , 87 , 105 , 119; Figure S2 ) . The presence of the respiration inhibitor AA requires the cells to metabolize more galactose molecules in order to form colonies , which serves to translate defects in the transcriptional activation of the GAL genes into stronger growth defects on galactose plates . The H10UbD58A mutant strain was also unable to grow on galactose plates in the absence of AA ( Figure 1A , line 5 ) , and it was transformed with a multi-copy library of S . cerevisiae genomic DNA fragments [36] . Gal3 was isolated by its ability to confer growth to the H10UbD58A mutant strain on galactose plates upon over-expression ( Figure 1A , line 6 ) . The over-expression of Gal3 also dosage-compensated the gal− phenotype of the other H10Ub mutant strains ( Figure S3 , compare odd and even lanes; the H10UbF4A mutant strain was barely viable and was excluded from further studies ) . Gal3 sequesters Gal80 in the cytoplasm upon galactose induction [10] , and our finding that the over-expression of Gal3 suppressed the gal− phenotype of the H10Ub mutant strains indicated that ubiquitin-mediated protein degradation of Gal80 could be required for galactose induction of the GAL genes and that the gal− phenotype of these H10Ub mutant strains might have been caused by excess Gal80 . Consistently , the additional gene deletion of GAL80 suppressed the gal− phenotype of the H10UbD58A mutant ( Figure 1A , line 7 ) and of the other H10Ub mutant strains ( Figure S4 ) . Reverse transcription coupled with real-time PCR quantification revealed that galactose induction of GAL1 mRNA relative to ACT1 mRNA was abolished in the H10UbD58A strain and that the over-expression of Gal3 and the additional gene deletion of GAL80 ( partially ) restored galactose induction ( Figure 1B ) . We performed chase assays with the protein biosynthesis inhibitor cycloheximide and found that HA-Gal80 was indeed degraded in galactose-induced H10Ub cells ( Figure 1C , lanes 5 to 8; Figure 1D , white bars ) . Importantly , HA-Gal80 had become stable in galactose-induced H10UbD58A mutant cells ( Figure 1C , lanes 13 to 16; Figure 1D , black bars ) as well as in the other gal− H10Ub mutants strains ( Figure S5 ) , suggesting that the galactose-stimulated protein degradation of Gal80 is necessary for transcriptional activation of the GAL genes . Our finding that the additional gene deletion of GAL80 suppressed the gal− phenotype of the H10Ub mutant strains provides genetic evidence that the failure to degrade Gal80 had been the cause ( and not the consequence ) of the gal− phenotype of the H10Ub mutant strains . E3 ubiquitin ligases add ubiquitin to proteins that are targeted for degradation by the 26S proteasome [18] , and Skp1 is an essential component of all SCF E3 ubiquitin ligases [19] . Previously , we had found that the Skp1 derivative Nub-HA-Skp1V90A , E129A ( Skp1dM ) causes the gal− phenotype when expressed in place of endogenous Skp1 [37] . We had isolated α2 as a multi-copy suppressor and shown that galactose-induced protein degradation of the repressor Mig2 , which—like α2 [38]—uses the co-repressor Tup1 , was abolished in the skp1dM strain [37] . The most likely explanation was that over-expression of α2 had titrated Tup1 away from GAL1 promoter-bound Mig2 , which—like Mig1 [39]—activated transcription in the absence of Tup1 . The additional gene deletion of MIG2 , however , had only partially suppressed the gal− phenotype of the skp1dM strain [37] , suggesting that Skp1 mediated the galactose-induced protein degradation of additional transcription factors . Therefore , we wanted to see if Gal80 was a functionally relevant target of SCF E3 ubiquitin ligases . HA-Gal80 protein was degraded in galactose-induced SKP1 wild-type cells ( Figure 2A , lanes 5 to 8; Figure 2B , white bars ) , while it was stable in galactose-induced skp1dM cells ( Figure 2A , lanes 13 to 16; Figure 2B , black bars ) , indicating that wild-type Skp1 was required for the galactose-induced protein degradation of Gal80 . We transformed the skp1dM mutant strain with multi-copy plasmids expressing Sgt1 ( which is required for Skp1-dependent cyclin degradation [40] ) , α2 , Ubp3 ( which dosage-compensates the gal− phenotype of cells expressing the proteolytically instable Tbp1E186D [41] ) , and Gal3 . The over-expression of Sgt1 suppressed the temperature sensitivity of the skp1dM strain ( Figure 2C , line 3 ) . The over-expression of Gal3 and α2 suppressed the gal− phenotype , but not the temperature sensitivity , of the skp1dM mutant strain ( Figure 2C , lines 4 and 6 ) , while the over-expression of Ubp3 had no effect ( Figure 2C , line 5 ) . Real-time PCR quantification revealed that galactose induction of GAL1 mRNA relative to ACT1 mRNA was abolished in the skp1dM mutant strain ( Figure 2D ) and that it was restored to some 550-fold in the presence of excess Gal3 and almost fully in the absence of Gal80 ( Figure 2D ) , providing genetic evidence that galactose-stable Gal80 had been the main cause for the gal− phenotype of the skp1dM strain . F-box proteins provide the substrate specificity to SCF E3 ubiquitin ligases [19] , and the deletion of the gene encoding the F-box protein Mdm30 causes a gal− phenotype [20] . Cycloheximide chase assays demonstrated that HA-Gal80 was degraded in galactose-induced BY4741ΔW wild-type cells ( Figure 3A , lines 5 to 8; Figure 3B , white bars ) , while it was stable in galactose-induced ΔMDM30 cells ( Figure 3A , lines 13 to 16; Figure 3B , black bars ) , suggesting that SCFMdm30 targets Gal80 for galactose-induced protein degradation . Importantly , and consistent with a recent report [42] , the additional gene deletion of GAL80 suppressed the gal− phenotype of the ΔMDM30 strain ( Figure 3C , line 4 ) . Gal80 was still degraded in galactose-induced ΔGAL11 cells ( Figure 3A , lanes 21 to 24; Figure 3B , grey bars ) and the additional gene deletion of GAL80 did not suppress the gal− phenotype of cells lacking Gal11 ( Figure 3C , line 6 ) , confirming that the suppression of the gal− phenotype of the ΔMDM30 strain by the additional gene deletion of GAL80 was gene-specific and that the F-box protein Mdm30 acts genetically upstream of the repressor Gal80 , while the Mediator component Gal11 ( Med15; which is a target of Gal4 [43] ) acts genetically downstream of the repressor Gal80 . Real-time PCR quantification of GAL1 mRNA relative to ACT1 mRNA confirmed that the additional gene deletion of GAL80 fully suppressed the transcriptional defect of the ΔMDM30 strain ( Figure 3D ) , suggesting that Mdm30 targets mainly Gal80 for galactose-induced protein degradation . Consistently , GST-Gal80 , but not GST , pulled down HA-tagged Mdm30 and Skp1 from yeast extracts ( Figure 3E , lanes 8 and 9; Figure S6 ) . Coomassie staining demonstrated that Gal80 and Mdm30 interacted at approximately equal amounts ( Figure 3E , lanes 8 and 9 ) . However , Gal80 interacted with Mdm30 ( and Skp1 ) not only in galactose-induced but also in glucose-grown cells ( Figure 3D , compare lanes 8 and 9; Figure S6 ) , possibly reflecting the ( slower ) protein degradation of Gal80 in cells grown with glucose ( Figure 3A , lane 4; Figure 3B , white bars ) . The half-life of Gal80 was calculated to be approximately 3 h in glucose-grown BY4741ΔW cells and approximately 1 h in galactose-induced BY4741ΔW cells . Gal80 had been completely stable in glucose-grown H10Ub cells ( Figure 1 , lines 1 to 4 ) , indicating that the N-terminal tail of 10 histidines might have interfered with the slow protein degradation of Gal80 in glucose-grown cells . In agreement with the hypothesis that Gal80 is not just degraded in galactose-induced but also in glucose-grown cells ( albeit with slower kinetics ) , Gal80 was poly-ubiquitinated in cells grown with glucose and in cells induced with galactose ( Figure 3F , lanes 6 and 7 ) . The amount of poly-ubiquitinated species of Gal80 was only slightly higher in galactose-induced cells as compared to in glucose-grown cells , suggesting that the generation of the poly-ubiquitinated species of Gal80 is rate-limiting , and once generated , poly-ubiquitinated Gal80 is immediately degraded . The ubiquitinated forms of HA-Gal80 are not visible in the input lanes , indicating that only a very small fraction of the Gal80 inside the cell is ubiquitinated at any point in time . The figure further shows that Gal80 was poly-ubiquitinated in wild-type cells as well as in cells lacking Mdm30 ( Figure 3F , compare lanes 7 and 9 ) , indicating that Mdm30 is not the only F-box protein targeting Gal80 . In order to identify additional SCF E3 ubiquitin ligases targeting Gal80 , we tested galactose utilization defective F-box protein gene deletion mutant strains [37] and found that Gal80 was also stable in galactose-induced cells lacking the F-box proteins Das1 and Ufo1 ( Figure S7A , B ) . Importantly , the gal− phenotype of cells lacking Das1 and Ufo1 was suppressed by the additional gene deletion of GAL80 ( Figure S7C ) and GST-Gal80 , but not GST , pulled down Das1 , and Ufo1 from yeast extracts ( Figure S7D , E ) , indicating that targeting of Gal80 by at least these three F-box proteins is required for the efficient galactose-induced protein degradation of Gal80 . Gal80 interacted with all three F-box proteins in cells grown with glucose and in cells grown with galactose . Consistently , the deletion of MDM30 , DAS1 , and UFO1 stabilized Gal80 also in glucose-grown cells ( Figures 3B and S7B ) . The signal observed for the pulldown of the F-box proteins with GST-Gal80 was higher in glucose-grown cells than in galactose-induced cells ( compare lanes 8 and 9 in Figures 3E and S7D , E ) . A possible explanation is that in galactose-induced cells , more than in glucose-grown cells , the protein-protein interaction between the F-box proteins and Gal80 resulted in the protein degradation of Gal80 , which means that the amount of the F-box protein pulled by GST-Gal80 does not necessarily reflect the strength of the protein-protein interaction . The over-expression of Mdm30 and Ufo1 suppressed the gal− phenotype of cells lacking Das1 ( Figure S7F , lanes 3 and 4 ) , indicating that galactose induction requires a critical threshold of Gal80-targeting SCF E3 ubiquitin ligases . SCF E3 ubiquitin ligases are enzymes that not only target Gal80 for ubiquitin proteasome-mediated protein degradation but also other proteins like Gal4 [20] and Mig2 [37] . It could be argued that defects in the protein degradation of some protein other than Gal80 had caused the gal− phenotype of the H10UbD58A , skp1dM , and ΔMDM30 mutant strains . We have shown that the additional gene deletion of GAL80 suppressed the transcriptional defects of all of these mutants , indicating that Gal80 is the only functionally relevant target , but in order to gain independent evidence that the galactose-induced protein degradation of Gal80 is required for the galactose induction of the GAL genes , we sought to generate a galactose-stable Gal80 derivative that would interfere with transcriptional activation of the GAL genes . Some degraded proteins contain an N-terminal degron , and we performed a series of small N-terminal deletions of Gal80 and tested them for causing defects in galactose utilization . The over-expression of wild-type HA-Gal80 reduced growth on a galactose plate in the presence of the respiration inhibitor Antimycin A ( Figure 4A , line 2 ) . The successive deletion of two amino acids increased the growth inhibition , with the deletion derivative lacking the 12 N-terminal amino acids of Gal80 showing the biggest growth inhibition ( Figure 4A , line 6 ) . N-terminal deletions of more than 12 amino acids resulted in less inhibition , with the Gal80 deletion derivative lacking the N-terminal 20 amino acids ( which removes the first four residues of the Rossmann-fold [44] ) having lost the ability to inhibit growth on the galactose plate ( Figure 4A , line 10 ) . Real-time PCR quantification of GAL1 mRNA relative to ACT1 mRNA showed that the over-expression of the HA-Gal80 derivative lacking the N-terminal 12 amino acids reduced galactose induction of the GAL1 gene 5- to 3-fold more than the over-expression of wild-type HA-Gal80 ( Figure 4B ) . Cycloheximide chase assays demonstrated that the HA-Gal80 deletion derivative lacking the N-terminal 12 amino acids was indeed stable in galactose-grown cells ( Figure 4C , lanes 19 to 24; Figure 4D , black bars ) , confirming our hypothesis that galactose induction of the GAL1 gene requires protein degradation of the repressor Gal80 . The essential Mediator subunit Srb7 ( Med21 ) plays a pivotal role in the regulation of transcription [45] , [46] . In order to identify human proteins interacting with the human Mediator component hSrb7 , we fused it to the C-terminal half of ubiquitin that was extended by the RUra3 reporter ( Cub-RUra3 ) and performed a Split-Ubiquitin screen [47] , [48] with an expression library of human cDNAs fused to the N-terminal half of ubiquitin ( Nub; Figure S8A ) . The Nub fusion of the human SCF E3 ubiquitin ligase component hSkp1 was isolated by its ability to confer FOA resistance to S . cerevisiae cells expressing hSrb7-Cub-RUra3 ( Figure S8B ) , indicating that both proteins interacted inside the yeast cells . E . coli–expressed GST-hSrb7 , but not GST , pulled down Nub-HA-hSkp1 from yeast extract ( Figure S8C , lane 6 ) , and E . coli–expressed GST-hSkp1 , but not GST , pulled down E . coli–expressed H6-HA-hSrb7 ( Figure S8C , lane 3 ) , demonstrating that both proteins interacted directly with each other also in vitro . The human Split-Ubiquitin system ( Figure S8D; [49] ) was used to demonstrate that both proteins interacted with each other also in vivo ( Figure S8E ) . hSrb7 and hSkp1 are subunits of distinct protein complexes , but the SCF component hSkp1 might play an additional role as a component of Mediator , while the Mediator component hSrb7 might moonlight as a component in an SCF complex . In order to distinguish between these possibilities , we performed co-immunoprecipitations with HeLa extracts and found that hSkp1 pulled down other Mediator components like hMed6 ( Figure S9A , lane 3 ) , while hMed6 pulled down other SCF components like hCul1 ( Figure S9A , lane 10 ) , indicating that hSrb7 and hSkp1 interacted with each other as components of their own respective complexes . We knocked down hSrb7 and hSkp1 in HeLa cells by RNA interference ( Figure S9B ) , which dramatically reduced the heat-shock induction of the human HSP70B' gene ( Figure S9C ) , indicating that hSrb7 and hSkp1 are functionally relevant for transcription in human cells . Skp1 is a component of the SCF E3 ubiquitin ligases , suggesting that protein degradation could be an important aspect of how Srb7 regulates transcription . The Split-Ubiquitin assay revealed that also the S . cerevisiae Srb7 and Skp1 proteins interacted with each other in vivo ( Figure 5A , line 2 ) . Interestingly , Skp1dM was defective for the protein interaction with Srb7 ( Figure 5A , line 4 ) . Our results showed that the Mediator of transcription interacts with SCF E3 ubiquitin ligases , and in order to see if Mediator plays a role in the galactose-induced protein degradation of Gal80 , we generated a gal− allele of SRB7 by replacing endogenous Srb7 with a GST fusion to a C-terminal fragment of Srb7 lacking the first 40 amino acid residues ( Figure 5B , line 2 ) . The over-expression of Gal3 and the deletion of GAL80 suppressed the gal− phenotype of the GST-Srb7Δ40 strain ( Figure 5B , compare lines 1 to 4 ) , indicating that excess Gal80 could have caused the gal− phenotype . The over-expression of Gal3 and the deletion of GAL80 did not suppress the gal− phenotype of cells lacking the Mediator subunit Gal11 ( Figure 5B , compare lines 5 to 7 ) , demonstrating that the suppression had been gene-specific and that the Mediator subunit Srb7 acts genetically upstream of Gal80 , while the Mediator subunit Gal11 acts genetically downstream of Gal80 . The over-expression of α2 and the deletion of MIG2 did not suppress the gal− phenotype of the GST-Srb7Δ40 strain ( Figure 5B , compare lines 8 to 11 ) , while the over-expression of α2 and the deletion of MIG2 had suppressed ( partially ) the gal− phenotype of the skp1dM strain ( Figure 2C , line 4 and [37] ) , suggesting that Skp1 acts genetically upstream of both Gal80 and Mig2 , while Srb7 acts genetically upstream of Gal80 only . Cycloheximide chase assays demonstrated that Gal80 was degraded in galactose-induced cells expressing wild-type Srb7 ( Figure 5C , lanes 5 to 8; Figure 5D , white bars ) , but stable in galactose-induced GST-Srb7Δ40 cells ( Figure 5C , lanes 13 to 16; Figure 5D , black bars ) , indicating that Mediator controls the galactose-induced protein degradation of Gal80 . Real-time PCR quantification confirmed that galactose induction of GAL1 mRNA relative to ACT1 mRNA was abolished in the GST-Srb7Δ40 strain and that it was almost fully restored by the over-expression of Gal3 and the deletion of GAL80 ( Figure 5E ) , suggesting that the failure of the GST-Srb7Δ40 strain to degrade Gal80 upon galactose induction had been the main cause for the failure to activate the transcription of the GAL1 gene . GST-Srb7 , but not GST , pulled down Skp1 from yeast extract , while GST-Srb7Δ40 failed to do so ( Figure 5F , lanes 5 and 6 ) , indicating that the protein-protein interaction with Skp1 is mediated by the N-terminus of Srb7 , which is the most conserved part of the protein [46] . Our results have shown that the degradation of Gal80 was abolished when endogenous Skp1 was replaced by a mutant Skp1 derivative that failed to interact with Srb7 and when endogenous Srb7 was replaced by a Srb7 mutant protein that failed to interact with Skp1 , suggesting that the protein-protein interaction between the Mediator component Srb7 and the SCF component Skp1 is required for the protein degradation of Gal80 . Mediator acts upstream of the activator Gal4 by controlling the galactose-induced protein degradation of the inhibitor Gal80 . But how does Mediator know about the switch in carbon source ? The protein kinase Snf1 is required for the transcription of glucose-repressed genes in S . cerevisiae , and the deletion of SNF1 resulted in the failure to degrade Gal80 ( Figure 6A , lanes 19 to 24; Figure 6B , grey bars ) , to utilize galactose ( Figure 6C ) , and to activate the GAL1 gene under inducing conditions ( Figure 6D ) . The activating gamma subunit Snf4 is required for the kinase activity of the SNF1 complex and Gal80 was also stable in galactose-induced ΔSNF4 cells ( Figure 6A , lanes 31 to 36; Figure 6B , grey bars ) , indicating that the kinase activity of the SNF1 complex is required for the degradation of Gal80 . The additional gene deletion of GAL80 fully suppressed the transcriptional defect of ΔSNF1 and ΔSNF4 cells ( Figure 6C , lines 3 and 4; Figure 6D ) , but no interaction was observed between Snf1 and Gal80 in a pulldown assay ( Figure S10 ) , indicating that Snf1 controls GAL1 expression mainly by targeting Gal80 via Srb7 and SCF E3 ubiquitin ligases . The Split-Ubiquitin assay did not reveal an interaction between Snf1 and Srb7 ( Figure S11 , line 23 ) , however Srb7 is a component of Mediator and Snf1 interacted with the Mediator components Med6 ( Figure S11 , line 6 ) , Med10 ( Figure S11 , line 11 ) , Srb6 ( Med22; Figure S11 , line 21 ) , and Srb11 ( CycC; 11 , line 27 ) . The protein interaction between the kinase Snf1 and the Mediator component Srb11 had been observed both in vivo and in vitro previously [50] , [51] . Srb11 is a cyclin-like cofactor for the protein kinase Srb10 ( Cdk8 ) , and the Mediator components Srb10 and Srb11 are both required for the full transcriptional activation of the GAL1 gene [52] . Gal80 was stable in galactose-induced ΔSRB10 and ΔSRB11 cells ( Figure S12A , lanes 7 to 12 and 19 to 24; Figure S12B ) , confirming that Snf1 might transduce the signal to degrade Gal80 via the Mediator subunit Srb11 . The additional gene deletion of GAL80 suppressed the galactose utilization defect of cells lacking Srb10 and Srb11 ( Figure S12C , lines 3 and 5 ) , providing genetic evidence that galactose-stable Gal80 had caused the gal− phenotype of ΔSRB10 and ΔSRB11 cells . The SNF1 kinase is activated by the absence of glucose , but transcriptional activation of the GAL genes requires additionally the presence of galactose , as transcription of GAL1 is not activated in cells grown with—for example—raffinose ( Figure S13B ) . Consistently , Gal80 was more stable in cells grown with raffinose than in cells grown with galactose ( Figure S14 ) . The half-life of Gal80 in BY4741ΔW cells was calculated to be approximately 3 h when the cells were grown with glucose , 2 h when the cells were grown with raffinose , 1 h when galactose-induced cells had been pre-grown with glucose , and half an hour when the galactose-induced cells had been pre-grown with raffinose . However , our observations also indicate that active SNF1 kinase is necessary but not sufficient for the galactose-stimulated protein degradation of Gal80 . An additional transducer that signals the presence of galactose is apparently required . A possible candidate for such a signal transducer is Gal3 , as it is known to bind both galactose and Gal80 [14] . Cells lacking Gal3 display a strong gal− phenotype ( Figure S15A , lines 3 and 4 ) , which is suppressed by the additional gene deletion of GAL80 ( Figure S15A , lines 5 and 6 ) , but the degradation of Gal80 in galactose-induced cells remained unchanged upon the deletion of GAL3 ( Figure S15B , lanes 5 to 8; Figure 15C ) , indicating that Gal3 does not play a role in the galactose-induced protein degradation of Gal80 and that galactose must utilize another transducer to stimulate the protein degradation of Gal80 .
Mediator was isolated by its ability to respond to transcriptional activators , and all studies published about Mediator have focused on the role of Mediator past its recruitment to the promoter by the activator [28] . Once recruited , Mediator is required to recruit the General Transcription Factors and RNA Polymerase II and to initiate transcription [29] . Mediator also affects post-initiation steps by affecting transcription elongation and chromatin structure [31] . We have shown here that Mediator additionally acts upstream of the activator Gal4 by controlling the degradation of the inhibitor Gal80 . In cells grown with glucose , Gal80 binds to the activation domain of Gal4 and prevents it from activating transcription . Upon galactose induction , Mediator initiates the degradation of Gal80 via its interaction with the SCF E3 ubiquitin ligase component Skp1 . Therefore , Mediator actually orchestrates its own recruitment to the GAL1 promoter by regulating the activity of Gal4 ( Figure 7 ) . SCFMdm30 targets not only Gal80 but also Gal4 in galactose-induced cells , leading to the mono-ubiquitination and subsequent poly-ubiquitination and protein degradation of Gal4 [20] . In galactose-induced cells lacking Mdm30 , Gal4 is no longer ubiquitinated and no longer degraded [20] . One could argue that changes in the proteolytic stability of Gal4 or in its mono-ubiquitination status might have been the cause for the gal− phenotypes that we have observed for the various H10UbD58A , skp1 , mdm30 , srb7 , and snf1 mutant strains described here . Therefore , it is important to note that our claim that the degradation of Gal80—and not the degradation of Gal4—is required for the transcriptional activation of the GAL genes is not just based on a simple correlation between the proteolytic stability of Gal80 and the inability of the cell to activate transcription of the GAL1 gene , but on functional suppression . The additional gene deletion of GAL80 fully suppressed the transcriptional defects of the H10UbD58A , skp1 , mdm30 , srb7 , and snf1 mutant strains . This means that in the absence of Gal80 , Gal4 activated transcription in all these mutant strains just fine , which demonstrates that any effects that these strain mutations might have had on Gal4 were not relevant for Gal4's function as a transcriptional activator . Therefore , while mono-ubiquitination of Gal4 was certainly affected in the H10UbD58A strain ( since endogenous wild-type ubiquitin had been replaced with H10UbD58A ) , Gal4-H10UbD58A fully activated transcription of the GAL1 gene in the absence of Gal80 , suggesting that H10UbD58A still protected Gal4 from the UAS-stripping activity of the 19S proteasome [25] . Furthermore , Gal4 fully activated transcription in cells lacking both Mdm30 and Gal80 , which argues that Gal4 does not have to be degraded to become transcriptionally active . In addition , we have generated a galactose-stable Gal80 derivative that inhibited galactose induction in otherwise wild-type cells , which means that we have presented evidence for our claim that galactose induction requires Gal80 degradation that did not rely on a mutant strain background . The deletion of the three F-box protein-coding genes MDM30 , DAS1 , and UFO1 completely abolished galactose induction of GAL1 mRNA ( Figure 3D and [37] ) . Das1 and Ufo1 ( but not Mdm30 ) also target the repressor Mig2 for galactose-induced protein degradation [37] . However , the additional gene deletion of MIG2 did not increase galactose induction of GAL1 mRNA in the ΔUFO1 strain and had only a very small effect on the galactose induction of the GAL1 mRNA in the ΔDAS1 strain [37] . Therefore , an additional target for Das1 and Ufo1 had been proposed , and we have now shown here that Gal80 is this functionally relevant target , as Gal80—like Mig2 [37]—became stable in galactose-induced cells lacking Das1 and Ufo1 ( Figure S7A and S7B ) , and the additional gene deletion of GAL80 suppressed the gal− phenotype of both the ΔDAS1 and the ΔUFO1 strains ( Figure S7C ) . We are proposing that a critical concentration of the three F-box proteins Mdm30 , Das1 , and Ufo1 is required for the galactose-stimulated protein degradation of Gal80 . If any one of these three F-box proteins is missing , the concentration of the remaining two F-box proteins is insufficient for targeting of Gal80; Gal80 is not degraded and excess Gal80 prevents Gal4 from activating the GAL genes under inducing conditions . In support of this model ( Figure 7 ) , we were able to show that the gal− phenotype of ΔDAS1 cells was suppressed by the over-expression of Ufo1 and Mdm30 ( Figure S7E ) . Gal3 sequesters Gal80 in the cytoplasm upon galactose induction [14] . The gene deletion of GAL3 had caused a gal− phenotype that was suppressed by the additional gene deletion of GAL80 , but the protein degradation of Gal80 was still stimulated in galactose-induced cells lacking Gal3 ( Figure S15 ) , indicating that instable Gal80 was not sufficient to allow Gal4 to activate transcription in the absence of Gal3 . On the other hand , sequestration of Gal80 into the cytoplasm by endogenous levels of Gal3 was not sufficient to allow Gal4 to activate transcription in the presence of stable Gal80 . Apparently , sequestration of Gal80 into the cytoplasm by Gal3 and ubiquitin-mediated protein degradation of Gal80 are both required for the galactose induction of the GAL genes . Contrary to a previous report [20] , we found that the deletion of the gene encoding the F-box protein Mdm30 abolished galactose induction of the GAL1 mRNA . We have grown the cells in glucose liquid media prior to the switch to galactose liquid media—which is consistent with the switch in carbon sources conducted for the plate assay—while Muratani et al . grew the cells in raffinose liquid media prior to the switch to galactose liquid media . In order to determine if this difference in protocols was the cause for the difference in results , we performed the galactose induction with cells that had been pre-grown in raffinose , and we found that in this case , galactose-induced protein degradation of Gal80 and galactose induction of GAL1 mRNA relative to ACT1 mRNA were restored in the ΔMDM30 strain ( Figures S13B and S14 ) . One other remarkable difference between the two growth protocols is the speed of induction . Galactose induction of GAL1 mRNA relative to ACT1 mRNA takes 4 h if the cells are pre-grown with glucose and only 1 h if the cells are pre-grown with raffinose ( Figure S16 ) . Consistently , cycloheximide chase assays demonstrate that Gal80 is more rapidly degraded in galactose-induced cells when the cells had been pre-grown with raffinose instead of with glucose ( compare Figures 3B and S14B ) . The half-life of Gal80 in galactose-induced BY4741ΔW cells was approximately 1 h when the cells had been pre-grown with glucose and only half an hour when the cells had been pre-grown with raffinose . The correlation of the kinetics of galactose-induced Gal80 destruction and GAL1 mRNA production suggests that the degradation of Gal80 is the rate-limiting step for the galactose induction of the GAL1 gene .
The S . cerevisiae strain SUB288 [53] has all chromosomal ubiquitin genes deleted and allows for the expression of plasmid-born ubiquitin derivatives in place of endogenous ubiquitin ( see Table S1 for the genotypes of the strains and Table S2 for the sequences of PCR primers ) . However , the strain fails to grow on galactose plates containing the respiration inhibitor Antimycin A ( AA ) . Transformation of the strain with single-copy vectors expressing Gal3 from its own promoter allowed the strain to grow on galactose AA plates and sequencing of the chromosomal GAL3 gene demonstrated that SUB288 carries a frame shift in the third codon of GAL3 . The TRP1 and LEU2 genes were deleted and the defective gal3 gene was repaired by homologous recombination with a wild-type GAL3 PCR fragment followed by selection on a galactose AA plate or with YIplac204-GAL3 , a derivative of the TRP1-marked integrative vector YIplac204 [54] containing the GAL3 gene , resulting in SUB288GAL3ΔWL+316-Ub and SUB288GAL3ΔL+316-Ub , respectively . DNA sequencing of PCR fragments derived from genomic DNA was used to confirm that the GAL3 gene had been repaired . The ubiquitin point mutants were generated by two-step PCR with degenerate primers and cloned into the LYS2-marked single-copy vector RS317 [55] containing the ACT1 promoter-terminator cassette and into RS317 expressing 10 histidines from the ACT1 promoter . The 317-Ub and 317-H10-Ub plasmids were transformed into SUB288GAL3ΔWL+316-Ub and 316-Ub was shuffled out on FOA plates . All ubiquitin mutant strains were confirmed by DNA sequencing . The GAL80 gene of SUB288GAL3ΔWL+317-Ub was knocked out with a derivative of NKY51 [56] , which carried the hisG-URA3-hisG cassette in the BglII site at nucleotide 612 of GAL80 . 317-Ub was replaced by 316-Ub via plasmid loss , and plasmid shuffle was used to generate the 317-H10-Ub and 317-H10-UbD58A strains carrying hisG integrated into GAL80 . The essential SRB7 gene of JD52 and JD52ΔGAL80 was knocked out with a PCR fragment containing the HIS3 gene flanked by 50 bp of SRB7 promoter and terminator in the presence of 33-SRB7 , a derivative of the URA3-marked single-copy vector YCplac33 [54] that expressed Srb7 from its own promoter . GST-Srb7 and GST-Srb7Δ40 were expressed from the TRP1-marked multi-copy vector YG1μ under the control of the ADH1 promoter . BY4741ΔW and BY4742ΔW and their gene deletion derivatives were obtained from the respective EUROSCARF strains by inserting hisG into the TRP1 gene with the help of NKY1009 [56] . YEp13-GAL3 was isolated from a LEU2-marked multi-copy YEp13-based genomic DNA library [37] as a multi-copy suppressor of the gal− phenotype of the H10UbD58A strain . YEp13-GAL3 contains a 2 , 643 bp genomic DNA fragment with the entire GAL3 gene , including 842 bp of promoter and 238 bp of terminator DNA . 112-GAL3 is a derivative of the TRP1-marked multi-copy vector YEplac112 [54] containing the genomic GAL3 fragment . 314-Gal3 is a derivative of the TRP1-marked single-copy vector RS314 [55] , expressing Gal3 from the ACT1 promoter . 315-Gal3 is a derivative of the LEU2-marked single-copy vector RS315 [55] , expressing Gal3 from the ACT1 promoter . 316-HA-Gal80 is a derivative of RS316 , expressing Gal80 from the ACT1 promoter . The N-terminal deletion derivatives of Gal80 were cloned into the same vector . 423-HA3-Mdm30 , 423-HA3-Das1 , and 423-HA3-Ufo1 are derivatives of the HIS3-marked multi-copy vector RS423 [55] , expressing Mdm30 , Das1 , and Ufo1 tagged with three HA epitopes from the ACT1 promoter . 424-GST and 424-GST-Gal80 are derivatives of the TRP1-marked multi-copy vector RS424 [55] , expressing GST and GST-Gal80 from the ACT1 promoter . YIplac128-Snf1c-HA3H10 is a derivative of the LEU2-marked integrative vector YIplac128 [54] , containing a C-terminal BglII-SalI fragment of SNF1 lacking the stop codon , and YIplac128-Skp1c-HA3H10 is a derivative of YIplac128 containing a C-terminal EcoRI-SalI fragment of SKP1 lacking the stop codon . Snf1-HA3H10 was expressed from the SNF1 promoter following digestion with MluI and integration into the SNF1 locus , while Skp1-HA3H10 was expressed from the SKP1 promoter following digestion with AvaI and integration into the SKP1 locus . A Clontech library derived from human B-cell cDNAs was partially digested with Sau3A and cloned into the BglII site of PADNX-Nub-IBC [57] in all three reading frames , resulting in 60 , 000 independent DH5α transformants . hSrb7 was cloned into Pcup1-Cub-RUra314 [57] and transformed together with the Nub library into JD52 [58] , resulting in 160 , 000 transformants , which were plated onto FOA plates containing 10 µM CuSO4 . The Nub plasmids from the 10 arising colonies were isolated and transformed back into JD52 containing hSrb7-Cub-Ura314 . Only one was plasmid-linked , and it contained the entire hSkp1 open reading frame fused to Nub-HA . HeLa cells were grown to 80% confluency and transfected with 2 µg of pSuper ( OligoEngine ) construct and 5 µl of lipofectamin in serum-free DMEM for 5 h before being transferred into regular DMEM . The three constructs used were an empty vector as a negative control , siRNA specific for hSKP1 , and siRNA specific for hSRB7 . 48 h after transfection , one set of cells was heat-shocked at 45°C for 15 min and allowed to recover for 1 h in a 37°C incubator . A non-heat-shock sample was also incubated at 37°C for an identical length of time . These cells were then harvested by trypsinization and their mRNA was extracted using a Qiagen RNA Easy Kit . 300 nM of mRNA was utilized for reverse transcription primed by random hexamers , and the cDNA was quantified using Sybr-Green in an ABI Prism . Primers for HSP70B' mRNA were 5′-ccccatcattgaggaggttg-3′ and 5′-gaagcagaagaggatgaacc-3′ . Primers for hSKP1 mRNA were 5′-gcaaagagaaccagtggtgtga-3′ and 5′-aggtttgggatctgtgctcaa-3′ . Primers for hSRB7 mRNA were 5′-aatgtggtcctcctgcctctt-3′ and 5′-ccagaagcatgtctcctcgata-3′ . Primers for GAPDH mRNA were 5′-ctctctgctcctcctgttcgac-3′ and 5′-tgagcgatgtggctcggct-3′ . S . cerevisiae cells were cultured in synthetic complete 2% ( w/v ) glucose medium at 28°C . At OD600 nm = 1 , the cells were collected by centrifugation . Galactose induction was performed by resuspending the cells in 2% galactose medium and incubation for the indicated amount of time . Total RNA was isolated using the RNAeasy Mini Kit ( Qiagen ) according to the manufacturer's protocol . cDNA was generated by reverse transcription PCR using Taqman MicroRNA Reverse Transcription Kit ( Roche Applied Biosystems ) . Quantitative real-time PCR was performed using SYBR Green PCR Master Mix ( Applied Biosystems ) . Primers used for ACT1 mRNA were 5′-gaccaaactacttacaactcca-3′ and 5′-cattctttcggcaatacctg-3′ . Primers used for GAL1 mRNA were 5′-acttgcaccggaaaggtttg-3′ and 5′-ttggtacatcaccctcacagaaga-3′ . All mRNA quantifications were performed three times , and the error bars represent the standard deviations . HeLa cells were grown to 80% confluency before they were transfected with 2 µg of pCMV-myc-hSKP1 or pCMV-myc vector and 5 µl of lipofectamin in serum-free DMEM for 5 h before being transferred into regular DMEM . The cells were harvested 48 h after transfection by trypsinization and lysed in 1× PBS by freeze-thaw . The cell lysate was subsequently agitated on a rotor with 2 µl of anti-myc affixed agarose beads ( Sigma ) in 500 µl of ice cold 1× PBS overnight . The beads were washed four times with 1 ml PBS prior to heat elution at 95°C for 15 min . Proteins were separated on a 12% gel , transferred to a nitrocellulose membrane , which was probed with anti-Med6 rabbit polyclonal antibody ( Abcam ) . HeLa cells were grown to 80% confluency before they were harvested by trypsinization and lysed in 1× PBS by freeze-thaw . The cell lysate was diluted 1∶5 with RIPA buffer ( 50 mM Tris-HCl ph 8 , 150 mM NaCl , 2 mM EDTA , 1% NP-40 , 0 . 5% Sodium deoxycholate , 0 . 1% SDS ) and incubated with 10 µl of rProtein G Sepharose ( GE Healthcare ) as well as 5 µl of anti-Med6 rabbit polyclonal antibody ( Abcam ) or anti-Cul1 mouse monoclonal antibody ( Abcam ) for 3 h . The sepharose was washed four times with 1 ml PBS and heat eluted at 95°C for 15 min . Proteins were separated on a 12% gel and transferred to a nitrocellulose membrane , which was probed with the reciprocal antibody ( anti-Cul1 mouse monoclonal antibody ( Abcam ) or anti-Med6 rabbit polyclonal antibody ( Abcam ) , respectively ) . GST pulldown assays were performed using whole cell S . cerevisiae extracts prepared by bead beating in yeast lysis buffer ( 100 mM Tris pH 7 . 5 , 50 mM KCl , 1 mM EDTA , 0 . 1% NP40 ) and whole cell E . coli extracts prepared by freeze-thaw in PBS ( Phosphate-Buffered Saline ) . 500 µl of whole cell extract was added to equilibrated glutathione beads ( Amersham Biosciences ) containing 2 mM PMSF and 1 mM DTT . The reaction mixture was incubated at 4°C for 1 h . The sample was centrifuged at 3 , 000 rpm and the supernatant was removed . The glutathione beads were washed five times before Western Blot analysis . S . cerevisiae cells were grown in 50 ml synthetic complete 2% glucose medium to OD600 nm = 1 and harvested by centrifugation . The cell pellets were suspended in 1 ml yeast breaking buffer ( Triton X-100 , 10% SDS , 5 M NaCl , 1 M Tris-Hcl pH 8 , 0 . 5 M EDTA; Figure 5D ) or yeast lysis buffer ( Figure S6 ) , pipetted into a screw-cap microcentrifuge tube containing acid-washed glass beads ( Sigma-Aldrich , USA ) , and 2 mM PMSF was added . The tubes were then subjected to homogenization with a bead beater for 1 min and then rested on ice for 3 min . This process was repeated for three times . The samples were then centrifuged for 15 min at 13 , 000 rpm , and the supernatants were incubated with 10 µl of equilibrated nickel beads for 1 h at 4°C . After incubation , the samples were centrifuged at 3 , 000 rpm for 2 min . The nickel beads were washed with 1 ml yeast breaking/lysis buffer containing 20 mM imidazole . This washing process was repeated five times . The bound protein was eluted from the nickel beads using 100 µl of yeast breaking/lysis buffer with 500 mM imidazole for 30 min . This process was repeated two times . The supernatant was collected and stored at −80°C . S . cerevisiae cells were grown in liquid drop out media containing 2% glucose or raffinose to OD600 nm = 1 . Half of the cultures were induced in liquid media containing 2% galactose for 1 h before the addition of 200 mg/l cycloheximide ( Sigma ) . Aliquots were taken at the indicated time points , and cellular proteins were analyzed by Western Blot with primary antibodies against hemagglutinin ( HA; Roche ) and carboxypeptidase Y ( CPY; Molecular Probes ) , followed by staining with a horseradish peroxidase-coupled secondary anti-mouse IgG antibody and by Coomassie Brilliant Blue ( Sigma ) staining . The intensities of the bands were quantified with Image J ( rsb . info . nih . gov/ij/index . html ) . The ratio of the band intensities before the addition of cycloheximide ( time = 0 ) was set as 1 , and the error bars represent the deviations between duplicates . Representative Western blots are shown . No significant differences were observed when the HA-Gal80 bands were normalized to CPY or to Coomassie staining . The half-life of Gal80 was calculated using trendline ( excel ) . | The expression levels of proteins are tightly regulated , not only via their production but also via their degradation . Genes are transcribed only if their encoded proteins are required by the environmental or developmental conditions of a cell , and once a certain protein is no longer needed , it is rapidly degraded by the ubiquitin proteasome system ( UPS ) . Transcriptional activators appeared to contradict this simple economic principle , as it had been claimed that they had to be degraded in order to function . The claim was based upon a correlation: if the degradation of an activator was prevented by drugs or mutations in the UPS , the activator became stable but also nonfunctional . We have now shown that it is not the activator itself but its inhibitor that is the functionally relevant target of the UPS . Furthermore , we have found that the degradation of the inhibitor is controlled by a protein complex called Mediator . The activator is known to recruit Mediator to gene promoters , where Mediator assists RNA polymerase in initiating transcription . Mediator was always considered to be completely under the control of the activator; however , we observe that by regulating the degradation of the inhibitor , Mediator is also able to control the activator and thereby to orchestrate its own recruitment to gene promoters . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"proteins",
"genetics",
"molecular",
"genetics",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Mediator Acts Upstream of the Transcriptional Activator Gal4 |
The Second Heart Field ( SHF ) has been implicated in several forms of congenital heart disease ( CHD ) , including atrioventricular septal defects ( AVSDs ) . Identifying the SHF gene regulatory networks required for atrioventricular septation is therefore an essential goal for understanding the molecular basis of AVSDs . We defined a SHF Hedgehog-dependent gene regulatory network using whole genome transcriptional profiling and GLI-chromatin interaction studies . The Forkhead box transcription factors Foxf1a and Foxf2 were identified as SHF Hedgehog targets . Compound haploinsufficiency for Foxf1a and Foxf2 caused atrioventricular septal defects , demonstrating the biological relevance of this regulatory network . We identified a Foxf1a cis-regulatory element that bound the Hedgehog transcriptional regulators GLI1 and GLI3 and the T-box transcription factor TBX5 in vivo . GLI1 and TBX5 synergistically activated transcription from this cis-regulatory element in vitro . This enhancer drove reproducible expression in vivo in the posterior SHF , the only region where Gli1 and Tbx5 expression overlaps . Our findings implicate Foxf genes in atrioventricular septation , describe the molecular underpinnings of the genetic interaction between Hedgehog signaling and Tbx5 , and establish a molecular model for the selection of the SHF gene regulatory network for cardiac septation .
Cardiac septation , the morphogenetic process that transitions the looped heart tube into the multi-chambered heart observed in mammals , is complex and often goes awry in Congenital Heart Disease ( CHD ) . Atrioventricular septation is the crucial process that separates the common atrioventricular canal into right and left compartments . Atrioventricular septal defects ( AVSDs ) are a common severe form of CHD . A novel paradigm for the developmental ontogeny of the atrioventricular septum has recently emerged [1]–[6] . This work describes atrioventricular septation as a process driven by molecular events in second heart field ( SHF ) cardiac progenitors rather than in the heart itself [1]–[6] . The identification of extracardiac lineages that generate the atrial and atrioventricular septum implies that the search for gene regulatory networks germane to cardiac septation should occur in SHF cardiac progenitors not in the heart itself . Hedgehog signaling is an essential developmental pathway conserved from flies to man [7] , [8] . Mutations in key Hedgehog pathway genes , including ligands such as Sonic hedgehog ( Shh; 20423 ) and downstream signaling cascade member Smoothened ( Smo; 319757 ) cause significant cardiac defects including complete atrioventricular septal defects [9] , [10] . Tissue specific knockout of Hedgehog signaling in the SHF recapitulates atrioventricular septal defects [4] , [5] and genetic inducible fate mapping showed that the atrial/atrioventricular septum is derived from Hedgehog-receiving SHF cardiac progenitors [5] . These observations laid the groundwork for identifying the Hedgehog-dependent SHF gene regulatory networks essential for atrial septation . Cardiogenic transcription factor genes Tbx5 ( 21388 ) , Nkx2 . 5 ( 18091 ) and GATA4 ( 14463 ) have been implicated in human atrial septation [11]–[14] . These transcription factors form a complex and can co-activate gene expression [12] , [15]–[17] . Tbx5 has been shown to be required in multiple contexts during cardiac development and adult function in mice . Tbx5 is required in the SHF for atrioventricular septation [6] , [15] , in embryonic cardiomyocytes for proliferation [18] , in adult myocardium for contractile function [19] , and in the adult cardiac conduction system for cardiac rhythm control [20] . Tbx5 target genes differ significantly between these distinct cellular and temporal contexts [6] , [21] . Yet the Tbx5-responsive cis-regulatory elements specific to these cellular contexts and the molecular cues that establish context dependent selectivity remain unknown . We previously described genetic interactions between Tbx5 and Hedgehog signaling in the SHF for atrioventricular septation in mice [6] . Mice haploinsufficient for both Tbx5 and the obligate Hedgehog signaling receptor gene Smo express AVSDs more frequently than mice haploinsufficient for either gene alone [6] . Furthermore , constitutive Hedgehog signaling in Tbx5-mutant SHF progenitors can rescue atrioventricular septation [6] . These studies predict that Hedgehog-dependent and Tbx5-dependent gene regulatory networks share vital , yet undescribed overlap in the SHF that is necessary for atrioventricular septation . In this study we attempted to define Hedgehog-dependent SHF gene regulatory networks and identify the molecular basis of the genetic interaction between Hedgehog signaling and Tbx5 . We characterized the Hedgehog-dependent SHF gene regulatory networks by in vivo whole genome transcriptional profiling and GLI-chromatin interaction studies . We found that Foxf1a ( 15227 ) and Foxf2 ( 14238 ) are downstream of Hedgehog signaling in the SHF . Mice haploinsufficient for both Foxf1a and Foxf2 compound heterozygotes have atrial septal defects , demonstrating the biological relevance of these Hedgehog targets . GLI3T ( 14634 ) binding data identified a candidate cis-regulatory element upstream of Foxf1a that contained an adjacent Tbx5 binding site . This enhancer binds to GLI1 ( 14632 ) , GLI3 and TBX5 in the SHF in vivo . In vitro and in vivo analysis demonstrated that this cis-regulatory element integrates Hedgehog signaling with Tbx5 activity and provides strong specific activity in the posterior SHF . This work identifies a novel role for Foxf transcription factors at the intersection of Tbx5 and hedgehog signaling in atrioventricular septation and describes a SHF gene regulatory network for cardiac morphogenesis .
Progenitor cells for the atrial and atrioventricular septum require Shh signaling in the posterior SHF ( pSHF ) between embryonic day 8 and embryonic day 10 ( E8–E10 ) to migrate into the heart to form the atrial septum between E9–E11 [4] , [5] . To identify the Hedgehog-dependent gene regulatory networks required for this process , we compared transcriptional profiling of the posterior SHF from wild-type and Shh ( MGI: 1932461 ) null embryos at E9 . 5 to identify differentially expressed transcripts . We isolated the pSHF by microdissection including the dorsal mesenchymal protrusion and closely associated surrounding ventral lateral plate mesenchyme . Our dissection included the attached foregut , but excluded the heart , dorsal lateral plate mesenchyme and neural tube ( Figure 1A ) . RNA was isolated and known Hedgehog-dependent transcripts were evaluated by RT-PCR to verify genotyping prior to whole genome transcriptional profiling . Shh , Ptch1 ( 19206 ) and Gli1 all demonstrated significantly reduced expression ( p>0 . 05 ) in the Shh null samples compared to wild-type micro-dissected samples ( Figure 1B ) . Specifically , Shh was reduced more than 90% , while Ptch1 and Gli1 were each reduced approximately 50% , consistent with significantly reduced Hedgehog signaling in the mutant samples and confirming the genotypic fidelity of the isolated samples . Transcriptional profiling of pSHF samples was performed on Agilent Mouse Whole Genome Arrays . Using a significance threshold with a multi-test adjusted p-value ( Q-value ) <0 . 005 and absolute fold change larger than 2 , comparing Shh−/− mutant mouse embryos ( n = 4 ) with wild-type embryos ( n = 3 ) identified a differentially expressed 560-gene signature ( Table S1 ) . Gene Ontology ( GO ) enrichment analysis of differentially expressed genes captured known processes disrupted in Hedgehog pathway mutants , such as pattern specification and organ morphogenesis ( Figure 1C ) [22] . To further identify the best candidates for an experimental validation , 65 genes were computationally evaluated according to more stringent criteria by three statistical tests ( non-parameter Wilcox-tested theoretical p<0 . 15 , empirical t-tested FDR<0 . 1 , and absolute fold change>3 , Figure S1 ) on the same data sets . From the Shh down-regulated candidates , we chose 21 targets and validated significant misexpression of 13 by qPCR ( p<2e-16 , Fisher's Exact test , FET ) ( Figure 1D ) . Eight others did not meet criteria for statistically significant misexpression primarily due to large expression variation , possibly related to the presence of non-SHF tissue isolated by our dissection process . To define loci directly downstream of Hedgehog signaling , we analyzed genome-wide chromosomal binding locations of the Hedgehog transcriptional regulator Gli3 in the embryonic SHF by chromatin immunoprecipitation with deep sequencing ( ChIP-seq ) . We performed ChIP using a Cre-inducible flag-tagged Gli3T expression line ( RosaGli3TFlag c/c MGI: 3828280 ) [23] combined with the SHF Cre driver Mef2c-AHF-Cre [24] ( MGI: 3639735 ) . The SHF tissue from 50 Mef2cAHF-Cre+; RosaGli3TFlag/+ embryos was micro-dissected and immunoprecipitated using an anti-FlagM2 antibody ( Sigma ) . To verify enrichment of Gli3T bound sequences by immunoprecipitation prior to sequencing , we tested a previously identified Gli3T peak upstream of Ptch1 ( Chromosome 13 , nucleotides 63577408–63579384 , mm9 ) , a known Gli3T-bound cis-regulatory element in the limb [23] . This sequence was 13 . 7-fold enriched in the SHF IP fraction by ChIP-PCR . We proceeded to sequence the IP library and apply Model-based Analysis for ChIP-Seq ( MACS ) [25] . We identified 1316 Gli3-bound peaks by comparing 68 million sequence tags in IP to 21 million sequence tags in input ( tag size = 36 bps , effective genome size = 2e+9 , band width = 200 , 2<model fold<200 , and p-value cutoff = 1e-05 ) [25] . From these peaks , we analyzed the distribution of the signal around the peak center and identified a typical distribution , confirming successful sequencing ( Figure 2A ) . The predominant GLI3T peak location from the binding data was intergenic and a considerable distance from the transcriptional start sites . We therefore considered the possibility that genes and up to 100 kbp in both directions from intergenic peaks may fall under control of GLI-mediated cis-regulatory elements , given that enhancers often reside thousands of base pairs away from their target of regulation and act independently of their orientation [26] , [27] . We therefore annotated GLI3T-bound regions to all transcription start sites within 100 kbp and to the nearest TSS if it resided outside the 100 kbp window [28] , [29] . This consideration resulted in mapping the 1316 peaks to 3296 neighbor genes ( Table S2 ) . The enrichment between GLI3T-bound and Shh-dependent genes was significant among approximately 22 , 000 mouse genes ( FET p<0 . 01 , Figure 2B ) . To define the direct Hedgehog-dependent SHF gene regulatory networks , we intersected the SHF Shh-dependent transcriptional profile signature with the SHF Gli3T chromatin contact results to define candidate Hedgehog-dependent Gli-target genes . This dataset intersection comprised 119 peaks annotated to 112 genes ( Figure 2B , Table S3 ) . The enrichment between Gli3T-bound and Shh-dependent genes was significant among ∼22k mouse genes ( FET p = 0 . 003 , odds ration = 1 . 4 , Figure 2B ) . The 119 Shh-dependent Gli3T-bound sites contained significant enrichment of the de novo and known Gli3-binding motif , as derived by ChIP-Chip ( CGTGGGTGGTCC ) [23] and by computational implication ( TRANSFAC database; Figure 2C , bottom panel ) [30] , [31] at a high degree of significance ( p≤1e-10; Figures 2C , top panel ) . A significant enrichment of transcription factors was observed in SHF Hedgehog target genes . Transcription factor activity and DNA binding were the two most significant gene-sets over-represented among the 112 Shh-dependent Gli3-bound genes . We directly analyzed our gene set for overrepresentation of transcription factors by searching TRANSFAC version 2013 . 1 [31] and identified 26 TFs among the 112 unique genes with significant Gli3T-bound peaks ( Figure 2D , Table S4 ) , representing a significant enrichment ( p = 0 . 0001 , odds ratio = 2 . 7 , Fishers exact test ) . Specifically , Shh transcriptional profiling and GLI3T chromatin interaction data both identified an enrichment of FOX gene family members , encoding Forkhead transcription factors , identifying FOX genes as potential SHF Hedgehog targets ( Figure S2 ) . The set of 112 Shh-dependent Gli3T bound genes included four Fox transcription factors , Foxb1 ( 64290 ) , Foxc1 ( 17300 ) , Foxd1 ( 15229 ) and Foxf1a , representing a significant enrichment ( Figure 2E , p = 0 . 0001 , odds ratio = 18 . 4 ) . We investigated the hypothesis that Foxf1a and Foxf2 expression was downstream of Hedgehog signaling in cardiac development . Shh-dependent expression of both genes in the SHF was confirmed by qPCR: Foxf1a expression was reduced by 50% ( p = 0 . 05 ) and Foxf2 was reduced by 80% in the SHF of Shh−/− versus wild-type controls ( p = 0 . 01 ) ( Figure 1D ) . In-situ hybridization to evaluate the patterning of expression showed that Foxf1a and Foxf2 were both expressed in the posterior SHF , but not in the heart , in wild-type embryos at E9 . 5 , with Foxf1a expression extending more ventrally than Foxf2 to include the DMP ( Figures 3A , A′ , E , E′ ) . Mesenchymal expression of both Foxf1a and Foxf2 demonstrated a severe decrement in shh−/− mutant embryos ( Figures 3B , B′ , F , F′ ) . In a search for common targets between Tbx5 and Hedgehog signaling in the SHF , we tested whether Foxf1a and/or Foxf2 SHF expression was Tbx5-dependent . We performed in situ hybridization for Foxf1a and Foxf2 in Tbx5+/− heterozygous mutant embryos ( MGI: 2387850 ) , which demonstrate 40% penetrance of AVSDs [15] . Foxf1a but not Foxf2 expression demonstrated significant reduction in Tbx5 heterozygotes at E9 . 5 . In Tbx5+/− embryos , Foxf1a expression was specifically decreased in the posterior SHF ( Figure 3C , C′ , D , D′ , arrow ) in the area of expression overlap with Tbx5 expression [6] . In regions where Foxf1a expression does not overlap with Tbx5 expression , such as the anterior SHF , Foxf1a expression appeared normal ( Figure 3D , D′ ) . Foxf2 expression in Tbx5+/− embryos appeared unaltered compared to wild-type embryos ( Figure 3G , G′ , H , H′ ) . Taken together , this analysis demonstrates that posterior SHF Foxf1a expression was Shh- and Tbx5-dependent whereas Foxf2 pSHF expression was Shh-dependent alone . We hypothesized that Foxf1a and Foxf2 were required in a dosage sensitive manner for atrioventricular septation . We analyzed the cardiac anatomy of embryos from an intercross between Foxf1a+/− and Foxf2+/− at E14 . 5 , when cardiac septation is normally complete . Foxf1a+/−; Foxf2+/− double-heterozygote embryos all exhibited atrioventricular septal defects ( Figure 4D , D′ asterisk; p = 0 . 03 ) . Primum-type atrial septal defects , characterized by absence of the dorsal mesenchymal protrusion , were observed in each case ( Figure 4D , D′ ) . Additionally , Foxf1a+/−; Foxf2+/− double-heterozygotes displayed larger than normal mesenchymal caps covering the primary atrial septum ( Figure 4D′ arrow ) , an observation in keeping with the known redundant requirement for Foxf1a and Foxf2 in limiting mesenchymal growth in other contexts [32] . Atrial septal defects were never observed in Foxf1a+/− ( Figure 4B , B′ ) or Foxf2+/− ( Figure 4C , C′ ) single-heterozygotes or wildtype control littermate embryos ( Figure 4A , A′ ) . We concluded that Foxf1a and Foxf2 are redundantly required for atrioventricular septation . We hypothesized that Foxf1a may represent a direct downstream target of Hedgehog signaling and/or Tbx5 in the SHF . We identified Foxf1a as a candidate direct target based on unbiased interrogation of GLI3T and TBX5 transcription factor chromatin interaction and transcriptional profiling data sets . We intersected our SHF GLI3T ChIP data set ( Figure 2B ) with a published TBX5 ChIP-seq data set generated from HL-1 cardiomyocytes [33] to define regions with potential co-occupancy of both transcription factors . The intersection of the ChIP-seq datasets identified a single overlapping interaction peak for Gli3T ( in the SHF ( Figure 2B ) ) and TBX5 ( in HL-1 cardiomyocytes ) [33] located approximately 90 kb upstream of the Foxf1a transcription start site ( Figure 5A and Figure S3 ) . The Foxf1a transcriptional start site is the closest protein-coding gene to the described peak . The transcriptional start site for a non-coding RNA is located approximately 1 . 3 kbp upstream of Foxf1a , oriented in the opposite direction [34] . Closer interrogation of the sequence underlying the interaction domains revealed a conserved canonical T-box binding site ( AGGTGTGG; chr 8 , nucleotides 123 , 517 , 714–721 , NCBI137/mm9 assembly ) and a conserved canonical Gli binding site ( GGACCACCCAGC; chr 8 , nucleotides 123 , 517 , 754–762 , NCBI137/mm9 assembly ) within 30 base pairs of one another ( Figure 5A ) . We evaluated the sequence information content for these sites from our SHF Gli3 ChIP-seq experiment and found close agreement with published binding sites for Gli3 [23] , [33] . This chromatin interaction data in combination with the Tbx5 and Hedgehog signaling-dependent Foxf1a SHF expression ( Figure 3 ) identified this conserved region ( mouse chromosome 8 , nucleotides 123 , 517 , 714–762 ) as a candidate Foxf1a cis-regulatory element . We evaluated the binding of TBX5 and the Hedgehog transcriptional regulators GLI1 and GLI3 to the candidate cis-regulatory element at Foxf1a in vivo in the SHF . We evaluated TBX5 binding in vivo by performing ChIP using an anti-TBX5 antibody on the micro-dissected wildtype SHF at E10 . 5 and observed 35-fold enrichment of the cis-regulatory element in the TBX5-immunoprecipitation fraction compared to the input fraction by qPCR ( Figure 5B ) . We evaluated GLI1 and GLI3T binding in vivo by performing ChIP on the micro-dissected SHF of mice carrying either a Cre-activated flag-tagged Gli3 ( RosaGli3TFlag c/c ) or Gli1 allele ( RosaGli1Flag c/c MGI: 4460761 ) in concert with the Nkx2 . 5-Cre ( MGI 2654594 ) , broadly expressed cardiac tissues and progenitors . We performed ChIP using an anti-flag antibody on the SHF from R26R-Gli3-flagNkx2 . 5-Cre/+ or R26R-Gli1-flagNkx2 . 5-Cre/+ embryos at E10 . 5 and observed , respectively , 6 . 8-fold and 7 . 1-fold enrichment of the Foxf1a cis-regulatory element in the GLI1- and GLI3T-overexpressing embryos over the input control by qPCR ( Figure 5B ) . We also evaluated two genomic loci between our identified binding site and the Foxf1a transcription start site to determine whether nonspecific pulldown occurred in our ChIP experiments . These loci were not significantly enriched in the IP'd DNA ( Figure S4 ) These results validate in vivo SHF binding of TBX5 , GLI1 , and GLI3 to the candidate cis-regulatory element at Foxf1a . We hypothesized that the conserved , adjacent , and functional in vivo Gli and Tbx5 binding sites may integrate Tbx5 and Hedgehog activity as a component of a downstream gene regulatory network . We evaluated the activity of TBX5 and GLI1 on the candidate Foxf1a enhancer in vitro . The conserved element was cloned into a pGL4 . 23 vector containing a minimal promoter driving luciferase as a transcriptional readout and was transfected into HEK293T cells along with expression vectors for Gli1 and/or Tbx5 . Co-transfection with the expression vector for Gli1 , a Hedgehog-responsive transcriptional activator , provided a 91 . 9-fold induction of luciferase activity ( p = 0 . 0017 ) . Co-transfection with the expression vector for Tbx5 alone provided a 3 . 9-fold increase of luciferase activity ( p = 0 . 039 ) . Co-transfection with both Gli1 and Tbx5 expression constructs provided a 171 . 6-fold increase in luciferase activity ( p = 0 . 00091 ) , demonstrating synergistic activity between these transcriptional co-activators ( Figure 5C ) . We assessed the requirement of TBX5 and GLI binding sites for transcriptional activation of the enhancer . To assess the requirement of TBX5-dependant transcriptional activation of the enhancer on TBX5 binding sites , a TBX5-mutant enhancer-luciferase construct with the 7 base pair core of 3 canonical TBX binding sites was generated by site-directed mutagenesis . This TBX5-mutant construct eliminated transcriptional activation by TBX5 alone ( p = 0 . 04 ) and limited transcriptional activation by TBX5 and GLI1 together ( p = 0 . 006 ) ( Figure 5C ) . A GLI-mutant enhancer-luciferase construct was also constructed with the 8 base pair core of 3 canonical binding sites altered by site-directed mutagenesis ( see materials and methods ) . This GLI-mutant construct profoundly diminished transcriptional activation by GLI1 alone ( p = 0 . 001 ) ( Figure 5C ) . Interestingly , transcriptional activation by TBX5 and GLI1 on the GLI-mutant enhancer construct was only modestly abrogated luciferase compared to the activity of GLI1 and TBX5 on the wild-type enhancer ( p = 0 . 003 ) . We hypothesized that the cis-regulatory element at Foxf1a may integrate Hedgehog signaling and Tbx5 activity as a SHF-specific enhancer in vivo . We cloned the Foxf1a genomic region into an Hsp68-LacZ expression construct , whose minimal promoter affords no intrinsic in vivo expression activity [35] . We evaluated the enhancer activity of the Foxf1a genomic fragment by evaluating LacZ expression in transient transgenic mouse embryos at E9 . 5 . The posterior SHF demonstrated strong lacZ expression and was the only anatomic region demonstrating consistent and robust expression in the 8 transgenic embryos genetically positive for LacZ ( Figure 5D ) . Interestingly , the SHF region with the most consistent and robust expression was the area of overlap between Hedgehog signaling and Tbx5 expression [6] , including the early dorsal mesenchymal protrusion and surrounding mesenchyme of the posterior SHF ( less frequent and intense expression was also observed in other anatomic locations that receive Hedgehog signaling outside of the Tbx5 expression domain , including the anterior SHF ( 5/8 ) , anterior lateral plate mesoderm ( 5/8 ) and somites ( 2/8 ) ( Figure 5D ) . These observations , in concert with the in vitro analysis suggested that Hedgehog and Tbx5 act synergistically to provide strong reproducible transcriptional activation of this enhancer in the posterior SHF .
Our observations identified a requirement for the Forkhead-box transcription factors Foxf1a and Foxf2 in heart development . Compound haploinsufficiency for both Foxf1a and Foxf2 caused an atrial septal defect of the primum type , an atrioventricular septal defect characterized by absence of the dorsal mesenchymal protrusion . Foxf1a and Foxf2 were expressed selectively in the SHF , not in the heart ( Figure 3 ) . The requirement for Foxf genes in atrioventricular septation ( Figure 4 ) provided further support for a model of atrioventricular septation as driven by molecular events in SHF cardiac progenitors as opposed to in the heart itself . We found that Foxf1a and Foxf2 are required downstream of Hedgehog signaling in atrioventricular septation , adding cardiac development to the previously described Hedgehog-dependent role for Foxf genes in murine gut development [32] , [36]–[37] . Atrioventricular septal defects are also observed in Shh-null mutant embryos [10] . Because Foxf1a and Foxf2 expression were each decreased in the SHF by more than 50% in shh−/− null embryos ( Figure 1D ) , Foxf1a+/−; Foxf2+/− double heterozygote embryos provided a reasonable developmental facsimile of their diminished expression levels in shh−/− embryos . The observation that Foxf1a and Foxf2 compound haploinsufficiency resulted in AVSDs is therefore consistent with the supposition that Foxf1a and Foxf2 are essential components of the Hedgehog-dependent SHF gene regulatory network . Foxf genes have also been previously implicated in cardiac specification in the ascidian Ciona intestinalis [38] , [39] . In ascidians , the single Foxf orthologue lies at the center of a pathway regulating numerous migration-related cellular processes , such as polarity , migration and membrane protrusion in trunk ventral cardiac progenitor cells [38] , [39] . Ciona trunk ventral cells with disrupted Foxf activity fail to migrate properly , but still differentiate into cardiac tissue at an improper location . Interestingly , removing Hedgehog signaling from the mouse SHF causes a migration failure of SHF progenitors [4] , [5] . Like the Ciona trunk cells without Foxf , SHF cells without Hedgehog responsiveness differentiate into cardiomyocytes , but their altered migration causes AVSDs [5] . Future efforts will determine whether the requirement for Foxf genes in cardiac progenitor migration is a conserved feature of mammalian cardiac development . Genetic interaction and rescue experiments investigating the requirement for Hedgehog signaling and Tbx5 in atrioventricular septation were consistent with Tbx5 acting either in parallel or upstream of Hedgehog signaling in atrioventricular septation [6] . Our interrogation of these pathways on a cis-regulatory element at Foxf1a provides molecular detail for their interaction . We observed that TBX5 and GLI1 , the Hedgehog-dependent transcriptional activator , synergistically activated the cis-regulatory element in vitro ( Figure 5C ) predicting strong activation of expression in areas of overlap between Tbx5 expression and Hedgehog signaling . This prediction held in vivo , where transcriptional activity of the enhancer was strong and reproducible only in the posterior SHF region , where Tbx5 expression and Hedgehog signaling overlap ( Figures 5D and 6 ) . This work is consistent and a model describing a SHF-specific gene regulatory network driven by GLI1 and TBX5 and essential for atrioventricular septation ( Figure 6 ) . This model provides specific predictions for the logic underlying enhancer choice in the SHF with ramifications for understanding the molecular and biochemical basis of atrioventricular septation and clinical AVSDs .
Mouse experiments were completed according to a protocol reviewed and approved by the Institutional Animal Care and Use Committee of the University of Chicago , in compliance with the USA Public Health Service Policy on Humane Care and Use of Laboratory Animals . The Shh− line was obtained from the Jackson laboratory . The Tbx5+/− mice have been previously reported [15] . Foxf1+/− and Foxf2+/− mouse lines were generated in the Kalinichenko lab ( Cincinnati Children's Hospital Medical Center ) by breeding Foxf1afl/fl and Foxf2fl/fl mice with EIIA-Cre transgenic mice ( Jackson Lab ) . Mef2c-AHF-Cre [24] , ROSA26-Gli1 [40] and ROSA26-Gli3T [23] were reported previously . For ChIP , transcriptional profiles and in-situ hybridizations , embryos were dissected in nuclease-free PBS on ice . For SHF microdissection procedures , head tissues anterior to the heart were removed , as were tail tissues posterior to the heart . Portions of these tissues were retained for genotyping if necessary . Neural tube tissues were also removed . The SHF mesenchyme was bisected into anterior and posterior portions when necessary , and then removed from the cardiac tissue ( Figure 1A ) . Shh+/+ and Shh−/− embryos were dissected as described above at E9 . 5 . SHF tissues from these embryos were pooled to isolate sufficient amount of RNA for synthesis of labeled cRNA . Transcriptional profiles were performed using Agilent Mouse Whole Genome Arrays mgug4122a . Microdissected SHF tissues were grouped into pools of approximately 50 . Tissues were briefly fixed in 1 . 8% formaldehyde , then washed and homogenized . Sonication was performed with a Misonix 4000 sonicator until the sheared chromatin was approximately 100–300 bp in length . Input control samples were reserved prior to overnight immunoprecipitation with the appropriate antibody bound to magnetic Dynabeads ( Invitrogen ) . Beads were precipitated and washed , the chromatin was eluted , de-crosslinked and purified using a PCR cleanup kit ( Qiagen ) . To determine fold enrichment , qPCR was performed using input controls compared with DNA bound to immunoprecipitated proteins , using primers specific to the site of interest as well as primers to two sites not expected to be enriched . ChIP-seq and microarray data were deposited in the Gene Expression Omnibus ( GEO ) database with a super accession number GSE44756 . In-situ hybridization was performed as in Moorman et al . [45] with slight modifications . Specifically , after post-hybridization washes with 50% formamide/2X SSC , specimens were incubated for 30 minutes at 37 degrees C in 20 ug/ml RNase A to remove unbound probe and reduce nonspecific staining . All in-situ hybridization experiments were performed on a minimum of three control and three experimental embryos . Expression vectors for Gli1 and Gli3T were obtained from the Vokes lab . Tbx5 was cloned into the pCDNA 3 . 1 expression construct [20] Foxf1a fragment was cloned into the pGL4 . 23 vector ( Promega ) . Expression and reporter vectors were transfected into HEK293T cells using FuGENE ( Promega ) . Cells were cultured for 48 hours after transfection , then lysed and assayed using the Dual-Luciferase Reporter Assay system ( Promega ) . The Foxf1a enhancer and minimal promoter used in the luciferase assays were subcloned from the pENTR vector into the Hsp68-LacZ vector [35] using the Gateway system ( Invitrogen ) . The resulting construct was digested with NotI enzyme to remove the pBlueScript backbone , gel-purified , injected into fertilized mouse eggs at the University of Chicago Transgenics Core Facility and implanted into female mice . Embryos were harvested at E9 . 5 and stained as described previously [5] . | Atrioventricular septal defects ( AVSDs ) are a common severe class of congenital heart defects . Recent work demonstrates that events in the second heart field ( SHF ) progenitors , rather than in the heart , drive atrioventricular ( AV ) septation . Our laboratory has shown that both Hedgehog signaling and the T-box transcription factor , Tbx5 , are required in the SHF for AV septation . To understand the molecular underpinnings of the AV septation process we investigated SHF Hedgehog-dependent gene regulatory networks . Transcriptional profiling and chromatin interaction assays identified the Forkhead box transcription factors Foxf1a and Foxf2 as SHF Hedgehog targets . Compound haploinsufficiency for Foxf1a and Foxf2 caused AVSDs in mice , demonstrating the biological relevance of this pathway . We identified a cis-regulatory element at Foxf1a that bound TBX5 and Hedgehog transcriptional regulators GLI1 and GLI3 in-vivo . Furthermore , TBX5 and Gli1 co-activate transcription of the identified cis-regulatory element in-vitro . The enhancer is expressed primarily in the pSHF in-vivo , where Tbx5 and Gli1 expression overlap . Our findings implicate Foxf1a and Foxf2 in AV septation and establish Tbx5 and Hedgehog signaling upstream of Foxf genes in a gene regulatory network for cardiac septation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"organism",
"development",
"heart",
"development",
"organogenesis",
"biology",
"and",
"life",
"sciences",
"morphogenesis"
] | 2014 | Foxf Genes Integrate Tbx5 and Hedgehog Pathways in the Second Heart Field for Cardiac Septation |
Dengue has become a major concern for international public health . Frequent epidemic outbreaks are believed to be driven by a complex interplay of immunological interactions between its four co-circulating serotypes and large fluctuations in mosquito densities . Viral lineage replacement events , caused for example by different levels of cross-protection or differences in viral fitness , have also been linked to a temporary change in dengue epidemiology . A major replacement event was recently described for South-East Asia where the Asian-1 genotype of dengue serotype 2 replaced the resident Asian/American type . Although this was proposed to be due to increased viral fitness in terms of enhanced human-to-mosquito transmission , no major change in dengue epidemiology could be observed . Here we investigate the invasion dynamics of a novel , advantageous dengue genotype within a model system and determine the factors influencing the success and rate of fixation as well as their epidemiological consequences . We find that while viral fitness overall correlates with invasion success and competitive exclusion of the resident genotype , the epidemiological landscape plays a more significant role for successful emergence . Novel genotypes can thus face high risks of stochastic extinction despite their fitness advantage if they get introduced during episodes of high dengue prevalence , especially with respect to that particular serotype . The rarity of markers for positive selection has often been explained by strong purifying selection whereby the constraints imposed by dengue's two-host cycle are expected to result in a high rate of deleterious mutations . Our results demonstrate that even highly beneficial mutants are under severe threat of extinction , which would suggest that apart from purifying selection , stochastic effects and genetic drift beyond seasonal bottlenecks are equally important in shaping dengue's viral ecology and evolution .
Dengue virus ( DENV ) is the most wide-spread arbovirus affecting human populations . During the last decades it has increasingly become a major public health problem with significant economic and social impact [1]–[3] . It is transmitted between humans in urban and peri-urban settings predominantly by the Aedes aegypti and Aedes albopictus mosquitoes vector [4] . Ae . aegypti is extremely well adapted to urban environments where it efficiently breeds in artificial water containers , such as flower pots , plastic bags or discarded car tires , near human habitations . Both vectors have undergone rapid expansion worldwide in the last couple of decades leading to DENV endemicity in more than 100 countries [5] . There are four closely related and potentially co-circulating serotypes of DENV ( DENV1-DENV4 ) [6] , [7] and recovery from infection is believed to provide life-long immunity to the infecting serotype but only a brief period of heterologous protection to all other serotypes [8] . Most primary infections are self-limited and clinically silent but can occasionally result in a short-lived febrile illness which is commonly known as dengue fever ( DF ) . In some cases this may progress to more severe and life-threatening illness such as dengue haemorrhagic fever ( DHF ) or dengue shock syndrome ( DSS ) [9] . While several risk factors for developing DHF/DSS have been described , including host genetic background , viral genotype , order of infecting serotype , time between infections or age of infection [1] , [9] , the most widely cited explanation is that of Antibody Dependent Enhancement ( ADE ) ( e . g . [10]–[13] ) whereby subneutralizing antibodies from primary infection can mediate viral entry into host cells leading to increased replication and disease manifestations [14]–[18] . The temporal epidemiological pattern of dengue is characterized by semi-periodic outbreaks whilst the inter-epidemic cycles in DF/DHF incidence highly correlate with the seasonal variations in vector population size ( see e . g . [19] ) . Furthermore , individual serotype prevalences show cyclical replacements in dominance ( Figure 1A ) which are believed to be induced by the immune profile of the human population [20] , [21] . Phylogenetic studies based on complete sequences of structural genes of all 4 serotypes have demonstrated the existence of multiple lineages in which different genotypes can be clustered [6] , [7] . Despite a general bias in the literature towards studies based on single-gene approaches , spatio-temporal patterns of genotype replacement in endemic regions have been widely recovered from data [6] , [7] , [22]–[24] . With the extrinsic pressures on DENV , such as seasonal or human-forced reductions in vector population size or abundance and mobility of susceptible hosts , it has been proposed that genetic drift plays a major role in the observed phylodynamics [22] , [25] . Furthermore , most studies have reported that DENV recent molecular evolution is marked by strong purifying selection , possibly due to the requirement of its two-host life cycle , and few reports have been able to show convincing evidence for positive selection either by the existence of non-synonymous mutations or in measures of fitness advantage in viral traits [6] , [7] , [23] , [24] , [26] . Following earlier reports of inter-serotypic difference in virulence ( see e . g . [27] ) one of the first convincing evidences for genetic determinants in disease outcome came from epidemiological studies suggesting that the DENV2 Asian genotype was associated with higher frequencies in DHF compared to the American genotype [28] . In vitro studies have since shown that the replication rate in both human monocyte-derived macrophages and dendritic cells as well as the vector's susceptibility were higher for the Asian genotype [29] , [30] . It was also found that the Asian genotype of DENV2 had a slightly higher replication rate within the mosquito and a shorter extrinsic incubation period [31] . These results provided a rational explanation for the replacement patterns observed in the Americas , where displacement of the American genotype by the Asian genotype has taken place in several countries in recent years [28] , [29] , [32] . A similar lineage replacement event has also occurred in SE Asia , with Asian-1 lineage viruses having displaced Asian/American viruses from Viet Nam ( Figure 1B ) , Cambodia and Thailand . This displacement was proposed to be due to difference in in vivo fitness , with higher viraemia levels observed in Asian-1 infected patients that could lead to an enhanced probability of human-to-mosquito transmission [33] . The study by Hang et al . [33] demonstrated some other intriguing aspects about the invasion dynamics of Asian-1 . A phylogenetic analysis suggested that the Asian genotype was introduced into the population years before it had been detected , and once it was detected it reached fixation within a relatively short period of time . The rate at which this genotype replaced the Asian/American type would suggest a significant fitness advantage not only over the resident genotype but possibly also over the other circulating serotypes; however , there was no discernible difference in the overall epidemiological dynamics in the period before or after fixation . Although these results suggested that a fitness advantage in a specific viral trait played a decisive role , the emergence of advantageous genotypes are as likely to be driven by the level of transmission and the underlying immune status of the human population . Here we have constructed an epidemiological model of dengue to qualitatively address the impact of immunity and transmission on the invasion and replacement patterns of a novel advantageous dengue genotype . Our results suggest that the observed replacement events can be explained by competition between genotypes of relatively small fitness differences which , although sufficient for displacement , do not interfere with the overall serotype dynamics . Furthermore , we show that invasion success and total time required for fixation are strongly influenced by inter- and intra-serotype competition at the time of introduction .
The model is an extension of the 4-serotype mathematical framework analysed by Recker et al . [34] and includes a mosquito vector component , temporary cross-immunity after primary infection and seasonal forcing in mosquito biting . In summary , we disregard the effect of maternal antibodies and instead assume that human individuals are born susceptible to all 4 serotypes . After recovery from primary infection they acquire life-long immunity to the infecting serotype and cross-immunity to any other serotype for a short period of time . As temporary immunity wanes , individuals become susceptible to secondary heterologous infection . For simplicity and because of the relative rarity of reported third and fourth infections we assume that after recovery from secondary infections individuals remain fully protected against further challenges [4] , [35] . The system can then be given by the following set of differential equations describing the rate of change in humans either susceptible , infected with , temporarily immune or recovered from dengue serotypes , DENV1 , DENV2 , DENV2′ , DENV3 or DENV4: ( 1 ) ( 2 ) ( 3 ) ( 4 ) ( 5 ) ( 6 ) with the force of infection of serotype affecting the human population , , given as ( 7 ) We denote as the mosquito biting rate and as the vector-to-human transmission probability; and are the respective durations of infection and cross-immunity . Given the short period of infection we do not account for the possibility of co-infections by two or more serotypes . We assume a constant human population size and further assume that infection has a negligible effect on the average death rate , . To account for seasonal variation we assume a periodically forced biting rate , that is we set ( 8 ) where is a positive integer influencing the ‘seasonality’ where results in shorter and more pronounced seasons . The dynamics of the mosquito population is given as follows: ( 9 ) ( 10 ) with the force of infection from humans to mosquitoes given as ( 11 ) In accordance with our previous model [34] we assume that antibody-dependent enhancement acts to increase both susceptibility to and transmissibility of secondary heterologous infection by factors and , respectively , with values described in Table 1 . To investigate the invasion patterns of a novel and fitter dengue genotype we assume that DENV2 is represented by two genotypes which differ in relative fitness but are antigenically equivalent . That is , individuals previously infected by DENV2 are immune to type DENV2′ and vice versa . We consider four different fitness traits which we can vary independently: ( i ) transmissibility from human to mosquito , e . g . through increased viral load , , ( ii ) longer life-expectancy of mosquitoes infected with DENV2′ to emulate a shorter extrinsic incubation period ( EIP ) , , ( iii ) longer infectious period in humans , , and ( iv ) an increased level of enhancement of secondary infections , . These can simply be given using: ( 12 ) ( 13 ) ( 14 ) ( 15 ) That is , , can be considered as the degree of the fitness advantage . In line with the suggestion by Hang et al . [33] , most of our analysis is concentrated on the fitness advantage due to increased viral load and thus transmissibility from the infected human individual to the mosquito vector , . In fact , we found that the results presented here are invariant to the actual viral trait that is enhanced; results obtained under changes to other viral traits can be found in the supporting material . To address certain aspects of the invasion process of a more probabilistic nature , such as invasion success rates and fixation events , we also implemented the above model as a stochastic framework using a tau-leap Gillespie algorithm [36] . Stochastic simulations were initialized with equilibrium population status derived from the deterministic framework with parameter values the same as given in Table 1 ( see Figure S7 and S8 for general model behaviour ) .
We examined the dynamics of a novel genotype introduced into a dengue endemic population by either an infected human individual or via an infected mosquito . The novel genotype is here denoted as DENV2′ , to represent the Asian-1 genotype of serotype 2 , whereas the resident type is denoted as DENV2 to represent the Asian/American type . Figure 3 shows the result of an invasion scenario where the invading genotype has a small fitness advantage over the resident type ( , corresponding to a fitness advantage of ) . In this case , higher viral fitness was realised through enhanced transmissibility from infected human individuals to the mosquito vectors , i . e . . In agreement with the data , two important features of the invasion dynamics can be observed and are highlighted in Figure 3B . Despite the eventual fast rate at which the advantageous genotype replaces the resident type , there is a significant lag between the point of introduction and the time when DENV2′ genotype would reach a detectable level of prevalence within the population; we refer to this level of prevalence as detection threshold . Furthermore , despite the expected temporary rise in dengue incidence , compared to the situation without invasion , the overall dynamics in both disease incidence and serotype prevalence remain largely invariant ( Figure 3A ) . This suggests that both the time lag between introduction and first detection and also the rapid exclusion of the resident genotype , such as reported by Hang et al . [33] , can be explained by a relatively small fitness advantage of the invading genotype . The same qualitative behaviour can be also found when changing other viral traits which could determine the fitness advantage . That is , shortening the extrinsic incubation period , , increasing the duration of infection , , or the level of enhancement of secondary infection , , have the same effect as increasing the transmission probability from infected humans to mosquitoes , . Notably , though , when considering low advantages , smaller differences in terms of viral fitness are required to achieve the same rate of fixation if the fitness advantage manifests itself in longer infectious periods compared to an increase in transmissibility ( Figure S3 ) . Interestingly , while similar levels of fitness advantages in either EIP or transmissibility result in the same fixation times ( Figure S4 ) , the disturbance on the epidemiological pattern of dengue is less severe when the fitness advantage is expressed in the mosquito ( Figure S5 ) . From now on , we concentrate only on a fitness advantage through the proposed increase in human-to-vector transmission . As shown in Figure 3 , a small increase in transmissibility from human to mosquito seems sufficient for a novel genotype to displace a resident type within a short period of time . The actual rate of competitive exclusion and overall time from introduction of the advantageous genotype to its fixation in the population is likely to depend on various factors including fitness advantage , rate of transmission and immune profile within the human population . As shown in Figure 4A , increasing viral fitness accelerates the rate at which the invading genotype drives the resident type , DENV2 , to extinction , resulting in a shorter period between introduction and fixation . For example , increasing the fitness advantage from to reduces the time to fixation from years down to years . However , this increase in viral fitness has a major effect on dengue incidence patterns and the dynamics of the other serotypes . In this case it leads to a significantly bigger epidemic outbreak at the time of replacement followed by a long period of low transmission and low prevalence of serotype 2 which could endanger its continuous persistence; this is highlighted in Figure 4B ( compare to Figure 3A ) . We next addressed the effect of the time of introduction on the invasion dynamics . This was simply motivated by the fact that serotype competition is not constant over time but is strongly affected by the level of transmission which itself is dependent on host immunity level and seasonal variation in mosquito densities . Not surprisingly , we found that the time of introduction can significantly alter the time taken for a novel genotype to reach fixation . Figure 5A shows the decrease in the frequency of DENV2 , relative to the fitter genotype DENV2′ , for two different time points of introduction . However , while the overall duration from invasion to fixation is dependent on the time when DENV2′ gets introduced , the actual rate of replacement remains constant . In other words , the time taken from DENV2′ passing a detection threshold , relative to DENV2 , to reaching fixation is independent of the time of introduction ( Figure 5B ) and therefore independent of the overall epidemiological dynamics . This , on the other hand , suggests that the time lag between introduction and the point when it has spread sufficiently for detection , or waiting time , is strongly influenced by the epidemiological profile at that time . To investigate further the determinants for fixation time we simulated a number of invasion events at various time points over a four year period and recorded the total time to fixation for each event with respect to ( i ) the number of naive individuals , ( ii ) serotype 2 susceptible individuals , ( iii ) disease prevalence and ( iv ) mosquito biting frequency . While we could not find a clear correlation between any of these population profiles and fixation time , we observed a trend for longer fixation times during the time window where the relative prevalence of serotype 2 was increasing ( Figure S6 ) . The results from our deterministic model suggest that novel genotypes can face long periods at very low prevalence before breaching a detection threshold and going to fixation . Within a more realistic setting these periods signify an enhanced risk of stochastic extinction of the novel type despite its fitness advantage over the resident type . To better address the invasion success of DENV2′ we used a stochastic formulation of our model ( see Methods ) and simulated a number of invasion events over a period of four years and recorded the success rate of invasion , here defined as the successful introduction into a population followed by competitive exclusion of the resident type . As demonstrated in Figure 6A we observed that invasion success shows an oscillatory behaviour whose phase seems negatively correlated to total dengue prevalence at time of introduction . This suggests that the invasion of a newly advantageous genotype can be hampered by serotype competition during epidemics and favoured during off-season periods . Moreover , the amplitude of oscillation , i . e . the maximum success rate , is dependent on and again negatively correlated to serotype 2 prevalence . Figure 6B shows the increase in relative prevalence of DENV2 over the 4-year period which clearly correlates with a decline in the success rate of DENV2′ . Since the time taken from passing a detection threshold to reaching fixation was shown to be independent of the time of introduction ( Figure 5B ) , we focused on the relationship between serotype 2 prevalence and the time to emergence , i . e . the period between introduction and reaching a prevalence threshold . Figure 7 clearly illustrates that a novel and advantageous genotype entering the population during periods of high DENV2 prevalence will face significantly longer emergence times than those introduced during periods of low prevalence . Together our results indicate that the fate of a novel genotype is strongly determined by both inter- and intra-serotype competition at the time of introduction .
We analysed the invasion pattern of a novel dengue genotype into an endemic population with 4 co-circulating serotypes . Within our framework we assumed that the invading genotype , representing the Asian-1 genotype of dengue virus serotype 2 , possesses a fitness advantage over the resident type , the Asian/American genotype , through enhanced transmissibility from infected human individuals to the mosquito vectors . This assumption was based on the findings by Hang et al . [33] which showed increased plasma viraemia levels in patients infected by Asian-1 DENV2 viruses . In contrast to other studies [30] , [41] , Hang and colleagues did not find increased infectivity of Asian-1 viruses to Ae . aegypti mosquitoes per se; however , it is easy to envisage how higher viral titers could enhance the ‘per bite’ probability of human-to-vector transmission . By thus focusing on the hypothesis of a small increase in transmissibility during primary and secondary infections , and in agreement with the data , we observed that the total time for genotype replacement is composed of a period during which the invading type can circulate at very low prevalence levels for several years , followed by a rapid shift in dominance and competitive exclusion after the invading genotype had emerged; here we defined ‘emergence’ as a threshold level of prevalence where widespread detection would be highly likely . Of particular interest is the time lag between introduction and emergence , or waiting time , when the detection of the new dengue genotype might be difficult by surveillance systems based on low viral sampling numbers and/or infrequent genotyping . Not surprisingly , we found that this period is strongly and positively affected by the difference in viral fitness between the resident and novel genotype . In the case of small fitness advantages several years could pass before the invading type has spread sufficiently to outcompete the resident type on a population-wide level . Furthermore , as the epidemiological pattern would remain largely invariant , passive surveillance systems based simply on case numbers could also easily fail to detect this intra-serotype replacement event . These results therefore support the findings of Hang et al . [33] who hypothesised that a small enhancement of human-to-mosquito transmission through increased viral load is sufficient to explain the observed invasion pattern in Southern Viet Nam where Asian-1 was first detected in 2003 despite the phylogenetic analyses dating the introductory event sometime during the late 1990's . Apart from increased transmission from infected humans to the mosquito vectors we also considered other viral traits that could be enhanced in the Asian-1 genotype , such as longer infectious periods or shorter extrinsic incubation periods ( EIP ) . The latter is of particular interest as it can potentially lead to a significantly increase in vectorial capacity [31] . While the actual viral trait which is enhanced does not alter the overall invasion pattern or results presented in this work ( Figures S3 , S4 , S5 , S9 , S10 , and S11 ) , we found that viral fitness traits have an additive effect ( Figure S4 ) . This means that even smaller individual enhancements are sufficient to explain the observed invasion dynamics of the Asian-1 genotype , especially under the assumption that this replacement event did not have a major effect on the sero-epidemiological pattern of dengue . Interestingly , though , our results suggest that dengue incidence and serotype dynamics are less disturbed when the fitness advantage is manifested through shorter EIP than increased infectivity or transmissibility ( Figure S5 ) . In addition to viral fitness , the time point at which a novel genotype enters a population is crucially important in determining its invasion dynamics and ultimately success . Whereas the relative fitness advantage affects the overall time between introduction and fixation , the epidemiological profile more strongly determines the period of low level prevalence before the advantageous genotype emerges . We tested various epidemiological factors for their influence on the waiting time but to our surprise only found the relative prevalence of DENV2 to have a strong effect . That is , whereas population susceptibility to either dengue in general or serotype 2 in particular had no immediate influence on the time between introduction and wide-spread detection , we found that the relative prevalence of DENV2 at the time of introduction positively correlates with extended periods during which the novel genotype circulates below a detection threshold . Therefore , while transmission intensities strongly affect the success of an invasion event , the dominance level of serotype 2 within the population determines both the invasion success rate and , independently , the period before the invading genotype would reach a sufficient level of prevalence to be widely detecable . Our results thus confirm that serotype interactions and the resulting epidemiological landscape can have a big influence on intra-serotype dynamics and thus viral evolution , as previously noted by Zhang and colleagues [23] . There is considerable interest in determining the evolutionary processes that underlie the observed structures and genetic variation of dengue virus populations ( both inter- and intra-serotypic ) . Overall , low estimates of selection pressure , in terms of average values , and the fact that dengue has a two-host life-cycle are commonly used to place purifying selection as the strongest selective force acting on dengue evolution [23] , [26] , [42] . However , it is also clear that dengue viruses exhibit strong spatio-temporal variations . Various phylogenetic studies have identified frequent DENV lineage turnover events which have resulted in the characteristic , ladder-like tree ( e . g . [24] , [42] ) and which are commonly ascribed to positive selection [24] , [32] , [43] . In addition , genetic drift has also been proposed to play a major part in dengue evolution such that the replacement of viral lineages or clades could be explained through stochastic processes alone . For example , repeated bottlenecks due to large seasonal fluctuations in mosquito densities imply that the emergence of novel and possibly advantageous genotypes could be a recurrent phenomenon followed by a strong probability for extinction in the subsequent circulating seasons which could explain the weak signature for positive selection in the data ( compared to purifying selection ) . This in turn would also suggest that the success of a genotype does not always reflect its viral fitness [7] . In fact , we have shown that novel genotypes , especially those that arise during large epidemic outbreaks , can face high risks of extinction despite possessing a fitness advantage . Furthermore , even successful genotypes , i . e . those that eventually reach fixation , potentially undergo prolonged periods of low frequency which can span for several transmission seasons independently of the epidemics therein . Therefore , low measures of adaptive selection in this case would not necessarily imply strong purifying selection but could equally be explained by other epidemiological factors . This , however , needs to be confirmed within a more rigorous framework . Dengue's two-host life-cycle implies a significant evolutionary constraint whereby the majority of newly arising variants are likely to be deleterious and selectively removed from the population . We have shown that even novel and advantageous DENV genotypes can undergo periods of several years prior reaching sufficiently large population sizes to escape the risk of extinction . Our results thus indicate that in addition to purifying selection , the epidemiological landscape and stochastic effects might be equally important determinants in shaping the viral evolutionary ecology . | Dengue fever and the more severe dengue haemorrhagic fever and dengue shock syndrome are mosquito borne viral infections that have seen a major increase in terms of global distribution and total case numbers over the last few decades . There are currently four antigenically distinct and potentially co-circulating dengue serotypes and each serotype shows substantial genetic diversity , organised into phylogenetically distinct genotypes or lineages . While there is some evidence for positive selection , the evolutionary dynamics of dengue virus ( DENV ) is supposed to be mostly dominated by purifying selection due to the constraints imposed by its two-host life-cycle . Motivated by a recent genotype replacement event whereby the resident American/Asian lineage of dengue virus serotype 2 ( DENV2 ) had been displaced by the fitter Asian-1 lineage we investigated some of the epidemiological factors that might determine the success and invasion dynamics of a novel , advantageous dengue genotype . Our results show that although small differences in viral fitness can explain the rapid expansion and fixation of novel genotypes , their fate is ultimately determined by the epidemiological landscape in which they arise . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases",
"infectious",
"diseases"
] | 2010 | Viral and Epidemiological Determinants of the Invasion Dynamics of Novel Dengue Genotypes |
Rift Valley fever ( RVF ) is a major zoonotic and arboviral hemorrhagic fever . The conditions leading to RVF epidemics are still unclear , and the relative role of climatic and anthropogenic factors may vary between ecosystems . Here , we estimate the most likely scenario that led to RVF emergence on the island of Mayotte , following the 2006–2007 African epidemic . We developed the first mathematical model for RVF that accounts for climate , animal imports and livestock susceptibility , which is fitted to a 12-years dataset . RVF emergence was found to be triggered by the import of infectious animals , whilst transmissibility was approximated as a linear or exponential function of vegetation density . Model forecasts indicated a very low probability of virus endemicity in 2017 , and therefore of re-emergence in a closed system ( i . e . without import of infected animals ) . However , the very high proportion of naive animals reached in 2016 implies that the island remains vulnerable to the import of infectious animals . We recommend reinforcing surveillance in livestock , should RVF be reported is neighbouring territories . Our model should be tested elsewhere , with ecosystem-specific data .
Rift Valley fever ( RVF ) is a major vector-borne , zoonotic , and hemorrhagic fever ( Phlebovirus , Family Bunyaviridae ) that severely affects human health , animal health and livestock production mainly in Sub-Saharan Africa [1–3] . Its potential for spread and emergence in current disease-free areas ( e . g . Europe , United States of America ) is of growing global concern . The emergence ( or re-emergence ) of a disease has been defined by Woolhouse and Dye ( 2001 ) [4] as an “increase [in its incidence] following first introduction into a new host population , or [an increase in its incidence] in an existing host population” following specific ecological changes , e . g . in anthropological factors , agricultural practices or climate [4–6] . Theoretically , the conditions leading to RVF emergence may result from a sudden increase in vector density , the availability of susceptible animals , and the presence of the virus . The virus could be newly introduced or already present locally; maintained in the mosquito population , or circulating at a low level in livestock or wild animal populations , although little evidence exists on the latter [7–9] . Due to little existing data on RVF , those mechanisms have not been fully quantified . Previous work on RVF emergence conducted for the Horn of Africa using ecological statistical modelling showed that above-average rainfall and vegetation density ( Normalized Vegetation Difference Index , NDVI ) over 3 to 4 months could lead to RVF re-emergence [10] , but it did not seem to be always the case , especially in Madagascar and Southern Africa , where other factors , such as the movements of animals or the level of livestock susceptibility may also play a role [11 , 12] . Finally , although a range of mathematical models have been developed to study RVF , they looked mainly at RVF spread , that is once the virus is introduced [13] or they assessed the impact of vaccination strategies [14] , and only few were fitted to data [15–17] . A large RVF epidemic started in Kenya in December 2006 , affecting humans and animals , and subsequently spread to East and Southern Africa over the following months , including Somalia , Tanzania , Sudan , Mozambique , the Union of Comoros [10 , 18–23] . In September 2007 , RVF virus was detected in humans on the island of Mayotte ( in the Mozambique Channel , Fig 1 ) , about 500km away from the African continent . Two Mayotte 2008 RVF virus isolates were sequenced and the results showed that they were related to the Kenyan 2006–2007 clade [20 , 24] . Subsequent serological studies carried out in Mayotte in livestock ( 2004–2016 ) showed that RVF had been present at least since 2004 , and re-emerged in 2008–10 , despite no symptoms in animals being detected [25 , 26] . Because of its proximity to the Kenyan 2006–2007 clade , the Mayotte 2008 isolates may have been introduced onto the island by animal trade from the African mainland through the Union of Comoros [6 , 27]; and very likely resulted in the re-emergence observed in livestock in 2008–10 [25] . Quantifying the main factors driving pathogen emergence can be approached using mathematical models , but accounting for the diversity of the processes underlying emergence remains a challenge due to the lack of existing data [28] . Because of its insular nature , its epidemiological connections to the African mainland and its 12-year ( 2004–2016 ) RVF serological dataset [25] , Mayotte offers an ideal setting to attempt to disentangle the impact of environmental and anthropogenic factors driving RVF dynamics in the livestock population . Here , we developed the first mathematical model for RVF emergence that accounts for livestock susceptibility , climate factors , and the import of infectious animals , while fitting serological data in a Bayesian framework . It allowed an estimate of ( i ) the most likely emergence scenario that could explain the past observed epidemic , and ( ii ) the likelihood of a future re-emergence , under different animal import scenarios .
Mayotte is a small island ( 374 km2 ) , with an estimated livestock population size of about 30 , 000 heads [29] ( 17 , 000 cattle , 12 , 000 goats and 1 , 000 sheep ) . The production system is agro-pastoral; animals are raised for family consumption or ceremonies . No official export or import of animals exists . The movements of people on small boats ( named "kwassa-kwassa" ) between Mayotte ( French overseas département ) and the nearest island of Anjouan ( Union of Comoros ) , 70 km apart , became illegal since the installation of French visa requirements to enter Mayotte in 1995 . Nevertheless , people attempt the journey with livestock on board [30] . Although the maritime border authorities attempt to control those entrants , they seize only a fraction of them ( estimated 16 , 000 people entering per year in 2007 [31] ) , allowing the introduction of potentially infected animals . The climate of Mayotte is marine tropical . The annual temperature varies between 25°C and 35°C , with a high annual rainfall ( 1500 mm ) and a peak rainy season ( December-March ) [32] . Despite rainfall seasonality , the normalized difference vegetation index NDVI ( a measure of vegetation density ) is high and shows little annual variation ( average Mayotte range 0 . 65–0 . 82 , NDVI ranges 0–1 ) [33] ( S1 Fig ) . A continually high NDVI potentially would allow the sustenance of mosquito breeding and therefore vector transmission even during the dry season [34 , 35] . Apart from RVF , other arboviral diseases reported are Dengue and Chikungunya [6] . The livestock population ( cattle , sheep and goats ) was modelled with an animal as a unit but without differentiating animals according to their species ( one livestock population considered ) , or their spatial location . Animals could pass through four successive and mutually-exclusive infection states of RVF infection: Susceptible ( S ) , Latent ( E ) , Infectious ( I ) and Immune ( R ) . Once infected , we assumed that animals remained in the ( R ) compartment , as natural infection is assumed to provide life-long immunity [14] . We accounted for deaths and births and assumed a constant population size ( N ) for the study period ( October 2004-June 2016 ) . The model was deterministic and discrete-time ( weekly time step ) . While we assumed homogeneous mixing , the livestock population was purposively stratified in 10 yearly age groups a ( a ∊[1–10] ) to allow fitting the model to age-specific IgG prevalence data ( see model fitting paragraph and S1 Text for details ) . The model is presented in S2 Fig , and in Eqs ( 1 ) to ( 7 ) . Indexing the state variables and parameters by yearly age-group a ( see S1 Text , methods section for the definition of age-groups ) , and time , t , we have the following equations: For ≤ 12 months-old animals ( i . e . age group a = 1 ) : S1 , t+1= ( 1−λt ) ( 1−δ ) αS1 , t+bt ( 1a ) E1 , t+1=λt ( 1−δ ) αS1 , t ( 1b ) I1 , t+1= ( 1−δ ) αE1 , t+Iimp_1 , t ( 1c ) R1 , t+1= ( 1−δ ) αR1 , t+ ( 1−δ ) αI1 , t ( 1d ) For > 12 months-old to ≤ 108 months-old animals ( i . e . age groups a ∊[2–9] ) : Sa , t+1= ( 1−λt ) ( 1−δ ) αSa , t+ ( 1−λt ) δαSa−1 , t ( 2a ) Ea , t+1=λt ( 1−δ ) αSa , t+λtδαSa−1 , t ( 2b ) Ia , t+1= ( 1−δ ) αEa , t+δαEa−1 , t+Iimp_a , t ( 2c ) Ra , t+1= ( 1−δ ) αRa , t+δαRa−1 , t+ ( 1−δ ) αIa , t+δαIa−1 , t ( 2d ) For > 108 months-old animals ( i . e . age group a = 10 ) : S10 , t+1= ( 1−λt ) α10S10 , t+ ( 1−λt ) δαS9 , t ( 3a ) E10 , t+1=λtα10S10 , t+λtδαS9 , t ( 3b ) I10 , t+1=α10E10 , t+δαE9 , t+Iimp_10 , t ( 3c ) R10 , t+1=α10R10 , t+δαR9 , t+α10I10 , t+δαI9 , t ( 3d ) With: λt=1−exp ( −βt∑a=110Ia , t ) ( 4a ) bt= ( 1−α ) ∑a=110 ( Sa , t+Ea , t+Ia , t+Ra , t ) −∑a=110Iimp_a , t ( 4b ) Where Sa , Ea , Ia , Ra , Iimp_a , Na are the age-specific number of Susceptible , Latent , Infectious , Immune , imported Infectious , and total number of animals in the ath yearly age group , with their sum over all ages denoted by St , Et , It , Rt , Iimpt and Nt , respectively . λt is the force of infection , and βt is the per-capita effective contact rate . The rate at which latent ( E ) become infectious ( I ) , and infectious ( I ) become immune ( R ) were fixed and equal to one , so that animals stay one time step ( i . e . one week ) in the ( E ) and ( I ) compartments . In the absence of vector data , the time spent in ( E ) is assumed to account for the extrinsic incubation period in the vector ( 3 days ) and the latent ( 1–6 days ) stage in the animal without explicitly modelling these processes , and the time spent in ( I ) accounted for the infectious stage in the host ( 3–6 days ) [34 , 36 , 37] . This was chosen because we were interested in fitting the model to the Immune ( R ) compartment only , whilst allowing the serial interval ( defined as the average time of infection between two consecutive cases , as per Wallinga and Lipstich [38] definition ) , at the animal level , being 2 weeks; which aligns with the 3 weeks estimated in South Africa at the farm level [39 , 40] . The rate at which animals are ageing at each time-step is noted δ; α is the survival rate for the age-groups 1–9 , and α10 for the age-group 10; and finally bt is the birth rate . No disease-related mortality was accounted for because such symptoms were not reported at that time in Mayotte , neither in the neighbouring Comoros and Mozambique RVF affected areas [23 , 41–43] . In addition , since the animal population was not fully susceptible at the beginning of our study period ( October 2004 ) , a proportion of immune animals ( imm_t0 ) was specified at t0 , such as: Rt=0=N×imm_t0 ( 5a ) St=0=N− ( Rt=0+Et=0+It=0 ) ( 5b ) We used NDVI as a proxy for climate conditions favouring mosquito habitat commonly used in RVF studies [10 , 11 , 44–46] , as no data on vector dynamics was available . NDVI data for Mayotte did not show any three-months NDVI anomaly over the study period as measured for the Horn of Africa [10] therefore it was not used in this model ( S1 Fig ) . Instead , we estimated the transmission parameter βt that varied over time as a function of the observed NDVI value at time t ( NDVIt ) [33] . In the absence of a known quantified relationship between βt and NDVI for RVF , or other vector-borne diseases , two models were tested . Model 1a assumed a linear relationship between βt and NDVIt , and Model 1b an exponential relationship , such as: βt=RstN ( 6a ) Model 1a: Rst=a ( NDVIt−NDVImin ) +b ( 6b ) and Model 1b: Rst=exp ( aNDVIt+b ) ( 6c ) Where Rst is the seasonal reproduction number , a and b are the coefficients of the linear and exponential functions . The linear function is defined such that Rs , t reaches its minimum value Rsmin = b when NDVIt is at its minimum ( NDVImin ) . Following the RVF outbreak in the Horn of Africa in 2006–2007 , it was assumed that infectious animals entered Mayotte in kwassa-kwassa , from a starting date timp and for a duration P . Those imported animals Iimp were added directly into the infectious compartments Ia ( Eqs ( 1c ) , ( 2c ) and ( 3c ) ) , and at a constant flow at each time-step t , for the length of the period P , such as: For timp< t< timp + 48P , Iimp= ( nseizedpikw ) / ( 48pseized ) ( 7 ) Where nseized is the number of animals seized by the maritime patrol per year , pikw is the proportion of these that tested positive to RVF recent infection ( S2 Table IgM positive ) , pseized is the proportion of kwassa-kwassas seized , and finally P the duration of importation , expressed in year fraction . In addition , to facilitate the aggregation of monthly estimates , a month was modelled as 4 weeks , and therefore a year was 48 weeks . A time step ( week ) corresponded to 1 . 08 calendar week . The parameters of the model were related to the natural history of disease and demographics , the climate-dependent transmission scenarios , or the viral introduction through animal import ( Table 1 ) . However , whilst some parameters were fixed input values , others were estimated by fitting the model predictions to the serological IgG prevalence data . Table 1 presents which parameters were fixed input values and which were estimated by fitting the model . Specifically , the fixed input parameters were the demographics parameters Na , α and α10 derived from demographic data [29 , 47 , 48] ( S1 Text , S1 Table and S3 Fig ) , as well as the number of animals seized per year by the maritime patrol ( nseized = 100 ) , and the proportion of them being infectious ( pkiw = 15% ) ( S2 Table ) . The proportion of imported animals that had been recently infected by RVF virus ( IgM positive , S2 Table ) was used as a proxy for pkiw . The other six parameters were estimated by fitting the model to the serological data ( Table 1 ) using a Bayesian framework . Each prior distribution of these parameters was a uniform function with the following lower and upper bounds: ( i ) the proportion of immune animals at t0 , imm_t0 , could be estimated between 5 and 20% , ( ii ) a and ( iii ) b , the parameters defining the functional relationship between βt ( and therefore Rst ) and NDVI were set to allow estimating Rs between 0 . 5 and 6 , ( iv ) the proportion of kwassa-kwassas seized pseized could take any value between 2 . 5% and 20%; and ( v ) the duration of import P varied from 1 month to 2 years . Finally , the date of the first import of infectious animals , ( vi ) timp , could be any time between January 2007 ( First report of RVF outside its initial confinement , i . e . in Kenya ) and September 2007 ( RVF detected in Mayotte ) , and was estimated from the data . Parameter estimation was done by fitting the age-specific simulated proportion of immune ( R ) animals , for each epidemiological year i , to RVF IgG prevalence ( Oct 2004-Jun 2016 ) , as presented in Metras et al . 2016 [25] . Note that for Oct 2004-Jun 2008 , age data was not available so we fitted to monthly prevalence ( blue dots on Fig 2 , S4 and S5 Figs ) . We sampled from the posterior distribution of all six parameters θ = {imm_t0 , a , b , timp , P , pseized} using a Monte Carlo Markov Chain Metropolis-Hastings algorithm [49] , assuming uniform priors ( Table 1 ) . Parameters were estimated for both exponential and linear models ( Models 1a and 1b ) , and the best model had the lowest deviance information criterion value ( DIC ) [50] . For details on parameter estimation , model fitting , and comparison see S1 Text . To estimate the probability of re-emergence , we simulated 5000 stochastic trajectories of the exponential model , sampling randomly from the posterior distribution . The simulations were done for Oct 2004-Jun 2020 , adding 48 months ( Jul 2016-Jun 2020 ) to the original model , additional time for which the long-term NDVI monthly average values were used ( however keeping a seasonal pattern ) . In “Forecast 1” infectious animals were only imported in 2007–09 . To estimate the impact of future infectious imports and their seasonal timing on RVF re-emergence , we simulated the import of 1 , 10 , 20 , 30 and 40 infectious animals in October 2016 ( low NDVI values , Forecasts 2–6 ) , and in April 2017 ( high NDVI values , Forecasts 7–11 ) . To test for the effects of animal imports on the probability of emergence , we fitted both linear ( Model 2a ) and exponential ( Model 2b ) models , without animal imports , therefore sampling from the posterior distribution of three parameters only θ = {imm_t0 , a , b} . Finally , to assess the impact of NDVI seasonality on transmission , we fitted Model 3 , a model with animal imports , but with a constant transmission parameter β , such as: β=R0N ( 8 ) In other words , we sampled from the posterior distribution of five parameters θ = {imm_t0 , β , timp , P , pic} . All models were compared using the DIC [50] . The data were collected under the under a national disease surveillance system ( Système d'Epidémiosurveillance Animale à Mayotte—SESAM ) with the approval of the Direction of Agriculture , Food and Forestry ( DAAF ) of Mayotte . Before 2015 , consent for blood sampling on a herd was obtained from its owner verbally after information in French ( official language ) or Shimaore ( local language ) was given . The animals were bled without suffering . No endangered or protected species were involved in the survey . From 2015 , all procedures were approved by the London School of Hygiene Animal Welfare and Ethical Review Board .
Both linear and exponential models with seasonal variations of the NDVI including animal imports ( Models 1a and 1b ) fitted equally well the data ( Table 2 , DICmodel1a = 522 . 8 and DICmodel1b = 523 . 7 ) and showed a good agreement with the IgG serological data ( Fig 2 and Fig 3A–3H ) . When transmission was not dependent on NDVI seasonality ( Model 3 ) , the IgG prevalence peak was also captured ( S4 Fig , black solid line ) , but the model did not fit the data as well as the NDVI seasonality-dependent models , the DIC being higher ( Table 2 , DICmodel3 = 543 . 6 ) . Models without any animal import ( Models 2a and 2b ) had the worst fit ( Table 2 ) , and failed to capture the IgG prevalence peak ( S4 Fig , blue and green solid lines ) , suggesting that the re-emergence of RVF virus in 2008–10 may have been due to the import of infectious animals . Since both models 1a and 1b with animal imports exhibited a similar fit , we present in the main manuscript the fitting and forecasts using the exponential model ( results obtained with the linear model are similar and provided in S5 Fig and S6A–S6H Fig ) . Fig 2 and Fig 3A–3H show the median and the 95% CrI of the 5000 stochastic trajectories of the proportion of IgG positive animals . The main discrepancies among model trajectories were observed for the first part of the study period ( Oct 2004-Jun 2008 , before the peak , Fig 2 ) , when the model was fitted to serological estimates supported by a smaller sample size . In contrast , in July 2008-June 2016 the model was fitted to age-specific IgG prevalence ( Fig 3A–3H ) , and simulations showed little variation . In the best models ( Models 1a and 1b ) , the import of infectious animals was estimated to have started in timp = June 2007 ( IQR [May 2007-Jul 2007] , or month number 33 ) , when 94 . 1% ( IQR [85 . 6–96 . 5] ) of the livestock population was estimated to be susceptible . The import scenario also estimated that 43 ( IQR [39–46] ) infectious animals were imported per month , during 23 months ( IQR [22–24] ) , which corresponded to 2 . 9% ( IQR [2 . 7–3 . 1] ) of the animals illegally imported caught by the maritime border ( Table 3 ) . Rst values ranged between 0 . 36 and 1 . 90 for the linear model , and 0 . 52–2 . 19 for the exponential model ( Fig 4A and S7 Fig ) , reflecting the seasonal variation of NDVI ( NDVImin = 0 . 59 and NDVImax = 0 . 85 ) . Finally , under those conditions , the proportion of immune animals at t0 ( imm_t0 ) , that is in October 2004 , was 12 . 9% ( IQR [11 . 7–14 . 1] ) . The seasonal variation of Rst over time reflected the seasonal NDVI values , and was compared to Rst values under the average-NDVI conditions ( Fig 4A ) . The effective reproduction number Re ( range 0 . 42–1 . 84 ) , and the monthly incidence ( monthly number of infectious cases ) are presented in Fig 4B . The incidence started to rise slightly in April-May 2007 , that is before the import of infectious animals , and despite the imports starting in June 2007 , the incidence remained stable and slightly decreased due to substantially below-average NDVI values . The highest incidence peak was reached the following year , in June 2008 ( 1558 cases , 95%CrI [707–2684] ) , and very likely resulted from the combination of infectious imports with above-average NDVI seasonal values; similarly to what is observed for the 2009 peak ( Fig 4A ) . Following that import period in 2007–2009 , the model predicted a very low probability of endemicity ( Fig 4B , Forecast 1 ) , with 99 . 74% of the trajectories indicating extinction in 2016 in a closed ecosystem . As of October 2016 , 97 . 9% ( 95%CrI [97 . 6–98 . 2] ) of the Mayotte livestock is estimated to be susceptible to RVF , such that the import of 40 infectious animals at that date ( low NDVI values , Fig 4C , Forecast 6 ) or in April 2017 ( high NDVI values , Fig 4D , Forecast 11 ) would result in an incidence peak similar to 2008 ( Forecast 6: Jul 2017: 1063 cases 95%CrI [242–2385]; Forecast 11: Jul 2018: 936 cases 95%CrI [297–1696] ) . However , if the number of infectious animals introduced into Mayotte is less than 40 , then the incidence peak remains substantially higher if importations take place in April , compared to importations taking place in October ( April: Forecasts 2–5 in S8A–S8D Fig and October: Forecasts 6–10 in S9A–S9D Fig ) .
Our work is the first dynamic mathematical model on RVF that accounts for climate and animal imports , and which is fitted to long-term epidemiological data [13] . The importance of livestock import was characterized as a major driver for RVF emergence , similarly to what has been described for Madagascar [9] . Our model narrowed the virus entry window in Mayotte to May-July 2007 , which represents a plausible 6 to 4-months delay following the first RVF report in Kenya and Tanzania ( Dec 2006 and Feb 2007 ) [10 , 18 , 19] . We also estimated the import of about 40 infectious animals per month over 23 months , which is possible back in 2007–10 . In the absence of animal movement data and epidemic curve in neighbouring territories , we assumed constant entrant flows of animals . While the actual entrant flows may have varied with time , due to climatic or anthropogenic factors ( such as political or economic factors ) , the proportion of boats seized may have also varied for the same reasons . Therefore , choosing a constant import flow was the least biased and most parsimonious option , that could be improved should better data be available . Rainfall and temperatures are known to have an impact on the dynamics of vector populations , and RVF virus can be transmitted by a large range of vectors species with different bio-ecologies [2 , 14] . The dynamics of rainfall and temperature may therefore result in a complex RVF vector multi-population dynamics for which no data are available in our case; and attempting to account for this without data would only increase the model uncertainty . In addition , studies on RVF vectors ( Culex pipiens and Aedes taeniorhynchus ) showed that temperature above 26°C increased virus transmission rates [51 , 52] , while in Mayotte the average temperature varies between 25°C and 35°C [32] , potentially allowing transmission year-round . If data on vector population dynamics were available , and if the ecosystem studied could bear cooler temperatures , both temperature and rainfall should be accounted for . Here , we used NDVI as a proxy for vector habitat and therefore vector density in common with many previous RVF studies did [10 , 11 , 44–46 , 53] . Since Mayotte has not reported any NDVI anomalies as in the Horn of Africa [10] , using monthly NDVI was the most relevant parameter to use . Furthermore , no previous dynamic models have used RVF transmission as a direct function of NDVI [13] , although NDVI is used in most spatial modelling works [10 , 11 , 44–46 , 53] . Our model allowed quantifying of a functional relationship between NDVI and transmissibility for RVF , with the highest Rs value being 2 . 19 , falling within the range of previously estimated R0 at 1 . 19 ( 95%CI [1 . 18–1 . 21] ) [54] in a theoretical endemic setting; or 1 . 18 ( range 0 . 5–2 . 1 ) [55] , and 1 . 17 ( range [0–3 . 68] ) [17] in an epidemic context . Finally , our model offers a benchmark for exploring RVF transmissibility without vector data , and should be tested in ecosystems with different NDVI dynamics . The credibility intervals of the estimated parameters were relatively narrow , and impacted only on the variability of the trajectories observed in 2004–08 , when monthly prevalence estimates were informed by a small number of sampled animals , generating large confidence intervals . Little information was available on how these samples were collected [25] , which could bias the model results . However , these samples retrospectively analyzed were randomly selected from a bank of sera collected under the annual veterinary services prophylaxis campaign , which attempted to be representative of the livestock population . Both models did not reach the peak prevalences in 2008–09 , and since these points corresponded indeed to recent infections ( IgM positive animals ) [25] , a biased serological sample in the data collected remains the most plausible explanation . From July 2008 onwards , trajectories showed only very little variability , and indicated a very low probability of a future re-emergence in the absence of new viral introductions; which is consistent with previous modelling conducted for Mayotte [34] . Finally , after 2008 , the model is fitted to the age-stratified IgG prevalence and is in good agreement with the observed IgG prevalence , even in the latest years ( 2015–2016 ) . The overall observed IgG prevalence in 2016 which appears higher than the simulated one is an artefact which can be explained by a high number of animals sampled in the oldest age-group . We assumed that animals were at equal risk of acquiring infection and becoming seropositive across species and age-groups . Indeed , in Mayotte all animals regardless of age and species are raised outdoors , and we therefore assumed that they were at equal risk of being exposed to mosquito bites . In addition , while some studies found differences in serological prevalence between livestock species [56 , 57] , a number of serological studies conducted in different study areas across Africa , such as Mozambique , Senegal , Tanzania , Kenya and Madagascar , did not show any difference in seroprevalence between livestock species during an epidemic or inter-epidemic period [23 , 41 , 58–64] . Finally , due to the island’s small size and the limited spatial variation of the ecosystem [65]; but also since herds are small ( about 5–7 animals ) [25 , 29] , and herds from all communes had been affected by RVF , we did not need to account for spatial heterogeneity . This also allowed implementing model fitting in a data-scarce environment . Stratifying per location would have resulted in data points supported by fewer samples , increasing uncertainty and precluding model fitting . In addition , since our livestock population did not experience the classical symptoms of RVF ( waves of abortions and high mortality in newborn ) , disease-induced mortality was not explicitly modelled . Sub-clinical forms were common for RVF in Mayotte , as well as in the neighbouring Union of Comoros and Mozambique [23 , 41–43] . Also , sheep , the most susceptible species to clinical symptoms , only represents 3–4% of the livestock population of the island . Finally , in the absence of vector data , nor evidence on human-to-animals RVF transmission , we assumed that the import of infectious animals was the most likely virus introduction pathway , although the introduction through infectious vectors and infectious humans cannot be ruled out . Model forecasts indicated a very low probability of RVF virus endemicity and therefore of re-emergence in a closed system . With a very high proportion of naive animals as reached in 2016 , the livestock population remains vulnerable to the introduction of infectious animals . Since 2011 , few RVF infections in Mayotte have been reported ( few young RVF IgG positive animals , or IgM positive animals in 2013–15 [25] , and no IgM positive in 2016 ) , whilst the surveillance system has been strengthened over years , giving weight to our model results . Ongoing surveillance including both active ( annual serological surveys ) , passive surveillance activities ( reports of animal mortality and abortions by farmers ) , but also the strict control measures for illegally introduced animals ( immediate euthanasia ) are still currently in place in Mayotte . Given that the animal population is naïve , our results suggest that such surveillance must be maintained , and reinforced should RVF be reported in neighbouring territories . This includes raising farmers’ awareness to report mortality and abortion events , and mitigating the risk of human exposure through communication and preventive messages ( best practices for abortion and raw meat handling , since most animals are still slaughtered at the farm with no individual protection equipment ) . Finally , assuming the availability of RVF , NDVI and animal movement data , our model framework could be adapted to other ecosystems to refine the ecosystem-specific relative role of livestock susceptibility , animal movements and NDVI-related transmissibility on RVF dynamics . | Rift Valley fever ( RVF ) is an arboviral hemorrhagic fever affecting primarily livestock in Africa and in the Arabian Peninsula . The conditions leading to RVF emergence are not fully understood , mainly because of data scarcity . Applied to the island of Mayotte ( our ecosystem under study ) , for which 12 years RVF serological dataset are available , and by using a mechanistic model , we demonstrate that RVF epidemics related mainly to the introduction of infectious animals . Our work confirms that anthropogenic factors , such as livestock movements , need to be accounted for in order to understand the epidemiology of this disease . Our model should be tested elsewhere , with ecosystem-specific data . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"livestock",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"animal",
"models",
"of",
"disease",
"tropical",
"diseases",
"microbiology",
"seasons",
"rift",
"valley",
"fever",
"age",
"groups",
"neglected",
"tropical",
"diseases",
"veterinary",
"science",
"research",
"and",
"analysis",
"methods",
"animal",
"models",
"of",
"infection",
"infectious",
"diseases",
"zoonoses",
"veterinary",
"epidemiology",
"serology",
"animal",
"studies",
"ecosystems",
"agriculture",
"people",
"and",
"places",
"ecology",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"viral",
"diseases"
] | 2017 | Drivers for Rift Valley fever emergence in Mayotte: A Bayesian modelling approach |
Amyloids are highly organized protein aggregates that are associated with both neurodegenerative diseases such as Alzheimer disease and benign functions like skin pigmentation . Amyloids self-polymerize in a nucleation-dependent manner by recruiting their soluble protein/peptide counterpart and are stable against harsh physical , chemical , and biochemical conditions . These extraordinary properties make amyloids attractive for applications in nanotechnology . Here , we suggest the use of amyloids in the formulation of long-acting drugs . It is our rationale that amyloids have the properties required of a long-acting drug because they are stable depots that guarantee a controlled release of the active peptide drug from the amyloid termini . This concept is tested with a family of short- and long-acting analogs of gonadotropin-releasing hormone ( GnRH ) , and it is shown that amyloids thereof can act as a source for the sustained release of biologically active peptides .
Amyloid fibrils , which are highly organized protein/peptide aggregates , are the hallmark of amyloid disease [1] . More than 20 human diseases are associated with disease-specific protein/peptide aggregation and amyloid fibril formation . For example , in Parkinson disease , the protein α-synuclein ( α-Syn ) and in Alzheimer disease , the amyloid peptide Aβ undergo structural alterations forming amyloid fibrils . These amyloid fibrils or conformational intermediates thereof are thought to be the toxic entity of the disease [1] . In striking contrast to the disease-associated amyloids , there are also amyloids with beneficial biological activities . For example , bacteria such as Escherichia coli express extracellular amyloid fibrils called curli that are involved in surface and cell–cell contacts promoting community behavior and host colonization [2] . The protein of chorion of the eggshell of silkworm is an amyloid that protects the oocyte and the developing embryo from a wide range of environmental hazards [3] . Yeast prions do not cause cell death , and are associated with enhanced survival of the host in certain environmental conditions [4 , 5] . The HET-s prion protein of the filamentous fungus Podospora anserina forms infectious amyloids that control a native function called heterokaryon incompatibility , believed to limit the spread of viral DNA [6] . Furthermore , the human protein Pmel17 forms a functional human amyloid , which appears to be important in the formation of skin pigmentation [7] . In light of these functional amyloids and other evidence , it has been suggested that the disease-associated amyloid fibrils also have a beneficial function by sequestering toxic oligomers of the amyloid protein [8] . Since proteins with nonhomologous sequences form amyloid fibrils , Dobson's group proposed that under certain conditions , many proteins and peptides can form amyloid-like fibrils [9 , 10] , and therefore , the formation of amyloid fibrils is a generic property of the polypeptide chain [11] . Therefore it is likely that amyloid fibrils share common properties , and indeed , electron microscopy ( EM ) studies suggest that they all form long filaments of several hundred nanometers in length and only a few nanometers in diameter [12] . The filaments are composed of several protofilaments , and within a protofilament , hundreds of protein/peptide molecules are aligned with a cross-β-sheet topology [12 , 13] , which enables them to bind Congo red ( CR ) [14] and Thioflavin T ( Thio T ) [15] . This structural arrangement makes amyloids inert against harsh treatment such as heat , pH , proteases , and physical forces . Furthermore , amyloids grow in a nucleation-dependent manner [16] . The properties outlined above make amyloids an attractive tool for nanotechnology and bioengineering . For example , amyloids may be used to make nanowires [17 , 18] , as scaffolds for bioactive materials [19 , 20] , and as a substrate for neurite outgrowth and synapse formation , as well as for tissue repair and tissue engineering [21] . Furthermore , the mechanical properties of insulin amyloids are comparable with steel [22] , indicating the super stability of amyloid organization . Here , we explore the use of amyloids in a pharmaceutical application for the formulation of long-acting drugs . One possibility to make drugs long acting is their storage in a depot from which the drug is released slowly in a controlled fashion [23] . It is our rationale that an amyloid-forming protein/peptide drug has the necessary properties of a long-acting drug: ( 1 ) an amyloid is a stable reservoir containing only the peptide drug; ( 2 ) the high structural organization of the amyloid aggregate guarantees a controlled release of the peptide drug from the fibril termini; and ( 3 ) the peptide drug is active upon release . This concept for drug delivery is in direct contrast to another approach in which the amyloid propensity of a peptide is reduced for better solubility and efficacy [24] . We examined the concept of using amyloids for the formulation of long-acting drugs to a family of analogs of gonadotropin-releasing hormone ( GnRH; p-Glu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH2 ) . GnRH is produced in the hypothalamus , stimulates the secretion of gonadotropins , and thereby plays a major role in the modulation of reproductive functions [25 , 26] . GnRH analogs are now recognized as potential drugs for the management of sex steroid-dependent pathophysiologies , such as hormone-responsive prostate cancers , management or treatment of breast and gynecological cancers , endometriosis , precocious puberty , uterine myoma , ovarian hyperandrogenism , premenstrual syndrome , and the induction of ovulation [26–28] . Most of these disorders can be treated with long-term applications of GnRH analogs . However , because GnRH analogs are peptides , they must be administered subcutaneously on a daily basis , frequently over long periods of time . Therefore , long-acting GnRH analogs were developed with a duration of action of up to 50 d compared to a few hours for normal GnRH [29 , 30] . Some of these long-acting GnRH analogs exhibit concentration-dependent aggregation and form liquid crystals or gels in aqueous solution [31 , 32] . Powell et al . [31] observed that the addition of electrolyte affects the liquid crystal stability , as well as the temperature at which birefringent liquid crystals formed in aqueous formulations of the GnRH analog deterelix . The GnRH analog leuprolide forms β-sheet–rich aggregates and gels with increasing peptide concentration or the addition of salts [33] , whereas the GnRH antagonist A-75998 exhibits aggregation and gelation in aqueous solution [32] . Furthermore , Jiang et al . [34] observed that subcutaneous injection of a solution containing the GnRH antagonist orntide forms gel at the injection site at high doses . Here , we show that gels of long-acting GnRH analogs are composed of amyloid fibrils and suggest that the duration of action of GnRH analogs depends on the ability of the amyloids to slowly release active peptide .
The kinetics of fibril formation in vitro is generally a slow process over several days or weeks , and can be monitored by the increase of fluorescence intensity of the cross-β-sheet–sensitive dye , Thio T [15] . Thio T fluorescence is thereby insignificant during the initial incubation period of amyloid aggregation due to the nonexistence of Thio T–sensitive β-sheet aggregates . After the initial phase , a time-dependent increase of fluorescence intensity is observed until a maximum intensity is reached , at which time maximal fibril formation has occurred . Hence , the Thio T binding study provides information about the relative rate of the formation of amyloid fibrils . For the Thio T binding studies , all GnRH analogs were incubated at room temperature at a concentration of 1 mg/ml in 5% D-mannitol , conditions identical to the formulation used in the animal experiments [29] . In addition , 0 . 01% sodium azide was added . The Thio T fluorescence was monitored daily for all analogs over a period of 21 d ( Figure 1 ) . With the exception of analog S10 , there is no significant Thio T binding by short-acting analogs during the entire incubation period . In contrast to short-acting analogs , all long-acting analogs show a significant increase of Thio T intensity indicative of fibril formation . The long-acting analogs L20 and L23 bind significantly more Thio T during the entire experiment , which suggests that these analogs form amyloids immediately upon sample preparation . The Thio T binding studies are summarized in Figure 1A and 1B: With the exception of S10 , all long-acting analogs produce three to four times more Thio T signal relative to short-acting analogs after 21 d of incubation , indicating that duration of action is correlated with amyloid formation . To confirm amyloid formation by GnRH analogs , CR binding was measured . CR , like Thio T , is a dye routinely used to detect β-sheet–rich assemblies [35] . CR binding can be measured by the “red shift” change of its absorption maximum from 490 to 540 nm , by an increase of the dye's molar absorptivity , and in favorable cases by eye . All three types of measurements were performed with GnRH analogs incubated for 30 d . The data ( Figure S1 and Table 1 ) show that among 12 of the short-acting analogs , seven do not bind CR , indicative of no amyloid formation . However , five short-acting GnRH analogs , i . e . , S3 , S4 , S6 , S10 , and S11 , bind CR , showing that some short-acting GnRH-analogs are able to form amyloids after prolonged incubation ( see also below ) . In contrast to short-acting analogs , all long-acting analogs bind CR . To formally determine the formation of amyloid fibrils by GnRH analogs and to characterize other possible assemblies present in the sample , EM was done on all of the GnRH analogs . Since the EM analysis is for a qualitative interpretation only , amyloid fibrils and other small aggregates might be observed under the electron microscope even if only a minority of the sample forms fibrils . Samples were studied after 8 and 30 d of incubation . The EM study at day 8 ( Figure 2 ) shows that among 12 short-acting GnRH analogs , only four ( S3 , S4 , S10 , and S11 ) form amyloid-like fibrils . After day 30 , the EM data suggest that the short-acting analogs S1 , S5 , S7 , S8 , S9 , and S12 do not form any detectable amyloid-like fibrils , but that all other short-acting analogs ( i . e . , S2 , S3 , S4 , S6 , S10 , and S11 ) form amyloid-like fibrils ( Figure S2 ) . In contrast , all long-acting GnRH-analogs form amyloid-like fibrils after 8 and 30 d of incubation ( Figures 2 and S2 ) . The fibrils show a wide variety of morphologies , comprising helical ultrastructures with varying degrees of lateral association . Furthermore , the fibrils are composed of one or more filaments , resulting in overall diameters ranging from approximately 6 to 12 nm . No oligomers are observed for the long-acting compounds at day 8 or at day 30 ( Figures 2 and S2 ) . Short worm-like protofibrils are observed only for L16 at day 8 ( Figure 2 ) . The combination of the Thio T , CR binding , and EM studies summarized in Table 1 show that all the long-acting GnRH analogs form amyloid fibrils within a short time period . Half of the short-acting analogs do not form fibrils under any of the conditions tested . However , the other half of the short-acting analogs is able to form fibrils after extended incubation times and S10 forms fibrils immediately . Therefore , amyloid fibrils of GnRH analogs appear to be required but not sufficient for a long-duration of action . Although amyloid fibrils are very stable , they must follow thermodynamic equilibrium conditions between fibril and monomer . The fibril–monomer equilibrium is of particular importance for the proposed pharmaceutical application , because fibrils must be both stable enough to guarantee a long duration of action , and the release of active peptide must be sufficient to generate effective drug concentrations . To address these two opposing tasks , peptide release for all GnRH analogs was measured by a dialysis experiment similar to a study of the GnRH agonist leuprolide [36] . A total of 200 μl of 30-d-old GnRH analog ( in ) was dialyzed 4 ml of 5% D-mannitol ( out ) . A 3 . 5-kDa cutoff , which is twice the size of the peptides studied , was chosen to ensure that the released particles are most likely monomeric . The peptide release and the remaining aggregates were monitored by fluorescence over 35 d at room temperature . All but two analogs ( i . e . , L16 and L19 ) produced good aggregation-independent fluorescence signal in the emission range of 290–500 nm ( excited at 280 nm ) and could be analyzed ( see Material and Methods for more details ) . Figure 3 shows representative examples of the dialysis experiment displaying normalized fluorescence intensities at wavelength maxima ( λmax ) both inside ( in ) and outside ( out ) of the membrane . In Figure 3A the monomer release profile of wild-type GnRH ( S12 ) and the short-acting analog S1 , neither of which form fibrils after 30 d of incubation ( Figures 1 and 2 ) , are shown . The fluorescence intensities inside and outside of the membrane decrease and increase , respectively , within a very short period of time and reach equilibrium within 3 d ( Figures 3A and S3 ) . The t1/2 and Koff of S1 and S12 are approximately 2 d and approximately 4 × 10−6 S−1 ( Table 1 ) , representing the intrinsic diffusion of monomeric GnRH analogs through the membrane . Similar release profiles are observed for all of the short-acting analogs , which do not form fibrils after 30 d of incubation , as well as the fibril-forming but short-acting analogs S2 and S6 ( Figure 3B ) and S11 ( Table 1 ) . The latter data suggest that S2 , S6 , and S11 form unstable fibrils . In contrast , the release profiles of fibril-forming , but short-acting , analogs S3 and S10 ( Figure 3C ) and S4 ( unpublished data ) show very little fluorescence intensity changes both inside and outside of the membrane at 35 d . Therefore , the t1/2 and Koff of S3 , S4 , and S10 are ≫35 d and ≪2 × 10−7 ( Table 1 ) , respectively . The data suggest that S3 , S4 , and S10 form very stable amyloids with very little peptide release . This hypothesis is further supported by the finding that S4 is present as amyloid at the injection site over a period much longer than its duration of action ( see below and Figure S6 ) . It can be concluded that the release profile of all short-acting analogs indicate that short-acting analogs are of three categories: ( 1 ) not forming fibrils , ( 2 ) forming very unstable fibrils that readily dissociate into monomers , and ( 3 ) forming very stable fibrils such as S4 which release peptides so slowly that an effective concentration of the drug for action eventually is not reached . In striking contrast , for all the long-acting GnRH analogs , a gradual decrease in fluorescence intensity inside and a corresponding increase outside of the membrane is observed between days 3 and 35 after the initial intrinsic diffusion of residual monomers within the first 3 d ( Figure 3D; the residual monomer comprises thereby only 1%–5% of peptide as estimated from the relative fluorescence intensities ) . The half-life ( t1/2 ) of all long-acting analogs ranges from day 15 to 30 d ( Table 1 ) . The Koff of the long-acting analogs ( 2 × 10−7 to 7 × 10−7; Table 1 ) suggest that long-acting analogs release monomer approximately ten times faster than short-acting , very stable , fibril-forming GnRH analogs such as S4 . Monomer release in vitro may be altered in vivo by amyloid binding components from the host , such as amyloid P and glycosaminoglycans , which may stabilize amyloid fibrils [37 , 38] . To test the effect of glycosaminoglycans on fibril formation and peptide release , the fibril formation and monomer release profiles of samples of compounds L13 and L14 were measured in absence and presence of low molecular weight ( LMW ) heparin ( 600 μM peptide in 5% D-mannitol , 0 . 01% sodium azide with or without equimolar LMW heparin ) . After overnight incubation in the absence of LMW heparin , both L13 and L14 samples bind Thio T indicative of amyloid formation . These data are in line with earlier experiments ( Figures 1 and 2 ) . Interestingly , the presence of LMW heparin results in much higher Thio T binding for both L13 and L14 peptides ( Figure S4 ) . These findings suggest that LMW heparin induces fibril formation of peptide analogs L13 and L14 . Furthermore , the dialysis experiment shown in Figure S5 suggests that the presence of LMW heparin results in the slower release of monomers and a significantly slower decay of the fibrillar material relative to the control . Therefore , host factors may influence the in vivo properties of amyloid fibrils of the GnRH peptide hormone analogs . In particular , they may stabilize the fibrils . To determine whether GnRH analogs form and/or stay as amyloids in vivo at the injection site , we injected them into adult rats and stained the potential amyloid with CR [39] . A fresh preparation of the long-acting analog L14 was subcutaneously injected at a concentration of 200 μg in 200 μl of 5% D-mannitol . The sample before injection was in a non-amyloid state as measured by EM and Thio T binding ( unpublished data ) . Also 5% D-mannitol was injected as a control into the same animal . Twenty-four hours after the subcutaneous injection , the tissues close to the injection sites were taken for cryo sections , and the tissue sections were stained with CR . Figure 4 shows that a portion of the tissue is stained by CR and produces strong birefringence indicating the presence of amyloid fibrils . No CR binding was observed with either the control ( 5% D-mannitol only; unpublished data ) or upon subcutaneous injection of a fresh preparation of the non-fibrillar short-acting analog S2 ( unpublished data ) . In contrast , CR binding was observed upon injection of an aged fibril-containing S2 sample , as well as aged amyloid-containing or freshly prepared S4 samples . The presence of amyloids of the S4 sample was observed after 24 h and 7 d of subcutaneous administration , respectively ( Figure S6 ) . These data show that GnRH analogs are able to form amyloids in vivo . The capacity of peptide drugs to form amyloids in vitro has been previously documented for insulin [40 , 41] . The correlation between fibril formation and a long duration of action suggests that a peptide administered in a fibril form might be longer-acting than in its soluble form if the latter does not form fibrils immediately upon injection . A good candidate to test this hypothesis is S2 , since it is a short-acting analog that does not form fibrils upon injection ( see above ) , but is able to form fibrils in vitro upon long incubation ( Figure S2 ) . Therefore , the duration of action of a fresh-prepared S2 and an aged fibrillar S2 sample was measured in vivo with the castrated male rat assay [29] . In short , 10 d after castration , rats ( six per group ) were injected subcutaneously with 50 μg of freshly prepared S2 and aged fibrillar S2 in 50 μl of 5% mannitol in parallel with a control ( vehicle only ) . Blood sampling was performed predose , then 1 , 6 , 24 , 48 , 72 , and 96 h after subcutaneous administration . The effects of the control and the two S2 samples on the gonadotropic axis were determined by measurement of plasma luteinizing hormone ( LH ) levels by radioimmunoassay ( Figure 5 ) [29] . Both samples suppress LH levels in the first 24 h as expected [42 , 43] . However , after 72 h , the freshly prepared S2 sample lost its action , whereas the aged fibrillar S2 sample is still highly active in suppression of LH . The loss of action after 96 h of the fibrillar S2 sample is attributed to the low stability of the S2 fibrils determined in vitro ( Table 1 ) . Since the only difference between the two S2 samples is the aggregation state ( soluble versus fibrillar ) , the difference in the duration of action is therefore due to their different aggregation state . In conclusion , the formation of fibrils of a peptide drug can prolong its duration of action . Since amyloid fibrils can be toxic to cells and are associated with several neurological diseases [1 , 44] , it is a concern whether or not amyloid fibrils of GnRH analogs are able to generate a toxic response as well . To measure the potential toxicity of GnRH analogs , we used the lactate dehydrogenase ( LDH ) release assay for cell killing [44] . The LDH release assay is a widely used assay to determine the cytotoxicity of chemicals or environmental toxic factors . LDH is a soluble cytosolic enzyme that is released into the culture medium following the loss of membrane integrity . For the LDH assay , we used mouse embryo fibroblast cells and aged peptide samples at 60 μM in 5% D-mannitol . Figure 6 shows that in contrast to the lysate as positive control , neither the peptide analog samples nor Aβ ( 1–40 ) generated toxicity in a dermal cell line . These toxicity studies suggest that amyloid fibrils of GnRH analogs are not toxic to dermal cells . Another concern for the application of amyloids in the formulation of long-acting drugs is the potential multiplication and spread of amyloids through seeding . Only a small amount of fibrillar seed is needed to induce and accelerate self-polymerization in the presence of monomeric amyloid protein/peptide [16] . Within the context of GnRH , perhaps amyloids of the injected GnRH analogs may act as the seeds and the host GnRH as the reservoir of the free monomer . To mimic this situation in vitro , a seeding reaction was set up with a small amount of matured , long-acting L20 fibrils ( 5% v/v ) in presence of a solution of wild-type GnRH ( 95% v/v ) . The fibril growth of the mixture as well as the wild-type GnRH alone ( S12 ) and mature L20 fibrils was followed over a period of 1 mo by Thio T binding ( Figure S7A ) . There is no significant Thio T binding during the entire seeding experiment , which suggests that L20 fibrils are unable to seed wild-type GnRH . Further confirmation was obtained by EM at day 0 and after 1 mo of incubation ( Figure S7B ) . No detectable amyloid fibrils in the mixture or wild-type GnRH alone were observed , further supporting the notion that L20 is unable to seed wild-type GnRH . Although the seeding of amyloids requires a high amino acid sequence similarity between the protein/peptide of the seed and its soluble host counterpart [1 , 45] , cross-seeding between different protein species have been documented [46] . To test whether amyloids of long-acting GnRH analogs are able to cross-seed with a protein associated with Parkinson disease , α-Syn , a seeding reaction was done with a small amount of matured L20 fibrils ( 5% v/v ) in the presence of α-Syn ( 95% v/v; 400 μM concentration of monomeric solution [pH 7 . 4] in PBS ) . The fibril growth of the mixture as well as α-Syn in PBS as control were followed over a period of 30 d by Thio T binding . As shown in Figure 7 , α-Syn aggregates spontaneously as reported [47] . The addition of a seed of L20 fibrils did not significantly influence the aggregation of α-Syn , which suggests that the cross-seeding capacity of L20 amyloids with α-Syn is low . In contrast to the L20 fibril seeding experiment , the addition of α-Syn fibril seeds accelerated α-Syn aggregation significantly ( Figure 7 ) .
Drugs of peptide or protein origin usually have low oral or transdermal bioavailabilities and short in vivo half-life , and therefore require delivery by infusion , frequent injections , or subcutaneous administrations . To overcome these delivery problems , novel formulations of the drugs have been designed that release continuously over an extended period to maintain an active drug concentration within the therapeutic window . Methods for controlled drug release include implantable devices such as infusion pumps , chips , and diffusion chambers , as well as biodegradable and easy to administer hydrogels , liposomes , microspheres , and polymer depots loaded with the drug of interest [23 , 48 , 49] . Another compelling approach proposed is the crystallization of the drug followed by the administration of drug crystals , which serve as a protected reservoir that releases the active compound relatively slowly [50 , 51] . Such depot designs may offer numerous advantages . These include protection of the drug from enzymatic degradation , the ability to deliver the drug locally to a particular site in the body , as well as increased patient comfort , convenience , dosage accuracy , and assurance of patient compliance [52 , 53] . However , improved technologies and new approaches are still needed that can deliver otherwise insoluble , unstable , or unavailable therapeutic compounds to reduce the amount of drugs used , to release the drug in a “smart” manner , and to maximize the drug load , which is the amount of drug per depot [54] . Here , we suggest the use of a peptide drug with the ability to form amyloids before or immediately after administration . This approach may simultaneously overcome two problems since the peptide drug itself forms a stable structurally organized depot , that releases monomeric active peptide slowly from the amyloid fibril termini ( Figure 8 ) . Hence , the long-acting amyloid drug consists of only one type of molecule and concomitantly the depot has a drug load of 100% . Another advantage may be that the amyloid fibril formulation may protect the biological , physical , and chemical integrity of the drug molecules in the very stable cross-β-sheet structure during processing , storage , and even upon delivery . Furthermore , amyloid binding components from the host organism such as amyloid P and glycosaminoglycans may stabilize the drug fibrils and protect them from degradation [37 , 38 , 55 , 56] , and may reduce the potential immune response of the drug by covering the drug fibrils . In addition , an amyloid drug may be easily administered by subcutaneous injection , it is highly concentrated since it is already an aggregate , and the amyloid depot can easily be located close to the injection site and removed if necessary . Therefore , an amyloid formulation may allow sustained release of the peptide drug for a long duration , thus avoiding the need of repetitive dosing , surgery for implanting depots , or/and complicated manufacturing procedures for making a depot filled with the drug . To test the concept that amyloid properties of peptides/proteins may be used for the formulation of long-acting drugs , we studied GnRH analogs because long-acting GnRH analogs form gels [29] , which is one of the indicators for fibril formation . Indeed , in vitro analysis of a family of 23 short- and long-acting analogs shows that all of the long-acting analogs studied form fibrils that slowly release the peptide monomer ( Table 1 ) . Furthermore , we demonstrated that the long-acting analogs also have the capacity to form amyloids in vivo upon subcutaneous injection ( Figure 4 ) . In contrast to long-acting analogs , the short-acting GnRH analogs either do not form fibrils , form unstable fibrils , which release peptide quickly , or form very stable fibrils , which do not release peptide within the period studied ( Table 1 ) . In addition , we show that the administration of a fibrillar sample results in a longer duration of action than its corresponding soluble counterpart ( Figure 5 ) . The findings strongly suggest that the long duration of action is directly related to the ability of the GnRH analog to form fibrils that release sustained peptide ( Figure 8 ) . The in vitro half-life time of amyloids of GnRH analogs varies between 2 d to more than 35 d , the Koff varies more than one order of magnitude as well ( Table 1 ) . This observation also shows that the release of monomer from the amyloid depot can be controlled by the chemical structure of the GnRH analog . Concomitantly , the duration of action can be manipulated through the peptide release property of the amyloid fibril by the chemical structure of the analog . Since amyloid fibrils of disease-specific peptides/proteins have been associated with a number of diseases , a potential pitfall of the use of amyloid as storage depot could be amyloid-associated toxicity of the depot as well as its potential seeding capacity to amplify amyloids by recruiting soluble host counterpart , i . e . , wild-type GnRH in the context of this study . Furthermore , the depot might cross-seed other amyloidogenic proteins . The in vitro seeding experiments suggested that amyloid of the GnRH analog L20 can seed neither wild-type GnRH nor α-Syn , a peptide associated with Parkinson disease ( Figures S7 and 7 ) . The rationale of this lack of seeding is based on the finding that seeding requires a host counterpart which has the ability to form fibrils and comprises a high sequence homology to the amyloid [1 , 45] . Since , GnRH does not form fibrils ( Table 1 , S12 ) and α-Syn has no significant sequence homology with L20 , seeding is unlikely . In addition , the potential capacity of cross-seeding of the depot in vivo is also unlikely due to the different compartmental localization of the seed and its host counterpart , i . e . , amyloids of GnRH analogs in the subcutaneous layer and α-Syn in the brain . The brain and the subcutaneous layer are additionally separated by the blood–brain barrier , inhibiting the crossing of peptides to the central nervous system ( CNS ) . Furthermore , amyloids are not toxic to dermal cells ( Figure 6 ) , and long-acting analogs show good properties in histamine release [43] . One of the long-acting , amyloid-forming GnRH analogs under study is degarelix ( L13 ) . Its pharmacological profile has been well studied . Degarelix ( L13 ) has a good safety margin with regard to histamine release and long duration of action up to 50 d [30] attributed here to the formation of amyloids . Degarelix ( L13 ) is in clinical trials for the treatment of prostate cancer . The results of the Phase IIb clinical trial show that treatment with degarelix ( L13 ) results in fast , profound , and sustained reductions in testosterone and prostate-specific antigen levels without a testosterone surge . Degarelix ( L13 ) is currently being tested in Phase III trials [57] . Another long-acting and amyloid-forming GnRH analog studied here is ganirelix ( L23 ) , which is an approved drug ( brand name , Antagon ) for use in assisted reproduction to control ovulation . Our data suggest that its long duration of action is also because of its ability to form amyloid fibrils . Since the formation of amyloid is probably a generic feature of the polypeptides , the concept of designing peptide/protein drugs that make amyloid depots for the formulation of long-acting drugs could be applicable to many drugs of peptide or protein origin . Primary requirements of this approach are that monomers that escape from the fibril end be functional and that a given designed segment , which may include , if necessary , an amyloidogenic tag , of the peptide/protein drug forms an amyloid core . The successful application to treat human diseases using long-acting , amyloid forming GnRH analogs such as degarelix and ganirelix underline the potential of the concept . It follows that in striking contrast to the original association of amyloids with diseases , amyloids might also be useful in the treatment of diseases .
Peptides were synthesized using the solid-phase approach and the Boc strategy as reported earlier [29] . GnRH analogs were dissolved in a glass tube in 1 ml of 5% D-mannitol and 0 . 01% sodium azide at a concentration of 1 mg/ml . The GnRH analogs were then incubated at room temperature without stirring . The fibril formation was monitored by EM , CR , and time-resolved by Thio T binding studies . Three independent experiments were performed for each sample . A 10-μl aliquot of peptide sample was diluted into 500 μl of water with 5% D-mannitol containing 0 . 01% ( w/v ) sodium azide . The solution was mixed with 10 μl of 1 mM Thio T prepared in the same solution . Fluorescence was measured immediately after addition of Thio T . The experiment was measured on a spectrofluorimeter ( Photon Technology International ) with excitation at 450 nm and emission at 482 nm . The fluorescence intensity at 482 nm was plotted against incubation time ( day ) . A rectangular 10-mm quartz microcuvette was used . Three independent experiments were performed for each sample . A 5-μl aliquot of peptide sample was mixed with 80 μl of PBS buffer containing 10% ethanol . Then 15 μl of a 100 μM CR solution ( filtered through 0 . 2-μm filter ) in PBS containing 10% ethanol was added . After mixing , UV was measured from 300–700 nm . For the measurement of the CR-only spectrum , 15 μl of CR solution with 85 μl of PBS containing 10% ethanol was prepared . As a control , a 5-μl aliquot of GnRH analog mixed with 95 μl of PBS containing 10% ethanol was measured . Three independent experiments were performed for each sample . In addition to the UV measurements , pictures of the cuvette containing the peptide sample and CR were taken to visually show the CR binding to the amyloids . A 5-μl aliquot of peptide sample was diluted into 50 μl of water to a peptide concentration of approximately 50 μM , spotted on a glow-discharged , carbon-coated Formvar grid ( Electron Microscopy Sciences ) , incubated for 5 min , washed with distilled water , and then stained with 1% ( w/v ) aqueous uranyl formate solution . Uranyl formate solutions were filtered through 0 . 2-μm sterile syringe filters ( Corning ) before use . EM analysis was performed using a JEOL JEM-100CXII electron microscope at 80 kV with nominal magnifications between 36 , 000 and 72 , 000 . Images were recorded digitally by using the SIS Megaview III imaging system . At least two independent experiments were carried out for each sample . The 600 μM L13 and L14 hormone analogs were mixed with 600 μM LMW heparin ( 5-kDa heparin from CalBioChem ) in 5% D-mannitol , 0 . 01% sodium azide ( 1:1 v/v ) . The mixtures were incubated overnight . For the control , 600 μM L13 and L14 hormone analogs were also incubated with 5% D-mannitol ( 1:1 v/v ) . The next day , a 10-μl aliquot of peptide sample was diluted into 500 μl of H2O with 5% D-mannitol containing 0 . 01% ( w/v ) sodium azide . The solution was mixed with 10 μl of 1 mM Thio T prepared in the same solution . Fluorescence was measured immediately after addition of Thio T with a spectrofluorimeter ( Photon Technology International ) using an excitation wavelength at 450 nm and emission at 482 nm . The fluorescence intensity at 482 nm was plotted . Three independent experiments were performed for each sample . A 200-μl aliquot of aged GnRH in solution conditions given above was transferred into a 0 . 6-ml Eppendorf tube with a hole in the cap . To study the effect of heparin on peptide release , a mixture of hormone analog and heparin ( 1:1 v/v ) was incubated overnight in parallel with a half-diluted peptide fibril sample in 5% D-mannitol and 0 . 01% sodium azide . The hole in the cap of the Eppendorf tube was sealed with a 3 . 5-kDa molecular weight cutoff ( MWCO ) membrane . Next , the Eppendorf tube was inserted upside down into a 15-ml Falcon tube containing 4 ml of 5% D-mannitol in water and 0 . 01% sodium azide . Care was taken to make sure that the membrane was exposed to solutions both in and outside and that any bubbles at the membrane interface were removed . The tubes were then incubated at room temperature . To measure the remaining peptide concentration inside the 0 . 6-ml Eppendorf tube , a 5-μl aliquot was diluted into 500 μl water with 5% D-mannitol , 0 . 01% sodium azide , and the fluorescence was measured with excitation at 280 nm and emission in the range of 290–500 nm . The excitation slit width was 2 nm , and emission slit width was 11 nm . To determine the peptide concentration outside the membrane in the 15-ml tube , an aliquot of 250 μl of solution was taken from the releasing medium , diluted into 500 μl final volume by 5% D-mannitol , 0 . 01% sodium azide , and fluorescence was measured . Following spectra recording , the dialysis system was reassembled as described above . No significant change ( ±5 nm ) of the wavelength maxima ( λmax ) of the fluorescence spectra from solutions inside and outside of the membrane was observed . Furthermore , an independent experiment , in which the intrinsic fluorescence during aggregation of GnRH analogs was measured over a period of 30 d , did not show significant changes of fluorescence intensity and wavelength maxima from freshly prepared solution to aggregates by 30 d of incubation ( unpublished data ) . The normalized intensity maxima from inside and outside the membrane were plotted as a function of time and fit by an exponential function to extract the half-life of the fibrils and the Koff , respectively . The data from day 0 and day 1 of incubation were not used in the analysis , because loosely bound monomer or monomers that are already in equilibrium with fibrils pass through the membrane during this initial incubation period , resulting in a bi-exponential function , which is difficult to fit . The extracted half-life , t1/2 , is the time taken , during which the fluorescence signal inside the 0 . 6-ml Eppendorf tube drops by 50% . The dissociation constants ( Kd ) of the aggregates given are equal to the final monomer concentration in accordance to O'Nuallain et al . [58] . Skin harvest: Adult male Sprague-Dawley rats ( 180–220 g ) were injected subcutaneously in shaved areas with the vehicle , freshly prepared ( i . e . , L14 , S2 , S4 , and vehicle-only; 200 μg/200 μl at four sites/rat ) , or aged GnRH analogs ( i . e . , S2 and S4; 200 μg/200 μl at one site/rat ) . Twenty-four hours or 7 d later , the animals were deeply anesthetized and tissues , including skin and muscle around the injection sites , were collected . They were then fixed in fresh 4% paraformaldehyde for 7 d , postfixed in 20% sucrose/4% PFA overnight , and divided into three sections per treatment . Each section was cut on cryostat at 10-μm thickness and stained with CR . LH levels: Adult male Sprague-Dawley rats ( 180–200 g were castrated under isoflurane anesthesia 10 d prior to the start of the experiment , and implanted with indwelling jugular cannulae 48 h prior to the assay for serial blood sampling [59] . The peptides ( 200 μg ) were dissolved either fresh in 200 μl of bacteriostatic water containing 5% mannitol or aged under the same conditions for 4 mo . The rats were injected subcutaneous with a total dose of 50 μl . Controls received the vehicle . Plasma LH levels were determined by radioimmunoassay ( RIA ) using reagents provided by the National Pituitary and Hormone Distribution Program of the National Institute of Diabetes and Digestive and Kidney Diseases ( NIDDK; Bethesda , Maryland ) [60] . All protocols were approved by the Salk Institute's Institutional Animal Care and Use Committee ( IACUC ) . The CR staining was performed using the diagnostic amyloidal stain kit HT60 from Sigma . Briefly , microscopic slides containing tissue were stained with hematoxylin solution ( Gill No . 3 ) for 10 min and rinsed in tap water for 5 min . Slides were placed in alkaline sodium chloride solution for 20 min and subsequently stained in alkaline CR solution for 20 min . The slides were rinsed three times with absolute ethanol and cleared in xylene before mounting . The samples were analyzed using a microscope equipped with a dual polarizer and a charge-coupled device ( CCD ) camera . For the LDH assay , we used mouse embryo fibroblast cells and peptide fibrils at 60 μM final concentration in 5% D-mannitol . For the buffer control , we used 5% D-mannitol in water . We also used 1 μM Aβ ( 1–40 ) and a cell lysate to compare our data . The assay was done using a kit ( Sigma ) as described by the manufacturer . The assays were done in triplicate and the data presented as the standard error of the mean . A total of 5 μl of matured L20 fibril solution at a concentration of 1 mg/ml in 5% D-mannitol , 0 . 01% sodium azide was mixed with 95 μl of freshly prepared GnRH solution in the same solution condition followed by incubation at room temperature without stirring . At day 0 and day 30 , aliquots of the mixture as well as the original solution were taken for EM . Once a week , a 10-μl aliquot was diluted into 500 μl water containing 5% D-mannitol , 0 . 01% sodium azide , and 10 μl of 1 mM Thio T to measure the Thio T fluorescence as described above . Wild-type α-Syn was expressed and purified according to Kessler et al . [47] . Solid σ-Syn was dissolved in PBS ( pH 7 . 4 ) , 0 . 01% sodium azide at a concentration of 10 mg/ml . The pH was adjusted to 7 . 4 by adding the appropriate amount of NaOH solution . The solution was briefly vortexed and then transferred to filters ( 50 , 000 MWCO , Microcon YM-50; Millipore ) , and centrifuged for 30 min at 16 , 000g using a benchtop microcentrifuge ( Eppendorf model 5417R; Brinkmann Instruments ) . The cutoff filters were washed three times by centrifuging with 200 μl of PBS ( pH 7 . 4 ) before use . The resulting solution should contain only monomer or dimer wild-type α-Syn . The protein concentration of the supernatant was measured by UV absorbance with ε280 of 6 , 500 M−1cm−1 . The 400 μM α-Syn solution was incubated with 5% ( v/v ) seed of matured L20 amyloid in 5% D-mannitol at 37 °C with vigorous stirring . α-Syn with only 5% D-mannitol was also incubated as a control . The monomeric solution of α-Syn was incubated with 5% ( v/v ) of sonicated α-Syn fibril seed as a positive seeding control . The formation of amyloid was monitored regularly for the mixture ( α-Syn+L20 ) and α-Syn by Thio T binding . | Amyloids are highly organized protein aggregates that are associated with both neurodegenerative diseases such as Alzheimer disease and benign functions such as skin pigmentation . Amyloids self-polymerize by recruiting their soluble protein counterpart and remain stable against harsh physical , chemical , and biochemical conditions . These extraordinary properties make amyloids attractive for applications in nanotechnology . Here , we suggest the use of amyloids in the formulation of long-acting drugs , which are active over extended periods of days and weeks . Long-acting drugs have been designed to increase patient comfort , convenience , dosage accuracy , and assurance of patient compliance for drugs that have a low oral bioavailability . It is our rationale that amyloids have the properties required of a long-acting drug because they are stable depots that guarantee a controlled release of the active peptide drug from the amyloid termini . This concept is tested with a family of short- and long-acting analogs of gonadotropin-releasing hormone , and it is shown that amyloids thereof can act as a source for the sustained release of biologically active peptides . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"biochemistry",
"neurological",
"disorders",
"pharmacology",
"biophysics"
] | 2008 | Amyloid as a Depot for the Formulation of Long-Acting Drugs |
HIV-1 entry requires the cell surface expression of CD4 and either the CCR5 or CXCR4 coreceptors on host cells . Individuals homozygous for the ccr5Δ32 polymorphism do not express CCR5 and are protected from infection by CCR5-tropic ( R5 ) virus strains . As an approach to inactivating CCR5 , we introduced CCR5-specific zinc-finger nucleases into human CD4+ T cells prior to adoptive transfer , but the need to protect cells from virus strains that use CXCR4 ( X4 ) in place of or in addition to CCR5 ( R5X4 ) remains . Here we describe engineering a pair of zinc finger nucleases that , when introduced into human T cells , efficiently disrupt cxcr4 by cleavage and error-prone non-homologous DNA end-joining . The resulting cells proliferated normally and were resistant to infection by X4-tropic HIV-1 strains . CXCR4 could also be inactivated in ccr5Δ32 CD4+ T cells , and we show that such cells were resistant to all strains of HIV-1 tested . Loss of CXCR4 also provided protection from X4 HIV-1 in a humanized mouse model , though this protection was lost over time due to the emergence of R5-tropic viral mutants . These data suggest that CXCR4-specific ZFNs may prove useful in establishing resistance to CXCR4-tropic HIV for autologous transplant in HIV-infected individuals .
For HIV to infect cells , the viral envelope ( Env ) protein must bind to the host protein CD4 and then to a coreceptor , most commonly CCR5 ( R5 HIV ) ( reviewed in [1] ) . The importance of CCR5 for HIV-1 pathogenesis is shown by the fact that individuals who are homozygous for an inactivating 32 base pair deletion in ccr5 ( ccr5Δ32 ) are highly resistant to HIV infection [2] , [3] , while heterozygotes typically live longer after HIV infection due to reduced CCR5 expression levels [4] , [5] . Recently , an HIV infected patient with acute myelogenous leukemia received a bone marrow transplant from a ccr5Δ32 homozygous donor [6] . This patient's viral load remains undetectable even in the absence of anti-retroviral therapy more than three years post-transplant , suggesting that this individual's HIV infection has been eradicated . In theory , the success of this approach could be recapitulated by inhibiting CCR5 with an orally bioavailable small molecule such as maraviroc , which binds to CCR5 and prevents its use by most R5 HIV-1 strains . However , virus strains that can utilize CXCR4 either in place of ( X4 HIV ) or in addition to CCR5 ( R5X4 HIV ) are found at significant levels in roughly 50% of late-stage infected individuals [7] , [8] , supporting the need for therapies targeted to CXCR4 [9] . Ideally , an approach to target CXCR4 would complement CCR5-specific therapy , but the broad expression pattern of CXCR4 has made systemic inhibition of this coreceptor by small molecules problematic [10] , [11] . In addition , resistance to CCR5 and CXCR4 antagonists can arise in patients by mutations in the viral envelope protein that enable it to utilize the drug-bound forms of these coreceptors [12]–[16] . The ability of HIV-1 to adapt to new selective pressures and the plasticity with which Env interacts with its coreceptors argues for approaches that reduce or eliminate coreceptor expression rather than simply altering coreceptor conformation . If approaches could be developed that specifically target expression of both CCR5 and CXCR4 on CD4+ T cells , virus entry should be inhibited more effectively . Several genetic approaches have been taken to reduce or eliminate CCR5 expression in human cells , including the use of ribozymes [17] , [18] , single-chain intracellular antibodies [19] , trans-dominant coreceptor mutants [20] , and RNAi [21] , [22] . However , these studies are limited by the requirement for stable expression of an exogenous gene . To circumvent this , a CCR5 specific zinc-finger nuclease pair ( R5-ZFNs ) has been developed [23] . Zinc finger proteins that recognize a specific 24 bp DNA sequence are fused with a monomeric cleavage domain from FokI endonuclease that functions only as a dimer ( Figure 1 ) . For DNA cleavage to occur , two zinc finger proteins must bind , each to specific , adjoining sequences in the CCR5 gene , leading to FokI dimerization and subsequent DNA cleavage resulting in a double strand break [24]–[26] . The double strand break then can be repaired by error-prone non-homologous end joining ( NHEJ ) often introducing insertions and deletions leading to a non-functional gene product when this break is placed within the coding region of the targeted gene [27] . Following introduction into human CD4+ T cells [23] or hematopoietic stem cells [28] via an adenovirus vector or DNA nucleofection , respectively , the ccr5 gene was efficiently and specifically disrupted . This confers protection in vitro and in humanized mice to infection by HIV-1 isolates that require CCR5 ( but not CXCR4 ) . Several early stage clinical trials using autologous infusions of ZFN-generated CCR5-modified CD4+ T cells are currently underway ( clinicaltrials . gov identifiers NCT00842634 , NCT01252641 , NCT01044654 ) . In this study we describe the design and pre-clinical evaluation of a CXCR4-specific ZFN pair ( X4-ZFNs ) that specifically and efficiently disrupts cxcr4 , rendering human CD4+ T cells permanently resistant to HIV-1 strains that require CXCR4 for infection . We also demonstrate that cxcr4 can be safely and efficiently disrupted in CD4+ T cells obtained from ccr5Δ32 homozygotes resulting in cells resistant to all strains of HIV-1 tested . This suggests that combined treatment of mature CD4+ T cells with X4-ZFNs and R5-ZFNs can provide permanent protection against HIV-1 infection .
We designed ZFNs specific to the human and rhesus CXCR4 and CCR5 genes using a previously described approach [29] . One ZFN pair was used to target both the human and rhesus macaque CXCR4 genes since the 24 bp target sequences are identical . Zinc-finger proteins were optimized against the target gene sequence and assembled as described [30] from an archive of in-vitro-selected modules [31] , [32] . The ZFP moieties ( target gene; ZFP name; target sequence ( 5′→3′ ) ; recognition α-helices ( finger number ) ) are as follows: CXCR4; X4-ZFN-L; GTAGAAGCGGTC , DRSALSR ( 1 ) , RSDDLTR ( 2 ) , QSGNLAR ( 3 ) , QSGSLTR ( 4 ) ; CXCR4; X4-ZFN-R; GACTTGTGGGTG , RSDSLLR ( 1 ) , RSDHLTT ( 2 ) , RSDSLSA ( 3 ) , DRSNLTR ( 4 ) . Rhesus CCR5; rhR5-ZFN-L; GATGAGGACGAC , RSDNLAR ( 1 ) , TSGNLTR ( 2 ) , RSDNLAR ( 3 ) , TSGNLTR ( 4 ) ; Rhesus CCR5; rhR5-ZFN-R; AAACTGCAAAAG; RSDNLSV ( 1 ) , QKINLQV ( 2 ) , RSDVLSE ( 3 ) , QRNHRTT ( 4 ) . , The human CCR5-specific ZFNs are described in Perez et al [23] . The Ad5/F35 adenoviral vectors were generated on an E1/E3 deleted backbone . The ZFNs targeting either the cxcr4 or ccr5 genes were linked via a 2A peptide sequence and cloned into the pAdEasy-1/F35 vector under control of the CMV TetO promoter , and the Ad5/F35 virus for each construct was generated using TREx 293T cells as described [33] . The Ad5/F35 vector encoding the X4-ZFNs is identical to that use by Nilsson , et al . [33] except for the ZFN inserts , promoter , polyA and linker sequences . Genomic DNA was extracted with the MasterPure kit ( Epicentre Biotechnologies ) according to manufacturer's instructions . Frequency of gene modification by NHEJ was evaluated as described previously [23] , [25] , [28] . Briefly , the purified genomic DNA was used as a template to amplify a fragment of the cxcr4 gene using the specific primers ( human CXCR4: 5′-CAACCTCTACAGCAGTGTCCTCATC -3′and 5′- GGAGTGTGACAGCTTGGAGATG -3′; rhesus CXCR4: 5′- GGTGGTCTATGTTGGAGTCTGG -3′and 5′- GGAGTGTGACAGCTTGGAGATG -3′ ) in the presence of a 32P-dATP and dCTP . The PCR products were then heated , allowed to re-anneal followed by treatment with the mismatch-sensitive Surveyor nuclease as described in order to detect insertions and deletions caused by NHEJ . For humanized mice samples , whole genome amplification using the REPLI-g Mini Kit ( Qiagen ) was conducted prior to the surveyor nuclease assay due to limiting cell numbers . Fresh CD4+ T cells from normal human donors , purified by negative selection , were obtained from the Center for AIDS Research Human Immunology Core at the University of Pennsylvania . 2 . 5 million CD4+ T cells were seeded at a density of 0 . 8×106 cells/ml in RPMI containing 10% fetal calf serum , 1% penicillin/streptomycin , and 100 U/ml interleukin-2 ( IL-2 ) . The cells were stimulated with anti-CD3/anti-CD28 coated magnetic beads at a 3∶1 bead to cell ratio [34] . Approximately 18 hrs post-stimulation , the cells were transduced with an Ad5/F35 vector encoding either the X4-ZFNs or R5-ZFNs at a multiplicity of infection ( MOI ) of 600 . Beginning 72 hours post-stimulation , cells were counted every 48 hours using trypan blue dye exclusion on an automated hemocytometer ( Countess , Invitrogen ) and split to 0 . 8×106 with fresh media containing 100 U/ml IL-2 . Five days post-stimulation , the magnetic beads were removed and washed twice in fresh media . Cells were counted and split until cell growth plateaued 10–14 days post stimulation . For longer experiments , cells were restimulated with beads and cultured for an additional 10–14 days . Five days post-stimulation the anti-CD3/anti-CD28 coated magnetic beads were removed from each of the three cultures ( non-transduced ( NTD ) , AdX4-ZFNs , and AdR5-ZFNs ) and 2 . 5 million cells were seeded in each of four cultures that were subsequently infected with either Bk132 ( primary X4 isolate ) , HxB2 ( lab-adapted X4 isolate ) , R3A ( R5X4 primary isolate ) , or media only ( mock ) . 100 ng p24 of HIV-1 was used per million cells . All staining was done at room temperature in FACS Wash Buffer ( 1 mM EDTA , 2 . 5% fetal calf serum in PBS ) and all antibodies were from BD Biosciences unless otherwise noted . 0 . 5–1 . 0×106 cells were washed in PBS and stained with Live/Dead Aqua ( Invitrogen ) for 10 min . Then , anti-CD4 PE Cy5 . 5 and anti-CXCR4 APC ( clone 12G5 ) were added and cells were stained for 20–30 minutes . Cells were then washed and permeabilized per manufacturer's protocol using Cytofix/cytoperm ( BD ) and stained intracellularly for HIV gag with KC57-RD1 ( Beckman Coulter ) . For compensation , ArC beads ( Invitrogen ) were used for live/dead , and CompBeads ( BD ) were used for all other fluorochromes . To detect wtCXCR4 and CXCR4Δ18 in 293T transient transfection experiments , anti-CXCR4 APC ( clone 12G5 ) and anti-CXCR4 PE ( clone 4G10 ) ( Santa Cruz Biotechnologies ) were used . All samples were run on an LSRII ( BD ) and analyzed using FlowJo 8 . 8 . 6 ( Treestar Inc ) . Events were gated as follows: singlets ( FSC-A by FSC-H ) , live cells ( SSC-A by Live/Dead ) , lymphocytes ( FSC-A by SSC-A ) , CD3+CD4+ ( CD3 by CD4 ) , and then events were divided into CXCR4+ and CXCR4- populations based upon a fluorescence minus one ( FMO ) control . Genomic DNA was isolated from CD4+ T cells using the QIAamp DNA Micro Kit ( Qiagen ) . For each condition , 200 ng genomic DNA was then PCR amplified using Platinum Taq High Fidelity ( Invitrogen ) using the following primers plus 454 adaptor sequences and 8 letter DNA barcodes: CAACCTCTACAGCAGTGTCCTCATC ( forward ) and GGAGTGTGACAGCTTGGAGATG ( reverse ) . Cycle conditions were 95° for 5 min , then 30 cycles of 95° for 30 sec , 55° for 3 sec , 68° for 30 sec , followed by 68° for 2 min . Following PCR amplification the PCR product was analyzed on a 2% agarose gel and then extracted and gel purified using Wizard SV Gel and PCR Clean-Up System ( Promega ) . Quant-iT dsDNA High-Sensitivity Assay Kit ( Invitrogen ) was then used to determine the concentration of each bar-coded amplicon . DNA samples were then pooled at an equimolar ratio and run on a Roche/454 GS FLX using standard chemistries at the University of Pennsylvania's DNA Sequencing Facility . Approximately 30 , 000–100 , 000 reads were obtained for each experiment . CXCR4 pyrosequencing data were assigned to samples by DNA barcode . Any reads containing ambiguous base calls or without a perfect match to barcode and primer were discarded . All remaining reads were aligned to the CXCR4 reference sequence using Mosaik ( http://bioinformatics . bc . edu/marthlab/Mosaik ) . All deviations from the CXCR4 consensus sequence 40 base pairs up or downstream from the ZFN binding site were determined . Any reads that did not extend across this region or that failed to align were discarded . Reads containing only two or fewer substitutions were not classified as mutations as these likely represent sequencing artifacts . Next , background pyrosequencing error , identified by an untransduced control sample , was subtracted from each group of reads . For frameshift analysis , the sequencing error was determined and subtracted for each individual insertion or deletion size . To ensure sufficient sampling of diverse amplicons , at least 200 ng gDNA was used for CXCR4 analysis and at least 400 ng gDNA was used for off-target site amplification , representing the genomic DNA content of approximately 70 , 000 and 140 , 000 alleles , respectively . Determining genetic disruption frequency by both the Cel1 and 454 assays require the assumption that wild type and disrupted alleles are not differentially amplified . To empirically determine the DNA binding preference of the X4-ZFNs , we employed SELEX as previously described [23] . Briefly , each ZFP was HA-tagged and incubated with randomized DNA oligonucleotides and anti-HA Fab fragments . Any DNA bound to the ZFPs was then isolated and amplified . The newly amplified DNA was then used to repeat this process for a total of four rounds of enrichment . The DNA pool was then sequenced at approximately 50× coverage to generate a positional-weighted matrix . This matrix was then aligned to the human genome with the following criteria: putative off-target sites could have up to six mismatches compared to the SELEX consensus sequence , the ZFP pairs must be separated by either 5 or 6 bps , and both ZFP homo- and heterodimers were considered . Off-target sites were ranked and scored by multiplying the probability of each nucleotide at each of the 12 positions of the positional-weighted matrix . The highest scores were then deemed most likely to be disrupted . 454 off-target site data was analyzed as discussed previously [23] . NSG ( NOD . Cg-PrkdcscidIl2rgtm1Wjl/Szj ) mice , 8–9 weeks old at time of initial injection , were derived from breeders purchased from The Jackson Laboratory ( Bar Harbor , ME ) . Animals were maintained in a defined flora animal barrier facility at the University of Pennsylvania's Stem Cell and Xenograft Core . Human CD4+ T cells were isolated and stimulated as previously described and then transduced with an Ad5/F35 vector expressing either the R5-ZFNs or the X4-ZFNs at an MOI of 600 . Cells were maintained as previously described . Ten days post stimulation 107 modified cells resuspended in 100 µL PBS were injected intravenously into the tail vein of each mouse . 23 animals received cells treated with X4-ZFNs and 22 mice received cells treated with R5-ZFNs . Animals were randomized by age , sex , and cage . Mice were maintained on the antibiotic Baytril ( Bayer ) for 24 hours post-injection . To infect the mice with HIV-1 , 105 autologous CD4+ T cells previously infected with X4 HIV-1 strain Bk132 were injected into the tail vein of each mouse . Autologous cells used to infect mice that were not transduced were obtained and stimulated simultaneously as the initially engrafted cells . Five days post-stimulation cells were infected with 100 ng p24/million cells and then were cryopreserved four days post-infection . Cell engraftment was assessed 27 days post injection , and mice were infected with HIV-1 the following day . To obtain whole blood , mice were anesthetized with isoflurane and a capillary tube was used to drain the retroorbital vein . Human CD4+ T cell counts were determined by staining 50 µl of whole blood in Trucount tubes ( BD ) with anti-CD45 FITC ( Biolegend ) , anti-CD3 Qdot 655 ( Invitrogen ) , anti-CD4 Alexa Fluor 700 , anti-CD8 Pacific Blue ( Biolegend ) , and anti-CXCR4 PE-Cy5 . Human CD4+ T cells were defined as CD45+CD3+CD4+CD8- . At the time of sacrifice , a cardiac puncture was performed to obtain maximal blood volume and then the spleen was harvested . Spleens were homogenized and erythrocytes were lysed with ACK lysis buffer ( Invitrogen ) before cell purification . Human CD4+ T cells were then isolated with the Human CD4 Positive Selection Kit using the Robosep robotic cell separator ( Stem Cell Technologies ) . Whole blood from rhesus macaques ( Macaca mulatta ) housed at the Tulane National Primate Research Center was used for CD4+ T cell isolation and ZFN treatment . Peripheral blood mononuclear cells were isolated by centrifugation with 96% Ficoll ( BD ) , followed by erythrocyte lysis with ACK lysis buffer . CD4+ T cells were then isolated by negative selection with a non-human primate CD4+ T cell selection kit ( Miltenyi ) . Cells were then stimulated with 1∶4 anti-CD3 ( clone FN-18 ) /anti-CD28 ( clone L293 ) M-450 tosylactivated beads ( Invitrogen ) at a ratio of 1 bead per cell [35] , [36] . Approximately 18 hours post-transduction , cells were transduced with an Ad5/F35 vector expressing either the X4-ZFNs or rhesus specific R5-ZFNs . Cells were maintained in culture as human CD4+ T cells . Surveyor nuclease assay was performed six-ten days post transduction to assess disruption efficiency . Human CD4+ T cells were obtained after written informed consent and approval by the University of Pennsylvania's institutional review board . All humanized mouse experiments were approved by the University of Pennsylvania's Institutional Animal Care and Use Committee ( Protocol 802436 ) , and were carried out in accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All rhesus macaque experiments were approved by the Tulane Institutional Animal Care and Use Committee approval ( Protocol P0085; Project 3520 ) The Tulane National Primate Research Center ( TNPRC ) is an Association for Assessment and Accreditation of Laboratory Animal Care accredited facility ( AAALAC #000594 ) . The NIH Office of Laboratory Animal Welfare assurance number for the TNPRC is A3071-01 . All clinical procedures , including administration of anesthesia and analgesics , are carried out under the direction of a veterinarian . Blood was collected while the animals were anesthetized with Tiletamine-zolazepam with Burprenorphine given as an analgesic . All possible measures are taken to minimize discomfort of all the animals used in this study . The University of Pennsylvania and Tulane comply with NIH policy on animal welfare , the Animal Welfare Act , and all other applicable federal , state and local laws .
To genetically disrupt the CXCR4 allele , we designed a pair of zinc-finger proteins ( ZFPs ) targeting the region of the cxcr4 gene that encodes residues Asp 187 to Val 196 in the second extracellular loop ( ECL2 ) of this seven-transmembrane domain receptor using methods previously described [29]–[32] ( Figure 1 ) . The ECL2 was chosen because this region is less well conserved amongst the CXC family of chemokine receptors , which should reduce the frequency with which other CXC receptors might be targeted , and because ECL2 is important in supporting interactions with the HIV-1 Env protein [37] , [38] . Two ZFPs were designed to bind each of two 12 bp targets separated by 6 bp in this region of CXCR4 . Each ZFP was then fused to a modified FokI cleavage domain , active preferentially as a dimer to reduce nonspecific DNA cleavage , resulting in zinc-finger nucleases ( ZFNs ) [25] . Upon binding of both X4-ZFNs , the FokI nuclease cleavage domains dimerize and then generate a double strand break that can subsequently be repaired by error-prone NHEJ resulting in mutations targeted to the cleavage site that can include missense mutations , deletions and insertions ( Figure 1 ) . To determine the efficiency and specificity with which the cxcr4 genes could be disrupted in human T cells , we produced a bicistronic Ad5/F35 vector to deliver the X4-ZFNs ( AdX4-ZFNs ) . The Ad5/F35 vector is a serotype 5 virus with the fiber protein from a serotype 35 adenovirus that utilizes CD46 for entry as opposed to the coxsackie and adenovirus receptor ( CAR ) , which is poorly expressed on human CD4+ T cells [39] . Primary human CD4+ T cells were stimulated with anti-CD3/anti-CD28 coated magnetic beads and transduced 18 hours later with AdX4-ZFNs , AdR5-ZFNs which expresses previously described CCR5-specific ZFNs [23] , or an Ad5/F35 vector that expresses green fluorescent protein ( AdGFP ) . To identify optimal disruption conditions , multiplicities of infection ranging from 100 to 1000 were employed . Cell growth was monitored every 48 hours post-stimulation for approximately two weeks and the efficiency of CXCR4 disruption was assessed at day five post-transduction by both the Surveyor nuclease assay and by deep-sequencing of the CXCR4 target site . As shown in Figure 2A , the Ad5/F35 vectors had a slight dose-dependent impact on cell growth at higher multiplicities of infection that was similar with the AdX4-ZFNs and AdGFP vectors . Cxcr4 allelic disruption efficiencies as determined by either deep sequencing or the Surveyor nuclease assay were comparable , and were approximately 10% at an MOI of 100 , 20% at an MOI of 300 , 34% at an MOI of 600 , and 38% at an MOI of 1000 ( Figure 2B ) . For subsequent experiments we used an MOI of 600 as this provided near-maximal disruption efficiency with limited impact on cell growth . Notably , this is also the MOI being used in an adoptive therapy phase I clinical trial with R5-ZFNs . Importantly , the level of cxcr4 disruption in cells from multiple donors was stable over nearly four weeks in culture ( Table S1 ) , indicating that CXCR4-disrupted cells continued to grow normally . Cell proliferation remained dependent on stimulation , and transformation has not been observed after treatment with ZFNs ( data not shown ) . Deep sequencing of the ZFNs target site 10 days after transduction made it possible to assess the mutations introduced by NHEJ reactions following cleavage with X4-ZFNs . Of the nearly 50 , 000 modified cxcr4 alleles analyzed across five independent experiments , 81 . 1% ( range 75 . 3–81 . 7% ) contained pure deletions from 1–64 bp in size with the most common deletions being 2 , 9 , 12 , 15 , 18 , and 25 bp , while 13 . 5% ( range 12 . 8–16 . 9% ) of cxcr4 alleles contained pure insertions ranging from 1 to 69 bp with more than 90% being 7 bp or less ( Figure 1B ) . The remaining 5 . 3% ( range 4 . 3–7 . 4% ) of disruption events contained multiple insertions and deletions that may be due to more extensive DNA end-processing or multiple cycles of ZFN-mediated cleavage and subsequent NHEJ . Surprisingly , frameshift mutations occurred at a ratio of 0 . 90 in-frame per out-of-frame mutation as opposed to the expected frequency of 0 . 50 ( 1 in-frame per 2 out-of-frame mutations; Table S1 ) . This unexpected bias likely resulted from microhomology-mediated joining that produced in-frame deletions . To our knowledge , preferential in-frame repair has not been reported or seen with other ZFNs [23] , [40] , [41] . To further characterize the consequences of disruption mediated by X4-ZFNs , we analyzed an unusually common lesion , an in-frame 18 bp deletion ( CXCR4Δ18 ) that results in the deletion of DNA encoding amino acids R188 to D193 ( Figure 1B ) . This deletion comprised 11 . 2% ( range 9 . 8 and 11 . 9% ) of all cxcr4 disruptions across five independent experiments with cells from five different donors . The resulting CXCR4Δ18 protein , containing a six-residue deletion in ECL2 , could potentially be expressed at the cell surface and support HIV infection . To examine this , we transiently expressed CXCR4Δ18 or wt CXCR4 as a control in 293T cells , which have low endogenous CXCR4 expression . CXCR4 cell surface and intracellular expression was detected by flow cytometry after co-staining with the N-terminal specific CXCR4 antibody 4G10 and the extracellular loop ( ECL ) specific antibody 12G5 whose epitope includes the CXCR4Δ18 deleted residues [42] . As expected , CXCR4 could be detected on the surface of control cells by both the N-terminal and ECL antibodies . However , CXCR4Δ18 was not detected at the cell surface , though it was detected intracellularly by the N-terminal antibody ( Figure 3 ) . In addition , cells expressing CXCR4Δ18 along with CD4 did not support HIV-1 infection . These findings indicate that CXCR4Δ18 , the most common in-frame deletion resulting from the X4-ZFNs , does not readily traffic to the cell surface and does not function as an HIV-1 coreceptor . Potential off-target genome modification comprises the predominant safety concern with ZFNs . Although ultra-deep full genome sequencing could best identify off-target effects , it is impractical and cost-prohibitive with current technology . Instead , we took a more targeted approach that used an experimentally derived binding site for each X4-ZFP to guide the identification of potential off-target cleavage sites . We conducted in vitro selection , or SELEX ( systemic evolution of ligands by exponential enrichment ) to determine the actual binding site preference of each X4-ZFP ( Figure S1 ) [43] , [44] . A positional-weighted matrix was then generated of the 12 bp binding site and 1 bp flanking region for each ZFP . A BLAST search against the human genome was then used to determine the top 15 off-target binding sites by allowing up to six mismatches per ZFP binding site , a 5 or 6 bp gap between ZFPs , and formation of hetero or homodimers ( Table S2 ) [23] . To assess low frequency disruption events , we conducted 454 deep sequencing on all 15 sites in both control CD4+ T cells and those treated with X4-ZFNs , yielding approximately 7 , 500–26 , 000 reads per site in the ZFN-treated samples ( Table S2 ) . In a sample with 26 . 9% of CXCR4 alleles disrupted , NHEJ events were detected at a frequency of 2 . 3% ( 170/7531 reads ) in an extragenic region on chromosome 12 and 0 . 8% ( 84/10531 ) in ADAMTS17 , a metalloprotease of unknown function [45] . The four mutations out of 20 , 312 reads found in DEC1 ( a putative tumor suppressor [46] ) and the single mutation out of 21 , 139 reads found in an extragenic region of chromosome 11 could be due to PCR and sequencing errors or to very low levels ( <0 . 02% ) of ZFN-mediated cleavage events . Overall , the X4-ZFNs are highly specific for cxcr4 with low frequency disruption clearly seen at 2 of 15 putative off-target sites with the highest homology to the intended target . Disruption of both cxcr4 alleles should render human CD4+ T cells resistant to X4- and perhaps some R5X4- viruses as well , while cells harboring a single disrupted allele might express lower levels of CXCR4 and so be more resistant to virus entry . To determine whether ZFN-mediated disruption of cxcr4 indeed protects CD4+ T cells from an in vitro HIV challenge , human CD4+ T cells from three different ccr5 wild type donors were stimulated and transduced with AdX4-ZFNs or an AdR5-ZFNs control . Four days post-transduction , the cells were infected with three diverse HIV-1 strains: BK132 ( primary X4 HIV ) , HxB2 ( lab-adapted X4 HIV ) , or R3A ( primary R5X4 HIV ) . Approximately two weeks post-transduction the cells were restimulated with anti-CD3/anti-CD28 beads , and cultures were maintained for an additional two weeks . In the absence of HIV infection , there was no detectable growth difference between the X4-ZFNs treated , R5-ZFNs treated , and non-transduced controls over the course of the experiment . However , upon infection with the X4- or R5X4- HIV-1 strains , X4-ZFNs treated cells maintained exponential growth compared to profound cell death seen in the R5-ZFNs and untransduced controls . Despite the ability of R3A to utilize both CCR5 and CXCR4 to infect cell lines , in human CD4+ T cells stimulated with anti-CD3/anti-CD28 coated magnetic beads , CCR5 is downregulated causing transient resistance to R5 HIV [47] . Thus , R5X4 HIV strains are likely to function predominantly as X4 HIV strains under these conditions [47] . The growth advantage conferred by treatment with X4-ZFNs in the presence of HIV was magnified upon restimulation . ( Figure 4A ) . This likely resulted from increased cell activation , which increases the ability of HIV to infect and replicate in CXCR4 positive cells . To determine whether the growth advantage conferred by X4-ZFNs treatment in the presence of X4- and R5X4- HIV resulted from a survival advantage of CXCR4 disrupted cells , we performed flow cytometry at various time points post infection as well as deep sequencing of the X4-ZFNs target site on HIV-infected and uninfected cultures . In the absence of HIV infection , the cxcr4 disruption frequency remained stable over time in four independent experiments testing four different ccr5 wild type donors as measured by deep sequencing . A representative experiment is shown in Figure 4B and CXCR4 disruption data from all experiments is shown in Tables S1 and S3 . While CXCR4 gene disruption remained stable over time at approximately 30% , CXCR4 gene disruption in HIV-infected cultures increased to 87% , 91% , and 88% in the presence of BK132 , HxB2 , and R3A respectively after 21 days of infection . FACS analysis showed that at day 19 post-HIV challenge , the frequency of CXCR4 negative cells amongst all live mock HIV-infected CD4+ lymphocytes was 13 . 0% in untransduced cells , 14 . 1% in cells transduced with R5-ZFNs , and 35 . 0% in cells transduced with X4-ZFNs compared to greater than 98% , 97% , and 99% of Bk132 , HxB2 , and R3A infected cultures transduced with the X4-ZFNs , ( Figure 4C ) . We also found that after 19 days post-HIV infection , reduced but significant cell growth was detectable in several of the HIV-infected control cultures , untransduced and treated with R5-ZFNs . However , greater than 95% of these cells , compared to approximately 10% of cells treated with X4-ZFNs , were CD3+CD4- suggesting that the surviving cell population was protected from HIV infection by down-regulating CD4 ( Figure S2 ) . Thus , CXCR4 disruption had no impact on cell viability , but conferred a significant survival advantage in the presence of HIV strains that can use CXCR4 to infect cells . Furthermore , in control cultures that were untransduced or treated with R5-ZFNs , viral titers exponentially increased until extensive cell death began approximately 8–10 days post infection . In contrast , in cultures treated with X4-ZFNs viral titers steadily decreased after peak viremia while cell growth remained exponential suggesting there was not significant viral production ( data not shown ) . Given the ongoing adoptive therapy trial of CD4+ T cells treated with R5-ZFNs and the anti-viral success of the recent ccr5Δ32 bone marrow transplant in an HIV-infected patient [6] , we sought to determine if cxcr4 could be genetically disrupted simultaneously with ccr5 . Human CD4+ T cells from a ccr5Δ32 homozygote were transduced with AdX4-ZFNs or AdR5-ZFNs and subsequently infected with HIV-1 strains Bk132 , HxB2 , and R3A as described above . Representative data from one of two independent experiments conducted in cells from the same donor is shown in Figure 5 and data from both experiments is shown in Tables S1 and S3 . As seen in ccr5 wild type CD4+ T cells , exponential cell growth was preserved in cultures treated with X4-ZFNs compared to control cultures that were untransduced or treated with R5-ZFNs ( Figure 5A ) . In addition , disruption frequency in cultures treated with X4-ZFNs as determined by deep sequencing remained remarkably stable between 32–33% from day 5 to day 26 post-transduction in the absence of HIV , which suggests that simultaneous disruption of ccr5 and cxcr4 does not adversely affect cell growth . However , in the presence of Bk132 , HxB2 , and R3A , cxcr4 disruption increased after 21 days of HIV challenge to 89% , 83% , and 90% , respectively ( Figure 5B ) , and was associated with markedly diminished virus replication ( data not shown ) , again consistent with significant protection conferred by cxcr4 disruption . Thus , treatment with X4-ZFNs of both wild-type and ccr5Δ32 CD4+ T cells confers stable cxcr4 disruption and a marked survival advantage in the presence of R5X4-HIV and X4-HIV in vitro without any detectable effect on cell growth or viability in the absence of HIV . This suggests that both ccr5 and cxcr4 can be genetically targeted simultaneously for the treatment of HIV infection , while preserving the replicative capacity of the CD4+ T cells . As a first step in evaluating the safety and efficacy of the X4-ZFNs in vivo , we employed a NSG humanized mouse model . Briefly , human CD4+ T cells were stimulated with anti-CD3/anti-CD28 beads and transduced with either AdX4-ZFNs or an AdR5-ZFNs control at an MOI of 600 . Cells were then expanded in vitro for ten days after which 107 CD4+ T cells treated with X4-ZFNs ( n = 23 ) or R5-ZFNs ( n = 22 ) were injected intravenously into each mouse . Engraftment was assessed by peripheral blood CD4+ T cell counts 27 days post-injection . All 45 animals successfully engrafted; however , one animal that received cells treated with the X4-ZFNs had a significantly higher but stable CD4+ T cell count and was thus excluded as an outlier from the remainder of the study . On day 28 post-engraftment , mice were intravenously injected with 105 autologous CD4+ T cells that were previously infected with the highly cytopathic X4 HIV-1 strain Bk132 or a mock control . CD4 counts , viral load , and CXCR4 disruption were then monitored to determine the effect of treatment with X4-ZFNs . To determine if X4-ZFNs impacted cell growth or viability in the absence of HIV , we first compared CD4 counts over time between the uninfected X4-ZFN and R5-ZFN control mice . There was no significant difference in CD4 counts between the two groups over the course of the 61 day experiment as determined by a generalized estimating equation ( GEE ) method ( p = . 88 ) ( Figure 6A ) . Next , we examined the frequency of CXCR4 DNA disruption over time with the surveyor nuclease assay . At the time of injection the percentage of cxcr4 alleles disrupted was 24 . 3% . This remained constant in both the blood ( p = . 32 ) and spleen ( p = . 70 ) over the course of the experiment suggesting that CXCR4 disruption did not significantly impact trafficking between these two compartments ( Figure 6B ) . Next , we characterized CXCR4 cell surface expression over time by FACS . In the R5-ZFN control group , with intact cxcr4 genes , 88% of CD4+ T cells expressed CXCR4 protein at day 27 post engraftment , compared to 84% of cells in the X4-ZFN mice ( ∼24% cxcr4 gene disruption ) as determined by a fluorescence minus-one ( FMO ) control . This difference persisted over time in the absence of HIV-1 infection ( p <0 . 001 ) ( data not shown ) . Together the stable disruption of CXCR4 as determined by both the surveyor nuclease assay and flow cytometry suggests that CXCR4 disruption did not negatively impact cell viability or growth in humanized NSG mice over a two-month period . As expected , xenogeneic graft versus host disease ( GVHD ) , assessed clinically by dermatitis and hair loss , was observed in mice receiving cells treated with both R5-ZFNs and X4-ZFNs in the absence of HIV challenge . The development of GVHD was equivalent between the two groups ( data not shown ) , suggesting that treatment with X4-ZFNs did not affect CD4+ T cell effector functionality . In response to X4 HIV challenge with HIV-1 Bk132 , CD4 counts decreased in both X4-ZFN and R5-ZFN mice . However , this rate of decline was slower in the X4-ZFN mice . The X4-ZFN group exhibited a mean 1 . 1 log CD4 count protection by day 14 post infection ( p = . 05 for a parametric t-test ) . However , this protective effect waned over time and there was no significant difference in CD4 counts by day 33 post infection ( p = . 88 ) suggesting that treatment with X4-ZFNs conferred only transient protection ( Figure 6A ) . One mechanism that could account for this would be if mutations arose in the viral Env protein to enable it to use CCR5 . To explore this possibility , we bulk cloned and sequenced the V3 loop of Env , the main determinant of coreceptor tropism [48] , from plasma isolated from three R5-ZFN mice and three X4-ZFN mice at the time of sacrifice . We identified a single amino acid substitution ( Y302N ) present in Env isolated from X4-ZFN mice but not R5-ZFN mice or the viral innoculum . Next , we cloned six distinct , functional Envs from the X4-ZFN mice and three distinct , functional Envs from the viral innoculum . As full length Bk132 Env would not pseudotype on an NL43 HIV core we truncated the cytoplasmic tail of the Envs [49] , [50] , and conducted tropism testing on NP2 cell lines expressing CD4 with either CCR5 or CXCR4 . Of the six functional Envs from X4-ZFN mice , four contained the Y302N mutation . Interestingly , these four Envs were able to utilize CCR5 and CXCR4 equivalently , similar to the R5X4-tropic control R3A . All clones with the wild type Tyr302 , including the Envs from the viral innoculum and two Envs from X4-ZFN mice utilized CXCR4 approximately 1000-fold more efficiently than CCR5 and comparably to the X4-tropic control TYBE ( Figure 6C ) . Thus , in an NSG humanized mouse model of HIV infection , the cells treated with X4-ZFNs engrafted , trafficked , and persisted comparably to control cells . In addition , treatment with X4-ZFNs resulted in significant transient protection of CD4+ T cell counts in response to X4-tropic HIV challenge , and HIV challenge provided cxcr4 disrupted cells with a survival advantage as determined by increase of cxcr4 disruption in the presence but not the absence of HIV . However , the extent of the protection conferred by the X4-ZFNs was mitigated by evolution or outgrowth of preexisting R5X4-tropic HIV . While humanized mouse models for HIV infection have utility , the model is limited due to incomplete immune reconstitution , development of xenogeneic graft versus host disease ( GVHD ) , and the absence of normal T cell homeostasis . For these reasons and others , the NSG model is suboptimal compared to non-human primate models to further elucidate the safety and efficacy of treatment with X4-ZFNs and R5-ZFNs . As a proof of concept for future clinical adoptive therapy studies , we attempted to disrupt the ccr5 and cxcr4 genes with ZFNs in rhesus macaque CD4+ T cells . Briefly , rhesus CD4+ T cells were isolated from whole blood , purified by magnetic bead negative selection , and then stimulated with anti-CD3/anti-CD28 coated beads as previously described [35] , [36] . As the 24 bp X4-ZFPs' binding site is identical between rhesus and humans , we were able to utilize the same ZFN pair . However , in order to target rhesus CCR5 , rhesus specific R5-ZFNs were developed . As for human cells , the ZFNs were delivered with an Ad5/F35 vector and disruption was assessed by the surveyor nuclease assay . Utilizing a range of MOIs of 600 , 1000 , and 2000 we observed mean ccr5 and cxcr4 disruption levels of 19 . 6% and 14 . 0% , respectively ( Figure 7 ) , which suggests that adoptive therapy of cells modified with ZFNs is feasible to model in rhesus macaques .
The apparent eradication of HIV resulting from a ccr5Δ32 homozygous allogeneic bone marrow transplant into an HIV-infected patient represents the first reported “cure” of HIV [6] . While an important proof-of-principle , few individuals could benefit from allogeneic ccr5Δ32 homozygous transplants due to toxicities of allogeneic rejection and limitations of finding sufficient HLA-matched ccr5Δ32 homozygous donors . However , coreceptor-specific ZFNs represent a novel therapeutic approach to recapitulate this success via autologous transplantation of gene-modified hematopoietic stem cells and mature CD4+ T cells . Ccr5 can be efficiently disrupted in both human CD4+ T cells and hematopoietic stem cells , conferring protection to HIV challenge in vitro and in humanized mice [23] , [28] . In addition , transgenic autologous hematopoietic stem cells can be successfully transplanted in HIV-infected individuals [18] and several phase I adoptive transfer trials of CD4+ T cells treated with R5-ZFNs in HIV infected individuals are currently underway . By design , this strategy addresses only viruses that require CCR5 to infect cells . Our long-term goal , therefore , is to explore the potential to genetically disrupt both ccr5 and cxcr4 for cell replacement therapies in HIV infected individuals , and in the case of cxcr4 do so in a way that specifically targets CXCR4 on T cells and not the many other cell types on which it is expressed . Unlike for ccr5 , there are no known humans with loss of function cxcr4 mutations that would provide insight into the safety and viability of cxcr4 disruption in mature CD4+ T cells . A concern associated with targeting CXCR4 is that it is broadly expressed , while CCR5 expression is largely limited to hematopoietic cells . CXCR4 , along with its natural ligand CXCL12 , plays a critical role in normal B cell , cardiovascular , and cerebellar development , though T lymphocytes appear to develop normally in cxcr4−/− mice [51] . Thus , it is possible that the selective disruption of cxcr4 in mature post-thymic CD4+ T cells may be tolerable . In addition to its role in development , the CXCR4-CXCL12 axis is a potent CD4+ T cell chemoattractant , and the broad expression of both proteins suggests that this axis may play a fundamental role in basal chemotaxis as opposed to a response to inflammation [52] . Indeed , inhibiting CXCR4 function systemically with the small molecule antagonist plerixafor results in the peripheral mobilization of hematopoetic stem cells , thus mitigating the potential of such therapy for long-term anti-retroviral therapy . However , plerixafor , which has not been reported to have adverse immunologic consequences resulting from inhibiting CXCR4 function in mature CD4+ T cells , provides proof of principle that inhibiting CXCR4 in mature CD4+ T cells may prove to be safe and viable [10] , [53] . This suggests that this essential gene can be targeted in a cell-type specific manner with CXCR4-specific ZFNs that limits the toxicities of systemic disruption . While we have demonstrated that CXCR4 is not essential for CD4+ T cell viability and function in vitro and in humanized mice in vivo , the redundancy of lymphocyte chemokine receptors and their ligands makes predicting the in vivo consequences of cxcr4 disruption in a normal host on CD4+ T cell function and trafficking difficult . We conclude that a logical next step will be to study the consequences of cxcr4 disruption in a non-human primate model of HIV infection , which will simultaneously permit the assessment of the consequences of this approach on T cell function and trafficking . A significant advantage of ZFN gene modification , compared to retrovirus based approaches , is that only transient transgene expression is required to permanently engineer an HIV resistant cell . As a result , adenovirus or other delivery mechanisms such as RNA transfection can be employed that avoid toxicities that can be associated with retroviral integration , such as cellular expansion or transformation . This “hit-and-run” approach limits the requirement of chronic transgene expression and the potential leakiness of other approaches including siRNA [21] , [22] , intrabodies [19] , and ribozymes [17] . However , like most gene transfer approaches a major concern with ZFN technology is the potential for oncogenesis due to off-target effects . While additional study is clearly needed , our current studies have clearly identified off-target disruption in two of the top 15 putative off-target sites: an extragenic site on chromosome 12 and in the metalloprotease ADAMTS17 , which is not expressed in CD4+ T cells . In addition , mature CD4+ T cells appear to be resistant to malignant transformation [54] , thus mitigating the potential concerns of off-target disruption . Consistent with this , more than 200 people have safely undergone adoptive transfer of genetically engineered lymphocytes with no reported cases of therapy-induced oncogenesis [55] . Reasons for resistance to transformation of mature lymphocytes are unclear , but may involve an unknown mechanism that ensures the diversity of the TCR repertoire and thus limits clonal outgrowth [54] . In contrast , the safety record of hematopoietic stem cell gene therapy is less clear , with a significant frequency of gene-therapy induced oncogenesis or clonal outgrowth reported in several hematopoietic stem cell trials [56] , [57] . One unexpected finding reported here is the predominance of in-frame mutations , particularly in-frame deletions , resulting from ZFN mediated cleavage of cxcr4 . This has not been observed in other ZFN studies reported thus far . The deep-sequencing approach we have taken makes it possible to comprehensively and accurately assess the types and frequencies of mutations that result from ZFN cleavage followed by DNA repair . The striking preponderance of in-frame deletions may have resulted from toxicities of frameshift mutations shortly after treatment with X4-ZFNs leading to decreased survival relative to in-frame mutants . However , this is unlikely given that the frequency of in-frame mutations remained stable over nearly four weeks in culture , that there was no significant increase in cell death between control cultures and those treated with X4-ZFNs , and that the most common in-frame mutant was not expressed on the cell surface and thus cannot maintain functionality . Rather , the preference for in-frame deletions is likely due to preferential in-frame DNA repair . The deletion in the most common X4-ZFN-induced lesion , cxcr4Δ18 , is flanked by a GTCA microhomology domain at the 5′ and 3′ ends consistent with a repair mechanism of microhomology-mediated NHEJ [58] . Similar microhomology sites are present in other common ZFN-induced cxcr4 mutants that we identified . Thus , it appears that the nucleotide sequence of the X4-ZFN binding site directs a preference for an in-frame repair mechanism . Our studies provide a fundamental demonstration that inactivation of cxcr4 by treatment with X4-ZFNs rendered human CD4+ T cells resistant to infection by X4 virus strains , while CXCR4 inactivation in the context of a ccr5Δ32 homozygous background rendered cells resistant to infection by both R5 and R5X4 strains . Genetic ablation of both CCR5 and CXCR4 will likely make CD4+ T cells entirely resistant to HIV-1 . Dual-disruption of CCR5 and CXCR4 will be needed for maximal therapeutic benefit since 46% of treatment-experienced individuals harbor R5X4 strains of HIV compared to 4% with only X4-HIV strains [59] . While virus strains have been identified that can infect cells in the absence of CD4 ( reviewed in [60] ) , none have been identified that can infect cells in the absence of a suitable coreceptor . In addition , virus strains that can use coreceptors other than CCR5 or CXCR4 to infect primary human cells are exceedingly rare . However , targeting CXCR4 alone could provide a selective advantage to CCR5-tropic virus strains . Suppression of CXCR4 by plerixafor in vitro can lead to the emergence of CCR5-tropic virus strains [61] , and highly active antiretroviral therapy can sometimes result in enhanced prevalence of R5 relative to R5/X4 virus strains in infected patients [62] . In the humanized mouse model under the conditions studied here , partial loss of cxcr4 in human T cells due to treatment with X4-ZFNs provided selective pressure for either the evolution or emergence of a pre-existing single amino acid mutation in the V3 loop of the infecting X4 HIV-1 strain that enabled it to use CCR5 as efficiently as CXCR4 . Thus , just as either genetic or therapeutic suppression of CCR5 can provide an advantage to virus strains that use CXCR4 , deletion of CXCR4 is expected to provide an advantage to CCR5-tropic viruses . However , this could provide a clinical benefit given the increased in vitro pathogenicity and correlation with progression to AIDS of X4-tropic HIV . While humanized mouse models provided a logical first approach to examine in vivo efficacy of CXCR4 disruption , this system does not make it possible to fully assess the functional impact of CXCR4 loss on CD4+ T cell function . To study this in the most rigorous way possible , we have explored the possibility of targeting CCR5 and CXCR4 in CD4+ T cells derived from rhesus macaques . Following re-design of the R5-ZFNs to account for sequence differences between the human and macaque alleles , we found that ZFNs could disrupt both alleles with reasonable efficiency in macaque CD4+ T cells . By inactivating CXCR4 singly and in combination with CCR5 , it will be possible to study the effects of CXCR4 loss on T cell function as well as virus infection in a more relevant animal model . | For HIV to enter T cells , the virus first binds to a primary surface receptor CD4 and then to a coreceptor , either CCR5 or CXCR4 . Previously we engineered zinc-finger nucleases ( ZFNs ) to specifically disrupt the CCR5 gene in primary human T cells , the predominant cell type infected and killed by HIV . This makes the cell permanently resistant to CCR5-tropic HIV; however , viruses that can utilize CXCR4 can still infect cells . ZFNs function as molecular scissors that cut a specific region of DNA . Then , the cell's own machinery repairs this cut , often introducing mutations that result in a non-functional protein . Currently , a clinical trial is underway in which HIV-infected individuals' own cells are removed from their blood , treated with the CCR5-ZFNs , and then infused back . Here , we report the use of novel zinc-finger nucleases that specifically and permanently disrupt the CXCR4 gene in T cells . This treatment results in resistance to CXCR4-tropic HIV . In addition , we combine CXCR4 and CCR5 genetic disruption to make cells resistant to all strains of HIV . Our long-term goal is to engineer HIV-resistant CD4+ T cells in infected individuals that can be reinfused and hopefully enable them to control infection in the absence of anti-viral drugs . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"biology"
] | 2011 | Engineering HIV-Resistant Human CD4+ T Cells with CXCR4-Specific Zinc-Finger Nucleases |
Network robustness is a crucial property of the plant immune signaling network because pathogens are under a strong selection pressure to perturb plant network components to dampen plant immune responses . Nevertheless , modulation of network robustness is an area of network biology that has rarely been explored . While two modes of plant immunity , Effector-Triggered Immunity ( ETI ) and Pattern-Triggered Immunity ( PTI ) , extensively share signaling machinery , the network output is much more robust against perturbations during ETI than PTI , suggesting modulation of network robustness . Here , we report a molecular mechanism underlying the modulation of the network robustness in Arabidopsis thaliana . The salicylic acid ( SA ) signaling sector regulates a major portion of the plant immune response and is important in immunity against biotrophic and hemibiotrophic pathogens . In Arabidopsis , SA signaling was required for the proper regulation of the vast majority of SA-responsive genes during PTI . However , during ETI , regulation of most SA-responsive genes , including the canonical SA marker gene PR1 , could be controlled by SA-independent mechanisms as well as by SA . The activation of the two immune-related MAPKs , MPK3 and MPK6 , persisted for several hours during ETI but less than one hour during PTI . Sustained MAPK activation was sufficient to confer SA-independent regulation of most SA-responsive genes . Furthermore , the MPK3 and SA signaling sectors were compensatory to each other for inhibition of bacterial growth as well as for PR1 expression during ETI . These results indicate that the duration of the MAPK activation is a critical determinant for modulation of robustness of the immune signaling network . Our findings with the plant immune signaling network imply that the robustness level of a biological network can be modulated by the activities of network components .
How network properties , such as robustness against network perturbations , emerge from biological networks has been a central question in systems biology [1] , [2] . Possible modulation of network robustness in a biologically relevant context and mechanisms underlying the modulation are areas of study that have rarely been explored . Innate immunity , in which defense responses are induced through signaling events initiated by recognition of pathogen attack , composes a major part of plant immunity [3] . PAMP/Pattern-Triggered Immunity ( PTI ) and Effector-Triggered Immunity ( ETI ) are modes of plant innate immunity defined by the way pathogens are detected [4] , [5] . PTI is triggered by recognition of microbe/pathogen-associated molecular patterns ( MAMPs/PAMPs ) by the cognate pattern-recognition receptors ( PRRs ) , which are typically receptor-like kinases or receptor-like proteins [6] . For example , Arabidopsis thaliana FLS2 is the PRR for flg22 , an elicitor-active epitope of flagellin from Gram-negative bacteria [7] . While most non-adapted pathogens cannot overcome PTI , adapted pathogens deliver effectors into the plant cell that manipulate plant cell functions to facilitate their infection by , for instance , interfering with PTI signaling [8] , [9] . ETI is triggered by specific recognition of effectors by resistance ( R ) proteins , which are often nucleotide-binding leucine-rich repeat ( NB-LRR ) proteins [10] . For example , the Arabidopsis intracellular NB-LRR R proteins RPS2 and RPM1 indirectly recognize perturbations of the PTI signaling component RIN4 by the effectors AvrRpt2 and AvrRpm1/AvrB , respectively , of a Gram-negative bacterial pathogen , Pseudomonas syringae [3] . In addition to proteinaceous effectors , some P . syringae strains deliver coronatine , which is a jasmonic isoleucine mimic , in order to suppress plant immunity [11] . Recently , it was shown that coronatine suppresses immune responses dependent on salicylic acid ( SA ) as well as independent of SA [12] , [13] . Thus , there are evolutionary arms races between hosts and pathogens . Pathogens evolve much faster than hosts , rapidly changing effector repertoires , thereby changing points of attack in host immune networks . As hosts cannot match the speed of pathogen evolution , it is important that hosts develop robust immune networks that remain functional in the face of effector attack . Mechanisms underlying network robustness are thus a critical aspect of immunity . SA is a signal molecule controlling a major portion of immunity against biotrophic and hemibiotrophic pathogens , including P . syringae [14] . SID2 encodes a key enzyme for SA biosynthesis in response to pathogen infection [15] . In Arabidopsis sid2 mutants , pathogen-induced SA accumulation is almost undetectable [14] . Hundreds of genes are transcriptionally regulated by SA signaling , mediated mainly by a positive regulator of SA signaling , NPR1 [14] . PR1 is one SA-inducible gene used as a canonical SA marker [14] . Arabidopsis has 20 mitogen-activated protein kinases ( MAPKs ) [16] , and four of them , MPK3 , MPK4 , MPK6 and MPK11 , have been described as immune signaling components [17] . MPK3 and MPK6 are associated with immune responses , such as reactive oxygen species ( ROS ) production , ET production/signaling , phytoalexin production and cell death [17] . For instance , ethylene production is positively controlled by dual regulation of enzymes ( ACS ) synthesizing the ethylene precursor 1-amino-cyclopropane-1-carboxylic acid . MPK6 stabilizes ACS2 and ACS6 by their phosphorylation , and MPK3 and MPK6 control gene expression through a transcription factor , WRKY33 , which is activated by the MAPKs [18] , [19] . The same cascade is required for production of a phytoalexin , camalexin , by controlling expression of a biosynthetic gene , PAD3 [20] . A double mutant deficient in MPK3 and MPK6 is embryonic lethal but the single mutants are viable , suggesting functional redundancy between them in development [21] . MPK3 phosphorylates the bZIP type transcription factor VIP1 whose phosphorylation is required for its nuclear translocation [22] . Transient over-expression of VIP1 led to weak induction of PR1 in Arabidopsis protoplasts although involvement of SA in this PR1 induction is not known [23] . The overall spectra of induced defense responses are overlapping between PTI and ETI whereas the kinetics and intensity of the responses seem different [4] , [24] . In Arabidopsis , knocking out the hub genes of four major signaling sectors abolished 80% of flg22-triggered PTI ( flg22-PTI ) and AvrRpt2-triggered ETI ( AvrRpt2-ETI ) , indicating extensively shared signaling network machinery between PTI and ETI [25] . Relationships among these signaling sectors are part compensatory and part synergistic in flg22-PTI but are predominantly compensatory in AvrRpt2-ETI , which explains a high level of robustness in the ETI level against network perturbations [25] . Single mutations ( dde2 , ein2 , pad4 and sid2 ) weakly but significantly compromised flg22-PTI but not AvrRpt2-ETI while the quadruple mutation largely abolished both . These observations demonstrated differences in the robustness of the highly overlapping signaling networks during the two modes of plant immunity . However , the molecular mechanism controlling modulation of the network robustness is not known . Here we report a molecular mechanism that affects the robustness of the plant immune signaling network . Although Arabidopsis MPK3 and MPK6 are activated during both PTI and ETI , the duration of the activation was much longer during ETI than PTI . Only sustained activation of the MAPKs supported expression of a majority of SA-responsive genes in the absence of SA . The roles of MPK3 and SA signaling during AvrRpt2-ETI were compensatory , contributing to network robustness against perturbations during ETI . Our findings demonstrate that a biologically important differential network property , robustness , can emerge from duration of the activity of a network component .
We previously reported that ETI is more robust against network perturbations than PTI due to a higher level of network compensation [25] . We hypothesized that this compensation occurred at the level of gene regulation . To test this hypothesis , we examined expression of a canonical SA marker gene , PR1 , during ETI . Transcriptional induction of PR1 was completely dependent on SID2 , which is a key SA biosynthetic enzyme , and hence completely dependent on SA signaling during PTI [26] . We found that PR1 induction was only partially dependent on SID2 and NPR1 at a late time point of 24 hours post inoculation ( hpi ) with ETI-triggering P . syringae pv . tomato DC3000 ( Pto ) strains expressing the effectors AvrRpt2 ( Pto AvrRpt2 ) or AvrRpm1 ( Pto AvrRpm1 ) ( Figure 1A and Figure S1 ) . While AvrRpt2 and AvrRpm1 are recognized by the CC-type NB-LRR proteins RPS2 and RPM1 , AvrRps4 is recognized by the TIR-type NB-LRR protein RPS4 [3] . We also observed SID2- and NPR1-independent PR1 induction during AvrRps4-triggered ETI although induction levels were lower compared to AvrRpt2- and AvrRpm1-ETI ( Figure S1 ) . In contrast , PR1 induction was completely dependent on SID2 in the case of the non-ETI triggering Pto strain carrying an empty vector ( Pto EV ) . Inoculation of the ETI-triggering strains at a high dose can trigger a form of programmed cell death called a hypersensitive response ( HR ) [3] . The inoculation dose used in this experiment was relatively low ( OD600 = 0 . 001 ) , and we did not observe a macroscopic HR within 24 hpi . To test the possibility that the SA level increased independently of SID2 during ETI , we measured the SA level in these tissues . The increased SA accumulation was completely dependent on SID2 in all conditions ( Figure 1B ) . These results indicate that some SA-independent mechanism ( s ) can activate PR1 during ETI . At an earlier time point of 6 hpi , only SA-dependent PR1 induction was observed with all three strains ( Figure 1A ) , suggesting that this SA-independent mechanism ( s ) during ETI requires more than 6 hours to be effective . SA-independent mechanism ( s ) for PR1 induction during ETI prompted us to investigate the possibility that other SA-responsive genes can also be transcriptionally regulated in an SA-independent manner during ETI . For this purpose , mRNA profiles were analyzed using a whole genome DNA microarray . Leaves of wild type ( Col ) or sid2 plants were inoculated with water ( mock ) , Pto hrcC , Pto EV , or Pto AvrRpt2 , and were collected at 24 hpi for mRNA profiling . The Pto hrcC strain is deficient in the type III secretion system used to transport effectors into plant cells . It elicits the PTI response by presenting various MAMPs [11] . Among 2828 genes that were significantly up- or down-regulated ( with q values<0 . 01 and more than 2-fold changes ) in both Pto EV and Pto AvrRpt2 infection in Col , regulation of 187 genes showed strong SID2-dependence in Pto EV infection ( Figure 2A and Table S1 ) . These genes are designated SA-responsive genes hereafter . Remarkably , regulation of most SA-responsive genes , including PR1 , at 24 hpi with Pto AvrRpt2 is largely SID2-independent although SA contributes to their full expression , indicating that SA-independent signaling mechanism ( s ) can regulate most SA-responsive genes during AvrRpt2-ETI . The SID2-dependency of gene regulation after Pto hrcC inoculation was similar to that after Pto EV inoculation , although the overall extent of up- or down-regulation was lower , and distinct from that after Pto AvrRpt2 inoculation ( Figure S2 and Table S2 ) . Thus , initiation of ETI appears to be the key for activation of this SA-independent mechanism ( s ) . We hypothesized that a kinetic difference in activation of network components is responsible for activation of SA-independent mechanism ( s ) . A prior study suggested that the duration of MPK3 and MPK6 activation is longer during ETI than non-ETI [27] . We compared the duration of MAPK activation in ETI and PTI . When wild-type seedlings in a liquid medium were treated with the PTI inducer flg22 , activation of the MAPKs was observed after 10 min and returned to the basal level within one hour ( Figure 3A ) , confirming previous observations [28] . The possibility that flg22 was rapidly degraded in the liquid culture was excluded since the MAPKs were activated similarly when fresh seedlings were placed in the liquid medium containing flg22 that had been incubated with other seedlings for 3 hours ( Figure 3A ) . Thus , MAPK activation is truly transient after flg22 treatment . We employed transgenic seedlings carrying an estradiol-inducible AvrRpt2 transgene ( XVE-AvrRpt2 ) to measure MAPK activation during ETI in the absence of PTI . The MAPKs were activated by three hours and remained active for at least 7 hours after estradiol treatment ( Figure 3B ) . This sustained MAPK activation was ETI-specific as no such activation was observed in the rps2 mutant background , which lacks the corresponding receptor ( Figure 3C ) . PR1 induction during AvrRpt2-ETI was independent of SA in XVE-AvrRpt2 transgenic seedlings ( Figure S3 ) , which is consistent with the results obtained using adult leaves inoculated with a Pto strain expressing AvrRpt2 ( Figure 1 ) . Similar trends in MAPK activation duration were observed when adult leaves were inoculated with Pto strains: sustained activation of the MAPKs was observed with Pto AvrRpt2 in a manner dependent on the R gene RPS2 , but not with the strains that do not trigger ETI ( Figure 4 ) . While the amounts of activated MPK3 and MPK6 were similar during AvrRpt2-ETI triggered in XVE-AvrRpt2 transgenic plants ( Figure 3 ) , there was more activated MPK3 than activated MPK6 during AvrRpt2-ETI triggered by Pto AvrRpt2 ( Figures 4 , S4 and S5 ) , suggesting that MPK3 plays a major role during AvrRpt2-ETI in bacterial infection . We also observed sustained MAPK activation during AvrRps4-ETI although levels of activation were weaker compared to AvrRpt2-ETI ( Figure S4 ) . Since there are 20 MAPKs in Arabidopsis [16] , we determined the identities of the activated MAPKs . Indeed , the activated MAPKs during AvrRpt2- and AvrRps4-ETI were MPK3 and MPK6 ( Figure S4 ) . Previously , Beckers et al ( 2009 ) reported that an SA analog , benzo ( 1 , 2 , 3 , ) thiadiazole-7-carbothioic acid S-methyl ester ( BTH ) , induced priming of MPK3 activation by inducing expression of MPK3 [29] . In contrast , sustained activation of MPK3 during AvrRpt2-ETI was independent of SA ( Figure S5 ) . The sustained activation was not due to an increased amount of MPK3 as we did not observe obvious changes in the MPK3 protein level during AvrRpt2-ETI ( Figure 3 ) . Taken together , our data show that sustained activation of the MAPKs is SA-independent and occurs during ETI but not during non-ETI responses . To test if sustained activation of MPK3 and MPK6 can induce PR1 in an SA-independent manner , transgenic plants expressing constitutively active forms of MKK4 ( MKK4DD ) or MKK5 ( MKK5DD ) under the control of a dexamethasone ( DEX ) -inducible promoter were employed ( DEX-MKK4DD and DEX-MKK5DD ) . MKK4 and MKK5 are MAP kinase kinases , whose activated forms phosphorylate and activate MPK3 and MPK6 [17] . DEX-induced expression of MKK4DD or MKK5DD leads to sustained activation of MPK3 and MPK6 ( Figure S6 ) [30] . Induction of PR1 was observed 9 hours after DEX treatment ( Figure 5A ) , suggesting that sustained activation of MPK3 and MPK6 is sufficient for induction of PR1 . Induction of FRK1 is thought to be a good marker for activation of MPK3 and MPK6 [31] and was observed 3 hours after DEX treatment while PR1 was not ( Figure 5B ) . FRK1 was strongly induced 30 minutes after flg22 treatment [32] , and the induction did not require SA accumulation ( Figure S7 ) . Thus , although transient MAPK activation of MPK3 and MPK6 is sufficient for FRK1 induction , sustained MAPK activation is necessary and sufficient for SA-independent PR1 induction . The sustained activation of MPK3 and MPK6 by DEX-induced MKK4DD or MKK5DD did not increase the level of SA ( Figure 6A ) . Furthermore , a wild-type-like PR1 induction 24 hours after DEX treatment was observed in plants deficient in SID2 or NPR1 ( Figure 6B ) . Since PR1 induction was not observed in a DEX-inducible ß-glucuronidase ( GUS , an arbitrary reporter gene ) line after DEX treatment , PR1 induction was not caused by the DEX-inducible system or DEX but by induced expression of MKK4DD or MKK5DD . Although MPK4 was activated as well as MPK3 and MPK6 during PTI and ETI ( Figure 4; [17] ) , expression of MKK4DD or MKK5DD does not lead to strong activation of MPK4 [30] . Therefore , it is unlikely that MPK4 plays a role . We conclude that sustained activation of MPK3 and/or MPK6 causes PR1 induction in an SA-independent manner . We tested whether mpk3 and mpk6 single mutations had effects on PR1 induction by MKK4DD or MKK5DD expression . PR1 induction was unaffected in mpk6 but strongly reduced in mpk3 plants ( Figure S8A ) . MKK4DD induction was also strongly reduced in mpk3 plants ( Figure S8B ) , so the reduction of PR1 induction in DEX-MKK4DD/mpk3 may be due to reduction of MKK4DD expression . MKK5DD induction in DEX-MKK5DD/mpk3 was reduced compared to DEX-MKK5DD/Col yet 10 times higher than MKK4DD induction in DEX-MKK4DD/mpk3 while PR1 induction was similarly compromised in both plant lines . Thus , these results suggest that MPK3 is required for SA-independent PR1 induction conferred by forced MKK5 activation while MPK6 is dispensable . We tested whether sustained activation of MPK3 and/or MPK6 also regulates other SA-responsive genes . Leaves of the DEX-MKK4DD transgenic lines in wild type ( Col ) or sid2 backgrounds were treated with DEX or mock control and were collected for mRNA profiling at 24 hours after treatment . The transcriptomic changes caused by DEX treatment were very similar between Col and sid2 ( Figure S9 and Table S3 ) , indicating that gene regulation by sustained activation of the MAPKs is mostly independent of SA . Therefore , only the mRNA profile from the DEX-MKK4DD sid2 line was included in the following analysis . The heatmap in Figure 2A shows that a majority of the SA-responsive genes responded in the DEX-treated DEX-MKK4DD sid2 line similarly to sid2 plants during AvrRpt2-ETI: most up-regulated or down-regulated SA-responsive genes in sid2 during AvrRpt2-ETI were up-regulated or down-regulated , respectively , in the DEX-treated DEX-MKK4DD sid2 line . This suggests that sustained activation of the MAPKs regulates a majority of SA-responsive genes in an SA-independent manner during AvrRpt2-ETI . Three gene clusters were selected for further analysis ( Clusters I–III in Figure 2A ) . The expression level changes of genes in each cluster were averaged and shown in Figure 2B–D . Clusters I and III include genes up- or down-regulated , respectively , in a SID2-independent manner during AvrRpt2-ETI and by sustained activation of the MAPKs . Thus , these genes appear to be regulated by sustained activation of the MAPKs during ETI . Cluster II includes genes that were up-regulated in a largely SID2-independent manner during ETI but not up-regulated by sustained activation of the MAPKs . Thus , up-regulation of the Cluster II genes during ETI is supported by a mechanism ( s ) other than the mechanism mediated by the MAPKs . When the GO terms associated with the clusters were examined , Cluster I , but none of the other clusters , was enriched with genes related to biological stresses ( response to biotic stimulus , P = 2 . 8×10−5; response to other organism , P = 1 . 1×10−4; multi-organism process , P = 5 . 9×10−4 ) . The results imply that genes induced by both SA and the MAPKs are important for biological stress responses . The regulatory trends for the clusters were confirmed by qRT-PCR analysis of one gene from each cluster ( Figure S10 ) . We investigated if compensation between MPK3/MPK6 and SA signaling could be detected in the PR1 expression level during ETI . Leaves of wild type ( Col ) , mpk3 , mpk6 , sid2 , mpk3 sid2 and mpk6 sid2 plants were inoculated with Pto AvrRpt2 or Pto AvrRpm1 , and PR1 expression levels were determined 24 hpi ( Figure 7A ) . While PR1 expression was compromised in sid2 but not in mpk3 or mpk6 during AvrRpt2-ETI , it was compromised in mpk3 sid2 more than in sid2 ( blue bar ) , suggesting compensation between MPK3 and SID2 on PR1 expression during AvrRpt2-ETI . To quantify the level of compensation between MPK3 and SID2 on PR1 expression , a signaling allocation analysis was applied [25] . In this analysis , the effects of the genes and their interactions were estimated for contribution to the PR1 expression level after inoculation . We estimated the individual contribution of MPK3 on the PR1 expression level as the difference in expression levels between sid2 and mpk3 sid2 , that of SID2 as the difference in PR1 expression levels between mpk3 and mpk3 sid2 and their combined contribution as the difference in PR1 expression levels between the wild type and mpk3 sid2 . The value of the genetic interaction between MPK3 and SID2 was calculated by subtracting the sum of the individual contributions of MPK3 and SID2 from their combined contribution . Their combined contribution in the wild type was less than the sum of the individual contributions of SA and MPK3 , which is signified by the negative interaction between them . We previously defined this less-than-additive combined contribution as compensation [25] . Such compensation was observed for AvrRpt2-ETI ( Figure 7B , top ) . Thus , signaling mediated by MPK3 and SA is compensatory on PR1 expression during AvrRpt2-ETI . No significant effects of MPK6 or the interaction ( MPK6:SID2 ) on PR1 expression were detected during AvrRpt2-ETI ( Figure 7A and B ) . No significant effects of MPK3 , MPK6 or their interactions ( MPK3:SID2 and MPK6:SID2 ) on PR1 expression ( Figure 7A and B , red bar ) or resistance ( Figure S11 ) were detected during AvrRpm1-ETI , suggesting a divergence in the mechanisms that modulate network robustness between different cases of ETI . A similar trend was observed with the effects of MPK3 and MPK6 on bacterial resistance in AvrRpt2-ETI ( Figure 7C ) . AvrRpt2-ETI is defined as the difference in in planta growth of Pto EV and Pto AvrRpt2 on a log10-scale [25] . The compensation between MPK3 and SID2 was clear from the signaling allocation analysis , as both had positive effects and their interaction was negative ( Figure 7D , left ) . We did not detect significant effects of MPK6 or the interaction ( MPK6:SID2 ) , although we observed a similar pattern to the case of MPK3 ( Figure 7D , right ) . Thus , compensation of SA signaling by a signaling mechanism involving MPK3 exists in inhibition of bacterial growth , as well as in PR1 expression , during AvrRpt2-ETI . Lethality of the double mutants mpk3 mpk6 [21] does not allow us to determine combined contributions of MPK3 and MPK6 to compensation of SA signaling during ETI . It is possible that MPK6 is not a major factor in SA signaling compensation during ETI and that a signaling mechanism ( s ) other than that involving MPK3 or MPK6 is important during AvrRpm1-ETI . Nonetheless , these results clearly demonstrate that at least during AvrRpt2-ETI , SA signaling can be compensated by MPK3-mediated signaling in regulation of SA-responsive gene expression and that this compensation increases the robustness of the network output . This allows immunity to be maintained even if the major network sector , SA signaling , is compromised .
A prior study implied that the duration of MPK3 and MPK6 activation is longer during ETI compared to during non-ETI upon P . syringae infection [27] . However , it did not rule out the possibility that the effector AvrRpt2 caused sustained MAPK activation through a mechanism independent of recognition of AvrRpt2 via RPS2 . We clearly demonstrated that sustained MAPK activation occurs when ETI is triggered ( Figures 3 and 4 ) . The duration of MPK3 and/or MPK6 activation is the determinant for activation of the SA-independent alternative mechanism to regulate the SA-responsive genes: only sustained MAPK activation results in activation of the alternative mechanism . One potential cause of the differential activation duration is rapid turnover of PTI receptors , PRRs . FLS2 is rapidly degraded and disappears within one hour upon exposure to flg22 [33] , [34] . Although turnover rates of other PRRs are not known , if many PRRs turn over rapidly upon activation , this could explain transient activation of the MAPKs by Pto hrcC ( Figure 4 ) , which presents multiple MAMPs [11] . The turnover rates of R proteins , the ETI receptors , upon their activation are largely unknown . Whether turnover rate is involved or not , this hypothesis that the duration of MAPK activation and , consequently , the robustness of the network can be tuned to each receptor is attractive because it would enable network robustness to be evolutionarily adapted according to what pathogen-derived signals are recognized by the receptors . Another potential but not mutually exclusive cause of the differential activation duration is involvement of protein phosphatases that dephosphorylate and inactivate the MAPKs: activation of the MAPKs may be negatively regulated by a phosphatase ( s ) during non-ETI responses while the phosphatase may be inactivated during ETI , resulting in the sustained activation of the MAPKs . Multiple types of such phosphatases including MAPK phosphatases are known in Arabidopsis [35] . Differential regulation of these phosphatases during ETI and non-ETI responses may explain the differential duration of MAPK activation . Switching of downstream signaling by differential duration of MAPK activation is known in animals and yeast [36]–[38] . In one case , it is explained by nuclear translocation of a MAPK that occurs only after its sustained activation [36] . In this way , sets of substrates available to the MAPK are distinct between its transient and sustained activation , which could lead to distinct downstream signaling . In plants , it has also been reported that MAPKs are translocated to the nucleus upon stimulation [39] , [40] . Investigation of potential subcellular localization changes of Arabidopsis MPK3 and MPK6 during PTI and ETI will provide insight into this possibility . Another appealing explanation is involvement of a feed-forward network motif [41] . For example , activation of a transcription factor TF-X may mediate the alternative mechanism regulated by sustained MAPK activation . The activation of TF-X may require signal Y in addition to active MPK3 and/or MPK6 . Signal Y may be slowly generated as a consequence of the activation of the MAPKs ( e . g . , 5 hours ) . The MAPKs would need to be activated for a long time to simultaneously have both signal Y and the active MAPKs to activate TF-X and regulate the SA-responsive genes . In either scenario , discovery of the signaling components downstream of the sustained MAPK activation will be the key to elucidate the mechanism that decodes duration of MAPK activation . Multiple transcription factors , such as TGAs , WRKYs , TBF1 and VIP1 [14] , [22] , [23] , [42]–[44] , are involved in regulation of PR1 . These transcription factors may provide a good starting point for a search for the decoding mechanism . Pto produces the small molecule coronatine , which is a molecular mimic of the JA-Ile conjugate and promotes virulence by suppressing SA signaling [13] . Pto is highly virulent on Arabidopsis plants while ETI-triggering strains of Pto , such as Pto AvrRpt2 , are much less virulent . Nevertheless , coronatine could suppress SA signaling . Therefore , SA-independent alternative mechanism ( s ) to regulate expression of the SA-responsive genes , such as that mediated by the MAPKs , may have a substantial role against perturbation of the immune signaling network by coronatine . This hypothesis is consistent with our observation that loss of both MPK3 and SA led to increased susceptibility to Pto AvrRpt2 ( Figure 7 ) . Pto DC3000 possesses type III effectors which directly or indirectly suppress MAPK activation [45]–[48] . However , we observed sustained activation of MPK3 and MPK6 during AvrRpt2-ETI when AvrRpt2 was delivered from Pto DC3000 ( Figure 4 ) . We speculate that the amounts of such MAPK-inhibiting type III effectors delivered and/or the kinetics of their delivery are not optimal to effectively suppress MAPK activation when the type III effectors are delivered from Pto DC3000 , which represents a relatively natural context . The effector HopAI1 from Pto DC3000 can physically interact with and inactivate MPK3 and MPK6 by removing the phosphate group from phosphothreonine via a phosphothreonine lyase activity [45] . HopAI1 also targets MPK4 and decreases MPK4 activity [48] . Decreased MPK4 activity appears to be monitored by the NB-LRR protein SUMM2 , resulting in triggering ETI . Overexpression of HopAI1 in wild-type Col-0 plants but not summ2 mutant plants leads to dwarfism and constitutive activation of immune responses [48] . However , Pto DC3000 does not trigger SUMM2-mediated ETI . Consistently , HopAI1 of Pto DC3000 is disrupted by an insertion in its promoter region [49] . Thus , the amount of HopAI1 delivered from Pto DC3000 appears insufficient for effective inhibition of MPK3 and MPK6 activation during AvrRpt2-ETI . Another effector , HopF2 , from Pto DC3000 can also suppress activity of MPK3 , MPK4 and MPK6 by targeting the upstream MKK5 and likely other MKKs as well [46] , [47] . When overexpressed in plants , HopF2 interferes with AvrRpt2-ETI by inhibiting AvrRpt2-mediated RIN4 degradation [50] . Again , the reason that HopF2 cannot suppress sustained activation of MPK3 and MPK6 triggered by AvrRpt2 when it is delivered from Pto DC3000 ( Figure 4 ) is likely insufficient HopF2 or inappropriate timing of its delivery . Delivery of AvrRpt2 may precede that of HopF2 [50] . One enigma is why plants need to make the robustness of the immune signaling network lower during PTI when the network itself has the capacity to be highly robust . If the network output during PTI were as robust as during ETI , the chance that “true” pathogens will overcome PTI would be much lower . We speculate that the lower robustness during PTI is selected through evolution as trade-offs with other requirements . Many MAMPs are shared among pathogens and benign microbes and provide low quality information about pathogen attack . It is probably not adaptive for plants to respond to a MAMP with strong and sustained immune responses similar to those during ETI since in many cases , plants encounter benign microbes and ETI-type responses cost fitness . A strategy apparently selected is to respond weakly first and wait to intensify the response until further information increases the probability that a true pathogen is present [24] . In contrast , since effectors are a hallmark of true pathogens and provide high quality information , during ETI plants can induce rapid and strong immune responses with a very low chance of needless fitness costs . The signaling sector activated by sustained activation of the MAPKs during ETI and the SA signaling sector can regulate the common set of genes . This is one of the mechanisms underlying robustness of the immunity level against network perturbations during ETI . This modulation of the network robustness is controlled by signaling kinetics of a network component . Our findings imply that properties of biological networks can be modulated through network component activities .
Arabidopsis plants were grown in a controlled environment at 22°C with a 12 h photoperiod and 75% relative humidity . Arabidopsis thaliana accession Col-0 was the background of all mutants used in this study . Arabidopsis mpk3-1 ( SALK_151594 ) [21] , mpk6-2 ( SALK_073907 ) [18] , npr1-1 [51] , rps2 101C [52] and sid2-2 [15] were previously described . We generated the double mutants mpk3 sid2 and mpk6 sid2 by standard genetic crosses . Estradiol-inducible AvrRpt2 transgenic lines [53] and the DEX-MKK4DD and -MKK5DD transgenic lines [30] were previously described . We crossed DEX-MKK4DD and -MKK5DD into the mutant backgrounds mpk3 , mpk6 , npr1 , sid2 and vip1 . Primers and restriction enzymes used for screening of the mutants are listed in Table S4 . Flg22 peptide was purchased from EZBiolab Inc ( Westfield , IN , USA ) . Estradiol ( E8875 ) and DEX ( D1756 ) were purchased from Sigma ( Saint Louis , MO , USA ) . Pto DC3000 strains ( or water for mock ) or 2 µM DEX ( or 0 . 1% ethanol for mock ) were infiltrated into leaves of 4-week-old plants . Leaves were collected at the indicated time points . Total RNA isolation and qRT-PCR analysis were carried out as described previously [54] , [55] . The following models were fit to the relative Ct value data compared to Actin2 using the lme function in the nlme package in the R environment: Ctgytr = GYTgyt+Rr+εgytr , where GYT , genotype:treatment:time interaction , and random factors; R , biological replicate; ε , residual; Ctgyr = GYgy+Rr+εgytr , where GY , genotype:treatment interaction; Ctgtr = GTgt+Rr+εgtr , where GT , genotype:time interaction . The mean estimates of the fixed effects were used as the modeled relative Ct values and visualized as the relative log2 expression values and compared by two-tailed t-tests . For the t-tests , the standard errors were calculated using the variance and covariance values obtained from the model fitting . Primers used in the study are listed in Table S4 . Four-week-old Arabidopsis Col-0 and sid2 leaves were infiltrated with Pto hrcC , Pto pLAFR ( EV ) , Pto AvrRpt2 or water ( mock ) . Independently , leaves of four-week-old DEX-MKK4DD plants in Col-0 or a sid2 background were infiltrated with 2 µM DEX or 0 . 1% ethanol ( mock ) . Samples were collected at 24 hpi . Total RNA was extracted as described previously [26] and profiled using the NimbleGen DNA microarray ( A . thaliana Gene Expression 12×135K array TAIR9 . 0 ) following the manufacturer's protocol ( Roche Applied Science , Indianapolis , IN , USA ) . Three independent experiments ( biological replicates ) were performed . The microarray data were submitted to Gene Expression Omnibus ( Accession , GSE40555 ) . Probe signal values were subjected to the robust multi-array average ( RMA ) summarization algorithm [56] using the standard NimbleGen software to obtain the expression level values of the transcripts . Among transcripts of a single gene , those with higher expression values were selected as the representative transcripts of the genes . The following models were fit to log2 expression values using the lmFit function in the limma package in the R environment: Sgyr = GYgyt+Rr+εgyr , where S , log2 expression value , GY , genotype:treatment interaction , and random factors; R , biological replicate; ε , residual . The eBayes function in the limma package was used for variance shrinkage in calculation of the p-values and the Storey's q-values were calculated from the p-values using the qvalue function in the qvalue package . First , genes whose expression was up-regulated or down-regulated ( q values<0 . 01 and more than 2 fold change ) in both Pto EV and Pto AvrRpt2-infected Col compared to mock were selected ( 2828 genes ) . Second , SID2-dependent genes in Pto EV infection ( inductions/suppression in sid2 are less than 20% compared to Col ) were selected ( 187 “SA-responsive” genes ) for the clustering analysis . Heatmaps were generated by CLUSTER [57] using uncentered Pearson correlation and complete linkage , and visualized by TREEVIEW [57] . The accession numbers for the Arabidopsis genes discussed in this article are as follows: Actin2 ( At2g18780 ) , Chitinase ( At1g02360 ) , CHS ( At5g13930 ) , FRK1 ( At2g19190 ) , MKK4 ( At1g51660 ) , MKK5 ( At3g21220 ) , MPK3 ( At3g45640 ) , MPK4 ( At4g01370 ) , MPK6 ( At2g43790 ) , NPR1 ( At1g64280 ) , RPM1 ( At3g07040 ) , RPS2 ( At4g26090 ) and SID2 ( At1g74710 ) . | Robustness of a network is defined by how consistently it performs upon removal of some of its components . It is a common strategy for plant pathogens to attack components of the plant immune signaling network in an attempt to dampen plant immunity . Therefore , it is crucial for the plant immune signaling network to have a high level of robustness . We previously reported that the robustness level of the plant immune signaling network is higher during Effector-Triggered Immunity ( ETI ) than Pattern-Triggered Immunity ( PTI ) . Here we discovered a molecular switch that determines two robustness levels during ETI and PTI . Salicylic acid ( SA ) is a major plant immune signal molecule that regulates many immune-related genes . SA-independent alternative mechanisms also regulated the majority of SA-responsive genes during ETI but not PTI . One of the SA-independent mechanisms was mediated by prolonged activation of MAP kinases ( MAPKs ) . MAPK activation was prolonged during ETI but transient during PTI . Thus , the duration of MAPK activation switches the robustness level of the plant immune signaling network . Our findings imply that the robustness level of a biological network can be modulated by activities of its components . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Dual Regulation of Gene Expression Mediated by Extended MAPK Activation and Salicylic Acid Contributes to Robust Innate Immunity in Arabidopsis thaliana |
Latent Epstein-Barr virus ( EBV ) infection contributes to both B-cell and epithelial-cell malignancies . However , whether lytic EBV infection also contributes to tumors is unclear , although the association between malaria infection and Burkitt lymphomas ( BLs ) may involve excessive lytic EBV replication . A particular variant of the viral promoter ( Zp ) that controls lytic EBV reactivation is over-represented , relative to its frequency in non-malignant tissue , in EBV-positive nasopharyngeal carcinomas and AIDS-related lymphomas . To date , no functional differences between the prototype Zp ( Zp-P ) and the cancer-associated variant ( Zp-V3 ) have been identified . Here we show that a single nucleotide difference between the Zp-V3 and Zp-P promoters creates a binding site for the cellular transcription factor , NFATc1 , in the Zp-V3 ( but not Zp-P ) variant , and greatly enhances Zp activity and lytic viral reactivation in response to NFATc1-inducing stimuli such as B-cell receptor activation and ionomycin . Furthermore , we demonstrate that restoring this NFATc1-motif to the Zp-P variant in the context of the intact EBV B95 . 8 strain genome greatly enhances lytic viral reactivation in response to the NFATc1-activating agent , ionomycin , and this effect is blocked by the NFAT inhibitory agent , cyclosporine , as well as NFATc1 siRNA . We also show that the Zp-V3 variant is over-represented in EBV-positive BLs and gastric cancers , and in EBV-transformed B-cell lines derived from EBV-infected breast milk of Kenyan mothers that had malaria during pregnancy . These results demonstrate that the Zp-V3 enhances EBV lytic reactivation to physiologically-relevant stimuli , and suggest that increased lytic infection may contribute to the increased prevalence of this variant in EBV-associated malignancies .
Epstein-Barr virus ( EBV ) causes infectious mononucleosis and is associated with a variety of different types of human malignancies , including B-cell lymphomas and nasopharyngeal carcinoma . EBV infects approximately 90% of the world’s population , and like all herpesviruses , EBV persists in the host for life . EBV establishes long-term viral latency in the memory B cell compartment of humans [1–3] , whereas lytic EBV infection occurs in antigen-stimulated B cells , plasma cells and oropharyngeal epithelial cells [4–6] . While the initial viral infection is often associated with the clinical symptoms of infectious mononucleosis , EBV does not usually cause any subsequent illness in immune competent hosts , although infectious viral particles continue to be periodically shed in the saliva . Nevertheless , EBV infection sometimes leads to the development of EBV-positive tumors , particularly in immunosuppressed individuals . EBV-induced tumors are thought to be primarily due to the latent form of viral infection . Latent EBV infection transforms primary human B cells in vitro into immortalized lymphoblastoid cell lines ( LCLs ) that induce lymphomas when injected into immunodeficient mice . Furthermore , the major EBV-encoded oncoproteins are expressed during latent EBV infection , and EBV-positive tumors in humans are composed largely of latently infected cells . Given the essential roles of the viral latency proteins in EBV-positive tumors , relatively little attention has been paid to the potential role ( s ) of lytic EBV infection in promoting EBV-induced tumors . Although widespread fully lytic EBV infection of tumor cells would likely be incompatible with tumor growth , increasing evidence suggests that lytic viral infection contributes to the formation of tumors caused by the related human gamma herpesvirus , KSHV ( Kaposi’s sarcoma associated herpesvirus ) [7] . Excessive lytic EBV infection in humans could potentially increase the likelihood of EBV-positive tumors by increasing the total number of virally infected cells ( including the number of latently infected cells ) , and/or by inducing paracrine effects that help support the growth and/or viability of latently infected tumor cells . For example , B cells with lytic EBV infection have enhanced secretion of the B cell growth factor , IL-6 [8] , the angiogenesis factor , VEGF [9] , and the immunosuppressive factors , viral and cellular IL-10 [10 , 11] . Consistent with a role for lytic EBV infection in promoting the development of EBV-associated tumors , increased lytic infection preceding tumor development seems to occur in a number of EBV-positive human cancers . Patients who develop EBV-positive nasopharyngeal carcinoma ( NPC ) almost universally have extremely high levels of antibodies directed against lytic EBV proteins even before their cancers are clinically symptomatic , and monitoring lytic EBV antibody levels is useful for detecting NPC at early stages [12] . Immunosuppressed organ transplant recipients , who are highly prone to developing EBV-induced lymphoproliferative disease , have a high level of lytic , as well as latent , EBV infection [13] . Malaria infection , which is thought to promote the development of EBV-positive Burkitt lymphoma ( BL ) , greatly increases the amount of lytic EBV infection in children [14–16] . Furthermore , higher levels of infectious EBV are present in the breast milk of malaria-infected , versus malaria-uninfected , Kenyan women [17] , and Kenyan infants more frequently become EBV-infected at very young ages ( less than 6 months ) when residing in areas with a high prevalence of malaria [18] . The latent-to-lytic switch in EBV-infected cells is mediated by the EBV immediate-early BZLF1 gene product ( Z ) . Z is a transcription factor that binds to , and transcriptionally activates , lytic viral gene promoters , resulting in the lytic form of viral DNA replication and assembly of infectious viral particles [19] . Regulation of the Z promoter ( Zp ) by cellular transcription factors determines whether EBV infection is latent or lytic . B-cell receptor ( BCR ) stimulation potently induces lytic EBV gene expression in certain Burkitt lymphoma cell lines in vitro , and BCR activation in response to antigen stimulation of EBV-infected B cells is thought to be a biologically important mechanism by which the EBV life cycle is regulated in humans [19] . In addition to its immunosuppressive effect [20] , malaria is thought to increase the amount of lytic EBV infection by inducing polyclonal B cell stimulation [16 , 21] . However , the precise cellular transcription factors that link the BCR signal to EBV lytic reactivation are only partially understood , and EBV-infected cell lines differ substantially in their ability to reactivate in response to BCR stimulation in vitro . A particular promoter variant of the EBV Z promoter ( Zp-V3 ) has been reported in two different studies to be over-represented in both EBV-infected NPCs [22] and EBV-infected AIDS-related lymphomas [23] in comparison to its frequency in EBV-infected non-malignant tissues obtained from patients in the same geographic regions . However , whether this promoter variant affects BZLF1 transcription or lytic EBV reactivation is not known . Here , we have examined the functional consequences of the malignancy-associated Zp-V3 variant . We show that in comparison to the prototype Zp variant ( Zp-P ) , the Zp-V3 variant responds much more strongly to BCR-ligation and ionomycin stimulation in B cells . Furthermore , we demonstrate that this difference results from a single nucleotide difference in the two promoter variants ( which creates an NFATc1 binding motif in the Zp-V3 form of the promoter ) , and is sufficient to confer greatly enhanced lytic viral protein expression in EBV-infected B cells . Importantly , we find that the Zp-V3 variant is highly over-represented in a set of EBV-transformed lymphoblastoid cell lines ( LCLs ) that were derived from EBV present in breast milk of Kenyan mothers that had malaria during pregnancy , versus a set of spontaneous LCLs that were derived from the blood of healthy Kenyan individuals living in malaria-high regions . In addition , we show that the Zp-V3 variant is also over-represented in both EBV-positive Burkitt lymphomas , and EBV-positive gastric carcinomas , relative to its frequency in healthy control patients . These findings suggest that the Zp-V3 version of the EBV BZLF1 promoter increases the likelihood of EBV-induced malignancies by increasing the level of lytic EBV infection .
To determine if the Zp-P and Zp-V3 variants of the EBV Z promoter have different activities in a B-cell environment , we inserted the two promoter variants upstream of the luciferase gene in the pCpGL luciferase vector and performed transient reporter gene assays . Although the two promoter variants had similar constitutive activity in BJAB cells ( derived from an EBV-negative B-cell lymphoma ) ( Fig 1A ) , the Zp-V3 variant was much more efficiently activated when cells were treated with an anti-IgM antibody to stimulate the BCR ( Fig 1B ) . In contrast to the effect of BCR stimulation , the two promoter variants responded similarly to the KLF4 transcription factor ( Fig 1C ) , which binds to and activates the Z promoter [24 , 25] . These results suggest that the Zp-V3 variant of the Z promoter is more responsive to BCR stimulation than the Zp-P variant . The promoter sequences located between -668 and +15 ( relative to the transcriptional start site ) of the Zp-P ( from B95 . 8 strain EBV ) and Zp-V3 ( from M81 strain EBV ) variants differ by only seven nucleotides ( Fig 2A ) . To examine how these differences contribute to the enhanced responsiveness of the Zp-V3 variant to BCR stimulation , we individually switched each of the variant nucleotides in the Zp-V3 luciferase construct to the nucleotides present in the Zp-P promoter . Altering the variant residues located at -100 , -106 , -274 , -365 , -460 , and -525 ( relative to the transcriptional start site ) had relatively little effect on the response of the Zp-V3 promoter to anti-IgM treatment in BJAB cells ( Fig 2B ) . However , switching the nucleotide located at -141 from a G to an A almost completely abolished the ability of the Zp-V3 variant to be activated by BCR stimulation , reducing it to the level seen with Zp-P . Thus , a G nucleotide at position -141 in the Zp promoter is required for efficient BCR activation . To determine if a G nucleotide at -141 is sufficient to increase the BCR responsiveness of the Zp-P version of the Z promoter , we switched the -141 nucleotide from an A to G in the Zp-P luciferase construct . As shown in Fig 2C , this single basepair change conferred strong responsiveness to BCR stimulation . These results suggest that a single nucleotide difference in the two Z promoter variants is largely responsible for the different responses of Zp-V3 and Zp-P to BCR stimulation . Comparison of the Zp-V3 versus Zp-P sequences surrounding the -141 Zp nucleotide revealed that the Zp-V3 promoter has a consensus NFAT ( Nuclear Factor of Activated T cells ) motif ( TGGAAA ) [26] that is not present in the Zp-P version of the promoter ( Fig 3A ) . The NFAT family consists of five cellular transcription factors; NFATc1 , expressed in lymphoid tissue , is translocated to the nucleus following T cell receptor and BCR activation [27–30] , and is constitutively activated in some B cell lymphomas , including the BJAB cell line [31–33] . Since BCR signaling is known to translocate and activate cellular NFAT transcription factors [30 , 34] , we used a luciferase assay to determine whether pretreating BJAB cells with cyclosporine ( an NFAT inhibitor [27] ) abolished the ability of BCR stimulation to activate the Zp-V3 promoter . As shown in Fig 3B , cyclosporine treatment prevented anti-IgM activation of Zp-V3 , suggesting that the BCR stimulatory effect may be mediated through a NFAT family member . Of note , NFAT activation of promoters commonly requires cooperative binding of NFAT with other transcription factors ( in particular , AP1 and Ets family members ) [35–38] , and the potential NFAT site on the Zp-V3 variant is adjacent to an AP1-like motif ( TGAGCCA ) known as ZIIIA that has previously been shown to be required for BCR activation of Zp [39 , 40] . To determine if NFATc1 can bind to the Zp-V3 , but not Zp-P , version of the Z promoter , we performed EMSAs using labeled oligonucleotide probes containing the -155 to -127 sequences of either Zp variant , and nuclear extracts harvested from untreated or anti-IgM treated BJAB cells . These probes contain the potential NFAT site but not the adjacent potential AP1 motif . A protein that binds to the Zp-V3 , but not Zp-P , version of the probes was observed in the presence and absence of anti-IgM treatment ( Fig 3C ) . Furthermore , two different unlabeled oligonucleotides containing different known binding sequences for NFAT competed for binding with this protein , while oligonucleotides containing Ets , AP1 , and ELK1 consensus sites did not ( Fig 3D ) . In addition , pre-incubating the nuclear extract with an antibody against NFATc1 blocked most of the binding to the Zp-V3 probe , while an antibody against another transcription factor , C/EBPα , had little effect ( Fig 3E ) . These results confirm that NFATc1 can bind to the Zp-V3 but not Zp-P form of Zp . Since NFATc1 is constitutively expressed in BJAB nuclear extracts , and its binding to Zp-V3 is not increased by anti-IgM treatment of cells , we next asked if BCR stimulation of BJAB cells enhances AP1 binding to the adjacent AP1-like ( “ZIIIA” ) motif , and if this effect is NFATc1-dependent . To confirm that crosslinking of the BCR ( which is known to induce expression of AP1 family members [24 , 25] ) results in enhanced AP1 activity in BJAB cells , EMSAs were performed using a labeled consensus AP1 probe and nuclear extracts harvested from untreated or anti-IgM treated cells . A protein binding to the probe was greatly increased in the anti-IgM treated extracts , and this binding was competed by cold oligonucleotide containing a consensus AP1 motif but not by an oligonucleotide containing the NFAT motif ( Fig 4A ) . These results confirm that BCR activation in BJAB cells results in strongly increased nuclear AP1 activity . We next determined if the Zp AP1-like motif located between -123 to -129 ( TGAGCCA versus the AP1 consensus binding sequence TGAGTCA[41] ) , can bind AP1 in the presence or absence of the adjacent NFAT motif . A previously published paper did not find AP1 binding to this motif [42]; however , this study did not use probes that also contain the adjacent NFAT motif . As shown in Fig 4B , when nuclear extracts from untreated or anti-IgM treated BJAB cells were incubated with a longer oligonucleotide probe that contains both the NFATc1 and ZIIIA motifs , an additional ( larger ) protein complex bound to the larger probe , and this complex was found only in the anti-IgM treated cells . Furthermore , this larger complex was competed away with both an unlabeled oligonucleotide containing the consensus AP1 motif , and an oligonucleotide containing the consensus NFAT motif . In addition , the anti-IgM dependent complex was super-shifted by pre-incubation with an anti-cFos ( AP1 ) antibody ( Fig 4C ) but not an anti-XBP1 antibody . These results confirm that BCR stimulation of BJAB cells allows AP1 to bind to the Zp ZIIIA motif in an NFATc1-dependent manner , and that this only occurs on the Zp-V3 version of the promoter . Given that AP1 is recruited to the Zp-V3 ZIIIA site in an NFATc1-dependent manner , we next asked if the combination of NFATc1 and cFos synergistically activates the Zp-V3 promoter in reporter gene assays . As shown in Fig 5A , the combination of cFos and NFATc1 activated the Zp-V3 luciferase construct much more effectively than either cFos or NFATc1 alone , although the level of transfected NFATc1 was similar with or without co-transfected cFos ( Fig 5B ) . Furthermore , mutation of either the -141 NFAT motif , or the ZIIIA motif , greatly decreased the ability of the NFATc1/cFos combination to activate the promoter . These results confirm that NFATc1 and cFos collaborate to activate the Zp-V3 promoter by binding to the -141 NFAT and ZIIIA sites , respectively . The EBV-encoded protein LMP2A mimics constitutively active BCR signaling to enhance B cell survival and proliferation [43–50] , and in some , but not all , studies has been reported to activate Zp [51] . To determine if LMP2A can activate the Zp-V3 form of Zp , BJAB cells were transfected with the Zp-V3 or Zp-P reporter gene construct in the presence or absence of a co-transfected LMP2A expression vector . As shown in Fig 6A , LMP2A expression greatly enhanced the activity of the Zp-V3 promoter , but only slightly increased that of the Zp-P promoter . This activation of the Zp-V3 promoter was abolished by mutation of either the NFATc1 or ZIIIA binding motifs ( Fig 6B ) . Furthermore , treatment of cells with the NFAT inhibitor cyclosporine also prevented LMP2A from activating the Zp-V3 promoter . Together these results suggest that , similar to the effect of the authentic BCR , LMP2A signals through NFATc1 to increase Zp-V3 but not Zp-P activity . We next asked if altering a single basepair of the Zp sequence ( -141 ) in the context of the intact approximately 172 Kbp type 1 B95 . 8 strain EBV genome ( which has Zp-P ) is sufficient to change the lytic phenotype of the virus . As shown in Fig 7A , B95 . 8 virus containing the mutated Zp -141 nucleotide ( Zp-V3 form ) expressed much more Z protein than the WT virus following infection of two different EBV-negative Burkitt lines , BJAB and Akata , although similar levels of the latent EBV protein , EBNA2 , were expressed in each cell type . To determine if the B95 . 8 Zp-V3 mutant is also more lytic following infection of normal B cells , we infected primary peripheral B cells with the mutant Zp-V3 virus or virus in which the mutation had been reversed back to the WT sequence ( revertant virus ) using an MOI of 0 . 1 . Since both viruses have the GFP gene inserted into their genomes , we used GFP flow cytometry analysis to examine the number of cells infected with each virus and the level of GFP expression per cell . Although a larger number of B cells ( approximately 8% of cells ) were infected with the revertant ( WT ) virus in comparison to the Zp-V3 mutant virus ( approximately 3% of cells ) , the level of Z expression assessed by immunoblot on day 3 after infection was greater in the Zp-V3 mutant infected cells ( Fig 7B ) . Together , these results confirm that a single nucleotide alteration of the BZLF1 promoter is sufficient to confer enhanced lytic gene expression following EBV infection of either primary B cells or Burkitt lymphoma cells . The expression of early and late lytic EBV genes in B cells requires that the incoming ( non-methylated ) viral genome becomes highly methylated ( since Z preferentially binds to and activates methylated viral promoters ) [52–54] , and does not occur until at least 2 weeks after EBV infection [53] . To determine if the Zp-V3 sequence enhances the ability of B95 . 8 virus to lytically reactivate in stably infected cell lines , we infected EBV-negative Mutu cells ( derived from a Burkitt lymphoma ) with WT , revertant , or Zp-V3 mutant B95 . 8 viruses , and used hygromycin selection to obtain stably infected cell lines . As shown in Fig 8A , stably infected Mutu cell lines all had type I EBV latency ( EBNA1-pos , LMP1-neg , EBNA2-neg ) , independent of whether cells were infected with the WT , revertant or Zp-V3 mutant viruses . However , when treated with the NFAT-inducing agent , ionomycin , Mutu cells infected with the Zp-V3 mutant had much more Z protein expression , as well as increased expression of an early lytic protein ( BMRF1 ) and a late lytic viral protein ( VCA-p18 ) , compared to cells infected with the WT or revertant viruses . Furthermore , the ability of ionomycin treatment to activate lytic EBV protein expression was reversed by cyclosporine treatment ( Fig 8A ) or NFATc1 siRNA ( Fig 8B ) , confirming that its effect is at least partially mediated through activated NFAT . Likewise , the ability of anti-IgG mediated BCR activation to induce lytic gene expression was increased in Mutu cells infected with the Zp-141G virus ( Fig 8C ) . In contrast , cells infected with the mutant , wildtype , or revertant viruses all induced similar levels of Z expression when treated with the combination of phorbol-12-myristate-13-acetate ( TPA ) and sodium butyrate ( NaBut ) , and the effect of TPA/sodium butyrate was not reversed by cyclosporine ( Fig 8D ) . Thus , the Zp-V3 variant specifically confers increased Z expression in response to NFAT-inducing agents , and does not globally increase Z expression in response to all lytic inducing agents . To confirm that NFATc1 binds more efficiently to the Zp-V3 form of the Z promoter in EBV-infected cells in vivo , we performed ChIP assays in ionomycin-treated Mutu cells infected with the wildtype or Zp-141G B95 . 8 viruses ( Fig 8E ) . Consistent with the in vitro binding assays , these results showed that endogenous NFATc1 binds more strongly to the Z promoter of the Zp-141G ( Zp-V3 type ) virus in comparison to the wildtype ( Zp-P type ) virus . Since excessively lytic EBV infection could be incompatible with the establishment of long-term viral latency and B cell transformation [55] , we determined if the Zp-V3 mutant B95 . 8 virus can transform primary B cells in vitro . Purified adult peripheral B cells were infected with the WT or Zp-V3 mutant viruses using 0 . 25 infectious EBV unit per cell , and the percentage of wells containing lymphoblastoid outgrowths at day 21 post-infection was examined . 10/10 wells infected with either mutant , wildtype or revertant viruses had such outgrowths , suggesting that the Zp-V3 mutant has similar transforming capacity as the WT B95 . 8 strain virus , at least when using an MOI of 0 . 25 ( Fig 9 ) . Therefore , the enhanced lytic protein expression that occurs following infection of B cells with viruses containing this form of the Z promoter is not sufficient ( at least in the context of B95 . 8 strain EBV ) to inhibit viral transformation of primary B cells . The frequency of the Zp-V3 variant in non-malignant tissues is variable depending upon the EBV type ( type 1 versus type 2 ) , and geographic region . Interestingly , type 2 EBV is most common in areas of the world where malaria and Burkitt lymphoma are endemic , and the Zp-V3 form of the BZLF1 promoter is present in all type 2 EBV genomes sequenced to date [56 , 57] . In contrast , Zp-V3 is relatively rare in type 1 EBV , except for type 1 EBV genomes isolated from Asian EBV strains [56] . Although the Zp-V3 variant has been shown to be over-represented in type 1 EBV genomes isolated from NPCs in China , and in AIDs-related lymphomas in Italy ( relative to its frequency in non-malignant samples obtained from humans living in the same geographic regions [22 , 23] ) , whether the Zp-V3 variant is over-represented in type 1 EBV genomes in Burkitt lymphomas ( BLs ) is not yet known . To investigate this , we examined the Zp status of EBV-infected BLs obtained from Africa or South America [56 , 58–60] ( detailed in S1 Table ) versus EBV genomes derived from non-malignant samples from the same geographic regions ( spontaneous LCL samples and infectious mononucleosis samples [56 , 60]; detailed in S2 Table ) . Zp sequence alignments from malignant and normal tissues for previously unanalyzed promoter variant sequences are shown in S6 Table . Type 2 EBV genomes were similarly represented in the Burkitt lymphoma and non-malignant samples ( Table 1 ) , and as expected all type 2 EBV genomes had the Zp-V3 form of Zp ( Table 2 ) . Importantly , we found that type 1 EBV genomes in BLs are much more likely to contain the Zp-V3 variant of Zp ( 37% ) versus type 1 EBV genomes obtained from non-malignant samples ( 4% ) ; p <0 . 003 by Fisher’s exact test ( Table 2 ) . This result reveals that Zp-V3 containing T1 EBV is over-represented ( and likely selected for ) in EBV-infected Burkitt lymphomas . Up to 10% of gastric cancers worldwide are EBV-infected [61–64] and the genomes of a number of EBV-infected gastric cancers are now available in Genbank and the TCGA database . To determine whether the Zp-V3 variant is also over-represented in gastric carcinomas , we examined the Zp status of 41 type 1 EBV-infected gastric carcinomas obtained worldwide [65–68] , versus 113 type 1 EBV genomes derived from non-malignant samples [22 , 56 , 69–71] ( including spontaneous LCL samples , infectious mononucleosis samples , saliva from healthy individuals , and EBV genome sequences detected in non-tumor control tissues ( of any kind ) in the whole exome sequence ( WXS ) TCGA database ) as detailed in S3–S5 Tables . Zp sequence alignments from malignant and normal tissues for previously unanalyzed promoter variant sequences are shown in S6 Table . Since the Zp-V3 variant is much more common in type 1 EBV isolates from Asia , and gastric carcinoma is also more common in Asia , we also compared the frequency of Zp-V3 in gastric carcinomas versus non-malignant tissues isolated from samples obtained from either Asian ( known Asian individuals or samples obtained from Asian countries ) or presumed non-Asian individuals ( known Caucasian individuals , or samples obtained from non-Asian countries ) . As shown in Table 3 , the Zp-V3 variant is significantly over-represented in type 1 EBV-positive gastric cancers worldwide ( 44% of tumors ) relative to its frequency in non-malignant tissues ( 19% ) ( p < 0 . 003 using a Fisher’s exact test ) . When only samples from Asian ( or presumed Asian ) patients are examined , Zp-V3 type 1 EBV is still over-represented in gastric tumors ( 68% ) relative to the non-malignant samples ( 27% ) ( p < 0 . 001 using a Fisher’s exact test ) ( Table 4 ) . Likewise , when only samples from presumed non-Asian patients are examined , Zp-V3 type 1 EBV is still over-represented in gastric tumors ( 17% ) relative to the non-malignant samples ( 0% ) ( p < 0 . 03 using a Fisher’s exact test ) ( Table 5 ) . Thus , the Zp-V3 variant may also increase the risk of developing EBV-positive gastric cancers . Finally , given the recent finding that infectious EBV in the breast milk of mothers that had malaria during pregnancy may serve as a source of EBV transmission to neonates [17] , we examined whether LCL lines derived from the breast milk of these mothers have a high prevalence of the Zp-V3 variant . For this analysis , we sequenced the Z promoter in the EBV genomes of LCLs transformed by breast-milk-derived EBV from 10 different Kenyan mothers that had malaria during pregnancy; we then compared the frequency of this variant in the breast-milk derived LCLs to that observed in 13 different spontaneous EBV-infected LCLs derived from the blood of healthy Kenyan individuals residing in areas with high frequency malaria transmission ( Table 6 ) . We excluded LCLs infected with type 2 EBV ( 6/19 of the spontaneous LCLs in healthy Kenyan donors , versus 0/10 LCLs derived from breast milk ) from this analysis since all type 2 isolates are known to contain the Zp-V3 variant . Surprisingly , all 10 of the breast milk-derived LCLs contained the Zp-V3 variant , and all also had type 1 EBV . In comparison , 0/13 of type 1 EBV-infected LCLs derived from the blood of healthy Kenya donors contained the Zp-V3 variant ( p < 0 . 001 ) . These results suggest that Zp-V3 containing EBV strains may be more prone to lytic reactivation in breast milk than Zp-P containing strains .
Latent EBV infection transforms primary B cells in vitro and clearly contributes to the development of a number of EBV-associated human malignancies . However , since the most transforming form of EBV latency ( type III ) is sufficient to induce proliferation and survival of B cells , whether lytic EBV infection also contributes to EBV-associated malignancies is less clear . Here we have examined whether a cancer-associated polymorphism of the viral BZLF1 immediate-early promoter affects lytic EBV gene expression . We demonstrate that this cancer-associated promoter variant contains an NFATc1 binding motif not present in the prototype promoter , and has enhanced lytic gene activation in response to BCR stimulation . Indeed , we find that altering only a single basepair of the BZLF1 promoter ( to confer NFATc1 binding ) in the context of the intact type 1 EBV genome is sufficient to increase the amount of lytic EBV protein expression in B cells . In addition , we show that the presence of high EBV titers in the breast milk of malaria-infected Kenyan women ( measured by the ability of the milk to transform B cells into LCLs in vitro ) is highly associated with the presence of the Zp-V3 variant in type 1 EBV strains . Importantly , we also show for the first time that Zp-V3 containing type 1 EBV is likewise over-represented in both Burkitt lymphomas , and gastric carcinomas , relative to non-malignant control samples . Together , the studies presented here suggest that enhanced lytic EBV gene expression increases the likelihood of at least some types of EBV-associated malignancies in humans . BCR stimulation has been known for some time to induce lytic EBV reactivation in many EBV-infected Burkitt cell lines , but the major BCR effect was previously reported to be mediated through post-translational modification of MEF2 family members that can also bind to the BZLF1 promoter [25 , 72] . However , many of these previous studies used the Zp-P form of the promoter , which is found in the prototype laboratory EBV strain ( B95 . 8 ) . While a previous paper reported that BCR-mediated stimulation of the Zp-P promoter is largely dependent on a feed-forward loop in which the BZLF1 protein binds to and activates its own promoter ( once a small amount of Z gene expression has been stimulated by cellular transcription factors such as MEF2 ) [39] , we show that BCR stimulation directly and strongly activates the Zp-V3 version of the promoter without the requirement for concomitant Z protein expression . We also found that the AP1-like “ZIIIA” motif is required for BCR-mediated activation of Zp-V3 , and binds AP1 in an NFAT-dependent manner . In contrast , previous studies using the Zp-P promoter variant found that the ZIIIA motif primarily serves as a binding site for the Z protein itself ( rather than AP1 binding ) [39 , 42 , 73] . Since the NFATc1 binding site in the Zp-V3 variant overlaps the “ZIC” motif previously reported to be important for TPA-induced activation of the Zp-P variant , we cannot exclude the possibility that additional transcription factors can also regulate Z transcription through this motif in one or both promoter variants . Given the long-appreciated epidemiologic association of malaria infection with EBV-induced Burkitt lymphomas in Africa , and the growing evidence that malaria increases the amount of lytic EBV in co-infected patients , we asked if the presence of infectious EBV in the breast milk of Kenyan mothers that had malaria during pregnancy correlates with the presence of the Zp-V3 form of the EBV promoter . This seems to be the case , since 10/10 of the type 1 EBV LCL lines derived from infectious EBV in the breast milk of these women had this form of the EBV promoter , versus only 0/13 type 1 LCL lines derived from the blood of healthy Kenyan patients in high malaria regions . B cells in breast milk are more highly activated , and much more likely to differentiate into plasmablasts or plasma cells , than the B cells in peripheral blood [74] . The known propensity of EBV to lytically reactivate in antigen-stimulated plasma cells , and our finding that the Zp-V3 variant is particularly responsive to BCR stimulation , may explain why the Zp-V3 variant is highly over-represented in LCLs derived from infectious EBV in the breast milk of Kenyan mothers that had malaria during pregnancy . Our results suggest that Zp-V3-containing EBV strains may be more easily transferred to nursing neonates than EBV strains containing the Zp-P variant , particularly if the breast milk does not contain enough maternal EBV-neutralizing antibodies to block infection of the neonate ( as may occur in malaria ) . If so , very early EBV infection in the context of an immature immune system may predispose infants to developing Burkitt lymphoma . Clearly , however , larger prospective studies are needed to confirm that Zp-V3 containing EBV strains are more likely than Zp-P containing strains to produce infectious viral particles in breast milk , and/or to be transmitted from breast milk to nursing infants . Whether the Zp-V3 variant also confers greater Z expression in epithelial cells is not yet clear . However , since we previously showed that B95 . 8 EBV-infected gastric AGS cells ( one of the few epithelial cell lines easily infected with this strain ) are remarkably lytic [75] , the Zp-P promoter variant is clearly quite active in this cell type . In addition , our finding here that KLF4 , a cellular transcription factor required for lytic EBV reactivation during epithelial cell differentiation [24] , activates the Zp-P and Zp-V3 variants with similar efficiency suggests that the Zp-P and Zp-V3 versions of the promoter may have similar activity in infected epithelial cells . Furthermore , a recent study showed that similar levels of EBV are contained in the saliva of United Kingdom college students infected with Zp-V3 containing EBV strains versus Zp-P containing strains [56] . Thus , over-representation of the Zp-V3 variant in nasopharyngeal carcinomas [22 , 23] and gastric carcinomas ( shown here ) may reflect increased hematogenous delivery of B-cell derived infectious EBV particles to nasopharyngeal and gastric epithelial cells . There are two major strains of EBV ( referred to as type 1 and type 2 ) , and essentially all type 2 EBV strains carry the Zp-V3 version of the BZLF1 promoter , whereas only a minority of the type 1 strains carry this version [56 , 57] . The major phenotypic differences between the type 1 and type 2 strains of EBV are thought to reflect differences in the sequences of the essential latent transforming genes , EBNA2 and EBNA3A/3C [76–80] . Although type 2 EBV transforms B cells in vitro less efficiently than type 1 EBV [81] , there is no evidence that type 2 EBV is less competent for causing EBV-associated Burkitt lymphomas in humans [82] . Indeed , type 2 strain EBV infection is particularly common in regions of Africa that have high rates of Burkitt lymphoma , although even in these regions type 1 strain is still more common [82] . Given the many differences between type 1 and type 2 EBV , it is important to stress that our studies here demonstrate that the Zp-V3 version of the Z promoter is more lytic than the Zp-P version in the context of a type 1 EBV genome ( B95 . 8 strain ) . Thus our results are not confounded by other potential important differences between the two strains due to alterations in the EBV latency proteins . Nevertheless , since we show here that the Zp-V3 promoter variant produces more lytic EBV reactivation than the Zp-P variant in response to BCR activation , and all type 2 EBV strains sequenced to date contain the Zp-V3 version of the promoter , it is interesting to speculate that increased lytic reactivation by type 2 EBV strains may partially compensate for the less transforming phenotype ( at least in vitro ) in terms of the ability of type 2 strains to promote Burkitt lymphomas . An interesting unanswered question is why EBV has evolved to have at least two different BZLF1 promoter variants . It is possible that the Zp-P variant common in the type 1 strain ensures that the virus can establish long-term latency in B cells , whereas the Zp-V3 variant that is universally present in type 2 strains ( as well as some type 1 strains ) increases the efficiency of lytic virus reactivation and helps ensure efficient horizontal transmission . Many individuals ( particularly when immunosuppressed ) are co-infected with type 1 and type 2 EBV strains [83] , and the breast milk of Kenyan women that had malaria during pregnancy was found to commonly contain both types of EBV [17] . Furthermore , a number of recombinant EBV genomes containing portions of both type 1 and type 2 EBV have now been found in tumors and LCLs [57] , suggesting that individual B cells may be simultaneously infected with both types of EBV . In B cells infected with more than one virus type , Z protein transcribed from the Zp-V3 promoter could simultaneously induce reactivation of the Zp-P carrying strain , since once made , Z can auto-activate either form of the Z promoter . However , an important finding in our studies was that type 1 EBV strains containing the Zp-V3 BZLF1 promoter variant are particularly over-represented in Burkitt lymphomas relative to either Zp-P containing type 1 EBV or Zp-V3 containing T2 EBV strains . Therefore , we speculate that Zp-V3 containing type 1 strains may be especially transforming because they incorporate both the increased lytic activity of Zp-V3 ( usually type 2 ) -containing EBV strains , and the enhanced transforming functions of type 1 EBV strains . Of note , although Zp-V3 incorporation into otherwise type 1 EBV genomes may result from recombination between type 1 and type 2 EBV strains , we identified some Burkitt lymphomas and gastric carcinomas in which only the Zp -141 nucleotide was switched to the Zp-V3 form of the promoter , and the other two variant nucleotides were the same as the Zp-P variant . In contrast , we found no tumors in which one of the other Zp-P specific nucleotides was mutated to the Zp-V3 variant without mutation of the Zp -141 nucleotide . This result suggests that in some instances , mutations which specifically convert the Zp-P -141 sequence to the Zp-V3 sequence may be selected for in EBV-infected tumors in the absence of recombination . Our results also raise the interesting possibility that certain EBV-associated cancers are particularly common in Asia due to the high frequency of Zp-V3 containing T1 EBV strains in this part of the world . Finally , our findings suggest that additional studies to examine whether the Zp-V3 version of the BZLF1 promoter is associated with increased lytic EBV infection in humans are warranted . If this proves to be the case , it would have important clinical implications , and would buttress the argument that anti-EBV vaccines that inhibit lytic infection without preventing the establishment of viral latency might be useful for preventing EBV-associated malignancies .
The EBV-negative B cell lines Mutu ( a gift from Jeff Sample ) , BJAB ( purchased from ATCC ) , and Akata ( a gift from Kenzo Takada ) , and EBV-positive Raji ( ATCC ) and Kem III ( a gift from Alan Rickinson and Jeff Sample ) were grown in RPMI-1640 media ( Gibco ) . All media was supplemented with 10–15% FBS and 1% penicillin-streptomycin ( pen-strep ) . The epithelial line HEK 293 ( from ATCC ) was maintained in DMEM ( Gibco ) . Mutu cells infected with the EBV p2089 Bacmid ( B95 . 8 ) were maintained under selection of 300ug/mL Hygromycin B . Primary human peripheral CD19+ B cells from healthy donors ( obtained from Stem Cell Technologies ( #70033 ) , who used Institutional Review Board ( IRB ) -approved consent forms and protocols ) were EBV-transformed and grown in RPMI . The NOKs cell line ( a gift from Karl Munger ) is a telomerase-immortalized normal oral keratinocyte cell line that was established and maintained as previously described [84] . Cells were treated with the following drugs for experiments: ionomycin ( Calbiochem ) at 2 . 5μg/mL , TPA ( Sigma ) at 20ng/mL , cyclosporine A ( Cell Signaling ) at 1μM , sodium butyrate ( Sigma ) at 3mM , anti-IgM ( Southern Biotech ) at 10μg/mL , anti-IgG ( Sigma I5260 ) at 10μg/mL . When cyclosporine was added to cells , it was done one hour prior to addition of any other drugs . Plasmid DNA was prepared using the Qiagen Maxiprep kit according to the manufacturer’s instructions . Plasmid pSG5 was obtained from Stratagene . SG5-LMP2A was a gift from Nancy Raab-Traub . pREP-NFAT2 ( NFATc1 ) was a gift from Anjana Rao ( Addgene plasmid # 11788 ) . pLX304-FOS-V5 was a gift from William Hahn ( Addgene plasmid # 59140 ) [85] . The promoterless luciferase reporter gene construct pCpGL-basic ( a gift from Michael Rehli ) was constructed as previously described [86] and contains no CpG motifs in the entire vector . The BZLF1 promoter ( -668 to +15 , relative to transcription start site ) was PCR amplified from the EBV B95 . 8 ( Zp-P ) and M81 ( Zp-V3 ) genomes and cloned upstream of the luciferase gene in pCpGL-basic using the SpeI and BglII restriction sites . The primer sequences are as follows: BZLF1 F SpeI 5’-GCGACTAGTAGGTGTGTCAGCCAAAG and BZLF1 R BglII 5’- GCGAGATCTCCGGCAAGGTGCAATG . Zp mutants in the pCpGL luciferase vector were constructed using the Stratagene QuikChange II XL site-directed mutagenesis kit . Primers are as follows: Zp-V3–100 5’- CTAATGTGCCTCATAGACACACCTAAATTTAGCACGTCC and 5’- GGACGTGCTAAATTTAGGTGTGTCTATGAGGCACATTAG; Zp-V3–106 5’- ACAGGCATTGCTAATGTACCTCAGAGACACACCTA and 5’- TAGGTGTGTCTCTGAGGTACATTAGCAATGCCTGT; Zp-V3–141 5’- CTGCCTCCTCCTCTTTTAGAAACTATGCATGAGCC and 5’- GGCTCATGCATAGTTTCTAAAAGAGGAGGAGGCAG; Zp-V3–274 5’- CTCCCCCCTGACCCCCGAACTTAATGAAATCTTGGA and 5’- TCCAAGATTTCATTAAGTTCGGGGGTCAGGGGGGAG; Zp-V3–365 5’- AGATGGACCTGAGCCACCCGCCCCC and 5’- GGGGGCGGGTGGCTCAGGTCCATCT; Zp-V3–460 5’- GGAGGACCCTGATGAAGAAACCAGTCAGGCC and 5’- GGCCTGACTGGTTTCTTCATCAGGGTCCTCC; Zp-V3–525 5’- CGGTGCCCCAGCCACTTGACCCGG and 5’- CCGGGTCAAGTGGCTGGGGCACCG; Zp-P -141 5’- CTGCCTCCTCCTCTTTTGGAAACTATGCATGAGCC and 5’- GGCTCATGCATAGTTTCCAAAAGAGGAGGAGGCAG . The ZIIIA/AP1 Zp-V3 mutant was mutated sequentially , first with Zp-V3 Ap1 mut 1 5’-TTGGAAACTATGCAGAAGCCACAGGCATTGCTAATGTGCCT and 5’-AGGCACATTAGCAATGCCTGTGGCTTCTGCATAGTTTCCAA , then with Zp-V3 AP1 mut 2 5’-TTGGAAACTATGCAGAATTCACAGGCATTGCTAATGTGCCT and 5’-AGGCACATTAGCAATGCCTGTGAATTCTGCATAGTTTCCAA . All constructs were verified by sequencing . The EBV p2089 Bacmid was a gift from Henri-Jacques Delecluse and contains the complete genome of the B95 . 8 strain of EBV in addition to a cassette containing the prokaryotic F-factor as well as the green fluorescent protein ( GFP ) and Hygromycin B resistance genes in the B95 . 8 deletion as previously described [87] . p2089 is the parental WT Bacmid to all mutants in this study . EBV Zp-P-141G Bacmid was constructed using the GS1783 E . coli–based En Passant method previously described [88] to change the -141 nucleotide in the B95 . 8 Zp to Variant 3 Zp sequence . Subsequently a revertant Bacmid , designated EBV Zp-P-141G . REV was constructed , reverting the altered nucleotide to wildtype B95 . 8 sequence . Finally , the Chloramphenicol cassette in the F-factor of each WT , Zp-P-141G , and Zp-P-141G . REV Bacmids was replaced with Kanamycin . Kanamycin resistance facilitated the transfer of all Bacmids to the Chloramphenicol-resistant BM2710 E . coli [89] used for infection of 293 cells . The integrity of each Bacmid was confirmed by analyzing the restriction digestion patterns with multiple enzymes . Furthermore , all mutations were confirmed by high fidelity PCR amplification and sequencing of the mutated junctions . The list of primers used for generation and confirmation of all mutants is as follows . Zp . T-141C Primer 1 5’- TGAGGTACATTAGCAATGCCTGTGGCTCATGCATAGTTTCCAAAAGAGGAGGAGGCAGTTTTAGGGATAACAGGGTAATCGATTT , Zp . T-141C Primer 2 5’-CTTATTTTAGACACTTCTGAAAACTGCCTCCTCCTCTTTTGGAAACTATGCATGAGCCACAGCCAGTGTTACAACCAATTAACC , Zp . T-141C . REV Primer 1 5’-TGAGGTACATTAGCAATGCCTGTGGCTCATGCATAGTTTCTAAAAGAGGAGGAGGCAGTTTTAGGGATAACAGGGTAATCGATTT , Zp . T-141C . REV Primer 2 5’-CTTATTTTAGACACTTCTGAAAACTGCCTCCTCCTCTTTTAGAAACTATGCATGAGCCACAGCCAGTGTTACAACCAATTAACC , Cam-Ff-Kan Primer 1 5’-CGGGCGTATTTTTTGAGTTATCGAGATTTTCAGGAGCTAAGGAAGCTAAAATGAGCCATATTCAACGGGAAAC , Cam-Ff-Kan Primer 2 5’-CAGGCGTAGCAACCAGGCGTTTAAGGGCACCAATAACTGCCTTAAAAAAATTAGAAAAACTCATCGAGCATC , Zp . -141-Confirm Primer 1 5’-CGGCAAGGTGCAATGTTTAG , and Zp . -141-Confirm Primer 2 5’-GTGTCAGCCAAAGAGGATCA . BJAB cells were nucleofected using the Amaxa Nucleofector 2b device ( Lonza ) and program M-013 ( with Buffer V ) in 12-well dishes with 500 ng of pCpGL-basic promoter construct and 500 ng of vector control , LMP2A , cFos , or NFATc1 plasmid . NOKs were transfected with Lipofectamine 2000 ( Thermo-Fisher Scientific ) . The cells were washed with PBS and harvested in 1× Reporter Lysis Buffer ( Promega ) at 24–48 h post-nucleofection or transfection . Lysates were subjected to three freeze-thaw cycles , and relative luciferase units were quantified with a BD Monolight 3010 luminometer ( BD Biosciences ) using Promega luciferase assay reagent . All luciferase assay figures represent two independent experiments , each performed in duplicate . BJAB cells were treated or mock-treated with anti-IgM for 30 minutes or 6 hours and then harvested . The cell pellet was resuspended in 100uL hypotonic buffer A ( 10mM HEPES-K+ pH7 . 9 , 10mM KCl , 1 . 5mM MgCl2 , 0 . 5mM DTT ) in the presence of protease inhibitor cocktail ( PIC , Roche ) and phosphatase inhibitor cocktail II ( Calbiochem ) , then incubated on ice for 10 min with vortexing . 1uL of 10% NP-40 was added and samples were vortexed to assist in the lysis for up to 1 min . The nuclei were centrifuged at 14 , 000 RPM for 5 min at 4°C . The nuclear pellets were resuspended in 50uL buffer C ( 20mM HEPES-K+ pH7 . 9 , 420mM NaCl , 0 . 2mM EDTA , 1 . 5mM MgCl2 , 0 . 5mM DTT , 25% Glycerol ) with PIC . Nuclei were incubated on ice for 40 min , and vortexed periodically . Supernatant containing nuclear protein was collected by centrifuging at 14 , 000 RPM for 10 min at 4°C and then aliquoted and snap frozen for use in EMSAs . EMSAs were performed as previously described [90 , 91] . Consensus binding probes ( oligonucleotides ) for AP1 ( #sc-2501 ) , Ets ( #sc-2549 ) , and NFAT ( 2 ) ( #sc-2577 ) were obtained from Santa Cruz . All other probes were custom designed and ordered from IDT . Their sequences are as follows: NFAT ( 1 ) consensus EMSA 5’- AGAAAGGAGGAAAAACTGTTTCATACAGAAGGCGTT and 5’- AACGCCTTCTGTATGAAACAGTTTTTCCTCCTTTCT; ELK1 consensus EMSA 5’- GGGGTCCTTGAGGAAGTATAAGAAGAAT and 5’- ATTCTTCTTATACTTCCTCAAGGACCCC; Zp-P -155 to -127 EMSA 5’-CCTCCTCCTCTTTTAGAAACTATGCATGA and 5’- TCATGCATAGTTTCTAAAAGAGGAGGAGG; Zp-V3–155 to -127 EMSA 5’-CCTCCTCCTCTTTTGGAAACTATGCATGA and 5’- TCATGCATAGTTTCCAAAAGAGGAGGAGG; Zp-P -155 to -120 EMSA 5’-CCTCCTCCTCTTTTAGAAACTATGCATGAGCCACAG and 5’- CTGTGGCTCATGCATAGTTTCTAAAAGAGGAGGAGG; Zp-V3–155 to -120 EMSA 5’-CCTCCTCCTCTTTTGGAAACTATGCATGAGCCACAG and 5’- CTGTGGCTCATGCATAGTTTCCAAAAGAGGAGGAGG . EMSAs were performed with binding buffer ( 50 mM KCl , 25 mM Hepes ( pH 7 . 6 ) , 10% glycerol , 1 mM EDTA , 0 . 5 mM spermidine , 0 . 5 mM PMSF , and 1 mM DTT ) with 2 μg of poly ( dI/dC ) :poly ( dI/dC ) ( Pharmacia ) and 2ug BJAB nuclear extract . The protein and binding buffer mixture was allowed to incubate for 5 min at room temperature , and then 20 , 000 cpm of γ-32P ATP labeled probe were added . The mixture containing labeled probes was allowed to incubate for an additional 20 min . For supershift conditions 1-2ug anti-NFATc1 ( Santa Cruz #sc-13033x ) , anti-XBP1 ( Santa Cruz #sc-7160x ) , anti-cFos ( Santa Cruz #sc-52x ) , or anti-C/EBPα ( Santa Cruz #sc-61x ) antibodies were added to the protein before addition of radiolabeled probe and allowed to incubate for 20 minutes . For cold competitor conditions 10X excess unlabeled probe was added to the protein before addition of radiolabeled probe and allowed to incubate for 20 minutes . Immunoblotting was performed as previously described [52] . The following primary antibodies were used: anti-EBNA1 ( Santa Cruz #sc-81581 ) , anti-EBNA2 ( Abcam #ab90543 ) , anti-LMP1 ( Abcam #ab78113 ) , anti-β-actin ( Sigma #A5441 ) , anti-GAPDH ( Cell Signaling Technology #D16H11 ) , anti-BMRF1 ( Millipore #MAB8186 ) , anti-p18 ( Thermo Scientific #PA1-73003 ) , anti-BZLF1 ( Santa Cruz #sc-53904 ) , anti-R rabbit polyclonal antibody directed against the R peptide ( peptide sequence EDPDEETSQAVKALREMAD ) , anti-NFATc1 ( Santa Cruz sc-17834 ) and anti-tubulin ( Sigma T5168 ) . The secondary antibodies used were horseradish peroxidase ( HRP ) –goat anti-mouse ( Thermo Scientific #31430 ) and donkey anti-goat ( Santa Cruz #sc-2056 ) . Image Studio Lite software was used to quantify levels of Z and R relative to loading control tubulin in Fig 7B . EBV-positive 293 WT , Zp-P-141G and Zp-P-141G . REV were derived using the BM2710 E . coli , which can mediate the transfer of intact recombinant DNA into mammalian cells due to expression of the invasin gene from Yersinia pseudotuberculosis and the listeriolysin O gene from Listeria monocytogenes [89] . Briefly , Bacmids were electroporated using a 0 . 1 cm gap cuvette ( 1 . 5 kV , 200 Ohms , 25 μF ) into BM2710 E . coli and selected with Kanamycin and Spectinomycin . BM2710 E . coli containing the respective Bacmid were used to infect EBV-negative 293 cells by co-incubation for 2 hours ( approximately 25 bacteria per cell ) . Cell lines were derived by single-cell cloning and screened for ability to complete the lytic cascade by immunoblotting for viral late protein VCAp18 ( product of EBV BFRF3 ) and titering cell-free virus on Raji cells . Cells were selected and maintained with 100–200μg/ml of Hygromycin B . Infectious viral particles were produced from 293 cell lines stably infected with the wildtype or mutant B95 . 8 viruses as previously described [92] . To determine the titer of the virus , Raji cells were infected with serial 10-fold dilutions of virus . After 24 hours , cells were treated with 50 ng/ml TPA and 3 mM sodium butyrate , and the number of GFP-expressing Raji cells was counted 24–48 hours later by fluorescence microscopy . B cells were centrifuged and resuspended in media containing wildtype , Zp -141 mutant , or revertant B95 . 8 virus ( produced by 293 cell lines and titered on Raji cells ) for a total volume of 500uL and an MOI of 0 . 1 or 0 . 25 ( primary B cells ) or 1 ( BJAB , Akata and Mutu cells ) . Cells and virus were incubated for 1–3 h with occasional stirring , then media was increased to 4 mL for overnight incubation . The next day cells were spun down and resuspended in fresh media , except for human peripheral blood CD19+ B cells , which were not centrifuged . Peripheral blood CD19+ B cells and Akata cells infected with EBV were harvested at day 3 post-infection , and BJAB cells at 6 days after infection , and extracts containing equal amounts of protein used for immunoblots . Mutu cells were selected with 300 ug/mL hygromycin B starting at day five post-infection to create stable cell lines; Mutu lines were under selection for at least two months before other experiments were performed . To determine the number of EBV-infected CD19+ B cells , and the mean fluorescence intensity , a portion of cells was analyzed with a LSRII flow cytometer ( BD Biosciences ) three days post-infection . Data analysis was performed using FlowJo software . EBV-infected Mutu cells were nucleofected with 120 pmol control siRNA ( Santa Cruz sc-37007 ) or NFATc1 siRNA ( 29412 ) using Amaxa program N-16 in Buffer V . Ionomycin or DMSO control was added after 48 hours . Cells were harvested 72 hours post-nucleofection and immunoblots performed using equal amounts of protein . EBV-infected Mutu cells treated with ionomycin for 3hrs were harvested and fixed with 1% formaldehyde in PBS for 8 minutes at room temperature followed by addition of glycine to 125mM for 5 minutes at room temperature to quench the reaction . Fixed cells were pelleted , washed once with PBS and twice with Cell Lysis/Wash Buffer ( 150mM NaCl , 50mM Tris pH 7 . 4 , 5mM EDTA pH 8 . 0 , 0 . 5% NP-40 , 1% Triton X-100 ) . Pellets were resuspended in ChIP Lysis Buffer ( 50mM Tris pH 8 . 0 , 10mM EDTA pH 8 . 0 , 1% SDS ) and chromatin was sheared by sonication using a QSonica Q700 sonicator ( 3 rounds of 10 cycles of 30sec on/30sec off at 95% amplitude in an ice water bath ) . Debris was cleared by centrifugation at 11 , 500 x g for 10 min at 4°C . Supernatant was then diluted 1:5 in ChIP Dilution Buffer ( 16 . 7mM Tris pH 8 . 0 , 167mM NaCl , 1 . 2mM EDTA , 1 . 1% Triton X-100 , 0 . 01% SDS ) and chromatin from approximately one million cells was incubated with 3ug of rabbit anti-NFATc1 antibody ( Bethyl Laboratories A303-508A ) or rabbit IgG control antibody ( Millipore 12–370 ) overnight at 4°C . Chromatin/antibody complexes were isolated with Magna ChIP Protein A+G magnetic beads ( Millipore 16–663 ) and subsequently washed with Low Salt Buffer ( 20mM Tris pH 8 . 0 , 150mM NaCl , 2mM EDTA , 1% Triton X-100 , 0 . 1% SDS ) , High Salt Buffer ( 20mM Tris pH 8 . 0 , 0 . 5M NaCl , 1% Triton X-100 , 0 . 1% SDS ) , LiCl Buffer ( 10mM Tris pH 8 . 0 , 0 . 25M LiCl , 1mM EDTA pH 8 . 0 , 1% NP-40 , 1% DOC ) , and TE ( 10mM Tris pH 8 . 0 , 1mM EDTA pH 8 . 0 ) . Crosslinks were reversed and DNA was isolated with an IBI Gel/PCR DNA fragment extraction kit ( IB47030; IBI ) and quantitated by qPCR using iTaq Universal SYBR Green Supermix ( 172–5124; Bio-Rad ) and primers to the Z promoter ( FWD 5′-GCCATGCATATTTCAACTGGGCTG-3′ and REV 5′-TGCCTGTGGCTCATGCATAGTTTC-3′ ) and analyzed using an ABI 7900HT real-time PCR system with SDS2 . 4 software ( Applied Biosystems ) . Transformation titration assays were performed by infecting human peripheral blood CD19+ B cells ( 10 , 000 cells/well in a 96-well microtiter plate ) with 0 . 25 infectious GFP Raji Units/cell ( 10 replicates per virus ) and culturing with RPMI complete medium . Wells with clearly growing LCLs ( lymphoblastoid cell lines ) were scored microscopically after at least 3 weeks of culture . A portion of the Zp sequences analyzed in this paper ( many of which were also previously analyzed for Zp type ) were derived from publicly available sequences of EBV genomes deposited in Genbank . Z promoter sequences that were not previously analyzed for the type of Zp variant present are aligned in S6 Table . Zp promoters were considered to have Zp-V3 if they contained the Zp-V3 specific nucleotide located at position -141 . In addition to Genbank EBV sequences , the TCGA database was interrogated for EBV-positive gastric cancers ( stomach adenocarcinoma or STAD ) by using the Genomic Data Commons Application program interface ( GDC-API ) to perform BAM slicing on harmonized TCGA data . Reads mapping to "chrEBV" were sliced and Samtools was used to quantify the number of reads with MAPQ of 20 or greater . Tumor EBV read count was a bimodal distribution with 27 tumors having 2 , 888–53 , 779 reads and 379 having 0–217 reads and one tumor with 534 reads . All tumors with >500 reads were treated as EBV-positive . Zp variant calling was performed on all 28 EBV-positive samples obtained from the TCGA database using any informative reads that could be obtained from RNAseq and whole exome sequencing ( WXS ) data for each tumor . The GDC-API was used to perform BAM slicing on GDC harmonized TCGA data from chrEBV in the region 91006–47 . Zp-P or Zp-V3 calls were made based on the following position: nt91006 "-100" Zp-P = A Zp-v3 = C; nt91012 "-106" Zp-P = T Zp-v3 = C; nt91047 "-141" Zp-P = T Zp-V3 = C . For 5 samples for which neither RNAseq nor WXS provided informative coverage of the Z promoter , whole genome sequencing data was obtained from the GDC legacy archive . Reads mapping to NC_007605 in the BAM files were manually reviewed in IGV 2 . 3 . 82 and Zp-P or Zp-V3 calls were made as above . To estimate the background prevalence of each Zp genotype in the TCGA database , blood and normal tissue WXS ( whole exome sequencing ) datasets from TCGA were interrogated using the GDC-API to perform BAM slicing for reads mapping to "chrEBV . " Samtools was used to identify samples containing reads from chrEBV: 91006–91047 with MAPQ of 20 or greater . These BAM files were then manually viewed in IGV 2 . 3 . 82 and Zp-P or Zp-V3 calls were made based as for tumor tissues . Additional criteria included requiring base quality of 10 or greater at the call position and the mate read had to also map to chrEBV . Available RNA-seq data from endemic Burkitt tumors were also downloaded from sequence read archive ( SRA ) PRJNA292327 [58] . Fastq files were then aligned to the type 1 and 2 EBV genomes ( AJ507799 . 2 and NC_009334 . 1 ) using BWA’s backtrack algorithm with default settings . Zp alignment for TCGA samples was performed as follows . BAM slices from GDC for each sample were evaluated in IGV v2 . 3 . 82 . Consensus quality scores were calculated by summing individual Phred scores when reads had different mates . Bases with consensus quality scores less than 20 or without read coverage were represented with a ( - ) . Bases with a consensus quality score of 20 or more were represented with an uppercase letter . LCLs were transformed by infectious EBV particles present in the breast milk of Kenyan women that had malaria during pregnancy living in a rural region of Kisumu County , as previously described [17] . DNA was isolated from these LCLs and the EBV Zp amplified by PCR using the following primers: Zp+34 primer 5’-GCAAAGATAGCAAAGGTGGC and Zp-561 primer 5’-GAACCGGTCGGATCCCTAAC . Sequencing of the Zp promoter was performed using the Zp +34 primer . The program Mstat , Version 6 . 1 , was used for all statistical analyses ( N . Drinkwater , McArdle Laboratory for Cancer Research , School of Medicine and Public Health , University of Wisconsin ) and is available for downloading ( http://www . mcardle . wisc . edu/mstat ) . The primary human peripheral CD19+ B cells from healthy donors used in this study are considered exempt by the University of Wisconsin-Madison Institutional Review Board ( IRB ) . These cells were purchased from Stem Cell Technologies ( #70033 ) , which obtained written donor consent using IRB-approved protocols . The anonymized EBV breast milk samples used for this study have been previously documented [17] . The original study received approval from the Kenya Medical Research Institute ( KEMRI ) , the University of Colorado COMIRB , and Upstate Medical University ( where R . Rochford was at initiation of study ) Review Boards . Written informed consent was obtained from all study participants before any sample collection . | Whether excessive lytic EBV infection increases the risk of EBV-induced cancers is not clear . A particular variant ( Zp-V3 ) of the viral promoter driving expression of the EBV immediate-early BZLF1 ( Z ) protein that mediates lytic viral reactivation has been reported to be over-represented ( relative to the prototype Zp-P form of the promoter ) in certain EBV-positive malignancies , but no functional difference between the two promoter variants has been reported . Here we show that the malignancy-associated Zp-V3 variant ( but not the Zp-P variant ) contains a binding site for the cellular NFATc1 ( nuclear factor of activated T cells c1 ) transcription factor that allows it to be activated by NFATc1-inducing stimuli such as B-cell receptor stimulation . Furthermore , we demonstrate that restoring this NFATc1-motif to the Zp-P variant in the context of the intact EBV genome greatly enhances lytic viral reactivation in response to the NFATc1-inducing stimuli . We also find that the Zp-V3 variant is over-represented in EBV-positive Burkitt lymphomas and gastric carcinomas , and in lymphoblastoid cell lines transformed by EBV-infected breast milk of Kenyan mothers that had malaria during pregnancy . These findings suggest that the Zp-V3 version of the EBV BZLF1 promoter increases the likelihood of EBV-induced malignancies by increasing lytic EBV infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"luciferase",
"milk",
"immune",
"cells",
"breast",
"milk",
"pathology",
"and",
"laboratory",
"medicine",
"body",
"fluids",
"enzymes",
"pathogens",
"gene",
"regulation",
"immunology",
"carcinomas",
"enzymology",
"cancers",
"and",
"neoplasms",
"microbiology",
"diet",
"parasitic",
"diseases",
"tropical",
"diseases",
"dna-binding",
"proteins",
"oncology",
"hematologic",
"cancers",
"and",
"related",
"disorders",
"viruses",
"nutrition",
"regulatory",
"proteins",
"dna",
"viruses",
"transcription",
"factors",
"lymphomas",
"herpesviruses",
"white",
"blood",
"cells",
"epstein-barr",
"virus",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"oxidoreductases",
"burkitt's",
"lymphoma",
"microbial",
"pathogens",
"gene",
"expression",
"hematology",
"antibody-producing",
"cells",
"biochemistry",
"cell",
"biology",
"b",
"cells",
"anatomy",
"viral",
"pathogens",
"beverages",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"malaria",
"organisms"
] | 2018 | A cancer-associated Epstein-Barr virus BZLF1 promoter variant enhances lytic infection |
In many animal embryos , a specific gene expression pattern is established along the animal-vegetal axis soon after zygotic transcription begins . In the embryo of the ascidian Ciona intestinalis , soon after the division that separates animal and vegetal hemispheres into distinct blastomeres , maternal Gata . a and β-catenin activate specific genes in the animal and vegetal blastomeres , respectively . On the basis of these initial distinct gene expression patterns , gene regulatory networks promote animal cells to become ectodermal tissues and vegetal cells to become endomesodermal tissues and a part of the nerve cord . In the vegetal hemisphere , β-catenin directly activates Foxd , an essential transcription factor gene for specifying endomesodermal fates . In the present study , we found that Foxd also represses the expression of genes that are activated specifically in the animal hemisphere , including Dmrt1 , Prdm1-r . a ( Bz1 ) , Prdm1-r . b ( Bz2 ) , and Otx . A reporter assay showed that Dmrt1 expression was directly repressed by Foxd , and a chromatin immunoprecipitation assay showed that Foxd was bound to the upstream regions of Dmrt1 , Prdm1-r . a , Prdm1-r . b , and Otx . Thus , Foxd has a dual function of activating specific gene expression in the vegetal hemisphere and of repressing the expression of genes that are normally expressed in the animal hemisphere . This dual function stabilizes the initial patterning along the animal-vegetal axis by β-catenin and Gata . a .
In many animal embryos , localized maternal factors create differential gene expression patterns along the animal-vegetal axis [1–3] , and the subsequent developmental program proceeds on the basis of this initial patterning . In ascidian unfertilized eggs , several identified and unidentified maternal factors are unequally distributed along the animal-vegetal axis [4] . At the 8-cell stage , the animal and vegetal hemispheres become separated into distinct blastomeres , and the difference along the animal-vegetal axis is clearly established; when blastomeres are experimentally isolated at the 8-cell stage , endomesodermal cells are differentiated from vegetal hemisphere cells [5–8] , and epidermal cells are differentiated from animal cells [9] . In 16-cell embryos of the ascidian Ciona intestinalis , the maternal transcription factor Gata . a activates Ephrina . d and Tfap2-r . b specifically in the animal hemisphere , and a complex of β-catenin and Tcf7 activates Foxd and Fgf9/16/20 in the vegetal hemisphere [10–15] . In the vegetal hemisphere , β-catenin/Tcf7 weakens the Gata . a-binding activity for target sites through a physical interaction , and thereby the animal hemisphere genes are not expressed in the vegetal hemisphere at the 16-cell stage [15] . In this manner , the initial difference between the animal and vegetal hemispheres is set up . Foxd and Fgf9/16/20 , which are activated by β-catenin/Tcf7 , encode a transcription factor and a signaling molecule , respectively . These molecules are required for expression of endodermal and mesodermal genes including Lhx3/4 , Zic-r . b ( ZicL ) , and Brachyury in the vegetal hemisphere [16–18] . In addition , Fgf9/16/20 signaling also induces expression of neural genes including Dmrt1 , Otx , Prdm1-r . a and Prdm1-r . b in the neural lineage of the animal hemisphere [11 , 19–22] . Animal hemisphere cells that are not induced by Fgf9/16/20 signaling give rise to epidermal cells under the control of Tfap2-r . b , which encodes a transcription factor [23] . Thus , the difference between the animal and vegetal hemispheres are critically important for subsequent developmental programs . However , the initial difference between the animal and vegetal hemispheres , which is established by Gata . a and β-catenin/Tcf7 , may not be sufficient for explaining differential gene expression patterns between them at the 32-cell stage and thereafter , because two animal hemisphere genes Dmrt1 and Dlx . b are expressed ectopically in the vegetal hemisphere of Foxd morphants at the early gastrula stage [19] . Dmrt1 is important for anterior neural and palp ( a placode-like structure ) fate specification [19 , 24] , and Dlx . b is important for neural and epidermal fate specification [23] . In the present study , we examined how the animal-vegetal axis is maintained at the 32-cell stage and thereafter , and showed that Foxd acts as a robust binary switch to stabilize the initial patterning along the animal-vegetal axis by Gata . a and β-catenin/Tcf7 .
Foxd is expressed under the direct control of β-catenin/Tcf7 in three vegetal cell pairs ( A5 . 1 , A5 . 2 , and B5 . 1 ) of 16-cell embryos ( Fig 1 ) . After the next division , among their daughter cells , cells with endodermal fate continue to express Foxd ( A6 . 1 , A6 . 3 , and B6 . 1 ) , and the expression becomes undetectable at the 64-cell stage . To identify genes regulated by Foxd in early embryos , we performed RNA-seq analysis at the 32-cell , 64-cell , and 112-cell stages to compare transcriptomes between unperturbed and Foxd knocked-down embryos . For Foxd knockdown , we used a morpholino oligonucleotide ( MO ) against Foxd . We picked up genes encoding transcription factors and signaling molecules that are known to be expressed zygotically between the 32-cell and 112-cell stages [25] , and compared expression levels between unperturbed and Foxd morphant embryos ( Fig 2 ) . We did not utilize biological replicates because we used these data for screening purposes and because we performed this analysis at three successive time points . Fourteen genes were identified to be differentially expressed at one or more stages by a computer program called NOISeq [26] ( > 80% , probability of differential expression by NOIseq-sim , which simulates technical replicates ) . Among them , nine genes were previously known to be regulated by Foxd: Zic-r . b ( ZicL ) , Brachyury , Fgf8/17/18 , Fgf9/16/20 , Foxb , Lhx3/4 , and Mnx were known to be positively regulated by Foxd , and Dmrt1 and Foxd itself are known to be negatively regulated [16 , 19 , 27] . These observations indicate that the RNA-seq experiments successfully identified genes under the control of Foxd . In addition to these nine previously characterized genes , there were five differentially expressed regulatory genes identified: Foxa . a was downregulated at the 64-cell stage , Nkx2-1 ( Ttf1 ) was downregulated at the 64-cell and 112-cell stages , Otx was upregulated at the 32-cell stage , and Prdm1-r . a ( Bz1 ) and Prdm1-r . b ( Bz2 ) were upregulated at the 32- and 64-cell stages in Foxd morphants . These genes were candidates for Foxd targets that had not yet been identified . To confirm downregulation of Foxa . a and Nkx2-1 in Foxd morphant embryos , we performed in situ hybridization . Foxa . a was normally expressed strongly in the vegetal blastomeres designated A7 . 1 , A7 . 2 , A7 . 3 , A7 . 5 , A7 . 7 , B7 . 1 , and B7 . 2 , and weakly in B7 . 3 at the 64-cell stage ( Fig 3A and 3B ) . Foxa . a expression was lost only in A7 . 3 , A7 . 7 , and B7 . 3 in Ciona Foxd morphants ( Fig 3C ) . Foxa . a expression begins at the 8-cell stage , and our data did not indicate downregulation of Foxa . a at the 32-cell stage ( Fig 2A ) , which is consistent with a recent study [16] . Nkx2-1 was normally expressed in the vegetal blastomeres designated A7 . 1 , A7 . 2 , A7 . 5 , B7 . 1 , and B7 . 2 at the 64-cell stage ( Fig 3D ) , whereas it was not expressed in Foxd morphants ( Fig 3E ) , as recently shown at the early gastrula stage [16] . Thus , Foxd positively regulated Foxa . a and Nkx2-1 . In addition to Foxa . a and Nkx2-1 , the genes Brachyury , Fgf8/17/18 , Fgf9/16/20 , Foxb , Mnx , and Zic-r . b , which were found to be positively regulated by Foxd ( Fig 2 ) , are all expressed in the vegetal hemisphere [17 , 25 , 27 , 28] . Namely , genes that were identified to be positively regulated by Foxd in early embryos were all expressed in the vegetal hemisphere . Prdm1-r . a , Prdm1-r . b , Foxd , Dmrt1 , and Otx were found to be negatively regulated by Foxd ( Fig 2 ) . While Prdm1-r . a and Prdm1-r . b are normally expressed in five pairs of animal cells ( a6 . 5 to a6 . 8 and b6 . 5 ) and a pair of vegetal cells ( B6 . 4 ) at the 32-cell stage [29] , these two genes were ectopically expressed in vegetal cells of Foxd morphants ( A6 . 1 to A6 . 4 , B6 . 1 , and B6 . 2 ) ( Fig 4A–4E ) . Foxd expression was examined in Foxd morphants ( S1A and S1B Fig ) . Foxd mRNA was detected in Foxd morphants at the 64-cell stage , while it was rarely detected in normal 64-cell embryos . This might suggest that Foxd negatively regulates itself , or alternatively , that Foxd mRNA was stabilized by binding the MO . To discriminate between these possibilities , we injected synthetic Foxd mRNA into Ciona eggs . Because the synthetic mRNA lacked the endogenous 3’-UTR , we were able to measure the amount of the endogenous Foxd mRNA by RT-qPCR with primers designed to its 3’-UTR . While levels of the maternal control mRNA Pou2 were unchanged , Foxd mRNA levels were greatly reduced by injection of synthetic Foxd mRNA ( S1C Fig ) . Therefore , Foxd indeed regulates itself negatively . We previously showed that Dmrt1 is expressed at the 64- and 112-cell stages in the anterior neural lineage of the animal hemisphere [25] , and the RNA-seq result of the present study suggested that this gene was expressed in 32-cell embryos under the control of Foxd . Indeed , upon careful re-examination , we detected a weak signal in the anterior animal cells ( a6 . 5 ) at the 32-cell stage of normal embryos . This expression pattern was expanded to the anterior vegetal cells ( A6 . 2 and A6 . 4 ) of Foxd morphants ( Fig 4F and 4G ) . Consistently , injection of Foxd mRNA reduced Dmrt1 expression ( S1C Fig ) . Otx is expressed in three pairs of vegetal cells ( B6 . 1 , B6 . 2 , and B6 . 4 ) and two pairs of animal cells ( a6 . 5 and b6 . 5 ) at the 32-cell stage in normal embryos [20] . This gene was expressed ectopically in the anterior vegetal cells ( A6 . 1 to A6 . 4 ) of Foxd morphants ( Fig 4H and 4I ) . Otx and Dmrt1 are activated by Fgf signaling [19 , 20] , and Fgf9/16/20 is downregulated at later stages in Foxd morphants [19] , which was consistent with the RNA-seq result at the 64-cell stage ( Fig 2B ) . On the other hand , Fgf9/16/20 is not downregulated at the 32-cell stage [16 , 17] , which was also consistent with the RNA-seq result at the 32-cell stage ( Fig 2A ) . Indeed , Fgf9/16/20 was not downregulated at the 16-cell stage in Foxd morphants ( S1D and S1E Fig ) . Therefore , it is likely that the earliest expression of Fgf9/16/20 , which is controlled by maternal β-catenin [15] but not by Foxd , induced Otx and Dmrt1 expression , even in Foxd morphants . Because Tfap2-r . b is regulated directly by a maternal factor [15] and expressed in the animal hemisphere at the 16-cell stage [25] , and because expression of Tfap2-r . b was not significantly changed in our RNA-seq experiment ( ~1 . 7 fold-increase ) , we examined the expression of this gene as a negative control . We confirmed by in situ hybridization that the expression of this gene was not affected in Foxd morphants ( S1F and S1G Fig ) . Our results showed that Foxd represses Prdm1-r . a , Prdm1-r . b , Dmrt1 , and Otx expression in vegetal cells at the 32-cell stage , although Otx is expressed in the posterior vegetal cells of normal embryos and Foxd morphants . In addition , Dlx . b , which is expressed in the entire animal hemisphere , is known to be regulated negatively by Foxd [19] , although this gene was not identified to be downstream of Foxd in our RNA-seq experiment ( Fig 2 ) ; this is probably because the number of cells with ectopic Dlx . b expression is much smaller than the number of animal hemisphere cells with Dlx . b expression . Because Prdm1-r . a , Prdm1-r . b , Dmrt1 , Otx , and Dlx . b play essential roles in the specification of epidermal and neural fates [11 , 19 , 20 , 22–24 , 29] , Foxd is likely to suppress ectodermal fates in the vegetal hemisphere . To understand the mechanism by which Foxd negatively regulates ectodermal fates , we analyzed the upstream regulatory sequence of Dmrt1 by introducing lacZ reporter constructs using electroporation . Experimental embryos were fixed at the 32-cell stage , and reporter expression was examined by in situ hybridization . The 924-base pair ( bp ) upstream sequence of Dmrt1 , which was slightly longer than the sequence used in previous studies [29 , 30] , drove reporter expression specifically in anterior neural cells ( a6 . 5 ) at the 32-cell stage ( Fig 5A; S2 Fig ) . A construct containing the 486-bp upstream sequence showed almost the same activity ( Fig 5A and 5B ) . Ectopic expression in vegetal cells was increased in constructs containing 386- , 343- , and 286-bp upstream regions , while expression in the a6 . 5 neural lineage was decreased in the constructs containing 343- , and 286-bp upstream regions . The construct containing the 186-bp upstream sequence did not drive reporter expression . This observation indicated that cis-elements important for expression in the animal hemisphere are present between bases -286 and -386 , and that cis-elements important for repression in the vegetal hemisphere are present between bases -343 and -386 . We searched candidate Foxd binding sites using the Patser program [31] and a position weight matrix for human FOXD2 [32] , which identified one putative Foxd binding site between -343 and -386 ( S3 Fig ) . This site was conserved in the genome of the closely related species Ciona savignyi ( S3 Fig ) . Therefore , we mutated this putative binding site . The mutant upstream sequence drove reporter expression in the vegetal hemisphere ( Fig 5A and 5C ) , suggesting that Foxd directly represses Dmrt1 expression via this site . To confirm if the identified site could bind Foxd , we performed gel-shift assays ( Fig 6 ) . Foxd binding was observed as a shifted band that disappeared upon incubation with a specific competitor ( competitor 1 ) but did not disappear upon incubation with competitors containing a mutation in the putative Fox binding site ( competitors 3 and 4 ) . Because the gel-shift probe contained an additional sequence similar to the Fox binding site ( AACA ) , we tested whether this sequence also bound Foxd . The competitor containing a mutation in this second site ( competitor 2 ) did not compete , suggesting that this site does not bind Foxd efficiently . Finally , we performed a chromatin-immunoprecipitation assay followed by high-throughput DNA sequencing ( ChIP-seq ) to confirm that Foxd bound to the regions containing the above putative Foxd binding site in vivo at the 32-cell stage ( Fig 7 ) . We electroporated an expression construct encoding a Foxd-Gfp fusion protein under the control of the Foxd upstream regulatory sequence , and performed a ChIP assay using 32-cell embryos with an anti-Gfp antibody . Two different computer programs identified 114 and 799 peaks , respectively ( false discovery rate < 0 . 1% ) , of which 63 peaks were common and considered in the subsequent analysis . Because Foxd-Gfp might be overexpressed above physiological levels , it is possible that the above binding interactions were stronger than interactions that would normally occur in normal embryos . However , among 52 , 518 of ‘GTAAACA’ sequences found in the genome , only 9 sites were included in the 63 peaks identified by the ChIP-seq assay , suggesting that Foxd-Gfp does not bind non-specifically to all potential binding sites . As shown in Fig 7A , the upstream region of Dmrt1 around the Fox binding site identified above bound Foxd , suggesting direct regulation of Dmrt1 by Foxd . In addition , we found peak regions in the upstream sequences of Prdm1-r . a , Prdm1-r . b , Otx , and Dlx . b ( Fig 7B–7E ) . All these peak regions contained Foxd binding motifs that were identifiable by the Patser program [31] and a position weight matrix for human FOXD2 [32] , although their scores were less than the scores of Dmrt1 ( S4 Fig ) . Meanwhile , the 63 significant peaks were not found in the upstream regulatory region of Tfap2-r . b , which is not regulated by Foxd as described above , although a weak , insignificant peak was observed ( S5 Fig ) . Therefore , it is conceivable that Dmrt1 , Prdm1-r . a , Prdm1-r . b , Otx , and Dlx . b are direct targets of Foxd .
In ascidian embryos , maternal factors establish differential gene expression patterns between the animal and vegetal hemispheres , which largely correspond to the ectodermal and endomesodermal lineages ( with the exception of part of the nerve cord , which is derived from the vegetal hemisphere ) . Gata . a and β-catenin/Tcf7 activate specific gene expression in these two domains at the 16-cell stage . However , our present results indicated that this segregation between the animal and vegetal hemisphere lineages was not robust enough to maintain this segregation alone . We found that Foxd activity commits vegetal cells to the endomesoderm fate by repressing ectoderm genes including Prdm1-r . a , Prdm1-r . b , Dmrt1 , Otx , and Dlx . b ( Fig 8 ) , although Foxd may not necessarily repress all genes that are expressed in the animal hemisphere . In other words , maternal factors generated a transient regulatory stage , which was maintained by Foxd activity . Thus , animal hemisphere gene expression is suppressed in the vegetal hemisphere continuously during early embryogenesis . First , Gata . a activity is suppressed by β-catenin/Tcf7 in the vegetal hemisphere of the 16-cell embryo [15] , and then Foxd , which is activated by β-catenin/Tcf7 , directly represses animal hemisphere genes in the vegetal hemisphere at the 32-cell stage and thereafter . In Foxd morphants , Prdm1-r . a and Prdm1-r . b were activated ectopically in both the anterior and posterior vegetal cells , while Dmrt1 and Otx were activated ectopically only in the anterior cells . Activators for Dmrt1 might not be present in the posterior vegetal cells . In normal embryos , Dmrt1 is activated only in the anterior neural cells , because Foxa . a , which encodes an activator for Dmrt1 , is not expressed in the posterior neural cells [19 , 25 , 33] . Foxa . a expression indeed begins in the anterior half of the 8-cell embryo , although it is expressed in posterior vegetal cells at the 16-cell stage and thereafter [33 , 34] . Meanwhile , in normal embryos , Otx is expressed in the posterior vegetal cells except the most posterior cells , in addition to the animal neural cells [20] . Different enhancers are responsible for expression in these two regions [11 , 35] . Therefore , even if the neural enhancer of Otx is ectopically activated in these posterior vegetal cells , this ectopic activation cannot be detected by in situ hybridization . Indeed , one of the peak regions in the Otx upstream region partly overlaps the neural enhancer identified in previous studies [11 , 35] ( Fig 7D ) . Activation of Otx in the vegetal hemisphere by different enhancers may explain why we detected differential expression of Otx only at the 32-cell stage by the RNA-seq experiments . In addition to the repressive function shown above , Foxd functions as an activator; it activates Zic-r . b and Lhx3/4 cooperatively with Foxa . a and Fgf9/16/20 [16 , 19 , 27] . A ChIP assay showed that Foxd binds to upstream regions of Foxd-regulated genes at the 64-cell stage [36] . Reporter assays also showed that two Fox-binding sites within the upstream sequence of Zic-r . b , which is activated by Foxd , are essential for its expression [37] . Lhx3/4 is also likely to be a direct target of Foxd , because Lhx3/4 is expressed at the 32-cell stage under the control of Foxd [16] , and because Foxd is bound to the upstream region of Lhx3/4 at the 64-cell stage [36] . Thus , Foxd is a dual-functional protein; it simultaneously promotes endomesodermal fates and inhibits ectodermal fates . It has been proposed that there are sub-circuits responsible for locking down regulatory states [38] . In Ciona early embryos , Foxd maintains the regulatory state of the vegetal hemisphere , and therefore this dual-functional protein may alone work like such a sub-circuit to lock down dynamic states . In Xenopus , FoxD4L1 . 1 has a dual role as a transcriptional activator and repressor in the neural ectoderm; it activates genes that keep cells in a proliferative state and represses genes that promote differentiation [39] . Xenopus FoxD4L1 . 1 is also involved in repressing BMP signaling , thereby suppressing epidermal fate . The activating function is mediated in Xenopus by an acidic domain near the N-terminus and the repressing function at least partly depends on an Engrailed homology region-1 ( Eh-1 ) located in the C-terminal region . Ciona Foxd also contains a putative acidic domain near the N-terminus and an Eh-1 motif in the C-terminal region ( S6 Fig ) . In both Ciona and Xenopus , Foxd acts as a robust binary switch that promotes one fate and suppresses the other fate . This might be an evolutionarily conserved function of Foxd . A previous study identified two critical Fox-binding sites to which Foxd might bind in the upstream region of Zic-r . b [37] . The sequences of these sites are slightly different from the sequence of the Foxd-binding site for Dmrt1 and those found in the peak regions in the upstream regions of Prdm1-r . a , Prdm1-r . b , Otx , and Dlx . b . In the ChIP-seq assay of the present study , we did not find clear binding peaks upstream of Zic-r . b , although our previous ChIP-chip assay using slightly older embryos exhibited peaks [36] . Therefore , the binding sites in the upstream regions of Prdm1-r . a , Prdm1-r . b , Otx , and Dlx . b might be stronger than the binding sites upstream of Zic-r . b . Indeed , at least one Foxd binding motif in each of the peak regions in the upstream regions of Prdm1-r . a , Prdm1-r . b , Otx , and Dlx . b gave a higher score than the Fox binding sites found in Zic-r . b ( S4B Fig ) . Such a qualitative difference might be important for Foxd to work as an activator or a repressor . In the ascidian embryo , Sox1/2/3 , and Gata . a are important for specification of ectodermal fate [11 , 12 , 15 , 23] . Because there are clear Sox and Gata binding motifs in the peak regions of Dmrt1 , Prdm1-r . a , Prdm1-r . b , Otx , and Dlx . b , it is possible that Sox1/2/3 and Gata . a help Foxd to act as a repressor .
Ciona intestinalis ( type A; this type is also called Ciona robusta ) adults were obtained from the National Bio-Resource Project for Ciona . cDNA clones were obtained from our EST clone collection [40] . Whole-mount in situ hybridization was performed as described previously [25] . Identifiers for genes examined in the present study are shown in S1 Table , according to the nomenclature rule proposed in a recent paper [41] . A morpholino oligonucleotide ( MO; Gene Tools , LLC ) for Foxd knock-down is designed to block translation of two paralogous Foxd genes , Foxd . a and Foxd . b ( 5′-GCACACAACACTGCACTGTCATCAT-3′ ) . This MO has been used previously , and its specificity has been evaluated [19 , 25] . The MO was introduced by microinjection under a microscope . The coding sequence of Foxd . b was cloned into pBluscript RN3 [42] , and Foxd mRNA was transcribed using the mMESSAGE mMACHINE T3 Transcription Kit ( Life technologies ) . Reporter constructs were introduced into fertilized eggs by electroporation . Chromosomal positions of the upstream sequences for reporter constructs and the mutated sequence are indicated in S2 Fig . We randomly chose embryos introduced with reporter constructs to examine reporter construct expression by in situ hybridization . We performed all gene knockdown experiments and reporter gene assays at least twice with different batches of embryos . Recombinant Foxd . b protein was produced as a fusion protein of the Foxd DNA-binding domain and glutathione S-transferase in Escherichia coli BL21 star DE3 strain ( Thermo Fisher Scientific ) , and the protein was purified under a native condition using glutathione Sepharose 4B ( GE Healthcare ) . After annealing two complementary oligonucleotides ( 5’-AAATAACAATAATGTTTACGTTGGT-3’ and 5’-AAAACCAACGTAAACATTATTGTTA-3’ ) , both protruding ends of the double-stranded oligonucleotides were filled with biotin-11-dUTP , and this biotin-labelled oligonucleotide was used as a probe . Proteins and the biotin-labeled probe were mixed in 10 mM Tris ( pH 7 . 5 ) , 50 mM KCl , 1 mM DTT , 1 mM EDTA , 50 ng/μL poly ( dIdC ) , 2 . 5% glycerol , and 0 . 05% NP40 with or without competitor double-stranded DNAs ( a 100 fold molar excess ) shown in Fig 6 . Proteins amounts were empirically determined . Protein–DNA complexes were detected using an AP-conjugated anti-biotin antibody ( Roche ) and CDP-star substrate ( Roche ) . For RNA-seq experiments , 50 unperturbed and Foxd-morphant embryos were collected at the 32- , 64- , and 112-cell stages . RNA was extracted using a Dynabeads mRNA DIRECT Purification Kit ( Thermo Fischer Scientific ) and libraries were made with an Ion Total RNA-Seq kit ver 2 ( Thermo Fischer Scientific ) . The libraries were sequenced with an Ion PGM instrument ( Thermo Fischer Scientific ) ( SRA accession number: DRA005206 ) . We did not utilize duplicates because we used this experiment for screening purposes , and the obtained results were confirmed using other methods , as explained in the Results section . NOISeq [26] was used to identify differentially expressed genes . We used a DNA construct encoding GFP-tagged Foxd under the control of the Foxd promoter [36] . Embryos were fixed at the 32-cell stage . The embryos were subjected to ChIP analysis using anti-GFP antibodies , and the immunoprecipitated DNA was amplified by ligation-mediated PCR [36] . Whole cell extract DNA was used as a control . Then , high-throughput DNA sequencing was performed with the Ion PGM instrument ( SRA accession number: DRA005285 ) . To identify peak regions , we used two different programs called Homer [43] with options “-style factor -F 4 -P 0 . 01 -L 4 -localSize 3000” and MACS2 [44] with an option “--nomodel -q 0 . 001" . For RT-qPCR , we extracted RNA from wild-type embryos and embryos injected with Foxd mRNA . The RNA was converted to cDNA by the Cells-to-Ct kit ( Thermo Fisher Scientific ) . The obtained cDNA samples were then analyzed by quantitative PCR with the SYBR green method . For each qPCR , the amount of cDNA used was equivalent to two-thirds of an embryo . Primers used were: Dmrt1 , 5’-CGCTGAACGACAACGAGTCAT-3’ and 5’-TTCGTTTTCCTCTTGTGCTTGTT-3’; Foxd . a , 5’-AGTTTCTTCCCCACAGTTCCAA-3’ and 5’-GGTTTGTTGTATCCGGGATGTT-3’; Foxd . b , 5’-GCAGTACGCATTCCGCAAT-3’ and 5’-CGGAACAAAAACACAAAAGTCAAA-3’; Pou2 , 5’- AAGATGGTTGCTGGATGCTAATAAT-3’ and 5’-TTGGATTGGAGTGGGAATAACAA-3’ . Ciona intestinalis is excluded from legislation regulating scientific research on animals in Japan . Although there is no scientific evidence that Ciona intestinalis can experience pain , discomfort or stress , we made our best efforts to minimize potential harm that Ciona individuals might experience when we obtained eggs and sperm from them . | In embryogenesis of most animals , a specific gene expression pattern is established along the animal-vegetal axis first . In the embryo of the ascidian Ciona intestinalis , the activity of the maternal factor Gata . a is suppressed by β-catenin , which is active only in the vegetal hemisphere , and thereby these two factors activate specific genes in the animal and vegetal blastomeres , respectively . We found that a gene encoding a transcription factor , Foxd , which is a direct target of β-catenin , works as a promoter for endomesodermal fate and an inhibitor for ectodermal fate . In the ascidian embryo , the animal-vegetal axis initially established by the maternal factors is not stable enough for subsequent developmental processes , and needs to be maintained by Foxd . Thus , the animal hemisphere fate is suppressed first by the maternal factor β-catenin , and then by Foxd , which is activated by β-catenin . The primary embryonic axis is not stable initially , and stabilized by a transcription factor , which is expressed differentially along the axis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"in",
"situ",
"hybridization",
"molecular",
"probe",
"techniques",
"gene",
"regulation",
"animals",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"experimental",
"organism",
"systems",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"embryos",
"sea",
"squirts",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"embryology",
"probe",
"hybridization",
"animal",
"cells",
"bioinformatics",
"gene",
"expression",
"molecular",
"biology",
"ciona",
"intestinalis",
"cell",
"biology",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"blastomeres",
"organisms"
] | 2017 | Differential gene expression along the animal-vegetal axis in the ascidian embryo is maintained by a dual functional protein Foxd |
CD8+ cytotoxic T-lymphocytes ( CTLs ) perform a critical role in the immune control of viral infections , including those caused by human immunodeficiency virus type 1 ( HIV-1 ) and hepatitis C virus ( HCV ) . As a result , genetic variation at CTL epitopes is strongly influenced by host-specific selection for either escape from the immune response , or reversion due to the replicative costs of escape mutations in the absence of CTL recognition . Under strong CTL-mediated selection , codon positions within epitopes may immediately “toggle” in response to each host , such that genetic variation in the circulating virus population is shaped by rapid adaptation to immune variation in the host population . However , this hypothesis neglects the substantial genetic variation that accumulates in virus populations within hosts . Here , we evaluate this quantity for a large number of HIV-1– ( n ≥ 3 , 000 ) and HCV-infected patients ( n ≥ 2 , 600 ) by screening bulk RT-PCR sequences for sequencing “mixtures” ( i . e . , ambiguous nucleotides ) , which act as site-specific markers of genetic variation within each host . We find that nonsynonymous mixtures are abundant and significantly associated with codon positions under host-specific CTL selection , which should deplete within-host variation by driving the fixation of the favored variant . Using a simple model , we demonstrate that this apparently contradictory outcome can be explained by the transmission of unfavorable variants to new hosts before they are removed by selection , which occurs more frequently when selection and transmission occur on similar time scales . Consequently , the circulating virus population is shaped by the transmission rate and the disparity in selection intensities for escape or reversion as much as it is shaped by the immune diversity of the host population , with potentially serious implications for vaccine design .
The cellular immune response mediated by CD8+ cytotoxic T-lymphocytes ( CTLs ) performs a critical role in the immune control of human viruses such as human immunodeficiency virus ( HIV-1 ) [1] and hepatitis C virus ( HCV ) [2] . Consequently , the major histocompatibility ( MHC ) loci that encode the human leukocyte antigen ( HLA ) class I molecules , which recognize and bind CTL epitopes in viral proteins , are among the most highly polymorphic genes in the human population [3] . Nevertheless , the CTL response often fails to control the infection completely because of mutations that occur within HLA-restricted CTL epitopes , enabling the virus to escape binding and recognition [4] . Because epitopes are often located in functionally conserved regions of the viral genome , escape mutations may become costly to maintain in the absence of a selective HLA allele [5 , 6] . Thus , when an escape variant is transmitted between HLA-mismatched individuals , competitive growth frequently selects for reversion of the mutation to wild-type , as demonstrated experimentally in simian immunodeficiency virus–infected rhesus macaques [7] and in a comparative study of HIV-1–infected human patients [8] . Consequently , host-specific selection for escape or reversion may play an important role in shaping genetic variation in the circulating virus population [1 , 2 , 5 , 9 , 10] . For instance , population-based analyses of HIV-1 [9] and HCV [11] sequences have found several significant associations between divergent sites within CTL epitopes and the selective HLA alleles in the host population , suggesting that the frequency of escape polymorphisms in the circulating virus population are directly shaped by the immune diversity of the host population . Furthermore , the viral load of HIV-1–infected individuals has been found to be positively correlated with the frequency of the corresponding HLA supertypes in the host population , implying that the total virus population is adapting to the most frequent HLA supertypes [12] . If escape variants are readily transmitted between hosts , then a host with a common HLA supertype is more likely to encounter a virus that has already escaped its immune response [13] , conferring a selective advantage to rare HLA supertypes . However , the virus genotype that becomes transmitted to the next host does not necessarily represent the ultimate outcome of adaptation to the previous host . Escape variants that have been transmitted into a host lacking a selective HLA allele can persist over long periods of time before reversion , or fail to revert at all over the duration of the study [8 , 14] . A delay or absence of reversion may be attributable to weak selection , when the fitness of the escape variant is either intrinsically high , or it has acquired compensatory mutations . To evaluate the role of CTL-mediated selection in shaping the genetic variation of human viruses , we have carried out a large-scale analysis of HIV-1 and HCV protein-coding sequences isolated from human hosts . Previous analyses of clonal HIV-1 subtype B envelope [5 , 15] and protease ( PR ) [16] sequences have shown that across codon positions , genetic variation within hosts is positively correlated with variation among hosts . These correlations suggest that the genetic variation at both levels of the virus population is being shaped by a common set of biological constraints . However , the use of clonal sequences to characterize within-host variation restricted these analyses to small samples of hosts ( n ≤ 12 ) . In addition , quantifying the influence of selection on genetic variation within and among hosts is potentially confounded by variation in mutation rates among codon positions . Because mutation is the ultimate source of all genetic variation , site-specific variation at either level will be roughly proportional to the local mutation rate , which can yield a positive correlation in the absence of selection [17] . Indeed , this effect constitutes the basis for several tests of non-neutral evolution in genetic sequences [18–20] . To address the problem of limited sample size , we exploit “sequencing mixtures” as a site-specific marker of genetic variation within hosts . A sequencing mixture occurs when multiple distinct peaks occur above the same position in a sequencing electropherogram [21]; by convention , mixtures are encoded in sequences by ambiguous nucleotide characters ( International Union of Pure and Applied Chemistry symbols “M” , “R” , “W” , “S” , “Y” , and “K” ) . Because mixtures can indicate the presence of a nucleotide polymorphism in the population , population-based ( or “bulk” ) sequencing is employed to detect minority variants that occur at frequencies above 10%–25% [21–23] . Although population-based sequencing may fail to detect mixtures below this threshold , transient polymorphisms under selection are more likely to be sampled at intermediate frequencies . This application of mixtures is particularly relevant to viruses with extremely high mutation rates such as HIV-1 and HCV , for which population-based sequences are exceedingly abundant . In this study , we use mixtures to quantify the effect of selection on within-host variation in population-based sequences of RT-PCR–amplified viral RNA from blood plasma isolated from over 4 , 000 HIV-1– or HCV-infected patients . To remove the confounding effect of variation in mutation rates , we normalized the nonsynonymous variation per codon position by the synonymous variation , for either level of the virus population . Thus , we calculated the site-specific difference between the frequencies of nonsynonymous ( mN ) and synonymous mixtures ( mS ) , and estimated the analogous difference between the rates of nonsynonymous ( dN ) and synonymous substitution ( dS ) . Our estimates of mN and dN were both scaled by the expected number of nonsynonymous sites at each codon position; likewise , estimates mS and dS were scaled by the expected number of synonymous sites in the codon . The difference in substitution rates ( dN − dS ) is a conventional summary statistic for diversifying selection among hosts , i . e . , host-specific selection causing nonsynonymous variation to accumulate among individual virus populations . We propose that the difference in mixture frequencies ( mN − mS ) can be employed as a summary statistic characterizing selection within each host . For instance , mN − mS > 0 can represent transient nonsynonymous polymorphisms undergoing directional selection ( which drives the fixation of a specific variant within the host ) . Using these quantities , we will show that the distribution of mixtures in our samples of HIV-1 and HCV sequences cannot be explained by variation in mutation rates alone , and that host-specific selection is an important force shaping variation at both levels of the total virus population . Because existing models of virus evolution seldom account for genetic variation both within and among hosts ( but see [24 , 25] ) , we formulate a novel yet simple model that invokes both host-specific selection and rapid transmission between hosts to explain the observed patterns of genetic variation within and among hosts infected by HIV-1 or HCV . Bolstered by stochastic simulations , our model specifies the conditions that yield this outcome , and quantitatively predicts the joint effect of selection and transmission on the genetic composition of the circulating virus population . We find that when host-specific selection for escape and reversion is unbalanced and the transmission rate is high , then the frequency of escape variants becomes considerably skewed from expectations derived from the immune diversity of the host population . Failing to account for this effect may lead to erroneous conclusions on the overall importance of CTL-mediated selection in directing the evolution of the total virus population , or the relative contribution of specific CTL epitopes . Furthermore , the design of an effective vaccine to human viruses such as HIV-1 or HCV is highly contingent upon our ability to anticipate the response of an infection to CTL-mediated selection .
We screened for sequencing mixtures in population-based sequences of HIV-1 PR ( n = 3 , 458 ) and reverse transcriptase ( RT , n = 1 , 997 ) isolated from 3 , 004 and 1 , 989 treatment-naive individuals , respectively , and HCV sequences of envelope protein E1 ( n = 2 , 691 ) and the hyper-variable region HVR1 of envelope protein E2 ( n = 346 ) . Although many sequences had at least one mixture ( 55% HIV-1 , 63% HCV ) , there were relatively few mixtures per sequence on average ( 0 . 015 mixtures per codon position in HIV-1 , 0 . 011 in HCV ) , suggesting that only a small number of codon positions had mixtures at detectable ( 20%–80% ) frequencies in a given host ( Figure S1 ) . We found substantial variation among codon positions in mixture frequencies ( Figure S2 ) , which was greater for nonsynonymous ( coefficient of variation = 1 . 98 HIV-1 , 1 . 28 HCV ) than synonymous mixtures ( 0 . 95 HIV-1 , 1 . 06 HCV ) . There was no significant correlation between nonsynonymous and synonymous mixture frequencies per codon position in either HIV-1 ( RT , Pearson's ρ = 0 . 04 , p-value = 0 . 52; PR , ρ = 0 . 13 , p-value = 0 . 21 ) or HCV gene sequences ( E1 , ρ = 0 . 01 , p-value = 0 . 75; E2 , ρ = −0 . 13 , p-value = 0 . 18 ) , indicating that the variation in mixture frequencies among codon positions was not simply due to local mutation rates . The difference between nonsynonymous and synonymous mixture frequencies ( mN − mS ) was highly correlated with the difference between nonsynonymous and synonymous substitution rates ( dN − dS ) per codon position for both HIV-1 and HCV gene sequences ( Figure 1A ) . This positive correlation between dN − dS and mN − mS remained significant for both E1 and E2 gene sequences even when different genotypes of HCV were analyzed separately . Overall , the quantity dN − dS assumed a negative value when averaged across the gene sequence , implying that nonsynonymous variation at the majority of codon positions was largely neutral or deleterious throughout the host population . Nevertheless , we detected significant diversifying selection ( dN − dS > 0 ) at nine codon positions in HIV-1 PR ( 12 , 13 , 19 , 35 , 37 , 63 , 64 , 77 , and 93 ) and eight positions in RT ( 35 , 39 , 102 , 122 , 135 , 200 , 211 , and 245 ) after correcting for the false-discovery rate [26] ( α = 0 . 05 ) ; likewise , significant diversifying selection was attributed to several codon positions in HCV E1 and E2 ( HVR1 ) sequences , which varied by genotype . For specific CTL epitopes in HIV-1 PR , RT , and HCV E1 sequences , we observed disproportionately higher frequencies of nonsynonymous mixtures at the anchor residues ( Figure 1B ) critical for MHC binding . In contrast , the profile of synonymous mixture frequencies within these epitopes lacked any distinct peaks in association with anchor residues . Overall , the median difference between the frequencies of nonsynonymous and synonymous mixtures was significantly greater at known HLA-B–restricted epitopes ( median mN − mS = −0 . 2% mixtures per sequence per site ) than in the remainder of the HIV-1 RT sequence ( −0 . 5%; Wilcoxon rank-sum test , p-value = 0 . 007 ) . We also found that mN − mS was greater at the anchor residues of HLA-B–restricted epitopes ( median = −0 . 2% ) than in an equivalent random sample of codon positions from HIV-1 RT on average ( median = −0 . 4% ) , but this difference was only marginally significant ( p-value = 0 . 11 ) . In contrast , the median was not significantly greater at the known HLA-A–restricted epitopes within RT ( Wilcoxon rank-sum test , p-value = 0 . 22 ) , consistent with previous studies suggesting that HLA-B alleles assume a dominant role in the CTL control of HIV-1 [9 , 27] . In HIV-1 PR , the median excess in nonsynonymous mixtures was considerably greater within the single known HLA-B–restricted epitope ( median = 0 . 7% ) than in the rest of the gene sequence ( median = −0 . 4% ) , but this difference was only marginally significant due to the small sample of codon positions ( Wilcoxon rank-sum test , p-value = 0 . 1 ) . Again , there was no significant difference in median values between HLA-A–restricted epitopes and the remainder of the PR sequence ( Wilcoxon rank-sum test , p-value = 0 . 55 ) . Similarly , in the HCV E1 sequences , we found that the median excess of nonsynonymous mixtures was significantly greater within the two known HLA-B–restricted epitopes ( median = 0 . 9% ) than in an equivalent random sample of codon positions ( median = −0 . 2%; Wilcoxon rank-sum test , p-value = 0 . 023 ) . However , the median value for known HLA-A–restricted epitopes in HCV E1 was significantly less ( median = −0 . 5% ) than that in the remaining codon positions ( median = −0 . 1%; Wilcoxon rank-sum test , p-value = 0 . 003 ) . There were only two known CTL epitopes in the HCV E2 HVR1 sequence , both classified as HLA-A–restricted . We found no significant association between the quantity mN − mS and codon positions located within these epitopes ( Wilcoxon rank-sum test , p-value = 0 . 87 ) . In sum , nonsynonymous mixtures tend to accumulate disproportionately at codon positions under CTL selection , preferentially within HLA-B–restricted epitopes . A surplus of nonsynonymous mixtures within CTL epitopes represents transient polymorphisms that are eventually driven to fixation in the host by selection for escape or reversion [28] . This implies that the probability of sampling nonsynonymous sequencing mixture should decline with the intensity of host-specific selection at that codon position . As a result , host-specific selection would produce negative correlation between mN − mS and dN − dS across codon positions in the range dN − dS > 0 , contrary to what we have observed in HIV-1 and HCV gene sequences . This paradox can be reconciled by incorporating the early transmission of unfavorable variants into a model of virus evolution ( Figure 2 ) . When selection and transmission act on similar time scales , the composition of the circulating virus population ( i . e . , the source of new infections ) will not necessarily match the diversity of HLA genotypes in the host population . Suppose that an escape variant is transmitted from a host with a rare HLA genotype to a new host with a common HLA genotype . If the escape variant cannot outcompete the wild-type virus in the absence of a CTL response , then selection will favor reversion [7 , 8] . But the selective advantage of the wild-type virus may be so narrow that a substantial probability remains of transmitting the original escape variant [8 , 14] . Under such conditions , the severe bottleneck upon transmission could fix either the wild-type or escape variant in the new individual population ( Figure 2 ) . Because the next host will likely have the common HLA genotype , this transmission event can recreate the selective conditions requiring a transient nonsynonymous polymorphism to occur . To investigate this hypothesis , we implemented a simulation of allele frequency evolution within individual virus populations with ongoing transmission through a succession of hosts . Each individual virus population was represented by a single locus containing either an escape variant ( at frequency p ) or the wild-type allele . We assumed that transmission of the virus to a new host involved a severe bottleneck , such that the next population was initially fixed for either the escape variant ( with probability p ) or wild-type allele . Viral fitness in a given host was determined by a single MHC locus , at which an allele restricting the wild-type virus ( HLA+ ) was present at a frequency q in the host population . We observed that the mean frequency of within-host polymorphisms fpoly:0 . 2 ≤ p ≤ 0 . 8 converged over time to an equilibrium value , which declined with stronger host-specific selection if the transmission rate was low ( Figure 3A ) . On the other hand , if the transmission rate was high , then fpoly increased with stronger selection and thereby became positively correlated with genetic variation among hosts . By sustaining high levels of polymorphism within hosts , a joint increase in selection and transmission rate may also cause the frequency of the escape mutation in the circulating virus population ( π = E ( p ) ) to depart substantially from the expected value at equilibrium in the absence of polymorphism ( π˄ = q , i . e . , individual virus populations fix alleles matching host HLA genotypes ) . In our simulations , if selection favoring escape in HLA+ hosts was sufficiently stronger than selection for reversion in HLA− hosts , then π˄ became substantially greater than q at equilibrium ( Figure 3B ) . On the other hand , if selection favoring reversion in HLA− hosts was greater , then the equilibrium value of π˄ was deflected in the opposite direction , below q ( not shown ) . This departure of π˄ from q became more pronounced with increasing transmission rates . Unequal mutation rates between the virus alleles could also contribute to this effect ( Figure S3 ) . An escape allele may therefore predominate the circulating virus population even when the selective HLA allele in the host population is rare . In other words , an individual possessing a rare HLA allele may nevertheless stand a high chance of becoming infected by a matched escape variant if selection for reversion is weak and the transmission rate is high . This process sustaining high levels of nonsynonymous polymorphism at codon positions under host-specific selection is related to the maintenance of genetic variation in a subdivided population by local adaptation [29 , 30] and can be illustrated with a simple deterministic model . We use the following differential equation [31]: to describe the mean rate of change in p within a given host , where s is the selection coefficient , and μ and ν are the forward and back mutation rates , respectively . Initial conditions for Equation 1 were defined to reflect the severe bottleneck imposed by transmission of the virus ( i . e . , p ( 0 ) = 0 or p ( 0 ) = 1 ) . Assuming that transmission occurs after a constant time interval ( τ ) , the expected value of π after n transmissions is obtained from the recurrence equation: where pHLA+ and pHLA− are approximate solutions of Equation 1 for evolution of p in HLA+ and HLA− hosts , respectively ( Protocol S1 ) . Equation 2 has an equilibrium solution: which reduces to π˄ = q when μ = ν and selection for escape and reversion is symmetric between host types ( sesc = srev ) . As τ approaches ∞ , π˄ also converges towards q because the evolution of the escape allele within hosts is resolved before transmission ( i . e . , and ) . Conversely , as τ approaches zero , π˄ converges towards a quantity determined by the ratio of ν and μ ( Protocol S2 ) . The behavior of π˄ at these limits implies the existence of an intermediate waiting time to transmission ( τmax ) , which maximizes the departure of π˄ from q . An approximation of τmax indicates that it is on the order of max ( sesc , srev ) −1 when selection is stronger than mutation ( Protocol S3 ) . Thus , our model confirms that the greatest departure of π˄ from the expectation q occurs when the mean transmission rate corresponds to the overall intensity of selection . We found a strong correspondence between this model and simulations ( Pearson's ρ = 0 . 92 , p-value < 10−15; Figure S4 ) with all incongruous cases being caused by stochastic effects due to effective population sizes within hosts of N = 102 or below . The effective population size for HIV-1 is estimated to be on the order of 103 and greater , while the total census population size is typically several orders of magnitude larger [32–34] , and the census size for HCV is approximately 10-fold greater still . Hence , this model is a reasonably accurate representation of evolution within realistic HIV-1 and HCV populations .
In this study , we have described a novel pattern in the genetic variation of two human viruses , and formulated a simple population genetic model , supplemented with stochastic simulations , to explain it . However , because of the limited availability of population-based sequences that have not been stripped of sequencing mixtures , we were required to restrict our analysis to the RT and PR coding region of HIV-1 , in which mixtures provide useful information on the evolution of resistance [21] . Although we focused our investigation on subtype B sequences isolated from treatment-naive individuals , we had no direct control over the sequencing and base-calling conditions of this data set . On the other hand , we obtained unprocessed sequencing electropherogram data of the HCV E1 envelope coding region , such that we could uniformly apply our own methods across all sequences . We were also unable to control for the circumstances under which sequences were isolated from either HIV-1– or HCV-infected patients , e . g . , days since infection or seroconversion , regionality of patient populations . Even so , these sampling issues would not bias inferences based on site-by-site comparisons of sequence variation ( e . g . , relative mixture frequencies ) . We were able to recover an exceptionally clear and consistent signal of a link between within-host and among-host genetic variation among codon positions in HIV-1 and HCV sequences . This pattern represents strong evidence for CTL-mediated selection in both viruses , specifically targeting with HLA-B–restricted epitopes . The rapid accumulation of genetic variation in HIV-1 and HCV enables these viruses to elude the immune system and forestalls the development of effective vaccines . Identifying the factors that shape genetic diversity in these human viruses remains a formidable challenge . Because these viruses possess exceptionally high mutation rates , extensive genetic variation accumulates within hosts that may be shaped by ongoing host-specific adaptation . However , the development of models of virus evolution within hosts has been largely independent of “dynamical” models of the transmission and spread of viruses across host cells and individuals [25] . As a result , few models of virus evolution integrate the evolution within hosts with viral dynamics at the level of the host population , which could otherwise reveal emergent properties of evolution within hosts . For example , there is an extensive literature characterizing selection in HIV-1 [10 , 35–47] by comparing inferred rates of nonsynonymous and synonymous substitutions , but these studies employ methods that do not explicitly distinguish between within- and among-host variation ( but see [19 , 48] ) . However , empirical evidence indicates that aspects of the host population can influence patterns of evolution within hosts , and vice versa . For instance , Ross and Rodrigo [10] found evidence that the magnitude and persistence of site-specific diversifying selection within patients was correlated with the rate of progression to acquired immune deficiency syndrome ( AIDS ) , which may influence long-term epidemiological dynamics in the host population . Moore et al . [9] found significant associations between divergent codon positions within CTL epitopes in HIV-1 RT and HLA allelic variation in the host population , which implied that CTL-mediated selection within hosts was influencing the evolution of the total virus population . More recently , Kosakovsky Pond et al . [48] developed a customized phylogenetic analysis to detect significant turnover in codon positions under diversifying selection in HIV-1 PR and RT sequences among human populations with distinct HLA frequencies . They also found that many nonsynonymous substitutions that were mapped to terminal branches of the tree ( i . e . , occurring within hosts ) were absent from internal branches , suggesting that adaptations within individual virus populations were not always maintained at the level of the total virus population [48] . These observations motivate the theoretical development of models of viral evolution that capture the interaction between the within-host and among-host levels of genetic variation . Recently , Grenfell et al . [24] sought to unify the characteristic shape of phylogenetic trees for different virus pathogens with the evolutionary processes within hosts . For instance , phylogenetic trees derived from HIV-1 or HCV sequences sampled from the host population tend to be more “balanced” , reflecting the epidemiological spread of the virus [24] . In contrast , trees derived from influenza A virus hemagglutinin sequences are less balanced , containing a persistent “backbone” that continually spawns short-lived lineages [49] . They proposed that this variation in tree shape , which indirectly manifests the genetic variation among hosts , was driven by the rate at which variants with a selective advantage in the previous host were being transmitted to the subsequent host . Our model complements this previous work by directly evaluating the influence of within-host evolution on the accumulation of nonsynonymous substitutions that differentiate individual virus populations , and the reciprocal effect of this divergence among hosts on variation within hosts . As a result , we can obtain quantitative predictions on how selection within hosts and the transmission rate will influence the frequency of escape variants in the total virus population . The model also predicts that variation in the mean surplus of nonsynonymous mixtures ( quantified by the summary statistic mN − mS ) per gene indicates divergent intensities of host-specific selection . Similarly , the characteristic transmission rates and overall intensity of selection of different viruses ( e . g . , HIV-1 , HCV , influenza A virus ) may revealed by a divergence in the mean surplus of nonsynonymous mixtures per virus . We did not attempt to infer differences between genes or viruses from the absolute frequencies of mixtures in the current data set due to the potential variation in sequencing protocols ( as discussed above ) . Nevertheless , our model should motivate investigators in viral evolution to provide access to raw sequencing data , including annotation of variables that could influence the detection of polymorphisms ( e . g . , lab sequencing protocol , automated sequencer type and manufacturer ) . Based on the distribution of relative mixture frequencies ( i . e . , site-by-site comparisons within genes ) , our model indicates that the genetic variation of HIV-1 and HCV is being shaped by the ongoing transmission of unfavorable variants , skewing the frequency of an escape variant in the total virus population towards the direction that host-specific selection is strongest . This unexplored imprint of within-host evolution , manifested as a site-specific surplus of nonsynonymous mixtures within CTL epitopes , can strongly influence the overall composition of the circulating virus population , in addition to founder effects . Because we observed this phenomenon in both HIV-1 and HCV , it may be a common feature of viruses that exhibit both prolific genetic variation within hosts and substantial rates of transmission .
We obtained treatment-naive HIV-1 subtype B sequences from the HIV Drug Resistance Database at Stanford University ( Stanford HIVDB ) [50] . At the time of analysis , there were 3 , 458 PR and 1 , 997 RT sequences meeting our criteria , representing 3 , 004 and 1 , 989 patients , respectively . By restricting the data set to treatment-naive individuals , we sought to minimize the confounding effects of selection for drug-resistant variants . Further screening for antiviral resistance was carried out by aligning each sequence to its closest subtype reference sequence ( obtained from the Los Alamos National Laboratory [LANL] HIV sequence database; [51] ) and scoring for resistance according to the Stanford HIVDB mutation scores using customized scripts in HyPhy [52] . Assuming worst-case resolution of ambiguous nucleotides ( i . e . , maximized scores ) , 149 RT and 58 PR sequences with at least low-level resistance ( score ≥ 15 ) were discarded from the data sets . All 297 nucleotide sites from PR sequences were included in our analyses . RT sequences were truncated to nucleotide sites 1 to 741 to exclude poorly sampled tail regions from the analyses . In addition , we obtained 2 , 691 chromatogram traces generated from ABI 310 and Beckman CEQ 8000 automated sequencers , covering the core E1 region of HCV . For the majority of traces , each corresponded to a unique isolate from a patient for the initial diagnosis and genotyping of an HCV infection . All trace files were converted to standard chromatogram format and processed with the base-calling program Phred [53] . Potential sequencing mixtures were identified by screening the uncalled peak output using a custom Python script . An uncalled peak was classified as representing a minority variant if: ( 1 ) it was located within ±1 trace points of a called peak; and the area under the uncalled peak was ( 2 ) at least 20% of the called peak area; ( 3 ) at least 10% the mean area of the last ten called peaks; and ( 4 ) at least two times greater than the mean area of the last five uncalled peaks . All sequences were truncated to the E1 coding region spanning the nucleotide sites 1 to 399 . We also obtained 346 published population-based RT-PCR sequences from Genbank ( see Accession Numbers ) spanning the hyper-variable region HVR1 of HCV envelope protein E2 [54–57] . Sequences were aligned using ClustalW [58] and manually adjusted with Se-Al version 2 . 0 [59] ( alignments available upon request ) . We used neighbor-joining [60] with Tamura-Nei [61] distance to reconstruct the phylogeny from each sequence alignment . Pairwise distances from each phylogeny indicated that the sequences were highly divergent ( Figure S5 ) . To estimate the number of nonsynonymous and synonymous substitutions with branch corrections at each codon position , we employed the single-likelihood ancestor counting method [62] as implemented in HyPhy [52 , 63] using the default settings . Ambiguious nucleotides were resolved to the consensus codon at that position in order to remove any possible influence of mixture frequencies on estimates of substitution rates . We tested for significant positive selection ( dN > dS ) by applying a continuous extension of the binomial distribution to model the probability that a given proportion of substitutions are nonsynonymous , given the proportion of sites that are nonsynonymous at the codon position [63] . For analyzing associations between nonsynonymous mixture frequencies and epitopes within HIV-1 PR and RT , we applied the CTL epitope definitions from the LANL HIV immunology database [64] . Similarly , we applied the CTL epitope definitions from the LANL HCV immunology database for analyzing associations within HCV E1 and E2 ( HVR1 ) [65] . We implemented a simulation of virus evolution in a host population using an iterative Moran process [66] . Both virus and host populations were each modeled by a single two-allele locus , representing the immune escape and HLA genotypes , respectively . Instantaneous rates for the unit increase and decrease of escape allele frequency within a host were: where j is the number of wild-type alleles in an ideal population of constant size N , and λ1 and λ2 are the wild-type and escape virus growth rates . If the host was HLA− , we set λ1 = 1 and λ2 such that the selection coefficient for reversion srev = ( λ1 − λ2 ) . Otherwise , we set λ1 < λ2 so that sesc = ( λ2 − λ1 ) > 0 . After an exponentially distributed waiting time ( τ ) with rate k , a randomly selected individual from K = 103 hosts was replaced . This new host was HLA+ with probability q ( and HLA− otherwise ) , and infected by wild-type virus with probability jτ/N , where jτ is obtained from the iterative application of j+ and j− , and the total number of events occurring in the time interval τ was determined by random draws from an exponential distribution with the rate ( j+ + j− ) . Otherwise , it was infected by the escape mutant virus . This new infection was therefore initially fixed for either the wild-type or escape virus genotype , assuming a severe bottleneck upon transmission between hosts . Simulations were run for 200 × K transmissions , which was sufficient for π to converge to an equilibrium for all parameter values evaluated . We recorded the frequency of the escape allele in the individual virus population ( p = 1 − j/N ) , from which we calculated the mean frequency among hosts ( π = E ( p ) ) . Given the empirical detection threshold of minority variants from population-based sequencing , an individual virus population was considered to be detectably polymorphic if 0 . 2 < p <0 . 8 . Unique parameter values were assigned to 100 replicate simulations by Latin hypercube sampling from their respective ranges: q = ( 0 , 0 . 5 ) ; μ , ν = ( 10−5 , 10−3 ) ; sesc , srev = ( 0 . 002 , 0 . 2 ) ; N = ( 102 , 105 ) ; and k = ( 0 . 00137 , 0 . 0137 ) , such that transmissions occur after approximately 0 . 2 to 2 years ( where τ is in units of days ) . To compare the simulation results to our deterministic model , we used the numerical integration function in Mathematica 5 . 1 ( Wolfram Research , http://www . wolfram . com ) to calculate the expectation of Equation 3 assuming that the waiting time τ was exponentially distributed with rate parameter k .
GenBank ( http://www . ncbi . nlm . nih . gov/Genbank/index . html ) accession numbers for the HCV E1 envelope protein-coding sequences used in our study are AY766700–AY768365 . GenBank accession numbers for the E2 envelope protein ( HVR1 ) sequences used in our study are the following: AY390002 , AY390005 , AY390008 , AY390010 , AY390013 , AY390016 , AY390019 , AY390022 , AY390024 , AY390027 , AY390030 , AY390032 , AY742960–AY743049 , AY309923–AY309954 , AY314963–AY314969 , AY390002–AY390035 , AY564735–AY564784 , and AY935999–AY936132 . | The rapid accumulation of genetic variation in human viruses , such as human immunodeficiency virus type 1 ( HIV-1 ) and hepatitis C virus ( HCV ) , enables these pathogens to elude the immune system and forestalls the development of effective vaccines . This variation may be shaped by selection due to host-specific immune responses , such that the total virus population mirrors the immune diversity of the host population . However , the often-neglected viral genetic variation within hosts may also play an important role in shaping variation in the total virus population . We carry out an innovative analysis of bulk HIV-1 and HCV sequences isolated from over 4 , 000 human patients , exploiting “mixtures” ( i . e . , ambiguous nucleotides ) as a site-specific marker of within-host genetic variation . We find that nonsynonymous mixtures are disproportionately abundant at codon positions under strong host-specific immune selection . Because existing models of virus evolution provide no explanation for this outcome , we have formulated a new model supplemented with stochastic simulations to demonstrate that the rapid transmission of viruses through diverse selective environments creates a positive correlation between nonsynonymous variation within and among hosts . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"viruses",
"infectious",
"diseases",
"immunology",
"evolutionary",
"biology",
"genetics",
"and",
"genomics"
] | 2007 | Adaptation to Human Populations Is Revealed by Within-Host Polymorphisms in HIV-1 and Hepatitis C Virus |
Rabies is traditionally considered a uniformly fatal disease after onset of clinical manifestations . However , increasing evidence indicates that non-lethal infection as well as recovery from flaccid paralysis and encephalitis occurs in laboratory animals as well as humans . Non-lethal rabies infection in dogs experimentally infected with wild type dog rabies virus ( RABV , wt DRV-Mexico ) correlates with the presence of high level of virus neutralizing antibodies ( VNA ) in the cerebral spinal fluid ( CSF ) and mild immune cell accumulation in the central nervous system ( CNS ) . By contrast , dogs that succumbed to rabies showed only little or no VNA in the serum or in the CSF and severe inflammation in the CNS . Dogs vaccinated with a rabies vaccine showed no clinical signs of rabies and survived challenge with a lethal dose of wild-type DRV . VNA was detected in the serum , but not in the CSF of immunized dogs . Thus the presence of VNA is critical for inhibiting virus spread within the CNS and eventually clearing the virus from the CNS . Non-lethal infection with wt RABV correlates with the presence of VNA in the CNS . Therefore production of VNA within the CNS or invasion of VNA from the periphery into the CNS via compromised blood-brain barrier is important for clearing the virus infection from CNS , thereby preventing an otherwise lethal rabies virus infection .
Rabies is a highly lethal disease caused by the neurotropic rabies virus ( RABV ) . It has been estimated that about 55 , 000 persons died from rabies each year mostly in Africa and Asia [1] . Successful vaccines have been developed for the prophylaxis of the disease . Timely post-exposure prophylaxis ( PEP ) can prevent the development of rabies , when individuals are exposed to the virus . Unfortunately , PEP is ineffective once clinical signs have appeared and virus replicates in the CNS [2] , [3] . It is generally believed that virus clearance is impossible once the virus reaches the brain [4] , [5] . However , there is now increasing evidence that non-lethal infection can occur in various animal species and in humans [6]–[8] . Up to date , six non-lethal human rabies cases have been documented in the US alone [9]–[13] . All these patients either had rabies specific antibodies in the cerebral spinal fluids ( CSF ) at the time of hospitalization or after treatment with the Milwaukee Protocol or a modification thereof [14] . In addition , recovery of laboratory animals from clinical rabies has also been reported [15] . It has been demonstrated in mice that the clearance of the virus from the CNS requires the induction of innate immune responses in the CNS , induction of RABV-specific adaptive immunity in the periphery , as well as infiltration of immune effectors across the blood-brain barriers ( BBB ) [16] . Roy et al . have demonstrated that the lethality of infection with the silver-haired bat RABV can be reduced by opening the BBB . Failure to enhance BBB permeability prevents the delivery of immune effectors to the CNS leading to the lethal outcome of rabies infection [17]–[19] . Although chronic natural RABV infection in vampire bats [20] , recovery from experimental RABV infection in dogs and ferrets , and recovery of humans from rabies has been documented [10] , [13] , [21] , the mechanism ( s ) involved in the prevention of lethal rabies are not completely understood . These observations have led to renewed efforts to obtain evidence of underlying mechanisms behind nonfatal rabies infections . One of the major findings is that non-lethal wt RABV infection or recovery from rabies correlates with the presence of VNA in the CSF that presumably crossed the BBB [8] , [19] , [22] . In the present report , we describe the observation of non-lethal infection in dogs after experimental infection with a wild type ( WT ) RABV that originated from a dog ( DRV-Mexico ) [23] . We found that the non-lethal infection correlated with the presence of high level VNA in the CSF , in contrast to lethal infection , where no or only little VNA ( <0 . 5 IU ) were detected in the CSF . On the other hand , vaccinated dogs resisted a challenge infection with no detectable VNA in the CSF but high VNA levels in the serum .
Dog RABV ( DRV-Mexico ) was originally isolated from a dog of Mexico origin [23] , [24] . Virus stocks were prepared by inoculating 10 µl of the virus by the intracerebral route into one-day-old suckling mice . When moribund , the mice were euthanized and brains removed . A 10% ( w/v ) suspension was prepared by homogenizing the brain in DMEM . The homogenate was centrifuged to remove debris and the supernatant collected and stored at −80°C . Healthy , non-rabies vaccinated , 5 month old clinically healthy female beagles were obtained from Covance , USA . All the experimental dogs were housed individually in temperature- and light-controlled quarters in the Animal Facility , College of Veterinary Medicine at University of Georgia . All animal experiments were carried out under Institutional Animal Care and Use Committee-approved protocols ( animal welfare assurance number A3085-01 ) . All the experimental dogs were pre-screened for the presence of maternal VNA to RABV using the Rapid Fluorescent Focus Inhibition Test ( RFFIT ) . Dogs were sedated with Acepromazine , an IM injection of phenothiazine derivative . Eight dogs were randomly selected and infected intramuscularly ( i . m ) with 100 µl viral suspension containing 200 MICLD50 ( 50% mouse intracerebral lethal dose ) of DRV-Mexico by direct inoculation into the left hemisphere of the temporalis muscle . Another group of 4 dogs were immunized with a RABV vaccine . The immunized dogs were challenged after 4 weeks post immunization with 100 µl of DRV suspension as described above . Dogs were observed at least once a day prior to challenge and two to four times a day for 30 days after challenge . Humane endpoint of the study is the appearance of hind limb paralysis of one or both limbs and the experimental endpoint of the study was decided on the basis of observed clinical signs for 30 days post challenge . Blood , CSF and brain samples were collected before infection and/or at the time of termination for various analyses including complete blood counts ( CBC ) , serum biochemistry , histopathology , immunohistochemistry , antibody titration , and CSF cytology . For histopathology and immunohistochemistry , brain samples were fixed in 10% neutral buffered formalin as described previously [25] . Brains were removed and paraffin embedded for coronal sections . For de-paraffinization , slides were heated at 60°C for 25 min and then dipped in CitriSolv ( Fisher Scientific , PA ) three times for 5 min and dried until chalky white . The slides were then stained with hematoxylin and eosin ( H&E ) . Slides were heated in antigen unmasking solution ( Vector Laboratories , CA ) above 90°C for 20 min and allowed to cool down to room temperature . Anti-RABV N monoclonal antibody Mab N42 and Anti-RABV G Mab 53 were used to detect the viral antigens N and G , respectively [26] . Biotinylated secondary antibodies were used for detection as described [27] , with avidin–biotin–peroxidase complex ( Vector Laboratories , CA ) and diaminobenzidine ( DAB ) as a substrate for color development . The intensity of DAB signals corresponding to CD3 antigen was quantified manually to obtain statistical analysis . Both the histopathology and immunohistochemistry slides were read and interpreted by the same pathologist . The blood and CSF samples were collected before and after infection and sent to the Clinical Pathology Laboratory at University of Georgia for analysis . Blood samples were collected in plastic tubes coated with or without EDTA anticoagulant from the jugular vein or cephalic veins . Whole blood and serum were analyzed for hematology parameters and chemistry parameters , respectively . CBCs were performed on Siemens Bayer Advia 120 using flow cytometry laser light scatter methodology . Serum chemistry is performed on Roche Hitachi P Module analyzer . The CSF samples were collected from the cerebellomedullary cistern site and were evaluated for various inflammatory parameters , such as , white blood cell counts ( WBC ) and total protein concentration . CSF cytology is performed by counting the cells on a Neubauer Hemocytometer and a cytospin smear is made to differentiate the cells counted . The resulting smear is stained on the Wescor stainer and evaluated microscopically . In addition , the color and transparency is recorded by visual inspection by the technologists . The CSF protein is measured on the Roche Hitachi P module . Blood and CSF samples were collected for measurement of VNA using the RFFIT ( Rapid Fluorescent Focus Inhibition Test ) as described previously [20] . Briefly , 50 µl of serial five-fold dilutions of serum were prepared in Lab-Tek Chamber slides ( Nalge Nunc International , Rochester , NY ) . Fifty FFD50 ( 50% Fluorescing Foci dose ) of CVS-11 was added to each chamber and incubated for 90 min at 37°C . NA cells ( 105 cells ) were added into each chamber and the slides were incubated at 37°C for 20 hr , fixed with ice-cold 80% acetone and stained with FITC-conjugated anti-RABV N antibodies for 1 hr at 37°C . Twenty fields in each chamber were observed under a fluorescent microscope , and the 50% endpoint titers were calculated according to the Reed-Muench formula . The values were compared with that of reference serum ( obtained from the National Institute for Biological Standards and Control , Herts , UK ) and normalized to international units ( IU/ml ) . The project AUP is entitled , “Virus clearance from the central nervous system” and the AUP number is A2011 03-016 . It was approved by the University of Georgia's Institutional Animal Care and Use Committee on 4 APR 2011 , and will expire on 4 APR 2014 . The University of Georgia's University Research Animal Resources unit is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care , International ( AAALAC-I ) . The registration number from the U . S . Department of Agriculture , Animal and Plant Health Inspection Service , Animal Care is ( USDA APHIS-AC ) . We have an assurance on file with the NIH-Office of Laboratory Animal Welfare ( NIH-OLAW ) , and are in compliance with the PHS Policy on Humane Care and Use of Laboratory Animals and the 8th edition of the Guide for the Care and Use of Laboratory Animals , 2011 . Statistical significance of the differences between groups was tested using student's T test with *** indicating a p value <0 . 0001 , ** a p value <0 . 001 , and * a p value <0 . 05 using GraphPad prism software .
Among the eight dogs infected with DRV-Mexico , three dogs showed typical signs of the dumb or paralytic form of rabies at 11 , 12 , and 21 days post infection ( dpi ) , respectively . Early signs consisted of subtle changes in behavior , including quietness , hiding in the corner of the cage , lethargy and loss of appetite . Later signs included agitation , poor coordination , tremors , trembling , persistent regurgitation and retching , excess salivation and paralysis . Among these dogs , only one showed signs of furious form of rabies such as aggressiveness , whining and barking . The rabid dogs reached the end point of the study ( hind limb paralysis ) and were sacrificed on 13 , 14 , and 22 dpi , respectively . The other 5 infected dogs exhibited only the subtle behavior changes , which began on 13 dpi , and appeared normal again by 21 dpi . At 30 dpi , all surviving dogs were sedated and after blood and CSF collection were euthanized and their brains removed for analysis . None of the dogs in the vaccinated group ( n-4 ) showed any clinical signs indicative of rabies infection during the 30 day observation period after which they were euthanized . All the serum and CSF samples were subjected to RFFIT analysis for VNA . None of the dogs had detectable VNA in either the serum or the CSF prior to infection or vaccination . Serum and CSF VNA titers detected in terminally ill dogs at the time of euthanization were only 0 . 42±0 . 41 and 0 . 1±0 . 15 IU/ml , respectively . VNA titers determined in the serum and CSF of the surviving dogs were 6 . 5±2 . 6 and 5 . 1±2 . 9 IU , respectively , at 30 dpi ( Fig . 1 & table 1 ) . The vaccinated group had serum VNA titers of 21 . 5±9 . 7 IU prior to challenge with 1 . 75±0 . 9 IU in serum and no detectable VNA in CSF at 30 dpi ( Fig . 1 & table 1 ) . The high VNA levels in the serum of the dogs recovering from clinical signs of rabies indicates that they were infected with the virus and the presence of VNA in CSF strongly suggests that the virus invaded their CNS , and VNA contributed to their survival . Dogs that developed lethal rabies did not develop detectable VNA levels in CSF . Total protein and WBC counts in the CSF were also analyzed . As shown in Fig . 2 , no WBC were detected in the CSF and total protein was 12 . 7±1 . 3 mg/dl in dogs prior to infection or in the immunized dogs , indicating that the BBB remained intact in these animals . Total proteins ( 64 . 4±9 . 3 mg/dl ) and WBC counts ( 342 . 3±78 . 1 cells/µl ) increased dramatically in the CSF of terminally ill dogs , indicating a strong CNS inflammatory response ( Fig . 2 & table 1 ) . In the surviving dogs , total proteins ( 29 . 5±8 . 2 mg/dl ) and WBC counts ( 29 . 0±7 . 3 cells/µl ) in CSF were higher than those in the dogs prior to infection or in the immunized animals , but were much lower than those in the terminally ill animals . The relatively large amount of WBC and proteins in CSF suggest that BBB permeability is enhanced in DRV-Mexico-infected dogs . To determine whether the apparent changes in BBB permeability were reflected in the infiltration of immune/inflammatory cells into the brain , brains were collected from the immunized , surviving and terminally ill animals for histopathology and immunohistochemistry . Increased cell infiltrates and perivascular cuffing of mononuclear cells with marked activation of microglial cells were found in the hippocampus , hypothalamus and cerebellum of dogs that succumbed to rabies . Only residual cell accumulation was observed in the recovered dogs while evidence of inflammation was absent in the immunized dogs ( Fig . 3A ) . To quantify immune cell infiltration into the CNS , CD3-bearing cells were assessed in various brain regions by immunohistochemistry . As shown in Fig . 3B , more CD3-positive cells were found in the hippocampus and hypothalamus of dogs that succumbed to rabies than in the same regions of recovered dogs . CD3 positive cells were not observed in the CNS tissues of immunized dogs ( Fig . 3B ) . Quantification of CD3-positive cells in different parts of the brains of the groups of animals indicates that there were indeed significantly more CD3+ cells in the CNS of the dogs that succumbed to rabies than in those that survived ( Fig . 3C ) . In order to correlate the development of rabies with virus loads in the CNS , brain tissues were evaluated for viral G and N antigens by immunohistochemistry . As shown in Fig . 4 , viral antigens were detected only the hippocampus , hypothalamus and cerebellum of the dogs that succumbed to rabies . Relatively high numbers of cells positive for rabies G ( Fig . 4A ) and N ( Fig . 4B ) were detected in the hippocampus and hypothalamus , whereas such were only sparsely detected in the cerebellum . No viral antigens were detected in the CNS of recovered or immunized dogs .
Despite the extensive progress in rabies research since the time of Pasteur , more than 55 , 000 people continue to die of rabies each year world-wide [28] . Once symptoms occur , rabies is almost always fatal . Recently , however , cases of non-lethal rabies infection and recovery from clinical rabies in laboratory animals and humans have been reported [6] , [10] , [21] , [29]–[32] . One of the major findings associated with these non-lethal infections is that many of the survivors had VNA in the CSF [6] , [10] , [22] , [29] , [32] , [33] . Consistent with these findings , we report in the present paper that experimental infection of dogs with a wt RABV is not invariably lethal , and that survival correlates with the presence of high VNA titers , evidence of WBC infiltration , and elevated levels of protein in the CSF . PEP is effective in preventing rabies after exposure providing that clinical signs of rabies have not appeared . Consisting of vaccination with RABV vaccines and administration of anti-rabies immunoglobulin at the site of exposure and systemically [34] , it is believed that PEP prevents the rabies virus from invading the CNS due to the long incubation period of the infection [35] . It has long been thought that it is difficult to clear the virus once it enters into the CNS [4] , [36] . This dogma was initially cast into doubt by the finding that RABV can be cleared form the CNS by VNA administered intravenously [37] . Recently it has been found that enhancement of the BBB permeability prevents rabies in the mouse model by allowing immune effectors from the periphery to enter into and clear RABV from the CNS [16] . Various studies have been conducted using different models to promote immune cell infiltration into the CNS in rabies including experimental allergy encephalitis , in which the disease enhances BBB permeability [38] , and infection with attenuated RABV such as recombinant viruses that express three copies of G [39] or immune stimulating agents [40] . All of these interventions result in an immune response to RABV in the CNS and reduce fatalities in mice infected with wt RABV [17]–[19] . However , simply targeting BBB integrity alone is not sufficient to protect mice from lethal rabies since administration of a chemokine , MCP-1 , enhanced BBB permeability but did not significantly increase the survival rate of mice infected with DRV-Mexico [41] . Neither does immunization with an inactivated RABV preparation 5 days after infection with DRV-Mexico despite inducing serum VNA . On the other hand , administration of MCP-1 to mice immunized with inactivated RABV significantly improved their survival from DRV-Mexico indicating that the combined effects of enhancement of BBB permeability and the production of rabies-specific VNA is protective [41] . In the present study , none of the dogs that succumbed to rabies had serum VNA = />0 . 5 IU and their CSF VNA were even lower ( ∼0 . 1 IU ) . This is despite the possibility that BBB permeability became enhanced at the end of their lives as suggested by the greater numbers of WBC and high protein levels detected in their CSF . In the dogs with non-lethal infection , BBB permeability had likely been enhanced at some stage since VNA , WBC , and high protein levels were detected in their CSF . Moreover , residual CD3 cell accumulation was observed but no virus antigen in the CNS tissues of these dogs at the time of sacrifice , suggesting that immune effectors acting in the CNS had cleared the virus . In the mouse model , lethal infection with wt RABV ( DRV-Mexico ) is not accompanied by enhanced BBB permeability ( cerebrum , cerebellum , or spinal cord ) at days 6 or 9 dpi when RABV antigens or RNA are detected in the CNS [19] . This is consistent with observations in mice of lethal infections with a variety of other wild-type RABV [42] . The apparent discrepancies observed between mice and dogs in the loss of BBB integrity indicate that the pathogenic processes in these two animal species may be somewhat different . Also , these discrepancies in BBB permeability may be due to the close co-adaptation of viral strain to their specific homologous host ( dogs ) than the spillover . Nevertheless , the correlation between the presence of VNA in the CSF and the clearance of RABV from the CNS is shared . In 3 of the humans that recovered from rabies , whether treated with the Milwaukee protocol or not , virus specific antibodies were detected in their CSF at the time of hospitalization [10] , [29] , [43] . In dogs recovered from infection with laboratory-attenuated virus , VNA was also detected in the CSF [44] . A ferret that recovered from rabies encephalitis also had detectable VNA in the CSF [32] . In our study , the dogs surviving wt RABV infection developed only mild disease without typical rabies symptoms yet developed high VNA levels in both serum and CSF . Particularly , the presence of VNA in CSF strongly suggests that the virus invaded their CNS . This is very different from immunized dogs in which VNA was produced in the serum , but not in the CSF . The detection of VNA in CSF and evidence of limited immune cell infiltration into CNS tissues makes it likely that the dogs survived the wt RABV infection by clearing the virus from the CNS . Thus we conclude that , like humans and other species [8] , [9]–[14] , [43] , infection of dogs with wt RABV is not invariably lethal even when the virus has reached the CNS . The mechanisms whereby certain dogs mediated an RABV-specific immune response that reached in the CNS and survived while others did not remain to be understood and may provide the foundations for the development of novel therapeutic intervention strategies for clinical rabies . | Inexorable lethality is still commonly attributed to rabies infection , although there is increasing evidence for non-lethal infection and even recovery from clinical rabies in various animal species and humans . This paper reports non-lethal infection in dogs . The striking difference between dogs that survived a wt RABV infection and dogs that succumbed to the infection is that the surviving dogs showed high level of VNA in the serum and in the CSF , as well as mild immune cell accumulation in the CNS , whereas dogs that succumbed to disease showed little or no VNA in the serum or in the CSF and developed severe CNS inflammation . Considering the role of VNA in clearing the virus from the CNS , production of VNA within the CNS or infiltration of VNA from the periphery into the CNS across the blood-brain barrier appears to be important for clearing the virus from CNS thereby preventing a lethal rabies infection . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Presence of Virus Neutralizing Antibodies in Cerebral Spinal Fluid Correlates with Non-Lethal Rabies in Dogs |
Adaptation to ecologically complex environments can provide insights into the evolutionary dynamics and functional constraints encountered by organisms during natural selection . Adaptation to a new environment with abundant and varied resources can be difficult to achieve by small incremental changes if many mutations are required to achieve even modest gains in fitness . Since changing complex environments are quite common in nature , we investigated how such an epistatic bottleneck can be avoided to allow rapid adaptation . We show that adaptive mutations arise repeatedly in independently evolved populations in the context of greatly increased genetic and phenotypic diversity . We go on to show that weak selection requiring substantial metabolic reprogramming can be readily achieved by mutations in the global response regulator arcA and the stress response regulator rpoS . We identified 46 unique single-nucleotide variants of arcA and 18 mutations in rpoS , nine of which resulted in stop codons or large deletions , suggesting that subtle modulations of ArcA function and knockouts of rpoS are largely responsible for the metabolic shifts leading to adaptation . These mutations allow a higher order metabolic selection that eliminates epistatic bottlenecks , which could occur when many changes would be required . Proteomic and carbohydrate analysis of adapting E . coli populations revealed an up-regulation of enzymes associated with the TCA cycle and amino acid metabolism , and an increase in the secretion of putrescine . The overall effect of adaptation across populations is to redirect and efficiently utilize uptake and catabolism of abundant amino acids . Concomitantly , there is a pronounced spread of more ecologically limited strains that results from specialization through metabolic erosion . Remarkably , the global regulators arcA and rpoS can provide a “one-step” mechanism of adaptation to a novel environment , which highlights the importance of global resource management as a powerful strategy to adaptation .
Adaptation to novel environments can proceed either through many mutations with small effects or through few mutations with large effects [1] . Adaptation to complex environments is the norm in biology , but a clear understanding of the adaptive processes employed by organisms in ecologically diverse environments is challenging . Ecological complexity can arise from increased species diversity , spatial or temporal heterogeneity or different resources . The availability of countless resources in complex environments can lead to a rapid and substantial increase in genetic diversity that can obscure broader biochemical principles of adaptation . For example , if resources are varied and plentiful , the population can accumulate mutations in genes that are not essential for survival in the selective environment [2] , [3] . Thus , in nutrient-rich environments , specialists with narrower niches can persist by using alternative resources without necessarily improving their fitness relative to the ancestor . Adaptation then occurs through specialization via fitness improvements or via metabolic erosion possibly without fitness improvements relative to the ancestor [3]–[5] . We hypothesized that , despite the tremendous opportunities for increased genetic diversity under conditions of plenty , consistent adaptive responses could be observed as parallel evolution across and within independently evolved populations . We further reasoned that such general and consistent adaptive responses could be driven by global metabolic regulators to provide an efficient reprogramming of metabolic networks with a minimal number of steps . Experimental evolution of model organisms under novel conditions is a versatile approach for understanding the evolutionary dynamics of adaptation and the functional constraints that shape the physiological evolvability of an organism . Typically , microbial model organisms are selected for adaptation to a single or a few distinct resources [6]–[11] , to antibiotics [12] , [13] , or to temperature [14] , [15] . Experiments in such relatively simple selective environments have shown that during adaptation to a single resource , the evolving population typically climbs a single-peak fitness landscape in an incremental manner with diminishing returns epistasis [16]–[18] . Even in such simple environments , resource partitioning or spatial or temporal heterogeneity can lead to the evolution of different specialists and complex ecological interactions [10] , [11] , [19]–[26] . To better understand the evolutionary and adaptive dynamics in ecologically complex environments , we focused on resource availability and conducted selection experiments in very nutrient-rich conditions . Unlike adaptation to a single limiting resource that is often conceptualized as a single fitness peak , a wealth of resources will potentially present abundant peaks in the fitness landscape . Because resources may differ only slightly , the selection differences can be very small and are reflected as very modest fitness peaks . As a consequence , selection will be weak and lead to an increase in genetic variation through the accumulation of mutations , though the fixation of any specific mutation would be unlikely . Identifying adaptive mutations in such genetically diverse populations can be difficult . However , we reasoned that adaptive mutations should evolve repeatedly in independent populations , while neutral or deleterious mutations should not show any discernible degree of parallelism . Parallel evolution has been readily observed in nature regardless of the ecological complexity [11] , [27] , [28] . Such adaptive , phenotypic convergence can be based on different underlying genetic changes , such that adaptive , parallel changes can occur in the same gene , or in different genes of the same pathway or functional group [12] , [15] . In an environment where selection is weak and selective differences are small , it is hard to imagine a scenario where an individual can quickly accumulate mutations along a multi-protein pathway that will lead to increased fitness . Instead , mutations in regulatory genes such as transcriptional or translational regulators that can simultaneously affect many operons or entire pathways could produce much larger benefits and circumvent potential complications from epistatic interactions among different mutations . One example of a gene with such large pleiotropic effects is the global stress response regulator rpoS , which is activated during late exponential and stationary phase [29] . Mutations in rpoS are often among the first mutations to evolve during experimental evolution of E . coli [30] and have been routinely observed in different selective conditions [31]–[33] . These mutations lead to changes in the stress response and nutrient acquisition [34] , change the stress induced mutation rates [35] , [36] and increase long-term viability [37] . Knocking out rpoS leads to a down-regulation of the starvation stress response and efflux pumps , and to increased nutrient efficiency via the up-regulation of proteins such as porins . The trade-off of stress resistance and nutritional competence was termed the SPANC balance ( self preservation and nutritional competence ) by Ferenci [34] . While the prevalence of rpoS knockout mutants is low in wild isolates [38] , considerable variation in rpoS expression has been observed among wild strains [39] . We therefore hypothesized that mutations in global regulators could be especially beneficial in ecologically complex , nutrient-rich environments that induce weak selection . In contrast to previous experiments where laboratory adapted strains were evolved in rich media commonly used in the laboratory [40] , we isolated naïve strains from their natural habitat , the gut of healthy humans , and used rich media as novel , selective environments . In the gut , E . coli and C . freundii interact with hundreds of other strains as well as their human host . More than 90% of these commensal gut bacteria are anaerobes , which convert non-digestible complex carbohydrates into short-chain fatty acids and produce the simple mono- and di-saccharides favored by E . coli [41] , [42] . In return , facultative anaerobes like E . coli and C . freundii play an important role in maintaining a low-oxygen environment . We chose two complex media ( BBL BHI and LB Miller ) that differed primarily in the composition and amounts of amino acids , vitamins and carbohydrates ( S1 Text ) . In addition to the populations selected in complex media ( two genotypes in two environments resulting in four treatments , Fig . 1 ) , we also performed a control experiment , where we reduced selection as much as possible by daily bottlenecking the population to a single cell , the approach commonly used for mutation accumulation ( MA ) experiments [43]–[45] . In mutation accumulation experiments , independently evolved lines are expected to accumulate a random set of mutations with little to no parallel evolution . We used a powerful combination of population proteomics and population genomics to reveal phenotypic convergence to identify potential biochemical mechanisms of adaptation . Despite the complexities imposed by the tremendous amount of underlying genetic diversity accumulated during adaptation to complex nutrient rich environments , we identified clear genomic signatures of adaptation across and within independently evolved populations . Strikingly , changes in the global regulators arcA and rpoS evolved consistently , while changes in other global regulators were largely absent . Subsequent proteomic and carbohydrate analysis of populations adapting to BHI showed increased abundance of enzymes associated with the TCA cycle and amino acid metabolism to make use of abundant amino acids , resulting in the secretion of the polyamine putrescine as a nitrogen sink . Thus in complex media , where the adaptive landscape is relatively flat and has many potential modest peaks requiring many changes to produce a substantive increase in fitness , the “go to” strategy may be to use global regulators such as arcA and rpoS to overcome epistasis by changing the regulation of whole metabolic pathways in a coordinated manner . This allows populations to rapidly reprogram resource utilization and to adapt to complex fitness landscapes in a much smaller number of moves .
We isolated E . coli RU1 and C . freundii RU2 ( S1 Figure ) de novo from the gut flora of healthy humans using only two overnight growths on agar to reduce any selection prior to our adaptation experiment . For each species , 12 populations were established and allowed to adapt over a minimum of 500 generations to media and conditions that were very rich in resources and substantially different than the environment of the human gut ( Fig . 1 ) . We chose LB and BHI as novel environments since they are likely to be very different from the gut resource base , but still support robust growth of both ancestral strains ( S1 Table ) . Over the course of the selection experiments , we observed modest but significant changes in various fitness components , consistent with adaptation under weak selection conditions . While lag time decreased in most treatments , maximum growth remained constant in LB and decreased in BHI ( S1 Table ) . In all but one treatment , LB-evolved C . freundii , the populations significantly increased their stationary phase density ( OD600 ) indicating enhanced abilities to utilize the resources efficiently . While we observed significant changes , the differences between ancestor and evolved populations were modest and consistent with permissive environments inducing weak selection pressures ( S1 Table , S1 Text ) . As a consequence of weak selection , considerable genetic variation evolved over the course of our experiment . This was evident both at the phenotypic as well as the genotypic level . We observed considerable phenotypic variation in colony size and in the ability to utilize arabinose ( Fig . 2 ) , in redox activity , in exopolysaccharide content and loss of motility ( S2A-D Figure , respectively ) . Interestingly , evolved E . coli populations had at least one colony among the 8 colonies assessed per population that lost motility , but only one single C . freundii colony out of all the colonies assessed ( two sets of 96 colonies in total ) lost motility ( S1 Text ) . To assess the evolved genetic variation and identify adaptive mutations , we sequenced the evolved populations and identified mutations in coding regions that occurred at a minimum frequency of 0 . 05 in a population . Two BHI-evolved populations ( E . coli BHI5 and C . freundii BHI20 ) could not be aligned properly and were omitted from further genomic analyses . The number of mutations ranged from 29 to 725 per population . The number of mutations per population did not differ significantly among the E . coli populations evolved in LB or BHI ( Fig . 3; Table 1 ) . Two populations evolved to become mutators in each environment ( LB4 , LB11 , BHI6 , and BHI10 , S1 Text ) . If the mutator populations are excluded , the average number of mutations between the LB and BHI-evolved populations was reduced , although there was still no significant difference in the number of mutations across environments . In contrast , the LB-evolved C . freundii populations accumulated significantly more mutations than the BHI-evolved populations . Overall , the number of mutations differed significantly both between media and species ( Full factor ANOVA with Media and Strain as fixed factors: Media F1 , 42 = 15 . 1 , p = 0 . 0004 , Species F1 , 42 = 4 . 5 , p = 0 . 039 , Media×Species F1 , 42 = 9 . 5 , p = 0 . 0036 ) . While synonymous mutations can have fitness effects [46] , [47] , we focused our analyses on non-synonymous mutations , which include SNPs , insertions , and deletions . The number of non-synonymous mutations ranged from 5 to 198 in a population , with more mutations arising in the LB than in BHI in the E . coli population ( Fig . 3 , Table 1 ) . Excluding the mutator populations reduced the average non-synonymous mutations per population further ( LB: 21±15 , BHI: 12±7 ) . Among the C . freundii populations , the average number of non-synonymous mutations was significantly higher in the LB-evolved populations than in the BHI-evolved populations . Again , we observed significant differences among media and species ( Media = F1 , 42 = 23 . 8 , p<0 . 0001 , Species: F1 , 42 = 13 . 2 , p = 0 . 0007 , Media×Species: F1 , 42 = 9 . 4 , p = 0 . 0037 ) . The accumulation of largely non-adaptive mutations complicates the identification of adaptive changes within a single , polymorphic population . However , we expected that important adaptive trajectories would exist across independently evolved populations . Therefore , we focused our analyses on parallel , non-synonymous mutations that evolved consistently across populations , both within and across species and media . Most mutations occurred in only one or a few populations , consistent with the presence of large non-adaptive genetic variation ( Fig . 4 ) . Strong parallel evolution across environments and species occurred in the global regulator arcA , which acquired mutations in all 24 LB-evolved populations and in nine of eleven BHI-evolved E . coli populations ( Fig . 5 ) . The probability that mutations evolved in the same gene in 24 independently evolved populations at random is very small considering that E . coli RU1 and C . freundii RU2 had 4565 and 5068 annotated genes , respectively ( p = ( 1/4565 ) 12 * ( 1/5068 ) 12 ) and suggests that these mutations are adaptive . Surprisingly , C . freundii adaptation to BHI did not implicate arcA . The second most commonly mutated gene was the global stress response regulator rpoS [29] , [48] , which had mutations in nine of 23 E . coli and eight of 23 C . freundii populations . None of the mutation accumulation lines had mutations in arcA or rpoS . Besides arcA and rpoS , only a few other genes acquired mutations in replicated populations , and unlike arcA and rpoS , these other mutations occurred only within a treatment and not across species and selective environment . No other mutation evolved with any degree of parallelism in the E . coli populations . Among the C . freundii populations , mutations in a gene encoding the Valine-Glycine Repeat Protein G , vgrG , a homolog to the tailspike of bacteriophage T4 , and in a gene encoding adenosylmethionine-8-amino-7-oxononanoate aminotransferase , an enzyme involved in biotin biosynthesis , occurred in almost all evolved populations in both selective environments , while mutations in different mobile elements , in the peptide deformylase , the methionyl-tRNA formyltransferase and the sodium/glutamate symport protein were only common among the LB-evolved C . freundii populations ( for further details see S1 Text ) . The highly parallel evolution of mutations in arcA and rpoS combined with their global effects suggests that these mutations are driving adaptation in these complex selective environments . Mutations in these genes were very common with multiple different alleles co-occurring within the same population . The cumulative frequencies of arcA mutations in particular reached high frequencies in LB ( average 0 . 75±0 . 08 ( mean and 95%CI ) across 24 populations ( Fig . 6 ) . In only one population did we observe the fixation of a single arcA mutation ( LB5 ) . We observed 46 unique mutations in arcA , both within and among populations ( Fig . 7 ) . Strikingly , none of these mutations introduced a stop codon or a frame shift; 44 of these 46 unique mutations were non-synonymous substitutions , one mutation resulted in a C-terminal deletion of three amino acids , and one mutation was an insertion of one amino acid . To independently confirm some of the mutations identified from population genomics , we directly sequenced arcA from eight single colonies isolated from six of the LB-evolved E . coli populations . We were able to confirm eleven of the 46 mutations identified in the whole population samples ( L2: I122M , Y137C , I22S , L4: N116T , L6: R16H , A76T; L8: E94K; L10: A25T , G59S , L50Q; and L12: L50Q ) . In addition , we identified two new mutations ( L8: G62D and 218ΔTPE; and L12: G62D ) suggesting that our cutoff of 5% in the deep sequencing population analysis still missed many arcA variants . Each clone had only one mutation in arcA , suggesting that the one mutation was sufficient to achieve a beneficial effect . The response regulator arcA is part of the two-component arcAB signal transduction system . The membrane bound sensor kinase ArcB phosphorylates ArcA ( ArcA-P ) in response to a variety of environmental challenges to maintain redox and metabolic homeostasis [49]–[51] . Mutations in the sensor kinase arcB could also affect the regulation of arcA . We therefore examined the whole genome sequencing data for mutations in the sensor kinase arcB and found mutations in six of the BHI-evolved ( BHI1 , BHI2 , BHI3 , BHI4 , BHI9 and BHI10 ) and in one of the LB-evolved E . coli populations ( LB8 ) . Among the C . freundii populations , mutations in arcB were less frequent , with only one population evolving mutations in arcB in each environment ( LB26 and BHI24 ) . As with arcA , all mutations were non-synonymous , though one mutation did result in a frame shift . Importantly , neither arcA nor arcB evolved mutations in the mutation accumulation lines suggesting that changes in arcAB are under selection and not random . The amount of variation in rpoS mutations was not quite as dramatic as in arcA , but substantial nonetheless . In the LB-evolved E . coli populations we predominately observed SNPs , while in the BHI-evolved E . coli populations mutations in rpoS resulted in stop codons , large deletions or frame shift mutations , suggesting a loss of function ( S3 Figure ) . Among the five LB-evolved and three BHI-evolved C . freundii populations with mutations in rpoS , only one population ( LB31 ) acquired a mutation resulting in a stop codon while the rest acquired substitutions . To confirm some of these mutations independently , we sequenced rpoS from eight single colony isolates for three of the LB-evolved populations ( LB6 , LB8 and LB12 ) and confirmed the A199T mutation in population LB8 as well as a new rpoS mutation ( Y283C ) in LB6 . Unlike arcA , we observed many loss of function mutations in rpoS , which is consistent with previous selection experiments where knock-out mutations evolved relatively rapidly under different selective conditions [31]–[33] . RpoS levels could also be attenuated through changes in the regulation of its expression . The expression of rpoS is repressed during exponential growth and activated upon starvation during late exponential and stationary phase using different transcriptional and translational mechanisms . Transcription of rpoS is up-regulated through spoT/ ( p ) ppGpp and BarA/UvrY and repressed by arcAB , while translation is up-regulated by two small RNAs , DsrA and RprA and repressed by a third small RNA ArcZ [29] . Presumably , mutations in any of these genes could also affect the up-regulation of rpoS and lead to reduced expression , resulting in similar phenotypes as the knockout mutants . We did not observe any mutations in spoT , though mutations in this gene evolve readily in minimal media [6] , [52] , [53] . Four of the LB-evolved C . freundii populations ( LB27 , LB32 , LB35 , LB36 ) , however , had a frame shift mutation in barA/uvrY . Three of these populations did not have a mutation in rpoS , suggesting that the effect of loss of function mutations in barA/uvrY could lead to reduced transcription of rpoS and result in a similar phenotype to rpoS knockout mutations . Mutations in the small RNAs DsrA and RprA could also lead to reduced translation of rpoS and result in a similar phenotype as rpoS knockout mutants . We looked for mutations in DsrA and RprA in E . coli and did not detect any mutations in these small RNAs . While we also searched for the small RNAs in C . freundii using sequences retrieved from Citrobacter , we were unable to locate the two small RNAs in our reference genome . Linking mutations to functional changes across regulatory sequences is more difficult as it requires excellent annotation and understanding of the transcriptional regulators . Nonetheless , it is certainly likely that changes in regulatory regions could provide adaptive changes . To test whether arcA or rpoS expression were altered by mutations within their regulatory regions , we examined the 500 nucleotides preceding the start codons of these two genes and found no mutation in any of the evolved populations or the MA lines . Population-level proteomic analysis of the BHI-evolved E . coli populations showed significant changes in protein abundance between the ancestor and evolved populations as well as a remarkable degree of parallelism among the evolved populations . We observed significant and highly parallel decreases of ArcA abundance in the evolved populations and increases of proteins of the TCA cycle , amino acid metabolism and transporters ( Fig . 8 ) . We identified 4469 unique peptides in 39 samples ( ancestor and twelve evolved populations with three replicates each ) , corresponding to 488 proteins ( see S1 Text for more details ) . Quantitative analysis of the 488 proteins revealed 166 proteins that were significantly different between the ancestor and the evolved populations ( p<0 . 01; log2-fold change>±0 . 7 ) . Of those , 58 proteins decreased and 108 proteins increased significantly over the course of the selection experiment ( S2 Table ) . All observed proteins associated with the TCA cycle ( aconitate hydratase , isocitrate dehydrogenase , 2-oxoglutarate dehydrogenase , succinyl-CoA ligase , succinate dehydrogenase , fumarate hydratase , and malate dehydrogenase ) and glyoxylate shunt ( isocitrate lyase ) significantly increased ( Fig . 8 ) . The up-regulation of the TCA cycle and the decrease in ArcA is consistent with previous studies that observed increased flux through the TCA cycles in arcA knock-out mutants [50] , [54] . RpoS was not detected in our proteomic analyses in any of the samples and therefore we cannot draw any conclusions about its abundance . Reduced starvation stress is associated with increased nutrient acquisition and metabolism and reduced stress responses [34] – conditions we expected in our resource rich environments . Altered resource utilization can be developed by increasing C/N acquisition via the up-regulation of porins , which allow nutrients to flow through the outer membrane , and a concomitant decrease of the effluxers that provide protection from toxins during starvation stress [34] , [55] . Consistent with increased C/N acquisition from amino acids and small peptides , we observed significant increases of peptidases ( alpha-aspartyl dipeptidase peptidase E , peptidase B , and methionine aminopeptidase ) and of proteins associated with ABC transporter systems responsible for the transport of amino acids or peptides ( glutamate aspartate ( GltI ) , lysine-arginine-ornithine ( ArgT ) , glutamine ( GlnH ) , histidine ( HisJ ) and oligopeptide sytems ( OppA ) ) , and carbohydrates ( galactose/methyl galactoside ( MglB ) , ribose ( RbsB ) , maltose/maltodextrin ( MalE ) ) . Genomic analyses suggested some loss of function among specific efflux pumps consistent with low stress conditions . We identified mutations in several RND efflux pumps including cmeA and cmeB that are found in multiple copies within the E . coli and C . freundii genomes . Mutations in cmeA and cmeB ranged in frequency from 0 . 05 to 0 . 41 and occurred in 15 of 24 populations across both environments and organisms , suggesting that decreases in CmeA and CmeB function are under selection during adaptation . Eight out of twelve LB-evolved E . coli populations acquired mutations in either cmeA or cmeB . Mutations in cmeA that resulted in likely loss of function ( all either insertions , deletions or SNPs to stop codons ) evolved in four populations ( LB5 , LB9 , LB11 and LB12 ) , while five different populations had mutations in one of the cmeB copies ( LB1 , LB4 , LB5 , LB7 and LB8 ) . Mutations in cmeA and cmeB were not as prevalent among the LB-evolved and completely absent among the BHI-evolved C . freundii populations . One LB-evolved C . freundii population acquired a substitution in cmeB ( LB25 ) , one had an insertion ( LB34 ) and a third population had an insertion in the RND efflux transporter ( LB2 ) . One MA line acquired an insertion in both cmeA and cmeB . The cmeA and B mutations resulting in loss of function mutations support the SPANC balance conditions of low starvation stress and increased nutrient uptake and decreased efflux . While rpoS mutants are predicted to have a decreased stress response , our proteomic data ( Fig . 8 , S2 Table ) suggested that changes in the stress response , were more nuanced and that some stress pathways , such as the starvation and acid stress responses were up-regulated while others such as protein unfolding stress were diminished . Across the twelve BHI-evolved E . coli populations , we observed decreases in chaperones associated with protein folding stress ( DnaK ) and heat shock proteins ( GroES ) , and in proteins involved in the oxidative stress response through glutathione ( glutaredoxin 2 and 3 , and glutathione peroxidase ) . Conversely , proteins involved in the oxidative stress response through thioredoxin ( thioredoxin reductase , universal stress proteins AEFG , superoxide dismutase , glutathione S-transferase ) , acid stress ( HdeAB ) and another heat shock protein ( HchA/Hsp31 ) increased . The up-regulation of the TCA cycle and the increased amino acid acquisition and metabolism could lead to increased production of ammonia or polyamines to maintain nitrogen homeostasis . To test this hypothesis , we determined whether the evolved populations produced and secreted more polyamines or ammonia . We began by testing the pH of spent media after 24 hours of growth , and observed a significant increase in pH from 8 . 1 to 8 . 4 in the BHI-evolved E . coli populations ( S1 Table ) compared to the ancestor . Similarly , the pH also increased significantly in the LB-evolved C . freundii populations , but not in the other two treatments . Increased pH is consistent with proteomic data that suggested significantly increased TCA cycle activity and amino acid metabolism . The breakdown of amino acids by the decarboxylation of ornithine or of arginine to agmatine can result in the production of the polyamine putrescine [56] . Indeed , putrescine was significantly higher in the spent media of the BHI-evolved E . coli population compared to the ancestor ( t-test: t = 6 . 08 , df = 22 , p<0 . 0001 ) , but not in the cell extract ( t-test: t = 0 . 3 , df = 20 , p = 0 . 76 ) ( S4 Figure ) . While we only have quantitative data for the BHI-evolved E . coli populations , the odor of the C . freundii populations at stationary phase suggested that they , too , all produced and secreted increased amounts of putrescine .
The natural world presents organisms with complex and variable environments . Resources often range from rich and varied , to poor and limiting . An abundance of new , but usable , resources may induce very weak selection pressures and result in a complex multi-peaked adaptive landscape , where most single nucleotide changes or mutations to specific components of a metabolic pathway would not generate enough fitness gains to facilitate rapid success . One path to adaptation would be to change the global regulators that control the management of metabolic flux to provide a simple “one-step” adaptation for the entire physiology of the organism . Adaptation by such a one-step mechanism constitutes a higher order ‘metabolic selection’ that allows the organism to capture larger gains in fitness and circumvent the complications of multi-gene epistasis . To test this idea , we used wild isolates of E . coli and C . freundii and investigated their adaptive responses under weak selection as they were moved from their natural habitat , the human gut , to a rich and markedly different resource base . We found that , as expected , weak selection induced by rich complex environments resulted in large genetic variation and likely allowed even deleterious mutations to persist . We observed a striking diversity of phenotypes across all populations . Underlying genetic diversity could be observed readily as a tremendous variation in colony sizes and physical appearance on different indicator agar plates , as well as loss of motility ( Fig . 2 and S1 Text ) . Whole population sequencing of the evolved populations identified arcA and rpoS as the targets of selection . Whole population proteomics of the BHI-evolved E . coli populations showed that these populations up-regulated several amino acid and carbohydrate transporters to move abundant nutrients into the cell and up-regulated the proteins of the TCA cycle needed to use them efficiently ( Fig . 9 ) . We also observed significantly increasing putrescine production consistent with increased utilization of amino acids as C/N sources . The combination of whole genome sequencing and whole population proteomics proved to be a powerful approach for the mapping of genotypic changes to biochemical mechanisms that , in turn , produce altered phenotypes . To identify common adaptive strategies , we focused on mutations that arose repeatedly in independently evolved populations . The two most common targets of selection were arcA and rpoS , both global regulators with large pleiotropic effects . Our overall picture for adaptation is one in which the adaptation through mutations in the global regulators arcA and rpoS drive the large metabolic changes essential for adaptation to nutrient rich environments under these selection conditions . Both of these global regulators affect up to 10% of the genes within their host genome [29] , [49] , [50] . Mutations in arcA evolved consistently in the majority of the populations . ArcA consists of two domains , the receiver domain ( residues 1–123 ) that includes the site of phosphorylation ( Asp54 ) [57]-[59] and a DNA binding domain ( 124–238 ) [59] . Phosphorylation stimulates formation of an ArcA-P dimer that binds to a variety of specific DNA sequence motifs with high affinity to repress or activate transcription of up to 229 operons directly or indirectly in response to the environment [49] , [50] , [54] . The majority of the diverse mutations in arcA were found in the receiver domain ( Fig . 7 ) . Mapping these mutations onto the three-dimensional crystal structure ( 1XHE ) of the receiver domain revealed that the vast majority of mutations are in surface positions and solvent accessible loops ( S5 Figure ) , with only a few mutations mapping to the hydrophobic core . While it is likely that some of these mutations could result in a complete loss of function , the likelihood that all 46 mutations do so is slim . It is interesting that all but two mutations in arcA were SNPs and the two exceptions were an insertion and a deletion at the C-terminus and likely resulting in a largely functional protein . This suggests that a complete loss-of-function that eliminates arcA function is not as beneficial as modifying its activity and is consistent with our proteomic data that showed a decrease in ArcA levels rather than a complete absence in ArcA . Mutations within the receiver domain could decrease ArcA signaling by a number of mechanisms including: 1 ) reducing ArcA stability; 2 ) decreasing the extent of ArcA phosphorylation during signaling; 3 ) increasing the rate of dephosphorylation; or 4 ) decreasing the extent of phosphorylation-dependent oligomerization . Only a few mutations mapped to the DNA binding domain but these could also alter ArcA function by decreasing DNA binding or any of the aforementioned mechanisms for altered receiver domain function . Mutations to arcA were also observed in previous selection experiments , notably during adaptation to glucose limited media [60] , [61] and LB [40] , [62] . Interestingly , in those studies mutations in arcA only evolved in the aerobic cultures , suggesting that oxygen deprivation and anaerobic stress were not the driver for arcA mutations . Unlike arcA , it was striking that all mutations to rpoS in the BHI-evolved E . coli populations and only one mutation in the other three treatments appeared to produce a complete loss of function . One explanation could be the differences in the composition of the media , mainly the presence of glucose in BHI . Carbohydrates and glucose in particular are utilized first before switching to amino acids [63]–[65] . This is reflected in a diauxic growth pattern with a second lag and exponential growth phase . The depletion of carbohydrates could induce the RpoS regulated starvation stress response , which might delay the transition to amino acid metabolism . Losing RpoS function might therefore be beneficial to a fast switching response to other nutrient sources . In LB we never observed diauxic growth patterns , which is consistent with the low concentration of carbohydrates in the media . Mutations in rpoS have also been shown to improve longevity during stationary phase [37] . The populations reached stationary phase within eight hours in LB and as a consequence , these populations remained in stationary phase much longer . SNPs that improve persistence in stationary phase could therefore be selected in LB . Proteomic analyses of BHI-evolved E . coli populations showed increased abundance of enzymes of the TCA cycle , which is consistent with the known phenotypes of arcA and rpoS knockout mutants based on flux analyses [50] , [54] , [66] . Knocking out either rpoS or arcA resulted in two-fold increases in metabolic flux through the TCA cycle , while knocking out arcB did not have an effect on the flux through the TCA cycle , consistent with our observation that mutations in arcB were rare [54] , [66] . While proteomics and flux analyses use different measurements , the effect of mutations in arcA and rpoS seem very similar . This is even more remarkable considering that our proteomic analyses are based on polymorphic populations and the small changes in proteomics are population averages . In addition , we see strong up-regulation of peptidases , amino acid metabolism , amino acid transporters and oxidative stress responses , indicating that these populations are metabolizing the media at an increased rate ( Fig . 8 and 9 , S2 Table ) . Again , this pattern is consistent with previous metabolic studies [54] , [66] . ArcA has been shown to either directly or indirectly regulate many operons such as amino acid and polyamine production , beta-oxidation of fatty acid and operons encoding pathways for the utilization of aromatic compounds and peptides [49] . Knock-out mutants of rpoS not only had increased TCA cycle activity but also increased amino acid metabolism [66] . The arginine , asparagine and glutamine metabolism pathways and the TCA cycle feed into the urea cycle . While we did not observe changes in expression of enzymes of the urea cycle , we observed a significant increase in the production and secretion of putrescine in BHI-adapted population . The polyamine putrescine can be produced during the breakdown of amino acids by the decarboxylation of ornithine or by the decarboxylation of arginine via agmatine [56] . Arginine decarboxylase has been proposed to localize in the cell envelope , where it converts exogenous arginine to putrescine via agmatine [67] . Because we observed a significant decrease of agmatinase in the evolved populations , it seems more likely that putrescine is produced from ornithine , a component of the urea cycle . In the adapted populations , putrescine might be acting as a nitrogen sink for catabolism of amino acids via the urea cycle or , alternatively , may help with the increased oxidative stress resulting from increased metabolism . There are many roles for polyamines in metabolism including as C/N sources , oxidative stress response , and signaling [68]–[73] , so an understanding of increased secretion of putrescine by our adaptive populations will require further biochemical studies . We expected that parallel evolution could also evolve along pathways and lead to phenotypic convergence by mutating different genes along the pathways [12] , [15] . Indeed , we did see some convergence in the regulation of rpoS , where not all populations had mutations in rpoS and instead had mutations in genes that regulate the expression of rpoS . We interrogated the genomic data for such phenotypic convergence by analyzing parallel evolution for different functional categories , but did not observe any evidence for phenotypic convergence at different levels of increasing complexity ( see S1 Text , S6–S8 Figure ) . Nonetheless , we did see a high degree of parallel evolution among the populations when we grouped the proteins with significant changes to pathways . This is even more remarkable considering that we performed our proteomic analyses on whole , polymorphic populations and as such only measure the average change of a populations compared to the ancestor . This parallelism shows a clear response to selection . The global up-regulation of metabolism and the lack of clear phenotypic convergence of mutations along the pathway further support our assertion that mutations in arcA and rpoS drive adaptation to the rich selective conditions and lead to the observed metabolic changes . These relatively small changes observed in the proteomic data also further support our previous observations that very small biochemical changes can have large fitness effects [74] and are likely very relevant to adaptation in nature . The consistent evolution of mutations in global regulators arcA and rpoS that each affect expression of about 10% of the genome supports the model in which adaptation evolved through the evolution of a few mutations with large beneficial effects . Instead of acquiring beneficial mutations in every gene involved in the TCA cycle and the various amino acid metabolism pathways , acquiring mutations in regulators affecting all these genes simultaneous is undoubtedly more efficient . Selection studies in other , less complex environments also implicated mutations in global regulators that lead to stable coexistence and polymorphic populations suggesting that mutations in global regulators are beneficial in different environments [60] , [61] . In contrast to those earlier studies under glucose-limiting conditions , mutations to arcA and rpoS arose very rapidly and repeatedly across our populations . Diverse phenotypes and genotypes suggest that our populations are polymorphic as well . For example , the sequential utilization of carbohydrates and amino acids [64] , [65] could select for different mutations specialized to either carbohydrate or amino acid utilizations , similar to the stable coexistence of different temporal specialist observed in previous studies [10] , [19] , [20] , [25] . It is possible that arcA and rpoS mutations provide selective benefits in different phases of the growth cycle and the coexistence of these mutations in the populations could be an indication of such temporal and potentially nutritional specialization . By investigating adaptation of wild organisms to resource rich environments we have shown that adaptation occurs within five hundred generations through mutations in global regulators , leading to increased rates of metabolism . Mutations to global regulators might be more common during selection in permissive environments . At niche boundaries such as a thermal limits , single mutations could greatly increase the fitness of an organism [75] , mostly because fitness at the niche boundaries can be dramatically reduced compared to the niche optimum [76] , [77] and thus a small number of mutations or even a single mutation can result in substantial fitness gains . Both rpoS and arcA have been linked to virulence [78] , [79] . Our findings suggest that it is important to appreciate the role of laboratory adaptation when evaluating strains for pathogenicity , especially in light of the fact that rpoS loss-of-function mutations evolve readily in the laboratory but are not found in natural populations [38] . The transition from the natural habitat to laboratory conditions suggests how we might improve experimental evolution studies that require handling and adaptation of pathogens as well as provide a starting point for forensic attribution of strains during outbreaks of novel pathogens .
E . coli RU1 ( hence forth referred to as E . coli ) and C . freundii RU2 ( hence forth referred to as C . freundii ) were isolated from the stool of healthy humans ( S1 Text , S5 Figure ) . To minimize any potential for adaptation during the isolation process , we plated stool samples on MacConkey agar plates . After a single overnight growth , half of a single colony was flash frozen at −80°C in Trypticase soy broth with 20% Glycerol ( BD , USA ) and the other half was used for phenotypic strain characterization . Strain identification by 16S sequencing was done from the frozen sample . All experimental evolution studies started from the frozen primary isolates by using a single clonalized colony derived from the initial snap frozen isolate . The identity of the wild "un-adapted" strains was based on the results of API 20 E ( Biomerieux , USA ) test strips and species-specific PCR ( using forward primer: AGAGTTTGATCMTGGCTCAG , reverse primer: GWATTACCGCGGCKGCTG ) . Assays were performed either at the population level or the single colony level . Cells or populations were grown in their selective media ( LB or BHI ) and grown in liquid media or plated on agar plates made with their selective media , unless otherwise stated . Adaptation to the selective environments was assessed as changes in lag time and growth rate by measuring OD600 over 24 hours of growth in liquid media following Walkiewicz et al . [74] . To test for changes in the pH of spent media , we grew the populations to stationary phase and measured the pH of the media after removing the cells . To test for genetic variation within the populations , we plated the populations at low density on tetrazolium arabinose plates and observed considerable variation in both colony size and in the ability to utilize arabinose . We plated the populations on the selective media supplemented with agar and isolated eight randomly chosen colonies from each of the 12 populations per treatment . These test sets of 96 individual isolates per species and environment were used for three phenotypic assays: 1 ) the redox state by plating on methylene blue ( 0 . 065 g/liter ) ; 2 ) differences in exopolysaccharides by plating on Congo Red ( 0 . 15 g/liter ) ; and 3 ) loss of motility by plating cells on soft agar ( 0 . 25% DIFCO ) . For more information see S1 Text . Cell extracts and spent media samples were prepared by growing the evolved populations and ancestral populations to stationary phase , separating the cells from the supernatant by centrifugation and inactivating remaining cells with 70% ( v/v ) ethyl alcohol . | Changing environmental conditions are the norm in biology . However , understanding adaptation to complex environments presents many challenges . For example , adaptation to resource-rich environments can potentially have many successful evolutionary trajectories to increased fitness . Even in conditions of plenty , the utilization of numerous but novel resources can require multiple mutations before a benefit is accrued . We evolved two bacterial species isolated from the gut of healthy humans in two different , resource-rich media commonly used in the laboratory . We anticipated that under weak selection the population would evolve tremendous genetic diversity . Despite such a complex genetic background we were able to identify a strong degree of parallel evolution and using a combination of population proteomic and population genomic approaches we show that two global regulators , arcA and rpoS , are the principle targets of selection . Up-regulation of the different metabolic pathways that are controlled by these global regulators in combination with up-regulation of transporters that transport nutrients into the cell revealed increased use of the novel resources . Thus global regulators can provide a one-step model to shift metabolism efficiently and provide rapid a one-step reprogramming of the cell metabolic profile . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"functional",
"genomics",
"metabolic",
"processes",
"parallel",
"evolution",
"evolutionary",
"adaptation",
"catabolism",
"microbial",
"genetics",
"microbial",
"genomics",
"amino",
"acid",
"metabolism",
"proteins",
"metabolic",
"pathways",
"proteomics",
"genetic",
"drift",
"biochemistry",
"genetics",
"biology",
"and",
"life",
"sciences",
"citric",
"acid",
"cycle",
"genomics",
"evolutionary",
"biology",
"metabolism",
"evolutionary",
"processes"
] | 2014 | Mutations in Global Regulators Lead to Metabolic Selection during Adaptation to Complex Environments |
Entamoeba histolytica causes intestinal disease in endemic settings throughout the world . Diagnosis of E . histolytica infection would be improved by the identification of biomarkers that are expressed by cysts of E . histolytica , but not by cysts of closely related commensal species of Entamoeba . Herein , we describe two novel monoclonal antibodies ( 1A4 and 1D3 ) produced against a spacer region of the E . histolytica Jacob2 lectin , an outer cyst wall protein . These reagents demonstrated no cross-reaction to E . dispar recombinant antigen and low picomolar molecular detection limits when paired in ELISA sandwich assays . In an immunofluorescence microscopy assay , the α-Jacob2 murine antibodies labeled cysts of three xenically cultured E . histolytica isolates but did not label cysts of three E . bangladeshi isolates . Monoclonal antibody 1A4 did not cross-react with xenic cultures of three E . dispar isolates , demonstrating specificity to E . histolytica , while monoclonal antibody 1D3 cross-reacted with two out of three E . dispar isolates . Both antibodies labeled cysts in formalin-fixed slides , a potential logistical advantage in some settings . The monoclonal antibody 1A4 was also used in an immunofluorescence microscopy assay with formalin-fixed stool specimens . Seven out of ten ELISA-positive stool specimens exhibited 1A4-labeled cyst-like objects , compared to one out of seven ELISA-negative specimens . These results demonstrate that antibodies generated against the flexible spacer of E . histolytica Jacob2 lectin recognize and bind to Jacob2 protein in whole cysts and are capable of differentiating Entamoeba species in fixed specimens . Thus , Jacob2 is a promising biomarker for use in diagnosing E . histolytica infection .
Entamoeba histolytica is a protozoan parasite that causes an estimated 30–50 million cases of illness and kills 100 , 000 people annually [1 , 2] . It has a dual-stage life cycle , consisting of motile trophozoites that colonize and invade the colonic epithelium and of robust , chitinaceous cysts that enter and exit the human body via fecal-oral transmission [3 , 4] . Infection is often asymptomatic , or it can lead to clinical manifestations that include dysentery , colitis , or extraintestinal abscesses [3–5] . Symptoms often resemble other enteric diseases caused by bacteria and viruses , as well as inflammatory bowel disease [3 , 4] . Control of this organism is particularly important for young children in endemic regions , whose nutrition , growth , and development are negatively impacted by enteric infections and repeated diarrheal episodes [6–8] . Detecting E . histolytica can be challenging due to the numerous commensal amoeba that colonize humans , some of which look morphologically identical to the pathogen [9–11] . Much work has been undertaken to identify and characterize superior biomarkers of infection in stool and of invasive disease in serum and abscess fluid . One such marker is the galactose/N-acetyl galactosamine ( Gal/GalNAc ) lectin , an adhesion factor that is important to E . histolytica trophozoite invasion [12] . This protein is the target of two antigen capture assays that have been widely and successfully used to specifically detect E . histolytica infections in fresh stool or liver abscesses [13–15] . However , these tests are unable to detect the cyst stage of the parasite , and they cannot be applied to formalin-fixed specimens [9 , 15 , 16] . The instability of shed trophozoites in unfixed specimens necessitates prompt transport and testing , a significant logistical limitation in some settings [2 , 9 , 15 , 16] . These limitations may have an impact on E . histolytica diagnosis , as one of the tests was found to be at best 79% sensitive relative to the more sensitive qPCR method [17] . A cyst wall lectin with diagnostic potential , named Jacob2 , was described recently [18] . It is one of a few proteins known to be expressed only in the Entamoeba cyst stage [19–21] . In the “wattle and daub” model of Entamoeba encystation , the Jacob2 protein first appears in intracellular vesicles and is secreted through the plasma membrane [19 , 22] . Then , it is tethered by the Gal/GalNAc lectin and binds to chitin to form the cyst wall via three , chitin-binding domains [18–20 , 22 , 23] . Between these domains is a flexible , serine-rich spacer sequence with an amino acid sequence dissimilar between E . histolytica and commensal non-pathogenic species E . dispar [18 , 19] . This sequence was also noted to be polymorphic between evaluated E . histolytica strains [18 , 19] . Nevertheless , Jacob2 and other cell wall components could potentially be utilized to distinguish E . histolytica from similar non-pathogenic species such as E . dispar . Here , we describe the generation of murine monoclonal antibodies against the variable spacer region of the E . histolytica Jacob2 protein . These reagents demonstrated excellent analytical sensitivity in a sandwich ELISA . More importantly , one of them bound E . histolytica cysts in xenic culture without cross-reacting to xenic isolates of the commensal species E . dispar and of the recently discovered species , Entamoeba bangladeshi , which was first described in 2012 based on genotypic and morphotypic criteria [11] . This same monoclonal antibody was effective at labeling cyst-like objects in formalin-fixed stool specimens . This study supports the feasibility of using the Jacob2 spacer region as an E . histolytica-specific cyst biomarker .
Stool specimens were collected from subjects in Dhaka , Bangladesh , with approval from the Ethics Review Committee of the International Centre for Diarrheal Disease Research , Bangladesh ( ICDDR , B ) and the Institutional Review Board ( IRB ) of the University of Washington . Informed , written consent was obtained from patients or from guardians of subjects ages 1–17 , and assent was obtained from subjects ages 8–17 . Monoclonal antibody production was approved by the Fred Hutchinson Cancer Research Center ( FHCRC ) Institutional Animal Care and Use Committee ( IACUC ) . Animal care protocols at FHCRC follow all federal guidelines , including the Public Health Service ( PHS ) Policy on Human Care and Use of Laboratory Animals , the United States Department of Agriculture ( USDA ) Animal Welfare Act , Code of Federal Regulations , Title 9 , Chapter 1 , Subchapter A—Animal Welfare , and the terms of the facility's PHS Animal Welfare Assurance . The Jacob2 sequences for E . histolytica strain HM-1:IMSS ( EHI_044500 ) and for E . dispar strain SAW760 ( EDI_246160 ) were stored on Geneious version 6 . 0 . 3 and aligned utilizing the ClustalW BLOSUM cost matrix , with a gap open cost of 10 and a gap extend cost of 0 . 1 . All cloning and expression reagents were obtained from Life Technologies ( Carlsbad , CA ) unless otherwise noted . Residues 159–481 of the E . histolytica strain HM-1:IMSS Jacob2 protein ( EhJacob ) were codon-optimized and TOPO-TA-cloned into the pET SUMO vector [24] . Residues 212–560 of the E . dispar strain SAW670 ( EdJacob ) were cloned into the Gateway pDEST17 vector with a N-terminal 6xHis tag . The constructs were expressed in BL21 ( DE3 ) Escherichia coli cells , and protein was purified via Ni-NTA resin ( Qiagen , Hilden , Germany ) . The SUMO tag on EhJacob was cleaved with SUMO protease ( Lifesensors , Malvern , PA ) and removed by Ni-NTA resin according to kit instructions . Murine monoclonal antibodies were generated by the Antibody Technology Core at the Fred Hutchinson Cancer Research Center ( Seattle , WA ) . EhJacob was dialyzed into 1x PBS and combined with three additional E . histolytica recombinant antigens ( EHI_101240 , EHI_070730 , and EHI_104630 ) for immunization . The multiplex antigen formulation was used to generate antibodies for multiple investigations including this one . Three sets of splenic cell fusions were performed: the first was from high-titer mice and was screened for specificity via an indirect ELISA , while the second and third were from lower-titer mice and were first selected for IgG secretion with a ClonePixII system ( Molecular Devices , Sunnyvale , CA ) , followed by target specific indirect ELISA screening . The isotypes were determined by indirect ELISA assays . IgG3 antibodies are not frequently found in outputs and thus were not screened , and IgM were not screened due to their typically low affinities and specificities . The 48 IgG1 and 16 IgG2 antibodies with highest binding activity ( OD >1 . 0 ) were screened as pairs in 48 x 16 matrix sandwich ELISAs . First , anti-IgG1 antibody ( 0 . 5 μg/mL , Life Technologies A10538 ) was coated overnight at 4°C to high binding , 384-well plates ( Greiner Bio-One , Kremsmünster , Austria ) . Next , the following materials were incubated for 1 hour at room temperature with subsequent three rounds of washing with PBS + 0 . 05% Tween 20 ( PBS-T ) : 1 ) 25 μL of 1:8 IgG1 supernatants; 2 ) 125 ng/mL EhJacob ( with SUMO tag ) or EdJacob antigen; 3 ) 25 μL of 1:8 IgG2 supernatants; and 4 ) 1:2000 HRP-conjugated anti-IgG2a or–IgG2b antibodies ( Life Technologies A10685 and M32507 ) . Wells were developed with ABTS ( Kirkegaard & Perry Laboratories , Gaithersburg , MD ) for 15 minutes and read at 405 nm , with threshold set at an optical density of 0 . 7 . The specificities of top antibodies were checked once again by indirect ELISA , this time incorporating two non-cognate antigens , one of which had a SUMO tag . Wells of a Nunc Maxisorp flat bottom plate ( Nunc , Rochester , NY ) were coated with 200 ng of antigen diluted in 50 mM bicarbonate buffer , pH 9 . 6 ( Sigma Aldrich , St . Louis , MO ) overnight at 4°C . After the plate was washed three times in PBS-T and blocked with Starting Block ( Life Technologies ) for 1 hour at room temperature , wells incubated with selected hybridoma supernatants diluted 1:5–1:20 in Starting Block for 1 hour at room temperature . The plate was washed again and incubated with 1:1000 of goat anti-mouse HRP ( H+L ) ( Bio-Rad Laboratories , Inc . , Heracles , CA ) for 1 hour at room temperature . One-step Ultra TMB substrate ( Life Technologies ) was added after four additional washes with PBS-T , and the reaction was stopped with 2 N hydrochloric acid upon development of color . The microplates were read at 450 nm . Wells in flat-bottom Maxisorp microplates were coated with 0 . 15 μg of purified monoclonal antibody 1A4 diluted in 50 mM bicarbonate buffer , pH 9 . 6 , overnight at 4°C . Simultaneously , an additional set of wells were coated with blank buffer to serve as a control with no capture antibody . After three washes in PBS-T , wells were blocked with 1% BSA in PBS-T for 90 minutes at room temperature . Next , a serial titration of SUMO-EhJacob , ranging 16 . 4 pg to 17 . 1 ng , was plated and incubated for 60 minutes at room temperature , followed by three washes in PBS-T . Monoclonal antibody 1D3 ( 1 . 2 μg/mL ) and 125 ng/mL HRP-conjugated goat anti-mouse IgG2a ( #ab97245 , Abcam , Cambridge , MA ) incubated for 1 hour at RT in succession , followed by three and four washes in PBS-T respectively . Finally , wells were developed with 100 μL of 1-Step Ultra TMB solution for 60–80 seconds and stopped with 2 N hydrochloric acid . The microplates were read at 450 nm . Stool specimens were obtained from children enrolled in an ongoing community-based prospective cohort study of enteric infections [25] . Diarrheal and monthly surveillance stools were examined by microscopy , and a 50 mg aliquot of any prospective Entamoeba positive feces was inoculated into a 7 mL glass McCartney bottle containing a sterile agar slant and a liquid overlay of Robinson’s media , bacto-peptone , erythromycin , and 10 mg of rice starch [26 , 27] . Bottles were incubated at 37°C for 24 hours . Next , a drop of culture from the fecal-starch layer was drawn up and examined by microscopy . If no amoeba were found , additional rice starch was added to the bottle , and the culture was incubated for an additional 48 hours before re-examination . If amoeba were identified , the culture was passaged every 48 hours by inoculation of 0 . 1 mL of fecal-starch layer into a fresh bottle of media . Entamoeba dispar and E . bangladeshi isolates were identified by PCR , whereas E . histolytica isolates were confirmed by both PCR and ELISA ( Techlab , Blacksburg , VA ) . Xenic cultures were briefly checked for the presence of Entamoeba-like organisms by light microscopy prior to slide preparation . Smears were first air dried onto Merifluor treated slides ( Meridian Bioscience , Inc . , Cincinnati , OH ) and then fixed in 10% neutral buffered formalin and in 100% methanol ( Merck , Darmstadt , Germany ) for 1 minute each . After quick washes in PBS and PBS-T , they were blocked in 1% BSA PBS-T for 1–2 minutes at room temperature . Next , the smears were stained with 2 . 9 μg/mL 1A4 or 2 . 3 μg/mL 1D3 purified monoclonal for 1 hour and with 20 μg/mL goat anti-mouse Alexa Fluor 488 ( Life Technologies ) for 30 minutes in a humid chamber at room temperature , with a brief PBS-T wash after each incubation . Finally , the smears were stained with 0 . 1% Calcofluor White M2R ( Sigma Aldrich ) for 5 minutes at room temperature in the dark . The smears were examined on an Olympus BX53 microscope under 400x magnification , and photos were acquired with an Olympus DP21 camera . Photos were taken under a UV filter at 10 millisecond exposure time and under a FITC filter at 167 millisecond exposure time . Stool specimens collected from subjects , including suspected E . histolytica cases in Dhaka , Bangladesh , were tested by using a commercial Gal/GalNAc ELISA assay ( Techlab , Blacksburg , VA ) , then archived by freezing at -70°C . For immunofluorescence microscopy using antibody 1A4 , frozen stool specimens were allowed to thaw for one hour , then were partially liquefied by the addition of PBS . The suspension was filtered through two-ply cotton gauze to remove large particulates , then smeared onto Merifluor treated slides . Dry smears were fixed in 10% neutral buffered formalin for 1 minute and then stained with primary and secondary antibodies as described above for xenic cultured amoeba . Finally , smears were stained with 1 μg/mL 4' , 6-diamidino-2-phenylindole ( DAPI; ThermoFisher Scientific , Waltham , MA ) for 5 minutes at room temperature in the dark . Stained smears were then subjected to a blinded evaluation as follows . Stained smears were examined by light and fluorescence microscopy on a Leica DMLB microscope under 400x magnification , coupled with a Leica DFC310FX digital camera . Three smears per specimen were scanned for cyst-like objects ( circular and between ten to twenty micrometers in diameter ) . The blinded examiner identified the five fields per smear that had cyst-like objects with the brightest visual Jacob2 staining ( by 1A4 primary antibody and FITC-labeled secondary antibody ) . Photos were taken under a FITC filter ( Leica 11513808 ) at an exposure time of 355 milliseconds , under a DAPI filter ( Chroma 310000v2 ) at an exposure time between 1 to 5 milliseconds , and under brightfield . This analysis yielded fifteen images per specimen ( 3 smears per specimen , 5 images per smear ) that had the strongest-staining potential cyst-like objects . Each photo taken under the UV filter ( Calcofluor or DAPI ) was sized to 1300 x 975 pixels on GIMP2 . 8 and converted to 8-bit grayscale on ImageJ ( National Institutes of Health ) [28] . Background was subtracted with a 50-pixel rolling ball radius , sliding paraboloid enabled and smoothing disabled . The image was then converted to binary after setting the threshold to ≥12 ( for cysts in xenic culture ) or ≥19 ( for raw stool specimens ) , and the watershed algorithm was applied to separate clustered particles . Resultant particles that were 1000–5000 square pixels in area and with circularity ranging 0 . 65–1 . 00 ( holes included , excluded on edge ) were labeled as cyst-like objects . Labels on the UV filter photos were transferred to the corresponding FITC filter photos , and the mean “gray value” for each labeled object was obtained as a measure of antibody ( 1A4 or 1D3 ) staining . This value is termed “mean fluorescence index” ( MFI ) in the results . Statistical analysis was conducted with Microsoft Excel and the Real Statistics Resource Pack software ( Release 3 . 5 ) . The mean gray value of each identified cyst-like object under the FITC filter , a proxy for Alexa Fluor 488 fluorescence intensity , was grouped by isolate . The overall mean fluorescence intensities of the E . histolytica isolates were compared to E . dispar isolates and E . bangladeshi isolates with the Mann-Whitney U test ( n = m = 3 , α = 0 . 05 ) . For analyzing antibody staining of cyst-like objects in stool specimens , maximum rather than mean MFI was compared between specimens , because many specimens had few or no cyst like objects , precluding calculation of means . Moreover , a t-test was used to compare the ELISA-positive and ELISA-negative groups .
The target of this study was the Jacob2 lectin , which was previously shown to be expressed only on the cyst cell wall of Entamoeba species . The spacer region of this protein was considered potentially useful as a diagnostic target , because its amino acid sequence diverges between Entamoeba species . Fig 1 is a protein sequence alignment of the Jacob2 lectin from E . histolytica HM-1:IMSS ( EhJacob ) and E . dispar SAW760 ( EdJacob ) . The highlighted regions were cloned for expression in this study . The full length-proteins have 64% identity and 22% gaps; however , the cloned regions ( which are mostly comprised of the spacer regions ) have 53% identity and 15% gaps . The cloned constructs expressed well in BL21 ( DE3 ) cells with codon optimization , and high purity protein was obtained with Ni-NTA affinity chromatography ( not shown ) . The SUMO tag was cleaved from EhJacob protein prior to immunization . EhJacob protein was combined in a cocktail of four immunogens , and immunization and fusion resulted in 108 hybridoma culture supernatants that were specific to EhJacob by indirect ELISA . This pool was narrowed based on activity and specificity , in a stepwise process summarized in Table 1 . Out of 108 antibodies that bound to EhJacob in the first step , 48 IgG1 and 16 IgG2 antibodies ( N = 64 ) with the highest activity in indirect ELISA ( ODs >1 . 0 ) were further screened against both EhJacob and EdJacob by using 48 x 16 matrix sandwich ELISAs . Of these , 41 bound EhJacob antigen as part of at least one pair . However , only 20 did not cross-react with EdJacob protein . Of these , 11 antibodies were examined in a third step involving further indirect ELISA . Six of the 11 clones were eliminated as SUMO-directed antibodies , based on their cross-reaction with an irrelevant , SUMO-tagged protein . This left five antibodies ( two IgG1 and three IgG2a ) that specifically bound to recombinant EhJacob protein . Out of the 5 EhJacob-specific antibodies , one IgG1 antibody ( designated 1A4 ) and one IgG2a antibody ( designated 1D3 ) were chosen for further investigation and purified . Their limit of detection as a pair in a sandwich ELISA was determined ( Fig 2 ) . When 1A4 capture antibody was eliminated from the sandwich , signals were OD450 < 0 . 2 , indicating that the Jacob antigens did not non-specifically bind to the plate . With 1A4 as the capturing antibody , the two antibodies could detect recombinant E . histolytica Jacob2 antigen in buffer at concentrations down to 9 . 8 pM ( 67 pg ) . The signal at this antigen concentration was more than three standard deviations greater than the signal with non-cognate EdJacob at 2500 pM concentration . Although the 1A4 and 1D3 antibodies exhibited excellent analytical sensitivity and selectivity in an ELISA format , it is not known whether Jacob2 can be detected as a free soluble protein in patient stool specimens . Therefore , we evaluated whether these antibodies can detect whole Entamoeba cysts in immunofluorescence microscopy assays conducted on xenic cultures of E . histolytica , E . dispar , and E . bangladeshi . Three independent isolates derived from stool were tested for each species . Xenic culture samples were smeared onto slides , fixed , blocked , and stained with 1A4 or 1D3 antibody , followed by labeling with goat anti-mouse Alexa Fluor 488 secondary . Calcofluor White M2R stain was used as a marker of encystation , as previously established [29] . Fig 3 is a panel of representative photos for each species with each antibody . By eye , IgG1 antibody 1A4 only produced fluorescent cyst-like objects when staining E . histolytica isolates . In contrast , cross-reaction was seen when Ig2a antibody 1D3 stained E . dispar isolates . Computer-assisted image analysis was used to quantify complete observations . Cyst-like objects were identified post-photo acquisition based on Calcofluor intensity , size , and circularity . Mean Alexa Fluor 488 fluorescence intensities of identified cyst-like objects were then calculated . Results are shown in Fig 4 by species and isolate . When measured by a Mann Whitney U hypothesis test ( n = m = 3 , α = 0 . 05 ) , the increased mean fluorescence of 1A4-stained E . histolytica cyst objects over 1A4-stained E . dispar and E . bangladeshi cyst objects was statistically significant ( one-tailed p-value = 0 . 025 ) ( Fig 4A ) . The higher fluorescence of 1D3 stained E . histolytica objects compared to 1D3-stained E . bangladeshi was also statistically significant ( one-tailed p = 0 . 025 ) , but the fluorescence of E . histolytica and E . dispar objects was not significantly different ( one-tailed p = 0 . 26 ) ( Fig 4B ) . Two of the three E . dispar isolates appeared to cross-react with this antibody . These quantitative results supported the results seen by eye . Although many cyst-like objects did not stain with either antibody regardless of species , strongly-staining objects were confined to E . histolytica in the case of 1A4 or to E . histolytica + E . dispar in the case of 1D3 . Antibody 1A4 was chosen for further evaluation using raw ( uncultured ) stool specimens collected from subjects in Bangladesh . Frozen specimens were thawed , filtered to remove large particulate matter , applied to microscope slides , formalin-fixed , and analyzed by immunofluorescence microscopy . Methods were similar to those in Fig 4 , except the following changes were made to accommodate the much greater complexity of raw stool samples: 1 ) DAPI was used in place of Calcofluor staining to identify cyst-like objects ( DAPI offered better specificity ) ; 2 ) stained slides were examined by a blinded reader who identified and photographed 15 fields per specimen that contained the most promising 1A4 antibody-stained cyst-like objects; and 3 ) a more stringent threshold value of 19 was used in image analysis of DAPI-stained cyst-like objects . The sample included ten specimens that had tested positive for E . histolytica by commercial Gal/GalNAc ELISA ( absorbance > 0 . 1 in ELISA assay ) , and seven specimens that had tested negative by Gal/GalNAc ELISA ( absorbance ≤0 . 1 ) . Among the ELISA-positive specimens , absorbance values ranged from 0 . 171 to 1 . 199 . Among the ELISA-negative specimens , the values ranged from 0 . 032 to 0 . 057 . For each of the 17 specimens , a visual analyst who was blinded to ELISA results generated 15 photographs containing the cyst-like objects that appeared to stain most strongly with antibody 1A4 . These photographs were then subjected to computerized image analysis that first identified all DAPI-staining cyst-like objects , and then quantified mean fluorescence index ( MFI ) of 1A4 staining ( FITC filter ) for each such object . Results are shown in Fig 5 . Seven of the ten ELISA-positive specimens exhibited at least one DAPI-stained cyst-like object . Each of these specimens included at least a subset of objects that also stained strongly with antibody 1A4 ( mean fluorescence index [MFI] > 40 ) . The remaining three ELISA-positive specimens ( specimen numbers 920 , 1084 , and 1344 ) had no detectable ( by DAPI staining and image analysis ) cyst-like objects . Among the seven ELISA-negative specimens , five specimens exhibited no detectable cyst-like objects by DAPI staining and image analysis . One ELISA-negative specimen ( #782 ) had a single cyst-like object that stained strongly with antibody 1A4 ( MFI = 78 ) . Another ELISA-negative specimen ( #763 ) had two small cyst-like objects that stained very weakly with antibody 1A4 ( MFI < 20 ) . S1 Fig shows a visual comparison between one of these weakly-staining objects in specimen #763 , and a strongly-staining object ( MFI >40 ) in an ELISA-positive specimen , #951 . Given the small number of cyst-like objects in the ELISA-negative group , statistical comparison of the two groups required that specimens with no cyst-like objects be assigned MFI values of zero . In a one-tailed t-test comparing maximum MFI in the ten ELISA-positive specimens to maximum MFI in the seven ELISA-negative specimens , staining with antibody 1A4 was significantly greater in the ELISA-positive group ( p = 0 . 038 ) . In contrast to Fig 4 , this analysis lacked specimens with confirmed cysts of non-pathogen Entamoeba species . However , the results demonstrate that antibody 1A4 can be used to stain naturally-occurring E . histolytica cysts in raw ( non-cultured ) , fixed stool specimens .
The Uniprot accession numbers for E . histolytica and E . dispar Jacob proteins aligned in this manuscript are C4LT72 and B0EEM4 . | Entamoeba histolytica is a prevalent human parasite requiring sensitive and specific detection . Assays available for Entamoeba detection utilize antibodies to detect parasite protein in stool , and they distinguish E . histolytica from nonpathogenic commensal amoeba . However , these tests have exhibited suboptimal sensitivity in some studies . This may have occurred because the tests do not detect the cyst stage of the parasite , which is more prevalent and stable in stool . Moreover , they cannot be used on formalin-fixed material , which creates logistical problems in some settings . Here , we have generated antibodies against a region of an E . histolytica cyst surface protein that has a unique sequence relative to nonpathogenic amoeba . One of these antibodies bound to E . histolytica cysts in fixed , xenic cultures without cross-reacting to nonpathogenic E . dispar and E . bangladeshi , while another antibody bound to both E . histolytica and E . dispar . The antibody that selectively bound to E . histolytica cysts also labeled cyst-like objects in fixed stool specimens from E . histolytica-positive patients . As they detected fixed organisms , these antibodies could ultimately make specimen collection and processing for Entamoeba diagnosis a simpler process . Additionally , this study indicates that the antibodies’ cyst protein target may be a useful biomarker for E . histolytica detection . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"nuclear",
"staining",
"immune",
"physiology",
"immunology",
"parasitic",
"protozoans",
"protozoans",
"antibodies",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"monoclonal",
"antibodies",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"immune",
"system",
"proteins",
"imaging",
"techniques",
"entamoeba",
"histolytica",
"proteins",
"immunoassays",
"antibody",
"isotype",
"determination",
"biochemistry",
"dapi",
"staining",
"lectins",
"physiology",
"biology",
"and",
"life",
"sciences",
"image",
"analysis",
"organisms"
] | 2016 | Species-Specific Immunodetection of an Entamoeba histolytica Cyst Wall Protein |
To better understand off-target effects of widely prescribed psychoactive drugs , we performed a comprehensive series of chemogenomic screens using the budding yeast Saccharomyces cerevisiae as a model system . Because the known human targets of these drugs do not exist in yeast , we could employ the yeast gene deletion collections and parallel fitness profiling to explore potential off-target effects in a genome-wide manner . Among 214 tested , documented psychoactive drugs , we identified 81 compounds that inhibited wild-type yeast growth and were thus selected for genome-wide fitness profiling . Many of these drugs had a propensity to affect multiple cellular functions . The sensitivity profiles of half of the analyzed drugs were enriched for core cellular processes such as secretion , protein folding , RNA processing , and chromatin structure . Interestingly , fluoxetine ( Prozac ) interfered with establishment of cell polarity , cyproheptadine ( Periactin ) targeted essential genes with chromatin-remodeling roles , while paroxetine ( Paxil ) interfered with essential RNA metabolism genes , suggesting potential secondary drug targets . We also found that the more recently developed atypical antipsychotic clozapine ( Clozaril ) had no fewer off-target effects in yeast than the typical antipsychotics haloperidol ( Haldol ) and pimozide ( Orap ) . Our results suggest that model organism pharmacogenetic studies provide a rational foundation for understanding the off-target effects of clinically important psychoactive agents and suggest a rational means both for devising compound derivatives with fewer side effects and for tailoring drug treatment to individual patient genotypes .
Neuropsychiatric disorders will effect 25% of all individuals at some point in their lives , with devastating social and economic consequences [1] . This constellation of diseases encompasses schizophrenia , depression , age-related memory and cognition decline , and the degeneration of neuromuscular function . Most prescribed psychoactive drugs are thought to primarily target neurotransmission pathways in the central nervous system , and thereby cause changes in perception , mood , consciousness , and behavior . Many of these therapeutics have been developed using in vitro assays and , as such , may have other unknown targets and unanticipated cellular effects in vivo . For example , side effects of antipsychotic drugs include tremors , hypotension , impotence , lethargy , and seizures [2] . In an effort to improve efficacy and to reduce side effects , new generations of drugs have been developed; among these are the so-called atypical antipsychotics such as clozapine . While clozapine is linked to a reduced risk of neuromuscular side effects , it is associated with new side effects such as life-threatening agranulocytosis in up to 1% of patients [3] , and , less frequently , fatal myocarditis [4]–[6] . As such , the therapeutic benefit of this and other new atypical drugs remains open to debate . For example , a comprehensive meta-regression analysis that compared both typical and atypical drugs concluded that atypical antipsychotics were neither more effective nor better tolerated than conventional agents [7] . Other classes of psychoactive drugs , such as the antidepressants , also cause numerous undesirable side effects and the broad usage of these medications have been questioned [8] . Surrogate genetics is an effective approach to interrogate heterologous gene function or drug mechanism of action using simpler model organisms [9] , [10] . The budding yeast Saccharomyces cerevisae has previously been used to help elucidate the basis of some psychiatric disorders [11]–[18] . For example , the expression in yeast of mutant and wildtype forms of the Huntington's disease gene revealed important factors regulating the toxicity of protein aggregates [11] , [15] , [19] , and a genome-wide suppressor screen in yeast uncovered kynurenine 3-monooxygenase as a potential new therapeutic target for the treatment of Huntington's disease [13] . In other studies , expression in yeast of the alpha-synclein gene associated with Parkinson's disease yielded a network of interacting genes that modulate cellular toxicity [11] , [15] , [19] . Recently , the genome-wide collection of yeast gene deletion strains has been used to generate genetic profiles of drug sensitivity and resistance [20]–[26] . These profiles have uncovered unexpected mechanisms of action for well-known drugs , such as for the anti-metabolite 5-fluorouracil in perturbation of rRNA processing [22] , [25] and for the anti-cancer agent tamoxifen in calcium homeostasis [26] . To better understand potential off-target effects of FDA-approved psychoactive drugs and their analogs , we profiled 214 psychoactive compounds in quantitative wildtype yeast growth assays and generated genome-wide deletion sensitivity profiles for the 81 drugs that caused overt growth defects . The sensitivity profiles for 49 of these drugs were overrepresented for core cellular functions such as chromatin organization , establishment of cell polarity , and membrane organization and biogenesis . Our results provide a rational foundation for personalized drug approaches and for understanding unwanted side effects in clinically important psychoactive agents .
To ask if psychoactive compounds can inhibit wildtype budding yeast growth , we challenged yeast with 76 high-purity psychoactives representing 16 ligand categories that encompass a broad spectrum of treatments for neurological disorders ( see Figure 1 for workflow and Table S1 for drug information ) . Despite the fact that yeast lacks the established neuronal targets of these compounds , 17/76 ( 22% ) drugs inhibited the growth of wildtype yeast ( when tested at 200 µM ) and are hereafter referred to as “bioactive” . This observation shows that in addition to their reported targets , many of these compounds also have secondary mechanisms of action . In fact , over half of the 16 tested ligand classes included compounds that were bioactive ( Figure 2A ) . Among these , serotonin uptake inhibitors were most effective; four of five tested molecules in this class inhibited yeast growth ( Figure 2A ) . Because our assay depends on growth inhibition in order to observe any effects on specific deletion strains , we proceeded with the 17 bioactive compounds and determined a drug dose that inhibited wildtype growth by ∼15% ( Figure 2B , Table S1 ) . In our previous genome-wide studies this level of inhibition best captured the ability to identify the known drug target while minimizing the number of generally sensitive strains [22] , [23] . Applying this drug dose , we subjected the bioactive compounds to genome-wide parallel fitness profiling . In this technique , pools of deletion strains are grown competitively for several generations in the presence of a sub-lethal concentration of drug , and genomic DNA is extracted . After PCR-amplification of the unique molecular barcodes incorporated into each gene deletion cassette , the relative role of each gene for growth in the presence of drug is determined by hybridization of the PCR products to a DNA microarray carrying the barcode complements [27]–[29] . The relative abundance of sequence tags in the drug experiments is compared to control experiments and fitness ratios and z-scores are calculated ( see Materials and Methods ) . We used two pools of diploid strains: i ) heterozygous deletion strains deleted for one copy of the essential genes ( 1158 strains ) , which often identifies compound targets through HaploInsufficiency Profiling ( HIP ) [22] , [30] , and ii ) homozygous deletion strains deleted for both copies of non-essential genes ( 4768 strains ) ; this HOmozygous Profiling ( HOP ) assay identifies genes that buffer the drug target pathway [24] . Using this combination of the HIP and HOP assays we found that only a few deletion strains ( ∼5 ) exhibited significant sensitivity to most of the 17 bioactive compounds ( Figure 2C ) . In contrast , several deletion strains ( ∼50 ) were scored as sensitive for the α1-adrenoceptor antagonist SR 59230A and the three selective serotonin re-uptake inhibitors fluoxetine ( Prozac ) , clomipramine , and fluvoxamine ( Figure 2C ) . Given this unexpected potency of the serotonergic drugs in our yeast assays , we extended our investigation to encompass pharmacologically related agents and screened two commercially available drug libraries encompassing 95 serotonergic and 55 dopaminergic compounds . These drug libraries contained the four FDA-approved serotonergics sertraline ( Zoloft ) , fluoxetine ( Prozac ) , paroxetine ( Paxil ) , and cyproheptadine ( Periactin ) , and the four FDA-approved dopaminergics bromocriptine ( Parlodel ) , clozapine ( Clozaril ) , haloperidol ( Haldol ) , and pimozide ( Orap ) . Based on our initial results , we anticipated a high rate of bioactivity on yeast for these two drug classes . Indeed , 66/150 ( 44% ) of the serotonergic and dopaminergic drugs were bioactive , a significant difference compared to the 22% of the initially screened drugs that represented the 16 different ligand sets ( p<10−7 ) . The high prevalence of bioactivity in yeast prompted us to ask if any particular psychoactive drug attribute correlated with the ability of these compounds to inhibit wildtype yeast growth . We first performed structural clustering of all ∼220 screened psychoactive compounds using chemical fingerprints in Pipeline Pilot ( Accelyrs , San Diego ) . As more than half of the resulting clusters contained both active and inactive drugs , chemical structure was not predictive of drug action on wildtype yeast growth for this selection of compounds ( data not shown ) . We next asked if any physiochemical properties , as predicted from the structures , were linked to drug activity . The parameters we tested included the number of H-bond donors and acceptors , molecular weight , and hydrophobicity as measured by AlogP ( the octanol-water partition coefficient ) . These measures are important descriptors used in the empirical parameter set known as Lipinski's Rule of Five [31] . In addition to the Lipinski descriptors , we tested six other parameters relevant to drug activity: van der Waals surface area , molecular surface area , molecular solubility , logD ( the octanol-water distribution coefficient; a combination of logP and pKa ) , number of rings and number of rotatable bonds . Principal component analysis revealed that a partition coefficient of AlogP>3 was best able to predict drug activity ( p<4 . 9e-13 , for details see Materials and Methods ) as shown in Figure 3 . A molecular weight of >260g/mole was also indicative of an active compound ( p<3 . 4e-05 , Figure 3 ) . If there is a correlation between human side-effects and conserved cellular pathways scored using our surrogate yeast system , it is possible that an additional study could help predict such effects based on structural features . To systematically interrogate compound mechanisms of action , we subjected the 66 bioactive serotonergic and dopaminergic compounds to genome-wide fitness assays using the approach described above ( Figure 1 ) . Combined with the initial set of 17 bioactive drugs , we screened a total of 81 unique drugs ( two drugs occurred in duplicate in the chemical libraries ) , eight of which are used therapeutically ( Table 1 ) . Fitness ratios and z-scores for all deletion strains are provided in Tables S2 and S3 , respectively ( raw data are available at ArrayExpress , EMBL-EBI , accession number E-MTAB-34 ) . The genome-wide fitness profiles were reproducible as the average correlation coefficient for the five replicated compounds was 0 . 83 , which is similar to the average correlation coefficient of 0 . 72 reported in a previous large-scale fitness study [23] . As an unbiased control , we calculated the average correlation coefficient between all possible random drug pairs in our assay . As expected , this value ( 0 . 44 ) was lower than the average correlation coefficient for duplicates , but well above the previously noted average correlation of zero for unrelated compounds ( Maureen Hillenmeyer , unpublished data ) . In agreement with this , two-dimensional hierarchical clustering [32] did not separate the dopaminergic and serotonergic profiles into two distinct groups , but clearly separated drugs from these two classes from most other compounds profiled ( Figure 4 ) . Further indicating the general similarities between dopaminergic and serotonergic drugs in our yeast screen , 25% of the significantly sensitive strains ( r>2 , z>3 , see Materials and Methods ) scored in both drug categories ( Table S4 ) . To ask which cellular functions and pathways were required for resistance to the tested drugs , we performed functional enrichment tests using Gene Ontology ( GO ) annotations specifically focusing on sensitive strains in the i ) essential heterozygous , ii ) homozygous or iii ) both collections ( see Materials and Methods ) . 32 drug sensitivity profiles were not enriched for any GO Process but the remaining 49 profiled drugs ( 60 . 5% ) interfered with 106 different processes ( multiple-testing corrected p-value<0 . 0001 , Table S5 ) . For visual clarity , we collapsed these 106 processes down to 22 ( Table S6 ) . The drug sensitivity profiles obtained with the combined set of heterozygous and homozygous strains were enriched for the highest number of condensed GO processes ( 119 processes , purple color in Figure 5 ) , while 12 processes were uniquely enriched among sensitive homozygous deletion strains ( blue color in Figure 5 ) . These processes likely reflect drug detoxification mechanisms ( e . g . “vesicle transport” and “response to drug” ) or other processes required for resistance to compound by an unknown mechanism ( e . g . “amino acid biosynthesis and metabolism” ) . Two processes were uniquely scored for essential genes ( red color in Figure 5 ) and are further discussed below . Investigating the general nature of our enrichment profiles , we found that the most frequently enriched processes across all drugs and genetic backgrounds were vesicle transport , protein localization , and telomere biology ( Figure 5 ) . Genes functioning in cell morphogenesis , establishment of cell polarity , cell cycle , amino acid biosynthesis , chromatin organization , RNA metabolism , and membrane organization were also needed for resistance to several ( >5 ) of the psychoactive drugs . A few GO Processes were unique to a single drug: protein glycosylation ( A77636 ) , methylation ( SB 216641 ) , cell wall organization and biogenesis ( GR 127935 ) , and membrane lipid metabolic process ( pimozide ) . In the subsequent sections we focus on the analysis of the FDA-approved drugs and summarize the most notable enrichments for these drugs in Table 2 . First , we discuss identified buffering pathways and drug detoxification mechanisms . Next , we concentrate on potential new drug targets identified for the therapeutically used psychoactive drugs . Vesicle transport was the most commonly overrepresented process among genes required for resistance to psychoactive drugs ( Figure 5 ) suggesting that uncompromised vesicle transport function is a general requirement for psychoactive drug detoxification . The enrichment of cellular transport genes was especially pronounced in response to clozapine treatment , where 9 of the 10 most required genes belonged to this category ( Table 3 ) . Protein sorting and localization accounted for the second most frequently enriched process ( Figure 5 ) . Deletion of vesicle trafficking and protein localization genes often resulted in very severe phenotypes ( bright yellow in Figure 4 ) . Gene products with protein localization roles include those involved in selecting cargo proteins for endosome-to-Golgi retrieval ( e . g . Vps29 ) , and those involved in sorting proteins in the vacuole ( e . g . Pep8 ) . Interestingly , the fitness profiles obtained with certain vesicle transport and protein localization deletions clustered with those obtained with strains deleted for genes functioning in actin filament organization/stabilization ( arc18Δ , tpm1Δ , vrp1Δ , ) , mRNA degradation ( lsm1Δ ) , and stabilization of membrane amino acid transporters ( npr1Δ ) ( Figure 4 , left text panel ) . A second , large group of strains mainly deleted for genes functioning in vesicle transport and protein localization exhibited similar phenotypes across the 81 drugs as ckb1Δ and ckb2Δ , which are deleted for genes functioning in regulation of transcription and mitotic cell cycle ( Figure 4 , right panel ) . Most of the drug sensitivity profiles were enriched for both protein localization and telomere biology ( Figure 5 ) . The apparent “linking” of these enrichments could be attributed to genes that are , in fact , involved in both these processes . Examples of such genes function in the three Endosomal Sorting Complexes Required for Transport , more specifically in ESCRT I ( VPS28 , STP22 ) , in ESCRT II ( SNF8 and VPS25 ) , and in ESCRT III ( SNF7 ) . These genes are , in addition , associated with telomere defects [33] , [34] . Because the more recently developed atypical antipsychotic drugs are still associated with side effects and their benefits are currently debated , we compared the phenotypic profiles of the atypical antipsychotic clozapine to two traditional antipsychotics , reasoning that if atypical drugs are more specific , they would exhibit fewer off-target effects in yeast . In contrast to this expectation , the atypical antipsychotic clozapine exhibited a similar number of significantly sensitive ( r>2 , z>3 , see Materials and Methods ) deletion strains ( 26 ) as the typical antipsychotic drugs pimozide ( 29 ) and haloperidol ( 20 ) . Comparing the fitness profiles of clozapine with the typical antipsychotics pimozide and haloperidol , we found that each drug was associated with unique functional enrichment profiles: clozapine for telomere biology and protein localization , pimozide for membrane lipid metabolic processes , and haloperidol for aromatic amino acid biosynthesis and metabolism ( Figure 5 ) . In contrast , vesicle transport was enriched in all three drug sensitivity profiles . The more detailed GO processes behind the condensed process vesicle transport were vesicle-mediated transport for all three drugs and , in addition , secretory pathway , secretion , post-Golgi vesicle-mediated transport and Golgi vesicle transport for haloperidol and clozapine ( Tables S5 and S6 ) . The distinct fitness profiles are consistent with the structural differences that exist between these drugs ( Figure S1 ) . For example , clozapine has substructures ( piperazine and diazepine ) that do not exist in pimozide and haloperidol , and haloperidol contains two benzene rings while pimozide has three . Compared to the other investigated therapeutics , the fitness profile in the anti-Parkinson drug bromocriptine pointed to a single potential off-target mechanism of action for this drug . The only overrepresented function among sensitive strains was amino acid biosynthesis and metabolism ( Figure 5 ) and the most sensitive strains were deleted for the aromatic biosynthesis genes TRP3 , TRP4 , TRP1 , ARO1 , TRP2 , and ARO2 . In addition to bromocriptine , six other dopaminergic drugs also interfered with amino acid biosynthesis and metabolism ( Figure 5 ) . The sensitivity profiles of all these seven drugs shared the enrichment for the detailed GO process aromatic compound metabolic process ( Tables S5 and S6 ) due to the sensitive phenotype of 13 strains in total . Among them , strains deleted for TRP1 , TRP2 , TRP3 , TRP4 , TRP5 , ARO2 , and ARO3 were scored in all 7 drugs and strains deleted for ARO1 and ARO7 in 6 drugs . Besides the notable enrichment for genes involved in aromatic compound metabolism , the sensitivity of strains missing other genes also contributed to the observed GO process enrichment . Such genes included the folic acid ( vitamin B9 ) biosynthesis gene FOL2 , the panthothenate ( vitamin B5 , precursor of coenzyme A ) biosynthesis gene FMS1 , and the protein kinase GCN2 , which induces amino acid biosynthesis genes in yeast in response to starvation and , in addition , restricts intake of diet lacking essential amino acids in rats [35] . The sensitivity profile of the typical antipsychotic pimozide showed a unique enrichment for membrane lipid metabolic processes not seen for any of the other 80 profiled drugs ( Figure 5 ) . In pimozide , the MCD4-deletion strain had the strongest phenotype and was 21-fold depleted compared to the control ( Table 3 ) . MCD4 is highly conserved among eukaryotes and functions in glycosyl-phosphatidylinositol ( GPI ) anchor synthesis . Because MCD4 is an essential gene , it may represent an additional , clinically relevant drug target for pimozide . The inositol-lipid-mediated signaling gene PIK1 and the spingholipid-mediated signaling gene YPK1 were also among the ten most required genes for resistance to pimozide ( Table 3 ) . They clustered with a group of other strains deleted for genes involved in lipid biology ( Figure 4 ) , such as the de novo lipid synthesis genes PAH1 and SUR4 . Eight drugs , among them the antidepressant fluoxetine , were enriched for the condensed term establishment of cell polarity ( purple or blue color in Figure 5 ) . In total , 51 genes were assigned to the detailed GO process establishment and/or maintenance of cell polarity and caused a sensitive phenotype when deleted ( Tables S5 and S6 ) . Many of these genes scored in the majority of the drugs , for example all four members ( CKA1 , CKA2 , CKB1 , and CKB2 ) of the casein kinase II-holoenzyme complex , and TPM1 , the major isoform of tropomyosin which directs polarized cell growth and organelle distribution . For the seven drugs where the enrichment for establishment and/or maintenance of cell polarity was scored using sensitive homozygous and essential heterozygous strains ( purple color in Figure 5 ) , six essential members ( EXO70 , SEC3 , SEC6 , SEC8 , SEC10 and SEC15 ) of the exocyst complex , which determines where secretory vesicles dock and fuse , were scored in all drugs except fluoxetine . Drug targets are often encoded by essential genes , thus essential genes scored in our assay may represent important additional targets of psychoactive compounds that may be useful in the development of therapeutics for other applications . In a given heterozygous strain , the reduced gene copy number of a potential drug target leads to a reduced level of the corresponding protein . When this strain is grown in the presence of a drug targeting the heterozygous locus , the result is a further decrease in “functional” dosage due to the drug binding to the protein target . If this protein is important for growth , the result will be drug sensitivity [22] . In our functional enrichment tests , two processes were uniquely overrepresented among sensitive essential genes ( red color in Figure 5 ) : mitotic and meiotic cell cycle for fluorophenyl-methoxytropane and chromatin organization for cyproheptadine . Examples of targeted essential genes in cyproheptadine treatment include chromatin-remodeling genes ( ARP4 , ARP7 , ARP9 ) , genes in the multisubunit ( NuA4 ) histone acetyltransferase complex ( EPL1 , ESA1 , SWC4 ) , and RSC4 and RSC6 in the RSC Chromatin remodeling complex . Although not revealed as a functional enrichment among sensitive strains deleted for essential genes , most of the other FDA-approved drugs also have potential secondary drug targets as infered by the presence of essential genes among the ten most required genes for drug resistance ( Table 3 ) . As judged by the high number of sensitive strains deleted for essential genes in paroxetine treatment ( 10 strains ) and sertraline treatment ( 9 strains ) , these selective serotonin re-uptake inhibitors are particularly rich in potential secondary drug targets . Essential genes required for resistance to the FDA-approved drugs include those involved in RNA processing , transcription and translation , genes functioning in the protein folding chaperonin complex , and the chromatin-remodeling/DNA repair gene ARP4 ( bold in Table 3 ) . Deletion of ARP4 resulted in some of the most sensitive phenotypes when cells were treated with cyproheptadine , sertraline , or with haloperidol ( Table 3 ) . ARP4 has a close human homolog , ACTL6B , which encodes a subunit of the BAF ( BRG1/brm-associated factor ) complex in mammals , functionally related to the SWI/SNF complex in S . cerevisiae . The SWI/SNF complex is thought to facilitate transcriptional activation by antagonizing chromatin-mediated transcriptional repression [36] . Another example of an essential gene required for drug resistance in several FDA-approved drugs is GSP1 , which functions in RNA-processing ( Table 3 ) . The mammalian homolog of Gsp1 , Ran ( BlastP E-value<E-261 ) is , as in yeast , a nuclear GTP-binding protein . Interestingly , the fitness profile of the ARP4-deleted strain was very similar to the strains deleted for the cytosolic chaperonin subunits CCT5 , CCT8 and TCP1 ( Figure 4 ) . The chaperonin complex is involved in protein folding ( primarily of actin and tubulin ) and cytoskeleton organization [37] . In our fitness assays , seven of eight CCT-strains scored as significantly sensitive in many of the probed psychoactive drugs . Some ( CCT3 , CCT4 , CCT7 and CCT8 ) were even among the top-ten required genes for resistance to cyproheptadine , fluoxetine , paroxetine , and sertraline ( Table 3 ) . Furthermore , several deletion strains with uncharacterized functions had similar fitness profiles as the chaperonins CCT5 , CCT8 and TCP1 ( Figure 4 ) . Among them were TVP23 and YIP5 which both localize to the late Golgi , YEL048C which is synthetic lethal with GCS1 ( involved in ER to Golgi transport ) , APM2 ( homologous to medium chain of mammalian clathrin-associated protein complex involved in vesicle transport ) and SWH1 ( similar to mammalian oxysterol-binding protein , localized to Golgi and nucleus-vacuole junction ) . To test if our findings in yeast might reflect drug action in human cells , we looked at the proportion of scored genes with human homologs . Among the strains significantly sensitive to at least one psychoactive compound , 58 . 4% were deleted for a gene with a close human homolog ( BlastP E-value<E-6 ) , as compared to 45 . 0% for all analyzed deletion mutants regardless of whether they had a fitness defect or not . To test if strains deleted for genes involved in core cellular processes are more sensitive in general , we compared our results obtained with the 81 psychoactive compounds to 81 randomly chosen chemically diverse compounds ( see Materials and Methods ) . We found that a similar proportion of genes with close human homologs ( 59 . 7% ) were scored for strains that were significantly sensitive to at least one of these diverse chemicals . Despite this similarity in proportion of sensitive strains with human homologs in the two datasets , conserved genes were scored much more frequently ( in >10% of the compounds ) in the psychoactive drug set than in the random drug set . In fact , considering only genes deleted in frequently scored strains , 64 . 1% of the psychoactive drugs had close human homologs ( BlastP E-value<E-6 ) while the corresponding proportion for the structurally diverse drug set was significantly ( p<0 . 006 ) lower ( 45 . 4% ) and similar as the fraction of human homologs for multi-drug resistance genes ( 47 . 1% ) in a recently published study [23] . This difference points to a significant enrichment of frequently scored sensitive strains with human homologs for the psychoactive drugs . Among the strains sensitive to the highest number of psychoactive compounds , seven of eight had close human homologs: NEO1 , SAC1 , PIK1 , VPS29 , PEP8 , ARP4 and VPS35 . The majority of these genes are involved in vesicle transport , which was the most frequently enriched function among strains sensitive to psychoactive drugs . Thus , the specific psychoactive drug detoxification mechanisms identified in yeast are likely to be of importance in humans treated with psychoactives .
Many psychoactive drugs are associated with adverse secondary effects in humans yet the mechanisms that underlie these off-target effects are poorly understood . To address mechanisms of drug action in a systematic manner , we profiled the genome-wide collection of budding yeast deletion strains for sensitivity to a broad spectrum of psychoactive compounds , of which dopaminergic and serotonergic drugs had a high bioactivity . Among 214 tested compounds , we uncovered 81 drugs that conferred a measurable growth defect on wildtype yeast . An appropriate dose of these active compounds was applied to the pooled heterozygous and homozygous yeast deletion sets to identify genes whose function is required for optimal growth in the presence of drug . Fifteen percent of all yeast strains ( deleted for non-dubious ORFs ) exhibited significant sensitivity ( r>2 , z>3 ) to these 81 psychoactive compounds and more than half of the drugs interacted with core cellular functions . Several clinically important drugs , such as fluoxetine , cyproheptadine , and clozapine were linked to diverse cellular processes . This observation may explain both the diversity of side effects observed in human patients and the therapeutic variability associated with these drugs . That is , polymorphisms in any of the conserved processes affected by a given drug are a likely source of the individual variation in response to drug . For instance , the response to the frequently prescribed antipsychotic clozapine is highly variable between individuals as the same dose can have markedly different efficacy and/or side effects in different patients [38] . Genes functioning in vesicle transport , protein localization , telomere biology , and catabolic processes were required for clozapine resistance in yeast . In another example , fluoxetine is associated with side effects such as seizures , nausea , sleepiness , anxiety , and serious allergic reactions . This antidepressant affects numerous cellular processes including establishment of cell polarity , protein localization , and cytoskeleton organization and biogenesis . Given the limited number of FDA-approved drugs within the set of 81 compounds analyzed here and the overlapping side effects associated with these drugs , it is not yet possible to correlate any single side effect to a particular perturbed pathway . The most frequently scored sensitivity for the 81 profiled antipsychotic drugs was due to loss of secretory pathway function , likely indicating the importance of vesicle transport ( e . g . to the vacuole ) for drug detoxification . The lysosome ( the mammalian vacuole equivalent ) is known as the major site of degradation of both exogenous and endogenous molecules . For FDA-approved drugs , the requirement for vesicle transport genes was reflected in the frequent sensitivity of the neo1 deletion strain as the most sensitive strain in six FDA-approved drugs . Neo1 is an essential , highly conserved type 4 P-type ATPase involved in intracellular membrane- and protein-trafficking . Members of this family of P-type ATPases are implicated in the translocation of phospholipids from the outer to the inner leaflet of membrane bilayers . Our data suggested that interference with membrane structure and transport through inhibition of Neo1 is an additional , unwanted mechanism of action for clozapine , cyproheptadine , fluoxetine , paroxetine , sertraline and haloperidol , and their drug analogs . The importance in humans of functional 4 P-type ATPases is well documented as hereditary cholestasis , caused by defects in biliary epithelial transporters , has been directly linked to mutations in a 4 P-type ATPase gene [39] . In addition to the frequently observed requirement for uncompromised vesicle transport for drug detoxification , several drug sensitivity profiles were enriched for more specific processes . Within the FDA-approved drug group , the antidepressant paroxetine was unique in targeting RNA processing genes , pimozide interfered with membrane lipid metabolic processes , cyproheptadine preferentially targeted essential genes with chromatin remodelling functions , and fluoxetine interfered with establishment of cell polarity . Furthermore , seven dopaminergic compounds including the anti-Parkinson drug bromocriptine resulted in sensitivity of strains deleted in aromatic amino acid biosynthetic genes . This sensitivity may be a result of that dopaminergic drugs block aromatic amino acid uptake in yeast , requiring yeast to activate the corresponding biosynthetic pathways . Given the fact that aromatic amino acids are precursors to dopamine and serotonin , this was an interesting observation suggesting that the levels of intracellular precursors may be important in the response to certain psychoactive drugs . Interestingly , interference with members of the chaperonin complex resulted in some of the most severe phenotypes . Seven of eight CCT-strains scored as significantly sensitive in several psychoactive drugs , among them CCT5 . The human homolog of this gene is associated with hereditary neuropathy [40] . Although it is unclear how mutated CCT5 causes this disease , it has been postulated that its mutation leads to accumulation of misfolded cytoskeletal proteins , leading to defective assembly of actin into microfilaments resulting in neuronal apoptosis [41] . In our yeast screens , CCT5 was needed for resistance to eight different compounds ( cyproheptadine , paroxetine , fluoxetine , indatraline , MDL72222 , CY208-243 , 2-Chloro-11- ( 4-methylpiperazino ) -dibenz[b , f]oxepin , N-Desmethyl-clozapine , and 3-alpha-[ ( 4-Chlorophenyl ) -phenylmethoxy]-tropane . We conclude that interference with tubulin and actin folding is an important , secondary mechanism of action of these compounds . As an example of how the information from our yeast assays may lead to testable drug-gene interaction hypotheses in humans , we found that the levels of the yeast strain heterozygous for ACC1 was eleven-fold reduced in ritanserin as compared to the control , indicating that the acetyl-CoA carboxylase Acc1 may be a secondary target of ritanserin . Like its yeast counterpart , the human homolog ACACA is required for de novo biosynthesis of long-chain fatty acids and its activity drops during fasting [42] . Because increased appetite is a reported side-effect during ritanserin treatment [43] , it is tempting to speculate that biochemically mimicking fasting would increase appetite . These studies raise several important issues for further consideration . Understanding the mechanisms that underlie adverse effects of clinically approved drugs is crucial for the development of next generation therapeutics with improved selectivity and efficacy . Moreover , knowledge of patient polymorphisms in off-target pathways may allow adverse effects of any given drug to be preempted by personalized pharmacogenomic strategies . It is also conceivable that some of the observed secondary drug effects are critical for therapeutic benefit . In summary , a number of cellular processes were associated with sensitivity to the dopaminergic and serotonergic classes of psychoactive compounds . This points to additional , previously uncharacterized mechanisms of action for these drugs in humans and suggests follow-up experiments aimed at understanding a drug's mechanism of action on a genome-wide level . Our results suggest that model organism pharmacogenetics can be used as a comprehensive and unbiased tool in initial studies aiming at unraveling secondary effects and mechanisms of action for therapeutic compounds and their analogs . A more rigorous understanding of the complete mechanism of drug action in humans would be beneficial in the development of a new generation of better tolerated psychoactive drugs , and in personalized medicine .
High purity compounds for genome-wide fitness profiles were obtained from Tocris BioScience ( http://www . tocris . com ) as ligand sets and as the serotonergic ( #1732 ) and dopaminergic ( #1718 ) collections . In total , these drug collections comprised 226 drugs , 12 of which overlapped between the collections . A complete list of drugs , catalogue numbers , solvents , and concentration used in the genome-wide screens is provided in Table S1 . For genome-wide fitness profiles , the complete sets of ∼4700 homozygous deletion strains and ∼1100 essential heterozygous deletion strains in the BY4743 and BY4744 backgrounds ( MATa/α his3Δ1/his3Δ1 leu2Δ0/leu2Δ0 lys2Δ0/LYS2 MET15/met15Δ0 ura3Δ0 /ura3Δ0 ORF::kanMX4 ) were used [29] , [44] . A strain in the same genetic background with YDL227C replaced by kanMX4 was used as the wildtype control for drug titration curves . Strains were stored in 7% DMSO at -80°C . Because all experiments were performed in rich media ( YPD [45] , without antibiotics ) , it is unlikely that the presence of auxotrophies had a major effect on our results , however , we cannot rule out that the disruption of the corresponding pathways in yeast may , in some cases , alter our findings . Beginning from an initial maximal concentration of 200 µM , the degree of growth inhibition was determined by exposing wildtype cells to a serial dilution of compound until only a slight inhibition ( ∼15% ) of wildtype growth was observed ( see Figure S1 ) . Cells were inoculated at an OD600 of 0 . 0625 in serial dilutions of drug and grown in a Tecan GENios microplate reader ( Tecan Systems Inc . , San Jose , USA ) at 30°C with orbital shaking . Optical density measurements ( OD600 ) were taken every 15 minutes until the cultures were saturated , and doubling time ( D ) was calculated as described [46] . Fitness assays using pooled deletion strains were performed as described [47] with the following modifications: i ) after growth , 350 µl from each of two independent cultures of the 5-generation homozygous pool and 350 µl from the 20-generation heterozygous essential pool were combined , thereby allowing for approximately equal representation of barcodes for PCR reactions and hybridization to the same DNA chip using the unique barcodes incorporated in each of these strains . ii ) for amplification of the tags , ∼0 . 2 µg genomic DNA was combined with a 1 µM mix of either up- or down-tags and 82% ( v/v ) Platinum High Fidelity PCR Supermix ( Invitrogen # 11306-016 ) containing anti-Taq DNA polymerase antibody , Pyrococcus species GB-D thermostable polymerase , recombinant Taq DNA polymerase , Mg2+ , and dNTPs , iii ) extension temperature was 68° , iv ) extension was for 2 min except for a final 10 min extension v ) 34 cycles of amplification were performed , vi ) after 10-16 h , the hybridization mix was removed from Affymetrix Gene Chips , replaced with Wash A ( 6x SSPE , 0 . 01% Tween ) , and chips were stained and washed using GeneChip Fluidics Station 450 ( Affymetrix ) according to the GeneFlex_Sv3_450 protocol with one additional wash A cycle before the staining . Intensity values for the probes on the chip were extracted using the GeneChip Operating Software ( Affymetrix ) . Quantile normalization , outlier omission , fitness defect ratio ( denoted as “r” ) and z-score ( denoted as “z” ) calculations were performed as previously described [47] , [48] . In short , fitness defect ratios were calculated for each deletion strain as the log2 of the ratio between the mean signal intensities of the control and the drug chips . The larger the ratio , the more depleted ( sensitive ) is the strain as compared to control condition without the drug . To include the variance in the control experiments , we also calculated z-scores for each gene by dividing the difference in mean intensity across the control chips and treatment with the mean standard deviation of the signal intensities for the given gene across all 18 control chips [48] . The larger the z-score , the more likely it is that a given strain is significantly depleted from the pool . In our analysis , we scored a deletion strain as significantly sensitive using a threshold for both z-score and log2 intensity ratio . A threshold of z>3 was selected based on our earlier observations that above this limit , 100% of 186 deletion strains detected as sensitive by microarray could be confirmed using individual growth assays [24] . This stringent threshold was chosen to minimize the number of false positives . In addition , we added a further requirement that a sensitive strain should display at least a fourfold depletion ( r>2 , i . e . log2>2 ) compared to the control condition . This criterion was added to avoid including z-scores which were artificially high due to a low standard deviation in the control chips . Due to the way the screens were performed ( at low drug concentration , i . e . an IC15 ) and analyzed [22] , [24] we have focused on sensitive strains in this work , as opposed to apparently resistant strains . Two-dimensional hierarchical clustering of the fitness ratios was performed using Pearson correlation [32] and the data was visualised using the MultiExperiment Viewer from the TM4 microarray software suite ( http://www . tm4 . org/index . html ) . In each of the 81 profiled drugs , sensitive deletion strains were tested for Gene Ontology Functional enrichment using the standard hypergeometric test provided by the GoStats Bioconductor modules for R [49] . For each drug , we performed three independent functional enrichment tests using i ) sensitive heterozygous strains deleted for essential genes ( z>2 ) , ii ) sensitive homozygous strains ( z>2 ) , and iii ) all sensitive strains in the given drug with z>2 . As the global control set , we used all yeast ORFs in the corresponding deletion background with chip intensity values above background . The background was determined as the average value of all unused tags on the chip ( ∼3600 tags×5 copies = 18000 values ) +2 standard deviations of the background tags . Obtained p-values were corrected for multiple testing by multiplying by the number of identified terms . Adjusted p-values<0 . 0001 were considered significant . GO processes linked to less than 20 or more than 300 genes in our background set were excluded from our tests . Two-dimensional hierarchical clustering of overrepresented GO processes was performed using binary data [50] . To test the robustness of our functional enrichment tests , we repeated the same analysis using each of the following thresholds: z>3 , r>2 , r>3 and found consistent functional enrichment profiles . In the calculations of the proportion sensitive strains deleted for genes with close human homologs ( Blastp E-value<E-6 ) , we used a set of 81 recently profiled ( our unpublished data ) compounds with potency against wildtype yeast . These compounds represent structurally diverse chemicals derived from Chemical Diversity Labs , Inc . repository of >500 , 000 compounds . Structure data files were obtained from Tocris and Pubchem for all compounds and Babel canonical smile strings were generated . In the chemical structure clustering , extended connectivity fingerprints based on functional classes in Pipeline Pilot were used [51] . In the physiochemical property clustering , ten descriptors representing important properties for potential drug candidates were calculated after salts were stripped , using Frowns and Openeye cheminformatic libraries [31] , [52] . PCA was used to find the strongest properties that separated active from non-active compounds . The revealed properties ALogP and molecular weight were validated to see how they correlated with the pattern of the other eight descriptor loadings . The non parametric Wilcoxon rank sum test supported LogP ( p-value 4 . 91e-13 ) and MW ( p-value 3 . 42e-05 ) as significant representative properties . All supplementary data can also be downloaded from our webpage , http://chemogenomics . med . utoronto . ca/Supplemental/psychoactives/ . | Neuropsychiatric disorders such as depression and psychosis affect one-quarter of all individuals during their lifetime , and despite efforts to improve the selectivity of psychoactive drugs , all are associated with side effects . Drug efficacy and tolerance are known to be linked to an individual's genetic profile , but little is known about the nature of this correlation due , in part , to the current emphasis on screening compounds against targets in vitro . Here we present a comprehensive , genome-wide effort to understand drug effects on the cellular level using an unbiased genome-wide assay to determine the importance of every yeast gene for tolerance to 81 psychoactive drugs . We found that these medications perturbed many evolutionarily conserved genes and cellular pathways , such as those required for vesicle transport , establishment of cell polarity , and chromosome biology . The 500 , 000 drug–gene measurements obtained in this study increase our understanding of the mechanism of action of psychoactive drugs . Specifically , this study provides a framework to assess the next generation of psychoactive agents and to guide personalized medicine approaches that associate genotype and phenotype . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"genetics",
"and",
"genomics/gene",
"function",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/functional",
"genomics"
] | 2008 | Off-Target Effects of Psychoactive Drugs Revealed by Genome-Wide Assays in Yeast |
Influenza virus particles are assembled at the plasma membrane in concert with incorporation of the virus genome , but the details of its spatio-temporal regulation are not understood . Here we showed that influenza virus infection induces the assembly of pericentrosomal endocytic recycling compartment ( ERC ) through the activation of Rab11a GTPase and cell cycle-independent maturation of centrosome by YB-1 , a multifunctional protein that is involved in mitotic division , RNA metabolism and tumorigenesis . YB-1 is recruited to the centrosome in infected cells and is required for anchoring microtubules to the centrosome . We also found that viral infection accumulates cholesterol in ERC and is dependent on YB-1 . Depletion of YB-1 shows reduced cholesterol-enriched ERC and prevented budozone formation at the plasma membrane . These results suggest that cholesterol in recycling endosomes , which are emanated from ERC , may trigger the virus assembly concomitantly with the packaging of the virus genome . We propose that the virus genome is transported to the plasma membrane by cholesterol-enriched recycling endosomes through cell cycle-independent activation of the centrosome by YB-1 .
Endocytic transport pathways are important to arrange the plasma membrane components for diversified cellular processes at the plasma membrane including virus budding . Endocytosed proteins are first delivered to the early/sorting endosomes , from where proteins are either recycled back to the plasma membrane or transported to late endosomes and lysosomes . Rab small GTPase family members show distinct intracellular localization and function as molecular switches to regulate vesicle carrier formation and fusion with target membranes . Rab11a-positive recycling endosomes are crucial for recycling and delivery of plasma membrane components to the cell surface [1–3] . The Rab11a-positive transport vesicles emerge from specific organelles called endocytic recycling compartments ( ERC ) . ERCs constitute a collection of tubular organelles that are close to the nucleus and associated with the microtubule organizing centre ( MTOC ) . However , the functional significance of ERCs is not fully understood . MTOC is a highly dynamic structure that achieves precise control of the microtubule array for the spatial and temporal regulation of several fundamental processes . Microtubule dynamics is controlled through continuous switching between phases of growth and shrinkage , as well as the level and timing of nucleation from the centrosome , which is the major MTOC in animal cells . The centrosome is composed of a pair of centrioles surrounded by pericentriolar material ( PCM ) , a matrix of more than a hundred different proteins . PCM proteins are organized radially around the centriole in a toroid-like arrangement [4–7] and PCM serves as a platform for microtubule nucleation . During mitosis , in a process known as centrosome maturation , PCM increases in size to promote the microtubule nucleation for mitotic spindle formation [8 , 9] . The influenza viral genome forms a viral ribonucleoprotein complex ( vRNP ) with viral RNA polymerases and nucleoprotein ( NP ) . After viral genome replication in the nucleus , the progeny vRNP is nuclear-exported and then accumulates around the centrosome [10] . vRNP is then transported to the budding site beneath the cell surface along microtubules through Rab11a-dependent recycling endosomes [11–13] . Recently , Y-box binding protein-1 ( YB-1 ) was reported to function as a porter to facilitate vRNP accumulation at the centrosome [14] . YB-1 is a major component of cellular mRNA ribonucleoprotein complexes and it regulates mRNA translation and degradation [15] . It is also reported that YB-1 accumulates in the centrosome during G2/M phases [16] and is required for the centrosome maturation [17] . Cholesterol is a major constituent of the plasma membrane in eukaryotic cells . It regulates the physical state of the plasma membrane and is involved in the formation of membrane microdomains , called lipid rafts . Lipid rafts are defined as small ( 10–200 nm ) , heterogeneous , highly dynamic , sterol- and sphingolipid-enriched domains that compartmentalize cellular processes [18] . Small rafts can sometimes coalesce to form larger platforms through protein-protein , protein-lipid , and lipid-lipid interactions . Three viral membrane proteins , HA , NA , and M2 , are embedded in the influenza virus envelope . M1 covers the inner viral membrane leaflet and binds to the cytoplasmic tails of HA and NA [19] . The assembly and budding of viral particles are coupled with the formation of functionalized raft domains , called budozone [20] . In the budozone , HA , possibly together with NA , is enriched by clustering several small rafts [21 , 22] . M2 possesses cholesterol-binding motifs [23 , 24] , but a relatively short transmembrane domain of M2 prevents complete immersion of the protein in the more ordered raft domains . Thus , M2 is thought to localize to the edge of the budozone to mediate the pinching off of virus particles from the plasma membrane [25] . Finally , vRNP is recruited to the budozone through the interaction of vRNP with M1 to initiate budding and release of virus particles . Here we showed that influenza virus infection induces the assembly of pericentrosomal ERCs through the activation of Rab11a and microtubule dynamics . Using three-dimensional structured illumination microscopy ( 3D-SIM ) , we found that YB-1 forms a toroid-like structure with a beads-on-a-string distribution pattern around the centriole . Knockdown ( KD ) analyses indicated that influenza virus stimulates the spontaneous centrosome maturation in interphase by recruiting YB-1 to anchor newly synthesized microtubules onto the centrosome . We also found that cholesterol accumulates in the pericentrosomal ERC with vRNP in an YB-1-dependent manner . Disruption of the cholesterol-enriched ERC formation by YB-1 KD results in defective viral budozone formation at the plasma membrane . Collectively , these results suggest that the recycling endosomes containing cholesterol and vRNP emanate from ERC , and cholesterol in recycling endosomes is a trigger for the viral budozone formation concomitantly with vRNP trafficking to the plasma membrane .
Transferrin is a typical marker to monitor the organization of active recycling endosomes during endocytosis and its return to the cell surface . To examine the dynamics of the recycling pathway in influenza virus-infected cells , cells were pulse-labeled for 30 min with transferrin Alexa fluor 568 , followed by a chase for 30 min without fluorescent transferrin . At 3 h post infection , transferrin-positive recycling endosomes were accumulated in ERC at a juxta-nuclear region , possibly near the centrosome ( Fig 1A and 1B , white arrowheads ) . Transferrin recycling proceeds with a t1/2 of approximately 20 min [26] , therefore the transferrin uptake should correspond to a steady-state distribution of the labeled ligand ( Fig 1A ) . We next performed an indirect immunofluorescence assay using anti-Rab11a antibody and FISH assay using a probe that hybridizes with the segment 1 virus genome ( Fig 1C , arrowheads ) . As is the case for transferrin , Rab11a was also present in the juxta-nuclear region and colocalized with the virus genome in approximately 40% of infected cells at 6 h post infection ( P<0 . 001 ) , suggesting that the virus genome is recruited to the pericentrosomal ERC after nuclear-export , as previously reported [10–14] . It has been shown that active Rab11a shows a marked accumulation of ERC at the centrosome [27] . To evaluate whether Rab11a is activated by influenza virus infection , we purified active Rab11a ( Rab11-GTP ) by GST pull-down assays using Rab11-binding domain of Rab11-FIP2 . Rab11-FIP2 acts as an effector molecule for Rab11-GTP through a highly conserved Rab11-binding domain ( RBD ) among Rab11-FIP family proteins [28] . Therefore , we can purify Rab11-GTP ( constitutive active mutant Q70L , lane 8 ) , but not the GDP form ( dominant negative mutant S25N , lane 9 ) , using GST-fused 41 amino acid peptide derived from RBD of Rab11-FIP2 ( GST-RBD ) ( Fig 2A ) . Next , we performed GST pull-down assays with lysates prepared from infected cells using GST-RBD at 8 h post infection ( at which the virus genome is actively transported ) and the co-purified Rab11a was analyzed by western blotting with anti-Rab11a antibody ( Fig 2B ) . The amount of Rab11a co-purified with GST-RBD from infected lysates was 4 . 5 ± 0 . 6 times more than that from mock-treated lysates ( Fig 2B; representative results from three independent experiments are shown ) , suggesting that a guanine nucleotide exchange factor ( GEF ) for Rab11a may be activated in response to infection . By interacting with a number of Rab11-FIPs , Rab11a associates with distinct motor proteins , enabling bidirectional transport along microtubules . Thus , recycling endosomes closely associate with microtubules , and their intracellular transport is fully dependent on the microtubule dynamics , which undergo cycles of nucleation , growing , and shrinking . The precise spatial and temporal regulation of the cycles is essential for the numerous cellular functions in which microtubules are involved . Previously , we reported that YB-1 accumulates in the centrosome with vRNP during interphase [as shown in Fig 1C , and [14]] . At 48 h post transfection of YB-1 siRNA , the expression level of YB-1 in KD cells decreased to 25% of that in control cells ( S1 Fig ) . The virus titer in YB-1 KD cells decreased to approximately 30% of that in control cells ( Fig 3A ) . We also found that Rab11a does not accumulate in the centrosome of infected YB-1 KD cells ( Fig 3B ) , suggesting that YB-1 is required for pericentrosomal ERC formation . Note that YB-1 is responsible for centrosome maturation in order to establish the polarity-dependent dynamic instability in the mitotic phase [17] . Thus , we hypothesized that YB-1 may stimulate pericentrosomal ERC formation through spontaneous centrosome maturation in infected interphase cells as it does in the mitotic phase . To test this hypothesis , we examined the centrosomal localization of YB-1 using 3D-SIM super-resolution microscopy ( Fig 3C , 3D , 3E and 3F ) . Note that only centrosomes showing a cross-sectional view of PCM during interphase were selected for this analysis . YB-1 formed a toroidal structure with a beads-on-a-string distribution pattern around GFP-centrin-2 , a marker protein of the centriole ( Fig 3C and 3E ) . The mean diameter of the YB-1 toroid at the peak intensity ( 545 ± 48 nm; n = 8 ) was similar to that of pericentrin toroid ( a marker for PCM; 581 ± 42 nm; n = 8 ) , suggesting that YB-1 localizes in PCM ( Fig 3D and 3F ) . However , YB-1 did not co-localize with pericentrin ( Fig 3E ) . It has been proposed that pericentrin exists as elongated fibrils that extend radially from the centriole [5 , 6] . The spatial domains separated by pericentrin are filled with a number of PCM proteins required for microtubule nucleation and anchoring , suggesting that YB-1 also regulates the microtubule nucleation and/or anchoring at PCM in response to infection at interphases . Next , we observed the dynamics of microtubule nucleation to examine the centrosome function in infected cells using EB1-GFP [8] , which interacts specifically with growing microtubule ends ( Fig 4 and S1 , S2 , S3 and S4 Videos ) . The time series of EB1-GFP were acquired at 1 . 56-sec intervals for 1 min . In image sequences , EB1-GFP comets continually emerged from the centrosome . In the control , the mean growth rate of nucleated microtubules in the infected cells was increased compared to that of the uninfected mock cells ( Fig 4B , P<0 . 001 ) . In contrast , EB1-GFP in infected cells treated with YB-1 siRNA mostly did not move in a straight line , but rather in a Brownian motion ( Fig 4A and S4 Video ) . Because growing microtubule ends decorated with EB1-GFP accumulated primarily in the centrosome of infected YB-1 KD cells ( Fig 4A , arrow head ) , it is likely that the microtubules nucleated from the centrosome even in infected YB-1 KD cells . Therefore , it is possible that the newly synthesized microtubules are released from the centrosome in infected YB-1 KD cells . Further , although most microtubules were still elongated radially from the centrosome ( Fig 4A ) , some of the EB1-GFP signals showed a faster migration rate in uninfected YB-1 KD cells ( Fig 4A and 4B ) . It has been reported that short microtubules released from the centrosome migrate faster than the centrosomal microtubules [29] , therefore YB-1 appears to be required , at least in part , for anchoring microtubules to the centrosome in uninfected interphase cells . To address whether YB-1 is involved in the anchoring of microtubules to the centrosome in response to infection , we carried out microtubule regrowth assays using nocodazole , a potent inhibitor of microtubule polymerization ( Fig 5 ) . After nocodazole treatment for 1 h , microtubules were depolymerized , and α-tubulin was dispersed throughout the cytoplasm ( Fig 5B , 5G , 5L and 5Q ) . After washing out the drug , cells were incubated at 37°C to allow the regrowth of the microtubules for 3 , 5 , and 15 min . As expected , the nucleation of microtubules from the centrosome was stimulated by infection in control cells at 5 min post release ( Fig 5I ) . In contrast , noncentrosomal microtubules were sporadically found at peripheral regions of the cytoplasm in infected YB-1 KD cells ( Fig 5R and 5S , arrowheads ) . These results suggest that YB-1 is required for anchoring newly polymerized microtubules to PCM when the microtubule nucleation is stimulated by infection . ERC is reported to be involved in intracellular sorting and polarized trafficking of apical plasma membrane components [26] . However , details regarding the roles of ERC remain to be clarified . Therefore , we next examined the loading of vRNP onto the recycling endosomes by using YB-1 siRNA to disrupt ERC formation . Cells constitutively expressing FLAG-Rab11a were subjected to immunoprecipitation assays with anti-FLAG antibody ( Fig 6A ) . We found that the amount of PB1 subunit of viral polymerase and NP coimmunoprecipitated with FLAG-Rab11a from YB-1 KD lysates were decreased to approximately 30% of those from control lysates ( Fig 6A , lane 6 ) . This result is supported by the fact that vRNP hardly colocalized with Rab11a in YB-1 KD cells as shown in the enlarged panel of Fig 3B . Furthermore , we examined the activation of Rab11a in YB-1 KD cells by GST pull-down assays using GST-RBD . The amount of Rab11-GTP was not changed between the control and YB-1 KD cells ( Fig 6B ) , suggesting that YB-1 KD does not influence the amount of active recycling endosomes . Thus , it is likely that the formation of pericentrosomal ERC is important to load vRNP onto the endosomal vesicles . Cholesterol is not uniformly distributed in the membrane , and 80–90% of total cellular cholesterol is enriched in the plasma membrane [30] . Although recycling endosomes contain considerably less cholesterol than the plasma membrane , it is known that the endocytic transport pathway through recycling endosomes is important for cholesterol trafficking and homeostasis in cells [31 , 32] . Therefore , we hypothesized that vRNP is transported to the plasma membrane via recycling endosomes with cholesterol . To test this hypothesis , we observed the intracellular localization of cholesterol in infected cells using the fluorescent cholesterol-binding polyene antibiotic , filipin . Some recycling endosomes were partially colocalized with cholesterol in uninfected cells ( Fig 6C ) . However , along with the formation of pericentrosomal ERC by infection , we found that cholesterol is highly enriched in ERC in an YB-1-dependent manner . Similar results were obtained in A549 cells infected with A/Panama/2007/99 , which is one of the representative strains of seasonal influenza A virus ( H3N2 ) ( S2 Fig ) . These findings suggest that vRNP is transported to the plasma membrane via recycling endosome vesicles that contain a higher concentration of cholesterol . Some viruses , including influenza virus , are known to utilize lipid rafts for budding from the plasma membrane [33] . Viral budozone formation is thought to be dependent on the spatial assembly of eight-segmented vRNP complexes and viral membrane proteins via clustering of lipid rafts . Although it has been reported that reorganization of cortical actin is required for the control of viral budozone formation [25 , 34 , 35] , the trigger to initiate the coalescence of lipid rafts is unclear . Thus , we examined whether the pericentrosomal ERC is required for budozone formation by using in situ proximity ligation assay ( PLA ) to detect the proximity between M2 and HA . In the in situ PLA system , the theoretical maximum distance between two target proteins is around 40 nm to yield amplified signals . At 8 h post infection , cells were subjected to in situ PLA using anti-HA and either anti-M2 or anti-M1 antibodies ( Fig 7 ) . Strong punctate PLA signals ( red ) between HA and M2 or between HA and M1 were observed at the plasma membrane in the infected control cells ( Fig 7A and 7B ) . Although HA and M2 were successfully transported to the plasma membrane in YB-1 KD cells ( Fig 7C ) , the intensity of PLA signals between HA and M2 was significantly decreased by YB-1 KD ( P<0 . 001; Fig 7B , left panel ) . In contrast , the signal intensity between HA and M1 was not decreased in YB-1 KD cells ( Fig 7B , right panel ) . This could be due to the direct binding of M1 with the cytoplasmic tail of HA [19] . Next , we examined whether cholesterol is required for the YB-1-dependent viral budozone formation using nonraft HA mutant virus , which has alanine substitutions at I533 , Y534 , and S535 in the transmembrane domain of HA . It is reported that this mutant HA rarely associates with lipid rafts and that the apical transport is delayed , but not blocked [36] . At 12 h post infection , a significant amount of HA was observed at the plasma membrane in nonraft virus infected cells ( Fig 7D , green ) . However , the intensity of PLA signals between HA and M2 was dramatically reduced in nonraft virus-infected cells compared with that in wild-type infected cells ( Fig 7D and 7E ) . Thus , as expected , it is likely that most of the signals observed in the in situ PLA system were mediated by lipid rafts . Furthermore , in contrast to wild type virus ( Fig 7B ) , the PLA signals between nonraft HA and M2 were nearly unaffected by YB-1 KD ( Fig 7E , compare lane 2 with lane 3 ) , suggesting that the interaction of HA with cholesterol is important for YB-1-mediated viral budozone formation .
The lipid-lipid , lipid-protein , and protein-protein interactions facilitate the formation of small raft domains into functional platforms for signal transduction , membrane trafficking , and cell adhesion [37–39] . Sphingolipids that have been enriched in these assemblies have saturated and longer acyl chains with larger polar headgroups , so cholesterol functions as spacers between sphingolipids through their acyl chains [40] . This cholesterol-sphingolipids interaction results in the packing and condensing of lipid rafts for their clustering . Fig 7 shows that YB-1 is important for clustering of viral membrane proteins at the plasma membrane through the interaction of viral raft protein with cholesterol . It is noteworthy that the amount of cholesterol at the plasma membrane was unchanged between the control and YB-1 KD cells ( S3 Fig ) , suggesting that small raft domains should be intact at the plasma membrane in YB-1 KD cells . This is possibly due to the fact that the recycling endosomes and TGN contain much less cholesterol than the plasma membrane [41] . However , it is known that moderate changes in the level of cholesterol transported through these compartments appear to have drastic effects on cellular homeostasis [41] . Taking these findings together , we propose that the fusion of cholesterol-enriched recycling endosomes with the plasma membrane induces the accumulation of sphingolipids that contain viral raft proteins which form viral budozone concomitantly with the arrival of vRNP beneath the plasma membrane ( Fig 8 ) . In general , cells acquire cholesterol mainly through receptor-mediated endocytosis of low-density lipoprotein ( LDL ) [30] . After LDL internalization , LDL-cholesterol is delivered to late endosomes and lysosomes to release the cholesterol molecules from LDL . The majority of cholesterol in late endosomes is then delivered to the plasma membrane . Although the itinerary of cholesterol from late endosomes to the plasma membrane is not clear , it is thought that cholesterol is transported through ER , TGN , and recycling endosomes . We found that influenza virus infection stimulates cholesterol accumulation in ERC ( Fig 6C ) . This could be due to a possibility that the accumulation of recycling endosomes in ERC ( Fig 1A ) may slow down the delivery of cholesterol to the plasma membrane . YB-1 is required for centrosome maturation during mitosis [17] , but little is known about the function of YB-1 in the centrosome . In infected cells , YB-1 was localized in PCM and formed a radial and toroidal structure around the centriole ( Fig 3 ) . It is proposed that the PCM proteins might be assembled based on the nine-fold radial symmetry of the centriole [5 , 6] . In which case , it is assumed that YB-1 is also a structural component of the PCM matrix for microtubule assembly . It is also reported that YB-1 interacts with microtubules and coats the outer surface of the microtubule wall in vitro [42] . Thus , YB-1 may connect microtubules to the PCM matrix by decorating the microtubules’ minus ends . The spatiotemporal regulation of Rab GTPase activity is of particular importance . Among the several GEFs known to regulate Rab GTPases , no GEF that activates Rab11a has been identified in mammalian cells despite a systematic characterization of the DENN domain subfamily of Rab GEFs [43] . It is necessary to identify the GEFs responsible for virus infection . Rab11a plays a role in the transport of M2 to the apical membrane [25] , although M2 is directly transported through TGN to the plasma membrane [44] . This is due to the fact that Rab11a also functions in constitutive exocytosis from TGN in addition to the recycling processes via ERC [45 , 46] . In YB-1 KD cells , HA and M2 were successfully transported to the plasma membrane ( Fig 7D ) , suggesting that centrosome maturation by YB-1 is required for their transport through ERC but not through TGN . It has been reported that the minus end of microtubules , which is released from the centrosome , could subsequently be captured by the Golgi membrane and then elongated into linear arrays [47] . Thus , even in the absence of YB-1 , the exocytic transport from TGN might be achieved along microtubules that are elongated from Golgi stacks . The majority of membrane proteins are sorted at TGN before their delivery to the appropriate cell surface domain . In addition to TGN , some other cellular lipid raft proteins , such as TLR4 and EGF receptor , are transported to the plasma membrane through the recycling endosomes [48 , 49] . Additionally , the transport rates of recycling endosomes are controlled in response to signaling pathways to increase or decrease the surface expression of molecules , such as insulin-regulated glucose transporter GLUT4 [50 , 51] . In this study , we propose that the recycling endosomes deliver cholesterol to the plasma membrane for not only cholesterol homeostasis , but also lipid raft clustering . Our findings contribute to the understanding of the molecular mechanism of lipid raft clustering in response to several signals that utilize lipid rafts as a platform .
Influenza virus A/Puerto Rico/8/34 strain and rabbit polyclonal antibodies against PB1 , NP , M1 , and YB-1 were prepared as previously described [14] . Mouse antibodies against HA ( TaKaRa; C179 ) , Rab11a ( BD; 47/Rab11 ) , Pericentrin ( Abcam; ab28144 ) , α-tubulin ( Sigma; DM1A ) , and a rabbit antibody against M2 ( Abcam; ab56086 ) were purchased . HeLa cells ( a gift from Dr . Masa-atsu Yamada of University of Tokyo ) were grown in minimal essential medium ( MEM ) containing 10% fetal bovine serum . Plasmids expressing GFP-centrin-2 and EB1-GFP were prepared as previously described [14] . To establish HeLa cell lines constitutively expressing either GFP-centrin-2 or EB1-GFP , cells were transfected with pSV2-Neo and either pCAGGS-GFP-centrin-2 or pCAGGS-EB1-GFP . The transfected cells were cultured in the presence of 1 mg/ml of G418 for 2 weeks , and then the G418-resistant colonies were isolated . For the construction of plasmid expressing GST-Rab-binding domain ( RBD ) of FIP2 , cDNA was amplified from pCAGGS-FIP2 ( provided by Dr . F . Momose , Kitasato University ) with primers 5ʹ-CCGGAATTCGAGCTGGTGAAACAC-3ʹ and 5ʹ-ACGCGTCGACTCACGGCACTCTGAG-3ʹ . The cDNA was cloned into pGEX-6P-1 . Nonraft HA virus was generously provided by Drs . Y . Morikawa and F . Momose ( Kitasato University ) [36] and amplified using MDCK cells constitutively expressing HA ( provided by Dr . N . Takizawa , Institute of Microbial Chemistry ) . Transferrin conjugated with Alexa 568 was purchased ( Life Technologies ) . Cells were incubated with 100 μg/ml of Transferrin for 30 min at 37°C . After washing with medium , cells were further incubated for 30 min at 37°C and then fixed in 4% paraformaldehyde ( PFA ) . Indirect immunofluorescence assays and fluorescence in situ hybridization ( FISH ) assays were carried out as previously described [14] . Briefly , cells infected with influenza virus at multiplicity of infection ( MOI ) of 10 were fixed with 1% PFA for 10 min and then pre-permeabilized on ice with 0 . 01% digitonin in PBS for 5 min on ice . After being washed with PBS , cells were fixed in 4% PFA for 10 min and permeabilized on ice with 0 . 5% Triton X-100 in PBS for 5 min . After incubation in PBS containing 1% bovine serum albumin for 1 h , coverslips were incubated with each antibody for 1 h and then with Alexa Fluor 488- , 568- , and 633-conjugated secondary antibodies , respectively ( Life Technologies ) . After indirect immunofluorescence assays , FISH assays were performed using an RNA probe complementary to the segment 1 virus genome . Images were acquired using confocal laser scanning microscopy ( LSM700; Carl Zeiss ) or super-resolution microscopy ( 3D-SIM ELYRA; Carl Zeiss ) . Cells were fixed in 4% PFA for 10 min and then incubated with 200 μg/ml of filipin ( Sigma ) . After washing with PBS , images were acquired by Axio Observer Z1 microscope using 63x Apochromat objective ( NA = 1 . 4 ) with AxioCam MRm camera ( Carl Zeiss ) . Observations were made with Axio Observer Z1 microscope using 63x Apochromat objective . Images were acquired at 1 . 56-sec intervals for 1 min with confocal laser scanning microscopy ( LSM700; Carl Zeiss ) . All experiments were carried out at 37°C under 5% CO2 in a temperature-controlled stage ( Carl Zeiss ) . Sequential images were processed using Image J digital image processing software ( National Institutes of Health , Bethesda ) . The average velocity of the punctate fluorescent signals of EB1-GFP was measured using a manual object tracking plugin , MTrackJ , for Image J . Cells were fixed with 4% PFA , followed by blocking with 1% milk for 30 min . The cells were incubated with mouse anti-HA antibody for 1 h and fixed again in 4% PFA . Cells were then permeabilized with 0 . 5% Triton X-100 for 5 min and incubated with either rabbit anti-M1 or anti-M2 antibody for 1 h . PLA was carried out using Duolink In Situ PLA kit ( Olink Bioscience ) according to the manufacturer’s protocol . The mean intensity of the PLA signals was measured using IMARIS software ( Carl Zeiss ) . Knockdown of YB-1 was examined as previously described [14] . Briefly , cells ( 5 x105 ) were transfected with 30 pmol of siRNA using Lipofectamine RNAi Max ( Life Technologies ) according to the manufacturer’s protocol . | Influenza virus particles are assembled at the plasma membrane in concert with incorporation of the virus genome , but the details of its spatiotemporal regulation are unknown . We found that the virus genome is transported to the plasma membrane using cholesterol-enriched recycling endosomes through cell cycle-independent activation of the centrosome by recruiting YB-1 , which is a mitotic centrosomal protein . We also revealed that the cholesterol-enriched endosomes are important for clustering of viral structural proteins at lipid rafts to assemble the virus particles . These results suggest that local accumulation of cholesterol , via fusion of endosomes to the plasma membrane , is one of the triggers for the virus assembly concomitantly with arrival of the virus genome beneath the plasma membrane . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Influenza Virus Induces Cholesterol-Enriched Endocytic Recycling Compartments for Budozone Formation via Cell Cycle-Independent Centrosome Maturation |
While the exopolysaccharide component of the biofilm matrix has been intensively studied , much less is known about matrix-associated proteins . To better understand the role of these proteins , we undertook a proteomic analysis of the V . cholerae biofilm matrix . Here we show that the two matrix-associated proteins , Bap1 and RbmA , perform distinct roles in the biofilm matrix . RbmA strengthens intercellular attachments . In contrast , Bap1 is concentrated on surfaces where it serves to anchor the biofilm and recruit cells not yet committed to the sessile lifestyle . This is the first example of a biofilm-derived , communally synthesized conditioning film that stabilizes the association of multilayer biofilms with a surface and facilitates recruitment of planktonic bystanders to the substratum . These studies define a novel paradigm for spatial and functional differentiation of proteins in the biofilm matrix and provide evidence for bacterial cooperation in maintenance and expansion of the multilayer biofilm .
Bacterial biofilm formation is the process by which bacteria attach to abiotic surfaces , the surfaces of other unicellular organisms , the epithelia of multicellular organisms , and interfaces such as that between air and water . Surface adhesion enables bacteria to arrange themselves favorably in their environment and , therefore , is critical to environmental adaptation and survival . Surface-attached bacteria may form either a single layered structure , known as a monolayer , or a multilayer biofilm [1] . Bacterial cells join monolayer and multilayer biofilms in response to distinct environmental signals , use distinct structures for adhesion in these two biofilms , and develop distinct transcriptional profiles within these two structures [2] , [3] . However , the critical difference between these two types of biofilms is the extracellular matrix that surrounds cells in the multilayer biofilm . This matrix is comprised of biological polymers such as exopolysaccharide , protein , and DNA [4] . The matrix not only mediates bacterial aggregation and surface attachment but may also serve as a reservoir for extracellular , degradative enzymes and the nutrients released by their function . Therefore , the multilayer biofilm affords the bacterium advantages that monolayer biofilm does not . Vibrio cholerae is a halophilic Gram-negative bacterium that causes the severe diarrheal disease cholera . V . cholerae makes two types of multilayer biofilms . One is dependent on environmental Ca2+ concentrations comparable to those found in seawater , while the other is dependent on the synthesis of an exopolysaccharide termed VPS [2] , [5] , [6] , [7] . The genes required to synthesize VPS are primarily found in two large operons within the VPS island , one of which encodes the proteins VpsA through VpsK and the other of which encodes VpsL through VpsQ [8] . Transcription of these operons is controlled by a complex regulatory network , suggesting that the ability to limit biofilm matrix synthesis to a highly specific environmental niche confers a survival advantage [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] . While most studies suggest that the VPS-dependent V . cholerae biofilm is not important for colonization of the human intestine [18] , [19] , [20] , this biofilm may be important for environmental persistence . Surface-attached V . cholerae predominate in the environment [21] , [22] . Multiple avenues of evidence suggest that the chitinaceous surfaces of arthropods are an important substratum for V . cholerae biofilm formation [23] , [24] , [25] , [26] , [27] . Furthermore , V . cholerae is especially well adapted to life on chitin because of its many chitinolytic enzymes , the marked modulation of its transcriptome by the degradation products of chitin [28] , and activation of its natural competence by chitin [28] , [29] , [30] , [31] . Our laboratory and others have identified several environmental signals that activate VPS-dependent V . cholerae biofilm formation [2] , [5] , [12] , [32] , [33] , [34] , [35] . Among these are sugars transported by the phosphoenolpyruvate phosphotransferase system or PTS . Chitobiose and N-acetylglucosamine , which are degradation products of chitinaceous surfaces of arthropods , are transported exclusively by the PTS [36] , [37] . Therefore , in the aquatic environment , association with arthropods is likely correlated with formation of a VPS-dependent biofilm . To gain insight into the role of biofilm matrix-associated proteins in V . cholerae surface attachment , we set out to define the proteome of the V . cholerae biofilm matrix . Here , we present evidence that the biofilm matrix selectively retains secreted proteins . Furthermore , we show that RbmA and Bap1 , two proteins of previously unknown function [3] , [38] , [39] , [40] , are present in the biofilm matrix . While RbmA functions similarly to previously identified biofilm matrix proteins in that it strengthens intercellular interactions [41] , [42] , [43] , Bap1 , which is jointly synthesized by biofilm-associated bacteria , is concentrated at the base of the biofilm where it reinforces the association of the biofilm with the surface and accelerates attachment of bystander bacteria not yet primed for biofilm matrix synthesis . These studies present evidence for specialization of proteins in the bacterial biofilm matrix and for bacterial cooperation in maintaining and expanding surface-associated biofilms .
Biofilm matrix proteins were isolated by a variety of methods . Briefly , biofilms were disrupted by vortexing in the presence or absence of 1 mm glass beads . Furthermore , biotinylation of extracellular proteins prior to biofilm disruption and subsequent neutravidin affinity purification were used to enrich for extracellular proteins . The protein mixtures prepared by these methods were analyzed by MS/MS . We then used in silico methods ( Genome Atlas ) to predict the subcellular localization of identified proteins . As shown in Figure S1 , the proportion of recovered proteins that were predicted to be extracytoplasmic increased with biotinylation . Gentler methods of biofilm disruption also resulted in isolation of a larger proportion of predicted extracytoplasmic proteins . However , the most gentle disruption methods yielded fewer proteins overall and , therefore , a smaller number of secreted proteins . The 74 predicted extracytoplasmic proteins identified by these methods are listed in Table S1 . Based on either known function or bioinformatics , we predicted that 10 of these proteins were secreted and , therefore , were candidate biofilm matrix-associated proteins ( Table 1 ) . In addition , 17 of these proteins were located in the outer membrane ( OM ) , and 26 of these proteins were located in the periplasm . The location of 18 proteins could not be predicted with certainty ( Table 2 ) . Citrate synthase ( VC2092 ) and a putative acetyl CoA synthase homolog ( VCA0139 ) , which were predicted to have transmembrane domains , are most likely in the inner membrane . No additional inner membrane proteins were identified . NusA ( VC0642 ) , a transcription elongation factor we identified in the proteomic analysis , was predicted to be secreted . However , because of its function , we hypothesize that it is cytoplasmic . Secreted proteins identified in our analysis included those forming bacterial appendages such as the mannose-sensitive hemagglutinin type IV pilus ( MshA ) and the flagellum as well as RbmA and RbmC , two proteins of unknown function that alter biofilm formation and are co-regulated with the VPS synthesis genes [3] , [38] , [39] , [40] . Three proteins not previously associated with biofilms were also identified , namely a hemolysin ( HlyA , VCA0219 ) , a chitinase ( VCA0027 ) , and the hemagglutinin/protease ( HAP; VCA0865 ) . We hypothesized that cell-associated proteins should be similarly represented in our analyses if they represented residual cellular material contaminating the biofilm matrix preparation . In fact , while 35% of the periplasmic and OM proteins identified were found in three or more biofilm matrix preparations , only 7% of all predicted cytoplasmic proteins were identified in 3 or more preparations . Furthermore , only two putative inner membrane proteins were identified . One possibility is that a common step in the purification process resulted in formation of spheroplasts , releasing outer membrane and periplasmic proteins into the supernatant during the purification process . Another possibility is that these OM and periplasmic proteins signify the presence of outer membrane vesicles in the biofilm matrix . RbmC ( 957 aa ) , which was identified in our proteomic analysis , and its homolog Bap1 ( 691 aa ) play uncharacterized and redundant roles in the observed colony morphology and biofilm phenotype of rugose V . cholerae variants [39] . The central portions of these proteins are 54% identical and 70% similar and include an EF hand domain , which is predicted to bind Ca2+ , and a β-prism lectin-like domain surrounded by six FG-GAP domains ( Figure 1A ) . RbmC is longer than Bap1 due to two N-terminal domains of unknown function that are also found in the E . coli mucinase StcE and a second C-terminal β-prism domain [44] , [45] . We first confirmed that these proteins also serve redundant roles in biofilm formation in our V . cholerae strain MO10 , which has a smooth rather than rugose colony morphology . As shown in Figure 1B , Δbap1 and ΔrbmC mutants formed a biofilm , while the double mutant did not . The biofilm defect of the Δbap1ΔrbmC mutant could be rescued by a plasmid encoding a wild-type allele of either Bap1 or RbmC ( Figure S2 ) . An rbmC allele with a truncation of the C-terminal β-prism domain not found in Bap1 ( RbmC-C140 ) also rescued the biofilm defect of the Δbap1ΔrbmC mutant ( Figure S2 ) . These results suggest that , as previously noted for a rugose variant of V . cholerae , Bap1 and RbmC perform redundant functions in the V . cholerae biofilm . Furthermore , Bap1 represents the minimal protein required to rescue the Δbap1ΔrbmC mutant phenotype . Our proteomic analysis identified ten candidate matrix-associated proteins ( Table 1 ) . ChiA-2 , MshA , Bap1 , RbmA , and the hemolysin HlyA were selected for further study . To determine whether these proteins were secreted by V . cholerae , the gene encoding each of these proteins was cloned between an inducible promoter and a C-terminal FLAG tag . As negative controls , we also cloned EIIAGlc ( VC0964 ) , a cytoplasmic component of the phosphoenolpyruvate phosphotransferase system , as well as Escherichia coli alkaline phosphatase ( AP ) and TcpG ( VC0034 ) , two periplasmic proteins . Each of these plasmids was introduced into V . cholerae . After culture in LB broth , the cells and supernatant were separated by centrifugation , and the presence of the tagged protein in each fraction was assessed by Western analysis ( Figure 2A ) . The negative controls EllAGlc , TcpG , and AP were found in the cell pellet only . The secreted proteins chosen for further study were all found in the supernatant to varying degrees . To determine if these secreted proteins were retained in the biofilm matrix , we formed biofilms with wild-type V . cholerae constitutively expressing affinity-tagged versions of each of these proteins . Biofilms were rinsed , and immunofluorescence was used to visualize the affinity-tagged proteins in the biofilm matrix . No fluorescence was observed for biofilms formed by strains carrying plasmids encoding the proteins EllAGlc , TcpG , or AP ( data not shown ) . As expected , the pilus-forming protein MshA was visualized in the biofilm matrix . In addition , RbmA , Bap1 , and HlyA were observed in the biofilm matrix . Although comparable amounts of ChiA-2 were secreted , much less was observed in the biofilm matrix ( Figure 2B ) . This suggests that RbmA , Bap1 , and HlyA are selectively retained in the biofilm , while ChiA-2 does not associate strongly with the biofilm matrix . Bap1 and RbmA were previously found to alter biofilm formation [3] , [38] , [40] . Therefore , we hypothesized that their role in biofilm formation might be a structural one . To compare the native distributions of Bap1 and RbmA in the biofilm , we fused a FLAG tag to the C-terminal end of Bap1 and RbmA on the chromosome and visualized these tagged proteins in the biofilm by immunofluorescence . As shown in Figure 3A , RbmA was evenly distributed in the vertical dimension , while Bap1 was concentrated at the base of the biofilm . To objectively assess this difference , we measured the total fluorescence intensity in each transverse section . For each biofilm , this measurement was normalized to the transverse section with maximum fluorescence intensity and plotted as a function of distance from the substratum . As shown in Figure 3B , these measurements confirmed that Bap1 was concentrated at the biofilm-surface interface . To determine whether the distinct vertical distributions of Bap1 and RbmA in the biofilm were the result of spatially heterogeneous transcription of bap1 and rbmA , we formed a biofilm with wild-type V . cholerae constitutively expressing Bap1-FLAG or RbmA-6XHis from a plasmid . As shown in Figure 4 , the vertical distribution of Bap1 and RbmA in these biofilms was similar to that in biofilms expressing Bap1 or RbmA from their respective native promoters . However , with constitutive expression , more Bap1 was observed within the biofilm , most likely due to increased levels of protein . Taken together , our data suggest that the vertical distributions of Bap1 and RbmA in the biofilm are not the result of heterogeneous transcription of bap1 and rbmA within the biofilm . Rather , we hypothesize that Bap1 migrates to the biofilm-substratum interface after secretion from the cell . To assess the transverse distribution of Bap1 and RbmA in the biofilm and the extent of co-localization of these two proteins , we combined equal numbers of a Δbap1 mutant expressing Bap1-FLAG from a plasmid and wild-type V . cholerae expressing RbmA-His from a plasmid . As shown in Figure 5 , in transverse sections close to the substratum , Bap1 and RbmA were both distributed around the perimeter of cells , and some co-localization was observed . However , RbmA was more evenly distributed , while foci of increased intensity were observed for Bap1 . Similar transverse distributions of each protein were observed in biofilms formed by a Δbap1 mutant expressing Bap1-FLAG from a plasmid alone and by a ΔrbmA mutant expressing RbmA-FLAG from a plasmid alone ( data not shown ) . Based on these observations , we hypothesized that Bap1 might play a different role than RbmA in biofilm formation . RbmA alters biofilm stability but not overall biofilm accumulation of rugose variants of V . cholerae [38] . In V . cholerae O139 strain MO10 , we observed that deletion of rbmA had a small , statistically insignificant effect on biofilm formation . Rescue of a ΔrbmA mutant with a wild-type rbmA allele produced a biofilm that was similar to that of wild-type V . cholerae but significantly increased as compared with the biofilm of the unrescued mutant ( Figure 6A ) . Vortexing completely dispersed the ΔrbmA mutant biofilm , while larger biofilm fragments were observed after similar treatment of the wild-type V . cholerae biofilm ( Figure 6C ) . This ΔrbmA mutant phenotype could be complemented by expression of a wild-type rbmA allele in trans . We hypothesized that , if secreted RbmA were essential for biofilm integrity , exogenous RbmA should rescue the biofilm defect of a ΔrbmA mutant . To test this , we first affinity purified RbmA ( Figure 6B ) . We then allowed the ΔrbmA mutant to form a biofilm in the presence of increasing amounts of purified RbmA . Purified RbmA was able to rescue the biofilm defect of the ΔrbmA mutant ( Figure 6C ) . We determined that rescue required an RbmA concentration of approximately 416 nM . Assuming all molecules of RbmA are functional , this corresponds to approximately 260 , 000 molecules per mutant cell . In a standard assay , the biofilm formed by a Δbap1ΔrbmC mutant was indistinguishable from that formed by a ΔvpsL mutant ( Figure 1B ) . However , we noticed that , unlike the ΔvpsL mutant , the Δbap1ΔrbmC mutant formed a pellicle on the liquid surface after 24 hours of static growth . Interestingly , mutation of bap1 and rbmC in a rugose variant of V . cholerae was not noted to preserve pellicle formation [39] . One possible explanation for this discrepancy is that , due to a difference in the surface chemistries of smooth and rugose variants , rugose variants do not interact as strongly with the air-water interface in the absence of Bap1 and RbmC . As shown in Figure 7 , the pellicle formed by the Δbap1ΔrbmC mutant was loosely associated with the glass surface . Gentle shaking dislodged the Δbap1ΔrbmC mutant pellicle from the substratum sending it to the bottom of the tube , while the wild-type pellicle remained attached . Furthermore , vortexing of the Δbap1ΔrbmC mutant pellicle caused it to fragment into many small pieces . However , these pieces were larger than those observed when a ΔrbmA biofilm was vortexed . These defects were rescued by a wild-type bap1 allele provided in trans but not by rbmA ( Figure 7 ) , again indicating that Bap1 and RbmA have distinct roles in biofilm formation . We hypothesized that if secreted Bap1 were responsible for adhesion of the biofilm to the surface , exogenously provided Bap1 should also rescue the Δbap1ΔrbmC mutant biofilm defect . To test this prediction , we used affinity chromatography to purify Bap1-FLAG as shown in Figure 8A . A Δbap1ΔrbmC mutant incubated in the presence of purified Bap1 formed a biofilm that was comparable to that of a Δbap1ΔrbmC mutant rescued by Bap1 expressed from a plasmid ( Figure 8B ) . To determine the concentration of Bap1 required to restore biofilm formation to the Δbap1ΔrbmC mutant , we titrated purified Bap1-FLAG into a Δbap1ΔrbmC mutant culture and measured biofilm formation after 24 hours . As shown in Figure 8C , an 8 . 8 nM solution of Bap1-FLAG was sufficient to restore surface attachment . Assuming all Bap1 molecules are functional , this corresponds to approximately 5 , 500 Bap1 molecules per bacterial cell . Therefore , approximately 47 times less Bap1 was required than RbmA to form a biofilm with properties similar to that of wild-type V . cholerae . To validate these quantifications in a native biofilm , we used Western analysis to estimate the relative quantities of Bap1-FLAG and RbmA-FLAG synthesized in biofilms formed with V . cholerae strains expressing either Bap1-FLAG or RbmA-FLAG from the native chromosomal location ( Figure 8D ) . Two bands were always observed for biofilm-associated RbmA , suggesting that RbmA undergoes proteolysis in the biofilm . Including both RbmA bands in the calculation , we determined that there was approximately 16 times less Bap1 in biofilm preparations as compared with RbmA , recapitulating our results with purified protein . We hypothesize that less Bap1 is required in the biofilm because it principally associates with the base of the biofilm , whereas RbmA is distributed evenly throughout . Because exogenously provided Bap1 restored biofilm surface adhesion to a Δbap1ΔrbmC mutant , we questioned whether Bap1 synthesis could be a joint venture in the biofilm community . To test this , we co-cultured a Δbap1ΔrbmC mutant with a ΔvpsL mutant . As shown in Figure 7 , this produced a biofilm that was comparable to that of wild-type V . cholerae . We rationalized that ( i ) this biofilm might be comprised chiefly of ΔvpsL mutant cells because the Δbap1ΔrbmC mutant was providing it with the requisite biofilm exopolysaccharide , ( ii ) the Δbap1ΔrbmC mutant might predominate because the ΔvpsL mutant was providing it with the requisite Bap1 and/or RbmC , or ( iii ) approximately equal numbers of these two mutants might be found in the biofilm because each was providing the other with the requisite materials for biofilm formation . To determine whether Bap1 , VPS , or both were shared resources within the biofilm , we performed a series of co-culture biofilm experiments using lacZ as a marker and determined the relative amounts of each species in the biofilm by plating on media containing 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside ( X-Gal ) . As shown in Figure 9A , although the lacZ+ strain always had a slight advantage in the biofilm , cells lacking Bap1 and RbmC were always found in the biofilm when co-cultured with cells that were able to produce these proteins . In contrast , cells that were unable to synthesize VPS were always excluded from the biofilm in spite of co-culture with cells that were able to synthesize VPS . Based on these findings , we hypothesize that Bap1 is a shared biofilm resource , but VPS is not . To document communal Bap1 in the V . cholerae biofilm , we then co-cultured a Δbap1ΔrbmC mutant with a ΔvpsL mutant expressing GFP from a chromosomal location and Bap1-FLAG from a plasmid . The biofilms harvested from these co-culture experiments were visualized by microscopy after immunofluorescent staining of Bap1-FLAG and DAPI staining of bacterial DNA . As expected , approximately one GFP-labeled ΔvpsL mutant cell was observed in the biofilm for every GFP-negative Δbap1ΔrbmC mutant cell ( Figure 9B ) . However , the perimeter of many Δbap1ΔrbmC mutant cells exhibited Bap1-FLAG-based immunofluorescence . To confirm that this observation was not the result of transfer of the plasmid from the ΔvpsL mutant to the Δbap1ΔrbmC mutant , we documented that all Δbap1ΔrbmC mutant cells in the biofilm remained sensitive to ampicillin ( data not shown ) . Our results confirm that Bap1-FLAG provided by a ΔvpsL mutant can be incorporated into the Δbap1ΔrbmC mutant biofilm . These findings indicate that Bap1 is a communal resource . In contrast , because ΔvpsL mutant cells were excluded from both Δbap1ΔrbmC mutant and wild-type V . cholerae biofilms , we conclude that VPS produced by neighboring cells is not available to the ΔvpsL mutant and , therefore , that unlike Bap1 , the biofilm exopolysaccharide VPS is not a communal resource but instead tightly associated with the cell of origin . We questioned whether Bap1 could also increase surface adhesion of bystander cells not yet committed to the sessile life style , as this would have implications for the role of Bap1 in biofilm expansion . We previously identified a medium in which V . cholerae does not synthesize enough of the biofilm matrix components to proceed past the monolayer stage of biofilm development [2] . We cultured wild-type V . cholerae , a Δbap1ΔrbmC mutant , or a ΔvpsL mutant in monolayer minimal medium with supplemented with purified Bap1 . As shown in Figure 10A and quantified in Figure 10B , Bap1 increased surface adhesion of wild-type V . cholerae , a Δbap1ΔrbmC mutant , and a ΔvpsL mutant in monolayer minimal medium , while a control protein , bovine serum albumin ( BSA ) , had no effect . This suggests that communal Bap1 secreted by nearby biofilm cells may also increase surface adhesion of bystanders that have not yet been reprogrammed for biofilm matrix synthesis .
The exopolysaccharide component of the bacterial biofilm matrix has been studied intensively [8] , [46] , [47] , [48] , [49] , [50] , [51] . More recently , components such as DNA and protein have been identified in the matrices of some bacterial biofilms . Here we provide the first proteomic analysis of a Gram-negative biofilm matrix . Our analysis revealed 10 secreted proteins , 43 periplasmic and outer membrane proteins , and 18 putative extracytoplasmic proteins whose location could not be predicted . OM and periplasmic proteins were much more likely to be identified in multiple matrix preparations than inner membrane and cytoplasmic proteins , suggesting that these proteins may not be artifacts caused by cell lysis but rather the contents of biofilm-associated OM vesicles . Outer membrane vesicles are retained in the biofilms of Pseudomonas aeruginosa and Helicobacter pylori , and these vesicles appear to play a role in biofilm formation [52] , [53] , [54] . Furthermore , there is evidence that the compositions of membrane vesicles derived from the biofilm and from culture supernatants are distinct [52] , [55] . V . cholerae has been reported to release outer membrane vesicles [56] , [57] , [58] . However , additional investigations are required to confirm the presence of these vesicles in the biofilm matrix and to determine their role in biofilm formation . We studied four secreted proteins identified in our preliminary analysis in addition to MshA . The chitinase , ChiA-2 , showed minimal retention in the biofilm matrix . However , three proteins of unknown function , Bap1 , RbmA , and HlyA showed extensive association with the matrix . RbmA has no conserved domains of known function . Bap1 , its homolog RbmC , and HlyA , which all contain at least one β-prism lectin domain , form a paralogous family in V . cholerae . We hypothesize that these secreted proteins are selectively retained in the biofilm , perhaps by binding to specific moieties in the polysaccharide scaffold . Bap1 and RbmA were previously shown to play an undefined role in V . cholerae biofilm formation [3] , [38] , [39] . Here we show that RbmA and Bap1 have distinct distributions in the biofilm matrix . RbmA surrounds biofilm-associated cells throughout the biofilm and reinforces intercellular contacts from this location . In contrast , Bap1 concentrates around cells that form the biofilm-surface interface and stabilizes adhesion of the biofilm to surfaces . The distinct distribution of these proteins is not the result of heterogeneous expression within the biofilm . Rather , we hypothesize that it is the result of self-segregation after secretion from the cell . This is the first example of spatial and functional differentiation of secreted structural proteins in a Gram-negative biofilm matrix . Biofilm matrix polysaccharide is considered to be a jointly synthesized , shared resource . We show here that this is not the case for the V . cholerae biofilm matrix . While the biofilm matrix protein Bap1 is a communal resource , VPS benefits only cells from which it is synthesized . Therefore , the V . cholerae biofilm exopolysaccharide is not freely secreted and available to the entire community . Lastly , our results suggest that matrix-associated proteins may play an important role in expansion of existing bacterial biofilms on surfaces . Exogenous Bap1 increases surface adhesion of planktonic bystanders as well . Because nutritional signals and surface attachment are strong activators of the biofilm matrix synthesis genes , in aquatic environments , it is unlikely that planktonic Bap1 and RbmC would be synthesized by planktonic cells in quantities sufficient to increase surface attachment . Rather , we envision that Bap1 and RbmC secreted from an existing biofilm would condition surrounding surfaces , increasing the probability of bystander cell attachment . These studies reveal a new paradigm for the bacterial biofilm matrix in which the biofilm exopolysaccharide forms a cell-associated scaffold to which communal biofilm matrix proteins adhere , possibly through carbohydrate-binding domains . These proteins may fulfill specialized structural roles or enable cooperative augmentation of the biofilm .
The bacterial strains and plasmids used in this study are listed in Table S2 . Vectors used for protein expression included either an IPTG inducible promoter and a FLAG-tag ( pFLAG-CTC , Sigma-Aldrich ) or an arabinose inducible promoter and a 6X-His tag ( pBAD-Topo , Invitrogen ) . Bacteria were cultivated either in Luria-Bertani broth ( LB ) or monolayer minimal media [2] . Where indicated , streptomycin ( 100 µg/ml ) , ampicillin ( 50 or 100 µg/ml ) , arabinose ( 0 . 04% wt/vol ) , and Isopropyl β-D-1-thiogalactopyranoside ( 1 mM ) ( IPTG ) were added to the growth medium . A 0 . 1 M phosphate-buffered saline solution ( PBS ) ( pH 7 . 0 ) was used in initial biofilm washes , and a 0 . 1 M Tris-buffered saline solution ( TBS ) ( ph 7 . 0 ) was used to wash biofilms after biotin labeling . 10 mls of LB broth supplemented with streptomycin was added to a Petri dish and inoculated with V . cholerae . A biofilm including a pellicle formed over 48 hours of static incubation at 27°C . After incubation , the associated planktonic cells were removed . The remaining biofilm was washed by addition of PBS , agitation on a rotary shaker for 5 minutes with PBS , removal of PBS and non-attached cells , and addition of fresh PBS . This procedure was repeated twice . Matrix proteins were then prepared using each of the following four protocols ( Figure S1 ) . In preparation ( i ) , the biofilm was disrupted in the presence of 1 . 0 mm glass beads ( Biospec ) and centrifuged to remove particulates . For preparations ( ii ) , ( iii ) , and ( iv ) , a cell surface biotinylation kit ( Pierce ) was used to biotinylate extracytoplasmic proteins in the washed biofilm according to the manufacturer's instructions . After biotinylation , the biofilm was transferred to a 50 ml conical tube containing 2 mls of PBS . Disruption of the pellicle was carried out by ten sonication cycles of 10 sec ( iii ) or by vortexing in the presence ( ii ) or absence ( iv ) of 1 . 0 mm glass beads ( Biospec ) for one minute . The mixtures were then centrifuged at 20 , 000× g for 30 min in the cold to remove particulates , the supernatants were applied to Neutravidin-agarose resin ( Pierce ) , and the resin was washed several times with PBS . Biotinylated proteins were eluted from the resin by incubation with PBS to which 50 mM DTT had been added . This disrupts the disulfide bonds bridging biotin residues to extracellular proteins . The four mixtures of proteins were precipitated with trichloroacetic acid , resuspended in SDS-PAGE loading buffer , run into a 4–20% gradient SDS-polyacrylamide gel , and then sent to the Taplin Mass Spectrometry Facility where the gel was cut into pieces and subjected to an in-gel trypsin digestion procedure . Peptides were extracted from the gel , dried in a speed-vac , and reconstituted in 5–10 µl of HPLC solvent A ( 2 . 5% acetonitrile , 0 . 1% formic acid ) . Each sample was loaded via a Famos auto sampler ( LC Packings ) onto a nano-scale reverse-phase HPLC capillary . Eluted peptides were subjected to electrospray ionization and then entered into an LTQ linear ion-trap mass spectrometer ( ThermoFisher ) . Peptide sequences were determined by matching protein databases with the acquired fragmentation pattern by the software program , Sequest ( ThermoFisher ) . The ORFs of interest were amplified by PCR using primers including the start and stop codons of each gene of interest . For cloning into pBAD-Topo , PCR products were inserted into the expression vector according to the manufacturer's protocol ( Invitrogen ) . For cloning into pFLAG-CTC , either NdeI and KpnI or NdeI and EcoRI restriction sites were included in the PCR primer pairs . The PCR products were then digested and ligated into the expression vector . The ligation products were transformed into E . coli TOP10 competent cells and selected on LB agar plates supplemented with ampicillin ( 100 µg/ml ) . The presence of the correct insert was confirmed by colony PCR and sequence analysis . Confirmed plasmids were electroporated into V . cholerae . V . cholerae strains harbouring a pBAD-Topo plasmid were grown in 0 . 02% arabinose C-terminal fragments of bap1 and rbmA were amplified from the pFLAG-bap1 and pFLAG-rbmA plasmids , respectively , by the polymerase chain reaction with the following primers: Bap1 A: ATCGTCTAGAGTGTACGCGGGTTACTACGC and B: GACTGCATGCCAGACCGCTTCTGCGTTCTG and RbmA: A: AGTCTCTAGAGCCAGTGATTGAAGCAAATC and B: GACTGCATGCCAGACCGCTTCTGCGTTCTG . The resulting PCR products were digested with XbaI and SphI and ligated into the multiple cloning site of the suicide plasmid pGP704 , and the sequence was confirmed . This plasmid was then integrated into the chromosome by single homologous recombination as previously described [37] . The ΔrbmC in-frame deletion mutant was constructed as previously described [12] . Briefly , the following primer pairs: Pair 1 A: TGGCGCCATATTCTATGACA and B: TTACGAGCGGCCGCATACACCCTTCGGCTTCATTC and Pair 2 A: TGCGGCCGCTCGTAATATTGGGCTCAACCCACTATG and B: GGCAGTTTAATGGCGATCAT were used to amplify two genome sequences spanning an in-frame deletion in the gene of interest . These DNA fragments were joined by the SOE technique [59] , cloned into pCR2 . 1-TOPO and then subcloned into the suicide vector pWM91 by ligation after digestion with XhoI and SpeI . This suicide plasmid was used to generate an in-frame deletion in rbmC by double homologous recombination [12] . A similar procedure was used for generation of the ΔrbmA in-frame deletion mutant using the following primer pairs: Pair 1 A: CGTACTCGAGCACCCACAATTAGTGATCGCT and B: TAACGAGCGGCCGCACAACCATTTGTTTTTACAACTGG and Pair 2 A: TGCGGCCGCTCGTTATAAATTTACCTAGTCACTTAGTCGT and B: TCGACACTAGTCAAACTCTAGAACGGAACAAAA . Biofilm quantification assays were performed as described previously with the following modifications [13] . Briefly , a single colony of V . cholerae was inoculated into 1 ml of LB broth and allowed to grow to mid exponential phase . The culture was then diluted in LB broth to yield an OD655 of 0 . 05 and divided into three disposable glass culture tubes ( 10 mm ×75 mm ) . These tubes were incubated statically at 27°C . After 24 hrs , planktonic cells were removed , and the OD655 of the cells was measured . Remaining biofilms were washed with PBS and then disrupted by vortexing in the presence of 1 mm beads . The OD655 of the resulting cell suspension was measured . For assays of biofilm integrity , biofilms were formed as described above and then either gently shaken or vortexed . All assays were performed in triplicate and statistical significance was determined by a student's t-test . To evaluate protein secretion , V . cholerae was inoculated into 2 mls of LB broth supplemented with IPTG and ampicillin and grown for 6 hours at 37°C with shaking at 200 rpm . The OD655 of the final culture was measured , and then the cells were centrifuged at 4°C for 15 minutes at 4500 rpm . The supernatants and cell pellets were separated . Cell pellets were resuspended in the volume of PBS required to yield a final OD655 of 1 . Five µl of this cell suspension were diluted in 20 µl 1x Laemmli buffer solution and boiled for 5 min . Supernatants were collected and filtered through a 0 . 25 µm filter . 10 µl of the supernatants were added to 2 µl 5x Laemmli buffer and boiled for 5 min . The protein mixtures in the cell pellets and supernatants were separated by electrophoresis on a 4–20 % precast SDS-PAGE gel ( Pierce ) and transferred onto a PVDF membrane ( Millipore ) with a semi-dry transfer apparatus using the Fast Semi-Dry Transfer Buffer ( Pierce ) . The affinity tagged proteins were visualized as follows . Membranes were incubated overnight in a blocking solution consisting of PBS with 0 . 05% Tween 20 ( PBS-T ) and 5% skim milk-PBS . The membranes were then incubated with a 1∶10 , 000 dilution of Anti-FLAG M2-Peroxidase antibody in PBS-T for 1 hour on a rotary shaker . Membranes were washed once for 15 minutes and twice for 5 minutes in PBS-T and then developed using the ECL Plus Western Blotting Detection Reagent ( GE Healthcare ) according to the manufacturer's instructions . To evaluate Bap1 and RbmA in the biofilm at native levels , a similar protocol was used with the following modifications: strains carrying either Bap1-FLAG or RbmA-FLAG on the chromosome were allowed to form biofilms for 24 hours in 2 ml of LB . After removal of planktonic cells and spent medium , biofilms were washed with PBS and resuspended in 500 µl of PBS . Biofilm cell extracts were prepared by sonication , and the protein concentrations of the extracts were determined by Bradford assay . 20 µg of each extract was diluted in 20 µL of MilliQ water and 5X Laemmli buffer and separated by SDS-PAGE . As a loading control , the RNA polymerase α-subunit was detected with an antibody raised against the α-subunit of E . coli RNA polymerase ( Neoclone ) . Relative amounts of Bap1 and RbmA in the gel were approximated by densitometry analysis using ImageQuant 5 . 2 ( Molecular Dynamics ) . Wells of a 12 well microtiter dish were filled with 2 mls of LB broth supplemented with ampicillin and arabinose , where noted , and a tilted 18 mm ×18 mm glass cover slip was placed in each well . After 24 hours of static culture , the cover slips were placed in 6 well microtiter dishes and washed twice for 5 minutes with 2 mls of PBS on a rotary shaker . The cover slips were then incubated on a rotary shaker for one hour in a blocking solution consisting of PBS supplemented with 3% BSA . This solution was replaced with blocking solution containing Anti-6X His ( 1∶1 , 000 dilution ) ( Abcam ) and/or Anti-FLAG M2 ( 1∶1 , 000 dilution ) ( Sigma-Aldrich ) , and the coverslips were then incubated for an additional hour . After this incubation , the cover slips were washed with PBS three times for 5 minutes each time . For labeling of FLAG-tagged proteins with DyLight549 , biofilms exposed to the unlabeled Anti-FLAG M2 antibody underwent an additional 45 min incubation with DyLight 549 AffiniPure Rabbit Anti-Mouse IgG H+L ( 1∶500 dilution ) ( Jackson ImmunoResearch ) . For His-tagged proteins , the same procedure was used with an Alexa Fluor 488 Goat Anti-Rabbit Antibody ( Invitrogen ) . The cover slips were then washed in PBS three times , for 5 minutes each time . Where indicated , the cover slips were also incubated with a 1 mg/ml DAPI solution for 5 min . Cover slips were mounted on concave glass slides filled with PBS and then sealed with nail polish . Confocal images were acquired at the Children's Hospital , Boston Imaging Core with a LSM700 microscope ( Zeiss ) equipped with a 63X objective and 405 , 488 , and 555 nm laser lines . A computer equipped with ZEN 2009 software was used to acquire and process images . As a control , a DAPI-stained biofilm was imaged before and after immunofluorescence manipulations . Very little change was observed after manipulation , demonstrating that the biofilm was not noticeably degraded by the immunofluorescence staining procedure ( data not shown ) . Wild-type V . cholerae carrying either a RbmA-FLAG or a Bap1-FLAG expression plasmid were grown overnight on an LB agar plate containing ampicillin . Several of the resulting colonies were inoculated into 100 mls of LB broth supplemented with ampicillin . When the culture reached mid-log phase , IPTG was added to a final concentration of 1 mM . After 4 hours of additional growth , the cells were pelleted at 5 , 000 rpm at 4°C ( Sorvall , rotor SLA-600TC ) , and the recovered supernatant was distributed into two 50 ml conical tubes . 200 µl of Anti-FLAG M2 Affinity Gel prepared according to the manufacturer's instructions ( Sigma-Aldrich ) was added to each tube , and the tubes were agitated for 1 hour at room temperature to allow the protein to adhere to the resin . The resin was collected in 10 ml chromatography columns ( Bio-Rad ) and washed with 2×10 ml PBS . Proteins bound to the resin were eluted with 300 µl of 0 . 1 M glycine , pH 2 . 5 and instantly brought to pH 8 by addition of 10 mM Tris , pH 8 . Protein concentration was determined by absorbance at 280 nm , and the eluate was analysed by SDS-PAGE using a 12% pre-cast gel ( Pierce ) . After separation , the gel was stained with Imperial Stain ( Pierce ) . For quantification , equal numbers of lacZ+ and lacZ− V . cholerae strains were inoculated into LB-filled wells of a microtiter dish , and biofilms were allowed to form at 27°C over 24 hours . Biofilms were then disrupted with 1 mm glass beads , and serial dilutions of the resulting cell suspensions were plated for isolation on LB agar plates containing X-GAL . In the morning , numbers of blue and white colonies were recorded . For microscopy , equal numbers of a ΔrbmCΔbap1 mutant and a ΔvpsL mutant carrying a chromosomally-encoded , constitutively expressed gfp allele and a plasmid-encoded bap1-FLAG allele were inoculated into LB-filled wells of a microtiter dish with a coverslip . Biofilms were allowed to form as described above . Biofilms formed on coverslips were subsequently removed and prepared for immunofluorescence as as described above . These biofilms were examined by confocal microscopy using the LSM700 microscope ( Zeiss ) . Cells were grown in a 24 well microtiter dish filled with minimal medium ( MM ) alone or supplemented with purified Bap1 or BSA . An Eclipse TE-2000-E phase contrast microscope ( Nikon ) equipped with a 20X objective and an Orca digital CCD camera ( Hamamatsu ) was used to obtain images . Surface area coverage was calculated using IP Lab software ( Nikon ) . Two randomly selected fields were measured in each of three biological replicates . Proteins listed in Tables 1 and 2 have the following Swiss Prot accession numbers . Table 1: MshA ( Q60074 ) , RbmA ( Q9KTH4 ) , RbmC ( Q9KTH2 ) , FlaB ( P0C6C4 ) , FlaD ( P0C6C6 ) , FlaC ( P0C6C5 ) , FlaA ( P0C6C3 ) , ChiA-2 ( Q9KND8 ) , HlyA ( P09545 ) , and HAP ( P24153 ) . Table 2: VC0174 ( Q9KVH2 ) , VC0430 ( Q9KUT5 ) , VC0483 ( Q9KUN2 ) , VC1101 ( Q9KT04 ) , VC1154 ( Q9KSV2 ) , VC1334 ( Q9KSC4 ) , VC1384 ( Q9KS75 ) , VC1523 ( Q9KRW1 ) , VC1834 ( Q9KR13 ) , VC1853 ( Q9KQZ4 ) , VC1887 ( Q9KQW1 ) , VC1894 ( Q9KQV4 ) , VC2168 ( Q9KQ36 ) , VC2517 ( Q9KP59 ) , VCA0026 ( Y2826 ) , VCA0058 ( Q9KNA7 ) , VCA0144 ( Q9KN22 ) , and VCA0900 ( Q9KL48 ) . | The bacterial multilayer biofilm consists of matrix-enclosed cells attached to each other to form large aggregates . The base of these aggregates may be attached to a living or non-living surface . The biofilm matrix most often contains at least one exopolysaccharide component and may also contain protein and DNA . While much is known about the exopolysaccharide component of the Gram-negative biofilm matrix , little is known about the function of biofilm matrix proteins . We hypothesized that the biofilm matrix might harbor proteins with diverse functions . Therefore , we undertook the first proteomic analysis of the biofilm matrix of a Gram-negative bacterium , V . cholerae . We subsequently focused on Bap1 and RbmA , two proteins that are abundant in the biofilm matrix . RbmA , which strengthens intercellular interactions , was found to be evenly distributed in the biofilm . In contrast , communally synthesized Bap1 was concentrated at the biofilm-surface interface and stabilized the association of the multilayer biofilm with the surface . Furthermore , the addition of purified Bap1 increased attachment of free-swimming cells to a surface . These studies provide evidence for spatial and functional differentiation of proteins in the biofilm matrix and suggest bacterial cooperation in stabilization of multilayer biofilm surface association and recruitment of new members . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"bacterial",
"biofilms",
"gram",
"negative",
"microbial",
"mutation",
"ecology",
"coastal",
"ecology",
"microbial",
"pathogens",
"marine",
"and",
"aquatic",
"sciences",
"biology",
"microbiology",
"marine",
"biology",
"bacterial",
"pathogens",
"microbial",
"ecology"
] | 2011 | A Communal Bacterial Adhesin Anchors Biofilm and Bystander Cells to Surfaces |
Social animals may share information to obtain a more complete and accurate picture of their surroundings . However , physical constraints on communication limit the flow of information between interacting individuals in a way that can cause an accumulation of errors and deteriorated collective behaviors . Here , we theoretically study a general model of information sharing within animal groups . We take an algorithmic perspective to identify efficient communication schemes that are , nevertheless , economic in terms of communication , memory and individual internal computation . We present a simple and natural algorithm in which each agent compresses all information it has gathered into a single parameter that represents its confidence in its behavior . Confidence is communicated between agents by means of active signaling . We motivate this model by novel and existing empirical evidences for confidence sharing in animal groups . We rigorously show that this algorithm competes extremely well with the best possible algorithm that operates without any computational constraints . We also show that this algorithm is minimal , in the sense that further reduction in communication may significantly reduce performances . Our proofs rely on the Cramér-Rao bound and on our definition of a Fisher Channel Capacity . We use these concepts to quantify information flows within the group which are then used to obtain lower bounds on collective performance . The abstract nature of our model makes it rigorously solvable and its conclusions highly general . Indeed , our results suggest confidence sharing as a central notion in the context of animal communication .
Animals living in groups sense their surroundings both directly , by environmental cues , and indirectly , through countless social interactions . There is an abundance of experimental evidence for the usefulness of social information in increasing both the range ( the “many eyes” principle ) [1]–[5] and the accuracy ( the “many wrongs” principle ) [6]–[9] at which environmental signals are perceived . Despite these advantages , there are many scenarios in which animals tend to prefer personal knowledge and direct environmental cues to social information [10] , [11] . Indeed , second hand information about the environment can become increasingly obsolete [12] , [13] , distorted [14] , and partial [15] as it passes from one individual to the next and , subsequently , lead to maladaptive responses [11] . These contradicting evidences call for a more comprehensive understanding of the usefulness of social information exchange and its limitations under noise . A distinction can be made between passive and active social messaging [16] . Passive information [17] , [18] is transferred as inadvertent cues [19] , i . e . , with no direct intention of signaling , evident by the behavior of one animal are perceived by others . As an example , models of complex flocking behaviors typically rely exclusively on passive interactions in which animals align their movements to those performed by their neighbors [6] , [20] . However , there is evidence that passive information is often accompanied by active , or intentional , signals that communicate part of the animal's internal state . In cooperative situations [21] active signals may enhance the effectiveness of passive cues and lead to faster and more accurate performance [13] , [14] . While elaborate active communication has its advantages , simplicity is , nonetheless , important . Indeed , it is required that communication remain energetically cheap [22] , cognitively manageable [23] , [24] and concise [21] . A main goal of this work is to identify simple active communication schemes that enhance the reliability and the benefits of social information . Animal groups , together with numerous other biological ensembles , are naturally described as entities that collect , share and process information . Unfortunately , with the exception of neuroscience [25] , the potential of information theory in providing rigorous descriptions of such ensembles remains , largely , unrealized [15] . For example , the term “information flow” is often used to describe the gradual process in which messages are being relayed between agents [26] , [27] . Although the speeds and directionality of information flows have been described for several systems [1] , [28]–[30] , it remains unclear how to rigorously analyze such flows to quantify the amount of transferred information . A second goal of this paper lies in introducing information theoretical tools as a means of quantifying information flows within a group of agents . In what follows , we use an algorithmic perspective [31]–[33] to tackle the question of information sharing within a population of cooperative agents . The agents use environmental cues intertwined with social interactions to obtain ever refined estimates of some fixed , unknown environmental target value [34] . Interactions include both passive and active components . A passive observation amounts to obtaining a noisy measurement of the observed agent's behavior . An active signal exposes some part of the observed agent's internal state . We are interested in how active signals may be economically used to best enhance the flow and benefits of passive communication . To study this question we compare two models . The non-restrictive model allows for infinite resources in terms of memory , active communication and individual computation . On the other hand , the compact model restricts active communication and memory to a single parameter and individual computation to a constant number of the basic arithmetic operations . We present recent experimental observations [14] , [35]–[37] as well as novel evidence regarding ant interactions that suggest that the communication of a self-confidence parameter is a relevant process within animal populations . Inspired by such observations , we propose a simple and natural algorithm for the compact model that relies on the sharing of confidence . This model can serve as a basic representative of the family of confidence-sharing algorithms . We show that the performances of this algorithm are highly competitive with those of the best possible algorithm for the non-restrictive case . One may be tempted to reduce active communication below what is permitted by the compact model , but we show that this may incur a heavy price in performance .
To evaluate the performances of algorithms , we compare them to Opt ( see section 2 in Text S1 ) , the best possible algorithm operating under the non-restrictive model . Being as liberal as possible , we further assume that active communication is completely reliable . This is since any definition of active noise must depend on a particular choice of a communication scheme which , in turn , may restrict an optimal algorithm . Moreover , here , agents are initially provided not only with the variances of the noise and initial distributions but also with their full functional forms . That is , the memory of an agent initially contains and . Without loss of generality , the memory of an agent further includes a vector that contains all prior moves and distance measurements it took . Following an interaction , the observing agent adds to its memory not only the new noisy distance measurement but also the full memory content of the observed agent . This leads to the accumulation of large nested data-structures . The agent may then perform arbitrarily sophisticated computations over its memory to adjust its location to its best possible estimate of . We stress that none of the proofs in this manuscript rely on the identification of an optimal algorithm . Nevertheless , for the sake of completeness , we specify Opt for independent meeting patterns ( section 1 . 1 . 3 in Text S1 ) , which are especially meaningful on short timescales or if the system is highly mixed . Indeed , in such cases , algorithm Opt can be precisely described ( section 2 . 2 in Text S1 ) . Specifically , each agent maintains a that represents the relative positioning of the target value with respect to its current location . The pdf is initialized to be . Upon observing another agent at time , agent performs the following operations: Algorithm Opt In general , as time passes , the description of the stored requires an increasing number of moments and its communication a more elaborate encoding scheme . Moreover , the calculations required for updates become increasingly complex . Algorithm Opt relies on the transmission and updates of probability functions and on relatively complex calculations . We wish to identify a simple algorithm whose performance is highly competitive with that of Opt . To do this one faces several difficulties . A first difficulty lies in the fact that the partial knowledge held by each agent is relative ( e . g . , an estimation to the distance between this agent and ) and hence may require the agents to carefully fuse other perspectives than their own . This difficulty is enhanced , as the agents are constantly on the move . We have shown how non-restrictive algorithms may overcome such difficulties if each agent encodes all its previous moves in memory and then uses this information to deduce absolute measurements ( section 2 . 1 in Text S1 ) . In compact models , such tactics lose their effectiveness and it is not clear how agent should treat distance measurements to an agent whose position constantly changes over time . It is known that a reasonable way to combine estimators is to form linear combinations in which each estimator is weighed by its inverse variance [43] . Although this is the best estimator that could be formed as a linear combination it is not overall optimal . Indeed , maintaining and communicating highly detailed memories can , in some cases , significantly improve an agent's assessment of the target value ( for example , see Figure 1 ) . This problem worsens in the context of an interacting population . Here , maintaining a high degree of detail requires storing an arbitrary number of moments which may grow with every interaction . Discarding this accumulating information by repeatedly using simple ( e . g . linear ) estimators could , therefore , lead to performances that deteriorate with time . Hence , it is not clear how to compress the information held by agents into few meaningful parameters while avoiding the accumulation of errors and runaway behavior . Another of the analysis difficulties corresponds to the fact that the held by an agent at time depends on many previous deviation measurements in a non-trivial way , and hence the variance of a realization of the does not necessarily correspond to the variance of the agents' opinion , when taking into account all possible realizations of all measurements . Hence , one must regard each as a multi-variable distribution . A further problem has to do with dependencies . The independent meeting pattern guarantees that the memory 's of two interacting agents are independent , yet , given the of the observing agent , the of the observed agent and the deviation measurement become dependent . Such dependencies make it difficult to track the evolution of an agent's accuracy of estimation over time . Indeed , to tackle this issue , we had to extend the Fisher information inequality [44] , [45] to a multi-variable dependent convolution case . Internal representations of confidence have been shown to affect animal behavior over a wide range of species [46]–[49] . Confidence as an internal parameter that builds up as a passive agent gathers external evidence has been measured in pre-decision neuronal responses ( see , for example , [50] ) . The notion of confidence as an internal parameter carries over into group contexts wherein animals were demonstrated to become more responsive to social information as their own certainty drops [37] , [51] , [52] . Furthermore , evidence also suggests that animals are capable of communicating their confidence as well as assessing that of their conspecifics [13] , [14] , [35] , [53] . One such example comes in the context of conflict , where threat behaviors may indicate the communication of confidence . While no single work directly binds all elements of confidence sharing many supportive evidences exist: Dominance hierarchies , like confidence , are constructed according to the accumulation of evidence [54] . Further , threats are correlated with large differences in dominance rank [55] and are often non-deceptive [56]–[58] and convey the animal's actual chances of winning the next fight . Moreover , threats are generated and perceived at different levels of intensity [55] , [59] to the extent of causing an opponent to back away from confrontation [53] , [60] . Other examples come from more cooperative scenarios such as house hunting behavior in honeybees ( Apis mellifera ) . It was shown that swarming bees collectively move towards a new nest site by communicating two-component messages: The direction in which bees fly encodes the desired direction towards the new site while the speed of flight determines the degree of responsiveness this message will elicit in others [61] , [62] . Furthermore , it was shown that high speed is associated with bees that have been to the new site ( streakers ) as well as bees that do not have first hand accounts but whose flight is generally directed towards the desired site [61] . These evidences are consistent with an analogy between flight speed and confidence regarding the correct direction to the new site . Another example occurs earlier in the house-hunting process . The messages which scouts convey regarding the location of prospect nest sites contain ( at least ) two components: While the direction to the advertised site is encoded by the waggle dance , the intensity of the message is encoded in the number of times the bee performs this dance [63] , [64] . The intensity of the message correlates with the quality of the advertised site and could be interpreted as the confidence of the bee that the site she advertises is the best of all options . This interpretation is strengthened if , similar to what has been shown for ants [65] , [66] , bees have some internal scale to the quality of a site . A further example for the role of confidence during interactions comes from recruitment behavior in the desert ant Cataglyphis niger [14] . Here , ants within the nest interact with their nest-mates to accumulate indirect evidence regarding the presence of food and towards an active decision to exit themselves ( recruitment ) . Similar to the accumulation of neuronal activity that proceeds a decision [50] , ants were observed to gradually increase their speed of movement before deciding to exit the nest [14] . Furthermore , ants which have been in direct contact with the food are certain of its presence and indeed maintain high speeds for extended periods of time [14] . These evidences suggest that an analogy between the speed of an ant and her confidence may be useful . In Figure 2 we present novel empirical evidence of the way ants update their speed following an interaction . This data confirms that speed ( confidence under this analogy ) is both transmitted and perceived by the ants . Moreover , the speed of an ant after the interaction is an increasing function of both her speed and the speed of her interacting partner prior to the interaction . Having identified confidence sharing as a relevant communication scheme in animal groups , we turn to study the compact algorithm Conf: a basic representative of the family of algorithms that rely on the active communication of confidence . This algorithm is basic in being both simple and natural: It is simple as it is highly economical in terms of communication , memory usage and internal computations . It is natural since it relies on linear combination information fusing techniques . Below , we describe Conf and show that it displays near optimal performance . In algorithm Conf each agent , , stores in its memory a single parameter that represents its confidence regarding its current distance from the target . The initial confidence of agent is set to . When agent observes agent , it receives both the passive noisy distance measurement and an active message containing the confidence parameter of . This information will then allow agent to relocate itself by using a weighted average procedure [34] , [43] . Then , a suitable update is made for to reflect 's confidence of its updated location . Specifically , upon receiving and , agent proceeds as follows: Algorithm Conf We provide a rigorous proof ( section 5 in Text S1 ) that the performances of Conf are very close to those of Opt when the meeting patterns are independent and active communication is noiseless . Specifically , we first show ( section 5 . 1 in Text S1 ) that under these conditions , the rules of Conf guarantee that the location of any agent serves as an unbiased estimator of and that the confidence satisfies: ( 1 ) We further show ( section 5 . 2 in Text S1 ) that although approximation errors that result from the information compression of Conf are inevitable , they do not accumulate with time and through repeated interactions . Indeed , the quotient between the variance of the population under Conf and its variance under Opt remains bounded , at all times , by the initial Fisher-deviation ( as defined in the Materials and Methods ) . More specifically , under algorithm Conf , the variance of any agent at time is bounded by times the corresponding variance under Opt ( see Figure 3A ) : ( 2 ) where denotes the location of agent at time under algorithm Opt . To prove Equation 2 , we relate the variance of Opt to a measure of information which we call the relative Fisher information . This measure , denoted ( formally defined in Text S1 , section 3 . 2 ) , quantifies the agent's current knowledge regarding . Intuitively speaking , this notion can be thought of as the Fisher information of the family that describes the random samples held by under algorithm Opt with respect to the translational parameter ( see Materials and Methods ) . We then use the Cramér-Rao bound to deduce that: ( 3 ) where the mean is taken over all possible random initial locations and communication noises , as well as , possibly , over all random choices made by the agents themselves . We then show that the confidence of an agent under algorithm Conf satisfies: which establishes Equation 2 and proves that the competitiveness of Conf with respect to Opt is , at most , the initial Fisher deviation . Note that for , the optimal algorithm Opt cannot , in fact , achieve the Cramér-Rao bound at all times ( t = 0 being a trivial example ) . Therefore the competitiveness of Conf with respect to Opt can be expected to be even tighter than . This is indeed verified by simulation ( see Figure 3A ) . Moreover , we show that in the case of Gaussian noise , and regardless of , the performance of Conf will approach that of Opt at large times ( section 5 . 2 . 3 in Text S1 and Figure 3A ) . Note that in the case in which the noise and initial distributions are all Gaussian , the Fisher deviation satisfies so that Conf is optimal ( Figure 3A ) . We next compare the Conf algorithm to even simpler algorithms that rely solely on passive communication . We first consider algorithms in which the interaction update rule is a simple linear combination of the observing agent's location , and the estimated location of the observed agent: for some constant ( note that in algorithm Conf , is not constant and is set according to the active message and 's current confidence ) . A simple average algorithm is obtained by setting . The performance of constant linear combination algorithms is of interest since they require minimal resources: agents are not required to store any memory of their current internal state . We find that , in general , when communication noise is substantial , linear combination algorithms do not perform well . They exhibit a speed accuracy tradeoff converging within a time scale of ( which diverges for small values of ) to a steady state with a variance that scales as ( section 7 in Text S1 and Figure 3B ) . On the other hand , in the case of uniformly informed populations and negligible communication noise , the performances of the simple average algorithm ( ) approach those of Conf in terms of both convergence rate and steady state variance . Intuitively , simple averaging functions well under these circumstances since the information held by two interacting agents , at any time , is of equal quality . Active communication can also become redundant when passive communication noise is very large with respect to the uncertainty of the agents . Indeed , in this case , a “passive” algorithm is obtained by translating the rules Conf into a high noise regime . The effective confidence of any observed agent becomes which is independent of its actual internal state . Conversely , “passive” algorithms are expected to fail in situations where noise levels are comparable to agent uncertainty and knowledge is non-uniformity distributed among the agents . In this case , the assumption that an observed agent's confidence is fails and could lead to irreparable mistakes in the observing agent's position and confidence after the interaction . For intuition , consider the extreme case in which a single agent has very accurate knowledge of the target value while all other agents have no information at all . In this case , Conf would allow for very fast convergence typical of rumor spread: roughly within rounds , where is the number of agents . On the other hand , if no active communication is allowed , it becomes difficult to distinguish the knowledgeable agent within a large population of anonymous agents ( see section 7 . 1 in Text S1 ) .
In this work we theoretically studied an abstract model of animal communication within a group which generalizes the work of McNamara and Houston [34] . Similar to their approach , we considered a basic model which enabled us to perform rigorous analysis , often impossible in more complex scenarios . We have shown that weighted averaging algorithms , previously known to be efficient for fusing multiple pieces of evidence [43] , naturally carry over to a scenario in which a group of agents share and aggregate information . The weights used may be interpreted as the agents' confidence in their opinion . We have theoretically shown , that , remembering and actively communicating confidence is , in fact , sufficient for near-optimal decisions in cooperative group contexts . Using the confidence measure is straightforward: individuals with high confidence are more persuasive while those with low confidence more fickle . Finally , the fundamental nature of our model makes our results potentially relevant to a large number of natural systems . We have used the framework of Fisher information to study information flows within cooperative groups . In particular , we have defined the Fisher Channel Capacity and demonstrated how it bounds collective reaction times . This opens the door for further rigorous quantifications of information flows within animal groups . We introduced Conf , a simple weighted-average based algorithm that uses compact memory and communication in a way that overcomes the anticipated shortcomings of information compression ( e . g . , see Figure 1 ) . We have shown that Conf is highly competitive when compared to an optimal algorithm that utilizes maximal memory , communication capacity , and computational resources . In fact , we bound the difference in performance by a constant factor - the initial Fisher-deviation . We have presented evidence that supports the relevance of Conf to actual biological groups and turn to suggest how this may be helpful for analyzing experimental data . A most intriguing result would be to utilize Equation 7 to obtain a lower bound on communication noise levels . Indeed , Equation 7 holds with respect to any algorithm operating in the corresponding setting , and with respect to any level of noise in active communication . If the setting is matched in an experiment , the initial variance is large , and the convergence time fast , Equation 7 would yield a lower bound on , the Fisher information in the noise corresponding to the passive communication . Such a result would demonstrate the usefulness of the indirect methodology , based on algorithmic lower bounds as suggested in [71] . Moreover , such a lower bound on the amount of noise seems to be difficult to obtain by other , more direct , methodologies . Further practical implications of our results include the identification of scenarios in which active communication is likely to be employed . These include cases in which the noise level is intermediate and situations of populations that are variable in terms of initial knowledge as is the case in effective leadership scenarios [36] , [38] . In such cases , our results suggest that it may be useful to search for the active transmission of “confidence” signals , which can be encoded e . g . , in the speed of agents [14] , [61] . Our analysis for the performances of Conf assumes independent meeting patterns . Such patterns are especially meaningful when agents rely on few interactions each , or when the system is highly mixed . We have used simulation to demonstrate that algorithm Conf continues to perform well for small groups in which interaction patterns are no longer independent . In addition , our simulations show that Conf is robust under active communication noise , heterogenic populations , and that simple extensions of this algorithm may be expected to perform well in dynamic environments . It is interesting to identify those scenarios in which active communication appears to be of lesser importance . When personal information is reliable and frequently updated there is , trivially , no requirement for any sort of communication . It is when personal information is less accurate that social information becomes useful . We have shown that simple averaging algorithms ( operating without long term memory ) behave well in uniform populations with communication noise that is negligible in comparison to the desired convergence state . We further showed that when communication noise is very large then an algorithm in which each agent maintains an internal confidence measure but does not communicate it [38] , [72] performs extremely well . This implies that in such cases , the system can perform well without resorting to active communication . Although our results were formulated in the language of animal group behavior they can readily be generalized to a large range of cooperative biological ensembles . For example , bacterial quorum sensing is mediated by both passive cues ( e . g . one cell senses another's waste products ) and active signaling mediated by designated quorum-sensing molecules [73] .
We consider parameterized probability density function ( ) families where is the functional form and is a translation parameter [45] . The Fisher information of a family is defined as: where denotes all variables on which depends . Note , that since is a translational parameter , the Fisher information is both unique ( there is no freedom in choosing the parametrization ) and independent of [45] . The Cramér-Rao inequality sets a lower bound on the variance of any unbiased estimator , based on a random sample taken from , for the parameter : To define the initial Fisher-deviation , denoted , we first define the Fisher-deviation of a distribution as Note that , by the Cramér-Rao bound , for any unbiased distribution . The initial Fisher-deviation is the supremum of the Fisher-deviations over all the ( unbiased ) distributions involved , namely , the distributions governing the initial locations and the noise distribution . Specifically , let and finally define Observe that if the distributions and are all Gaussians then . | Cooperative groups are abundant on all scales of the biological world . Despite much empirical evidence on a wide variety of natural communication schemes , there is still a growing need for rigorous tools to quantify and understand the information flows involved . Here , we borrow techniques from information theory and theoretical distributed computing to study information sharing within animal groups . We consider a group of individuals that integrate personal and social information to obtain improved knowledge of their surroundings . We rigorously show that communication between such individuals can be compressed into simple messages that contain an opinion and a corresponding confidence parameter . While this algorithm is extremely efficient , further reduction in communication capacity may greatly hamper collective performances . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"collective",
"animal",
"behavior",
"computer",
"and",
"information",
"sciences",
"animal",
"behavior",
"zoology",
"cognition",
"population",
"modeling",
"network",
"analysis",
"social",
"networks",
"information",
"theory",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"cognitive",
"science",
"neuroscience",
"animal",
"cognition",
"animal",
"signaling",
"and",
"communication"
] | 2014 | Confidence Sharing: An Economic Strategy for Efficient Information Flows in Animal Groups |
In many immunological processes chemoattraction is thought to play a role in guiding cells to their sites of action . However , based on in vivo two-photon microscopy experiments in the absence of cognate antigen , T cell migration in lymph nodes ( LNs ) has been roughly described as a random walk . Although it has been shown that dendritic cells ( DCs ) carrying cognate antigen in some circumstances attract T cells chemotactically , it is currently still unclear whether chemoattraction of T cells towards DCs helps or hampers scanning . Chemoattraction towards DCs could on the one hand help T cells to rapidly find DCs . On the other hand , it could be deleterious if DCs become shielded by a multitude of attracted yet non-specific T cells . Results from a recent simulation study suggested that the deleterious effect dominates . We re-addressed the question whether T cell chemoattraction towards DCs is expected to promote or hamper the detection of rare antigens using the Cellular Potts Model , a formalism that allows for dynamic , flexible cellular shapes and cell migration . Our simulations show that chemoattraction of T cells enhances the DC scanning efficiency , leading to an increased probability that rare antigen-specific T cells find DCs carrying cognate antigen . Desensitization of T cells after contact with a DC further improves the scanning efficiency , yielding an almost threefold enhancement compared to random migration . Moreover , the chemotaxis-driven migration still roughly appears as a random walk , hence fine-tuned analysis of cell tracks will be required to detect chemotaxis within microscopy data .
Upon maturation , T lymphocytes continuously circulate in the blood and secondary lymphoid organs such as LNs and spleen . When they encounter dendritic cells ( DCs ) that present cognate antigen , the T cells become activated and subsequently start to proliferate . Before such an immune response is mounted , the fraction of T cells specific for any antigen is about – [1] . Because LNs are packed with T cells that have irrelevant specificities , it seems a challenge to establish a contact between a specific T cell and a DC carrying cognate antigen . Over the last decade , two-photon microscopy ( 2 PM ) experiments applied to living lymphoid tissues have offered a wealth of insight in T cell migration characteristics and T cell-DC interactions in LNs [2] , [3] . In the absence of cognate antigen , T cells move at high speeds in an approximately constant direction for up to several minutes , whereas in the long run their migration pattern roughly resembles a random walk [4]–[8] . During their journey through the LN , T cells engage in brief contacts with DCs , lasting a few minutes on average [5] , [7] , [8] . DCs migrate much more slowly than T cells , and continuously extend and retract long , thin dendrites , thereby greatly increasing the LN volume that they are able to scan [8] . T cell behaviour changes in the presence of activated DCs presenting cognate antigen: After an initial phase of rapid migration and brief contacts , similar to their behaviour in the absence of cognate antigen , T cells form stable contacts with DCs lasting several hours [7]–[9] . Subsequently , T cells resume migration , exhibit signs of activation and start proliferating [8] . In the presence of cognate antigen , it has been shown that ‘licensing’ of DCs by either T cells [10] , T cells [11] or NKT cells [12] increases their ability to recruit naive T cells [13] , and this is mediated by chemoattractant ligands produced by the licenced DCs [12] , [14] . In contrast , 2 PM experiments showed that T cell migration patterns in vivo resemble a persistent random walk , suggesting that the migration process does not involve chemotaxis or that chemotaxis only plays a marginal role . Furthermore , it was proposed that the random walk would represent an optimal search strategy [6] , [15] . The alternative strategy of chemoattraction of T cells towards DCs was thought to be counterproductive , because nonspecific T cells would also be attracted and subsequently block the DC from scanning other T cells [15] . However , this notion is in conflict with the fact that chemoattraction has been observed in vivo , at least when cognate antigen is present , and that such chemoattraction promotes effective cytolytic as well as CD8 T cell memory responses [12] , [14] , together suggesting an important functional role for chemotaxis . The question whether chemoattraction is expected to help or hamper scanning of T cells by DCs was further addressed by Riggs et al . [16] using a theoretical framework , i . e . a 2D agent based model of the LN T-cell zone , in which T cells could either migrate in a random fashion , or in addition react chemotactically to a local chemokine gradient around DCs . In those simulations , the presence of chemoattraction led to a reduction of the number of unique T cells contacted per DC and therefore to a less efficient immune response , supporting the view that chemotaxis towards DCs is detrimental . However , this result need not be representative because the model formalism that the authors chose may lead to unrealistic blocking of cell migration in crowded lattices [17] . To alleviate this problem , the authors performed simulations at low cell densities [16] . Nevertheless , when multiple T cells are attracted to the same location via chemotaxis , cell densities become locally high and blocking of cell migration may again arise . Therefore , the model may have generated an answer to the issue that is biologically non-reasonable . Moreover , it has been pointed out that the typical analysis of T cell migration used in the 2 PM studies , i . e . , deriving the type of migration process from a mean ( square ) displacement plot , is insufficient to distinguish between a random walk and migration amongst several local sources of chemoattractant , as would be the case when multiple DCs in the LN are producing chemoattractant [2] . Taken together , it cannot be excluded yet that chemotaxis enhances the likelihood of establishing interactions between T cells and DCs . Here , we therefore readdress the question of the expected impact of chemoattraction on T cell scanning by DCs , using the Cellular Potts Model ( CPM ) . We have opted for the CPM because it allows for a mesoscopic description of cell shape , cell migration , chemotaxis and cellular interactions within complex tissue environments [18] . The CPM is a spatial grid-based model formalism that has initially been developed to describe the biophysics of cell sorting due to differential adhesion [19] , [20] . Within the formalism , cell motion comes about from the overall minimisation of the energy of deformation and stretching of the cell membrane through stochastic fluctuations , in which the global and local forces upon a cell edge are resolved [21] . For single cells and small tissues , extensions have been made to describe the detailed biophysics and regulation of cortical tension [22] , tissue deformation [23] , and cell migration and chemotaxis [24] , [25] . In contrast , to capture cell migration and chemotaxis within the dynamics of larger and more intricate cell populations , more phenomenological descriptions of those processes have been developed , in which the detailed biophysics were replaced by effective forces along the membrane ( for an overview , see [21] ) . Previously we have shown that such a more phenomenological description of cell migration can be used to realistically and quantitatively capture T cell and DC dynamics within a densely packed LN , with approximately persistent motion on short timescales and random motion on long timescales [26] , [27] . We here extend this existing framework with a frequently used CPM extension for chemotaxis [18] , [28]–[30] to describe the chemotactic response of T cells to chemokines produced by DCs . Besides the impact of chemoattraction , we here investigate the potential role of T-cell desensitization to the chemoattractant in scanning efficiency . We show that chemoattraction of T cells towards DCs increases the T cell scanning efficiency and thus the probability of T cells to find a rare DC carrying cognate antigen .
We performed most of our simulations using a 2D model of a part of the LN T cell zone around the high endothelial venules ( HEVs ) through which T cells enter the lymph node , and where T cells and DCs come into contact with each other . Most simulations were done in 2D to be able to more directly compare our results to those of the model by Riggs et al . [16] , which was also simulated on a 2D lattice . However , we have also performed 3D simulations to confirm our 2D results within a more realistic spatial setting , and to test whether dimensionality plays a role in the relationship between chemotaxis and search efficiency . Our model contained in silico T cells ( blue and yellow in Figure 1A ) , DCs ( red ) , reticular network ( green ) and the capsule ( cyan ) . The latter two elements were included to capture a realistic LN structure . We modelled entry into the LN by introducing new T cells at random positions in the indicated region in Figure 1A , and exit by allowing T cells to leave the simulations at the bottom . We kept the tissue densely packed with cells , to realistically mimic the situation in LNs . Entry and exit of T cells were balanced such that the number of T cells in the simulation was kept constant , as is approximately the case in vivo [31] . In a LN of , there are roughly one million T cells . Translating this to the volume of our 3D simulations , this would amount to approximately 1000 cells , significantly lower than the 5000 cells used in our simulations . However , because T cell zones largely consist of T cells , the true density of T cells in these zones is closer to our simulated densities than reflected by the average density over the whole LN . By default , our in silico T cells performed a persistent random walk and parameters were tuned such that T cells moved with realistic speeds and motility coefficients [26] ( see the section on T cell migration without chemotaxis ) . In contrast to the T cells , DCs were kept in our simulations at predetermined and more or less fixed mean positions and were distributed approximately evenly throughout the space , as observed in vivo [32] . Although some DCs are migratory and carry antigen to the LN , they transfer their antigen to static , lymphoid resident DCs which then activate T cells [33] . Furthermore , it is thought that most DCs die within the LN , and consistent with this , only 0 . 1% of cells leaving via efferent lymphatic vessels is estimated to be DC . Therefore , we did not include entry and exit of DCs in our model . The simulated DCs continuously extended and retracted dendrites from their centre of mass ( as described previously [27] ) , giving them a large surface area to be able to contact T cells . DCs could also produce a chemoattractant , which subsequently diffused through the tissue and decayed . The combination of multiple chemokine sources in the field , together with diffusion and decay , generated a complex concentration profile in which the steepness and orientation of the chemokine gradient was highly variable in space . T cells in our simulations could be in two different states , either sensitive or insensitive to the chemokine gradient created by the DCs . We hypothesized that a natural point in time for T cells to become desensitized would be the moment they establish an interaction with a DC ( referred to as the ‘DC-contact’ mechanism ) . Such a desensitization potentially allows the T cell to leave and other cells to approach the DC . After a recovery period , the length of which we varied , T cells resensitized to the chemokine gradient . To determine the importance of desensitization , we also ran simulations in which T cells did not become insensitive at all ( the ‘no-desensitization’ mechanism ) . Furthermore , we varied the strength of the reaction of T cells to the chemokine gradient ( see Methods ) . To do so , we varied a T cell parameter ( ) which determined the amplification of the chemokine gradient signal ( i . e . , effectively corresponds to random migration , while is the strongest chemotactic response we used ) . Without any obstacles or other cells present , an in silico T cell migrates on a short timescale of a few minutes in a more or less straight direction , but in the long run it migrates randomly . When cell density is high , the T cells in the simulation self-organize into large streams of coherently migrating cells , because colliding cells force each other to move into the same direction , and the same holds for streams that bump into each other , until one is left with a single , global stream of cells [21] . Furthermore , our previous simulations predicted that the obstacles formed by RN , DCs and other T cells prevent the formation of such a global stream of T cells , instead triggering the formation of many small , dynamically changing streams , and we confirmed that such microstreams indeed occur in vivo by a detailed analysis of 2 PM imaging data [26] . In this study we observe again the formation of such microstreams ( Video S1 ) . Moreover , the combination of cell entry at random positions within the tissue and cell exit at the bottom causes a slow overall downward cell flux on top of the chaotic microstreams ( see Videos S1 and S2 ) . The downward flux is sufficiently small to not cause any visible bias in the cells' trajectories over short time intervals ( of 10 min , see Fig . 1B ) . We first set out to examine the effect of chemoattraction on T cell migration without distinguishing between antigen-bearing and non-antigen bearing DCs , or between specific and nonspecific T cells . Thus , all DCs in the simulation produced chemoattractant and all T cells responded in the same manner . This is similar to the simulations by Riggs et al . [16] , because in those simulations all DCs entering the lattice were quite quickly licensed by T cells and produced chemokine . Since our first aim was to study whether the negative effects of chemoattraction that were reported by Riggs et al . , remain valid in a system where cells can deform , align and squeeze between other cells , we first closely mimicked their situation . Note that , because we did not explicitly model antigen recognition , these simulations would be equivalent to DCs producing chemokine regardless of the presence of cognate antigen . We compared the migration behaviour in simulations without chemoattraction ( , video S1 ) to simulations with strong chemoattraction ( , video S2 ) , both for the no-desensitization mechanism and for the DC-contact desensitization mechanism with a fixed recovery time ( which was set to 15 min ) . The mean speed of T cells migrating without chemotactic cues was tuned to about 11 . 0 , i . e . , close to the typical speeds in 2 PM experiments [4] , [6]–[8] , [34] . The mean speed ( standard deviation within a simulation ) slightly increased to when T cells migrated chemotactically with the no-desensitization mechanism , and to a very similar with the DC-contact mechanism . As expected , overlays of normalized cell tracks suggested that in our simulations there was no apparent preferred direction of migration for either randomly migrating cells or chemotactically moving cells ( Fig . 1B ) . The mean square displacement plots varied slightly between the three modes of migration ( Fig . 1C ) . However , the shapes of the curves were very similar , and therefore it is unlikely that chemotactic attraction of T cells towards DCs can be detected with a mean square displacement plot based upon 2 PM imaging data , confirming the suggestion of Germain et al . [2] . We next investigated whether it is possible to distinguish between random migration and chemoattraction to DCs from the simulated cell tracks . To do so , we determined the angle ( between a vector in the direction of T cell migration as measured between two consecutive time points and a vector pointing from the T cell towards the nearest DC ( Fig . 2A ) . For random migration in two dimensions , every angle is expected to occur with equal probability , but when T cells are attracted towards DCs , acute angles ( less than 90 degrees , see Fig . 2A ) are expected to occur most frequently [35] . Without chemotaxis , the overall distribution of those migration angles for in silico T cells that were less than 100 away from their nearest DC was close to uniform ( Fig . 2B , C ) . In the presence of strong chemotaxis , close to DCs we indeed more often observed acute angles compared to intermediate angles , both for the DC-contact mechanism and the no-desensitization mechanism , indicating that attraction of T cells towards DCs could be detected in our simulations . Surprisingly , when T cells migrated chemotactically , close to DCs also obtuse angles ( more than 90 degrees ) were observed more frequently than intermediate angles , suggesting that a subpopulation of the T cells was effectively repelled from the DCs . This phenomenon was actually a consequence of spatial exclusion: for every T cell that approached a DC another T cell had to make room for it by leaving the area . In a more detailed analysis we distinguished between T cells that had recently contacted a DC and the remaining T cells . ( Note that in the DC-contact method this corresponded to the insensitive and sensitive cells , respectively . ) This analysis showed that both in the simulations with and without chemotaxis , the cells which were in recent contact with a DC were the ones that were effectively being repelled , although the effect became more pronounced with chemotaxis ( Fig . 2B , C ) . In fact , cells with recent DC contact seemed to be part of a micro-stream of T cells moving away from the DC ( Video S2 ) , suggesting that the process was similar to convective flow . T cells which were at larger distances from the nearest DC were often at similar distances from other DCs as well , which could strongly influence their migratory behaviour , resulting in a drop in the migration bias towards that nearest DC . When T cells were more than 100 away from the nearest DC , not only the bias towards the nearest DC had disappeared , but T cells were even preferentially moving away from the nearest DC ( Fig . 2D ) . This reversal of migration bias was due to the slow background flow of cells from top to bottom: all cells within the simulated tissue are slowly pushed downward by each other , which explains the migration away from the nearest DC . These results show that measuring migration angles of T cells that are close to a DC may allow to distinguish between random migration and chemoattraction towards DCs using 2 PM imaging data acquired in the absence of cognate antigen . ( Note that it has already been successfully applied to show that T cells are chemotactically attracted towards licenced DCs [36] . ) We showed above that current 2 PM data are consistent with the notion of chemotactic attraction towards DCs . In light of the ongoing debate on the importance of chemoattraction ( see [2] ) , we next used our in silico environment to examine whether chemotaxis enhances T cell scanning by DCs , compared to randomly migrating T cells . This was done by varying the strength of chemoattraction of T cells towards DCs ( i . e . , , see Methods ) , and measuring the scanning rate . Focussing on unique contacts only ( i . e . , contacts established with T cells which had not been seen before by this DC ) , we found that DCs scan about 300 randomly migrating T cells per hour ( Fig . 3A , top ) . With the no-desensitization mechanism , we found a substantial , 40% increase in the number of unique T cells that contacted a DC at the highest chemotactic strength that we simulated . The total number of contacts per DC ( i . e . , when repeated T cell contacts to the same DC were counted as well ) also increased strongly ( Fig . S1 ) . Because many more cells visited each DC within the same timespan , the interactions between T cells and DCs lasted on average shorter for strong chemoattraction than for random migration ( Fig . 3B , top ) , indicating that there was fierce competition between T cells for contacting DCs . For the DC-contact mechanism we found that chemoattraction very strongly increased the number of unique contacts between T cells and DCs , with a threefold higher number of unique contacts at maximal chemotaxis strength compared to random migration ( Fig . 3A , bottom ) . Although the average contact duration also decreased in these simulations , the effect was less strong than for the no-desensitization mechanism ( Fig . 3B , compare top with bottom ) , suggesting that desensitization reduced competition between T cells around the DC . Furthermore , the reduction in the contact duration was slightly weaker for longer recovery times , despite the fact that longer recovery times led to the scanning of more unique T cells than shorter recovery times ( Fig . 3A ) . This was because a sufficiently long recovery time allowed insensitive cells to ‘escape’ from the chemokine attraction field of a recently contacted DC and subsequently contact other DCs . Moreover , longer recovery times allowed for longer T cell-DC interactions because competition around DCs was reduced . In short , the DC-contact mechanism caused a strong coordination in T-cell movement ( Fig . 2 ) , leading to a higher motility coefficient ( Fig . 1 ) and allowing for cells to escape the chemoattractant field around the DC , all together causing the high scanning efficiency of this mechanism . In contrast , Riggs et al . [16] used a different method of desensitization . To make a better matching comparison , we also tested this alternative mechanism , even though it might be biologically less reasonable . This method decoupled T-DC contact from desensitization , by letting T cells become insensitive after having been in contact with the chemokine for a certain fixed time period ( the desensitization time , see Methods ) . In these simulations we observed an optimal duration of the desensitization time , which depended on the concentration threshold at which the T cell started to sense the chemokine gradient ( Fig . S2 ) . The higher this threshold , i . e . , the closer to the DC the T cell had to be to sense the gradient , the shorter this desensitization time had to be in order to achieve efficient scanning . This is because scanning is most efficient when desensitization occurs around the time the T cell gets into contact with the DC . Moreover , it was less efficient to have an overly short desensitization time ( e . g . , 1 min ) than to have an overly long desensitization time ( e . g . , 30 min; Fig . S2 ) . This confirmed our finding that chemotaxis always increases the efficiency of DC scanning , while it is an even more efficient strategy to become insensitive soon after being in contact with a DC . Regardless of the exact mechanism of chemoattraction used , we consistently observed more efficient T cell scanning for chemoattraction towards DCs compared to random migration . Therefore , these results clearly showed that chemoattraction is expected to promote T cell scanning by DCs . Because in our model the entry of new cells into the tissue occurred only when other cells left the tissue , chemoattraction might decrease the influx of new cells by keeping cells within the tissue that otherwise would have left . Alternatively , chemoattraction could increase the influx because of the larger motility coefficient . In the latter case , our observation of chemoattraction increasing the T cell scanning efficiency by DCs might be due to an increased influx of new T cells instead of a more effective search amongst the cells that were already locally present . However , the number of cells entering the simulation hardly changed in the presence of chemoattraction ( Fig . 4A ) , so this scenario could be excluded . About half the T cell population left our simulated space over the timespan of an hour ( Fig . 4A ) , and the average residence time of a T cell in our simulations was approximately 1 . 5 hours . To investigate whether chemotaxis remains efficient over timescales longer than one hour ( our typical simulation time ) , we also ran some simulations lasting for 24 hours . Figure 4B shows that the number of unique first contacts increased linearly with time , demonstrating that scanning in the presence of chemoattraction remained efficient at long timescales , during which numerous cells entered and left the simulated area . In conclusion , independent of the small variability in the entry rate of T cells , chemotaxis leads to efficient T cell scanning by DCs at both short and long timescales . Our conjecture is therefore that the negative effects of chemotaxis reported by Riggs et al . [16] were due to their model formalism . However , it is not immediately clear whether the positive effects we observed in our model still hold in the context of rare cognate antigen , and we turned to a more realistic , 3D version of our model to investigate this . We next addressed the question whether there are differences in the efficacy of scanning with chemotaxis compared to random migration when the DCs producing chemokine are rare . To capture this situation , we allowed only a single DC ( located centrally in the field ) to produce chemokine . To increase the realism of our simulations , we simulated a 3D LN tissue with a similar layout as used for the 2D simulations ( Fig . 5A , B ) . Because it is unlikely that there is a difference in cellular properties ( migratory or otherwise ) between antigen-specific T cells and nonspecific T cells prior to contact with the cognate DC , we scored for a large number of T cells ( representing antigen-specific T cells ) in each simulation whether they were able to find the DC bearing cognate antigen . Specifically , we followed 100 T cells per simulation entering the tissue after an initialization period . For each of these T cells we recorded whether they managed to come into contact with the antigen-bearing DC or left the tissue without such a contact . The output of each simulation was the percentage of T cells finding the single chemoattracting DC . Using the DC-contact mechanism , the percentage of T cells arriving at the chemoattracting DC increased markedly ( more than 3-fold ) with increasing chemotaxis strength ( Fig . 5C ) . We also investigated the time it took until T cells found the DC presenting cognate antigen , as well as their starting distance to the DC in case they managed to find it . As shown in figure 5D , it took T cells slightly less time to find the cognate DC in the presence of chemoattraction compared to random migration . Thus , chemoattraction modestly speeds up the search process for T cells that successfully find the chemokine-producing DC , and greatly increases the probability of establising cognate interactions between T cells and DCs . Random search processes depend strongly on the dimensions of the space considered , because cells are more likely to revisit previously searched regions in 2D than in 3D , which makes a 2D random search less effective [37] . Additionally , crowding effects due to chemoattraction might be less prominent in 3D than in 2D , because there exist more ‘escape’ routes in 3D . Therefore , we performed similar simulations with a single attracting DC in 2D . Consistent with the 3D simulations , the the effects of chemotaxis on the efficiency of scanning increased very strongly in 2D ( Fig . S3 A ) . However , in contrast to the 3D simulations , chemotaxis did not speed up the search process ( Fig . S3 B ) . Despite these small differences , it is clear that in both 2D and 3D the scanning efficiency is enhanced by chemotaxis of T cells towards dendritic cells , thus contributing to an effective and timely immune response .
2 PM imaging experiments have shown that T cell migration in LNs roughly conforms to a persistent random walk [4]–[8] . However , subtle chemotactic migration could well be hidden in such data . For example , we recently discovered a small yet biologically relevant directed migration component amongst germinal centre B cells [38] , for which migration had earlier been described as random [39] . Similarly , Textor et al . [40] recently showed that a uniform ( e . g . , from LN ingress to egress points ) chemotactic migration component of considerable size could be present in LN T cells in the absence of cognate antigen . Hence , the same is likely true for a chemoattraction component towards DCs . Here , we used computational modelling to address the question whether chemoattraction of T cells towards DCs is expected to promote or to hamper the scanning efficiency of DCs . Our simulations showed that , in the absence of cognate antigen , chemoattraction towards DCs enhanced their T cell scanning efficiency about three-fold compared to random migration . Furthermore , when T cells in our simulations had to find a single , centrally located DC that produced a chemoattractant , this search was most successful when there was both strong chemoattraction towards that DC and desensitization upon arrival of a T cell . From these simulations we also learned that the efficiency of the search mechanism hardly depends on the dimensionality of the tissue and is thus very well suited for the complex LN environment . Consistent with our finding that chemoattraction of T cells towards DCs is more efficient than a random search process , 2 PM imaging experiments have revealed that chemotaxis towards licenced DCs indeed occurs in vivo [11] , [12] , [36] . Chemotactic migration amongst multiple local sources of attractant is not detectable by a mean ( square ) displacement analysis [2] , [35] , which we here confirm for our simulation data . Rather , measurements of T cell migration angles relative to the vector towards a nearby DC can be used to detect potential chemoattraction ( note that this is also how chemotaxis towards licenced DCs was demonstrated experimentally [36] ) . It might be possible to make such an angle analysis more powerful by distinguishing between cells that had a recent contact and those that did not . The angle measurements of the T cells in our simulations offer an explanation for why especially desensitization upon DC contact renders a very efficient search process: the migration pattern of cells that just desensitized gave the impression that these cells are moving away chemotactically from the DC ( Fig . 2 ) . However , we did not include such chemorepulsion in our model and the pattern must therefore purely result from the pushing away of insensitive T cells by sensitive T cells . Insensitive cells are pushed away more easily than sensitive cells . Sensitive cells have a bias towards the DCs which the insensitive cells do not have . Therefore , when a sensitive cell on its way to a DC collides with an insensitive cell , it is likely that the insensitive cell is pushed into a different direction of migration while the sensitive cell continues , eventually causing the formation of a small stream of sensitive cells moving towards the DC . Furthermore , when those sensitive cells reach the DC and become insensitive cells , they persist in migrating in the direction into which they are pushed by the sensitive cells behind them , causing them to form a stream that moves away from the DC . In this way around the DCs convective streams are formed , with sensitive cells moving towards the DC at one spot and insensitive cells moving away from the DC at another spot , analogous to the organization of crowds of pedestrians in a busy city centre or subway [41] . This process reduces competition near DCs when the DC-contact mechanism is used , allowing for longer contacts than in the simulations without desensitization . Furthermore , the efficient displacement of insensitive T cells away from DCs that they had already contacted allows these cells to swiftly establish contact with other DCs , thus resulting in efficient T cell scanning . Although desensitization of individual receptors has been shown to occur in leukocytes [42] , [43] , it is currently not known whether T cells can desensitize to a chemokine gradient . The presence of such desensitization upon DC contact would require fast signalling between T cells and DCs . Indeed , signals between T cells and DCs may be transferred in the course of seconds [44] and therefore it is possible that desensitization occurs even for brief T-DC contacts . An alternative could be that T cells become sensitive to other chemokines after contact with a DC , allowing them to move away from that DC . However , this seems unlikely because each DC should then produce a different chemokine . If alternatively the other chemokine is produced by an entity outside the T cell zone , T cells would leave the T cell zone soon after their contact with the first DC , and would likely miss DCs carrying cognate antigen , although such a mechanism could make sense after a T cell made contact with a DC carrying cognate antigen . Thus , although there is currently little experimental evidence for desensitization and resensitization , i . e . , the loss and recovery of sensitivity to the chemokine gradient , our results suggest that loss and recovery are expected to lead to more efficient scanning . This is because it allows T cells to more easily reach DCs different from those already seen , and as such have a higher probability to find rare antigens . Once a T cell has recognized cognate antigen presented by a DC , other pathways need to be induced to stabilize the interaction . This likely does involve other chemokines [45] as well as formation of an immunological synapse [46] . Contrary to our findings , earlier simulations using a 2D agent-based model of T-DC interactions in the presence of a chemoattractant gradient suggested that chemotaxis hampers the scanning efficiency of DCs [16] . The negative effect of chemoattraction in those simulations was a consequence of chemotactic T cells blocking access to the DCs [16] . These results , however , may stem from the fact that agent based models have the intrinsic property that cells cannot move into lattice positions that are already occupied . Therefore , in these models cell migration cannot be properly captured when cell density is high without the use of additional assumptions: the in silico cells cannot squeeze past each other [17] and cannot push each other , whereas 2 PM imaging studies have shown that T cells in the LN are highly flexible , readily change their shape and migrate rapidly despite the densely packed environment [4] . In the simulations by Riggs et al . [16] the authors attempted to alleviate this problem by reducing the T cell density to below-physiological values . However , in the scenario with chemoattraction of T cells towards DCs , T cell densities would still become locally high , thereby reintroducing the problem of blockage . Thus , their result that chemoattraction reduces the number of unique T cells that are scanned by DCs seems to be a consequence of the model formalism . It would be interesting to attempt to solve the problem of blockage in such agent-based models on a lattice by either allowing two cells to temporarily occupy the same lattice site such that they can pass each other [47] , or to allow for swapping of cells [48] , [49] . ( Note that ‘convective flow’ as we observed is unlikely to occur in such simulations , because cells would still not be able to push each other . ) Another CA-based model , which combines persistent motion and chemotaxis , has been proposed for for B cell activation in the lymph node [50] . As an alternative approach to reinvestigate the question whether chemoattraction of T cells towards DCs is expected to promote or obstruct scanning , we employed the CPM formalism [19] , [20] . Using this formalism , we were able to show that even in a densely packed field combined with chemoattraction of cells towards DCs , no blocking occurs and T cells can keep on migrating by forming a convective flow around the DCs . We conjecture that the difference in model behaviours is because the CPM is able to properly describe the shape and flexibility of biological cells as well as their interactions with other cells within a densely packed area ( e . g . , [26] , [27] ) . In the CPM cells are represented by multiple lattice sites , allowing them to undergo complex shape changes . Combined with the ability to push and pull each other , microstreams are generated that organize the circulation of T cells near DCs . For these simulations , we chose not to explicitly model the subcellular processes that play a role in chemotaxis . Instead , we applied a phenomenological shortcut to capture these processes , which allowed us to study the consequences of many interacting cells responding chemotactically to a very complex and dynamically changing chemotactic field . Our approach was further simplified in the sense that we did not study the disturbance of the chemokine gradients in the LN by the migrating cells themselves . Indeed , it is difficult to imagine how a gradient could be maintained in the presence of numerous , frantically moving T cells . As a possible mechanism , it has been proposed that secreted chemokines may be rapidly immobilized on the reticular network [51] , forming a gradient for T cells to follow . Although we did not study this in our CPM simulations , we expect that such a role of the network would give similar results as the scenario we considered here . In conclusion , we have shown that chemoattraction of T cells towards DCs is expected to increase the efficiency of T cell scanning by DCs , thereby greatly contributing to a timely immune response .
In the CPM model formalism [19] , [20] , cells consist of multiple lattice sites ( with 2D coordinates i and j , or 3D coordinates i , j and k ) , and have a type and identification number . Lattice sites of the cell in contact with the surrounding environment ( other cells , medium , RN ) have a surface energy which depends on the type of the cell ( ) and that of its neighbour ( ) . Cells are assumed to minimize their surface energy while at the same time trying to maintain their volume at a target value . During updates of the lattice , the probability of a randomly chosen neighbour to extend into the site under consideration depends on the so-called Hamiltonian ( given for the 2D case only ) : ( 1 ) The first term represents the sum of all surface energies , where is the Kronecker delta and sums over all 8 neighbours in the neighbourhood . The second term keeps the actual volume close to the target volume , where is the inelasticity of cells . The probability that a neighbouring site extends into the lattice site under consideration is 1 if ¡ 0 , and e otherwise , where is the change in the Hamiltonian due to the considered modification , and T represents the membrane fluctuation amplitude of cells . The model was implemented using the C programming language , and the cell migration measurements were performed using customized Perl scripts . In silico T cells exhibit a target direction , and extension of lattice sites into that direction occurs with a higher probability than extension into the opposite direction . This was incorporated by extending for T cells: ( 2 ) where is the ‘directional propensity’ of cells , and is the angle between the target direction and the displacement vector under consideration ( i . e . , the vector given by the mean position of the cell and the coordinates of the position to be modified ) . The target direction is updated every seconds according to the actual displacement of the cell . Apart from this baseline motility that gives rise to a persistent random walk [26] , T cells in our simulations respond chemotactically along a local gradient . Extension into sites with a high chemokine concentration is favoured , and this depends on local subcellular chemokine concentration differences . This is implemented into as follows ( shown for the 2D case ) : ( 3 ) where is the chemokine concentration at the lattice site under consideration and is the concentration at the neighbouring site that attempts to extend . An increase in causes cells to react more strongly to a gradient . However , there is a limit to the extent to which can be increased , because at some point the term driving chemotaxis becomes stronger than the volume conservation term , which can cause T cells that are pushed against DCs by other T cells to reduce their volume to zero and ‘die’ . In all simulations , we keep below the point where this non-biological behaviour occurs . In some simulations , T cells remain sensitive to the chemokine gradient for the entire duration of their stay in the simulation , which we call the no-desensitization mechanism . In simulations with T cells that are insensitive to the chemokine gradient , the second extension of ( equation ( 3 ) ) is only taken into account for the sensitive cells . We implemented two manners in which T cells could become insensitive , referred to as the DC-contact method and the gradient-contact method . In the DC-contact method , T cells become insensitive immediately upon contact with a DC . In that case , they remain insensitive to the chemokine gradient for the duration of a ‘recovery period’ ( ) counting from the time of first contact . The gradient-contact method is independent of contact with DCs . Instead , T cells become insensitive after a ‘desensitization time’ , i . e . , they desensitize minutes after the first chemotactic response , which is initiated when the cell has sensed a chemokine concentration above a threshold value . They subsequently resensitize after a recovery period of minutes . As described earlier [27] , DC dendrites are modelled by defining multiple actin bundles protruding from positions within the DC and retracting after a pre-set time period . In brief , the bundles grow in a random , straight direction , provided that the sites to be copied into belong to the DC . Protrusion of the membrane is achieved by increasing the likelihood of DC membrane elements copying into positions adjacent to a bundle ( in that case is decreased with ) . To prevent breaking of dendrites , membrane elements at or adjacent to a bundle cannot be copied into . When bundles cannot extend for timesteps due to obstacles or chance processes , they retract . Otherwise , the bundle retracts after a maximum of timesteps . Retraction occurs with a probability per time step , and as soon as a bundle has completely retracted , a new bundle starts to grow out in a random direction . Each DC has dendrites at a time . In silico DCs produce chemokine , and the chemokine concentration c is followed over time in each lattice site of the grid ( measured in arbitrary units ) . The chemokine diffuses with a diffusion coefficient , irrespective of the occupancy of lattice sites by cells , RN or extracellular medium . This gives the following equation:in which the chemokine production term is limited to the DCs only , while the decay , given by , takes place all over the tissue . Unless mentioned otherwise , we use the default parameters defined here . Our 2D model consists of a wrapped square of , and the ( wrapped ) 3D model space is . One site of the lattice represents 1 ( or 1 ) . Cell flow across the upper boundary is blocked by an obstacle representing the lymph node capsule , which spans the width of the field ( Fig . 1A ) . When T cells touch the lower boundary their target area is set to zero so that they shrink and leave the simulation . When the cell has disappeared , a new T cell immediately enters at a random position . In this manner , the number of cells in the simulation remains constant , and there is a weak flow of T cells downward . The number of T cells in the 2D field is close to 6700 , and close to 5000 in the 3D field , resulting in fields that are nearly completely packed with cells and RN . Small changes in T cell density do not have a qualitative effect on the T cell scanning efficiency of DCs . However , at very low densities chemoattraction retains T cells in the field , which leads to a decreased long-term scanning efficiency compared to random migration . We previously found that in the absence of obstacles , densely packed T cells in CPM simulations tend to form massive streams [26] , which are not observed in a real lymph node . In the presence of an in silico reticular network the streams turn into more realistic local , highly turbulent microstreams . We therefore included a representation of the reticular network by incorporating 900 randomly placed circular objects with a radius of 4 into the 2D field ( about 18% of the space ) and 3000 randomly oriented rods with a 1 radius and a length of 20 into the 3D field ( about 17% of the space ) . T cells are initialized at random positions , whereas DCs are placed according to predefined coordinates that are the same for every simulation ( Fig . 1A ) . During a simulation DCs are not allowed to move large distances , which is achieved by having their dendrites grow out from a block around the initial position of the DC . All cells are initialized as a 9 block in 2D or a 27 block in 3D , after which they quickly grow out to their target area ( 30 or 150 for T cells and 100 or 1400 for DCs ) . T cells are considered to have a slight preference to adhere to DCs , and there is no differential adhesion between other cell types . Preferential adhesion is implemented as a negative surface tension ( ) between cell types x and y , and is calculated from the surface energies as follows: . The default surface energy and surface tension parameters are shown in Table 1 . Other default parameters are , , , , , , , , , , for 2D simulations; and , , , , , , , , , and for 3D simulations . After one Monte Carlo timestep , all sites in the lattice have been considered for updating , which corresponds to 1 sec in real time . Measurements start after 100 sec , defined as time 00∶00 ( min∶sec ) . The mean position of each T cell is registered every 10 seconds and is used to calculate displacements , speeds and migration angles . Motility coefficients and persistence times were estimated from mean square displacement plots by fitting Frth's equation for a persistent random walk ( , where is the mean square displacement , is the dimension of the space , is the motility coefficient , is the persistence time and is the elapsed time period since the start of the trajectory ) [40] , [52] to the data using the software package R ( freely available at http://www . r-project . org/ ) . Interactions between T cells and DCs are registered every second and are considered contacts when they touch each other at at least one grid point . | T lymphocytes are important actors of the immune system that find and kill infected cells . Before a T cell can mount such an immune response , it has to be activated through contact with a dendritic cell ( DC ) carrying antigen relevant to the specificity of the T cell receptor . This process typically takes place in secondary lymphoid organs such as lymph nodes and spleen , where DCs scan many T cells at a time . However , the fraction of T cells specific for any antigen is about – , and therefore establishing a contact between a DC carrying cognate antigen and the correct T cells seems quite a challenge . Here , we show with a computational model that despite the presence of millions of competing non-specific T cells , the probability of such a cognate interaction greatly increases when DCs produce a chemokine ligand to attract T cells . The scanning process becomes even more efficient when T cells become insensitive to the chemokine after contacting the DC . These findings oppose the earlier notion that chemoattraction is counterproductive due to blocking of DCs by T cells of irrelevant specificities . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"immunology",
"biology",
"computational",
"biology"
] | 2012 | Chemotactic Migration of T Cells towards Dendritic Cells Promotes the Detection of Rare Antigens |
Identifying genetic factors responsible for serious adverse drug reaction ( SADR ) is of critical importance to personalized medicine . However , genome-wide association studies are hampered due to the lack of case-control samples , and the selection of candidate genes is limited by the lack of understanding of the underlying mechanisms of SADRs . We hypothesize that drugs causing the same type of SADR might share a common mechanism by targeting unexpectedly the same SADR-mediating protein . Hence we propose an approach of identifying the common SADR-targets through constructing and mining an in silico chemical-protein interactome ( CPI ) , a matrix of binding strengths among 162 drug molecules known to cause at least one type of SADR and 845 proteins . Drugs sharing the same SADR outcome were also found to possess similarities in their CPI profiles towards this 845 protein set . This methodology identified the candidate gene of sulfonamide-induced toxic epidermal necrolysis ( TEN ) : all nine sulfonamides that cause TEN were found to bind strongly to MHC I ( Cw*4 ) , whereas none of the 17 control drugs that do not cause TEN were found to bind to it . Through an insight into the CPI , we found the Y116S substitution of MHC I ( B*5703 ) enhances the unexpected binding of abacavir to its antigen presentation groove , which explains why B*5701 , not B*5703 , is the risk allele of abacavir-induced hypersensitivity . In conclusion , SADR targets and the patient-specific off-targets could be identified through a systematic investigation of the CPI , generating important hypotheses for prospective experimental validation of the candidate genes .
Identifying genetic risk factors responsible for serious adverse drug reactions ( SADRs ) is one of the priorities in pharmacogenetics [1] . As it is not practical to perform genome-wide association study due to the lack of samples [2] , candidate gene selection has been an important strategy . Challenges arise when the primary mechanisms for many SADRs are unclear . Consequently , the candidate genes selected are generally limited to those coding therapeutic targets [3] , transporters or metabolic enzymes [4] . We named this strategy as the “known interaction driven selection” , for it is driven by the known interactions between drugs and proteins . However , the drug-protein interactions at this level cannot explain why an SADR is only induced by certain medications but never caused by other drugs . For example , Stevens-Johnson syndrome ( SJS ) is often caused by diclofenac , didanosine and tenoxicam but never caused by propoxyphene . It is undisputed that direct interaction between a chemical and a protein , for example , noncovalent binding of a drug to the active center of an enzyme , is a fundamental step in drug effect . Hence , we hypothesized that drugs causing the same type of SADR might share a common mechanism by targeting unexpectedly on the same SADR-mediating protein . But questions still arise as why after strict assessment before the drugs came to the market , SADRs still and only happen to some certain individuals , especially the type B or idiosyncratic SADR [5] . Evidences showed that some rare polymorphisms within these SADR targets made them more sensitive to the drug . For example , oseltamivir ( Tamiflu ) , an anti-flu drug , whose active form binds to the active site of human cytosolic sialidase . A rare polymorphism near the binding pocket may enhance this unexpected binding and might increase susceptibility to oseltamivir-induced neuropsychiatric disorders [6] . Another case concerns the A1555G mutation in mitochondrion DNA , which enhances the unwanted binding of aminoglycosides to human 12s rRNA , mediating the susceptibility of aminoglycosides-induced deafness [7] . Thus , the candidate gene selection of SADR genetics can be tackled by exploring the unexpected chemical-protein bindings . To harvest them at high throughput , we established the first chemical-protein interactome ( CPI ) in the form of the interaction strength among FDA-approved drugs and human proteins . Each of the drugs was reported to cause at least one of the four major SADRs including SJS/TEN , cholestasis , rhabdomyolysis and deafness . We designed a data-mining strategy against the CPI to explore whether the common SADR targets existed . In brief , if different drugs that share the common outcome of SADR “S” all interact with a particular protein “P” , whereas drugs that do not cause the “S” outcome do not interact with it , then the common target “P” can be considered as a mediator of “S” and prioritized for association studies and mechanism investigations .
Several techniques such as BIACORE biosensors [8] and drug affinity pull-down [9] can be used to assess the chemical-protein interactions and to identify the unexpected chemical-protein bindings . However , in order to test the utilities of CPI in a low cost and high throughput manner , we intended to choose a mature technique . Dock programs [10] appeared to be a good option . The DOCK [11] has been under development and improvement for over 20 years and is widely used to evaluate the interaction strength between drug candidates and proteins targets . Particularly , it has previously been used to identify the unexpected binding . A classical case is haloperidol , an anti-psychiatric drug , which was found to bind unexpectedly to HIV protease and had become a template for developing anti-HIV drugs [12] . The discovery was made with DOCK , whose later version [13] was applied in our research reported here . Since human knowledge of the four SADRs is limited , neither did we know any of the unintended drug-protein interactions in them nor did we not know how many protein targets should be enough to cover the mechanism space of them without bias . So we selected targets from literature and third-party targetable protein database [14]–[17] and then applied quality control steps as described in the methods . Considering our productivity of pockets preparation and the urgency of solving the SADR problem , a set of 845 proteins ( Table S1 ) finally passed the quality control . Although the protein set was incompetent to cover the whole SADR targets , if some unexpected and valuable information could be mined from it , the methodology of CPI would enlighten the following research and thus lead to the construction of a large scale target set . We constructed a test CPI to test the feasibility of using DOCK in evaluating the interactions . In our protein set , there were 12 proteins which had been the therapeutic targets as they were listed in DrugBank [18] . We extracted the known direct interactions with the 12 proteins reported in DrugBank or the literature for a total of 46 drugs . We then ran the DOCK for 46×891 times , resulting in a docking score matrix with 46×891 elements . The matrix was then shrunk to 46×845 elements before we converted the matrix into the Z-score [19] matrix , where binding affinities for each drug across the 891 binding pockets were normalized to a mean of zero and standard deviation of one , since it has been reported that the normalization of the docking score matrix can improve the hit ratio [20] . Docking score distribution is dependent on sizes , charges and other characters of the drug . The normalization could address this inconsistency among drugs . Each drug-protein interaction was classified into one of the two categories , depending on whether ( group 1 ) or not ( group 2 ) the interaction was previously reported in DrugBank or in the literature . Statistical test of Z-score showed that the two groups belonged to different population ( Table S2 ) . The area under ROC curve was 0 . 74 ( 95% CI: 0 . 68–0 . 80 , Figure S1 ) , indicating that the Z-score was valuable in measuring true bindings . The 50th percentile of Z-score in group 1 interactions was −1 . 240 while the 50th percentile of group 2 was only −0 . 47 . We thus set a Z-score threshold of −1 . 2 in order to distinguish the known bindings ( group 1 ) from the unidentified bindings ( group 2 ) . Note that some of the unidentified bindings might exist per se , which would reduce the difference between two groups . However , this misclassification of unidentified interactions into known bindings did not affect the specificity of highlighting the true bindings from the unidentified ones . So we concluded that Z-score could tell whether a binding will occur at the interval of high absolute value . Based on the reliability of the dock program and the data processing strategy that effectively separate known bindings from unidentified ones , we introduce the first release of CPI in the form of a Z-score matrix . The chemicals selected here were FDA-approved drugs , each of which was reported to cause at least one of the four major well-known SADRs mentioned above . The derivatives of these drugs , such as the known major metabolites and their known isomers were also included . In summary , the CPI consists of the binding affinity data between 162 chemicals and 891 binding pockets . The interaction strengths were converted into a Z-score matrix . We did not choose SADRs like hepatotoxicity , for it is a relatively big concept compared with cholestasis and can be induced by almost every drug . The four SADRs included in this research were not only the major SADRs usually reported in the FDA's Adverse Event Reporting System ( AERS ) , but each of which was also appeared to be triggered by a particular set of medications . This type of SADR allowed us to identify “case” and “control” drugs from which clear differences in the pattern of binding to many proteins were observed from the CPI . In the following section , we used the SJS/TEN SADR as an example to illustrate the construction and utilities of the CPI . The SJS and TEN are two forms of the same life-threatening cutaneous reactions that cause rash , skin peeling , and sores on the mucous membranes triggered by particular types of medications [21] with primary mechanism unknown . No significant association was observed between the metabolite enzyme genes and the SJS/TEN [22] , implying that “known interaction driven” genes might not be the fundamental element . We first selected drugs that were reported to be associated with this SADR in peer-reviewed publications . All of the drug-SJS/TEN relationships were confirmed in the FDA's AERS . In total , 32 drugs along with their 21 major derivatives served as the case group , whereas 17 drugs were verified to be unrelated to SJS/TEN in both publications and AERS and served as the control group , which did not contain the derivatives . To avoid biases in the following assessment , we also confirmed that they did not share the same chemical features . After docking all 70 molecules into all 891 binding pockets of our set of 845 proteins ( step 1 of Figure 1 ) , we obtained a Z-score matrix of the binding affinities . We then split it into the case matrix ( 53×845 relations ) and the control matrix ( 17×845 relations ) . We performed a hierarchical clustering [23] on the resulting zero-floored Z-score matrix , and found that three sub-groups of case drugs clearly interacted selectively with three different sub-groups of proteins ( step 2 Figure 1 ) , implying that the three different sub-groups of case drugs might trigger the SJS/TEN through three different mechanisms . Then we divided the case matrix into three sub-CPIs , and performed a trimming procedure to exclude the redundant case chemicals if multiple forms of a drug were clustered into the same sub-CPI . To identify proteins preferentially interacting with the case drugs , we performed Fisher's exact tests followed by false discovery rate ( FDR ) corrections [24] for every sub-CPI in comparison to the control group . This non-parameter test only required binding information in a binomial pattern , which could be specifically measured by Z-score . In sub-CPI 1 ( step 3 of Figure 1 ) , p and q values [24] for each proteins were calculated . As a result , HLA-Cw*4 heavy chain ( 1QQD ) together with other proteins were highlighted . In sub-CPI2 and sub-CPI3 , leukotriene A4 hydrolase ( 1HS6 ) , CD26 ( 2G5P ) and Fab′ fragment of IgG ( 1DBJ ) together with other proteins were identified ( step 4 , 5 of Figure 1 ) . We further enriched the biological process ( BP ) terms of Gene Ontology ( GO ) from the gene co-citation network ( GCCN ) [25] . We downloaded 18566 SJS/TEN-related PubMed entries , calculated the citation rates for each of the human genes in SJS/TEN related and nonrelated PubMed entries . Genes cited more specifically in the SJS/TEN topic were defined as core genes . Two genes were connected if they were co-cited in a PubMed entry ( step 6 of Figure 1 ) . To retrieve a more delicacy set of GO BP terms for SJS/TEN , we extended the core gene set through indexing their neighbors in GCCN . For example , LTA was one of the neighbors of the core gene HLA-C [26] . Though had not been investigated in SJS/TEN , LTA was co-cited with HLA-C [27] , implying a putative functional linkage of this gene to SJS/TEN through HLA-C . Both the extended and the core genes were used in the enrichment analysis of BP terms [28] . As shown in step 6 of Figure 1 , genes annotated with “immune response” ( p = 8 . 32E-07 ) , “inflammatory response” ( p = 2 . 48E-18 ) and “T cell activation” ( p = 1 . 02E-41 ) tended to occur more frequently in SJS/TEN-oriented GCCN than random selection at a significance level of 0 . 01 . Finally , the BP terms were assigned manually to proteins highlighted from sub-CPIs . Proteins were defined as the class I candidates if they shared the same BP terms enriched from GCCN . Otherwise they were defined as class II candidate . Class I candidates were involved in the known biological processes of SJS/TEN while class II proteins did not . However , it does not mean that class II candidates are less important since human knowledge on SJS/TEN is still limited . MHC I protein heavy chain Cw*4 ( 1QQD ) showed the lowest p value in the candidate list . It is assigned as a class I candidate , as it was annotated with the BP terms “immune response” ( GO:0006955 ) and “antigen processing” ( GO:0019882 ) , which were highlighted from the GCCN . When investigated the interaction strength among all case-control drugs and HLA-Cw*4 ( Table S3 ) , we found that 85% ( 11 of 13 ) case drugs including 9 sulfonamides bind strongly to it . Table S4 showed significant differences between case and control interactions with HLA-Cw*4 either in docking scores or in Z-scores . By visualizing the binding conformations at the lowest energy , we found that all sulfonamides tended to “root” at MHC I's antigen presentation groove through the hydrogen bond interaction of sulfuryls to the two arginine residues ( Figure 2B ) . This identification of HLA-C ( w*4 ) as the mediator of sulfamethoxazole ( SMX ) -induced TEN was validated by other studies , as it was confirmed that the immune response and the TEN will only be triggered by SMX in presence of MHC I ( Cw*4 ) [26] , [29] . Following the same data-mining and text-mining pipeline , we identified candidate genes from the other two sub-CPIs ( Table 1 ) . Two representative proteins highlighted were leukotriene A4 hydrolase ( 1HS6 ) and Fab fragment of IgG ( 1DBJ ) . The former is the rate-limiting enzyme in formation of leukotriene , which transduces the signal of inflammation in skin reactions [30] . Case drugs tend to bind to its peptidase active center and might interfere the suicide regulation [31] of the enzyme itself when the enzyme is over-expressed . Case drugs in sub-CPI 3 tend to bind to the variable region of a certain IgG . We could not deduce the downstream events of this binding , but it has been known that the binding of antigens to the IgG induces the release of leukotriene , activate alexin system and the type III hypersensitivity . The CPI would not only tell which protein to mediate the SADR , but would also tell which allele of this protein would be more sensitive to the unexpected drug attack . To our knowledge , HLA-B*57 is the only reliable susceptibility gene of SADRs [32] , [33] of which structures with both risk and non-risk alleles are available . We constructed an interactome including interaction strength among abacavir , allopurinol and four structures of risk and non-risk alleles of abacavir-induced hypersensitivity ( Table 2 ) . No specificity of allopurinol to any of the proteins was found . This result was in coordination with the fact that none of these alleles was the risk allele of allopurinol-induced hypersensitivity , In contrast , abacavir did not accommodate the binding site of B*5703 , but appeared to have the high affinity with B*5701 . The major difference between the two alleles lay in two polymorphisms ( N114D , Y116S ) from B*5703 to B*5701 . When Y116S substitution appeared in B*5703 , a better compatibility of geometry shape between drug and binding pocket as well as several hydrogen bonds were formed . As a result , abacavir molecule fixed deeply into the antigen presentation groove of HLA-B*57 ( Figure 3 , Video S2 ) . Comparatively , the N114D seemed to be less important for the substitution of a nitrogen atom to an oxygen atom would neither affect the steric hindrance nor did the buildup of hydrogen bonds . Given the premise that direct binding of drug to HLA-B*57 protein mediates abacavir-induced hypersensitivity , we deduced that B*5701 tended to be the risk allele compared to B*5703 . The discovery of the fact that abacavir interacts directly with the 114th and 116th residues of MHC I B*57 which mediates SJS is consistent with the genetic evidences [32]–[35] . This newly identified molecular mechanism has also been validated at the cell biology level [36] . In the presence of abacavir and MHC I ( B*5703 ) , the percentage of responding IFNg+ CD8+ T cells was only 1 . 34% , and the percentage remained unchanged when N114D mutation was introduced . However , this percentage suddenly rose to 28 . 4% when another mutation Y116S , which formed the risk allele B*5701 , was introduced . The result fit the drug-MHC I direct binding model for that Y116S was essential to the binding of abacavir , and N114D tended to be less important . A systematic insight into the CPI led to the identification of the SADR targets common to case drugs , and the risk alleles of them as well . We thus explored SADR targets for three other SADRs using the same methodology ( Table 1 ) . The relationships among drugs , SADR outcome and the corresponding sub-CPI were listed in Table S5 . Vasopressin receptor was highlighted from cholestasis-related CPI . Being the native ligand , administration of vasopressin could result in a reduction in bile flow and then induce cholestasis [37] . Troponin T and PAK1 protein kinase were harvested as well . The former regulates muscle contraction while the latter , inhibited by most statin drugs in the rhabdomyolysis-oriented CPI , takes a vital part in the polymerization and depolymerization of actin . Energy dysfunction plays an important role in pathogenesis of hearing loss [38] . We found that , compared with other SADRs , deafness is significantly associated with proteins contributing to energy metabolism ( see note of Table 1 ) . The consistently higher than random recall rates of proteins known to be related to SADRs indicates that all these results could not have been achieved by chance . Network pharmacology [39] pointed out that many drug effects are mediated by the chemical-protein interactions of drugs towards multi-protein set . The SADR might also be triggered by the combination effect of drug-SADR target interactions . So we hypothesized that drugs causing the same SADR not only share the same SADR targets , but might also possess the similar binding strength profile towards multi-protein set . If this similarity can be detected , we may infer that the CPI not only represent the binding situation of a drug to a protein , but also reflect the interacting character of it towards multi-protein set . Here we utilized support vector machine ( SVM ) model to see whether drugs could be correctly classified as case or control drugs based on their binding profile vector against 845 proteins . If they could be , there was a similarity in high-dimensional space among case drugs or control drugs towards 845 protein set . The effectiveness of the classifier was measured by the binary classification accuracy using cross validation . For each SADR , categorical attributes of case or control drugs were labeled as “1” or “0” . Z-score of a drug towards 845 proteins was used as the attribute vector . The cross validation accuracy ( CVA ) varied from 85% to 91% among four SADRs ( Table 3 ) . To evaluate whether such CVAs were achieved by chance , we permutated the position of the case and control drugs randomly for 100 runs , and recalculated the mean CVAs . The mean CVAs of permutated data turned out to be much lower ( Table 3 ) . So we concluded that there were similarities among drugs with the same outcome of an SADR , which were found in the CPI profile of them towards multi-protein set .
The hypothesis that drugs with similar phenotypic effects tend to interact with same targets is similar to a recent study done by Campillos et al [40] . However , the target spaces and the aims of the two studies are different . The work of Campillos et al managed to construct new connections among drugs and known therapeutic targets , which are a small portion of protein spaces whose functional information is clearly identified . Our research tried to construct new connections among drugs and human proteins , which is a step into a larger protein space whose function needs further exploration . However , our methodology is hampered by the lack of the structurome information of the human proteins . The aim of the former research is to explore the off-targets . For our research , the major aim of finding the off-targets is to figure out the key interacting residues and the risk allele for each individual . Unexpected drug-protein interaction is the vital step in pathogenesis of SADRs . Although drug response is a complex trait [1] mediated by multiple genes and their interactions , some well-known cases of polymorphism within a gene have pronounced effects on drug response . To our knowledge , all of these polymorphisms alter the pattern of direct chemical-protein interactions . Examples include the T790M mutation in the gefitinib binding pocket of EGFR [41]; the T164I mutation within the epinephrine binding pocket of β2-adrenergic receptor [42]; and the polymorphism within binding pocket of STI-571 to c-Abl [43] . Although multiple genes take part in immune response only the HLA genotypes are significantly associated with SJS [22] , [32] , [33] , [44] , [45] . Such a strong linkage suggests that a direct binding of SADR-causing drugs to MHC I may be the primary event . In the case of HLA-Cw*4 , all sulfonamides bind to the antigen presentation groove . Several “wet” observations support this direct binding model . Firstly , the presentation of the sulfamethoxazole ( SMX ) parent drug displayed a direct , noncovalent binding fashion to the MHC–peptide complex [46] . Secondly , the antigen peptide within the MHC I groove does not appear to be essential , since the elution of the peptide did not affect the presentation of SMX [47] , thus there might be competitive binding between the drugs and the peptide to the groove . Thirdly , von Greyerz et al [29] found that most T cell clones showed a “MHC-allele restricted drug-specific recognition” that was stimulated by the parent drug rather than its derivatives . Fourthly , Nassif [26] discovered that blister fluid T lymphocytes , which were derived from a patient suffering SMX-induced TEN , were cytotoxic only when SMX is present and the cells share HLA-Cw*4 . This HLA allele-drug specific cytotoxicity was confirmed in another study [48] , and can be abrogated with the change of a single residue in the groove . Finally , the S116Y within HLA-B*5701 was shown to hamper completely the presentation of abacavir [36] , suggesting that the drug itself , or a metabolite , might be accommodated in the groove and the residue is essential for this binding . All these facts corroborate the direct drug-protein interaction model in which the strong MHC allele-drug specificity could be best explained by a steric complementarity together with other strong non-covalent interactions between the drug molecule and the antigen presentation groove . The groove contains a variable region where MHC I molecules coded by thousands of HLA alleles differ and the antigen peptide-MHC I recognition takes place . In our model , a similar drug-MHC I recognition occurs when the drug binds to its specific “port” of a particular MHC allele at the variant region . This specificity could also be found at the genetic level . For example , severe cutaneous adverse reactions are found to be triggered by allopurinol in the presence of B*5801 [44]; abacavir-induced skin reaction requires the parallel genotype of B*5701 [32]; and carbamazepin-induced SJS is linked to B*1502 , but not to HLA-A*1101 [45] . All these markers have a pronounced predictive capability of SADRs , leading the U . S . FDA's recommendation for their implementation for personalized medicine . The growing body of evidence suggests that the direct chemical-protein binding may enable the identification of more promising markers for SADR genetics , especially for predicting the specific HLA alleles that may be responsible for other drug-induced hypersensitivity , and finally , for a better design of new drugs or the modification of existing drugs to prevent these unintended interactions . Unexpected drug-protein interactions should be explored with any available technologies such as drug affinity pull-down [9] or compare the similarities of pocket shapes between known drug target and the off-target [49] . However , in a drug affinity pull-down experiment , it remains a challenging task to prioritize the candidate proteins and to explain the biological significance when hundreds of proteins are identified to be compatible to the drug . The CPI methodology of finding the common SADR target could enlighten the design of these “wet” experiments . For example , a pull-down through a mixture of resins immobilized by different case drugs might help to enrich the common targets , while the follow-up pull-down using resins immobilized by control drugs might exclude the false positives . Though DOCK could tell true bindings from the unidentified ones , no strict assessment had been put forward as to the resolving power of its scoring functions to evaluate the degree of the interaction strength of CPI . The lack of human protein structures is another problem . An ideal chemical-protein interactome would include the 3D structures of all human proteins whose structural flexibility can be effectively handled by more advanced docking programs so that the binding affinity can be better estimated . This would lead to a whole structurome-wide study with more unexpected interactions to be identified . Because of the biases in the structural and functional coverage in PDB and our pocket preparing criteria , the 845 proteins might not be representative , some of which were even redundant . The lack of randomization might introduce biases in the statistical model . Instead , preparing a reference protein set from other structural databases , e . g . , choosing one representative structure for each SCOP super family [50] , might improve the model . However , our pocket preparing criteria could guarantee all the proteins were targetable . Preserving the redundancy also enabled us to disclose more flexibility information of the protein cavities . Although this protein set needed improvement , with these structures , we were still able to i ) highlight the true bindings from the unidentified bindings; ii ) identify the susceptibility gene ( HLA-Cw*4 ) for the SMX-induced TEN without any prior knowledge of the underlying mechanism; iii ) identify the candidate risk allele of the susceptibility gene ( HLA-B*5701 ) based on the direct drug-MHC I interaction model which had never been proposed in all drug-induced hypersensitivity models . In addition , many candidate SADR targets prioritized from the CPI tended to be linked to the known SADR mechanisms compared with the random selection . As “wet” techniques for building up the CPI is not mature , the in silico approach appears to be the only means feasible to apply the CPI at the prescreening level , considering the urgency of the global SADR problem . Though false positive candidate genes might also be proposed , they can be controlled in different data transformation steps . Firstly , if one ligand tends to give low docking scores , it is usually caused by the ligand factor which could be eliminated in the normalization steps . Secondly , if one target tends to give low docking scores , it could raise low score for both case and control drugs . So this target cannot achieve a low p value in Fisher's exact test , and cannot be highlighted in the CPI . Thirdly , the false positive given by p value judgment could be controlled by the FDR correction . Lastly , they can be eliminated through association studies of the SADR patients or the functional studies of the SADR mechanisms , just as the docking procedure for identifying drug candidates is always followed by the binding affinity experiments . One limitation of our SVM classification model is the number of control samples in SJS/TEN group . We did not managed to find enough SJS- drugs to construct a sample set with case-control ratio at 1∶1 . SJS seems to be particular that only a few drugs do not linked with it . We could not find enough independent validation set either due to our strict criteria of the sample selection . This is the first try of using CPI profile to predict ADR outcome of a drug , the prediction performance of the CPI thus needs further validation and the model needs to be optimized . However , the permutation result added some reliability to the classification model . Compared with the permutation result , the classification result showed there were similarities in the CPI profiles of the case drugs , which provided hint for the construction of the prediction model based on this methodology . Although case drug molecules in Fig . 2B shared some structure similarity , the high-dimension information provided by CPI extends beyond simple structural comparison . Take fluoxetine 4 and fluoxetine in Fig . 2A for example , only a simple change in the ionization state between the two molecules will give different CPI profiles , with the distance between the two molecules far on the clustering tree . The chemical-protein interactomic analysis might also be complementary to trascriptomics in toxicogenomics . The latter provides a rich description of cell status [51] , whereas the CPI strategy seems to provide more direct and interpretable biological understandings . The primary interactions of a drug to proteins are the causes of biological events , whereas the trascriptome strategy only detects the resulting phenotypes . Knowing which proteins' function are disabled and which alleles tend to be disabled by the drugs are vital , for they might lay a direct solution to SADR at the source . Usually we could only access “wild-type” protein structures , and it is not only possible but also necessary to simulate genetic variability in 3D structures and thereby discover patient-specific off-targets so that we can predict one's SADR or drug effect from their “structypes” . However , we cannot construct a CPI that only uses the modeled “structypes” , since the type I error might accumulate significantly . So our strategy is to find proteins tend to be highlighted in a CPI containing “wild-type” structures , and then investigate whether the allele of the highlighted ones could constitute a risk allele . The SADR genetics requires worldwide collaboration [52] , [53] . At a time when samples are rare and the primary mechanisms are unclear , both the sample information and the hypothesized candidate genes should be shared by the community for prospective experimental validation . Thus , in anticipation of this global collaboration , we made the statistically significant candidate genes highlighted from CPI ( Table S6 ) publicly available for SADR consortia's consideration , verification and the improvement of CPI methodology . We expect that the perfection of CPI will eventually benefit the public by understanding , minimizing and predicting the occurrence of SADRs .
We selected targets from literature and from third-party targetable protein databases [14]–[17] . Every pocket had been examined manually and was screened according to the criteria pre-defined: I ) In order to identify unexpected drug-protein interactions , the target space should not be confined to the narrow space of the targets for the marketed drugs , which are merely a small portion of all protein spaces . II ) To utilize the structure space to the max , the species of the protein should not be confined to Homo Sapiens , and the homolog protein can be considered . Targets with greater than 30% in protein sequence similarity to the corresponding human protein at the bioactive site could be chosen . III ) The PDB structure should contain the co-crystallized ligand to define the bioactive site and to indicate the protein is targetable , PDB entries whose ligand is at the surface of the protein are not acceptable . IV ) The ligand embedded in the PDB structure should achieve certain rigidity and some specificity towards the target , e . g . , compounds with a large portion of rotatable bonds were not acceptable . V ) No missing residue should be around the bioactive site . Residues within 10 Å departed from the ligand were defined as the bioactive pocket of the protein , and balls with radius ranging from 1 . 1–1 . 4 Å were generated to fill in the pocket . Grid box was made 3–5 Å departed from the ‘cloud’ of the balls . The SMILES information of the small molecules was retrieved from PubChem . The minimal energy conformations of chemicals were generated with CORINA . All structures of proteins and chemicals were prepared using Chimera [54] and PyMOL . All the above procedures were performed manually with a strict quality control . An intersection operation was performed between all drug targets in DrugBank and the proteins in our pocket set using PDB id or Uniprot id as the identifier . We confirmed that each of the proteins in intersection had at least two FDA-approved drugs bind directly to them with a clear pharmacology to insure that they were credible drug targets . The running of the DOCK program and the extraction of the results were controlled by Perl or shell scripts on a Ubuntu™ Linux cluster . The docking score was calculated as the sum of intramolecular and intermolecular energy . A docking score greater than zero was treated as a missing value . There were a total of 845 proteins with 891 pockets in our pocket set . When a protein has multiple pockets to bind with , only the lowest docking score was chosen as the reference score , so the docking score matrix was shrunk to 46×845 elements . Here Xij represents the docking score of drug j to protein i . The Z-score is calculated as:wherewhere Nj equals 845 minus the number of missing docking value of drug j to the protein set . So a Z-score matrix of 46×845 elements was generated . All direct bindings verified in literature or in the “description , indication , pharmacology and mechanism_of_action” fields of DrugBank database were defined as group 1 , whereas other unidentified bindings were designated as group 2 . The nonparametric Mann-Whitney test was performed on the Z-scores . All accessible AERS raw data from Jan 2004 to Mar 2008 were downloaded from FDA website and then deposited into a relational database ( MySQL 5 . 1 . 22 ) . In an SADR report , only the primary or the secondary suspected drugs were regarded as linked to the reported SADR . Drugs reported in the literature to have caused a SADR were further examined in AERS; only those reported in the AERS more than three times were considered as case drugs . The candidates for control drugs were collected on condition that they had never been co-cited with this SADR in the literature . They were classified into control group only if the report number was zero or less than 5% of its total reports when jointly used with the case drugs . The Z-score matrix was generated using the same pipeline as described above . However , for the SADR oriented-CPI , the Z-score matrix was further trimmed using the formula:where Zij was the interaction strength between drug j and protein i . The Pearson correlation coefficient r between the two columns of binding affinity for drug j1 and drug j2 was measured as:where N represents the number of Z′ value pairs of drug j1 and drug j2 with no missing value . and are the average Z′-score of drug j1 and j2 , whereas and are standard derivation of Z′-score of the two drugs . The hierarchical clustering was then performed based on the r values between each pair of drugs . In the trimming procedure within each sub-CPI after clustering , only molecule with the lowest mean Z-score was chosen when multiple forms of a drug were clustered into the same sub-CPI . For protein i , ai , bi , ci , di values , representing the number of binding ( ai or bi ) and non-binding ( ci or di ) by case drugs or control drugs respectively , were counted and the relative risk ( RR ) value was calculated as follows:Protein targets with a RR value exceeding 1 were chosen for Fisher's exact test , as the expected values in any of the ai , bi , ci , di value sometimes is below 10 . The p values were then corrected using fdrtool package [55] of R to control the false positives . Although a more reliable network might be constructed using the STRING server [56] , it is still uncertain that whether its physical and functional relationships can denote the true situation in SADR pathology . However , it could be more certain that a gene is involved in an SADR if it is cited specifically in this SADR related literature . So we only chose GCCN to organize the SADR related knowledge in a gene-oriented fashion . Four sets of PubMed entries were retrieved through the following four querying terms: Cholestasis; Deafness OR “hearing loss”; rhabdomyolysis OR myalgia OR myopathy OR myositis; rash OR Stevens-Johnson syndrome OR toxic epidermal necrolysis . The records were downloaded in XML format through the eSearch and eFetch APIs , and were deposited into a MySQL database . The index file of human genes to PubMed entries was downloaded from the Entrez Gene ftp site . A core gene of SADR “S” must meet one of the two criteria: the citation rate of this gene in this “S”-related corpus must exceed its citation rate in corpus under other topics; the number of citation must exceed four . A connection between two genes was established in GCCN if they were co-cited in more than two PubMed entries . Both core genes and extended genes of “S” were included in enrichment analysis of biological process ( BP ) GO terms using EASE [57] . The p value generated from Bonferroni correction was used as a measure in choosing the significant BP terms . The GO terms of each candidate proteins highlighted from CPI were assigned through querying the Gene Ontology Annotation database [58] with their UniProt ID as the identifier . Attributes of candidate class ( I or II ) were assigned manually depending on whether the candidate targets had shared the BP terms enriched from the GCCN . For each SADR , categorical attribute of a drug was labeled as “1” or “0” if it could ( SADR+ ) or could not ( SADR− ) trigger this SADR . Z-scores were chosen as a measure of the interaction strength . Then for each protein , Z-scores were linearly scaled to the range of [−1 , 1] . We chose a nonlinear RBF kernel function to build the model , because the relations between class labels and interactome attributes are nonlinear , and this kernel function , could map vectors onto a higher dimensional space nonlinearly . Here C and γ were two essential parameters for RBF function , but it was not known beforehand that which C and γ fitted best for our model . We performed the exhaustive searching for the best ( C , γ ) pair each time we performed the 5-fold cross-validation test . | Why do tragedies caused by Vioxx or Avandia only happen to certain individuals ? The unexpected bindings among drugs and human proteins might play important roles in such serious adverse drug reactions ( SADRs ) . To mine these unexpected chemical-protein interactions , 162 drug molecules known to cause SADRs are ‘hybridized’ onto 845 proteins to construct a chemical-protein interaction matrix , from which two aspects of the information , the binding strength and the binding conformation , are disclosed . Followed by the data-mining strategies , the unexpected bindings that mediate SADRs are identified . For example , abacavir is found to bind to the antigen presentation groove of MHC I molecule in patients carrying the B*5701 allele but not B*5703 , which explains why HLA-B*5701 , not B*5703 , is the risk allele of abacavir hypersensitivity . This research could explain to the public that SADR happens when some of the innocent proteins are attacked by drugs unexpectedly , and variances in certain people's genome make their proteins more sensitive to the drug . By pre-therapy screening , the susceptible people could be protected . Furthermore , new drugs or modified drugs will be designed to avoid these patient-specific unintended bindings , in a step toward realizing personalized medicine . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/pharmacogenomics",
"computational",
"biology/macromolecular",
"structure",
"analysis",
"biochemistry/bioinformatics",
"pharmacology/adverse",
"reactions"
] | 2009 | Harvesting Candidate Genes Responsible for Serious Adverse Drug Reactions from a Chemical-Protein Interactome |
In spite of the importance of hyaluronan in host protection against infectious organisms in the alveolar spaces , its role in mycobacterial infection is unknown . In a previous study , we found that mycobacteria interact with hyaluronan on lung epithelial cells . Here , we have analyzed the role of hyaluronan after mycobacterial infection was established and found that pathogenic mycobacteria can grow by utilizing hyaluronan as a carbon source . Both mouse and human possess 3 kinds of hyaluronan synthases ( HAS ) , designated HAS1 , HAS2 , and HAS3 . Utilizing individual HAS-transfected cells , we show that HAS1 and HAS3 but not HAS2 support growth of mycobacteria . We found that the major hyaluronan synthase expressed in the lung is HAS1 , and that its expression was increased after infection with Mycobacterium tuberculosis . Histochemical analysis demonstrated that hyaluronan profoundly accumulated in the granulomatous legion of the lungs in M . tuberculosis-infected mice and rhesus monkeys that died from tuberculosis . We detected hyaluronidase activity in the lysate of mycobacteria and showed that it was critical for hyaluronan-dependent extracellular growth . Finally , we showed that L-Ascorbic acid 6-hexadecanoate , a hyaluronidase inhibitor , suppressed growth of mycobacteria in vivo . Taken together , our data show that pathogenic mycobacteria exploit an intrinsic host-protective molecule , hyaluronan , to grow in the respiratory tract and demonstrate the potential usefulness of hyaluronidase inhibitors against mycobacterial diseases .
Infectious diseases caused by mycobacteria are serious threats to human health . Tuberculosis is caused by infection with mycobacteria , most frequently with Mycobacterium tuberculosis but also with Mycobacterium bovis , Mycobacterium africanum , Mycobacterium microti , and Mycobacterium canetii and kills around 2 million people annually . Leprosy is caused by Mycobacterium leprae and the globally registered prevalence of leprosy was around 22 , 000 cases at the beginning of 2006 . The major portal of entry for mycobacterial pathogens is through the respiratory tract . The primary phase of the infection begins with inhalation of bacteria , which are then phagocytosed by alveolar macrophages in the periphery of the lungs . In addition , several lines of evidence indicate that mycobacteria interact with epithelial cells in the respiratory tract [1]–[4] . The recent reports show the significant role of type II pneumocytes in the pathology of tuberculosis [3] , [5] , [6] . The onset of mycobacterial diseases frequently occurs after a long latent phase . Mycobacteria are an intracellular bacterium , multiplying within host cells , but also grow extracellularly [7] , [8] . Macrophages phagocytose mycobacteria through interaction with several cell surface receptors , including complement receptors , mannose receptors , surfactant protein A , scavenger receptors , and Fc receptors [9] . By contrast , mycobacteria attaches to or invades lung epithelial cells through interactions with glycosaminoglycans ( GAG ) [10] . M . tuberculosis , M . bovis bacillus Calmette-Guerin ( BCG ) , and M . leprae produce two types of GAG interacting adhesins , heparin-binding hemagglutinin ( HBHA ) [10] , [11] and mycobacterial DNA-binding protein 1 ( MDP1 , also called histone-like protein and laminin-binding protein in M . leprae ) [1] , [12] . HBHA is secreted to the extracellular milieu from mycobacteria [13] , whereas MDP1 is tightly attached on the mycobacterial cell wall [14] . We previously demonstrated that hyaluronan is a major portal for infection of mycobacteria into A549 human lung epithelial cells by interacting with MDP1 [1] . Hyaluronan is a nonsulfated linear GAG composed of thousands of repeating units of GlcNAc- ( beta-1 , 4 ) -GlcUA- ( beta-1 , 3 ) and is synthesized by 3 isoforms of hyaluronan synthases ( HAS ) , designated HAS1 , HAS2 , and HAS3 in both mice and humans [15]–[18] . In vertebrates , hyaluronan is a ubiquitous structural component of the extracellular matrix , and is abundant in the chondral and vitreous tissues . Recent findings demonstrated that hyaluronan has a pivotal role in diverse dynamic biological functions such as embryonic development [19] , cell migration [20] , [21] , tumor transformation , [22] , [23] , wound healing [24] , and inflammation [25]–[27] . On the mucosal surface of the airway , hyaluronan retains bactericidal enzymes so that they are “ready-to-use” , protecting mucosal tissues from invading pathogens [28] . Furthermore , in the alveolar tracts , released fragmented HA stimulates innate immune responses by activating Toll-like receptor 2 and 4 dependent pathways and initiating lung inflammation [25] . By contrast , during resolution of respiratory inflammation , immuno-stimulatory hyaluronan is taken up via the hyaluronan receptor CD44 on alveolar macrophages [26] . Thus hyaluronan plays a pivotal role in host defenses in the respiratory tract , but its role in mycobacterial infection had not been elucidated so far . In this study , we analyzed the role of hyaluronan after mycobacterial infection was established .
A549 cells , a type II human lung epithelial cell line , were exposed to recombinant BCG expressing luciferase ( rBCG-Luc ) under the control of the HSP60 promoter [14] at a multiplicity of infection ( MOI ) of 10 for 16 hours . Cells were then washed and various doses of hyaluronan added into the culture . Growth of BCG was monitored by luciferase activity at each time point , which is indicative of viable bacteria [14] , [29] . We found that exogenously added hyaluronan enhances bacterial growth in a dose-dependent manner ( Figure 1A ) . We also confirmed this effect by counting viable bacteria using a colony forming units ( CFU ) assay ( Figure 1C ) . In our experimental setting , around 60% of the bacteria adhere to the cell surface and the remaining 40% are internalized by the cells [1] . Therefore , we next examined whether hyaluronan enhances extracellular or intracellular growth by treatment with gentamicin , which kills extracellular but not intracellular bacteria . After infection , we added gentamicin ( 50 µg/ml ) into the culture for 6 hours and then added hyaluronan after removing gentamicin . The results showed that gentamicin treatment abrogated the growth of BCG ( Figure 1B ) , indicating that bacterial growth occurred extracellularly . The enhanced effect of hyaluronan on bacterial growth was also abolished by gentamicin treatment ( Figure 1B ) . This suggests that hyaluronan enhances growth of BCG attached to these cells . We next examined if the same effects of hyaluronan can be seen in M . tuberculosis growth after infection to A549 cells . We infected M . tuberculosis H37Rv to A549 cells , then added hyaluronan , and monitored growth by counting colony-forming units ( CFU ) . Similar to the case of BCG , we found that presence of hyaluronan enhances the growth of M . tuberculosis in a dose dependent manner ( Figure 1D ) . Gentamicin treatment also abrogated the growth of M . tuberculosis and growth-enhancing effect of hyaluronan . To determine why hyaluronan enhances the growth of BCG , we hypothesized that BCG can utilize it as a carbon source because hyaluronan is a polymer of disaccharides . We cultured BCG-Luc in 7H9 based carbon-starved broth in the presence ( 0 . 5 mg/ml ) or absence of hyaluronan . As expected , in the carbon-starved media BCG did not grow , while the addition of hyaluronan supported the growth of BCG ( Figure 2A ) , demonstrating that BCG can utilize hyaluronan as a carbon source . We next compared hyaluronan with other GAG in terms of their growth supporting effect . BCG-Luc was cultured in 7H9-based carbon starved media or media including 0 . 5 mg/ml of each GAG as the sole carbon source . The results showed that BCG did not grow in the media supplemented with heparin or heparan sulfate . Both hyaluronan and chondroitin sulfate encouraged the growth , but hyaluronan sustained higher growth rates of BCG than chondroitin sulfate ( Figure 2A ) . We also demonstrated that the growth supporting effect of hyaluronan is comparable to an equivalent amount of glucose ( 0 . 5 mg/ml ) ( Figure 2B ) . In order to evaluate uptake of hyaluronan during hyaluronan-dependent growth of mycobacteria , we cultured BCG in the presence of 3H-labeled hyaluronan in the media containing hyaluronan as a sole carbon source . As shown in Figure 2C , live BCG incorporated hyaluronan , whereas heat-killed bacteria did not , showing actual uptake of hyaluronan into bacteria . We next assessed the action of hyaluronan in the growth of virulent M . tuberculosis ( strain H37Rv ) , and environmental mycobacterial species such as M . smegmatis ( strain mc2155 ) and M . avium ( ATCC25291 ) . In carbon-starved media , none of the three strains grew . However , M . tuberculosis H37Rv , along with BCG , multiplied in the media containing hyaluronan as a sole carbon source while neither M . smegmatis nor M . avium proliferated . After 12 days culture , optimal density ( OD ) at 630 nm of M . tuberculosis culture increased to 0 . 32±0 . 038 from 0 . 01 ( day 0 ) . We then compared hyaluronan and other GAGs in terms of growth supportive effects on M . tuberculosis . Similar to the case of BCG , hyaluronan most effectively enhanced the growth of M . tuberculosis among tested GAGs ( Figure 3 ) . Because hyaluronan is a long chain consisting of the repeat of two monosaccharides at over 2×105 Da , we hypothesized that extracellular cleavage of the polymer would be required before taken up by cells . Therefore , we next assessed hyaluronidase activity in mycobacteria . Hyaluronan was incubated in the presence or absence of cell lysates derived from BCG before precipitation by phenol/chloroform extraction . Precipitates were then fractionated by polyacrylamide gel electrophoresis ( PAGE ) and visualized by alcian blue staining as described previously [30] . Hyaluronan was separated into discrete ladder-like bands by electrophoresis after incubation with BCG lysate ( Figure 4A ) , demonstrating that BCG possesses hyaluronidase activity . We then addressed whether hyaluronidase activity is crucial for hyaluronan -dependent growth of mycobacteria . L-Ascorbic acid 6-hexadecanoate ( Vcpal ) is shown to be a potent inhibitor of hyaluronidase [31] . We investigated the effect of Vcpal on hyaluronidase activity of BCG and found that hyaluronidase activity was abolished in the presence of 25 µM Vcpal ( Figure 4A , lane 4 ) . We next examined the effects of Vcpal on the growth of BCG . BCG-Luc was cultured in modified 7H9 media containing hyaluronan ( 0 . 5 mg/L ) as the sole carbon source or 7H9-ADC complete media , which contains Tween 80 , glycerol , and dextrose as carbon sources and BSA . We found that 25 µM Vcpal did not change the growth rate of BCG in 7H9-ADC complete media , while it abolished the growth of BCG in the media containing hyaluronan as the sole carbon source ( Figure 4B ) . We also examined the effect of Vcpal on the growth of M . tuberculosis . M . tuberculosis H37Rv was cultured in the media with or without Vcpal ( 50 and 100 µM ) . Vcpal suppressed the growth of M . tuberculosis in the media containing hyaluronan as a sole carbon source but not the growth in conventional 7H9-ADC media ( Figure 4C ) . Other hyaluronidase inhibitors , such as apigenin and quercetin [32] , also inhibited hyaluronan dependent growth of M . tuberculosis as shown in Figure S1 . These results indicate that hyaluronidase activity is essential for both BCG and M . tuberculosis when utilizing hyaluronan as a carbon source . We next examined whether Vcpal suppresses the enhancing effect of hyaluronan on the growth of BCG after attachment to A549 epithelial cells . After exposure to BCG-Luc , hyaluronan was added with or without Vcpal ( 25 µM ) into the culture and growth of BCG was monitored by measuring luciferase activity . After 6 days culture , RLU values of BCG-Luc increased to 36 . 6±7 . 5 RLU or 52 . 6±18 . 7 RLU in the absence or presence of hyaluronan , respectably . Adding Vcpal abrogated the enhanced effects of hyaluronan ( 29 . 3±2 RLU ) , demonstrating that BCG utilized exogenously added hyaluronan as a carbon source after infection to A549 cells . This work so far on the growth of mycobacteria has been performed with hyaluronan purified from human umbilical cord ( Sigma ) . In order to elucidate whether mycobacteria can use hyaluronan actually synthesized in situ by mammalian cells , we employed the previously established stable human HAS1–3 expressing rat 3Y1 fibroblasts [15] . 3Y1 rat fibroblasts do not produce detectable hyaluronan themselves but each transfectant produces different sized hyaluronan . Both HAS1 and HAS3 transfectants secret hyaluronan with broad size distributions with molecular masses between 2×105 to ∼2×106 Da , while the HAS2 transfectant secretes extremely large hyaluronan at an average molecular mass of >2×106 Da [15] . We analyzed the level of hyaluronan production by utilizing a hyaluronan-binding protein ( HABP ) -based ELISA assay and confirmed that the HAS2 transfectant produced high levels of hyaluronan ( 235 . 7 µg/mL in the culture media ) , while the HAS3 transfectant synthesized the smallest amount of hyaluronan ( 15 . 9 µg/mL ) . The HAS1 transfectant produced moderate levels of hyaluronan ( 85 . 3 µg/mL ) , and the empty vector transfectant did not produce detectable amounts of hyaluronan . Each human HAS transfectant was exposed to BCG-Luc and the growth kinetics of the bacteria were monitored by luciferase activity . The results showed that BCG grew after attachment to 3Y1 cells transfected with HAS1 and HAS3 but not with HAS2 or empty vector ( Figure 5A ) . In addition , we found that hyaluronidase treatment of HAS1 transfected cells enhanced the growth of BCG ( Figure 5B ) . These results suggest that shorter sized chains of hyaluronan are preferential for BCG growth . We also monitored the growth of M . tuberculosis H37Rv after infection to these HAS transfectant cells . Along with the case of BCG , HAS1 and HAS3 but not HAS2-tranfectants supported the growth of M . tuberculosis ( Figure 5C ) . To see if hyaluronan is present at the site of infection of M . tuberculosis , we assessed the expression of hyaluronan synthases ( HAS1 , HAS2 , and HAS3 ) in the lungs of BALB/c mice infected with the M . tuberculosis H37Rv strain , using the low-dose aerosol infection model . Total RNA was extracted from the lungs after 1 , 3 , 5 , 7 , 14 , and 21 days of infection , and analyzed for HAS1 , HAS2 , and HAS3 mRNA transcription by reverse transcriptase-polymerase chain reaction ( RT-PCR ) ( Figure 6A ) . The data showed that HAS1 mRNA expression increased after infection and was maintained at all time points ( Figure 6A ) . We next determined if hyaluronan is present in alveoli using biotin-conjugated hyaluronan-binding protein ( HABP ) and histochemical analysis . Before infection , hyaluronan was located on the surface of the airways and alveoli ( Figure 6B ) . After M . tuberculosis infection , hyaluronan levels were profoundly increased and accumulated in the granulomatous legion ( Figure 6B ) . Taken together , these data indicate that the major hyaluronan synthase in the lungs is HAS1 both before and after M . tuberculosis infection and hyaluronan accumulates in the tuberculosis lesion . M . tuberculosis-infected mice had numerous sites of granulomatous inflammation in their lungs but in primates , tuberculosis granulomas are well-organized and tighter . We next studied hyaluronan in the lung granuloma of M . tuberculosis H37Rv-infected rhesus monkeys by staining with alcian blue , which is commonly used dye to detect GAG . The dye stained the surrounding region of well-organized granuloma ( Figure 7A ) and the staining was largely abolished by treatment with hyaluronidase ( Figure 7B ) , showing that hyaluronan is a major GAG surrounding granuloma . Acid-fast bacilli ( arrow heads in Figure 7C ) were located in alcian blue stained areas , thus suggesting a strong correlation between the localization of the tubercle bacilli and hyaluronan . Finally , we addressed the effect of Vcpal on the growth of BCG in BALB/c mice . Mice were infected with BCG intravenously through their tail veins . One day after BCG challenge , the hyaluronidase inhibitor Vcpal ( 0 . 4 or 1 . 64 mg/dose ) was injected every day thorough the tail veins for 14 days . Two days after the final injection , the mice were euthanized and viable bacteria counts were determined by the CFU assay . As a positive control , we also treated mice with amikacin ( Amk ) , which kills extracellular but not intracellular mycobacteria , by an intramuscular injection . The results showed that Vcpal apparently suppressed growth of BCG in the lungs , similar to Amk ( Figure 8 ) .
Although hyaluronan is crucial for both structural and physiological properties in the alveolar spaces , its role in mycobacterial infection was previously unknown . We demonstrated before that hyaluronan is the major attachment site of both BCG and M . tuberculosis in the infection of A549 cells , which itself produced hyaluronan [1] probably depending on HAS3 and HAS2 ( Figure S2 ) . In this study , we further extended our research and studied the role of hyaluronan after infection was established . First , we examined the effect of hyaluronan on the growth of BCG after infection of A549 cells . BCG is an attenuated strain of the virulent M . bovis and is a live vaccine against tuberculosis . Because BCG bacilli share biological and pathological characteristics [33] and over 99 . 5% of their genome with that of M . tuberculosis [34] , BCG is frequently utilized for the analysis of virulence of M . tuberculosis . Utilizing BCG , we first found that exogenously added hyaluronan enhances the growth of BCG after incubation with A549 cells . We found that gentamicin treatment abrogated the growth of both BCG and M . tuberculosis , showing that these mycobacteria grow outside A549 cells . By contrast , this BCG strain ( Pasteur ) and M . tuberculosis H37Rv grew inside J774 mouse macrophages . These data apparently suggest that intracellular spaces in A549 cells are not suitable for the growth of mycobacteria . Mycobacteria are intracellular pathogens and survive in macrophages by blocking phagosome-lysosome fusion ( P-L fusion ) at the stage of Rab5–Rab7 conversion [35]–[37] . Mycobacteria can infect non-professional epithelial cells in addition to alveolar macrophages . However , the exact mechanisms of how mycobacteria invade and persist or are killed in epithelial cells are unknown . Clemens and Horwitz demonstrated that mycobacterial phagosomes acquired Rab7 in HeLa epithelial cells , suggesting that P-L fusion is not efficiently blocked . Furthermore , Takeda's group recently found that type II pneumocytes produce antimicrobial peptides , secretory leukocyte protease inhibitor and Lipocalin 2 , which have potent anti-mycobactericidal activities [5] , [6] . Such bactericidal molecules may contribute to the inhibition of intracellular growth of mycobacteria within type II pneumocytes . These data suggest that intracellular trafficking of mycobacteria-containing vacuoles and intracellular states of mycobacteria are different from that in macrophages . We found that both BCG and M . tuberculosis grew in the media containing hyaluronan as the sole carbon source ( Figure 2A and 3 ) . In addition to hyaluronan , mammals synthesize several GAGs , but hyaluronan most strongly supported the growth of BCG among GAGs and is comparable with glucose ( Figure 2 ) . By contrast , environmental mycobacteria , such as M . smegmatis and M . avium , failed to use hyaluronan as a carbon source . These data help us to understand why pathogenic mycobacteria have the ability to adhere to hyaluronan and metabolize it . It is reasonable to assume that this property is a great advantage , allowing them to grow in the hyaluronan-rich respiratory organs of their hosts . Because hyaluronan is a long carbon chain , we considered that cleavage must be an essential step for its use as a carbon source , and indeed found hyaluronidase activity in BCG ( Figure 4 ) . Although certain other species of bacterial pathogens , such as Streptococcus , Staphylococcus , and Streptomyces , produce hyaluronidases [38] , there has been no report of hyaluronidase of mycobacteria . This is the first report showing hyaluronidase activity in mycobacteria . There are two main groups of hyaluronidases identified to date . One group is endo-β-N-acetyl-hexosaminidase or endo-β-glucuronidase , which degrades hyaluronan by hydrolysis [39] . These enzymes are distributed in some vertebrates including mouse and human . Others are lyase type hyaluronidase that degrade hyaluronan by β-elimination [39] . Bacterial hyaluronidases are lyases , which are unstable but have stronger activity than those of vertebrates , and generate unsaturated products , which is more suitable for energy supply than saturated hyaluronan . Therefore , it is reasonable to consider that mycobacteria have the lyase type of hyaluronidase . Although hyaluronidase is not yet described in the genome of either M . tuberculosis [33] or BCG [34] , there are approximately 40 lyases . One of these lyases may be responsible for degradation of hyaluronan . Defining which enzyme is responsible for cleavage of hyaluronan is next important issue . Most hyaluronidases in mammals and bacteria display redundancy in recognition of their GAG substrates . Our data show that chondroitin sulfate also supported the growth of BCG ( Figure 2 ) . This may imply that hyaluronidase ( s ) of BCG cleave chondroitin sulfate as well . Hyaluronan possesses many properties in vivo and it is believed that these biological activities are dependent on its size [40]–[42] . Although hyaluronan is composed of simple repeating disaccharides , its secondary structure is flexible . It is affected by the numbers of intramolecular hydrogen bonds , their location , and hydrophobic interactions [43] , [44] , all of which are increased as the size of the chains increase . Dynamic laser light-scattering analysis showed that the rod-like structure of low molecular weight hyaluronan changes to a stiff coil structure beyond a molecular weight of 1×105 Da [45] . Taken together , it is conceivable that hyaluronan synthesized by HAS1 and HAS3 exhibits a different structure from that synthesized by HAS2 . Employing HAS transfectants , we found that both BCG and M . tuberculosis utilize hyaluronan synthesized only by HAS1 or HAS3 for multiplication ( Figure 5A and 5C ) . The fact that BCG and M . tuberculosis grow when co-cultured with HAS1 and HAS3 but not HAS2 transfected cells ( Figure 5A and 5C ) suggests that HAS1 and HAS3-synthesized hyaluronan supports the growth of mycobacteria in the human body . We founds that HAS1 is the major hyaluronan synthase in M . tuberculosis-infected mouse lungs ( Figure 6A ) . HAS1 is expressed in immune cells , such as dendritic cells and T cells [46] . To clarify what kind of cell expresses HAS1 during mycobacterial infection is the next important issue . In spite of the importance of hyaluronan in host protection in the lungs , its role in mycobacterial diseases had not been elucidated . In this study , we demonstrated that BCG and M . tuberculosis can utilize it as a carbon source . Hyaluronan was observed in the granulomatous region of mice lungs infected with M . tuberculosis ( Figure 6 ) . Furthermore , M . tuberculosis bacilli were residing in the region where hyaluronan was located in the lungs of monkeys that had died from tuberculosis ( Figure 7 ) . We also showed that blocking hyaluronidase inhibited in vivo multiplication of BCG ( Figure 8 ) . These results suggest that pathogenic mycobacteria have evolved to exploit the intrinsically host-protective molecule , hyaluronan as a nutrient to grow . Similar behavior of pathogenic mycobacteria was observed during infection of macrophages , that is , BCG is phagocytized in a cholesterol-dependent manner [47] and utilizes cholesterol as a carbon source to survive in activated macrophages [48] . It is likely that mycobacteria developed several strategies to obtain nutrients under nutrient-limited conditions . After digestion of hyaluronan , it must be incorporated into mycobacteria through specific receptors or membrane proteins . Based on our results and consideration , hyaluronidase and a potential transporter of fragmented hyaluronan of pathogenic mycobacteria are potential drug targets .
All animals were maintained under specific pathogen-free conditions in the animal facilities of Osaka City University Graduate School of Medicine and in a biosafety-level-3 facility at The Research Institute of Tuberculosis according to the standard guidelines for animal experiments at each institute . RPMI 1640 media , L-glutamine , fetal bovine serum , HEPES , hyaluronan from human umbilical cord , heparin from porcine intestinal mucosa and heparan sulfate from bovine kidney were purchased from Sigma-Aldrich ( St . Louis , MO ) . Chondroitin sulfate A and C were purchased from Calbiochem ( Gibbstown , NJ ) . For conventional culture of mycobacteria , Middlebrook 7H9 medium ( Becton Dickinson ) supplemented with 0 . 085% NaCl , 10% albumin-dextrose-catalase ( BD Biosciences ) , 0 . 2% glycerol , and 0 . 05% Tween 80 ( 7H9-ADC ) or 7H11-agar supplemented with 0 . 085% NaCl , 10% oleic acid-albumin-dextrose-catalase ( BD Biosciences ) , and 0 . 2% glycerol ( 7H11-OADC ) were used . 7H9 medium ( Becton Dickinson ) supplemented with 0 . 085% NaCl and 0 . 1% albumin was used as a carbon-starved 7H9 medium . A549 cells were grown in RPMI 1640 medium containing 10% heat-inactivated fetal bovine serum , 2 mM L-glutamine , 25 mM HEPES and 5 . 5×10−5 M 2-mercaptoethanol ( complete culture medium ) at 37°C in an atmosphere of 5% CO2 . Cells were suspended at 2×105/ml in complete culture medium and 1 ml of cell suspension was dispensed into individual wells of a 24-well polystyrene plate ( BD Biosciences , San Jose , CA ) . Plates were incubated at 37°C for 24 h and were washed with serum-free RPMI 1640 medium to remove nonadherent cells . Wells were then refilled with 1 ml of complete culture medium . M . bovis BCG or M . tuberculosis cell suspension was prepared as described previously [1] . The bacterial cell suspension was added to A549 cells at multiplicities of infection ( MOI ) of 10 . After 16 ( BCG ) or 4 ( M . tuberculosis ) h incubation , unbound bacteria were removed by washing with serum-free RPMI 1640 three times . After adding 1 ml of fresh complete culture medium to each well , hyaluronan solution was added to final concentrations ranging from 5 to 500 µg/ml . Cells were collected periodically for luciferase or CFU assays . Construction of BCG expressing luciferase was described previously [1] . Luciferase activity was measured using the luciferase assay system from Promega ( Madison , WI ) according to the manufacturer's protocol on a Wallac 1420 manager as described previously [14] . A549 cells in 96-well polystyrene plates ( 8×104/well ) were infected with BCG-Luc or M . tuberculosis at MOI of 10 at 37°C . After 16 ( BCG ) or 4 ( M . tuberculosis ) h , the monolayers were washed three times with RPMI 1640 medium to remove extracellular bacteria . Fresh complete culture medium containing 1 mg/ml of hyaluronan and 50 µg/ml of gentamicin were added to each well ( 200 µl/well ) and incubated at 37°C . Cells were collected periodically for detection of luciferase activity of BCG-Luc or CFU assay of M . tuberculosis . BCG-Luc or M . tuberculosis was adjusted to a concentration of 1×104 CFU/ml in carbon-starved 7H9 medium described previously [14] , and 200 µl of bacterial cell suspension was added to 96-well polystyrene plates . Heparin , heparan sulfate , chondroitin sulfate , hyaluronan or glucose was added to appropriate wells to a final concentration of 500 µg/ml . Plates were incubated at 37°C and bacterial cells were collected periodically for detection of luciferase activity of BCG-Luc or CFU assay of M . tuberculosis . BCG Pasteur was grown aerobically in 7H9-ADC medium at 37°C . Cells were then collected by centrifugation and half of the cells were heat-killed by heating at 65°C for 30 min . Then bacteria were washed , resuspended by carbon-starved 7H9 medium and adjusted to an optical density at 600 nm of 0 . 07 . One hundred microliters of cell suspension was added to 100 ml of carbon-starved 7H9 with or without 6 mg of 3H-labeled hyaluronan and 14 mg of non-labeled hyaluronan ( final concentration of 100 mg/L of total hyaluronan ) . Cells were then incubated at 37°C . After incubation , cells were harvested by use of a Scatron Harvester ( Scatron ) onto a glass fiber filter . The incorporated radioactivity was measured in a gamma counter ( ALOKA ARC-2000 ) . M . tuberculosis strain H37Rv , M . smegmatis strain mc2155 and M . avium strain type4 were grown in carbon-starved 7H9 medium containing 0 . 5 mg/ml of hyaluronan , and the cultures were monitored periodically for their optical density at 600 nm ( M . tuberculosis and M . smegmatis ) or CFU ( M . tuberculosis and M . avium ) . BCG was grown in 7H9-ADC medium to mid-log phase . After incubation , bacterial cells were harvested , washed three times with ice-cold PBS ( pH 6 . 0 ) and resuspended in the same buffer . To disrupt bacterial cells , the cell suspension was added to a screw-capped tube containing glass beads ( diameter , 1 . 0 mm ) and the tube was oscillated on a Mini-Bead Beater ( Cole-Parmer ) . The tube was centrifuged at 10 , 000×g for 10 min , and the supernatant containing the bacterial protein extract was collected into a new tube . The protein solution was then mixed with 1 mg/ml of hyaluronan in PBS ( pH 6 . 0 ) at 37°C . After incubation for 24 h , the solution was mixed with an equal volume of phenol to remove protein . The mixture was centrifuged at 10 , 000×g for 10 min and the supernatant was collected for PAGE analysis . PAGE analysis of hyaluronan was performed as previously described by Ikegami-Kawai et al . [30] with minor modifications . The PAGE mini-slab gels contained 12 . 5% acrylamide , 0 . 32% N , N′-methylene bis-acrylamide in 0 . 1 M Tris-borate-1 mM Na2EDTA ( TBE , pH 8 . 3 ) . For the electrophoretic run , samples containing hyaluronan were mixed with one-fifth volume of 2M sucrose in TBE and 10 µl of the mixtures was applied directly to the gel . Bromophenol blue in TBE containing 0 . 3 M sucrose was used as a tracking dye , but was generally applied to a well with no sample . The gels were electrophoresed at 300 V for approximately 70 min using TBE as a reservoir buffer . After electrophoresis , the gels were stained with alcian blue as described previously [30] . Briefly , the gels were soaked in 0 . 05% Alcian blue in distilled water for 30 min in the dark and destained in water for 30 min . BCG-Luc or M . tuberculosis H37Rv was suspended in 7H9-ADC , carbon-starved 7H9 or carbon-starved 7H9 containing 0 . 5 mg/ml of hyalurona to a final concentration of 1×104 CFU/ml and 200µl of each suspension was added to 96-well polystyrene plates . Vcpal was added to each well . Bacterial cells were then incubated at 37°C and were collected periodically for detection of luciferase activity for BCG-Luc or CFU assay for M . tuberculosis . Similarly , M . tuberculosis H37Rv was incubated in the media containing 0 . 5 mg/ml hyaluronan in presence or absence of 0 . 1 or 0 . 5 mM of apigenin or quercetin . After incubation for 7 days , living bacterial number was determined by CFU assay . The expression of hyaluronan synthase genes in the lung tissues of mice aerogenically challenged with the virulent M . tuberculosis strain H37Rv was determined by RT-PCR . Seven-week-old of female BALB/c mice were aerogenically infected with the M . tuberculosis strain H37Rv ( 2×102 CFU/mouse ) using a Glas-Col chamber . At different time points , 3 mice per group were euthanized and , the lungs were homogenized in PBS containing 0 . 05% Tween 80 . The homogenates were centrifuged , and the pellets were processed to isolate total RNA using the RNeasy mini kit ( QIAGEN , West Sussex , UK ) according to the manufacturer's instruction . One microgram of total RNA was reverse transcribed using Super Script II RNase H reverse transcriptase ( Invitrogen ) . The cDNA was then subjected to RT-PCR . The following primer pairs were used: β-actin , 5′-TGGAATCCTGTGGCATCCATGAAAC-3′ ( F ) and 5′-TAAACGCAGCAGCTCAGTAACAGTCCG-3′ ( R ) ; HAS1 , 5′-GCTCTATGGGGCGTTCCTC-3′ ( F ) and 5′-CACACATAAGTGGCAGGGTCC-3′ ( R ) ; HAS2 , 5′-TGGAACACCGGAAAATGAAGAAG-3′ ( F ) and 5′-GGACCGAGCCGTGTATTTAGTTGC-3′ ( R ) ; HAS3 , 5′-CCATGAGGCGGGTGAAGGAGAG-3′ ( F ) and 5′-ATGCGGCCACGGTAGAAAAGTTGT-3′ ( R ) . The amplification procedure involved initial denaturation at 94°C for 4 min followed by 35 cycles of denaturation at 94°C for 1 min , annealing of primers at 57°C for 1 min and primer extension at 72°C for 3 min . After completion of the 35th cycle , the extension reaction was continued for another 7 min at 72°C . Total RNA was extracted from A549 cells by RNeasy mini kit ( QIAGEN ) and then 1 µg of total RNA was reverse transcribed using Super Script II RNase H reverse transcriptase ( Invitrogen ) . The cDNA was then subjected to RT-PCR . The following primer pairs were used: β-actin , 5′-GATCATTGCTCCTCCTGAGC-3′ ( F ) and 5′-CACCTTCACCGTTCCAGTTT-3′ ( R ) ; HAS1 , 5′- ACTCGGACACAAGGTTGGAC -3′ ( F ) and 5′- TGTACAGCCACTCACGGAAG -3′ ( R ) ; HAS2 , 5′- ATGCATTGTGAGAGGTTTCT -3′ ( F ) and 5′- CCATGACAACTTTAATCCCAG -3′ ( R ) ; HAS3 , 5′- GACGACAGCCCTGCGTGT -3′ ( F ) and 5′- TTGAGGTCAGGGAAGGAGAT-3′ ( R ) . The amplification procedure involved initial denaturation at 94°C for 10 min followed by 40 cycles of denaturation at 94°C for 1 min , annealing of primers at 56°C for 1 min and primer extension at 72°C for 2 . 5 min . The M . tuberculosis H37Rv challenge infection study of in rhesus male monkeys was performed previously [49] . The lung of non-vaccinated monkeys that died of tuberculosis 3 month after intratracheal challenge of 3 , 000 CFU/lung of M . tuberculosis H37Rv were immediately removed and fixed with 15% formalin for 10 days . Three animals' lungs were embedded in paraffin blocks and used in this study as well . After deparaffinization by washing with xylene and ethanol , the tissue sections were washed in TBS and incubated with fresh TBE containing 0 . 05 mM of Pronase K ( Dako ) for 60 min at room temperature . After washing with TBS containing 1% bovine serum albumin , the slides were incubated with 3% bovine serum albumin in TBS for 30 min at room temperature to block non-specific binding sites . The slides were then washed with TBS twice for 10 min and incubated with the biotinylated hyaluronan-binding protein ( HABP ) probe at a concentration of 2 mg/ml in TBS for 60 min at room temperature . Following washing in TBS , the slides were incubated with a streptavidin-peroxidase reagent and the staining developed using DAKO Cytomation LSAB-system AP ( Dako ) . The slides were then washed with distilled water and counterstained with Mayer's hematoxylin . Paraffin sections were also stained with alcian blue ( Sigma ) pH 2 . 5 ( 3% acetic acid ) for 5 min . The slides were counterstained with nuclear fast red ( Biomeda ) and mounted with Gel/Mount ( Biomeda ) . For GAG digestion , 0 . 5 mg/ml ( 10 U/ml ) Streptomyces hyaluronidase was added for 30 min at 37°C before alcian blue staining . The slides were stained by Ziehl-Neelsen technique using carbol-fuchsin and malachite green ( Sigma ) . | Mycobacterium tuberculosis and Mycobacterium bovis are major bacterial pathogens that kill approximately 2 million people annually by causing tuberculosis . The M . tuberculosis complex has several strategies to parasitize the host . After infection is established , these pathogens are rarely eliminated from the host , and nowadays approximately a third of the world's human population is infected with the Mycobacterium tuberculosis complex . The elucidation of the parasitic mechanisms of the M . tuberculosis complex is important for the development of novel strategies against the disease . The major portal entry of M . tuberculosis complex is through the respiratory tract . On the surface of the airway , hyaluronan retains bactericidal enzymes so that they are “ready-to-use” , protecting tissues from invading pathogens . Furthermore , fragmented hyaluronan produced as a result of infection is used by the immune system as a sensor of infection . Thus , hyaluronan plays a pivotal role in host defenses in the respiratory tract . However , in this study , we observed that the M . tuberculosis complex utilizes hyaluronan as a carbon source for multiplication . We also found that the M . tuberculosis complex has hyaluronidase activity and showed that it is critical for hyaluronan-dependent growth of the M . tuberculosis complex . This study demonstrates a novel parasitic mechanism of the M . tuberculosis complex and suggests that mycobacterial hyaluronidase is a potential drug target . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbiology/medical",
"microbiology"
] | 2009 | Mycobacteria Exploit Host Hyaluronan for Efficient Extracellular Replication |
Functional magnetic resonance imaging ( fMRI ) measures brain activity by detecting the blood-oxygen-level dependent ( BOLD ) response to neural activity . The BOLD response depends on the neurovascular coupling , which connects cerebral blood flow , cerebral blood volume , and deoxyhemoglobin level to neuronal activity . The exact mechanisms behind this neurovascular coupling are not yet fully investigated . There are at least three different ways in which these mechanisms are being discussed . Firstly , mathematical models involving the so-called Balloon model describes the relation between oxygen metabolism , cerebral blood volume , and cerebral blood flow . However , the Balloon model does not describe cellular and biochemical mechanisms . Secondly , the metabolic feedback hypothesis , which is based on experimental findings on metabolism associated with brain activation , and thirdly , the neurotransmitter feed-forward hypothesis which describes intracellular pathways leading to vasoactive substance release . Both the metabolic feedback and the neurotransmitter feed-forward hypotheses have been extensively studied , but only experimentally . These two hypotheses have never been implemented as mathematical models . Here we investigate these two hypotheses by mechanistic mathematical modeling using a systems biology approach; these methods have been used in biological research for many years but never been applied to the BOLD response in fMRI . In the current work , model structures describing the metabolic feedback and the neurotransmitter feed-forward hypotheses were applied to measured BOLD responses in the visual cortex of 12 healthy volunteers . Evaluating each hypothesis separately shows that neither hypothesis alone can describe the data in a biologically plausible way . However , by adding metabolism to the neurotransmitter feed-forward model structure , we obtained a new model structure which is able to fit the estimation data and successfully predict new , independent validation data . These results open the door to a new type of fMRI analysis that more accurately reflects the true neuronal activity .
Functional magnetic resonance imaging ( fMRI ) measures brain activity by detecting associated changes in blood oxygenation through the blood-oxygen-level dependent ( BOLD ) response . The BOLD response is caused by time-dependent changes in deoxyhemoglobin ( dHb ) concentration [1] . Although the BOLD response reflects neuronal activity through the neurovascular coupling [2] , the mechanisms causing the response are still not fully understood . Here , we investigate these mechanisms by mathematical modeling using a systems biology approach . The BOLD response endures approximately 15 seconds after a short neural stimulus and it has several characteristic features ( Fig 1A ) [3]: ( i ) During the first couple of seconds a shallow undershoot , referred to as the initial dip , is sometimes observed in activated areas of the brain [4][5] . The initial dip is hypothesized to reflect an increased cerebral metabolic rate of oxygen ( CMRO2 ) that is followed by an increase of dHb content in the blood . ( ii ) At 6–8 s after the stimulus , the BOLD response peaks as a result of increased cerebral blood volume ( CBV ) and/or increased cerebral blood flow ( CBF ) . ( iii ) After the peak , the BOLD response decays and shows a post-peak undershoot before returning to baseline . The mechanisms controlling these processes ( i-iii ) remain unresolved , and there are at least three different approaches to understand these mechanisms . The first approach is centered around mathematical modeling . One of the most common approaches is to model the hemodynamic response function ( HRF ) using the so-called Balloon model [6][7][8][9] , which has been of paramount importance in the development of fMRI image analysis [10] . The Balloon models describe the interplay between CMRO2 , CBV , and CBF . The dynamics of these three entities are described in part by purely phenomenological descriptions , such as convolutions with the covariate gamma functions , and in part by physical models e . g . describing the dynamics between CBV and CBF in an expanding balloon . In other words , these Balloon models typically do not incorporate intracellular and biochemical mechanisms involved in cell metabolism or intra-cellular signaling processes related to the BOLD response . Nevertheless , there do exist mathematical models that also incorporate intracellular metabolism [11] , but ultimately even these models explain the actual BOLD response via the Balloon model , which appears as a sub-model in the complete model . Other models , which are not extensions of the Balloon model , describe e . g . spatiotemporal properties of the BOLD response as hemodynamic traveling waves [12] or oxygen transport in the brain by modeling CBF with a linear flow model and CMRO2 using a gamma function [13] . The second approach to understanding the BOLD response is centered around the so-called metabolic feedback hypothesis . According to this classical hypothesis ( Fig 1B , left ) , the BOLD response is the result of a tight connection between glucose metabolism and blood flow; when the brain is activated , the neurons consume more energy , resulting in decreased blood glucose and oxygen levels [14][15] , which trigger a feedback signal increasing CBF to meet metabolic demands . In other words , the metabolic hypothesis is centered around a feedback control to keep glucose level constant . The third and final approach relevant to this paper is referred to as the neurotransmitter feed-forward hypothesis . This hypothesis is reviewed in e . g [16] , and it is today more actively discussed than the metabolic feedback hypothesis . The neurotransmitter feed-forward hypothesis ( Fig 1B , right ) suggests sequential feed-forward signaling where neurotransmitters , especially glutamate , cause neurons and astrocytes to activate a chain of intracellular events , involving the release of nitric oxide ( NO ) or arachidonic acid ( AA ) metabolites , which in turn control constriction and dilation of the blood vessels . In this way , the feed-forward system “anticipates” the increased need , and goes directly from increased neural activity to increased blood supply . Of the three approaches mentioned above , only the first involves mathematical modeling and these models are focused mainly on the phenomenological description of the HRF . However , although the metabolic feedback and the neurotransmitter feed-forward hypotheses have been extensively studied through purely experimental approaches , these two hypotheses have never been implemented as mathematical models . It is therefore not known whether the proposed mechanisms of the metabolic feedback and the neurotransmitter feed-forward hypotheses actually would produce a BOLD response or not . Model-based testing of mechanistic hypotheses has been done in biological research for many years , and has gained increased interest through the rise of systems biology . As mentioned above , mathematical models are already standard when analyzing fMRI , but these models are partially phenomenological and have not been developed to test intracellularly centered hypotheses such as the metabolic feedback and the neurotransmitter feed-forward hypotheses . In contrast , intracellular mechanistic models are the main focus in systems biology , and here hypotheses are formulated as direct representations of the assumed biochemical reactions [17][18] . This formulation allows for a new type of data analysis , which revolves around two steps: ( i ) rejections and ( ii ) uniquely identified core predictions ( Fig 1C ) . This model-based approach provides a more comprehensive , correct , and verifiable analysis , compared to analyses based on inspection and reasoning . In other words , while it sometimes may seem logical to draw a certain conclusion based on visual inspection of some given data , we and others have repeatedly shown that such manual inspections often lead to incorrect , or at the very least incomplete , interpretations of the data [19 , 20] . In this paper , we provide a first systems biology analysis of the metabolic feedback and the neurotransmitter feed-forward hypotheses ( Fig 1B ) with regards to their ability to describe the BOLD response . We show that neither of the two hypotheses alone can satisfactorily describe the response . In contrast , a feed-forward mechanism with added oxygen metabolism can provide a satisfactory explanation of the BOLD response measured in fMRI .
Mechanistic modeling has within systems biology evolved into an iterative process , which alternates between model-based data analysis and the collection of new experimental data . This process is outlined in Fig 1C . In Step 1 , existing hypotheses are reformulated into a set of mathematical equations . In this paper , the two main hypotheses are the metabolic feedback and the neurotransmitter feed-forward hypotheses . In Step 2 , data are collected , which in this paper correspond to BOLD responses during visual stimulation . The model is then fitted to the data by optimization of the model parameters . Step 3 involves the formulation of biological criteria which might not be found in the collected data , but which have been previously reported in the literature and which the model must fulfil . Step 1–3 compose the hypothesis testing , which includes formulation , fitting and testing of the models . Model testing investigates which sets of equations can and cannot explain the given data , and whether or not these explanations are biologically realistic . This analysis leads to either rejection , which bring the process back to the model formulation step , or acceptance and further analysis . Step 4 , minimization , is a simplification of the model where as many states and parameters as possible are removed in order to identify key mechanisms of the system and facilitate computation and overview of the model . In Step 5 , further analysis of the explanations consists of the identification of relevant core predictions , i . e . uniquely identified predictions with uncertainty [21] . These predictions may sometimes be suitable for experimental testing , and this leads to collection of new data , Step 6 , which in turn leads to the final testing and analysis of the model . Sometimes , interesting model behaviors are discovered in this final step , which might lead back to the hypothesis testing step for further investigation . In this way , systems biology modeling has the potential to be a never-ending cycle , but for each step that is passed , new information about the system is obtained . The iterations end when the model is satisfyingly detailed or no more suitable data can be gathered .
Visual stimulation during both the intensity and the frequency experiment elicited significant activation in bilateral primary visual cortex , p < 0 . 05 family wise error ( FWE ) corrected for multiple comparisons ( Fig 3C ) . FWE is a Bonferroni-correction applying the random-field theory ( RFT ) to control the FWE rate by assuming that the data follow certain specified spatial patterns [28] . The Montreal Neurological Institute ( MNI ) co-ordinates of the activation peaks were: [-2 , -96 , 4] and [-10 , -84 , 2] for the intensity and the frequency experiments , respectively . The mean BOLD response to the primary visual stimuli in both experiments had a characteristic response peak at approximately 6 s after the stimuli ( Fig 4 ) . Peak amplitude was 23 . 5 ( 2 . 12% signal change ) in the intensity experiment and 19 . 8 ( 1 . 95% signal change ) in the frequency experiment . We also observed a post-peak undershoot , but neither of the resulting BOLD responses in any subject displayed a clear initial dip . Results described above show that the increased oxygen metabolism in the metabolic feedback hypothesis can produce an initial dip and that the neurotransmitter feed-forward hypothesis can give a realistic description of the blood flow increase during the BOLD response . Therefore , the neurotransmitter model structure was extended with a metabolic module , Mnm1 ( Fig 7 ) . In Mnm1 , the neuronal activity increases metabolism in the metabolic module and at the same time triggers glutamate release in the neurotransmitter module . The levels of dHb and oHb are controlled by the metabolic module and the blood flow is controlled by the neurotransmitter feed-forward module . Mnm1 has no feedback control of the blood flow . The output of this final model structure is the ratio of dHb and oHb .
The initial dip is stated as a biological criteria that the model must be able to perform in order to be accepted . The existence of the initial dip is debated [4][5]; it has been observed in some studies [30][31] , but not in others , at least not in all subjects [30][32] . Several factors can explain the absence of an initial dip in experimental data: ( 1 ) The initial dip is reported to be only 1–2% of the baseline signal [30][31]; in other words , given the low signal to noise ratio in fMRI , the shallow dip could easily be undetected . ( 2 ) The intersubject variability is considerable in fMRI [33][34] , and thus , the subject selection could be decisive for observing an initial dip or not . ( 3 ) The observance of an initial dip could be dependent on the experimental design . For instance , Hu et al . [30] found that the magnitude of the dip was reduced for brief stimuli; the minimum stimulus duration in their study was 1 . 5 s , and we used a duration of only 0 . 5 s , which could possibly explain the absence of the initial dip in our study . Even though the data largely lacked the initial dip , we included the initial dip as a constraint in our models . The leading hypothesis of the mechanism underlying the initial dip is an uncoupling of the oxygen metabolism from the CBF [4][5] , where the increased stimulus-induced oxygen metabolism leads to increased dHb levels and following decreased early-phase BOLD response . This hypothesis has been described previously by a model using the gamma variate curve [13] , and is also supported by optical imaging studies showing early stimulus-related dHb increases [35][36] . Based on these previous studies and the simulations of the metabolic feedback model structure in our work , the oxygen metabolism ( but not the feedback ) of the metabolic feedback hypothesis is necessary , but not sufficient , to reproduce the shape of the BOLD response . However , the metabolic feedback model structure has difficulties explaining the initial dip in combination with a post-peak undershoot ( see Fig 5 ) . The neurotransmitter feed-forward model structure , on the other hand , predicts an initial dip caused by initial vasoconstriction , a prediction that is not supported by any previous data ( e . g . [4][5] ) leading to rejection of that model structure . According to our extended model structure , Mnm1 , the initial dip is caused by changes in dHb/oHb ratio due to increased oxygen metabolism occurring during a CBF delay period , as described above . The most noticeable feature of the BOLD response is the large stimulus-induced peak of the fMRI signal . Originally it was suggested that reduced blood-oxygen levels increased the CBF [1] , and consequently the fMRI signal . This suggestion formed the basis for the model structure Mm1 . However , we showed that the oxygen-triggered feedback cannot cause such large overshoot ( Fig 5A ) . Later on , Fox et al . [15] showed that CBF and CMRO2 correlate strongly in the brain at rest , but not in response to stimuli; they found that CBF increased by 50% , CMRglu by 51% , but CMRO2 only by 5% when the brain is activated . They also concluded that the brain metabolism is aerobic during rest and anaerobic in response to stimuli , to cover the excess energy needed , which is reflected later in the astrocyte-neuron-lactate-shuttle model proposed by Magistretti and Pellerin [37] . Fox et al . [15] also showed that regional CBF is not driven by oxidative metabolism , but that stimulus-induced CBF is driven by increased glucose demand ( see review by Paulson et al . [29] ) . Prichard et al . [38] hypothesized that aerobic glycolysis might be close to its maximum capacity in the resting brain . Therefore , stimulus-induced activity requires quick energy increases via anaerobic glycolysis , causing the uncoupling of glucose and oxygen metabolism and CBF . In line with previous experimental research [15][29] , we show that if metabolism is the driving agent for stimulus-induced CBF then blood flow needs to be controlled by glucose , and stimulated metabolism must be partly anaerobic to obtain both an initial dip and a peak in the BOLD response . However , we also show that the metabolic feedback model structures Mm2 and Mm3 overstate predicted glucose reduction in response to stimuli , as they predict almost total depletion of glucose to trigger CBF feedback leading to rejection of the metabolic feedback model structure , Mm . This suggets that the metabolic feedback hypothesis plays a limited , if any , role in the neurovascular response , a conclusion supported by results from Lindauer et al . [39] and Wolf et al . [40] who showed that an increased CBF response will still occur even when hemoglobin is fully oxygenated and that CBF remains unchanged at hypoglycemia . The neurotransmitter feed-forward model structure Mn suggests that CBF is regulated in response to neuronal signaling itself and determined by the intensity and duration of the input signal . In our work , Mn predicts all characteristic features of the BOLD response and has acceptable fit to data . By minimizing the model , we could show that the main mechanism in Mn is a balance between vasoconstriction and vasodilation . One caveat with the neurotransmitter feed-forward model is that it predicts vasoconstriction to cause the initial dip , as discussed above . According to the final model structure Mnm1 , which has neurotransmitter control of the blood flow combined with metabolism of glucose and oxygen , CBF is primarily controlled by neurotransmitters that initiate processes in neurons and astrocytes causing release of vasoactive agents . The final model also predicts glucose response with similar shape as post-stimulus glucose levels measured by optical methods in rat [41] ( Fig 8 ) . The final characteristic of the BOLD response is the post-peak undershoot . According to our final model structure , Mnm1 , the post-peak undershoot is dependent solely on neurotransmitter-triggered changes in CBF . This result is supported by previous literature that suggests that the post-peak undershoot is caused by a post-stimulus CBF undershoot ( [42][43][44] reviewed in [3] ) . In our data , the post-peak undershoot appears in most of the individual data sets . However , the undershoot seems to become deeper as the peak grows higher in the intensity data sets , and the models tested in the current work cannot predict this behavior . Furthermore , the model cannot describe the deeper undershoot in the 1 s IPI dataset , although it can predict the increased amplitude of the peak . There are other hypotheses of the mechanisms of the post-peak undershoot that are not investigated here . For example , it is suggested that the post-peak undershoot is caused by slow post-stimulus baseline return of CMRO2-related oxygenation or venous CBV [3] . Mandeville and coworkers [45] found slow recovery of CBV that matched the post-peak undershoot duration suggesting a biomechanical rather than a metabolic effect [46] . It is worth noting that the balloon model explains the post-peak undershoot as a slow CBV recovery [6] . It has also been suggested that the post-peak undershoot is modulated by post-stimulus neural activity [47][48] . Future studies incorporating models for CBV changes and/or post-stimulus neural activity might clarify the neurovascular mechanisms behind the post-stimulus undershoot . The neurotransmitter feed-forward model with metabolism Mnm1 contains several experimentally undetermined parameters , such as glutamate and glucose levels . In future studies , the model parameters can be evaluated and optimized using magnetic resonance spectroscopy ( MRS ) in combination with BOLD-fMRI . In proton MRS , it is possible to obtain time-dependent variations of the neurotransmitters glutamate and GABA and metabolites such as glucose and lactate [49][50][51] . In these recent high-field ( 7 T ) MRS studies , it has been shown that glutamate , GABA , and lactate levels increase during visual stimulation and motor activation , whereas the glucose levels decreases during the same period . If volume was added to the model , a closer estimation of some parameters would also be possible , using experimental values from current literature . This would open the door to prediction of parameter values , in addition to the current predictions of model behaviour . As with all models , the model structures evaluated in this work depend on a number of underlying assumptions , which in turn limit the conclusions that can be drawn from the results . Nevertheless , such assumptions are essential in order to build a comprehensible model . We are well aware that there are several different hypotheses of the mechanisms behind specific features of the BOLD response , of which some are described above . In this work we chose to focus on two fundamental hypotheses . In the final model structure , Mnm1 , the metabolic and the neurotransmitter module run in parallel . One of the consequences of this structure is that the metabolism is directly controlled by the input signal ( see Fig 7 ) . A more physiologically relevant model would be to integrate the metabolism into neurons and astrocytes . That is to say , to model the glycolysis and oxidative metabolism as actually occurring in the neuronal cells in response to stimuli . Another part of the model structure that lacks physiological details is the action of smooth muscle cells and effects of cortical vessel elasticity . These mechanisms are in our model represented by delay states , which will not accurately represent the possible non-linearities of receptor actions and muscular contraction or relaxation . In addition , the current model does not differentiate between capillaries and arterioles . Recent research has found that cerebral hemodynamics has a spatiotemporal dependence related to the effective blood viscosity and cortical vessel stiffness [52] and mechanical restrictions on blood vessels depending on cortical depth [53] . A compartmentalized model that takes spatiotemporal hemodynamics into account would provide a physiologically more accurate description of the BOLD response . Finally , the output signal of the model is oHb/dHb , a simplification suggested by Ogawa et al . [1] . However , the BOLD signal equation provides a more correct description of the output signal . Δ S S 0 = e - Δ R 2 * · T E - 1 ≈ - Δ R 2 * · T E ( 9 ) where S0 is the MR signal at baseline and ΔS is the BOLD signal change with activation . Δ R 2 * is the difference in transversal relaxation rate between the activated state and baseline . Δ R 2 * is linearly related to dHb concentration . Following ideas from Davis et al . [54] , several improvements of the BOLD signal description have been published [8][55][56] . In the current work , volumes are not included in the model and therefore it is not possible to implement an output dependent on dHb concentration . However , by dividing the model into a tissue compartment and a blood compartment , as has been done in e . g . [13] , a more realistic expression for the output BOLD signal can be obtained . When comparing models with each other , it is important to keep track of model complexity and watch out for potential problems with overfitting . We approach these issues first by choosing a model framework not designed to be as flexible as possible , but instead based on the actual mechanisms believed to be present in the system . Second , we do visual inspection of the plots comparing data and model fits ( Figs 9A and 10A ) . As can be seen in both Fig 9A and 9B , in the time-window 12–18 seconds the mean values in the data show minor fluctuations , which most likely are noise . The model is not following these minor variations , which argues that we do not have problems with overfitting . Nevertheless , in the early time-points ( t = 0–3 s ) , the extended model does an initial dip , stays down a while , and then rises . Since a similar delayed rise can be seen in the data , this could in principle be a sign of overfitting . However , our third approach to test for overfitting—core prediction analysis—argues against that . In the core prediction analysis , we study a representative sub-set of all parameters that describe the data in a statistically acceptable way . In other words , since the core prediction analysis includes both optimal and less optimal parameters , it does not matter if there are some parameters that are overfitted , as long as parameters that are not overfitted are included in the prediction uncertainty analysis . Furthermore , this core prediction analysis shows that all found parameters show an initial dip and a delay ( Figs 9A and 10A , middle columns ) , arguing that this property is a necessary consequence of the model structure and the data , i . e . a uniquely identified core prediction . Our fourth approach for checking for unnecessary over-parametrization is model minimization . This punishes for unnecessary complexity in the sense of parameters that can be removed without significantly worsening the agreement with the estimation data ( Figs 9A and 10A , middle and right columns ) . Finally , the perhaps most important approach to check for overfitting is to use independent validation data . As can be seen in Figs 9B , 9C and 10B , 10C , both the extended model ( middle columns ) and the minimized model ( right columns ) , agree with this independent data , to which they have not been fitted . Furthermore , as can be seen in e . g . Fig 10C , the original extended model actually agrees slightly better with the data than the minimized model . All these facts argues that our models—although over-parametrized in the sense that many parameters have non-unique values—still are based on realistic biological mechanisms that capture the main features seen in the data , and not on too flexible model structures that are fitting to a specific noise realization . Although the BOLD response has been extensively studied and systems biology is a well established method , no one has so far investigated the BOLD response using this type of modeling . In this article , a model based on current physiological hypotheses of the mechanisms behind the BOLD response is presented . The model structures Mnm1 and Mnm2 can describe the time course of the BOLD response in the human visual cortex and correctly predict the response to several variations of the original stimulus not present in the estimation data . In contrast , the individual hypotheses Mm and Mn cannot alone describe the BOLD response in a realistic manner . Systems biology opens the door to a new type of fMRI analysis , which is firmly based in the physiology of the neurovascular coupling behind the measured signal . Systems biology also gives us the opportunity to obtain information about neural activation beyond what we can measure and may thereby help deepen our understanding of the complex system that is the brain . | Functional magnetic resonance imaging ( fMRI ) is a widely used technique for measuring brain activity . However , the signal registered by fMRI is not a direct measurement of the neuronal activity in the brain , but it is influenced by the interplay between the metabolism , blood flow and blood volume in the active area . This signal is called the blood-oxygen-level dependent ( BOLD ) response and occurs when the blood supply to the active area increases in response to neuronal activity . The mechanisms that the cells use to influence the blood supply are not fully known , and therefore it is difficult to know the true neuronal signalling only from inspection of the fMRI signal . In this article , we present a new mathematical model built on the physiological mechanisms thought to underlie the BOLD response . We could successfully fit the model to data and predict the activity caused by new stimuli . By using the validated model we investigated physiological mechanisms that cause different parts of the BOLD response . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"carbohydrate",
"metabolism",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"chemical",
"compounds",
"oxygen",
"carbohydrates",
"neuroscience",
"glucose",
"metabolism",
"organic",
"compounds",
"glucose",
"magnetic",
"resonance",
"imaging",
"oxygen",
"metabolism",
"systems",
"science",
"mathematics",
"brain",
"mapping",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"imaging",
"techniques",
"animal",
"cells",
"chemistry",
"hematology",
"blood",
"flow",
"systems",
"biology",
"biochemistry",
"cellular",
"neuroscience",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"blood",
"anatomy",
"organic",
"chemistry",
"cell",
"biology",
"physiology",
"neurons",
"monosaccharides",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"metabolism",
"chemical",
"elements"
] | 2016 | Mechanistic Mathematical Modeling Tests Hypotheses of the Neurovascular Coupling in fMRI |
Mutations in dystrophin can lead to Duchenne muscular dystrophy or the more mild form of the disease , Becker muscular dystrophy . The hinge 3 region in the rod domain of dystrophin is particularly prone to deletion mutations . In-frame deletions of hinge 3 are predicted to lead to BMD , however the severity of disease can vary considerably . Here we performed extensive structure-function analyses of truncated dystrophins with modified hinges and spectrin-like repeats in mdx mice . We found that the polyproline site in hinge 2 profoundly influences the functional capacity of a microdystrophinΔR4-R23/ΔCT with a large deletion in the hinge 3 region . Inclusion of polyproline in microdystrophinΔR4-R23/ΔCT led to small myofibers ( 12% smaller than wild-type ) , Achilles myotendinous disruption , ringed fibers , and aberrant neuromuscular junctions in the mdx gastrocnemius muscles . Replacing hinge 2 of microdystrophinΔR4-R23/ΔCT with hinge 3 significantly improved the functional capacity to prevent muscle degeneration , increase muscle fiber area , and maintain the junctions . We conclude that the rigid α-helical structure of the polyproline site significantly impairs the functional capacity of truncated dystrophins to maintain appropriate connections between the cytoskeleton and extracellular matrix .
Duchenne muscular dystrophy ( DMD ) is a lethal X-linked recessive disease caused by mutations in the 2 . 2 MB dystrophin gene [1]–[3] . In skeletal muscle , dystrophin provides a flexible connection between the cytoskeleton and the dystrophin-glycoprotein complex at the sarcolemma , myotendinous junction ( MTJ ) and neuromuscular junction ( NMJ ) [4]–[6] . Mutations that affect the mechanical integrity of this molecular scaffold render muscles more susceptible to contraction-induced injury leading to cycles of necrosis and regeneration [3] . As a general rule , most frame-shift mutations in dystrophin lead to DMD whereas internal truncations ( in-frame deletions ) lead to a milder form of the disease called Becker muscular dystrophy ( BMD ) [7]–[14] . The severity of BMD can also vary depending on whether a critical region of dystrophin is deleted and the amount of dystrophin being expressed [7]–[14] . Dystrophin consists of a N-terminal actin-binding domain , a large central rod domain , a cysteine rich region and a C-terminal domain ( Figure 1A ) [15] , [16] . The central rod domain contains 24 spectrin-like repeats , 4 hinges and a second actin-binding domain [15]–[20] . The locus encoding the N-terminal actin-binding domain and the region near hinge 3 of dystrophin are more susceptible to deletion mutations [7]–[13] . In-frame deletions of the central rod domain typically lead to a mild BMD [8]–[13] . However , in-frame deletions at the “hot spot” near hinge 3 can lead to more variable phenotypes [8]–[13] , [21] . The role of dystrophin in vivo has been largely defined by the structure-function relationship of truncated dystrophins in humans and mice [8]–[13] , [22]–[24] . Rational design of dystrophin mini-genes has been highly effective in preventing and reversing functional abnormalities of dystrophic muscles [22]–[29] . In particular , we previously developed a microdystrophin ( ΔR4-R23/ΔCT; defined as those with 4 or fewer spectrin-like repeats [24] ) that accommodates the limited cloning capacity of recombinant adeno-associated viral vectors ( rAAV ) [24] . Intravenous injection of rAAV vectors pseudotyped with serotype 6 capsid ( rAAV6 ) expressing microdystrophinΔR4-R23/ΔCT can prevent and reverse most aspects of dystrophic pathology in mdx muscles [24] , [28] , [30]–[35] . MicrodystrophinΔR4-R23/ΔCT also significantly protects muscles from contraction-induced injury [24] , [28] , [30]–[35] . While the microdystrophinΔR4-R23 transgene provides a clear benefit to dystrophic muscles [24] , more detailed analyses have revealed a potentially serious abnormality in some muscle groups . The microdystrophinΔR4-R23/mdx transgenic mice have chronic Achilles myotendinous strain injury , which leads to the formation of ringed fibers and fragmentation of the neuromuscular junctions [33] , [36] . In the present study we examined whether the domain composition or the small size of microdystrophinΔR4-R23/ΔCT led to this myopathy in mdx mice . We found that the hinge regions of microdystrophin , rather than its small size can profoundly influence skeletal muscle maintenance , maturation and structure .
We initially screened several truncated dystrophins and found that inclusion of hinge 2 , but not hinge 3 could lead to the structural abnormalities we observed in some muscles of the microdystrophinΔR4-R23 transgenic mice ( Text S1; Figures S1 , S2 , S3 ) . We subsequently compared the efficacy of two microdystrophins that differ only in their inclusion of hinge 2 ( microdystrophinΔR4-R23/ΔCT ) or hinge 3 ( microdystrophinΔH2-R23+H3/ΔCT ) ( Figure 1A ) to examine whether the hinge composition of microdystrophin could influence various aspects of muscle disease . We administered a sub-optimal dose of 2×1012 vector genomes of a rAAV6 pseudotyped vector expressing either microdystrophinΔR4-R23/ΔCT or microdystrophinΔH2-R23+H3/ΔCT intravenously into 2 week-old mdx4cv mice . We used a sub-optimal dose of rAAV6-microdystrophins so that we could examine whether changing the hinge domain increased or decreased the functional capacity of microdystrophin . Six months after treatment , both microdystrophins were expressed in a similar percentage of gastrocnemius and tibialis anterior ( TA ) muscle fibers ( ranging from approximately 61% to 71%; P = 0 . 238 when comparing between the microdystrophins; Figure 1B and 1D ) . Western blots confirmed similar expression levels of truncated dystrophins in treated gastrocnemius muscles ( Figure 1C ) . Both microdystrophins restored dystrophin-associated proteins to the sarcolemma except for nNOS ( Text S1; Figure S4 ) . MicrodystrophinΔR4-R23/ΔCT containing hinge 2 significantly prevented muscle degeneration ( ∼11% central nuclei for treated muscles verse ∼78% for untreated mdx muscles; P<0 . 001 ) , and limited the fiber area of skeletal muscles ( 12% smaller than wild-type; P<0 . 05; Figure 1E ) , consistent with previous studies [24] , [32] , [33] . MicrodystrophinΔH2-R23+H3/ΔCT containing hinge 3 was significantly better able to prevent muscle degeneration ( 1–2% central nuclei; P<0 . 05 compared to microdystrophinΔR4-R23/ΔCT ) , and surprisingly increased average muscle fiber cross sectional area ( 34% larger than wild-type; P<0 . 001; Figure 1E ) . Thus , replacing hinge 2 of microdystrophinΔR4-R23/ΔCT with hinge 3 significantly improved its capacity to prevent muscle degeneration and promote skeletal muscle maturation . The tendon extends deep folds into wild-type skeletal muscles to minimize membrane stress under shear ( Figure 2 ) [37] . Most of the folds in the mdx junctions did not extend as far into the gastrocnemius muscles ( Figure 2 ) . rAAV6-microdystrophinΔR4-R23/ΔCT severely disrupted the Achilles myotendinous junctions in mdx mice . Many of the junctional folds were missing and myofibril degeneration was evident ( Figure 2 ) . Approximately 17% of the adjoining mdx gastrocnemius muscles had ringed fibers . In contrast , rAAV6- microdystrophinΔH2-R23+H3/ΔCT with hinge 3 retained the normal architecture of the Achilles myotendinous junction and we found no ringed fibers in the adjoining gastrocnemius muscles ( Figure 2 ) . Thus , the hinge domains influenced whether microdystrophin was capable of maintaining the myotendinous junction and myofibril structure in mdx gastrocnemius muscles . We also examined neuromuscular synapses in mdx mice treated with rAAV6-microdystrophins . Most neuromuscular synapses in wild-type mice ( ∼97% ) form a continuous tertiary structure as shown by staining whole muscle fibers with α-bungarotoxin ( Figure 3A ) . Neuromuscular synapses in mdx mice begin to fragment temporally coincident with muscle degeneration [38] . Approximately 89% of neuromuscular synapses were fragmented in the gastrocnemius muscles of mdx mice ( Figure 3B ) . We had previously shown that the neuromuscular synapses in transgenic microdystrophinΔR4-R23/mdx gastrocnemius muscles fragmented temporally coincident with the formation of ringed fibers [36] . In the present study we found that rAAV6- microdystrophinΔR4-R23/ΔCT containing hinge 2 maintained continuous synapses in only 46% of the mdx gastrocnemius muscles ( Figure 3A and 3B ) . In contrast , approximately 84% of synapses were continuous in mdx gastrocnemius muscles treated with rAAV6-microdystrophinΔH2-R23+H3/ΔCT containing hinge 3 ( Figure 3A and 3B ) . Neuromuscular synapses also contain folds in the postsynaptic membrane that align directly adjacent to vesicle release sites ( active zones ) in the pre-synaptic nerve terminal ( arrows; Figure 3C ) . The number of synaptic folds in mdx mice was significantly reduced compared to wild-type ( P<0 . 01; Figure 3C and 3D ) as previously described [4] , [39] . The number of folds was restored in microdystrophinΔR4-R23/ΔCT and microdystrophinΔH2-R23+H3/ΔCT treated muscles ( Figure 3C and 3D ) . The synaptic folds extended significantly further into microdystrophinΔR4-R23/ΔCT treated mdx muscles compared to wild-type muscles ( P<0 . 001; Figure 3C and 3E ) , as previously described in transgenic microdystrophinΔR4-R23/mdx mice [36] . In contrast , the number and length of synaptic folds in microdystrophinΔH2-R23+H3/ΔCT treated mdx muscles was similar to wild-type ( Figure 3C–3E ) . Thus , microdystrophinΔH2-R23+H3/ΔCT containing hinge 3 can maintain the structure of neuromuscular junctions in mdx muscles . Contraction-induced injury can initiate muscle degeneration in mdx mice [40] . Skeletal muscles from mdx mice have a lower force producing capacity than wild-type muscles and are more susceptible to contraction-induced injury ( Figure 4 ) . We found that sub-optimal doses of both rAAV6-microdystrophinΔR4-R23/ΔCT and rAAV6- microdystrophinΔH2-R23+H3/ΔCT maintained the peak force producing capacity of mdx gastrocnemius and tibialis anterior muscles ( Figure 4A ) . Both microdystrophins also significantly improved the specific force ( force per cross sectional area of muscle ) production in mdx muscles ( P<0 . 05; Figure 4B ) . The specific force was not restored to wild-type partly because the sub-optimal dose of rAAV6-microdystrophin did not prevent the pseudo hypertrophy normally found in mdx muscles ( P = 0 . 454 when comparing the muscle mass between mdx and treated mdx muscles; one-way ANOVA ) . Each microdystrophin significantly protected the treated limb muscles from contraction-induced injury ( P<0 . 001; Figure 4C and 4D ) . However , we found no significant difference between the peak force , specific force or protection from contraction-induced injury when comparing between the two microdystrophins with either hinge 2 or hinge 3 . Together , our results suggested that the structural abnormalities observed in some treated mdx muscles could be traced to the presence of hinge 2 within the microdystrophin . We next examined the molecular composition of the hinges to define what was unique about hinge 2 . The hinges in dystrophin are defined as such because of the higher concentration of proline residues , which function to limit the continuation of the α-helical coiled-coils of the spectrin-like repeats through the entire length of the dystrophin rod domain [19] . Both hinge 2 and hinge 3 have six proline residues and the lengths of these hinges are similar [19] . We hypothesized that the placement of the prolines most likely results in their different functions [5] , [19] . Hinge 2 has 5 consecutive proline residues ( polyproline; Figure 5A ) whereas the proline residues in hinge 3 are more evenly distributed throughout the hinge [19] . Polyproline residues are thought to have their own defined rigid helical structure [41] , [42] , and this could affect the functional capacity of microdystrophinΔR4-R23/ΔCT . To test this hypothesis we compared muscles expressing the original microdystrophinΔR4-R23/ΔCT with a newly developed microdystrophinΔpolyP/ΔR4-R23/ΔCT that lacks the polyproline site in hinge 2 ( Figure 5A ) . We delivered 6×1010 vg of each microdystrophin into mdx gastrocnemius muscles at 2 days of age and examined the mice 7 weeks after treatment . Both microdystrophins were expressed in a similar percentage of muscle fibers ( Figure 5B; 59–68% ) , and were expressed at similar levels ( Figure 5C ) . Each microdystrophin significantly reduced muscle fiber degeneration ( Figure 5D ) . As expected , the original microdystrophinΔR4-R23/ΔCT limited muscle fiber cross-sectional area ( Figure 5E ) , was associated with disrupted myotendinous junctions ( Figure 5F ) , led to the formation of ringed fibers ( Figure 5F ) , and perturbed neuromuscular junctions ( Figure 5G–5I ) . In contrast , the mdx muscles treated with microdystrophinΔpolyP/ΔR4-R23/ΔCT did not show any abnormalities in muscle fiber maturation or structure ( Figure 5 ) . Thus , the presence of this polyproline site in hinge 2 of microdystrophinΔR4-R23/ΔCT prevented the appropriate integration of muscles into the nerve-tendon environment .
MicrodystrophinΔH2-R23+H3/ΔCT with hinge 3 significantly increased peak force , specific force and protected muscles from contraction-induced injury . However , the morphological improvements of microdystrophinΔH2-R23+H3/ΔCT treated muscles did not translate into a functional improvement compared to microdystrophinΔR4-R23/ΔCT treated muscles . This could result from the molecular and cellular responses to myotendinous strain injury that help protect the rAAV6-microdystrophinΔR4-R23/ΔCT treated muscles from contraction-induced injury [33] . Another possibility is that the presence of some dystrophin negative fibers masked any functional difference between the two proteins . The inclusion of hinge 2 in microdystrophin limited muscle fiber area whereas the inclusion of hinge 3 increased muscle fiber area ( Figure 1 ) . Larger muscle fibers in microdystrophinΔH2-R23+H3/ΔCT treated mice could have two distinct advantages: They could replace some of the muscle mass lost in advanced stages of disease and they could be better protected from contraction-induced injury [44] . However , the sub-optimal dose of either rAAV6-microdystrophin did not prevent the pseudo hypertrophy in mdx mice and no mechanical advantages could be discerned when comparing treatments . Saturating levels of rAAV6-microdystrophins or transgenic mice will most likely be required to detect minor differences in the mechanical properties of muscles expressing various truncated dystrophins . Our most effective truncated dystrophins developed for gene therapy have been designed to maximize functional interactions between specific spectrin-like repeats and hinge domains . This design has been influenced by genetic studies in mice and man as well as biophysical studies in vitro on the structure , folding and physical properties of both individual and tandemly expressed spectrin-like repeats and hinge domains [24] , [45]–[52] . Individual spectrin-like repeats are not all interchangeable , and ones adjacent to hinges have distinct properties from those flanked by other spectrin-like repeats [21] , [24] , [47] , [51] , [52] . Also , spectrin-like repeats rarely function as isolated units [15] , [24] , [50]–[53] . Instead , they appear to fold into nested domains interrupted by various insertions ( hinges ) that disrupt the uniformity and rigidity of the spectrin-like repeat rod domain [24] , [45]–[48] , [53]–[55] . These interruptions appear important for the elastic and flexible structure that dystrophin requires in its role as a force transducer and shock absorber in muscle [56]–[59] . Our studies suggest that the most functional truncations of dystrophin retain a central hinge domain that is flanked by spectrin-like repeats found adjacent to a hinge in the wild-type dystrophin [24] . Disruption of this linkage could influence protein folding , stability and function leading to the variable phenotypes in patients associated with deletions at or near hinge 3 , which is encoded on exons 50–51 [21] . Individual spectrin-like repeats are composed of 3 helical domains connected by non-helical linkers , which fold into a triple helical coiled coil structure ( Figure 5A; [45] , [47] ) . The linker regions between discreet repeats are also typically short and relatively unstructured to allow a smooth connection between the third helix of a preceding repeat and the first helix of the next repeat ( Figure 6A ) . However , hinge domains interrupt the nested nature of adjacent spectrin-like repeats and allow more flexibility in the rod domain ( Figure 6B ) . This degree of flexibility appears to be significantly different when hinge 2 or hinge 3 is present . While both hinges contain 6 prolines , which act to disrupt alpha helical structures , in hinge 3 they are dispersed whereas 5 of the 6 prolines in hinge 2 are clustered together ( Figure 5A , Figure 6C and 6D; [10] ) . Polyproline residues form a rigid α-helix [41] , [42] , much like a molecular ruler [60] . We suggest that the location of this polyproline sequence within a highly truncated rod domain induces a severe structural disruption that can affect the ability of dystrophin to form a mechanically flexible connection between F-actin and β-dystroglycan . Spectrin-like repeats 1-3 have been shown to associate with the sarcolemmal membrane , while the WW domain in hinge 4 forms a critical portion of the β-dystroglycan binding domain [45] , [61] . A rigid rod domain induced by polyproline in hinge 2 may directly impair the ability of microdystrophin to form a flexible interaction with either or both of these structures ( Figure 6C ) . In contrast , when hinge 2 is present in full-length dystrophin , a significantly greater number of spectrin-like repeats are present between the hinge and the β-dystroglycan binding domain , allowing greater flexibility in the overall structure . It is difficult to predict the function of the polyproline site from patients with in frame deletions of exon 17 ( hinge 2 ) of dystrophin . The described deletions ( Leiden Muscular Dystrophy Pages ) usually encompass larger regions of dystrophin than the polyproline site and it is not clear how these deletions affect protein stability . Our finding that hinge 3 microdystrophin can prevent muscle degeneration suggests that the polyproline site is not a necessary component of dystrophin , similar to previous reports on longer forms of truncated dystrophins [24] , [62] . Flanigan et . al . , 2009 has proposed that approximately 62% of all DMD patients could be treated with oligonucleotides that skip exons 45–55 ( from spectrin-like repeat 18–22 ) [14] . This would create a truncated dystrophin that contains hinge 2 but not hinge 3 , similar to , but much larger than our microdystrophinΔR4-R23 transgene . It will therefore be of interest to determine whether the polyproline site in hinge 2 can influence the functional capacity of larger , truncated dystrophins . It will also be of interest to examine whether the polyproline site affects the functional capacity of truncated utrophin constructs that are designed for gene therapy of DMD [63] , [64] .
We utilized C57Bl/10 wild-type mice and mdx4cv mice . All experiments are in accordance with the institution of animal care and use committee ( IACUC ) of the University Of Washington . The expression vector CMV-ΔR4-R23/ΔCT which uses the cytomegalovirus immediate early promoter and enhancer to drive expression of a microdystrophin cDNA was generated as previously described [32] . We generated the ΔH2-R24/ΔCT , ΔR2-R23+R18-H3/ΔCT , ΔH2-R23+H3/ΔCT and ΔPolyP/ΔR4-R23/ΔCT constructs using recombination PCR with CMV-ΔR4-R23/ΔCT as the template [65] . The primers used to generate ΔH2-R24/ΔCT , ΔR2-R23+R18-H3/ΔCT , ΔH2-R23+H3/ΔCT and ΔPolyP/ΔR4-R23/ΔCT are found in Table S1 . The resulting expression vectors were sequenced and co-transfected with the pDGM6 packaging plasmid into HEK293 cells to generate recombinant AAV vectors comprising serotype 6 capsids that were harvested , purified , and quantitated as described previously [29] . The resulting titer was determined by comparison to previously known concentrations of rAAV6-CMV-lacZ and ΔR4-R23/ΔCT by Southern analyses with a probe to the CMV promoter . The rAAV6-microdystrophins were delivered intravenously by tail vein injection at two weeks of age or directly into the mdx gastrocnemius muscles at 2 days of age while the mice were anaesthetized . Gross muscle morphology was analyzed as previously described [24] , [32] . Primary antibodies included the N-terminus of dystrophin ( 1∶800; [23] ) , utrophin A ( 1∶300; gift from Stanley Froehner , University of Washington ) , mouse monoclonal anti-α-dystrobrevin ( Transduction laboratories; 1∶200 ) , rabbit polyclonal anti-Syn17 ( α-syntrophin; 1∶200; [66] ) , rabbit polyclonal anti-nNOS ( Alexis; 1∶200 ) . Secondary antibodies included Alexa 488 , Alexa 594 rabbit polyclonal or Alexa 488 mouse monoclonal secondary antibodies ( Molecular Probes; 1∶800 ) . The sections were mounted in anti-fade mounting media containing DAPI ( Vector Labs ) . Fluorescent sections were imaged using a Nikon eclipse E1000 fluorescent microscope ( Nikon; NY ) and captured using a DeltaVision fluorescence microscope . Muscle fiber areas were quantified using Image J ( NIH ) . For immunoblots , n = 4 gastrocnemius muscles from mdx mice and mdx mice treated with rAAV6-microdystrophinΔR4-R23/ΔCT or rAAV6-microdystrophinΔH2-R23+H3/ΔCT were thawed from OCT blocks and placed into extract buffer ( 50 mM Tris-HCl , 150 mM NaCl , 0 . 2% sodium dodecyl sulfate , 10% glycerol , 24 mM Na Deoxycholate , 1% NP40 , 47 . 6 mM Na Fluoride , 200 mM Na orthovanadate , Roche ) . Protein concentrations were determined by Coomassie Plus Bradford Assay ( Peirce ) . Equal amounts of protein ( 15 mg ) were resolved on a 4–12% SDS polyacrylamide gel . The blots were incubated in rabbit polyclonal antibodies to dystrophin ( 1∶500; kind gift from James Ervasti , University of Minnesota ) and mouse monoclonal antibodies to α-sarcomeric actin ( 1∶500; SIGMA ) . We also performed immunoblots on frozen tissue sections from n = 4 gastrocnemius muscles treated with rAAV6-microdystrophinΔR4-R23/ΔCT and microdystrophinΔPolyP/ΔR4-R23/ΔCT as previously described [67] , with minor modifications . Briefly , we cut twenty-five 20 µm sections and diluted the sections into 200 µl lysis buffer ( 4% SDS , 25 mM Tris pH 8 . 8 , 40% glycerol , 0 . 5 M phenylmethylsulfonyl fluoride , 100 mM dithiothreitol and bromophenol blue ) . Samples were briefly sonicated ( 10 sec at 4°C ) , heated to 95°C for 5 minutes , centrifuged for 5 minutes at 13 , 200×g and electrophoresed on a 4–12% SDS-polyacrylamide gel . The blots were incubated in primary rabbit polyclonal antibody against the N-terminus of dystrophin ( 1∶500; kind gift from James Ervasti , University of Minnesota ) . All blots were developed with ECL Plus ( Pierce ) and scanned with the Storm 860 imaging system ( Amersham Biosciences ) . Electron microscopy was performed as previously described [33] . The junctional fold number and lengths were measured from n = 4 mice at 6 months of age using Image J ( NIH ) and compared using Students t-test ( Prism ) . The counts represent the fold numbers and lengths from all fibers ( dystrophin positive and negative ) . We quantitated the number of ringed myofibers in EM images and thick ( 1 µm ) toluidine blue sections from at least 4 animals per group . At least 300 muscle fibers from n = 4 gastrocnemius muscles were examined from wild-type , mdx4cv and mdx4cv mice expressing the various microdystrophins . Neuromuscular synapses were analyzed in whole mount immunofluorescence stained muscles and quantitated as previously described [36] . The acetylcholine receptor clusters were stained with TRITC conjugated α-bungarotoxin ( αBTX; 1∶800; Molecular Probes ) . Synapses were classified as continuous if they presented with 3 or less continuous regions of AChR clustering and discontinuous if they presented with more than 3 regions of AChR clustering . More than 50 synapses were analyzed from treated and untreated gastrocnemius skeletal muscle fibers from n = 4 mice . The counts in treated muscles include both dystrophin positive and negative fibers . We compared the proportion of continuous synapses using a Students t-test . Muscle physiology was performed as previously described for tibialis anterior [29] and gastrocnemius [33] muscles . We examined six-month-old wild-type , mdx , and mdx mice treated with rAAV6-microdystrophinΔR4-R23/ΔCT or rAAV6-microdystrophinΔH2-R23+H3/ΔCT ( n = 5 ) . | Dystrophin functions like a large molecular spring between the muscle cytoskeleton and the extracellular matrix in order to protect the membrane from contraction-induced injury . Mutations in dystrophin can lead to a severe muscle wasting disease called Duchenne muscular dystrophy ( DMD ) in young boys . DMD patients are typically wheelchair bound by 9–13 years of age and die at approximately 30 years . There are also mutations within the dystrophin gene that lead to internal truncations of non-essential regions , such as the internal rod domain that leads to a mild form of the disease called Becker Muscular Dystrophy . However , these internal truncations frequently occur at a “hot spot” within the rod domain where the resulting disease severity is difficult to predict . Here we found that consecutive proline residues , that function much like a molecular ruler , can dramatically influence the function of these internally truncated dystrophins within skeletal muscles . Using this information , we designed a dystrophin mini-gene that can accommodate the limited packaging size of recombinant adeno-associated virus . This virus can deliver the dystrophin mini-gene to most muscles throughout a dystrophic mouse to prevent muscle degeneration and partially restore muscle function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/animal",
"genetics",
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/gene",
"function",
"genetics",
"and",
"genomics/gene",
"therapy",
"genetics",
"and",
"genomics"
] | 2010 | The Polyproline Site in Hinge 2 Influences the Functional Capacity of Truncated Dystrophins |
CellProfiler has enabled the scientific research community to create flexible , modular image analysis pipelines since its release in 2005 . Here , we describe CellProfiler 3 . 0 , a new version of the software supporting both whole-volume and plane-wise analysis of three-dimensional ( 3D ) image stacks , increasingly common in biomedical research . CellProfiler’s infrastructure is greatly improved , and we provide a protocol for cloud-based , large-scale image processing . New plugins enable running pretrained deep learning models on images . Designed by and for biologists , CellProfiler equips researchers with powerful computational tools via a well-documented user interface , empowering biologists in all fields to create quantitative , reproducible image analysis workflows .
Image analysis software is now used throughout biomedical research in order to reduce subjective bias and quantify subtle phenotypes when working with microscopy images . Automated microscopes are further transforming modern research . Experiments testing chemical compounds or genetic perturbations can reach a scale of many thousands of perturbations , and multidimensional imaging ( time-lapse and three-dimensional [3D] ) also produces enormous data sets that require automated analysis . In light of this data scale , computer algorithms must deliver accurate identification of cells , subcompartments , or organisms and extract necessary descriptive features ( metrics ) for each identified object . Racing to keep up with the advancement of automated microscopy are several classes of biologist-focused image analysis software , such as companion packages bundled with imaging instruments ( e . g . , MetaMorph—Molecular Devices , Elements—Nikon ) , stand-alone commercial image processing tools ( e . g . , Imaris—Bitplane ) , and free open-source packages ( e . g . , ImageJ/Fiji , CellProfiler , Icy , KNIME ) . Commercial software is often convenient to use , especially when bundled with a microscope . Although cost and lack of flexibility may limit adoption , there is a focus on usability , particularly for applications of interest to the pharmaceutical industry . Still , the proprietary nature of the code in commercial software limits researchers from knowing how their data is being analyzed or modifying the strategy of a given algorithm , if desired . The open-source biological image analysis software ecosystem is thriving [1] . ImageJ [2] was the first and is still the most widely used package for bioimage analysis; several other packages are based on its codebase ( most notably , Fiji ) . ImageJ excels at the analysis of individual images , with a user interface analogous to Adobe Photoshop . Its major strength is its community of users and developers who contribute plugins , although an associated drawback is the sheer number of plugins , with varying degrees of functional overlap , usability , and documentation . Multitasking toolboxes like KNIME [3] offer a more modular approach , which is better suited to automated workflows . KNIME equips users with a wide breadth of powerful utility , from performing image analysis to data analytics . CellProfiler , our open-source software for measuring and analyzing cell images , has been cited more than 6 , 000 times , currently at a rate of more than 1 , 000 per year . The first version of CellProfiler was introduced in 2005 and published in 2006 [4] . It is widely adopted worldwide , enabling biologists without training in computer vision or programming to quantitatively measure phenotypes robustly from thousands of images . A second major version of CellProfiler , rewritten in Python from its original MATLAB implementation , was published in 2011 [5] and included methods for tracking cells in movies and measuring neurons , worms , and tissue samples . In 2015 , a laboratory unaffiliated with our team rigorously compared 15 free software tools for biological image analysis: CellProfiler was ranked first for both usability and functionality [6] . CellProfiler provides advanced algorithms for image analysis , organized as individual modules that can be placed in sequential order to form a pipeline . This pipeline is then used to identify and measure cells or other biological objects and their morphological features . CellProfiler’s modular design and carefully curated library of image processing and analysis modules benefits biologists in several ways: Reproducibility at scale: CellProfiler is designed to produce high-content information for each cell or other object of interest in each image and to apply the same objective analysis in high-throughput , e . g . , across thousands or millions of images . Flexible feature extraction: Individual modules measure standard morphological features such as size , shape , intensity , and texture . Customized combinations of modules can extract even more complex information . As such , CellProfiler is commonly used for morphological profiling experiments such as Cell Painting [7 , 8] , which is being adopted in pharmaceutical companies to speed several steps in drug discovery [9] . Easy to learn: Each of the 70+ modules includes carefully crafted documentation , curated by both imaging and biology experts , to make image processing more approachable and understandable for the average scientist . Further , each individual setting is explained in practical terms to aid researchers in configuring it . The number of modules and settings is carefully limited to avoid overwhelming users , while a plugin system allows the flexibility of a larger array of contributed modules . Community: CellProfiler has an active community of more than 3 , 000 people on its online question and answer forum . With more than 15 , 000 posts , users provide feedback that fuels improvements to CellProfiler , find pipelines related to their area of research , interact with developers , get input on challenging problems , and improve image analysis skills and knowledge by helping other users design solutions .
This new version of CellProfiler has support for analysis of 3D images in many of its modules ( S1 Fig ) . Although open-source software tuned to 3D problems exists ( e . g . , Vaa3D , BioImageXD , Slicer ) [10] , it often emphasizes visualization and rendering; these new 3D capabilities of CellProfiler meet the community’s demand for modular high-throughput 3D analysis . CellProfiler 3 . 0 can apply image processing , segmentation , and feature extraction algorithms to entire image volumes ( volumetric analysis ) , in addition to the more typical iterative and separate analysis of two-dimensional slices from a 3D volume ( “plane-wise” analysis ) . Whole-volume algorithms consider 3D neighborhoods and incorporate information from surrounding planes , yielding more accurate results , but require more available memory , particularly for large files . CellProfiler’s volumetric algorithms can be configured to account for anisotropic data ( in which the distance between Z planes does not match the distance between pixels in the X and Y dimensions ) . While we focused on adding 3D capability to most of our image processing and feature extraction modules , we will continue increasing the number of CellProfiler modules that support image volumes for situations in which it is not computationally prohibitive . We developed 3D pipelines to identify cells and subcompartments of cells for a number of experimental situations and sample types across a number of laboratories . We identified nuclei based on a DNA stain ( Fig 1A ) in 3D image stacks of human induced pluripotent stem cells ( hiPSCs ) . After processing by several CellProfiler modules ( Fig 1C ) , the final results agree well with manually annotated nuclei ( Fig 1D ) . Results for a variety of images with a range of complexity are shown in Fig 2 , with more detailed views in S2–S5 Figs . We characterized CellProfiler’s segmentation accuracy in two ways: in the first , we used real microscopy images ( Fig 1A , Fig 2A , Fig 2B ) whose ground truth was manually annotated by an expert image analyst; such images are realistic , but the manual annotation introduces some subjectivity . We therefore also used synthetic images ( Fig 2C , Fig 2D ) [11 , 12] , which , depending on the model used to create them , may not perfectly represent real microscopy images but whose ground truth can be unambiguously known . To determine how well the segmented objects agreed with ground truth , CellProfiler’s “MeasureImageOverlap” module was used to calculate the plane-wise Rand index [13] , a performance metric of accuracy ( Fig 1B , Fig 2E ) . Rand index values showed good agreement ( 0 . 919–0 . 976 ) between each tested image and its ground truth . The results produced by CellProfiler 3 . 0 were comparable to results produced by the commonly used Fiji plugin MorphoLibJ ( 0 . 930–0 . 977 ) ( Fig 1B , Fig 2E and S2–S5 Figs; the MorphoLibJ macro codes are provided in S1 Table ) . We demonstrate several kinds of analysis , including analyses of cell count in a time series that was synthetically generated [11 , 14] ( S5 Fig ) ; identification and quantification of children objects inside parent objects , such as speckles of transcripts within cells ( Fig 3 ) ; and measurement of various features of hiPSCs located at the center and the edge of the cell colony ( Fig 4 ) . All pipelines , annotated with notes to understand the function of each module , are provided at https://github . com/carpenterlab/2018_mcquin_PLOSBio . All raw images , together with ground truth annotations used to test CellProfiler 3 . 0 performance , are publicly available for further community algorithm development in the Broad Bioimage Benchmark Collection [15] , as indicated in the legends for Fig 1 and S2–S5 Figs . Convolutional neural networks ( CNNs ) are a type of deep learning model that transforms input images into outputs specified by the problem type [16] . For instance , image classification models transform images into categorical labels [17] , while image segmentation models transform images into segmentation masks [18] . CNNs are now widely used to solve many computer vision tasks , given their ability to produce accurate outputs after learning from examples . CellProfiler now can be configured to make use of cutting-edge CNNs to analyze biomedical images . While CellProfiler does not yet incorporate user-friendly functionalities to train neural networks , various models that have been already trained by researchers can be run inside CellProfiler . Running neural network models requires the installation of certain deep learning frameworks that are distributed separately , such as TensorFlow or Caffe . TensorFlow [19] is an open-source software library for machine learning that interfaces with Python and is compatible with CellProfiler when installed from source on Linux , Mac , and more recently , Windows . Caffe [20] is a deep learning framework designed for high-performance neural networks and is primarily available for Linux systems . Some network models may need special graphics processing units ( GPUs ) installed and configured in the system to run the computations efficiently , but this is not always required . Fortunately , both TensorFlow and Caffe can easily switch between running on GPUs and traditional central processing units ( CPUs ) just by changing the corresponding configuration . We created the CellProfiler 3 . 0 module ClassifyPixels-Unet to segment nuclei in images stained with DNA labels ( https://github . com/CellProfiler/CellProfiler-plugins ) . This plugin implements a U-Net[18] model using TensorFlow and can be run on CPUs . We have also provided the network architecture with training routines in case users have their own annotated images to learn a segmentation model for different images and objects of interest ( https://github . com/carpenterlab/unet4nuclei ) . The ClassifyPixels-Unet module classifies pixels into one of three classes: background , nucleus interior , or nuclear boundary ( S7 Fig ) . A pretrained network for nuclei segmentation is available for download and is automatically loaded by the plugin; a pipeline and image to run this are available as S4 File . We also created a CellProfiler 3 . 0 module , MeasureImageFocus , in collaboration with Google Accelerated Science , who trained a model to detect focus in images [21] . The module displays a table with the predicted focus score and certainty for the whole image , as well as a figure with the focus scores and corresponding certainties of individual 84 × 84 patches represented by color and opaqueness . It uses TensorFlow as its underlying deep learning framework . Independently , Sadanandan and colleagues created a CellProfiler 2 . 2 . 0 module—CellProfiler-Caffe bridge—that enables running a pretrained model for cell segmentation within a CellProfiler pipeline [22] . We created Distributed-CellProfiler ( https://github . com/CellProfiler/Distributed-CellProfiler ) , a script-based interface that allows running thousands of batches of images through CellProfiler in parallel on Amazon Web Services ( AWS; S8 Fig ) . While Distributed-CellProfiler does require basic knowledge of AWS and interaction with the command line , it is well documented and has been successfully run by biologists without formal computational training . The script handles infrastructure creation and removal as well as creation and storage of logs , allowing users without access to a local cluster computing environment to analyze large data sets with only minimal time devoted to having to set up those resources . Sample pipelines and configuration files are available as S5 File . Plug-ins: CellProfiler-plugins is a new repository for the community to share and distribute new CellProfiler modules ( https://github . com/CellProfiler/CellProfiler-plugins ) . Documentation: All of CellProfiler’s documentation was updated for content and readability; detailed help is available for 100% of module configuration options ( excluding plugins ) . New image processing features: CellProfiler 3 . 0 introduces an extended suite of modules for feature detection , feature extraction , filtering and noise reduction , image processing , image segmentation , and mathematical morphology operations . Infrastructure improvements: The project team reengineered major core components of CellProfiler . CellProfiler’s codebase was trimmed down , in part because of better integration with Python’s scientific community . We have adopted and contributed to the standard libraries of the scientific Python community , including NumPy , SciPy , and scikit-image . CellProfiler’s code is now 100% Python , which improves interoperability with the robust Python scientific ecosystem and simplifies third-party contributions . As well , we upgraded support to 64-bit on Linux , MacOS , and Windows , and a continuous integration process ensures the software is well tested on a variety of platforms . We made substantial progress simplifying CellProfiler’s installation . In addition to our previously existing Mac and Windows builds , a Python wheel is now available from the Python Package Index , and a Docker image is now available from Docker Hub . In an effort to expand CellProfiler’s flexibility , we made CellProfiler much simpler to compile on a variety of familiar and unusual platforms by requiring fewer dependencies and only using ubiquitous build systems . Educational resources: CellProfiler’s many examples and tutorials are now publicly available on GitHub ( https://github . com/CellProfiler/examples and https://github . com/CellProfiler/tutorials ) and have been updated for compatibility with CellProfiler 3 . 0 . Speed: CellProfiler 3 . 0’s processing speed is faster than version 2 . 2 on the most common types of pipelines; the degree of difference depends on the exact modules involved: CellProfiler 3 . 0 ran at a comparable or faster speed than CellProfiler 2 . 2 for 11 of 16 example pipelines tested ( S9 Fig ) . While the total amount of time needed to run the five pipelines shown in S9 Fig was comparable between CellProfiler and MorphoLibJ ( 482 versus 542 seconds ) , the relative speed was highly specific to the individual pipeline ( S6 File ) , ranging from 2× faster in CellProfiler to 6× faster in MorphoLibJ ( S2 Table ) . In addition , CellProfiler can run multiple images in parallel , depending on the individual’s number of threads , computing power , and access to cloud computing resources , making it suited to large-scale experiments . As well , CellProfiler’s modules enable more readily configurable complex analyses than MorphoLibJ , such as associating cytoplasm regions ( as in Fig 3 ) , transcripts ( as in Fig 3 ) , and other entities to nuclei and measuring a wide variety of morphological properties of each , including intensities , shapes , textures , colocalization metrics , and neighborhood relationships ( as in Fig 4 ) .
CellProfiler is mature software serving a large community and making an impact through its thousands of users’ biological discoveries . It has been involved in the discovery of potential life-saving drugs for infectious diseases , leukemia , and cerebral cavernous malformation [23–27] and in clinical trials for hematological malignancies [28] and will continue to fuel basic and applied research around the world . CellProfiler can readily generate a large amount of morphological information for each biological entity that is measured . We see advancements in data mining , downstream and apart from CellProfiler , as blossoming in the coming years . Already , 20 laboratories in the field of morphological profiling have gathered for two annual meetings/hackathons ( now called CytoData ) [29] , collaborated to outline best practices [30] , and begun a community library ( Cytominer , https://github . com/cytomining/cytominer ) . In addition to our user-friendly tool for classical machine learning based on measured features , CellProfiler Analyst [31] , we have begun creating Deepometry ( http://github . com/broadinstitute/deepometry ) , a tool that enables scientists without training in machine learning to perform single-cell phenotype classification using deep learning and other advanced downstream data analytics . Interoperability of CellProfiler with popular notebook tools like Jupyter would allow seamless workflows involving other complementary software tools . Finally , deep learning has revolutionized computer vision and other fields in the past few years [16 , 32] , and bioimaging will be no exception . As noted , already some models trained for specific tasks can be used via CellProfiler , and we expect that over time , more generalizable models will be created that can accomplish useful tasks such as detecting common cellular structures across diverse types of images and experimental setups , as in , for example , the 2018 Data Science Bowl challenge . Community-driven collections of images and ground truth , as well as “model zoos , ” will be instrumental for this . We have also begun creating libraries ( Keras-ResNet [https://github . com/broadinstitute/keras-resnet] and Keras-RCNN [https://github . com/broadinstitute/keras-rcnn] ) that will provide the foundation for interfaces that allow biologists to annotate , train , and use deep learning models . We expect that over time , these models will reduce the amount of time biologists spend tuning classical image processing algorithms to identify biological entities of interest in images .
Images were kindly provided by Javier Frias Aldeguer and Nicolas Rivron of Hubrecht Institute for Developmental Biology and Stem Cell Research and Li Linfeng of MERLN Institute for Technology-Inspired Regenerative Medicine . As per Rivron and colleagues [33] , mouse embryos ( 3 . 5 dpc ) were fixed right after isolation from the mother’s uterus . Fixation was performed using 4% PFA in RNAse-free PBS containing 1% acetic acid . ViewRNA ISH Cell Assay kit ( cat# QVC0001 ) was used for performing smFISH on the embryos . The protocol includes steps of permeabilization and protease treatment as well as probes , preamplifier , amplifier , and label hybridizations . Embryos were then mounted in Slowfade reagent ( Thermofisher cat# S36937 ) and directly imaged in a PerkinElmer Ultraview VoX spinning disk microscope in confocal mode by using a 63×/1 . 40 NA oil immersion lens . Images were acquired by collaborators from the Allen Institute for Cell Science , Seattle , as per Roberts and colleagues [34] . Briefly , wild-type C ( WTC ) hiPSCs were cultured in a feeder-free system on tissue culture dishes or plates coated with GFR Matrigel ( Corning ) diluted 1:30 in cold DMEM/F12 ( Gibco ) . Undifferentiated cells were maintained with phenol red containing mTeSR1 media ( 85850 , STEMCELL Technologies ) supplemented with 1% ( v/v ) penicillin-streptomycin ( P/S; Gibco ) . Cells were not allowed to reach confluency greater than 85% and are passaged every 3–4 days by dissociation into single-cell suspension using StemPro Accutase ( Gibco ) . When in single-cell suspension , cells were counted using a Vi-CELL Series Cell Viability Analyzer ( Beckman Coulter ) . After passaging , cells were replated in mTeSR1 supplemented with 1% P/S and 10 μM ROCK inhibitor ( Stemolecule Y-27632 , Stemgent ) for 24 hours . Media is replenished with fresh mTeSR1 media supplemented with 1% P/S daily . Cells were maintained at 37°C and 5% CO2 . Cells were maintained with phenol red–free mTeSR1 media ( 05876 , STEMCELL Technologies ) 1 day prior to live cell imaging . Three to four days after cells are plated and mature and healthy colonies are observed on 96- and 24-well imaging plates , the cells are stained with NucBlue Live ready probe reagent ( R37605 , ThermoFisher ) and CellMask Deep Red plasma membrane stain ( C10046 , ThermoFisher ) to visualize DNA and plasma membrane , respectively . The protocol is available online: http://www . allencell . org/uploads/8/1/9/9/81996008/sop_for_cellmask-and-nucblue_v1 . 0_1 . pdf . Phenol red–free mTeSR1 is preequilibrated to 37°C and 5% CO2 . 1X NucBlue solution made in preequilibrated phenol red–free mTeSR1 is spun for 60 minutes at 20 , 000 g . The 2X and 10X working stocks of CellMask Deep Red lot #1730970 and #1813792 , respectively , are made in 1X NucBlue solution . All solutions are kept at 37°C and 5% CO2 until used . The 100 μL and 400 μL of NucBlue solution are added per well of 96-well imaging plates and 24-well imaging plates , respectively , and incubated at 37°C and 5% CO2 for 20 minutes . An equal amount of CellMask Deep Red working stock is added to the wells containing NucBlue solution . Final dye concentrations in the wells are 1X NucBlue and 1X and 5X CellMask Deep Red lots #1730970 and #1813792 , respectively . Cells are incubated at 37°C and 5% CO2 for 10 minutes and gently washed with preequilibrated phenol red–free mTeSR1 . Fields of view as shown in Fig 4 that are acquired near the edge ( and the center as a control ) of hiPSC colonies receive an additional photoprotective cocktail treatment which serves to minimize singlet oxygen and free radical formation . The photoprotective cocktail is used at a working concentration of 0 . 3 U/ml ( 1:100 ) OxyFluor as defined by the OxyFluor product insert , with the addition of 10 mM sodium lactate and 1 mM ascorbic acid ( OxyFluor OF-0005 , Oxyrase ) . As per Roberts and colleagues [34] , cells were imaged on a Carl Zeiss spinning disk microscope with a Carl Zeiss 20×/0 . 8 NA plan APOCHROMAT or 100×/1 . 25 W C-APOCHROMAT Korr UV Vis IR objective , a CSU-X1 Yokogawa spinning disk head , and Hamamatsu Orca Flash 4 . 0 camera . Microscopes were outfitted with a humidified environmental chamber to maintain cells at 37°C with 5% CO2 during imaging . Cells are imaged immediately following the wash step and for up to 2 . 5 hours after dye addition on a Zeiss spinning disk microscope at 100× with the following general settings: 405 nm at 0 . 28 mW , 200 ms exposure; 638 nm at 2 . 4 mW , 200 ms exposure; acquiring each channel at each z-step . Experienced bioimage analysts drew outlines around nuclear boundaries on each slice of the 3D images and labeled background regions in a different color with GIMP ( https://www . gimp . org ) , an open-source drawing and annotation software . These annotated layers were then exported from GIMP as an image . This outline image is converted to 3D objects via a CellProfiler pipeline ( https://github . com/CellProfiler/tutorials/tree/master/Annotation ) , and an object label matrix image is exported , in which each object’s voxels are assigned a unique integer value . These label images are referenced as ground truth . | The “big-data revolution” has struck biology: it is now common for robots to prepare cell samples and take thousands of microscopy images . Looking at the resulting images by eye would be extremely tedious , not to mention subjective . Thus , many biologists find they need software to analyze images easily and accurately . The third major release of our free open-source software CellProfiler is designed to help biologists working with images , whether a few or thousands . Researchers can download an online example workflow ( that is , a “pipeline” ) or create their own from scratch . Pipelines are easy to save , reuse , and share , helping improve scientific reproducibility . In this release , we’ve added the capability to find and measure objects in three-dimensional ( 3D ) images . We’ve also made changes to CellProfiler’s underlying code to make it faster to run and easier to install , and we’ve added the ability to process images in the cloud and using neural networks ( deep learning ) . We’ve also added more explanations to CellProfiler’s settings to help new users get started . We hope these changes will make CellProfiler an even better tool for current users and will provide new users better ways to get started doing quantitative image analysis . | [
"Abstract",
"Introduction",
"Results",
"Future",
"directions",
"Materials",
"and",
"methods"
] | [
"open",
"science",
"blastocysts",
"methods",
"and",
"resources",
"engineering",
"and",
"technology",
"signal",
"processing",
"applied",
"mathematics",
"biologists",
"simulation",
"and",
"modeling",
"algorithms",
"developmental",
"biology",
"scientists",
"mathematics",
"science",
"and",
"technology",
"workforce",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"embryology",
"staining",
"computer",
"and",
"information",
"sciences",
"imaging",
"techniques",
"people",
"and",
"places",
"professions",
"cell",
"staining",
"science",
"policy",
"image",
"processing",
"careers",
"in",
"research",
"computer",
"software",
"software",
"engineering",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"open",
"source",
"software",
"physical",
"sciences",
"software",
"tools",
"image",
"analysis"
] | 2018 | CellProfiler 3.0: Next-generation image processing for biology |
The fungal pathogens Aspergillus fumigatus and Candida albicans are major health threats for immune-compromised patients . Normally , macrophages and neutrophil granulocytes phagocytose inhaled Aspergillus conidia in the two-dimensional ( 2-D ) environment of the alveolar lumen or Candida growing in tissue microabscesses , which are composed of a three-dimensional ( 3-D ) extracellular matrix . However , neither the cellular dynamics , the per-cell efficiency , the outcome of this interaction , nor the environmental impact on this process are known . Live imaging shows that the interaction of phagocytes with Aspergillus or Candida in 2-D liquid cultures or 3-D collagen environments is a dynamic process that includes phagocytosis , dragging , or the mere touching of fungal elements . Neutrophils and alveolar macrophages efficiently phagocytosed or dragged Aspergillus conidia in 2-D , while in 3-D their function was severely impaired . The reverse was found for phagocytosis of Candida . The phagocytosis rate was very low in 2-D , while in 3-D most neutrophils internalized multiple yeasts . In competitive assays , neutrophils primarily incorporated Aspergillus conidia in 2-D and Candida yeasts in 3-D despite frequent touching of the other pathogen . Thus , phagocytes show activity best in the environment where a pathogen is naturally encountered . This could explain why “delocalized” Aspergillus infections such as hematogeneous spread are almost uncontrollable diseases , even in immunocompetent individuals .
The frequency of invasive mycoses due to opportunistic fungal pathogens has increased significantly over the past decades . These infections are associated with high morbidity and mortality . They directly correlate with increasing patient populations that are at risk for developing invasive fungal infections . These populations include individuals undergoing solid-organ transplantation , bone marrow transplantation , and major surgery , as well as those with AIDS , neoplastic disease , immunosuppressive therapy , advanced age , and premature birth [1–3] . In this context , Aspergillus fumigatus can be regarded as the primary mold pathogen . Conidia of A . fumigatus are constantly inhaled . However , in immunocompetent individuals , mucociliary clearance and phagocytic defense prevent the disease . Alveolar macrophages ( AMs ) are the major phagocytes of lung alveoli . They , along with polymorphonuclear neutrophils ( PMNs ) , which are recruited during inflammation , are responsible for phagocytosis of A . fumigatus [1 , 4–6] . AMs kill conidia by producing reactive oxygen species [7] . The conidia that escape from the AMs germinate , but are attacked by PMNs that adhere to the surface of the hyphae and kill them by secretion of reactive oxygen species and degranulation [8–14] . In addition , PMNs are also able to kill resting or swollen conidia [1 , 6] . Furthermore , A . fumigatus antigens induce the activation and maturation of dendritic cells ( DCs ) [8] . A dysfunctional immune system , however , provides an opportunity for conidia to germinate and invade lung tissue [6 , 9] . Candida albicans is a normal component of mucosal flora , which in immunocompromised individuals can transform itself into an invasive pathogen [10] . The primary defense of the mucosal flora is formed by PMNs [11] , which are rapidly recruited to the sites of infections [12] and phagocytose the fungus in the tissue environment of microabscesses [13] . The process of A . fumigatus phagocytosis has been analyzed in detail using AMs from mice [7] or humans [14 , 15] , macrophage cell lines [15 , 16] , or DCs from the mouse [17] . In addition , the interaction of PMNs with hyphae in the presence of platelets has been studied [18] . However , none of these studies enabled the direct observation of the process itself by live cell imaging , since the available techniques did not allow the direct distinction between phagocyte and fungal components . Also , live cell imaging of phagocytosis of Candida has not been performed to our knowledge . Cell motility , however , is an essential requirement for the function of phagocytes . Defects in the cellular dynamics as a basis for defects in function are well known . One such example is that inhibition of the cytoskeleton of DCs [17] or AMs [15] severely inhibits the ability of these cells to phagocytose Aspergillus conidia . And genetic defects in actin polymerization are associated with increased susceptibility to infections [19] . Furthermore , it is generally assumed that physical contact between a phagocyte and a conidium inevitably leads to phagocytosis of the latter . Thus , it is unclear why previous studies addressing phagocytosis of A . fumigatus in vitro only found 50% of the phagocytes carrying conidia , while 50% of cells did not [7] . In addition , although PMNs are not very efficient at phagocytosing A . fumigatus in vitro , patients suffering from neutropenia are at a higher risk for developing invasive aspergilloses . Thus , mechanisms distinct from phagocytosis must exist , which allow PMNs to control fungal growth in the lung . Finally , until now studies on phagocytosis have ignored the location in the body at which this process occurs . One of the major features of in vivo phagocytosis is its almost ubiquitous three-dimensionality , in other words , the spatial arrangement of extracellular matrix proteins that acts as a scaffold for the attachment and movement of cells [20] . While conventional experimental setups , where cells are cultured in dishes covered with a liquid medium , constitute a two-dimensional ( 2-D ) system that does not allow migration in space , we mimic more physiological conditions by providing a three-dimensional ( 3-D ) collagen matrix as a substratum for cell migration and cell–cell communication [21 , 22] . We have previously used this system to demonstrate strong differences in the interaction of T cells with antigen presenting cells in 3-D versus 2-D environments [23] . Subsequently , the results have been confirmed by us [24] and others [25–27] in true lymphatic tissue . However , in vivo 2-D movements can also be observed , and phagocytosis of pathogens at the surface of a lung alveolus [28] is a prominent example of this . The inner part of an alveolus is a surface composed of type-I epithelial cells , which is covered by a water-based hypophase and a thin film of surfactant produced by type-II epithelial cells . Hypophase and surfactant form the surface-lining layer ( SLL ) [29] . The SLL is not a solid , migration-supporting scaffold but rather is thin ( in the range of 50 nm [30] ) and devoid of prominent internal structures [30–32] . Inhaled particles such as A . fumigatus conidia are drawn beneath the SLL , pressed to the epithelial surface by surface tension , and then are phagocytosed [30 , 33] . Normal cellular motility is essential for this process , as impaired motility in the thick mucus isolated from the lungs of individuals with cystic fibrosis is associated with decreased phagocytic activity of PMNs [34] . In contrast , alveolar DCs , the major antigen-presenting cells of the lung , must be able to phagocytose conidia from the interstitial tissue between alveoli , which is composed of collagenous fibers [35] , and thereafter migrate from the lung into draining lymph nodes to present pathogenic antigens to T cells [17] . Thus , this study was performed to obtain a dynamic picture of the phagocytosis of A . fumigatus and C . albicans by the main immune-effector cells . The goal was to directly compare the phagocytic efficiency on a per cell basis to elucidate the fate of individual fungal elements and finally study the role of environmental cues on the observed mechanisms .
The labeling of C . albicans with carboxyfluorescein-succinimidyl-ester ( CFSE ) was a reliable procedure and yielded bright and stable fluorescent cells useful for live cell imaging ( Figure 1H ) . However , the same protocol did not work sufficiently for A . fumigatus . Thus , we generated a strain of A . fumigatus with bright and stable endogenous fluorescence by producing the red fluorescent protein DsRed2 using the acuD promoter , which gave conidia with strong red fluorescence , and which is differentially regulated in hyphae of A . fumigatus [36] . The 5′ sequence , including the ATG start codon and codons encoding some N-terminal amino acids of the isocitrate lyase gene acuD , was fused in frame with the DsRed2 gene . All strains with an integration of the acuDp-DsRed2 plasmid exhibited detectable fluorescence well . One of the strains , designated AcuD-DsRed2–9 , was selected for further investigations . Southern analysis revealed that the strain contained two copies of the plasmid integrated at ectopic sites into the genome ( unpublished data ) . The expression of the DsRed2 gene fusions was monitored by growing the transformants in Aspergillus minimal medium ( AMM ) for 7 h ( germlings ) , or 16 h ( hyphae ) , and on AMM agar plates for 5 d with different carbon sources at 37 °C . Different developmental stages of A . fumigatus , i . e . , conidia , germlings , and hyphae , were studied . Conidia of AcuD-DsRed2–9 were derived from sporulating cultures on AMM agar plates and displayed strong fluorescence of stably deposited DsRed2 irrespective of the carbon source used ( Figure 1A shows glucose , unpublished data with ethanol as carbon source ) . In contrast to conidia , fluorescence of germlings and hyphae was dependent on the carbon source . Strong fluorescence was observed in germlings and hyphae during cultivation on ethanol ( Figure 1B and 1E ) . Cultivation of the fungus on glucose only led to faint , residual fluorescence in germlings ( Figure 1C ) whereas hyphae showed no fluorescence ( Figure 1F ) , as did conidia or hyphae of the wild-type controls ( Figure 1D ) . The observed slight fluorescence of germlings was possibly due to diffusion of the stable DsRed2 from red conidia into germlings . The carbon source-dependent expression pattern of DsRed2 observed here , verified that the isocitrate lyase promoter is exclusively active during growth conditions that require the glyoxylate cycle , i . e . , with ethanol or C2-generating carbon sources . Furthermore , bright fluorescence was observed when the conidia germinated in macrophages ( Figure 1G ) , suggesting that isocitrate lyase plays a significant role while growing in macrophages . Analysis of living cells by time-lapse microscopy and single-cell tracking showed that despite dynamic lamellipodia at the cell perimeter , neither AMs nor J774 cells migrated with a high velocity in either 2-D or 3-D environments ( Figure 2A ) . This is in accordance with our earlier data on macrophage migration [21] . PMNs and DCs , in contrast , efficiently migrated in both 2-D and 3-D environments; however , migration in the 3-D environment was somewhat slower than that in the 2-D environment . The percentage of cells in a tracked population migrating at any given time ( activity ) was significantly higher for AMs , J774 cells , and DCs in 2-D versus 3-D systems . Conversely , a significantly higher number of PMNs was found to be more mobile in 3-D systems than in 2-D systems ( Figure 2B ) . By analyzing tracks of single cells , we found that the migration of immune cells was almost perfectly random in all directions ( Figure 2C and 2D ) . The interaction of phagocytes with conidia of A . fumigatus was a highly dynamic process . In 2-D environments , actively migrating PMNs and DCs could be observed touching and phagocytosing multiple conidia within 1 . 5 h of observation ( Figure 3A; Videos S1 and S2 ) . Phagocytosis by AMs was less dynamic . Here , cells produced slight membrane protrusions towards nearby conidia before ingesting them ( Figure 3A; Video S3 ) . The phagocytosis of the macrophage cell line J774 was morphologically very similar to that of AMs . Here , we also observed that a single successful phagocytosis event could prime cells for more rapid additional phagocytosis events ( Video S4 ) . However , when transferred into a 3-D tissue-based environment , PMNs , AMs , and J774 cells were impaired in their ability to phagocytose conidia . This was observed despite frequent touching that could even lead to a slight displacement of individual conidia ( Figure 3B; Video S5 ) . DCs , in contrast , were equally efficient at ingesting conidia in 3-D as compared with the 2-D environment ( Figure 3B; Video S6 ) . A quantification of these phenomena showed that DCs were the most efficient cells for the phagocytosis of conidia , both in 2-D and 3-D environments , while all other phagocytes efficiently phagocytosed in a 2-D environment but were severely limited in their capacity for phagocytosis in 3-D ( Figure 3C ) . A frequent characteristic of phagocyte–conidium interaction was the dragging of conidia over long distances without obvious phagocytosis . This was especially pronounced in PMNs in 2-D environments and , to a lesser extent , also with DCs ( Figure 4A; Videos S7 and S8 ) . The dragging of multiple conidia by a single PMN often led to the formation of large aggregates of nonphagocytosed conidia in the center surrounded by several PMNs on the periphery ( Figure 4A; Video S7 ) . High-resolution electron microscopy showed that dragging cells could generate surface extensions that were reminiscent of phagocytic cups to individual conidia , while other conidia on the same cell were merely attached to the cell surface without the induction of membrane protrusions ( Figure 4B ) . Three dimensional reconstructions of confocal z stacks of PMNs incubated with conidia for 1 h clearly showed conidia completely internalized , and those conidia simply attached to the surface of a cell ( Figure 4C; Video S9 ) . Due to their inherently low mobility , neither AMs nor J774 cells dragged conidia . To study the frequency of conidium–phagocyte encounters , we analyzed the physical interaction of single cells with individual conidia in time-lapse video sequences ( Figure 5 ) . These analyses showed that touching rates between all types of phagocytes and conidia were higher in 2-D than in 3-D environments , reaching mean values between one to three contacts per cell per hour ( Figure 5A ) . Individual DCs were found touching as many as ten conidia in 1 h in 2-D systems ( Figure 5A ) . The percentage of cells touching conidia was consistently higher in 2-D environments , although , with the exception of AMs , these differences were not large enough to reach the level of significance ( Table 1 ) . The continuous filming of cells interacting with conidia also allowed the analysis of the fate of touched conidia . Thus we calculated the rate of touches ( phagocytosis touching index [PTI] ) that finally ended in successful phagocytosis as shown in Videos S5–S8 . These analyses showed , that for J774 cells , the PTI was lower in 3-D compared with that in 2-D environments . For AMs , the PTI in the 3-D environment was zero , meaning that none of the observed touches between AMs and conidia led to phagocytosis . For PMNs , fewer cells with an intermediate PTI were found in 3-D environments as compared with those in 2-D environments . At the same time , cells with a PTI of one remained equally as frequent . These differences , however , were not significant . Also , the PTI values of DCs did not change significantly with the environment ( Figure 5B ) . In the same way , we also analyzed the rate of touches that were followed by dragging the conidia rather than by phagocytosis ( dragging touching index [DTI] ) . Here , the efficient dragging of conidia by PMNs in the 2-D environment was almost completely lost in the 3-D environment ( Figure 5C ) . These data indicated that after transfer from a 2-D into a 3-D environment , physical contact formation between PMNs , AMs , and J774 cells with conidia was impaired , leading to strong defects in dragging and phagocytosis by these cells . AMs are exclusively located and function in the lungs , whereas PMNs can exert their activity in all tissues . Thus , we asked whether the defect in productive interaction with A . fumigatus conidia in 3-D environments was a general inability of PMNs to function in 3-D environments , or whether the type of pathogen determined cellular activity . Therefore , we measured the phagocytosis of the tissue-invading fungus C . albicans by PMNs and the macrophage cell line RAW 264 . 7 , which is derived from peritoneal ( i . e . , tissue-associated ) macrophages [37] . Live cell imaging showed intense contacts of both PMNs and RAW cells with yeast cells of C . albicans in 2-D ( Figure 6A; Videos S10 and S11 ) as well as in 3-D environments ( Figure 6B; Videos S12 and S13 ) . Nevertheless , only a few PMNs or RAW cells were found carrying C . albicans intracellularly in the 2-D environment , while the number of cells carrying single or multiple yeast cells in 3-D systems was almost four times higher when compared to the 2-D systems ( Figure 6C ) . Videos also showed the tendency of both PMNs ( Video S12 ) and RAW cells ( Video S13 ) to catch C . albicans cells even from great distances , while frequent contacts between phagocytes and C . albicans in 2-D environments led to no apparent physical association or a directional change of morphology of the cells . Due to the previous observation that one completed phagocytosis event could increase the frequency of subsequent events ( Video S4 ) , we reasoned that successful interaction with one pathogen might enable PMNs to phagocytose the other pathogen en route , even within the “wrong” environment . To test this assumption , we set up competitive phagocytosis assays , where PMNs were allowed to interact with a mixture of A . fumigatus conidia and C . albicans yeast cells in both 2-D and 3-D environments . Again , imaging showed intensive contacts of PMNs with both fungal elements in both environments . However , in 2-D environments , PMNs selectively chose A . fumigatus conidia in a manner indistinguishable from the situation with conidia alone ( Figure 7A; Video S14 ) , while in 3-D environments very efficient phagocytosis of C . albicans occurred despite clearly detectable contacts to nearby A . fumigatus conidia by the same PMN ( Figure 7B; Video S15 ) . A quantification of these results showed that even in competitive phagocytosis assays , individual PMNs could phagocytose one pathogen very efficiently while simultaneously ignoring the other ( Figure 7C ) . It was possible that the contact with collagen fibers , rather than the dimensionality , was responsible for the observed differences in phagocytosis . To test this assumption , we analyzed the phagocytosis of either A . fumigatus or C . albicans by PMNs on slides coated with collagen . Here , cells were in intimate contact with collagen fibers , but were not embedded three-dimensionally within the matrices . Interestingly , on collagen-coated slides , PMNs were as efficient in interacting with A . fumigatus conidia as observed before in our conventional ( plastic dish–based ) 2-D system . Accordingly , C . albicans yeast cells could not be phagocytosed by PMNs on collagen-coated slides ( Figure 7D ) . This suggested that it was indeed 3-D embedding within collagen that caused the observed differences in interaction/phagocytosis . The next question was whether this was a specific function of 3-D gels based on type-I collagen only . To test this , we measured interaction of PMNs with A . fumigatus conidia or C . albicans yeast cells in 3-D matrices composed of Matrigel ( a basement membrane entirely devoid of type-I collagen ) . Its main constituents are laminin and type-IV collagen . Confirming our previous assumptions , the interaction of PMNs with both fungal pathogens within 3-D matrices composed of Matrigel was indistinguishable from what had been observed before within 3-D collagen ( Figure 7E ) . In addition , when incubated on slides coated with Matrigel ( analogous to the experiment depicted in Figure 7D ) , the cellular behavior again was identical to what had been observed in the other 2-D systems ( unpublished data ) . These results suggest that it is indeed the dimensionality of the environment that critically influences the interaction of phagocytes with fungal pathogens . Finally , it was important to investigate whether the observed phenomena were restricted to mouse phagocytes only . To analyze this , we investigated the interaction of human peripheral blood PMNs from three independent healthy donors with both fungal pathogens and in 2-D and 3-D environments . We found that the behavior of human cells closely mimicked the murine PMNs , in that phagocytosis of A . fumigatus was much better in the 2-D compared to the 3-D environment . We also noted that the interaction with C . albicans was exactly the opposite ( Figure 7F ) . Thus , the dependence of phagocytes , at least of PMNs , on the dimensionality of the environment appears to be a general phenomenon not restricted to only mouse cells .
In this study we analyzed the cellular dynamics underlying the interaction and phagocytosis of two important fungal pathogens , A . fumigatus and C . albicans , using the major phagocytes of the mammalian body . Continuous imaging , which is indispensable for these studies , requires the reliable identification of both cells and fungi under constant light exposure . Expression of DsRed2 under control of the acuD-promoter resulted in bright , stable fluorescence in conidia of A . fumigatus , which is necessary for long-term live-cell imaging analyses . At the same time , this transgenic fungus confirmed our earlier data on the environmental conditions leading to activity of the acuD-promoter [36] , and thus adds additional proof to the efficacy of the fluorescent reporter gene strategy for studies on the activity of novel promoter elements in filamentous fungi [38] . The fact that the acuD-promoter was active during germination of conidia and outgrowth of macrophages made this promoter a valuable tool for following phagocytosis of conidia in cells . The acuDp-DsRed2 transgenic A . fumigatus made it possible to capture dynamic pictures of the phagocytosis process of conidia by the main cell types that interact with the fungus during natural infection [39] . This also made it possible to evaluate the fate of individual conidia and cells over a period of several hours and , finally , an assessment of phagocytic efficiency on a per cell basis could be conducted . The data indicate that the majority of cells were not able to phagocytose each conidium they touched . Although a number of cells , especially among AMs , were 100% successful ( each contact led to phagocytosis of the touched conidium ) , a larger fraction of contacts between phagocytes and conidia did not end in phagocytosis , but rather in the release of conidia by the phagocytes . This could be observed not only with cells that had not phagocytosed any conidium , but also with cells that had already successfully phagocytosed conidia before . Hence , phagocytosis is not always the consequence of physical contact between a phagocyte and a pathogen , but could result from other , yet undisclosed , factors within the phagocyte and pathogen that need to coincide for phagocytosis to occur . This might explain why , in conventional phagocytosis assays of conidia , only ∼50% of AMs finally carry conidia [7] , and after infection in vivo only 40%–50% of all airway DCs showed ingested conidia [17] . In addition , live microscopy also detected a novel , previously unrecognized means of phagocyte–pathogen interaction: the dragging of large numbers of conidial elements by individual cells without phagocytosing them . This was especially prominent among PMNs . The dragging of conidia by highly motile PMNs could lead to the collection of almost all conidia visible within a given field of view and to the formation of aggregates of conidia surrounded by highly motile PMNs . By this means , PMNs , which are not as efficient in phagocytosing conidia as AMs , might take control over large numbers of inhaled conidia . These aggregates have been isolated recently from infected murine lungs and shown to be associated with large amounts of reactive oxygen species [40] , which could be a possible mechanism employed by PMNs to control Aspergillus infection [41] . Future work is needed to show whether such aggregates between PMNs and conidia can also be isolated from the lungs of humans exposed to Aspergillus conidia . The lack of PMN–Aspergillus aggregates in patients succumbing to the infection might explain why neutropenia is an especially dangerous condition for infection with A . fumigatus . The most important finding of this study was the strong dependence of the phagocytic efficiency of PMNs and AMs ( excluding DCs ) on the dimensionality ( i . e . , 2-D or 3-D ) of the environment , which was retained irrespective of the presence or absence of collagen . It is well known that in vivo DCs must take up conidia from airways and then migrate through the interstitial lung tissue to reach draining lymph nodes for antigen presentation to T cells [17 , 35 , 42] . Thus , DCs must be able to function in both environments . However , PMNs and AMs interact with inhaled conidia in the lung alveolus on the surface of the alveolar epithelium beneath a very thin , liquid SLL [30] , a situation representing a typical 2-D environment . Nevertheless , the ability to phagocytose C . albicans shows that PMNs are not generally unable to function in 3-D environments . Strikingly , hyphae , the tissue invasive form of A . fumigatus , were also efficiently attacked by a large number of PMNs in our 3-D environment ( unpublished data ) . In the light of these observations , it is tempting to speculate that phagocytes may have co-evolved with different pathogens or even with their morphotypes in a way to optimally recognize them in the environment where the encounter within the mammalian body is most likely to occur . The ability to differentiate between C . albicans morphotypes also was previously demonstrated for DCs [43] , albeit only for 2-D environments . The inability of PMNs to recognize pathogens in the inappropriate environment may have important consequences . For example , mice , which can accept a lung infection with 2 × 108 A . fumigatus conidia without notable dysfunction [44] , are killed by 5 × 106 conidia when administered intravenously [45] . Thus , mice are >40 times as sensitive to the same infection when the infection occurs in a non-natural environment despite the presence of a large number of PMNs in the bloodstream . This might also explain why the transition from a local to systemic infection , which occurs by hematogeneous spread of A . fumigatus , shows such high mortality [9] . Currently , the molecular explanation for these findings is puzzling . It has been shown that human macrophages can more efficiently phagocytose and kill bacteria [46] or Candida [13] in contact with a type-I collagen gel . This was associated with increased Fc-receptor or complement-receptor–mediated uptake and induction of phagolysosomal fusion after contact with collagen . However , the same effects obviously cannot account for the decrease in activity of PMNs against A . fumigatus in collagen . Generally , the cellular receptors responsible for ingestion of A . fumigatus by phagocytes are poorly defined . It has been shown that surfactant is important for this process [47] . However , since purified cells were used , participation of surfactant in the processes described here can be excluded . Previously it had been shown that PMNs were able to rapidly change their transcriptional profile in response to the environment [48] . Thus , the de novo expression , or down-regulation of receptors in 3-D compared to 2-D environments is likely . The same holds true for the pathogens themselves . For example , a simple temperature shift was shown to alter the expression of several hundred genes in A . fumigatus [49] , so it is very likely that genes are differentially expressed in A . fumigatus after transfer of the culture from liquid culture into collagen . Such an adaptation might account for the inability of phagocytes to recognize the fungus in this environment . The ability to adapt the transcriptional profile to the environment was also shown in C . albicans [10 , 50] . Thus , future work needs to define genetic alterations in both the pathogen and the phagocyte in relation to the environment in order to obtain a complete understanding of the cellular processes underlying dragging , phagocytosis , and the killing of fungal pathogens in 2-D and 3-D environmental niches .
CEA17 is a uracil-auxotrophic A . fumigatus strain , which encodes a mutated pyrG gene [51] . The A . fumigatus strain AcuDp-DsRed2 was derived from the CEA17 strain following transformation with plasmid pacuDp-DsRed2-pyrG . The strain carries two copies of the acuDp-DsRed2 gene fusion integrated ectopically into the genome . Vectors and plasmids were propagated in Escherichia coli XL1 Blue MRF′ ( Δ ( mcrA ) 183 Δ ( mcrCB-hsdSMR-mrr ) 173 endA1 supE44 thi-1 recA1 gyrA96 relA1 lac [F′ proAB lacIq ZΔM15 Tn10 ( TetR ) ] ) ( Fermentas , http://www . fermentas . com ) . For the cultivation of A . fumigatus strains , AMM with 1% ( w/v ) glucose or 50 mmol ethanol as a carbon source was used [52] . Conidial suspensions were obtained from AMM agar plates after 5 d of cultivation [52] . Fungal and bacterial strains were grown at 37 °C . Spore suspensions were prepared as described without the addition of antibiotics [52] . E . coli strains were grown on LB agar plates or in LB medium at 37 °C . Ampicillin was added to give a final concentration of 100 μg/ml . C . albicans ( based on SC5314 , obtained from R . Calderone , Georgetown University ) was grown to stationary phase in YPD medium ( Sigma , http://www . sigmaaldrich . com/Brands/Sigma . html ) at 30 °C with orbital shaking at 160 rpm . For fluorescence labeling , 1 × 108 yeasts were harvested by centrifugation ( 16 , 000g , 5 min , 24 °C ) , washed twice in 1 ml PBS , and stained with CFSE ( 0 . 5 μmol in 1 ml PBS/0 . 1% DMSO ) ( Invitrogen , http://www . invitrogen . com ) for 1 . 5 h at 37 °C . Yeast cells were washed three times in PBS to remove remaining dye before use . An A . fumigatus sequence containing the promoter from the acuD gene encoding isocitrate lyase [36] was used . The pDsRed2 vector was obtained from Clontech ( http://www . clontech . com ) . The A . fumigatus pyrG gene was used as a selection marker [53] . The pyrG gene was synthesized by PCR amplification using the oligonucleotides pyrG_Afum_notI_1 ( 5′-GCGGCCGCACAGCTATGCGACCG-3′ ) and pyrG_Afum_notI_2 ( 5′-GCGGCCGCATATCTCTGGTTGGAG-3′ ) , which encode NotI restriction sites ( underlined ) and chromosomal DNA of A . fumigatus as the template . For generation of the acuDp-DsRed2 gene fusion the 5′ sequence of the A . fumigatus isocitrate lyase gene ( 942 bp ) , including the ATG , was amplified by PCR , using oligonucleotides AfAcuD_upst_Bam ( 5′-CGGATCCGAAGGACAGGAAC-3′ ) and AcuD_rev_KpnI ( 5′-CTGGATCCAAACCCATTGTGACAGGTATGAAGAGG-3′ ) and chromosomal DNA of A . fumigatus as the template . The oligonucleotides encoded BamHI and KpnI sites at the 5′ and 3′ ends , respectively ( underlined ) . The PCR fragment was cloned into the pCR2 . 1 vector . The resulting plasmid was digested with BamHI and KpnI . The DNA fragment obtained was ligated into plasmid pDsRed2 , which had also been digested with BamHI and KpnI , to give the plasmid pacuDp-DsRed2 . The PCR amplified pyrG gene was integrated into plasmid pacuDp-DsRed2 by ligation into its single NotI site to yield plasmid pacuDp-DsRed2-pyrG ( Figure 1B ) . Transformation of A . fumigatus was performed as described [54] . Microscopic analyses shown in Figure 1 were performed with a Leica DM4500 B microscope with a filter set of BP 546/12 for excitation and BP 605/75 for emission . Images were obtained and processed with Leica Application Suite 2 . 3 . 4 R2 ( Leica Microsystems , http://www . leica-microsystems . com ) . Fluorescence analysis of Figure 1G was performed with an Axiovert 200 M/LSM 510 META laser scanning confocal microscope ( Zeiss , http://www . smt . zeiss . com ) . DsRed2 and 4′ , 6-diamidino-2-phenylindole-dihydrochloride ( DAPI ) were excited by a laser line of 542 and 364 nm , respectively . Fluorescence signals were detected using a 385-nm long pass filter for DAPI and by 560–615-nm band pass filters for DsRed2 . Images were acquired using the LSM-510-META 3 . 2 software ( Zeiss ) . Figures were assembled with Adobe Photoshop ( http://www . adobe . com ) . PMNs and DsRed A . fumigatus conidia were mixed at a ratio of 1:5 and incubated for 1 h over 12-mm poly-L-lysine–coated cover slips . Cells were fixed in 4% paraformaldehyde ( PFA ) ( Sigma ) at ( pH 7 . 4 ) for 20 min at room temperature and then washed three times with prewarmed PBS . The cells were then permeabilized with 4% PFA and 0 . 1 % Triton X-100 ( Sigma ) . After washing with PBS , the cells were blocked with a solution of PBS containing 1% BSA and 5% horse serum ( Sigma ) . Staining for actin cytoskeleton was done using Alexa 488 labeled phalloidin ( 2 U/ml ) ( Molecular Probes , http://probes . invitrogen . com ) for 45 min to 1 h . Cover slips were mounted on clean glass slides with Mowiol ( Calbiochem , http://www . emdbiosciences . com ) with 0 . 01% paraphenylene diamine ( Sigma ) . Cells were imaged with an Olympus LSM confocal microscope ( Fluoview 1000 ) with a 100-× objective . A 3-D rendering of confocal z stacks was done using the Volocity software package ( version 4 . 0; Improvision , http://www . improvision . com ) . BALB/c bone marrow DCs were generated in 8-d cultures as described [55] . Cell lines secreting murine granulocyte-macrophage colony-stimulating factor ( GMCSF ) or IL-4 were kindly provided by Thomas Blankenstein from the Max Delbrück Center for Molecular Medicine ( MDC ) , Berlin ( http://www . mdc-berlin . de ) . PMNs were obtained by positive selection from mouse bone marrow ( BM ) . BM cells were prepared by flushing the femurs and tibiae of BALB/c mice with PBS + 1% FCS ( v/v ) . Following erythrocyte lysis , the cells were incubated with Fc–Block ( BD Biosciences , http://www . bdbiosciences . com ) and then subjected to cell sorting by Gr-1–labeled magnetic particles ( clone RB6-8C5 , BD Biosciences ) following the manufacturer's instructions . The purity of the cells was >97% as determined by FACS analysis . AMs were obtained by washing the trachea and lungs of BALB/c mice with PBS through a 22G plastic catheter ( Braun , http://www . bbraun . de ) to obtain bronchoalveolar lavage fluid . After erythrocyte lysis , the cells were resuspended in complete medium supplemented with glutamine , penicillin , and streptomycin . The cells were kept on ice until use . J774 cells were cultured in BioWhittaker's X-Vivo 15 medium ( Cambrex , http://www . cambrex . com ) . Before use , the cells were stimulated overnight with 2 . 5 U/ml interferon-γ ( Boehringer Ingelheim , http://www . boehringer-ingelheim . com ) . RAW 264 . 7 macrophages [37] ( American Type Culture Collection , http://www . atcc . org ) were maintained in RPMI ( Gibco , http://www . invitrogen . com ) containing 10% FCS at 37 °C and harvested by scraping with a rubber policeman . The cells were subjected to no more than 20 passages . Human PMNs were derived from the peripheral blood of healthy volunteers . Briefly , freshly drawn blood was diluted with HBSS without CaCl2 and MgCl2 ( Gibco ) and layered over PolymorphPrep ( Axis-Shield PoC AS , http://www . axis-shield . com ) according to the manufacturer's instructions . PMNs were carefully removed and resuspended in RPMI supplemented with 5% pooled human serum ( Chemicon , http://www . chemicon . com ) following washing and erythrocyte lysis with ACK buffer ( Cambrex ) . All experiments with human cells were done with 5% ( v/v ) pooled human serum . For outgrowth experiments of germinating conidia from macrophages ( Figure 1G ) , J774 macrophages were cultured in RPMI complete medium ( Cambrex ) + 5% ( v/v ) FCS ( = RPMIF ) . Macrophages were incubated with a ratio of two conidia per macrophage for 2 h . The cells were washed extensively with RPMIF and incubated for 4 h with RPMIF containing 25 mmol imidazole . The cells were then fixed with 3 . 8% ( v/v ) para-formaldehyde for 10 min at room temperature and stained with DAPI . The samples were analyzed with fluorescence microscopy . A total of 1 × 106 purified cells were mixed with 0 . 5–1 × 107 filtered ( BD Falcon cell strainer ) conidia of strain AcuDp-DsRed2 or 3 × 106 Candida yeasts in 66 μl complete medium ( CM ) containing RPMI 1640 supplemented with NEAA ( 1 × ) , FCS ( 10% , v/v ) , L-glutamine ( 2 mmol ) , HEPES ( 10 mmol ) , sodium pyruvate ( 1 mmol ) , β-mercaptoethanol ( 50 μmol ) , and penicillin/streptomycin ( 100 U/ml ) . The suspension was mixed with 133 μl of type-I collagen stock solution ( Vitrogen-100; Nutacon , http://www . nutacon . nl ) to a final collagen concentration of 1 . 7 mg/ml and poured into a tracking chamber as described [23 , 24] . Fluorescence and cell interactions were monitored simultaneously at 37 °C at two frames/min using an Olympus BX61 microscope with a 60 × LUMPLFL W/IR ( NA 0 , 9 ) lens , together with the cellR software ( version 2 . 1 ) from Olympus Biosystems ( http://www . olympus-europe . com ) . For observations in a liquid medium , immune cells and conidia were mixed at a ratio as mentioned above to a final volume of 200 μl in complete medium . These cell suspensions were poured into glass chambers , and microscopy was performed focusing on the bottom of the chamber . Alternately , imaging of cells and pathogens was carried out over collagen-coated surfaces that were prepared by applying a very thin coat of type-I collagen at a concentration of 1 . 7 mg/ml on a glass slide to rule out the effect of collagen on 2-D systems . Matrigel basement membrane matrix ( BD Biosciences ) , containing laminin as a major component , was used as an alternative for the type-I collagen matrix and was diluted with a mixture of cells and respective pathogens in PBS with 1% BSA at the ratio as mentioned above , so that its final concentration was 1 . 7 mg/ml . Cell migration was analyzed by computer-assisted cell tracking using a software program developed for this study as described [21 , 56] . Briefly , the time-lapse video was displayed on a computer screen . For videos in a collagen matrix , this was a 2-D projection of a 3-D image . From 40 to 60 cells were then randomly marked to give an unbiased representation of the cell population . The cell movements were then followed with a trackball in both media and collagen matrices . Tracking was stopped in case a cell disappeared within the collagen matrix or left the field of view . From these tracks cell velocities were calculated and normalized to the true dimensions of the field of view and expressed as micrometers per minute . In addition , the percentage of cells migrating at a given time point were expressed as activity . The efficiency of phagocytosis and dragging was obtained as follows: all cells visible in the video sequences were analyzed for their physical association with conidia ( touching ) . Then it was further observed whether this touch led to phagocytosis of the conidium , which could last from a few seconds up to several hours . The ratio of completed phagocytosis events over the number of observed touches was calculated as PTI for each cell . The DTI was calculated in a similar manner for interactions that did not lead to phagocytosis but nevertheless showed considerable displacement of conidia , often over hundreds of micrometers . Samples were fixed in 5% formaldehyde and 2% glutaraldehyde in cacodylate buffer ( 0 . 1 M cacodylate , 0 . 01 M CaCl2 , 0 . 01 M MgCl2 , and 0 . 09 M sucrose [pH 6 . 9] ) for 1 h on ice and washed with cacodylate buffer . We coated 12-mm cover slips with poly-L-lysine ( Sigma ) for 10 min , and then they were washed in distilled water and air dried . We placed 30 μl of the fixed samples on a cover slip and allowed it to settle down for 10 min . Cover slips were then fixed in 2% glutaraldehyde in cacodylate buffer ( 5 min ) and washed with TE-buffer ( 20 mM TRIS and 1 mM EDTA , [pH 6 . 9] ) before dehydrating in a graded series of acetone ( 10% , 30% , 50% , 70% , 90% , and 100% ) on ice for 15 min for each step , critical-point dried with liquid CO2 ( CPD 30; Balzers , http://www . oerlikon . com ) and covered with a gold film by sputter coating ( SCD 40 , Balzers ) , before being examined in a field emission scanning electron microscope ( Zeiss DSM 982 Gemini ) using the Everhart Thornley SE detector and the inlens detector in a 50:50 ratio at an acceleration voltage of 5 kV . The values for cells phagocytosing conidia in a collagen culture versus in a liquid culture were analyzed for significant differences using the nonparametric Mann-Whitney U-test . Several data contained only zero values , which precluded the use of the Mann-Whitney U-test . In such cases , the Wilcoxon rank sum test was used to estimate significance . All other statistical analyses were done using the Student's unpaired t-test . p-Values < 0 . 05 were considered significant ( * ) , values < 0 . 01 highly significant ( ** ) .
The National Center for Biotechnology Information ( NCBI ) CoreNucleotide ( http://www . ncbi . nlm . nih . gov ) accession numbers for the genes and gene products discussed in this paper are acuD ( gi|44844013 ) and acuD promoter ( AJ620297 ) . | Aspergillus fumigatus and Candida albicans are the most common of all human pathogenic fungal germs . Normally , inhaled Aspergillus spores are destroyed by alveolar macrophages and polymorphonuclear neutrophils ( PMNs ) , both of which are lung phagocytes , i . e . , cells that kill inhaled microbes by ingestion . In contrast , C . albicans is a normal constituent of the human gut flora that is controlled by tissue-resident PMNs . If immune control is lost , both fungi grow into the surrounding tissue and cause life-threatening infections . To investigate how phagocytes function in the disparate environments of lung air sacs ( lacking a definite matrix-composition [two-dimensional ( 2-D ) ] ) or mucosal tissues ( providing a three-dimensional [3-D] space ) , the authors mimicked 2-D and 3-D environments and analyzed the process of ingestion , called phagocytosis , by PMNs and other phagocytes . Phagocytosis was a dynamic cellular process where distinct cells showed vastly different behavior . The environmental setup ( 2-D versus 3-D ) had a profound impact on the cell's ability to phagocytose . Aspergillus conidia were much better ingested in 2-D systems , while Candida yeasts were only ingested in 3-D systems , even if the other pathogen was present . This was true for different 2-D and 3-D systems and for both cells of mice and humans . Besides providing a comprehensive analysis of the cellular movements underlying phagocytosis , the results also suggest an evolution of phagocytes to optimally recognize fungal pathogens in the environment of natural infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"infectious",
"diseases",
"cell",
"biology",
"yeast",
"and",
"fungi",
"in",
"vitro",
"immunology",
"microbiology",
"mus",
"(mouse)"
] | 2007 | Environmental Dimensionality Controls the Interaction of Phagocytes with the Pathogenic Fungi Aspergillus fumigatus and Candida albicans |
Activation-induced deaminase ( AID ) converts C to U and 5-methyl-C to T . These mutagenic activities are critical to immunoglobulin ( Ig ) gene diversification and epigenetic reprogramming , but they must be tightly controlled to prevent compromising cell fitness . AID acts in the nucleus but localizes predominately to the cytoplasm . To address this apparent paradox , we have carried out time-lapse imaging of AID in single living B cells and fibroblasts . We demonstrate that AID enters the nucleus in brief ( 30 min ) pulses , evident in about 10% of cells in the course of a single cell cycle ( 24 hr imaging ) . Pulses do not depend on AID catalytic activity , but they are coordinated with nuclear accumulation of P53 . Pulsing may protect cells from pathologic consequences of excess exposure to AID , or enable AID to synchronize its activity with transcription of genes that are AID targets or with nuclear entry of factors that act at sites of AID-catalyzed DNA deamination to promote Ig gene diversification or epigenetic reprogramming .
Activation-induced deaminase ( AID ) is essential for the three processes that diversify immunoglobulin ( Ig ) gene sequence and structure: somatic hypermutation , class switch recombination , and gene conversion [1–6] . In B cells of the germinal center and other lymphoid tissues , AID deaminates C to U at the rearranged and transcribed Ig genes and factors from the base excision repair or mismatch repair pathways then remove the U , thereby creating a DNA nick that can initiate somatic hypermutation , gene conversion or class switch recombination [7 , 8] . AID also erases methylation marks to reprogram transcription in very early development [9–12] and in germinal center B cells [13] by deaminating 5-meC to T , which is then replaced by C lacking a methylation mark . Even though AID must act in the nucleus , it localizes primarily to the cytoplasm . Cytoplasmic localization of AID has been documented in the vast majority ( >90% ) of cells in a variety of tissues and cell populations , including fixed germinal center B cells that express endogenous AID and cultured cells of other lineages that express fluorescent tagged AID ( e . g . [14–18] ) . AID can cause genomic instability or chromosomal translocations [19–23] and the cytoplasmic localization of AID protects cells from deleterious consequences of nuclear AID activity . AID is sequestered in the cytoplasm [24] and requires active transport to enter the nucleus [25] . AID nuclear persistence is limited both by ubiquitin-dependent proteolysis [26 , 27] and export by the CRM1-dependent nuclear export pathway [15 , 16 , 28] . Nuclear AID is potentially toxic , and mutations that impair export or proteolysis of nuclear AID can compromise cell viability [22 , 27 , 29] . A great deal is known about AID regulation , but most current understanding derives from analyses of cell populations . Analyses of single living cells can provide new insights into regulation , as gene expression and protein activity and localization can vary dynamically among cells within a nominally homogeneous population [30] . We have therefore carried out live cell imaging of cells expressing AID tagged with fluorescent protein . Here , we demonstrate that AID enters the nucleus in brief pulses , of about 30 min duration . Pulses are observed in about 10% of cells in asynchronous culture over the course of 24 hr imaging . They are independent of AID catalytic activity . AID nuclear accumulation stimulates nuclear accumulation of P53 , but while AID appears to be purged from the nucleus at the end of each pulse , nuclear P53 persists in cells that have pulsed . These results identify a new pathway of AID regulation .
AID is expressed in B cells and in non-lymphoid cells . To facilitate imaging , we first examined AID-mCherry transfectants of human HT1080 cells , which derive from a fibrosarcoma and form stable attachments to the surface of a slide that maintain cell position during extended periods of imaging . Cells were imaged at 10 min intervals , using a DeltaVision microscope equipped with an environmental chamber to maintain the cells at 37°C , 5% CO2 . Time-lapse imaging produced a striking result . The AID-mCherry signal was cytoplasmic in most cells , as predicted by a number of studies documenting cytoplasmic localization of AID; and dispersed throughout the cell during mitosis ( S1 Fig , S1 Movie ) , as previously reported [6] . However , in a fraction of cells , AID-mCherry appeared to localize to the nucleus in brief pulses ( S1 Movie ) , illustrated by two representative pulses of 40 ( above ) and 30 min ( below ) duration ( Fig 1A ) . We then analyzed AID-GFP transfectants , to confirm that results were not an artefact of the mCherry tag . In order to distinguish the nucleus and cytoplasm in these living cells , we examined HT1080 cells expressing both AID-GFP and the mKO2-CDT1 fusion protein , which confers a red signal to nucleus during G1 phase [31] . Like the AID-mCherry signal , the AID-GFP signal was cytoplasmic in most cells and appeared in brief pulses in the nuclei of a minority of cells . Consecutive time-lapse images of two representative cells at high resolution illustrate how the green AID-GFP signal is initially exclusively cytoplasmic ( t = 0 ) ; then nuclear , overlapping the red nuclear mKO2-CDT1 signal ( t = 10 , 20 min ) ; then cytoplasmic once again ( t = 30 min; Fig 1B; S2 and S3 Movies ) . To establish whether pulses occur in B cells as well as fibroblasts , we analyzed AID-mCherry transductants of the human B cell line , Ramos . Ramos cells derive from a Burkitt lymphoma and express endogenous AID , and they carry out ongoing somatic hypermutation of their endogenous VH and VL regions . The small size of B cells makes them very sensitive to radiation , so cells were imaged at 15 min rather than 10 min intervals , to help preserve viability . In Ramos AID-mCherry transductants , the AID-mCherry signal appeared in the nucleus in transient pulses like those observed in HT1080 transfectants expressing fluorescent-tagged AID ( S2 Fig , S4 and S5 Movies ) . We conclude that AID localizes to the nucleus in short pulses in B cells as well as fibroblasts . Pulses have previously been shown to regulate nuclear access of about a dozen proteins in yeast , all of them transcription factors [32–34]; and five human proteins , including P53 [35] and NF-κB [36–38] . These factors all respond to specific signals that synchronize pulses among cells in a population . To determine if AID nuclear pulses occur in response to analogous cues , we carried out detailed characterization of pulsing in individual HT1080 AID-mCherry transfectants . By time-lapse imaging at 10 min intervals over the course of 24 hr , we recorded 289 individual cells ( 948 hr total imaging time ) . We observed one or more AID nuclear pulses in 36 cells ( 12 . 5%; S1 Table ) , but pulses were not synchronous among cells in a population . Thus , pulses do not occur in response to subtle perturbations of culture conditions . The timeline for pulsing in each cell was diagrammed to indicate when pulses occurred with respect to cell division , and individual timelines were ranked based on the interval between capture of the first image ( t = 0 ) and cell division ( Fig 2A ) . Most of the cells in which pulses were observed carried out cell division during the observation period ( 79%; S1 Table ) , evidence of their viability . Pulses occurred at variable intervals before and after cell division , so cell division does not regulate pulsing . A total of 138 pulses were observed in the 36 cells that pulsed , or an average of 3 . 8 pulses in each cell that pulsed ( S1 Table ) . Individual timelines contained from 1–15 AID-mCherry pulses ( Fig 2B ) . Pulse duration averaged 27 ± 9 min ( N = 138; S1 Table; S3A Fig ) . Duration did not change over the course of imaging and thus appeared to be independent of the number of pulses before or after a given pulse ( S3A Fig ) . Intervals between pulses were irregular , although in some cells several pulses occurred in short succession , followed by an interval with no pulses . In 17 of the 26 timelines in which cell division was documented , both daughter cells could be followed during subsequent imaging . In most cases ( 15 of 17 , or 88% ) , only one of the two daughter cells pulsed . In the two cases in which both daughter cells pulsed , pulses did not appear synchronized ( Fig 2A ) . The timer that initiates the pulse thus appears not to be heritable . Some proteins that pulse are under feedback regulation that maintains constant protein abundance despite changes in gene dosage or expression [32 , 33 , 39 , 40] . If this is the case for AID , then the GFP and mCherry signals are predicted to be lower in HT1080 AID-mCherry AID-GFP double transfectants than in single transfectants . We established that the GFP and mCherry signals in double transfectants could be readily distinguished by flow cytometry ( S3A Fig ) . We then showed that expression of AID-mCherry in AID-GFP transfectants did not diminish the GFP signal; and conversely , that expression of AID-GFP in AID-mCherry transfectants did not diminish the mCherry signal ( S4B Fig ) . Thus , abundance of AID is not subject to feedback regulation . AID localizes to the nucleus in response to DNA damage induced by ionizing radiation and other agents [28 , 41 , 42] . This raised the possibility that AID nuclear pulses are a response to damage caused by DNA deamination catalyzed by nuclear AID present at levels too low to detect by imaging . To test this , we asked if AID catalytic activity was necessary for nuclear pulses by analysis of HT1080 transfectants expressing AIDH56A-mCherry . This derivative bears a mutation at a conserved histidine that coordinates an active site zinc ion essential for deaminase activity . This site has previously been targeted to impair deaminase activity because mutation of His to Ala at the corresponding residue of the active site of the highly related murine adenosine deaminase reduces catalytic activity without altering secondary or tertiary structure [27 , 43] . Live cell imaging of AIDH56A-mCherry transfectants identified clear nuclear pulses ( Fig 3; S6 and S7 Movies ) . Thus , nuclear pulses are independent of damage initiated by AID deaminase activity . By time-lapse imaging of 506 individual HT1080 AIDH56A-mCherry transfectants proliferating in asynchronous culture , at 10 min intervals over the course of 24 hr , we documented one or more nuclear pulses in 52 cells ( S1 Table; S6 and S7 Movies ) . Pulses were not synchronous among the cells in a population . Timelines ( 46 total ) of cell division and pulsing were generated in a total of 1543 hr of imaging ( Fig 3A ) . Most of the cells analyzed ( 30/46 , 65% ) divided during the observation period ( S1 Table ) . Individual cells exhibited from 1–10 pulses , with a significant fraction of the cells ( 26/52 , 50% ) pulsing only once ( Fig 3B ) . Pulse duration averaged 34 ± 13 min ( N = 124; S1 Table; S3B Fig ) . The average number of pulses ( among cells that pulsed ) was 2 . 4 per cell ( 124 pulses/52 cells ) , significantly fewer than the 3 . 8 pulses per cell in AID-mCherry transfectants ( p = 0 . 03 ) . We previously showed that mutation of AID catalytic activity ( H56A or H56R ) reduced both nuclear and cytoplasmic levels of AID to about 25% those of wild type AID [27] , as is evident in representative images of individual examples of AIDH56A-mCherry nuclear pulses ( Fig 3C ) . That fewer pulses per cell were documented in HT1080 AIDH56A-mCherry transfectants could reflect difficulty in scoring pulses with a fainter nuclear signal . Alternatively , AID catalytic activity may be a positive regulator of pulse frequency , even though it is not required for pulsing . Pulses limit AID access to the nucleus and this will limit DNA deamination , providing a physiologic check on DNA damage potentially caused by AID . High content screening ( HCS ) microscopy has identified roles for both nuclear export and proteolysis in regulating AID nuclear levels in Ramos human B cells [27] . We asked if both mechanisms are also active in HT1080 cells by using HCS microscopy to analyze nuclear and cytoplasmic signals in populations of cells treated with MG132 , which inhibits ubiquitin-dependent proteolysis; LMB , which inhibits CRM1-mediated nuclear export; or both ( S8 and S9 Movies ) . Treatment of HT1080 AID-mCherry transfectants with MG132 alone had little effect . Treatment with LMB alone caused nuclear accumulation of AID-mCherry in all cells . Treatment with both LMB+MG132 caused a much greater increase in nuclear abundance than treatment with LMB alone , but did not affect cytoplasmic abundance . In HT1080 cells , the rate of AID nuclear degradation was comparable in all phases of cell cycle ( S5C and S5D Fig ) . This contrasts with Ramos B cells , where nuclear proteolysis of AID is slightly slower in G1 phase than in S-G2/M phase [27] . Thus , both nuclear export and nuclear proteolysis regulate AID in HT1080 cells , and either process could in principle purge nuclear AID to terminate a pulse . To further examine the role of nuclear export in terminating AID pulses , we carried out live cell imaging of HT1080 transfectants expressing AIDF193A-mCherry , which carries a mutation in the NES that prevents nuclear export via the CRM1-mediated pathway . In normally proliferating cells , AIDF193A-mCherry was absent from the cytoplasm and produced a strong nuclear signal on fluorescence imaging [27 , 44] . The mCherry signal in cell populations expressing AIDF193A-mCherry was lower than in cell populations expressing AID-mCherry ( S6A Fig ) . By time-lapse imaging at 10 min intervals over the course of 24 hr , we recorded 568 individual HT1080 AIDF193A-mCherry transfectants ( 1384 hr total imaging time; S8 and S9 Movies ) . A strong nuclear mCherry signal was evident in most cells , and there were no transient increases in signal as observed in the pulses; instead , transient attenuation of the AIDF193A-mCherry signal was evident , occurring one or more times in 49 of the single cells imaged ( 8 . 6%; S1 Table ) . The reduced nuclear levels of AIDF193A-mCherry could reflect regulation by several different mechanisms , including enhanced cytoplasmic retention , impaired nuclear entry , nuclear exit by passive diffusion or a CRM1-independent active pathway , or enhanced nuclear proteolysis . A total of 105 signal attenuation events were documented . Timeline analysis of individual cells ( Fig 4A ) showed that most cells in which signal attenuation occurred also divided during the course of observation ( 78%; S1 Table ) . Attenuation events occurred at variable intervals after cell division , so division does not set the timer for signal attenuation . Individual cells exhibited 1–7 attenuation events ( Fig 4B ) , an average of 2 . 1 per cell ( 105 events/49 cells ) . The average duration of AIDF193A-mCherry signal attenuation events was 24 ± 8 min ( N = 42; S1 Table; S3C Fig ) . Individual examples of transient attenuation of the AIDF193A-mCherry nuclear signal are shown in representative images ( Fig 4C ) . The duration and frequency of AID-mCherry nuclear pulses and AIDF193A-mCherry nuclear attenuation events are similar , suggesting that they may be regulated by shared mechanisms . If so , then pulses and attenuation events are predicted to occur in synchrony . To test this , we imaged HT1080 AID-GFP AIDF193A-mCherry double transfectants , after first confirming by flow cytometry that the signals from these two fluorescent proteins could be readily distinguished , and that there was no change in mCherry signal in AIDF193A-mCherry + AID-GFP double transfectants relative to AIDF193A-mCherry transfectants; nor any effect on GFP signal in AIDF193A-mCherry AID-GFP double transfectants relative to AID-GFP transfectants ( S6B Fig ) . Time lapse imaging of live cells showed that AID-GFP nuclear pulses coincided with attenuation of the AIDF193A-mCherry nuclear signal , as evident in representative examples from three different cells ( Fig 5A; corresponding movies: S10 and S11 Movies ) . To quantify the changes in the AID-GFP and AIDF193A-mCherry nuclear signals , we determined the average mCherry and GFP signals in the nucleus and the cytoplasm ( pixel intensities/area ) in HT1080 AID-GFP and AIDF193A-mCherry double transfectants . The GFP and mCherry signals were used to define the cell perimeter and nuclear boundary , respectively . Analysis of a total of 15 HT1080 AID-GFP AIDF193A-mCherry double transfectants showed that of the 31 nuclear pulses/attenuation events that were observed , 94% were in synchrony . This synchrony is illustrated by representative tracings that quantify the nuclear signals of AID-GFP and AIDF193A-mCherry for two pulses in each of the three cells shown in Fig 5A . The tracings reveal that over the course of a pulse , the AID-GFP nuclear signal increased 25% or more , then dropped; while the AIDF193A-mCherry signal dropped by 20% or more , then recovered . Timing of AID-GFP pulses and AIDF193A-mCherry attenuation events appears to be coordinated . The maximum AID-GFP signal coincided with the minimum AIDF193A-mCherry ( Fig 5B ) . Similarly , the peak in the ratio of nuclear to cytoplasmic signal ( N/C ) for AID-GFP coincided with the minimum for AIDF193A-mCherry ( S7 Fig , above ) . Tracings of cytoplasmic signals revealed a typically modest ( ≈20% ) increase in the cytoplasmic signal of AID-GFP and AIDF193A-mCherry during each pulse or attenuation event ( Fig 5B; S7 Fig , below ) . This increase could reflect cytoplasmic accumulation of protein that has exited the nucleus or cytoplasmic retention of newly synthesized protein . The AIDF193A mutant is deficient in CRM1-dependent nuclear export , so if exit from the nucleus is responsible for AIDF193A-mCherry signal attenuation and cytoplasmic protein increase , exit would need to occur via a CRM1-independent transport pathway or by passive diffusion . Nonetheless , it is important to recognize that synchrony of AID pulses and AIDF193A extinction events does not necessarily imply a common mechanism . The evidence for temporal coordination of AID-GFP pulses and AIDF193A-mCherry attenuation events raised the question of whether AID pulses might also coordinate with nuclear import of P53 . P53 is a key regulator of apoptosis . P53 nuclear levels rise in response to DNA damage and other oncogenic stresses , such as stress caused by impaired nucleolar function , and elevated levels induce a transcriptional program that results in apoptosis and cell death . To ask if accumulation of P53 might be coordinated with AID pulses , we generated AID-mCherry P53-GFP double transfectants of the SV40-transformed normal human fibroblast line , GM639 . Live cell imaging of GM639 AID-mCherry P53-GFP double transfectants showed that nuclear P53-GFP levels increased in coordination with the initiation of the AID-mCherry pulse ( S12 and S13 Movies ) . We analyzed 63 AID-mCherry P53-GFP double transfectants and identified AID nuclear pulses in 8 of them ( 12 . 6% ) . All of the cells in which AID-mCherry pulses occurred also exhibited an increase in nuclear P53-GFP signal . The nuclear P53-GFP signal was constant in 74 . 5% of cells that did not pulse in the course of imaging ( 41 of 55 ) and increased in 25 . 5% ( 14 out of 55 ) . Fig 6A diagrams AID-mCherry and P53-GFP signals over the course of 24 hr imaging in one representative cell that did not pulse ( Cell 0 ) and three cells that exhibited one , three or five AID-mCherry pulses as quantified by Image J analysis ( Cells 1–3 , respectively ) . Nuclear P53-GFP levels increased 4- to 6-fold in cells that pulsed ( Fig 6B ) . Representative time-lapse images document temporal coordination between AID-mCherry pulses and P53-GFP accumulation ( Fig 6C ) . Nuclear accumulation of P53-GFP is therefore temporally coordinated with AID nuclear pulses .
By imaging single living cells , we have shown that AID appears in the nucleus in brief ( 30 min ) pulses in about 10% of cells . Pulses are a general rather than cell type-specific event , as they were observed in human HT1080 fibrosarcoma cells , GM639 transformed fibroblasts , and Ramos B cells . AID pulses do not depend upon AID catalytic activity , as the AIDH56A catalytic mutant pulsed with a frequency and duration comparable to wild type AID . Pulses occurred asynchronously among populations of proliferating cells , so pulses do not reflect a response to subtle changes in culture conditions , and they appear not to be cell cycle regulated . Nuclear pulses are well-documented regulatory process , with the potential to coordinate nuclear entry of factors with complementary functions and to limit nuclear access of potent factors that regulate sensitive processes [32–34] . Pulsing controls nuclear access and persistence of about a dozen different transcriptional regulators in yeast and five from human cells , including P53 [35] , NF-κB [36–38] , ERK [45] , Smad [46] , and NFAT [47] , all of which are key regulators that respond to and integrate inputs from multiple other factors and networks . AID fits naturally into this constellation of proteins , making it unlikely that AID pulses are an artefact of overexpression or tagged protein . The other proteins this far shown to pulse do so in response to specific stimuli or treatments: P53 pulses in response to DNA damage caused by ionizing radiation or drugs [35]; NF-κB in response to TNF [36–38]; ERK in response to EGF [45]; Smad in response to TGF-β [46]; and NFAT in response to changes in calcium concentration [47] . Some proteins that pulse do so in response to feedback regulation that maintains constant protein nuclear abundance [32 , 33 , 39 , 40] , but feedback regulation appears not to regulate AID pulses . Analysis of 1363 individual cells over 3875 hours of imaging identified 367 pulses or attenuation events in 137 cells expressing wild-type AID and its AIDH56A and AIDF193A mutant derivatives , defective in catalysis and CRM1-dependent nuclear export , respectively ( summarized in S1 Table ) . The average duration of events ranged from 24–34 minutes , and the frequency of cells in which events were observed ranged from 9–12% . The temporal coordination of nuclear pulses and attenuation events in double AID-GFP AIDF193A-mCherry transfectants suggests that there is a common underlying cause for both pulses and attenuation events , but does not necessarily mean that the mechanism is shared . Altered regulation of cytoplasmic retention , nuclear import , nuclear export and proteolysis could all contribute to characteristic cycle of pulses and attenuation events . The results reported here do not enable us to distinguish among these mechanisms . AID nuclear pulses are consistent with the physiological challenges of carrying out limited and highly regulated mutagenesis , although the results presented here do not identify a direct mechanistic connection between AID pulsing and function . The short duration of AID pulses will limit the interval during which DNA deamination can occur and thereby limit the potential for AID to cause DNA damage that exceeds its physiological mission . Intervals between pulses may provide time for DNA repair , or for assessment of whether sufficient deamination has occurred to support robust gene diversification . The relatively infrequent occurrence of AID pulses could also explain the paradoxical predominately cytoplasmic localization of AID evident in a variety of cells and tissues [14–18] . In addition , the episodic nature of AID pulses may influence repair pathways available to AID-induced damage . This should be taken into account in engineering applications that use AID as a mutagen , or when AID-induced damage is used as a model for repair at DNA nicks or strand breaks . P53 accumulation in the nucleus accompanied AID pulses . This suggests that P53 may respond to DNA damage caused during a pulse , but P53 also enters the nucleus in response to other signals , including metabolic and oxidative stress and altered ribosome biogenesis . P53 accumulated in the nucleus of 25 . 5% of cells independent of an AID pulse , raising the possibility that accumulation may have occurred as a result of some other stress signal , perhaps even as a side-effect of imaging . Better understanding of regulation of AID pulses will clarify the role of DNA damage in initiating AID pulses . AID is both a mutagen and a transcriptional regulator , as its deaminase activity can be deployed not only to initiate gene diversification , but also to erase epigenetic marks to alter gene expression patterns [9–13] . The other proteins that have been shown to pulse , in human cells and in yeast , all bind specific targets in duplex DNA to regulate transcription [32 , 33 , 39 , 40] . The mechanisms that target AID to specific genes are not clearly defined . AID deaminates single-stranded but not double-stranded DNA , making actively transcribed genes targets for AID [48–50] . Pulsing adds a temporal cue to combinatorial regulation of transcription [34] , and coordination of AID pulses and transcriptional activation could enhance targeting of some genes , such as Ig genes; or protect other genes from AID attack .
The pEGFP-N3 construct for expression of AID-GFP was a gift from Dr . Javier Di Noia ( Department of Microbiology and Immunology , University of Montreal , Montreal , Quebec , Canada ) . We substituted mCherry for a region of GFP flanked by ApaI and BsrGI restriction sites in the pEGFP-N3 construct to generate an AID-mCherry expression construct , pAID-mCh . pAID-mCh CSII and pAID-GFP CSII: We amplified AID-mCherry and AID-GFP from pAID-mCh and pAID-GFP , respectively , with primers PQL31 , 5’-ATATCAATTGAGATCCCAAATGGACAGCC-3’ and PQL32 , 5’-ATATTCTAGATTACTTGTACAGCTCGTCCATGC-3’ , and inserted the fragments between EcoRI and XbaI sites in p-mAG-GEM CSII , thereby replacing AG-GEM with AID-mCherry or AID-GFP . p AIDF193A-mCh CSII and p AIDH56A -mCh CSII: F193A and H56A mutants were generated using QuikChange II XL Site-Directed Mutagenesis Kit ( Agilent , Cat # 200521 ) with primer set , F193A FOR 5’-CTTACGAGACGCAGCTCGTACTTTGGGAC-3’ and F193A REV 5’-GTCCCAAAGTACGAGCTGCGTCTCGTAAG-3’; H56A FOR 5’- GAACGGCTGCGCCGTGGAATTGCTC -3’ and H56A REV 5’- GAGCAATTCCACGGCGCAGCCGTTC -3’ . The P53-GFP expression construct was purchased from Addgene ( plasmid #12091 ) . Lentiviral particles were produced using second-generation packaging plasmids in 293T cells . 293T cells were transfected with transfer vector , viral packaging vector ( psPAX2 ) , and viral envelope vector ( pMD2G ) at 4:2:1 ratio using Lipofectamine LTX ( Life Technologies , Cat # 15338100 ) transfection as directed by manufacturer’s protocol . Viral particles were collected at 24 hr and 48 hr after transfection and passed through 0 . 22 μm membrane ( Steriflip; EMD Millipore; Cat # SCGP00525 ) . Virus particles were used without further concentration . The human fibrosarcoma line HT1080 and SV40-transformed fibroblast line GM639 ( ATCC ) were cultured in DMEM media supplemented with 10% FBS , 2 mM L-glutamine , and penicillin/ streptomycin . HT1080 and GM639 cells were transfected by 4-D nucleofector ( Lonza , Cat# PBC2-00675 ) and Lipofectamine LTX reagent ( ThermoFisher Scientific , Cat# 15338100 ) , respectively , as directed by the manufacturer . The human Burkitt lymphoma cell line , Ramos , was cultured in RPM 1640 media supplemented with 10% FBS , 2 mM L-glutamine , penicillin/ streptomycin , 1X non-essential amino acids ( Gibco , Cat# 11140–050 ) , 1 mM sodium pyruvate ( Gibco , Cat# 11360–070 ) , and 10 mM HEPES ( Gibco 15630–080 ) . Lentiviral transductions used 2x105 Ramos B cells grown in medium containing 8 μg/ml of polybrene per ml . Cells were grown for 3–4 days after transduction and then sorted for mCherry+ to enrich for transduced cells expressing the desired fluorescent protein prior , typically constituting 0 . 1–10% of the population . Cells were treated with leptomycin B ( LMB; LC Laboratories , Cat# L6100 ) at 50 ng/ml or MG132 ( Z-Leu-Leu-Leu-aldehyde; Sigma-Aldrich ) at 50 μM for indicated time . High content screening microscopy and analysis of cell cycle were carried out as previously described [27] . Time-lapse microscopy was carried out using a DeltaVision microscope equipped with an environmental chamber to control for temperature ( 37°C ) and CO2 ( 5% CO2 ) . Transfected HT1080 and GM639 cells were grown in a 35 mm glass bottom dish ( MatTek , sample part # P35G-0-10-C ) at 37°C/5% CO2 at least 24 hr before imaging . HT1080 and GM639 cells were imaged at 10 min intervals for duration of 24 hr . Ramos B cells expressing AID-mCherry were grown at 37°C/5% CO2 at least 24 hr before imaging on LiveCell Array Microscope slides ( Nunc , Cat # 130505 ) , which limited but did not prevent tumbling . B cells were imaged at 15 min intervals for 3–4 hr; more frequent or more extended imaging compromised cell survival . Timelines illustrate pulses and attenuation events of less than 60 min duration relative to cell cycle . Longer events were occasionally observed in dying cells , and these were eliminated from the dataset . To indicate when events occurred with respect to cell division , individual timelines were ranked based on the interval between capture of the first image ( t = 0 ) and separation of the daughter cells . ImageJ software with “RGB measure” plugin was used to quantify nuclear and cytoplasmic signals of GFP and mCherry . HT1080 AID-mCherry transfectants were grown in 96-well μclear microplate ( Greiner Bio One ) for 24 hr prior to treatment with LMB , MG132 , or both at indicated times . Following treatment , cells were fixed in 3 . 7% formaldehyde and stained with whole cell stain ( HCS CellMask , Invitrogen ) and DAPI . Fixed cells were then washed with PBS and were imaged by Thermo Scientific ArrayScan VTI HCS reader . Cells with very low or very high mCherry signals were eliminated , gating based on the mock transduction control ( low ) and eliminating cells with signals more than 5 SD from the mean ( high ) . The HCS Colocalization BioApplication protocol was used to determine nuclear and whole cell boundaries in individual cells as defined by DAPI and HCS CellMask , respectively , thereby defining the cytoplasmic region as the region between nuclear and whole cell boundaries . The average signal in the nuclear and cytoplasmic compartments was determined in individual cells by measuring the total intensity of mCherry signal divided by area within each compartment . The ratio of nuclear to cytoplasmic signal ( N/C ) was calculated as the ratio of the average signals of nuclear and cytoplasmic mCherry . Statistical significance was determined using two-tailed , unpaired Student’s t-test , assuming unequal variances | Activation-induced deaminase ( AID ) is a mutagenic factor that plays a critical role in immunoglobulin gene diversification and also functions in early development to reprogram methylated regions of the genome . AID must be tightly regulated to prevent compromising cell fitness , but how this occurs is not thoroughly understood . One level of regulation is known to be nuclear entry and exit , but spatiotemporal regulation of AID had not been examined at the level of single cells . We have carried out time-lapse imaging of individual cells expressing fluorescent-tagged AID . Our movies show that AID enters the nucleus in brief pulses , of about 30 minutes duration . Pulses occur in only a fraction of cells in the course of one cell cycle and appear to respond to a metronome intrinsic to individual cells . AID pulses are coordinated with pulses of P53 , which regulates the cellular response to DNA damage . Pulsing may enable AID to synchronize both with factors that respond to AID-initiated damage and with factors that regulate transcription of AID target genes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"blood",
"cells",
"fluorescence",
"imaging",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"gene",
"regulation",
"cell",
"cycle",
"and",
"cell",
"division",
"metabolic",
"processes",
"cell",
"processes",
"immunology",
"biological",
"cultures",
"in",
"vivo",
"imaging",
"ht1080",
"cells",
"dna",
"damage",
"proteolysis",
"dna",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"transcriptional",
"control",
"imaging",
"techniques",
"animal",
"cells",
"proteins",
"gene",
"expression",
"cell",
"lines",
"antibody-producing",
"cells",
"biochemistry",
"cell",
"biology",
"b",
"cells",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"metabolism"
] | 2019 | Activation-induced deaminase (AID) localizes to the nucleus in brief pulses |
Bioluminescence techniques allow accurate monitoring of the circadian clock in single cells . We have analyzed bioluminescence data of Per gene expression in mouse SCN neurons and fibroblasts . From these data , we extracted parameters such as damping rate and noise intensity using two simple mathematical models , one describing a damped oscillator driven by noise , and one describing a self-sustained noisy oscillator . Both models describe the data well and enabled us to quantitatively characterize both wild-type cells and several mutants . It has been suggested that the circadian clock is self-sustained at the single cell level , but we conclude that present data are not sufficient to determine whether the circadian clock of single SCN neurons and fibroblasts is a damped or a self-sustained oscillator . We show how to settle this question , however , by testing the models' predictions of different phases and amplitudes in response to a periodic entrainment signal ( zeitgeber ) .
The circadian rhythm of organisms ranging from cyanobacteria to humans beats at the cellular level; it is a remarkable manifestation of celestial mechanics mirrored in molecular biology . The standard view is that the mammalian circadian clock is a hierarchically organized system , governed by the suprachiasmatic nuclei ( SCN , consisting of about 20 , 000 neurons ) in the hypothalamus . The SCN neurons are coupled to each other and are entrained by light to oscillate in synchrony to the 24 h earth rotation , and in turn entrain cells and organs in the rest of the body . In studies of dissociated SCN neurons , typically most of the cells are classified as being self-sustained oscillators [1]–[4] . However , as also predicted theoretically [5] , [6] , dissociated individual oscillating SCN neurons can vary greatly in their precision , and many have been suggested to be damped oscillators [1] , [7] , especially if synaptic input or mediated signaling is compromised [8] , [9] . At the SCN tissue level , mutant neurons that are arrhythmic when dissociated from each other can interact to generate a collective coordinated self-sustained rhythm [10] . Peripheral tissues contain independent clocks [11] , thought to be synchronized by the SCN via neural and hormonal pathways , as well as via more indirect routes such as body temperature and feeding behavior [12]–[14] . In the last decade it became clear that the circadian rhythm of immortalized fibroblast cell lines [15] as well as peripheral tissues such as liver , lung , and muscle [11] , has its origin at the single cell level . These rhythms are damped at the cell population level , but recent studies employing single-cell techniques suggested that the rhythms in peripheral tissues actually are self-sustained at the single-cell level [16]–[17] . It is not clear how the peripheral rhythms may differ from the SCN rhythm at the single-cell level . Rapid progress has been made during the last decade in unraveling the molecular components of the clock , although the picture is not yet complete . The consensus view is that the core clockwork consists of several interlocked negative and positive feedback loops [18] . There are numerous theoretical models for how these combine into a ticking molecular clock , all of which assume a self-sustained oscillator that essentially relies on the negative feedback loop consisting of the Period ( PER1 , PER2 ) and Cryptochrome ( CRY1 , CRY2 ) proteins inhibiting their own production once translocated into the nucleus , by attenuating the action of their transcriptional activators CLOCK and BMAL1 [19]–[22] . Corresponding knockout mutants have been studied at the cellular level [10] , although a quantitative characterization of these mutants is lacking . Using two simple canonical oscillator models and time-series analysis , we have extracted general parameters , such as oscillator damping rates and biochemical noise levels , from previously published single-cell bioluminescence measurements in SCN neurons and fibroblasts [9]–[10] . Considerable progress has recently been made in the understanding of the origin of noise in gene expression and protein concentrations [23] , and such noise was considered in our analysis since it is readily noticeable and quantifiable from the data at hand . We report here that the circadian clock in both wild-type single dissociated SCN neurons and in fibroblasts is well approximated as a damped ( non-self-sustained ) oscillator driven by biochemical noise . This model assumes that the underlying dynamics of the oscillator are those of a damped system , but that the damped oscillator via the filtering and molding of biochemical noise exhibits a circadian rhythm . An alternative minimal self-sustained ( limit-cycle ) oscillator model described SCN neurons as noisy small-amplitude oscillators with noise levels comparable to the oscillator amplitudes . Similar low-dimensional self-sustained models have previously been applied successfully to analysis of circadian single-cell fluorescence and bioluminescence time series [24]–[26] , as well as to measurements of circadian body temperature cycles [27] , [28] . We were also able to characterize , , and mutants , to each of which we could confer a unique quantitative signature . This paves the way for a characterization of mutants that goes beyond the mere classification as “rhythmic” or “arrhythmic” . Finally , based on a comparison with bioluminescence recordings of single neurons within cultured SCN tissue slices , we show how to use the fitted models to understand entrainment properties of single cells . Importantly , the entrainment properties differ markedly , depending on whether the noise-driven damped model or the self-sustained model is assumed , pointing to future experimental tests that would discriminate between the two concepts .
We used raw data from a recent investigation [10] , where bioluminescence imaging was employed to track the relative amount of functional mPer2-luciferase fusion protein in single dissociated mouse SCN neurons and fibroblasts , for wild-type ( WT , fibroblasts: , SCN neurons: ) and for ( fibroblasts: , SCN neurons: ) , ( fibroblasts: , SCN neurons: ) , and ( fibroblasts: , SCN neurons: ) mutants . Example WT time-series ( detrended and mean-centered ) are shown in Figures 1A ( fibroblast ) and 1D ( SCN neuron ) . From the bioluminescence time-series , we calculated empirical autocovariances and fitted these to the theoretical autocovariance formula of a linear damped oscillator with additive noise ( Methods ) . This model is comprised of two variables that oscillate , with a given period , towards a stable equilibrium point . However , the additive noise keeps the variables from settling at the equilibrium point , continuously offsetting them , which makes the oscillator appear to be self-sustained . In effect , the biochemical noise is filtered by the damped oscillator so that the circadian frequency components are amplified , while other frequency components are suppressed . This principle is illustrated in Figure 2 . The model is specified by three parameters only: the frequency , the damping rate , and the noise intensity . A representative example of the fitting of the autocovariances is given in Figures 1G and 1H , where the experimental data autocovariances ( black lines ) were calculated from the time-series in Figures 1A and 1D , respectively . In general , excellent fits were obtained with the linear damped oscillator model . With the fitted parameters , the model can be used to simulate time-courses ( see Methods and Text S1 for details about the model and simulations ) , as shown in Figures 1B ( fibroblast ) and 1E ( SCN neuron ) . The simulations illustrate that the noisy damped linear oscillator model indeed produces time courses qualitatively similar to the experimental ones ( Figures 1A and 1D ) . Note that the seemingly lower simulated noise level in Figure 1B compared to the experimental data in Figure 1A is due to the fitting procedure being able to selectively filter out and reject measurement noise ( see Text S1 ) . In order to test whether we could reject the damped oscillator model we adopted a bootstrapping approach related to the method proposed by Hall and Wilson [29]: we made 1000 simulations of the model ( Methods ) for each cell , using the parameters extracted from the data . For each simulated time-series , we calculated autocovariances and again fitted these to the analytical autocovariance function ( Methods ) as outlined above . Thus , for each cell , we have one autocovariance function calculated from experimental data , and 1000 autocovariance functions calculated from simulations . We then calculated the fraction of the 1000 simulations that produced better fits ( in the least-squares sense ) than the experimental data . We took this fraction as a measure of how reasonable the model is: if the simulations give better fits than experimental data in more than 95% of the cases , we consider the experimental data to be different enough from the simulated time series to reject the damped model . We could reject the damped model in this way for only 5 . 1% of the SCN neurons . For fibroblasts , the percentage was higher , 17% . The full results are given in Table S1 . To summarize and visualize the data fits for all cells in the Liu et al . [10] study , the fitted damping rates are plotted against the average noise-driven relative oscillation amplitudes in Figure 3A . Relative amplitudes are the oscillation amplitude divided by the overall mean . Key results of this parameterization are: ( 1 ) We can characterize WT , , , and mutants , i . e . separate them in parameter space . mutants were more weakly damped than WT cells and the other two null mutants . mutants have a more disrupted circadian rhythm than mutants , since their average relative oscillation amplitude is distinctly lower . This difference was not noted earlier [10] . ( 2 ) SCN neurons and fibroblasts cluster together , both for WT and mutant cells . Thus , given this model , we can find no significant difference between the circadian rhythms of these two cell types at the uncoupled single-cell level . Although fibroblasts generally give rise to dimmer bioluminescence due to lower PER2 expression level , cf . panels 1A and D , this peculiarity is not reflected in our analysis , since we measure amplitudes in relative units . ( 3 ) According to this model and data fit , WT cells constitute a more heterogeneous population than any of the mutants , in that the parameters of the model exhibit a larger spread . We could quantify this , cf . Table S2 . Ordered according to heterogeneity , were less heterogeneous , mutants even less heterogeneous , while the mutants were the most homogeneous . ( 4 ) The damping time of the cells , i . e . the time it would take a given perturbation to decay to half its magnitude , lies on the order of 5 to 100 hours with a median of 21 hours . As discussed below , this has implications for the synchronization and entrainment of the cells . To probe the generalizability of the method and the results , we further analyzed hitherto unpublished bioluminescence data from an earlier study by Yamaguchi et al . [9] , where transgenic mice with a luciferase reporter gene driven by the mPer1 promoter were used . The data we analyzed come from the smaller ventral part of two mouse SCN slices each cut in two . The cells in the smaller parts were not synchronized , and we hence infer that they may have been in a state similar to the dissociated SCN neurons from the Liu et al . [10] study . Again , we calculated the autocovariance for each cell ( ) , and fitted these to the theoretical autocovariance function for the linear damped oscillator ( Methods ) . The results are summarized in Figure 3B , where we have plotted the results of these fits together with those of the earlier fits to the WT cells of the Liu et al . [10] study . It is most reasonable to consider relative amplitudes when comparing data from different studies , since relative units are insensitive to differences in expression level of reporter genes , laboratory equipment , and other systematic factors . In general , although different reporter genes were used in the two studies , the cells occupy the same region in the parameter space of our model . We also performed the statistical test outlined above for these cells; in this case , we could reject the damped model for 1 . 9% of the cells . We sought to use the same methodology as above to evaluate the consensus picture of the circadian clock as a self-sustained oscillator at the single-cell level . This can be modeled most simply as a two-variable oscillator with a given amplitude , a given noise intensity , and a given amplitude relaxation rate to the stable oscillation ( limit-cycle , Methods ) . Also in this case , one obtains a theoretical autocovariance formula from the model , which was fitted to the autocovariances estimated from the experimental time-series , as in the representative example in Figure 1G and H , light gray dashed line . The autocovariance formula for the self-sustained oscillator model has two degrees of freedom more than that of the linear damped oscillator ( Methods ) , which is why we always expect a slightly better fit in the self-sustained case . Again , this model can be simulated ( see Text S1 ) , and typical simulated time-courses are shown in Figures 1C and F ( fibroblast and SCN neuron , respectively ) , which show that the self-sustained model also generates realistic time-courses . As for the damped model , we took the bootstrapping approach described above and made 1000 simulations for each cell and tested for how many cells we could reject the self-sustained model . This we could do for 4 . 2% of the SCN neurons and for 18% of the fibroblasts . The full results can be found in Table S1 . We summarize the results from these data fits in Figure 4 . On the abscissa are the coefficients of variation ( CV ) for the amplitudes , which are the standard deviations ( SDs ) of the fluctuations in the oscillator amplitudes , divided by the amplitudes themselves . On the ordinate are the oscillator amplitudes divided by the means of the time-series , as for the fits to the damped model . Of special note is that if the CV is greater than 1 , the stationary probability density ( see Text S1 ) of the two variables of the oscillator is qualitatively the same as for a damped oscillator–it is unimodal , i . e . has one single maximum . However , if the CV is less than 1 , the probability density forms a “crater ridge” around a local minimum . This “crater ridge” represents the self-sustained oscillation . The key results for the data fit to the self-sustained oscillator model are: ( 1 ) In this model , as in the damped model , the WT cells and mutants are also clearly separable . The mutant cells generally have clear oscillator characteristics ( CV less than 1 ) , while the majority of the WT ( 58% ) and other mutant cells have a CV greater than 1 , meaning that the amplitude fluctuation SDs are greater than the oscillator amplitudes . Thus the stationary probability densities are unimodal just like for damped oscillators . Again , mutants generally had a weaker rhythm than mutants . ( 2 ) As above , the fits to the self-sustained model cluster the fibroblasts together with the SCN neurons . ( 3 ) For the parametrization chosen in Figure 4 , WT cells are again the most heterogeneous . Ordered according to heterogeneity , mutants were less heterogeneous , neurons were even less heterogeneous , while mutants were the most homogeneous ( see Table S2 ) . ( 4 ) The damping time for relaxation to the oscillation cycle is on the order of 1 to 10 hours with a median of 3 . 2 hours . We repeated the same fitting procedure for the neurons of the Yamaguchi et al . [9] study . The results , overlayed by the results of the fit to the WT cells of Liu et al . [10] , are shown in Figure 4B . Just as for the damped model , the cells from the two different studies occupy the same region in parameter space , although we see here that the neurons of the Yamaguchi et al . [9] study generally are less noisy; most of them have a CV less than one . Also , there is a trend towards lower relative amplitudes , which at least partially could be due to a higher background glow from non-synchronized neighboring neurons of the intact slice . Statistically , we could reject the self-sustained model ( using the method outlined above ) for 26% of these neurons .
We have made an in-depth analysis of bioluminescence time-series data of the circadian rhythm in mouse SCN neurons and fibroblasts . The main conclusions are , first , that it is possible to estimate fundamental parameters of the oscillators , such as damping rate and noise intensity . Second , we show that the question of whether the circadian clock is self-sustained or damped is not settled , since we could reject neither hypothesis . Third , we predict frequency response curves , experimentally obtainable via e . g . temperature entrainment , that will look quite different depending on whether the rhythm is self-sustained or damped . The parameters extracted in this study ( damping rate , noise level , etc . ) typically span an order of magnitude for the cell populations we study . We find it remarkable that fibroblasts and SCN neurons cluster similarly in parameter space even for the different mutants studied , which hint at a universality of the clockwork behavior . Even SCN neurons from an earlier , independent investigation [9] where a different reporter gene was used , clustered together with the WT neurons of the Liu et al . [10] study . This suggests that the parameters reflect general properties of the circadian oscillator , and that the extraction procedure we employed is robust . Another interesting property of the distributions of the estimated parameters ( except for the period ) is that they appear far less skewed ( or even Gaussian ) when viewed in logarithmic coordinates rather than linear coordinates . This is a more or less universal property of diverse biological data , such as species abundance , gene expression , and mRNA and protein copy numbers [32]–[35] . Biophysically detailed models are helpful for the conceptual understanding of how molecular properties influence the dynamics of larger integrated systems , the circadian clock being no exception [20] , [21] , [36] . Our modeling approach here , however , is complementary in the sense that it is top-down: it takes as starting point the dynamical features observed in single cell time-series , rather than precise knowledge of the molecular species of the system . This allows unambiguous fitting of the few parameters to data , whereas it has been shown that already moderately more complex circadian models do not allow this [37] . It is noteworthy that although the circadian clock is a very complex system , the time-series analyzed here exhibit only a few degrees of freedom and are well described by very sparse models with five or fewer parameters . On the other hand , the parameters we estimate are observational . This means that they describe the characteristics that can be observed from time-series of merely one component ( PER protein ) , but in reality are compound parameters of a system that is much larger . A challenging problem to be solved is how the few parameters quantified here relate to kinetic and thermodynamic quantities that biophysically detailed models typically are based upon . Such quantities include transcription and translation rate constants , mRNA and protein degradation rates , protein phosphorylation rates , and equilibrium constants . It is necessary to find such relations in order to systematically observe or manipulate molecules , which will allow us to understand or alter systems level properties like circadian phase and jet-lag response , rather than relying on serendipitous discovery . On a general level , Indic et al . [38] have already shown that it indeed is possible and reasonable to reduce two different detailed biophysical circadian clock models [20] , [21] to two-dimensional limit cycle models of the type considered in this study . Therefore , the results obtained in the present study should serve as constraints for biophysically detailed models . Previously , a few experimental and theoretical studies have mapped transcription rates and PER degradation rates to oscillation frequency [39]–[42] . In order for a negative feedback oscillator to be self-sustained , either strong nonlinearities or a great number of intermediate reaction steps are required . The celebrated Goodwin model [43] , for instance , is never a self-sustained oscillator if the Hill coefficient is smaller than 8 , independent of other parameters of the system [44] . An increased number of intermediate reaction steps lowers this Hill coefficient threshold [45] . More recently , Morelli and Jülicher [46] related the fidelity of a noisy negative feedback oscillator to the number of its constituent elementary reaction steps . In effect , the presence of noise relaxes such conditions , making the system oscillate for a broader range of parameters , i . e . also when it is damped in the absence of noise , thus enhancing robustness in this respect . Turning to our data , it is quite possible that what we observe is actually a mixture of self-sustained and damped cellular oscillators that take advantage of this principle . A consequence of this design is that noisy oscillators can have large amplitude fluctuations and peak-to-peak time variations ( phase diffusion ) . However , it has been shown that coupling a large population of less precise SCN neurons results in a markedly more precise synchronized rhythm [7] . Virtually all theoretical descriptions of the single-cell circadian clock so far have postulated that it is a self-sustained rhythm . However , the present study suggests there is no particular reason to prefer a limit-cycle model to a damped model . Rather , neither the damped nor the self-sustained model could be rejected . Furthermore , in more than half of the WT cells , the fitted self-sustained model described an oscillation with such small amplitude in relation to the magnitude of fluctuations ( in Figure 4 ) , that the distinction between damped and self-sustained oscillators becomes blurred . The original estimate of 66% rhythmic WT cells by Liu et al . [10] would , using our method , imply classifying cells with as arrhythmic , although it may be more reasonable simply to characterize cells using a continuous parameter like CV than to impose a binary classification ( rhythmic or arrhythmic ) . In any case , a distinct difference between the predictions of the scenarios remains: the different time-scales of the amplitude relaxation rates . This is the prediction that we propose to test by measuring the frequency responses to different entrainment frequencies . In light of the fact that the observed phase spreads of intact SCN slices seems to lie in between the ones predicted by the damped model and the self-sustained model respectively , a conservative conclusion of the present study would be that the SCN is composed of a mixture of damped and self-sustained oscillators , as suggested by Aton et al . [1] in a study of the electrical activity of SCN neurons . That such a heterogeneous mixture of oscillators indeed can synchronize and entrain to an external circadian forcing has been demonstrated in a modeling study [47] . Biochemical oscillators that are damped in the deterministic sense , but driven by biochemical noise to appear self-sustained , have for a long time been hypothesized to exist [48] , [49] . It is only very recently that experimental support for such ideas has begun to emerge [50] . Here , we show that time series data on the cellular circadian clock are consistent with this principle . Both the damped and the self-sustained model predict frequency response curves for the oscillation amplitude when entrained to a periodic forcing , a zeitgeber . Experiments should be able to settle the question as to whether these frequency response curves are actually exhibited by the circadian clock , and give further clues favoring either the damped or the self-sustained model . Such experiments will have to combine high-resolution imaging with entrainment , which can be achieved by temperature cycles [12] or light pulses [51] . We would then have a solid explanation for the phase spread of single circadian oscillators in the intact SCN [9] , [10] . Together with recent advances in the theoretical understanding of the synchronization of noisy oscillators [52] , one has a basis for the theoretical understanding of the SCN and the entrainment of the circadian oscillators at the organism level .
A generic linear damped oscillator with additive white noise can be described by the following linear system of stochastic ( or Langevin ) differential equations: ( 3 ) where , and . Here , the noise terms are white noise sources; denotes time average is Dirac's delta function , and the parameter is the noise intensity . Without noise terms , this system is a damped oscillator with damping rate and angular frequency . With noise terms , the system is continuously perturbed and exhibits the behavior seen in Figure 1B and E . Formally , this system could be interpreted as a negative feedback loop , where positively influences , while the latter negatively influences . The variables and clearly must not be interpreted as absolute concentrations of chemical species , but can be interpreted as differences or distances to reference steady state concentrations . This Langevin approach belongs to the standard methods for models of stochastic gene expression , see e . g . [53] . We cannot , given our data , separate intrinsic and extrinsic noise , and hence put no specific constraints on the noise intensity , which we instead estimate from the experimental data . Our approach to fit Equations 3 to the data is to fit the autocovariance function , i . e . the expected values , of the model to autocovariances estimated from the data . From Equations 3 one obtains ( see Text S1 ) : ( 4 ) This lets us extract the damping rate , the frequency , and the noise intensity from the fit . A generic self-sustained oscillator with linear relaxation to a limit cycle can be described in polar coordinates , i . e . radius and angle , by the following system of stochastic differential equations: ( 5 ) where , , and . This model has two noise intensities , one for perturbations perpendicular to the limit cycle ( ) , which decay with a rate , and one for perturbations along the limit cycle ( ) . The self-sustained oscillation has an amplitude , and cycles with frequency . Any perturbation away from this cycle of amplitude will relax back to it with a damping rate . This model is a Taylor expansion to the first order around a symmetric limit cycle , and thus the parameter is equivalent to a Floquet exponent . The symmetry of the model is reflected by radial isochrones [54] and constant angular speed . This model class has been extensively studied by , among others , Winfree [54] , also in the context of circadian rhythms . It is important to note , that without noise and with , Equations 5 are just Equations 3 in radial coordinates . Again , we calculate the autocovariance function in order to perform data fits . Transforming to Cartesian coordinates , the following autocovariance function is obtained ( see Text S1 ) : ( 6 ) This is a cosine function multiplied by the sum of two exponential functions , which implies that in addition to a slowly relaxing term with exponent , we also have a faster relaxing term with exponent . Our approach to a distinction between self-sustained and damped oscillators rests upon the difference between one time scale of autocovariance decay , as in Equation 4 , or two time scales , as in Equation 6 . Crucial for this distinction is the absence of additional time scales in a damped scenario , which requires a symmetric dynamical system , as defined above . Evidence for this is the close-to-sinusoidal shape of oscillations , as validated by the good fits of the models' autocovariances to the data autocovariances . Further evidence is the fairly symmetric type 1 phase response curves [54] , [55] of moderate magnitudes that are typically found in SCN neurons [56]–[60] Thus , the dynamical system is relatively symmetric , and there is probably no fixed point in the close vicinity of a limit cycle . Such a fixed point would be indicative of a highly asymmetric dynamical system ( cf . e . g . [49] ) , which potentially could invalidate our models . Also in mammalian fibroblasts , symmetric type 1 phase response curves have been measured [51] , [61] , [62] , although also a type 0 phase response curve was observed by Nagoshi et al . [16] , perhaps reflecting the smaller absolute amplitudes of the fibroblast circadian oscillator [61] . However , this type 0 phase response curve also appeared symmetric . When looking at entrainment properties of the cells , we consider the damped linear oscillator driven by a periodic forcing term , i . e . ( 7 ) For simplicity , we here neglect noise , and we consider only the driving force acting directly on one of the variables ( ) . The analysis would be similar if we considered driving of both variables . A convenient framework for deriving properties of this system is provided by control theory ( see Text S1 for some further details ) . One can thus show that the system will oscillate with the frequency of the driving force , and that the y-variable of the entrained oscillator is described by the equationwhere is the phase of entrainment , and is the entrained amplitude of the variable . The amplitude depends on the forcing frequency according to: ( 8 ) For , has a maximum at the resonance frequency . The phase of entrainment is given by: ( 9 ) where is the four-quadrant inverse tangent function , so that will lie in the interval . This means that the y-coordinate of the entrained oscillator lags behind the entrainment signal . When studying the forced self-sustained oscillator , we apply the forcing to the Cartesian x-direction , in order to be able to compare to Equation 7 . Thus , we consider the system ( 10 ) when studying the entrainment of the self-sustained model . | Earth's 24-h-rotation around its axis is mirrored in the circadian clock that resides within each of our cells , controlling expression of ∼10% of all genes . The circadian clock is constructed as a negative feedback loop , in which clock proteins inhibit their own synthesis . During the last decade , a picture has emerged in which each cell is a self-sustained circadian oscillator that runs even without synchronizing cues . Here , we investigated state-of-the-art single-cell bioluminescence recordings of clock gene expression . It turns out that these time series are very well described by low-dimensional models , enabling us to extract descriptive parameters that characterize each cell . We find that different cell types do not differ much in their dynamics . However , different mutations in core clock genes yield different dynamic characteristics . Furthermore , we could not statistically reject the idea that the cells are in fact damped oscillators driven by noise . We thus declare the question of whether the circadian clock is a damped or self-sustained oscillator still unresolved . Further , we propose a way to resolve this question by examining the frequency-dependent response of single cells to periodic stimuli . We will then be in a better position to understand how cells coordinate and synchronize their circadian rhythms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"expression",
"biophysics/theory",
"and",
"simulation",
"computational",
"biology/systems",
"biology",
"cell",
"biology/cell",
"signaling"
] | 2009 | Quantification of Circadian Rhythms in Single Cells |
Human fasciolosis is a re-emerging disease worldwide and is caused by species of the genus Fasciola ( F . hepatica and F . gigantica ) . Human fasciolosis can be diagnosed by classical coprological techniques , such as the Kato-Katz test , to reveal parasite eggs in faeces . However , although 100% specific , these methods are generally not adequate for detection of acute infections , ectopic infections , or infections with low number of parasites . In such cases immunological methods may be a good alternative and are recommended for use in major hospitals where trained personnel are available , although they are not usually implemented for individual testing . We have developed a new lateral flow test ( SeroFluke ) for the serodiagnosis of human fasciolosis . The new test was constructed with a recombinant cathepsin L1 from F . hepatica , and uses protein A and mAb MM3 as detector reagents in the test and control lines , respectively . In comparison with an ELISA test ( MM3-SERO ) the SeroFluke test showed maximal specificity and sensitivity and can be used with serum or whole blood samples . The new test can be used in major hospitals in hypoendemic countries as well as in endemic/hyperendemic regions where point-of-care testing is required .
Fascioliosis ( = fascioliasis ) is a plant-borne zoonosis caused by infection with trematode species of the genus Fasciola . Human fasciolosis is a re-emerging disease present worldwide , and can be produced by Fasciola hepatica and Fasciola gigantica . In Europe , the Americas and Oceania , only F . hepatica is present , but in several areas of Africa and Asia the geographical distribution of both species may overlap [1] and even hybridize in some cases [2] , [3] . Overall , it is considered that between 2 . 4 [4] and 17 million [5] people from 55 countries are infected by Fasciola species and that 180 million people are at risk of infection [1] . Fasciolosis is an important health problem in some South American countries ( Bolivia , Peru , Chile and Ecuador ) , in the Caribbean ( Cuba ) , northern Egypt , and in the Caspian region ( Iran ) , although human infections by Fasciola are not infrequent in European countries such as Portugal , Spain , France and the United Kingdom [6] . Furthermore , it is expected that this trematode-induced disease will increase over time as a result of climate change and the advent of milder , wetter weather [7] , [8] . Human fasciolosis can be diagnosed by classical coprological techniques , such as the Kato-Katz test [9] , to reveal parasite eggs in faeces . However , although 100% specific , these methods have some limitations including: i ) non usefulness during the prepatent period ( first 3–4 months after infection [10] ) ; ii ) poor sensitivity in patients with a low degree of parasitization , or in patients with intermittent egg shedding; iii ) non usefulness in ectopic infections or when parasites do not reach maturity . To avoid the above problems , some immunological techniques based on the determination of circulating secretory antigens [11] , [12] , testing of coproantigens [11] , [13] , or determination of serum anti-Fasciola antibodies [14]–[18] have been reported in the last two decades . However , for detection of early stages of Fasciola infections in humans , or ectopic infections , antibody determinations are preferable to coprological tests as circulating antibodies are produced early on and remain detectable for long periods . Several antigenic fractions of Fasciola [19] , [20] , purified antigens [21] , [22] and recombinant antigens [15] , [23] , [24] have been successfully used for the serodiagnosis of fasciolosis in human and animal species . Nevertheless , cathepsins L are the most frequently used target antigens for detecting anti-Fasciola antibodies [15] , [21] , [25] , [26] , as circulating antibodies to these molecules remain at high levels for long periods [27] . An ELISA test ( MM3-SERO ) that had proven useful for serodiagnosis of fasciolosis in several animal species with maximal sensitivity and specificity [27]–[30] , was recently found to involve binding to cathepsins L1 and L2 from both species of Fasciola [31] . Despite the usefulness of serological ELISAs for the diagnosis of fasciolosis , human infections frequently occur in non-developed countries where access to diagnostic laboratories is not always possible . In this sense , the development of rapid lateral flow immunoassays ( LFIAs ) , also known as immunochromatographic tests , would be of great interest . In this study we report the development and evaluation of a new LFIA for serodiagnosis of human fasciolosis , which can be used for point-of-care ( POC ) testing .
The study protocol was approved by the Ethics Committee of the Universidad de Santiago de Compostela , Spain . The serum samples used in this study were obtained as part of public health diagnostic activities , were already collected before the start of the study , and were tested as anonymous samples . Control ( negative ) blood samples were obtained from volunteers after they had provided written informed consent . The serum samples ( n = 203 ) used in this study were obtained from serum collections stored in the Centro Nacional de Microbiología ( ISCIII , Madrid , Spain ) , Laboratorio de Parasitología ( Facultad de Farmacia , USC , Spain ) , Centro de Investigación de Enfermedades Tropicales de la Universidad de Salamanca ( CIETUS , Salamanca , Spain ) and the Centre for Parasite Immunology and Biology ( CSPGF-INSA , Porto , Portugal ) . We analyzed serum samples from 39 patients with fasciolosis , 27 patients with schistosomosis , 20 patients with filariosis , 9 patients with hydatidosis , 15 patients with anisakiosis , 22 patients with toxocariosis , 12 patients with Chagas' disease , and sera from 59 patients with non-infectious pathology ( mainly from surgical interventions for non-painful processes such as cataracts and hernias ) . The above infections were diagnosed in the country of origin of the patients by one or more of the following methods: i ) routine coprological techniques for detection of ova and parasites ( Kato-Katz and Ritchie tests , using 3 stool samples ) , ( ii ) examination of urine for Schistosoma haematobium eggs , after sedimentation , ( iii ) Knott's test for detection of microfilaremia in blood , ( iv ) the ICT Filariasis test ( Binax , Portland and Maine ) for detection of Wuchereria bancrofti , ( v ) an “in house” ELISA test with soluble antigens present in sheep hydatidic cysts as target [32] , [33] , ( vi ) detection of IgE antibodies to the allergens Ani s 1 and Ani s 7 from Anisakis [34] , ( vii ) Enzymun-test IgE ( Boehringer-Mannheim Lab . , Mannheim , Germany ) for detection of anti-Toxocara antibodies in serum , and ( viii ) the Chagatest-ELISA recombinante ( Wiener Laboratorios S . A . I . C . , Rosario , Argentina ) for detection of serum antibodies in Chagas' disease . Fasciola seropositive patients from Portugal were tested by using Fasciola excretory-secretory antigens ( ESAs ) as target in ELISA and/or by Western-blotting analysis [35] , while seropositive patients from Salamanca were tested by an ELISA test with Fasciola ESAs , according to Hillyer et al . [36] , and patients from Madrid were diagnosed by use of a haemagglutination assay ( Fumouze Diagnostics , Paris , France ) . All Fasciola positive patients from Santiago de Compostela , and the above positive sera were confirmed for the presence of anti-Fasciola antibodies by use of the MM3-SERO ELISA ( see below ) . Negative whole blood samples were obtained from 12 volunteers ( 8 women and 4 men , aged 20–55 years ) by fingertip puncture with a lancet ( Glucoject , Menarini Diagnostics , Firenzi , Italy ) , at the Faculty of Pharmacy , University of Santiago de Compostela , Spain . An L1-cathepsin from F . hepatica ( gb|FR848428 ) was cloned as previously reported [31] . Briefly , the encoding gene without the putative N-terminal fragment ( signal peptide ) was cloned into the pQE expression vector ( QIAGEN , QIAGEN Iberia S . L . , Madrid ) with the primers 5′-pCL1 TCGAATGATGATTTGTGGCATCAGTGGAAGCG ( forward ) and 3′-pCL1 CGGAAATCGTGCCACCATCGG ( reverse ) , and further transformed into the M15 [pREP4] strain of E . coli ( QIAGEN , [37] ) . Fasciola hepatica recombinant rpCL1 expression was induced by addition of 1 mM IPTG . After induction , cells were harvested by centrifugation and the insoluble recombinant proteins were purified with B-PER reagent ( Thermo Fisher ) , solubilized and purified by affinity-chromatography with HIS-Select Nickel Affinity Gel ( Sigma-Aldrich ) under denaturing conditions , as indicated by the supplier ( 8 M urea ) . The protein solution containing the rpCL1 was then refolded by dispersing the eluate at a ratio of 1∶ 50 in PBS containing cysteine ( 10 mM ) and cystine ( 1 mM ) , followed by membrane-filtration concentration in an Amicon Stirred Ultrafiltration Cell equipped with a Filtron Omega Series membrane ( 10 K nominal molecular weight limit; Pall Filtron Corporation ) . Finally , the protein concentration was measured with the Micro BCA Protein Assay Kit ( Pierce , Rockford , IL ) , adjusted to 2 mg/ml in PBS , and the sample was stored at −80°C until use . Human serum samples were analyzed by MM3-SERO ELISA , a capture immunoassay that detects antibodies against the highly specific antigens recognized by mAb MM3 [29] . Polystyrene microtitre F16 plates ( Greiner Bio-One , Sigma-Aldrich , Madrid , Spain ) were coated for 2 h at 37°C with mAb MM3 ( 100 µl/well of a solution containing 5 µg/ml protein in PBS ) , and the uncoated sites were blocked with a 1 . 5% solution of buffered sodium caseinate for 1 h at room temperature ( RT ) . After washing once with PBS containing 0 . 2% Tween-20 ( PBS-T ) , 100 µl of either F . hepatica ESAs ( 0 . 4 µg/ml total protein ) in PBS-T , or 100 µl of PBS only , were added to odd ( Ag+ ) and even ( Ag− ) plate rows , respectively , and then incubated for 1 h at RT . After washing 4 times with PBS-T , 100 µl of each serum sample , diluted 1/100 in PBS-T containing 1% of skimmed milk , were added to each Ag+ and Ag− ELISA well , in duplicate . Plates were then incubated for 2 h at 37°C , washed 5 times with PBS-T , and incubated at 37°C for 1 h with 100 µl/well of anti-human IgG1 monoclonal antibody ( Sigma-Aldrich ) labelled with FITC and diluted 1/3000 . The reaction was revealed with peroxidase-conjugated rabbit anti-FITC IgG ( Serotec; dilution 1/5000 ) . Plates were washed again 5 times with PBS-T , and 100 µl of substrate ( SigmaFast OPD , Sigma-Aldrich ) were added to each well . The plates were then incubated for 20 minutes , and the reaction was stopped with 25 µl of 3N H2SO4 . The optical density ( OD ) was measured at 492 nm in an ELISA reader . The OD value for each sample was calculated as OD1 minus OD2 , where OD1 is the mean for the 2 Ag+ wells , and OD2 is the mean for the 2 Ag− wells . A cut-off value of OD = 0 . 05 for positive responses was calculated as the OD mean plus four times the standard deviation obtained for 200 human serum samples ( age range: 20–76 years ) from patients with eosinophilia of unknown origin but negative for Fasciola , by use of the Fumouze's haemagglutination test . However , to avoid false positive reactions due to batch variations in secondary ELISA reagents , a doubtful range ( OD = 0 . 05–0 . 099 ) was arbitrarily established for this test , while OD values equal or higher than OD = 0 . 1 were considered positive .
The HF135 ( Millipore ) NC membranes , which allow flow rates of 135±34 sec/4 cms , a pad conjugate containing rpCL1 bound to colloidal gold at pH 8 . 6 , together with protein A in the test line , and mAb MM3 in the control line , were used to construct a lateral diffusion test for serodiagnosis of human fasciolosis . Typical results for a positive and a negative serum are shown in Fig . 3 . As can be observed , both positive and negative lines were strongly coloured , and therefore suitable for screening with the naked eye . The data in Fig . 3 also show that negative samples produced a strong signal in the control line with no background in the test zone . To investigate the detection limit of the assay [38] , we compared the signals obtained with two-fold serial dilutions of three positive pooled sera in the LFIA , with those obtained in the MM3-ELISA used as gold standard . The data in Fig . 2 show that the new LFIA is highly sensitive , as it was able to recognize as positive the pooled sera at a higher dilution ( dilution 9; 1/25600 ) than the MM3-SERO test ( dilution 7; 1/6400 ) . With respect to the stability of the SeroFluke test , we observed that the detection limit did not change after storage for three months at 37°C , or 6 months at room temperature , when the strips were maintained in plastic bottles with a desiccant ( data not shown ) . In order to evaluate the sensitivity and specificity of the new SeroFluke test , we compared the results obtained with this test and the MM3-SERO ELISA test , using sera from patients positive for fasciolosis and sera from patients with the other parasitic infections described above . The data in Table 1 show an almost perfect concordance of results obtained by both methods , for positive ( n = 39 ) and negative samples ( n = 164 ) . In fact , the only partial discrepancy was one serum sample containing a very small amount of anti-Fasciola antibodies , which tested doubtful ( OD = 0 . 08 ) in the MM3-SERO ELISA and positive in the SeroFluke test ( Table 2 ) . The mean value for the MM3-SERO ELISA in the group of Fasciola positive sera was OD = 1 . 61 ( OD range: 0 . 08–2 . 71 ) , all of which produced a clear signal in the test line of the SeroFluke device , according to the AUs scale indicated in the Material and Methods section . The data in Table 2 also indicate that although some positive sera produced higher signals with SeroFluke than with MM3-SERO , more than 82% of sera that produced strong signals in the SeroFluke test ( ranks 3–5 AU ) also produced good signals ( >OD = 0 . 5 ) in ELISA . Once we had evaluated the usefulness of the SeroFluke test for diagnosing human fasciolosis in sera , we checked the validity of the test for diagnosing human Fasciola infection by using whole serum samples . Because of the non availability of whole blood from positive patients , we mixed blood samples from 12 healthy volunteers with pooled positive or negative sera . As indicated above , the test was run for 10 min , and the device was then washed in fresh buffer to remove the haemoglobin , thus facilitating the visualization of positive lines . The practical use of this test with whole blood samples and the results obtained after the washing step for two positive and one negative serum samples are shown in Fig . 4 .
The SeroFluke LFIA described in this study is the first one-step LFIA developed for serodiagnosis of human fasciolosis and , to our knowledge , it is also the first LFIA device developed for serodiagnosis of a trematode-induced disease . As reported above , human fasciolosis can also be correctly diagnosed with several ELISA tests [17] , [18] , [23] , or the MM3-SERO described here , but in general these immunoassays must be carried out in specialized centres , by trained personnel . However , the SeroFluke test can be used by personnel with minimal training , and more importantly , it can be used for POC testing , which is of great importance for medical attention of populations in several countries where this trematodosis is endemic [9] . In such countries , the possibility of using whole blood samples for the SeroFluke test ( obtained by e . g . taking a drop of blood by fingertip puncture with a lancet , as in this study , or even collected on filter paper [39] ) is also advantageous , as this is cheaper and saves the time and expense involved in obtaining serum samples . However , even considering only major laboratories in hypoendemic countries , the new LFIA may offer advantages over ELISA methods , as it allows individual testing of patients , whereas ELISA methods are designed to test several patients at once and some robustness may be lost when they are used only sporadically [40] . Furthermore , the LFIAs can be produced in small quantities ( e . g . in the range of 10–25 tests ) , which would reduce the costs associated with the shelf-life of the kits . In this respect it should be considered that the use of serological methods to detect anti-Fasciola antibodies in non-endemic countries is frequently limited to patients with eosinophilia of unknown origin , or to confirm infection suspected after radiological analysis . In the present study , only one Fasciola serum sample tested doubtful by MM3 SERO . The sample was obtained from a patient who had tested positive by MM3 SERO two years before and was subsequently treated with triclabendazole . Assuming such serum to be truly positive , the SeroFluke test displayed 100% sensitivity and 100% specificity . Moreover , as demonstrated by testing serial serum dilutions ( Fig . 2 ) , the SeroFluke test showed a lower detection limit , which enables clear positive signals to be obtained with sera from patients with very low circulating anti-Fasciola antibodies . The high sensitivity of the SeroFluke test was at first glance surprising , since the sensitivity of LFIAs designed to be read by the naked eye is often lower than the equivalent laboratory-based ELISA tests [41] . Moreover , this may be even more surprising if we take into account that the antigen used in the MM3-SERO ELISA includes several L-type cathepsins ( L1 and L2 ) from Fasciola [29] , [31] , whereas the SeroFluke test was constructed with a single-labelled artificially-refolded rpCL1 from F . hepatica , lacking some of the conformational epitopes that are present in mature cathepsins from this trematode [31] . Nevertheless , this may be explained by considering that positive sera from infected humans predominantly recognized the MM3 epitope present in the rpCL1 , as revealed by ELISA inhibition studies in our laboratory ( unpublished results ) . Thus , while in the MM3-SERO ELISA , the MM3-recognized epitope on the cathepsins is blocked by mAb MM3 during the process of antigen capture , in the SeroFluke test this epitope is available to react with human antibodies . The special preference for the MM3-epitope of the antibodies present in serum from infected humans may also explain the maximal specificity obtained in the present study , in which all patients infected with other related or unrelated parasites tested negative ( Table 1 ) . It is also possible that , in addition to the immunodominant MM3 epitope , other epitopes not present in mature cathepsin L from Fasciola contribute positively to recognition of the antigen by human sera . This may be the case of the B-cell epitopes , which were described in the prosegment region of the Fasciola cathepsin L [42] . However , the relevance of such epitopes as targets of anti-Fasciola serum antibodies in humans has not yet been investigated . With respect to the specificity of the LFIA , it should also be noted that sera from patients infected with other liver food-borne trematodes ( e . g . Clonorchis sinensis and Opisthorchis viverrini ) were not available for study , which prevented us from establishing whether these pathogens may provoke false positive results . Nevertheless , as indicated above , since most of the anti-Fasciola antibodies are directed to the specific MM3-recognized epitope , cross-reactions with these species are not expected . This supposition is also consistent with a previous study reporting no recognition of a cathepsin L from Clonorchis sinensis by sera from patients infected with Fasciola hepatica [43] . Another aspect of the SeroFluke test to be considered when interpreting the results is that for some patients with large amounts of anti-Fasciola antibodies , a decrease in the intensity of the control line may be observed . This is due to competition between anti-Fasciola antibodies and the mAb MM3 immobilized in the control line for the MM3 epitopes present on the rpCL1 . Therefore , a decrease in the signal in the control line often indicates that the sera tested have a high concentration of anti-Fasciola antibodies , rather than a failure of the device ( for example due to incomplete liquid migration ) . Since the SeroFluke test was designed for use without a plastic housing , interpretation of the results is straightforward , as liquid migration can be observed directly . This is even more evident when testing whole blood samples because the migration of haemoglobin clearly stains the absorbent pad red ( see Fig . 4 ) . Nevertheless , if interpretation were doubtful , the test could be repeated with a more diluted sample ( e . g . 1/500–1/1000 ) which would lead to an increase in signal intensity in the control line . The presence of mAb MM3 in the control line , rather than any other anti-rpCL1 antibody , may be advantageous for its use by trained personnel in hospitals , as this design enables qualitative estimation of the concentration of anti-Fasciola antibodies in serum by comparing the intensity of colour in the positive and control lines . In previous studies with the MM3-SERO test in sheep [27] , we have observed that such a test can be used to detect infections by either F . hepatica or F . gigantica . In this study we did not have the opportunity to test sera from patients infected with the latter species . However , although experimental confirmation is required , on the basis of our previous experience with the MM3-SERO ELISA , it appears likely that the new SeroFluke test will also be able to be used to detect human infections by F . gigantica . In summary , we present a new highly stable , sensitive , and specific LFIA for serodiagnosis of human fasciolosis . It is expected that such a test will be useful for hospitals in hypoendemic regions as well as in hyperendemic regions where POC testing is required . Although serological tests cannot differentiate between current and past infections , these methods are very useful for detecting early infections , which may be then confirmed with data from coprological tests , when the patent period of infection has been reached , or with other complementary data . The suitability of the SeroFluke test for detection of antibodies in animal species is also being investigated . | Fasciolosis is an important plant-borne trematode zoonosis . This disease is of both clinical and veterinary relevance and , according to the WHO , is considered a re-emerging disease that is spreading around the world . Fasciolosis has a serious impact on health because of the large size of the parasite and the effects of the parasite in down-regulating the host immune response . Human fasciolosis can be distinguished by an acute phase , in which the parasite migrates through different tissues , and a chronic phase in which it invades the bile ducts . Here we describe the development of a rapid , simple and inexpensive immunochromatographic diagnostic method , based on the use of a recombinant cathepsin L1 protein , which performs better than other more complex indirect methods , providing similar specificity and higher sensitivity . The simplicity of the method represents a great advantage for the intervention systems applied in different endemic areas by WHO , such as passive case finding ( e . g . Vietnam ) and selective treatment ( e . g . Egypt ) . Because of its characteristics , the system can be applied to both phases of the disease , and in holo , meso and hyperendemic areas where point-of-care testing is required . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"neglected",
"tropical",
"diseases",
"fascioliasis"
] | 2011 | Development and Evaluation of a New Lateral Flow Immunoassay for Serodiagnosis of Human Fasciolosis |
Neonatal meningitis due to Escherichia coli K1 is a serious illness with unchanged morbidity and mortality rates for the last few decades . The lack of a comprehensive understanding of the mechanisms involved in the development of meningitis contributes to this poor outcome . Here , we demonstrate that depletion of macrophages in newborn mice renders the animals resistant to E . coli K1 induced meningitis . The entry of E . coli K1 into macrophages requires the interaction of outer membrane protein A ( OmpA ) of E . coli K1 with the alpha chain of Fcγ receptor I ( FcγRIa , CD64 ) for which IgG opsonization is not necessary . Overexpression of full-length but not C-terminal truncated FcγRIa in COS-1 cells permits E . coli K1 to enter the cells . Moreover , OmpA binding to FcγRIa prevents the recruitment of the γ-chain and induces a different pattern of tyrosine phosphorylation of macrophage proteins compared to IgG2a induced phosphorylation . Of note , FcγRIa−/− mice are resistant to E . coli infection due to accelerated clearance of bacteria from circulation , which in turn was the result of increased expression of CR3 on macrophages . Reintroduction of human FcγRIa in mouse FcγRIa−/− macrophages in vitro increased bacterial survival by suppressing the expression of CR3 . Adoptive transfer of wild type macrophages into FcγRIa−/− mice restored susceptibility to E . coli infection . Together , these results show that the interaction of FcγRI alpha chain with OmpA plays a key role in the development of neonatal meningitis by E . coli K1 .
Professional phagocytes , including neutrophils and macrophages ( MØ ) express a specific set of phagocytic receptors that recognize , bind to and mediate internalization of microbial pathogens [1] , [2] , [3] . Although MØ receptor-mediated phagocytosis generally leads to the destruction of the pathogen , certain receptor-ligand interactions allow for a permissive environment in which the pathogen can thrive and even proliferate . MØ provide a barrier that pathogens must overcome to adhere to and penetrate into tissues . Nonetheless , diverse strategies are used by different bacterial pathogens to subvert phagocytes . Escherichia coli K1 causes meningitis in neonates , which remains a significant problem for the last few decades with case fatality rates ranging from 5 to 40% of infected neonates [4] , [5] , [6] , [7] . Despite treatment with advanced antibiotics , up to 30% of survivors exhibit neurological sequelae such as hearing impairment , mental retardation , and hydrocephalus . Furthermore , due to the emergence of antibiotic resistant strains , mortality rates may significantly increase in future [8] . The crossing of the mucosal epithelium and the invasion of small subepithelial blood vessels by E . coli K1 represent critical early steps in the pathogenesis of meningitis . During initial colonization , E . coli K1 encounters several host defense mechanisms such as complement , neutrophils , and MØ on its path to the blood-brain barrier ( BBB ) . However , very little is known about the mechanisms by which E . coli K1 finds a niche to avoid these host defenses . Our previous studies demonstrated that E . coli K1 evades complement attack by binding to the complement pathway regulator C4bp via outer membrane protein A ( OmpA ) , which subsequently cleaves C3b and C4b complement proteins [9] , [10] . In addition , lack of significant quantities of C9 , a terminal complement component necessary for the formation of the membrane attack complex , in neonatal population gives an additional opportunity for E . coli K1 to survive in the blood [10] . However , our studies have shown that an inoculum of >103 CFU/ml of E . coli K1 is required to resist serum bactericidal activity [11] , indicating that the bacteria must take a refuge in certain cells to survive and multiply during the initial stages of infection , when fewer bacteria are present in the blood . Despite the importance of MØ in innate and adaptive immunity , the interaction of E . coli K1 with these cells is poorly defined . MØ phagocytose a broad range of pathogens by recognizing pathogen-associated molecular pattern ( LPS and peptidoglycans ) via pattern recognition receptors ( PRR ) , which include TLRs , the mannose receptor and the scavenger receptor [12] , [13] . Opsonin-dependent phagocytosis involves complement receptors and antibody-dependent phagocytosis requires Fcγ receptors . Studies from our lab have shown that E . coli K1 enters and multiplies in both human and murine MØ , either in the presence or absence of opsonization . OmpA expression is critical for these processes [14] . Therefore , it is important to determine whether E . coli OmpA interacts with any cell surface proteins of MØ for entry . Numerous studies have shown that the expression of FcγRI is increased during septicemia and meningitis caused by a variety of pathogens [15] , [16] , [17] , although its importance in any of these infections has not been explored . Fcγ receptors ( FcγR ) recognize the Fc region of IgG and play a pivotal role in linking the cellular and humoral arms of the immune response . FcγR comprises a multigene family divided into 3 classes ( FcγRI , II and III ) , which are defined by their affinity for IgG [18] , [19] , [20] , [21] , [22] , [23] . FcγRI is a transmembrane receptor , which binds IgG with high affinity and induces the association of the γ-chain for signal transduction and triggering of effector responses such as MØ phagocytosis [23] . The ligation of FcγRI with IgG also mediates antibody-dependant cellular cytotoxicity induced transcription of cytokine genes and release of inflammatory mediators [24] . The cytoplasmic domain of the γ-chain contains an immunoreceptor tyrosine activation motif ( ITAM ) , which is necessary for the signaling cascade of FcγRs . The cytoplasmic domain of FcγRI has been shown to modulate receptor function , although it does not contain any recognized signaling motif [25] , [26] . In this study , we examined the role of MØ and FcγRI α-chain ( FcγRIa ) in the pathogenesis of neonatal E . coli K1 meningitis by depleting MØ from newborn mice and utilizing a FcγRIa−/− knockout mouse model . Our studies provide evidence of a role for a novel interaction between FcγRIa and OmpA in the onset of meningitis due to E . coli K1 .
Our previous studies have shown that OmpA+ E . coli enters and survives in human and mouse MØ [14] . To determine the role of MØ in the pathogenesis of E . coli K1 induced meningitis , MØ were depleted in newborn mice by the administration of carrageenan [27] , [28] . MØ readily ingest carrageenan in contrast to lymphocytes , which are not actively phagocytic and lack a well-developed lysosomal complex . Due to its unique secondary and tertiary structure , carrageenan is resistant to biochemical degradation by lysosomal glycosidases . Carrageenan containing phagolysosomes eventually rupture due to osmotic swelling . The consequent release of hydrolytic enzymes into the cytosol causes irreversible damage and eventual lysis of MØ [29] . Following three days of carrageenan administration starting at day 1 after birth , the animals showed >95% depletion of MØ from livers and spleens , as shown by flow cytometry after staining with F4/80 antibody ( 5 . 33%±0 . 4% before and 0 . 17%±0 . 1% after carrageenan treatment ) ( Figure 1A ) . However , treatment with carrageenan did not affect B cells ( 39 . 81%±0 . 7% before and 40 . 19%±0 . 9% after carrageenan treatment ) , CD4+ T cells ( 17 . 56%±0 . 5% before and 18 . 02%±0 . 6% after carrageenan treatment ) , CD8+ T cells ( 2 . 11%±0 . 4% before and 2 . 53%±0 . 3% after carrageenan treatment ) , DCs ( 5 . 67%±1 . 2% before and 6 . 09%±0 . 9% after carrageenan treatment ) , or PMNs ( 3 . 98%±1 . 2% before and 4 . 13%±1 . 4% after carrageenan treatment ) in spleens of MØ-depleted mice compared with untreated mice ( Figure S1 ) . The MØ-depleted mice were then infected with 103 CFU of E . coli K1 by intranasal instillation and examined for progression of the disease as previously described [28] . Control animals ( n = 15 for each group ) developed bacteremia at 6 h post-infection , which was increased to 5 . 5 log10 CFU per ml of blood by 48 h ( Figure 1B ) . In contrast , the MØ-depleted mice , despite having a similar number of bacteria in the blood at 6 h , cleared these bacteria from the circulation by 48 h post-infection . In agreement with the bacteremia levels , >90% of control mice developed meningitis at 72 h after infection , whereas none of the MØ-depleted animals showed signs of meningitis and all survived beyond 7 days ( Figure 1C ) . Determination of serum cytokine levels at various times post-infection revealed that control animals produced an initial burst of IL-10 , which peaked at 12 h , and then declined to basal levels by 48 h ( Figure 1D ) . In contrast , the pro-inflammatory cytokines , TNF-α , IFN-γ , IL-1β , IL-6 and IL-12p70 only became detectable in the blood at 12 h post-infection and peaked by 72 h ( Figure 1D and Figure S2 ) . Of note , although the MØ-depleted mice had early production of pro-inflammatory cytokines , their levels were significantly lower than those in the control mice . In these mice , IL-10 levels progressively rose during the initial stages of infection and peaked at 72 h at which time the bacteria were completely cleared from the circulation ( Figure 1D ) . Histopathological examination of control mice infected with E . coli K1 revealed marked infiltration of PMNs in the leptomeningeal and ventricular spaces ( Figure 1E ) . The hippocampus was also inflammed and there was apoptosis of neurons , as indicated by pkynotic nuclei in Ammon's horn . Acute hemorrhage and inflammation was observed , most prominently in the white matter of the brain . The cortex and molecular layer had increased cellularity due to inflammatory exudates . The MØ-depleted mice , however , did not reveal such pathological changes . Blood brain barrier ( BBB ) leakage is the hallmark of neonatal meningitis . Therefore we used the Evans blue extravasation method to quantify BBB leakage in both the control and MØ-depleted mice [28] . The dye was injected intraperitoneally at 68 h post-infection and after four hours , the brains were removed and Evans blue concentration determined . A marked increase in the permeability of the BBB was observed in infected WT animals , which was significantly reduced in MØ-depleted mice , ( p<0 . 001 by student's t test ) ( Figure 1F ) . Furthermore , the number of E . coli K1 entering the brain was approximately 6 . 0 log10 CFU in control animals , whereas the brains of the MØ-depleted animals contained very few bacteria ( Figure 1G ) . These results demonstrate that MØ may be important for E . coli K1 to reach a required level of bacteremia , which is critical for the establishment of neonatal meningitis . Our previous studies have shown that OmpA+ E . coli binds and enters MØ in vitro irrespective of opsonization status of the bacteria [14] . OmpA− E . coli , although entered in lower numbers but failed to survive inside MØ . This indicates that OmpA mediated entry into MØ enables OmpA+ E . coli K1 to resist the normal antimicrobial mechanisms of MØ . Therefore , to understand the nature of the macrophage surface structures that interact with E . coli K1 , biotin-labeled cell surface proteins of THP-1 cells differentiated into MØ ( THP-M ) and RAW 264 . 7 cells were incubated with OmpA+ E . coli , OmpA− E . coli or a laboratory E . coli HB101 . Bound proteins were then released and analyzed by western blotting with streptavidin peroxidase . A small number of proteins bound to all the bacteria from both the cells . However , OmpA+ E . coli prominently bound to the 110 and 70 kDa proteins from both THP-M and RAW 264 . 7 cells , whereas OmpA− E coli bound only to the 110 kDa protein ( Figure 2A ) . Although some proteins bound to HB101 were of similar molecular mass to those bound to OmpA+/OmpA− E . coli , other proteins showed different binding patterns . Based on their molecular masses , we speculated that the proteins binding to E . coli K1 could be Toll like receptor-4 ( 110 kDa ) and FcγRIa ( CD64 , 72 kDa ) . Since OmpA+ E . coli specifically bound to the 70 kDa protein in contrast to OmpA− E . coli , the blots were reprobed with an anti-FcγRI antibody , which reacted with the 70 kDa protein , suggesting that OmpA+ E . coli binds to FcγRIa . Of note , treating the bacteria with 40% pooled human serum did not alter the binding , indicating that opsonization with complement and/or with non-specific antibody did not alter bacterial interaction with macrophage surface proteins . Next , we used blocking antibodies to determine the contribution of OmpA-FcγRIa interaction in E . coli entry into MØ . OmpA+ E . coli was incubated with Fab fragments of anti-OmpA antibody ( polyclonal ) prior to addition to MØ . In other experiments , the RAW 264 . 7 cells were pre-treated with antibodies to FcγRI , CR3 , TLR2 , TLR4 or the mannose receptor prior to addition of OmpA+ E . coli . Isotype matched antibodies or anti-S-fimbria antibodies were used as controls . Both anti-OmpA and anti-FcγRI antibodies reduced the number of bound and intracellular E . coli K1 by ∼80% , whereas other antibodies showed no significant inhibition ( Figure 2B ) . To verify that the anti-FcγRI antibody actually inhibited FcγRI–mediated phagocytosis , the effect of this antibody on the entry of zymosan coated with fluorescent-labeled IgG2a that occurs via FcγRI was also determined . The internalized zymosan particles were counted per 100 cells after quenching the external fluorescence by Trypan Blue [30] . As predicted , anti-FcγRI antibodies significantly inhibited the entry of opsonized zymosan ( Figure 2C ) . MØ pretreated with the anti-FcγRI antibody were also infected with Group B streptococcus ( GBS ) pre-treated with C8-deficient serum ( for deposition of C3 and to avoid bacterial killing by serum ) , which is known to enter MØ through the CR3 receptor [31] , [32] , [33] . The internalization of GBS , however , was not affected by pretreatment with anti-FcγRI antibody , suggesting that it did not interfere with CR3 receptor function in MØ ( Figure 2D ) . However , as expected , anti-CR3 antibodies significantly blocked the binding and entry of GBS into RAW 264 . 7 cells . To further confirm the role of OmpA interaction with MØ in E . coli entry into MØ , OmpA was purified from OmpA+ E . coli and reconstituted into liposomes as previously described [34] , which were used to pre-treat RAW 264 . 7 cells prior to adding the bacteria ( Figure 2E ) . The liposomes containing OmpA blocked both binding and intracellular survival of E . coli K1 by approximately 50% , whereas liposomes containing outer membrane proteins from OmpA− E . coli did not show such inhibition . Increasing concentrations of OmpA liposomes showed no further increase in the inhibition , indicating that the structure of OmpA in liposomes may not be optimal to that of OmpA on E . coli K1 to bind to FcγRIa . The fate of OmpA+ E . coli after phagocytosis by RAW 264 . 7 cells was examined by immunocytochemistry after differential staining . Extracellular bacteria were stained with FITC labeled secondary antibody ( green ) and the intracellular bacteria were stained with a TRITC labeled secondary antibody ( red ) after incubation with primary anti-S-fimbria antibody . As shown in Figure 2F , a number of OmpA+ E . coli bound to RAW 264 . 7 cells , whereas very few OmpA− E . coli bound at 30 min post-infection . Analysis of intracellular bacteria over time revealed that OmpA+ E . coli multiplied , whereas OmpA− E . coli were degraded inside the cells . Collectively , these studies suggest that the OmpA of E . coli K1 interacts with regions of FcγRIa similar to those involved in the binding of Fc and that this interaction enables the organism to enter MØ . In addition , the data suggest that other receptors that recognize pathogen-associated molecules may not play a significant role in MØ binding and entry of E . coli K1 . However , entry through other receptors in the absence of OmpA-FcγRIa interaction renders the bacteria susceptible to macrophage killing . To confirm the role of FcγRIa in OmpA+ E . coli entry of MØ , short hairpin RNA ( shRNA ) sequences for murine FcγRIa and CR3 in pGeneClip Neomycin vectors were used to transfect RAW 264 . 7 cells . Suppression of FcγRIa and CR3 gene transcription and expression was verified by RT-PCR and flow cytometry , respectively . The respective shRNA suppressed the transcription of FcγRIa and CR3 considerably , but had no effect on GAPDH , TLR2 or TLR4 mRNA transcript levels ( Figure 3A ) . On par with changes in transcription levels , the surface expression of FcγRIa and CR3 was significantly reduced , while TLR2 and TLR4 expression was unaltered ( Figure 3B ) . There was >90% reduction in the OmpA+ E . coli phagocytosed by FcγRIa-shRNA/RAW cells compared to control or CR3-shRNA/RAW cells ( p<0 . 001 by two-tailed t test ) ( Figure 3C ) . This reduction was due to inefficient binding of E . coli K1 to these cells , as less than 30% of bacteria were bound by the FcγRIa-shRNA/RAW cells compared to non-transfected or control-shRNA transfected cells . In contrast , both binding and intracellular survival of GBS were not affected in FcγRIa-shRNA/RAW cells , whereas CR3-shRNA transfection caused significant reduction in both of these processes ( Figure 3D ) . Immunocytochemistry of E . coli K1 infected FcγRIa-shRNA/RAW cells revealed that very few cells ingested bacteria and were killed within 2 h post-infection ( Figure 3E , fragmented bacteria ) . However , E . coli K1 entered and replicated in CR3-shRNA/RAW cells similar to control RAW cells . Comparable results were also obtained with THP-M cells transfected with shRNA specific to human FcγRI ( data not shown ) . To further confirm that lack of FcγRIa expression rendered bacteria susceptible to macrophage killing , FcγRIa-shRNA/RAW cells infected with E . coli K1 were examined by transmission electron microscopy . Although few numbers of FcγRIa-shRNA/RAW cells engulfed E . coli K1 , several of them were either degraded or in the process of degradation by 1 h post-infection and were completely killed by 8 h post-infection ( Figure 3F ) . In contrast , CR3-shRNA/RAW cells showed intact bacteria in endosomes undergoing significant multiplication by 8 h post-infection . Taken together these results demonstrate that OmpA-FcγRIa interaction is critical for E . coli K1 to bind to , enter and survive in MØ . The activation of FcγRI in phagocytic cells by the binding of the Fc region of IgG requires the association of FcγRIa with the IgG γ-chain [19] . To examine whether γ-chain association with FcγRI is also necessary for E . coli K1 invasion , COS-1 cells were transfected with pcDNA3 plasmids containing Myc-tagged human FcγRIa ( Myc-hFcγRIa ) , C-terminal truncated Myc-hFcγRIa which lacks the cytoplasmic tail ( CT ) or Myc-hFcγRII . Expression of these proteins was verified by Western blotting using the anti-Myc antibodies ( Figure 4A ) and flow cytometry ( Figure 4B ) . OmpA+ E . coli binding to , and invasion of , hFcγRIa+/COS-1 cells was significantly greater compared to that of mock-transfected cells ( Figure 4C ) . OmpA− E . coli showed very negligible binding to , and invasion into , both FcγRIa transfected and mock-transfected COS-1 cells ( data not shown ) . The invasion of E . coli K1 into Myc-hFcγRIa-CT/COS-1 cells was significantly reduced , although the binding of bacteria to these cells was decreased by only 30% compared to Myc-hFcγRIa+/COS-1 cells . In contrast , overexpression of FcγRII did not increase E . coli binding to , or invasion of , COS-1 cells . These data suggest that FcγRIa acts as receptor for OmpA mediated entry of E . coli K1 into COS-1 cells and that the C-terminal portion is required for this invasion . Next , to examine whether FcγRIa interacts with OmpA , recombinant hFcγRIa ( rhFcγRIa ) was purified by Myc-affinity column chromatography from COS-1 cells and incubated with OmpA+ or OmpA− E . coli . The bound proteins were released and subjected to Western blotting with antibodies to Myc or FcγRI . The purified rhFcγRIa bound to OmpA+ E . coli but not to OmpA− E . coli , whereas BSA , used as a control , did not interact with the bacteria ( Figure 4D ) . rhFcγRIa used to pre-treat bacteria prior to adding them to COS-1 monolayers in the invasion assays resulted in much more significant inhibition of E . coli K1 binding to , and entry into the cells in a dose dependent manner when compared to the BSA control ( Figure 4E ) . These results suggest that the alpha chain of FcγRI is sufficient for E . coli K1 to bind to , and invade , COS-1 cells . One important question to address in these studies is how OmpA of E . coli K1 binds to FcγRIa at the same region as the Fc-portion of IgG in the context of whole blood . Generally , specific or even non-specific IgG in circulation binds invading bacteria and thereby presents the pathogen to FcγR receptors on MØ . Therefore , it is possible that OmpA+ E . coli may be displacing IgG for binding to FcγRI . We tested this hypothesis by performing two different competitive binding experiments . First , OmpA− E . coli were coated with anti-S-fimbria antibody and added to FcγRIa+/COS-1 cells treated with cytochalasin D to prevent internalization . The cells were washed and then various quantities of OmpA+ E . coli were added and incubated for 10 min . After washing the monolayers , the number of OmpA− E . coli that remained bound to COS-1 cells were determined by plating on agar containing tetracycline ( OmpA+ E . coli is sensitive to tetracycline ) . As shown in Figure 5A , IgG2a opsonized OmpA− E . coli bound COS-1 cells in significantly greater numbers compared to unopsonized bacteria and progressively more bacteria were released from the cells as more OmpA+ E . coli were added to the wells . In contrast , OmpA− E . coli could not displace bound OmpA− E . coli . In separate experiments , peritoneal MØ were incubated with FITC-IgG2a ( 1 µg ) for 1 h in the presence of cytochalasin D , washed and then various quantities of OmpA+ E . coli or OmpA− E . coli were added . The cells were incubated for 10 min , washed and the amount of FITC-IgG that remained bound to the MØ was determined by flow cytometry . As shown in Figure 5B , the amount of FITC-IgG2a bound to peritoneal MØ was decreased when OmpA+ E . coli were added , whereas addition of OmpA− E . coli had no effect . These results indicate that the interaction of E . coli K1 with FcγRIa via OmpA can displace bound IgG2a . Binding to the γ-chain of FcγRIa is crucial for inducing the anti-microbial activity of MØ [20] . Since OmpA binding to FcγRIa prevented the killing of the bacteria , we hypothesize that E . coli K1 interaction with FcγRIa avoids the association of the γ-chain . Consistent with this assumption , OmpA+ E . coli interaction with MØ in the presence or absence of IgG2a opsonization induced far less γ-chain association with FcγRIa in comparison to OmpA− E . coli , as shown by immunoprecipitation studies ( Figure 5C ) . Similarly , OmpA+ E . coli induced a distinct tyrosine phosphorylation pattern of macrophage cytoplasmic proteins compared to OmpA− E . coli opsonized with IgG2a ( Figure 5D ) . Taken together , these studies suggest that the interaction of E . coli K1 with FcγRIa can displace the bound IgG2a , which is mediated by OmpA . They also indicate that OmpA-FcγRI interaction induces novel signaling patterns , which may abrogate the normal antimicrobial response of these cells . To confirm the role of FcγRIa in the pathogenesis of E . coli K1 meningitis , FcγRIa−/− mice were used for infection studies . MØ isolated from FcγRIa−/− mice did not express FcγRIa but had unchanged expression of other FcγRs , TLRs , mannose receptor and CR3 were unchanged compared to normal littermates ( data not shown ) . The newborn animals were intranasally infected with E . coli K1 and examined for disease progression . Of note , the FcγRIa−/− animals did not develop bacteremia even at a 100 fold higher infectious dose , even though E . coli K1 entered the circulation within two hours of infection ( Figure S3A ) . In contrast , wild type ( WT ) animals showed 7 . 0 log10 CFU of E . coli K1 in blood at 72 h post-infection ( Figure 6A ) . The FcγRIa−/− mice did not develop meningitis even when infected with a 100-fold greater inoculum ( data not shown ) . These mice did not show any signs of meningitis even after 7 days of infection , whereas 90% of WT mice showed positive CSF cultures by 72 h post-infection ( Figure 6B ) . Cytokine analysis in the sera of these animals demonstrated that infected WT animals generated significant amounts of TNF-α , IL-1β , IL-6 , IFN-γ and IL-12 , but FcγRIa−/− mice did not ( Figure 6C and Figure S3B–D ) . On the other hand , IL-10 production peaked at 24 h post-infection and subsequently returned to basal levels in WT mice , whereas FcγRIa−/− showed increased IL-10 production at 72 h post infection ( Figure 6D ) . We next examined blood-brain barrier leakage in FcγRIa −/− mice . Infection with E . coli K1 caused no leakage in FcγRIa−/− mice , whereas WT animals had significant leakage of Evans blue dye ( Figure 6E ) . Furthermore , no bacterial colonies were detected in the brains of FcγRIa−/− mice , while WT animals had a high bacterial load ( Figure 6F ) . Similarly , the pathology of the brains from FcγRIa−/− mice revealed no infiltration of neutrophils , neuronal damage or gliosis , which are the characteristic pathological features of E . coli K1 meningitis observed in WT bacteria infected mice ( Figure 6G ) . In contrast , infection of FcγRIa−/− mice with GBS resulted in significant bacteremia and development of meningitis ( Figure S4A-C ) . Together these results suggest that FcγRIa expression is critical for E . coli K1 to achieve high-grade bacteremia and for subsequent development of meningitis in newborn mice . Our studies have shown that MØ isolated from E . coli K1 infected mice exhibit increased expression of FcγRI and TLR2 , as well as increased production of nitric oxide ( NO ) due to iNOS activation [28] . We also observed that upregulation of CR3 expression on MØ led to enhanced killing of E . coli K1 , whereas this effect was completely abrogated in CR3 siRNA transfected MØ in vitro . Other investigators have also demonstrated that CR3 , TLR2 and TLR4 play important roles in the phagocytic ability of MØ [35]-[42] . Therefore , we examined whether the inability of E . coli K1 to survive in FcγRIa−/− mice was due to altered expression of surface receptors using flow cytometry . Peritoneal MØ isolated from infected FcγRIa−/− mice exhibited increased expression of CR3 and TLR4 , but lower expression of TLR2 ( Figure 7A ) . These cells also produced lower or negligible quantities of inducible NO upon challenge with E . coli K1 , whereas MØ from WT mice generated six-fold higher amounts of NO at 6 h post-infection ( Figure 7B ) . Furthermore , E . coli K1 binding to , and entry into , bone marrow-derived MØ ( BMDMs ) from FcγRIa−/− mice were significantly lower compared to WT MØ ( Figure 7C and D ) . Some bacteria entered FcγRIa−/−BMDMs , but they were killed within a short period of time as determined by immunocytochemistry ( data not shown ) . To substantiate the role of FcγRI in E . coli K1 entry , FcγRIa−/−BMDMs were transfected with hFcγRIa , FcγRIa-CT or FcγRII and then used for binding and invasion assays . As shown in Figure 7C , E . coli K1 binding to FcγRIa−/−BMDM/FcγRIa and FcγRIa−/−BMDM/FcγRIa-CT increased significantly compared to FcγRIa−/−BMDMs and FcγRIa−/−BMDM/FcγRII . However , entry was limited to binding to FcγRIa−/−BMDM/FcγRIa cells only . These results suggest that FcγRIa expression is critical for E . coli K1 binding to , and entry into , MØ and that the C-terminal domain plays a significant role for the entry . FcγRIa−/−BMDM transfected with a FcγRIa construct exhibited decreased expression of TLR4 and CR3 and increased expression of FcγRI and TLR2 in comparison with FcγRIa−/−BMDM after challenge with E . coli K1 ( p<0 . 01 ) ( Figure 7E ) . Transfection with FcγRIa-CT , however , resulted in only a partial increase or decrease of these surface molecules . In contrast , FcγRIa−/−BMDM transfected with FcγRII showed basal level expression of these molecules . Confirming the requirement of FcγRIa interaction with E . coli K1 to induce NO production , FcγRIa−/−BMDM/FcγRIa cells generated greater quantities of NO by 6 h post-infection as compared to FcγRI−/−BMDM/FcγRI-CT and FcγRI−/−BMDM/FcγRII cells ( Figure 7F ) . Taken together , these data suggest that FcγRIa interaction with OmpA of E . coli K1 is necessary for suppression of CR3 and TLR4 expression and to enhance the expression of FcγRI and TLR2 , and maximal NO production . To confirm the role of FcγRIa expression on MØ in the pathogenesis of E . coli K1 meningitis , FcγRIa−/− mice were reconstituted with FcγRIa+/+ or FcγRIa−/− MØ and then infected with E . coli K1 . FcγRIa−/− mice that received FcγRIa+/+ MØ showed higher blood bacterial numbers compared with animals replenished with FcγRIa−/− MØ ( Figure 8A ) . 94% of CSF cultures were positive for E . coli K1 in FcγRIa+/+ MØ reconstituted mice , whereas all cultures were sterile in animals that received FcγRIa−/− MØ ( Figure 8B ) . BBB disruption was significant in FcγRIa+/+ MØ-replenished animals compared to FcγRIa−/− MØ reconstituted mice ( Figure 8C ) . Higher numbers of bacteria were also recovered from the brains of mice replenished with FcγRIa+/+ MØ compared to animals those received FcγRIa−/− MØ ( Figure 8D ) . These results confirm that FcγRIa expression on MØ is critical for the onset of E . coli K1 meningitis .
The host response to infection starts with the identification of invading microorganisms via innate immune surveillance systems [43] . Nonetheless bacterial pathogens utilize very effective mechanisms to avoid host defenses in order to promote successful replication and dissemination [44] . MØ provide an important innate and adaptive immune coverage in the host , although their importance in E . coli K1 meningitis is unexplored . In the present study , we demonstrate that the expression of FcγRIα-chain in MØ is critical for the survival of E . coli K1 inside these immune cells by using MØ-depleted and FcγRIa−/− mice . It is tempting to speculate that the ability of E . coli K1 to survive inside MØ might enable these bacteria to infect the central nervous system via a “Trojan horse” mechanism . Pathogens that naturally infect the central nervous system , such as Brucella , Listeria , and Mycobacterium , have been demonstrated to use this mode of entry [45] , [46] . We observed that the interaction of OmpA with FcγRIa in MØ is critical for bacterial binding to , entry into , and subsequent survival in these cells . Generally various FcγRs recognize microbes coated with either specific or non-specific antibodies . However a select number of microbes have developed methods to avoid this recognition . Protein A of S . aureus is known to bind to the Fc portion of the antibodies so that it avoids interacting with FcγRI , whereas most other microbes either downregulate phagocytic mechanisms or avoid phagocytosis entirely [47] , [48] . This study therefore depicts the first evidence that a bacterial protein binds directly to FcγRIa to divert anti-microbicidal mechanisms . Our competitive inhibition studies demonstrated that OmpA interacts with FcγRIa and can displace the binding of Fc portion of IgG . Therefore , it is possible that the bacteria in circulation , despite being coated with non-specific IgG , interact with MØ via FcγRIa for binding to and entering the cells for subsequent multiplication . OmpA− E . coli could not survive in MØ , suggesting that the interaction of OmpA with FcγRIa induces survival strategies or suppresses anti-microbial pathways in MØ . However , OmpA− E . coli has been shown to express reduced levels of type 1 fimbriae and susceptible to chemical stresses [49] , [50] . Therefore , it is possible that OmpA− E . coli could be less capable of dealing with macrophage-induced stresses . Listeria , Shigella , and Rickettsia escape from the phagosome to the cytosol to avoid destruction in phagolysosomes [51] . Other pathogens interfere with the normal biogenesis of phagolysosomes , thus leading to the formation of replicative vacuoles [52] , [53] . Since E . coli K1 continue to multiply inside phagosomes , one can speculate that phagosomes containing OmpA+ E . coli avoid lysosomal fusion by blocking phagosome maturation . The receptors expressed on the surface of MØ play a decisive role in the course of infection , whether pathogens are killed or the MØ machinery is taken over by the microbes [54] . Receptors like TLR2 , TLR4 and CR3 have been implicated in the phagocytic ability of MØ [55] , [56] , [57] . Downregulation of CR3 expression on the surface of MØ has been associated with the decrease in the phagocytosis of pathogens and hence survival inside MØ [58] . TLR2 expression has been shown to prolong survival of Staphylococcus aureus inside phagosomes in MØ , which may be a strategy adopted by this pathogen to evade innate immunity . On par with this concept , TLR2 or MyD88 KO mice have been demonstrated to be resistant to sepsis , indicating that TLR2 mediated signaling is playing an important role in the survival of bacterial pathogens [59] . Activation of MØ through TLR4 has been shown to direct the induction of Th1 and Th-17 cells , which mediate protective cellular immunity to Bordetella pertussis by enhancing the bactericidal activity of MØ [60] . It is still to be determined whether TLR2 expression upon E . coli K1 infection has any role in the pathogenesis of meningitis . We recently demonstrated that iNOS−/− mice and aminoguanidine ( iNOS specific inhibitor ) treated MØ showed enhanced expression of CR3 and TLR4 and very low levels of TLR2 and FcγRI , indicating that iNOS suppression results in decreased expression of FcγRI [28] . In agreement with these studies , we showed here that lack of FcγRI in MØ prevented the production of inducible NO and increased the expression of CR3 and TLR4 , indicating that OmpA-FcγRIa interaction is critical for manipulating the surface expression of CR3 and TLRs in MØ . Our current results indicate that in E . coli K1 pathogenesis , FcγRI interaction with OmpA enhances the expression of TLR2 , which in turn can be utilized by the bacteria as a receptor to modulate the efficiency of phagosome formation . Alternatively , E . coli K1 interaction with FcγRIa activates non-microbicidal mechanisms for the bacterial survival in MØ . Our studies have demonstrated that E . coli K1 infected MØ also exhibit increased expression of gp96 , a known chaperone for TLR2 and TLR4 [28] . These interactions may also induce effector proteins into MØ by E . coli K1 that eventually are responsible for the control of macrophage environment . Further studies are in progress to examine these possibilities . As cytokines are known to modulate MØ microbicidal activity , it is also possible that the surface expression of TLRs and CR3 could be controlled by the circulating cytokines in E . coli K1 infection . Of note , we have demonstrated that IL-10 administration suppressed the expression of FcγRI and enhanced the expression of TLR4 and CR3 , which in turn prevented the survival of E . coli K1 in MØ [61] . In contrast , for several other pathogens , circulating IL-10 supports intracellular replication , indicating that E . coli K1 pathogenesis is distinct from that induced by other bacterial pathogens [62] . Previous studies have shown that the cytoplasmic ( CY ) domain of FcγRIa plays an important role in phagocytosis and antigen presentation [63] . However , lack of the CY domain neither alters the association of γ-chain with FcγRIa nor influences the tyrosine phosphorylation of γ-chain in response to receptor specific cross-linking [63] . In contrast to these findings , we observed that OmpA binding to FcγRIa did not induce the association of γ-chain despite the presence of the CY domain . This binding also induced a different tyrosine phosphorylation response in MØ . Therefore , the CY domain of FcγRIa induces signaling events independent of γ-chain during the invasion of E . coli K1 . Similarly , Qin et al demonstrated that the CY domain induces different gene expression in murine MØ compared to MØ stably transfected with CY-deleted FcγRIa [64] . Alteration of signal transduction pathways to impair FcγR-mediated phagocytosis has also been observed in HIV infected MØ , which have downregulated the expression of the γ-subunit [65] . Moreover , direct interaction of periplakin with the CY domain of human FcγRIa can confer unique properties on this receptor [66] . It should be noted that there are significant differences in the cytoplasmic regions of human and murine FcγRIa . However , our data demonstrate that the interaction of OmpA induced a similar response in both human and murine MØ . In summary , our studies provide the first evidence that a bacterial protein interacts directly with FcγRIa in order to bind to and enter MØ and manipulates the intracellular signaling for bacterial survival and multiplication . The new repertoire of interaction also suggests that MØ function may be manipulated by targeting additional epitopes without activating MØ microbicidal function . This strategy will be useful for devising novel methods of therapy for other diseases involving FcγRIa in addition to neonatal E . coli K1 meningitis .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal care and Use Committee ( IACUC ) of The Saban Research Institute of Childrens Hospital Los Angeles ( Permit number: A3276-01 ) . All surgery was performed under sodium pentobarbital anesthesia , and all efforts were made to minimize suffering . E . coli E44 , a rifampin-resistant mutant of E . coli K1 strain RS 218 ( serotype O18:K1:H7 ) , has been isolated from the cerebrospinal fluid of a neonate with meningitis and invades human brain microvascular endothelial cells ( HBMEC ) [34] . E91 , a derivative of E44 in which ompA gene is disrupted ( designated as OmpA− E . coli ) and HB101 ( a laboratory E . coli strain that expresses K-12 capsular polysaccharide ) are noninvasive in HBMEC [34] . Group B streptococcus type III strain COH-1 used in these studies was provided by Dr . Craig Rubens of Seattle Children's Hospital , Seattle [67] . All bacteria were grown in brain heart infusion broth with appropriate antibiotics as necessary . Bacterial media were purchased from Difco laboratories ( Detroit , MI ) . Murine MØ cell line RAW 264 . 7 , human macrophage like cells , THP-1 and COS-1 cells were obtained from American Type Culture Collection ( Manassas , VA ) . COS-1 cells were stably transfected with cDNA encoding human FcγRIa , a mutant form of FcγRIa containing a stop codon after first amino acid of the cytoplasmic domain ( Lys315→Stop 315 ) ( FcγRI-CT ) , or with human FcγRII [67] . Anti- FcγRI ( blocks the binding of Fc-portion of IgG to FcγRI ) , anti-CD11b , anti-CD32 , anti-MR , anti-TLR2 , anti-TLR4 and anti-Myc antibodies were obtained from Cell signaling . Purified IgG2a and FITC-IgG2a were obtained from Sigma ( St . Louis , MO ) . Anti-gp96 antibody was raised in our lab as previously described [34] , [68] . Anti-phospho-tyrosine antibody ( 4G10 ) was obtained from BD Sciences and all secondary antibodies coupled to various fluorophores were obtained from Bio-Rad Labs ( Hercules , CA ) . Confluent MØ monolayers in 24-well plates were incubated with 1×106 E . coli K1 ( multiplicity of infection of 10 ) in experimental medium ( 1:1 mixture of Ham's F-12 and M-199 containing 5% heat-inactivated fetal bovine serum ) for 60°min at 37°C , whereas COS-1 cell monolayers were infected with E . coli K1 at an MOI of 100 for 1 . 5 h . The monolayers were washed three times with RPMI 1640°and further incubated in experimental medium containing gentamicin ( 100 µg/ml ) for 1 h to kill extracellular bacteria . The monolayers were washed again and lysed with 0 . 5% Triton X-100 . The intracellular bacteria were enumerated by plating on sheep blood agar . In duplicate experiments , the total cell associated bacteria were determined as described for invasion except that the gentamicin step was omitted . SureSilencing shRNA plasmids to mouse FcγRIa and CR3 ( CD11b ) in the pGeneClip Neomycin Vector were obtained from Super Array Inc . , ( Frederick , MD ) . RAW 264 . 7 cells were transfected with shRNA plasmids using Lipofectamine 2000 and later selected for G418 resistant colonies . The cell surface proteins of THP-1 cells differentiated into MØ ( THP-M ) and RAW 264 . 7 cells were biotinylated by adding to 0 . 1 M sodium bicarbonate buffer ( pH 8 . 0 ) containing 0 . 5 mg/ml NHS-LC-Biotin ( Pierce Co , Rockford , IL ) at a final protein concentration of 2 mg/ml in tissue culture flasks . The flasks were incubated on ice for 1 h , the cells were extensively washed with ice-cold PBS and solubilized in 5% Triton X-100 in PBS . Total membranes from the cells were isolated following extensive dialysis against PBS and then were concentrated using Centricon tubes ( Millipore , Bedford , MA; 10-kDa cut-off ) . Biotinylated proteins ( 2–5 µg ) were incubated with various bacteria from a 5-ml overnight culture in a volume of 0 . 5 ml at 37°C on a rotator for 1 h . The bacteria were then centrifuged and the pellets were washed three times with PBS containing 0 . 1% Triton X-100 . After a final wash , the bound proteins were released with Laemmli buffer in the presence of β-mercapto-ethanol and analyzed by SDS-PAGE . The separated proteins were transferred to nitrocellulose and immunoblotted with streptavidin coupled to peroxidase . The protein bands were visualized by ECL reagent ( Amersham Biosciences , Piscataway , NJ ) . Total RNA was isolated from various transfected RAW 267 . 4 cells with TRIZOL-LS-reagent ( Gibco BRL , Gaithersburg , MD ) and quantified using a nanodrop machine . RT-PCR was performed using the following primer sequences: FcγRIa ( 321 bp ) FP 5′-TCCTTCTGGAAAATACTGACC-3′ and RP 5′ GTTTGCTGTGGTTTGAGACC-3′; TLR2 ( 459 bp ) FP 5′-TGAGAGTGGGAAATA TGGAC-3′; RP 5′-CCTGGCTCTATAACTCTGTC-3′; TLR4 ( 506 bp ) FP 5′- TGGAT ACGTTTCCTTATAAG-3′ and RP 5′-GAAATGGAGGCACCCCTTC- 3′; GAPDH ( 479 bp ) , FP 5′-CACAGTCCATGCCATCACTG-3′ and RP 5′- TACTCCTTGGAG GCCATGTG -3′ . Negative control assays without primers were performed in parallel for every reaction . The amplified products were separated on a 1% agarose gel and were stained with ethidium bromide . Expression of FcγRI , CR3 , TLR2 and TLR4 was detected by staining with appropriate FITC- , phycoerythrin ( PE ) - , PE-CY5 . 5- , or allophycocyanin ( APC ) -coupled mouse monoclonal antibodies ( eBiosciences , San Diego , CA ) . Cells were first pre-incubated for 20 min with IgG blocking buffer to mask non-specific binding sites and then further incubated with the indicated antibodies or an isotype control antibody for 30 min at 4°C . The cells were subsequently washed three times with PBS containing 2% FBS and fixed with BD Cytofix ( BD Biosciences ) . Cells were then analyzed by four-color flow cytometry using FACS calibur Cell Quest Pro software ( BD Biosciences , San Jose , CA ) . Side and forward scatter parameters for which F4/480 was used as a MØ-gating marker , which formed the collection gate and at least 5000 events within this gate were collected for analysis . Newborn C57BL/6 mice were injected intraperitoneally with ( 20-mg/Kg body weight ) α-carrageenan ( Sigma , St . Louis , MO ) on days 1 , 2 and 3 before infecting with E . coli . In control groups , mice were treated with equal volumes of saline . Three-day old mice were randomly divided into various groups and infected intranasally with 103 CFU of bacteria . Control mice received pyrogen free saline through the same route . Blood was collected from the tail or facial vein at designated times post-infection and plated on LB agar containing rifampicin to assess bacteremia and level of infection . CSF samples were collected aseptically under anesthesia by cisternal puncture and directly inoculated into broth containing antibiotics . Mice were perfused intracardially with 0 . 9 % saline to remove blood and contaminating intravascular leukocytes . Brains were aseptically removed and homogenized in sterile PBS . Bacterial counts in all tissues were determined by plating ten-fold serial dilutions on rifampicin LB agar plates . Growth of E . coli in rifampicin containing LB broth from the CSF samples was considered positive for meningitis [28] . Determination of leukocytes in livers and spleens of untreated and carrageenan treated mice was done using flow cytometry [61] . PMNs were identified by staining with anti-Ly6-G ( GR-1 ) followed by goat anti-rat- phycoerythrin ( PE ) . CD4+ and CD8+ T lymphocytes were stained with rat anti-mouse-CD4 followed by goat anti-rat-PE and anti-CD8-FITC . DCs were stained with APC conjugated anti-CD11c antibody . B lymphocytes were detected by staining with anti-CD45R ( B220 ) -FITC . Flow cytometry was performed on a FACScan instrument ( BD Biosciences , CA ) and the data were analyzed with Cell Quest Software . Total cell lysates of RAW 264 . 7 cells infected with bacteria for varying time periods were centrifuged at 16 , 000 X g for 20 min at 4°C . The supernatants were collected and the protein contents determined . For immunoprecipitation studies , 300–500 µg of protein was incubated with the appropriate antibody overnight at 4°C , washed and further incubated for 1 h with protein A-agarose . The immune complexes were washed four times with cell lysate buffer and the proteins bound to agarose were eluted in SDS sample buffer for further analysis by Western blotting . Portions of the cell lysates were subjected to electrophoresis on a 10% SDS-polyacrylamide electrophoresis gel . The proteins were transferred to a nitrocellulose membrane , which was then blocked with 5% bovine serum albumin ( BSA ) in Tris-buffered saline containing 0 . 05% Tween 20 ( TBST ) for 2 h at room temperature . The blot was then incubated with the primary antibody overnight at 4°C in 5% BSA/TBST . The blot was washed with TBST and further incubated with the horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature . Subsequently , the blot was washed four times with TBST for 1 h , developed with SuperSignal chemiluminescence reagent , and exposed to x-ray film to visualize the proteins . RAW 264 . 7 cells were incubated with E . coli K1 at an MOI of 10 for varying times , washed and then fixed with 2% glutaraldehyde in 0 . 1 M cacodylate buffer , pH 7 . 1 . All samples were washed three times in 0 . 1 M cacodylate buffer for 15 minutes each . The cells were then post-fixed for 20 minutes in 1% osmium tetroxide at 4°C followed by addition of EtOH ( 60% ) . Samples were dehydrated through 70 , 80 , 95 , and 100% EtOH ( two times , 15 min each ) , then into propylene oxide ( two time , 15 min each ) , and into a 1:1 propylene oxide/Eponate , left overnight , capped , at room temperature . The propylene oxide/Eponate mixture was decanted off and replaced with 100% Eponate mixture . The samples were polymerized at 70°C for 48 h . Thin sections ( ∼80 nm ) were cut using a diamond knife , mounted on un-coated 300 mesh copper grids and stained with 5% uranyl acetate for 20 min . Photographs were take with a transmission electron microscope ( JEOL JEM 2100 LaB6 ) equipped with a Gatan Ultra Scan 1000 CCD camera . COS-1 cells were grown in 24-well tissue culture plate to confluence and then treated with 0 . 5 µg/ml of cytochalasin D for 30 min prior to addition of bacteria . OmpA− E . coli were incubated with anti-S-fimbria antibody for 1 h on ice , washed , and then added to the COS-1 monolayers at an MOI of 100 for 1 h . OmpA− E . coli alone infected monolayers served as controls in these experiments . The monolayers were then washed to remove unbound bacteria and incubated with OmpA+ E . coli at an MOI of 10 and 100 for 10 min , washed the monolayers , and then dissolved in 150 µl of PBS containing 0 . 3% Triton X-100 . Serial dilutions were made and plated on agar containing tetracycline ( 12 . 5 µg/ml ) in which only OmpA− E . coli grow . The number of CFU was counted and determined the percent displacement by OmpA+ E . coli . In some experiments , FITC-IgG2a ( 1 µg ) was incubated with cytochalasin-D treated peritoneal MØ while rotating the test tube at a low speed for 30 min and washed to remove unbound IgG . Various inocula of OmpA+ E . coli or OmpA− E . coli were added to the cells and incubated for 10 min , washed and the bound FITC-IgG was determined by flow cytometry . RAW 264 . 7 cells were grown in eight-well chamber slides and infected with E . coli K1 as described above . The monolayers were then washed with PBS and fixed in 2% paraformaldehyde for 10 min at room temperature . Subsequently , anti-S-fimbria antibody ( 1:1000 dilution ) was added to the cells and incubated for 1 h at room temperature . The cells were then washed with PBS and incubated with secondary antibodies conjugated to FITC for 30 min at room temperature . The monolayers were washed four times with PBS and incubated with excess amounts of secondary antibody coupled to horseradish peroxidase for 1 h at RT to block the external primary antibody sites . After thorough washing of the cells , the monolayers were permeabilized with 5% normal goat serum in phosphate-buffered saline containing 1% Triton X-100 ( NGS/PBST ) for 30 min . The cells were again incubated with anti-S-fimbria antibody for 1 h in Triton/NGS/PBST buffer , washed and further incubated with secondary antibody coupled to Cy3 for 30 min . The cells were washed again , the chambers removed , and the slides mounted in Vectashield ( Vector Laboratories ) anti-fade solution containing 4′ , 6-diamidino-2-phenylindole . Cells were viewed using a Leica ( Wetzlar , Germany ) DMRA microscope with Plan-apochromat ×40/1 . 25 NA and ×63/1 . 40 NA oil immersion objective lenses . Image acquisition was with a SkyVision-2/VDS digital CCD ( 12-bit , 1280×1024 pixel ) camera in unbinned or 2×2-binned models into EasyFISH software , saved as 16-bit monochrome , and merged as 24-bit RGB TIFF images ( Applied Spectral Imaging Inc . , Carlsbad , CA ) . The images were assembled and labeled using Adobe PhotoShop 7 . 0 . BBB permeability was quantitatively evaluated by detection of extravasated Evans blue dye [28] . Briefly , 2% Evans blue dye in saline was injected intraperitoneally into infected or uninfected mice and after 4 h , mice were deeply anesthetized with Nembutal and transcardially perfused with PBS until colorless perfusion fluid was obtained from the right atrium . Brains from infected animals were harvested , weighed and homogenized . Tissue supernatant was obtained by centrifugation and protein concentration was determined . Evans blue intensity was determined on a microplate reader at 550 nm . Calculations were based on external standards dissolved in the same solvent . The amount of extravasated Evans blue dye was quantified as micrograms per milligram protein . Peritoneal MØ were isolated from mice according to the method of Mittal et al [28] , [69] , [70] . Briefly , the mouse peritoneal cavity was exposed carefully without disrupting blood vessels and 2–3 ml of RPMI was slowly injected . The lavage was collected and cultured in tissue culture flasks for 2 h at 37°C under 5% CO2 to allow adherence of MØ . Non-adherent cells were removed and the flasks washed three times with Hanks' solution . The adherent cells were harvested from the flasks using a rubber policeman and were resuspended in 10% FCS-RPMI 1640 medium . MØ were then positively selected using Miltenyi biotech kit and percentage purity examined by FACS analysis using F4/80 antibody , which was >97% . Viability of MØ following interaction with bacteria was assessed using an Annexin V kit ( BD Biosciences , San Diego , CA ) . Mouse bone marrow cells were isolated from the tibias and femurs of 6- to 10-wk-old WT and FcγRI−/− mice [71] . After euthanasia of mice by CO2 asphyxiation , femurs were harvested and bone marrow cells aseptically flushed from the marrow cavities with ice-cold PBS . Cells were collected by centrifugation and erythrocytes were lysed by resuspending in 0 . 15 M NH4Cl for 3–5 min . Celle were washed with PBS and resuspended in complete DMEM medium supplemented with M-CSF ( 10 ng ml−1 ) and IL-3 ( 10 ng ml−1 ) , plated and allowed to differentiate into MØ . After 5–7 days in culture , adherent MØ were washed with PBS , scraped gently into suspension and counted . The purity of the MØ was determined by flow cytometry using F4/80 antibody and found to be >95% . Fresh bone marrow derived MØ ( 5×106 cells ) were transferred by intraperitoneal injection into mice 6 h before infecting with E . coli K1 . NO production was determined in MØ supernatants by a modified Griess method as described earlier [28] , [72] , [73] . Briefly nitrate was converted to nitrites with β-nicotinamide adenine dinucleotide phosphate ( NADPH , 1 . 25 mg ml−1 ) and nitrate reductase followed by addition of Griess reagent . The reaction mixture was incubated at room temperature for 20 min followed by addition of TCA . Samples were centrifuged , clear supernatants were collected and optical density was recorded at 550 nm . The amounts of NO produced were determined by calibrating standard curve using sodium nitrite . Half of the brain was fixed in 10% buffered formalin , routinely processed and embedded in paraffin . 4–5 µm sections were cut using a Leica microtome and stained with hematoxylin and eosin ( H & E ) . Pictures were taken using a Zeiss Axiovert Microscope connected to a JVC 3-chip color video camera and read by the pathologist in a blinded fashion . Cytokine ( TNF-α , IL-1β , IL-6 , IL-12 p70 and IL-10 ) levels in sera from various animals were determined using Biosource ELISA kits ( Invitrogen , Carlsbad , CA ) according to the manufacturer's instructions . For statistical analysis of the data , two tailed Fischer test , Wilcoxon signed rank test and Student's t-test were applied and p value <0 . 05 was considered statistically significant . | Escherichia coli K1 is the most common cause of meningitis in premature infants; the mortality rate of this disease ranges from 5% to 30% . A better understanding of the pathogenesis of E . coli K1 meningitis is needed to develop new preventative strategies . We have shown that outer membrane protein A ( OmpA ) of E . coli K1 , independent of antibody opsonization , is critical for bacterial entrance and survival within macrophages . Using a newborn mouse model , we found that depletion of macrophages renders the animals resistant to E . coli K1 induced meningitis . OmpA binds to α-chain of Fcγ-receptor I ( FcγRIa ) in macrophages , but does not induce expected gamma chain association and signaling . FcγRIa knockout mice are resistant to E . coli K1 infection because their macrophages express more CR3 and are thus able to kill bacteria with greater efficiency , preventing the development of high-grade bacteremia , a pre-requisite for the onset of meningitis . These novel observations demonstrate that inhibiting OmpA binding to FcγRIa is a promising therapeutic target for treatment or prevention of neonatal meningitis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/bacterial",
"infections",
"infectious",
"diseases/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"microbiology/cellular",
"microbiology",
"and",
"pathogenesis",
"microbiology/innate",
"immunity"
] | 2010 | Fcγ Receptor I Alpha Chain (CD64) Expression in Macrophages Is Critical for the Onset of Meningitis by Escherichia coli K1 |
Identifying the factors that determine microbial growth rate under various environmental and genetic conditions is a major challenge of systems biology . While current genome-scale metabolic modeling approaches enable us to successfully predict a variety of metabolic phenotypes , including maximal biomass yield , the prediction of actual growth rate is a long standing goal . This gap stems from strictly relying on data regarding reaction stoichiometry and directionality , without accounting for enzyme kinetic considerations . Here we present a novel metabolic network-based approach , MetabOlic Modeling with ENzyme kineTics ( MOMENT ) , which predicts metabolic flux rate and growth rate by utilizing prior data on enzyme turnover rates and enzyme molecular weights , without requiring measurements of nutrient uptake rates . The method is based on an identified design principle of metabolism in which enzymes catalyzing high flux reactions across different media tend to be more efficient in terms of having higher turnover numbers . Extending upon previous attempts to utilize kinetic data in genome-scale metabolic modeling , our approach takes into account the requirement for specific enzyme concentrations for catalyzing predicted metabolic flux rates , considering isozymes , protein complexes , and multi-functional enzymes . MOMENT is shown to significantly improve the prediction accuracy of various metabolic phenotypes in E . coli , including intracellular flux rates and changes in gene expression levels under different growth rates . Most importantly , MOMENT is shown to predict growth rates of E . coli under a diverse set of media that are correlated with experimental measurements , markedly improving upon existing state-of-the art stoichiometric modeling approaches . These results support the view that a physiological bound on cellular enzyme concentrations is a key factor that determines microbial growth rate .
Traditional metabolic modeling techniques involve the reconstruction of kinetic models based on detailed knowledge on enzyme kinetic parameters for all enzymes in a certain system [1] . These models are limited to small-scale systems due to lack of sufficient data on kinetic constants and the highly complex nature of these models . An alternative approach called Constraint-Based Modeling ( CBM ) predicts certain steady-state cellular metabolic phenotypes in microorganisms on a genome-scale by relying solely on simple physical-chemical constraints , without requiring enzyme kinetic data [2] , [3] , [4] . This approach identifies steady-state flux rates ( in units of mmol/ ( g[DW]*h ) through a metabolic network , satisfying stoichiometric mass-balance as well as reaction directionality constraints , such that nutrients taken up with a certain measured rate ( in units of mmol/g[DW]*h ) are transformed into biomass . The metabolic network includes a biomass production reaction that consumes essential biomass metabolites , with its stoichiometric coefficients representing the molar quantities required for generating a unit mass of cells ( in units of mmol/g[DW] ) . This reaction's flux activity represents the growth rate ( in units of 1/h ) . CBM is now commonly used for metabolic engineering in microorganisms , predicting the effect of gene knockouts on organism viability [2] . Flux Balance Analysis ( FBA ) is a commonly used CBM approach that enables to predict biomass production yield ( in units of gram biomass/gram nutrient ) based solely on reactions' stoichiometry and directionality ( i . e . without measurements of nutrient uptake rates ) . Given information only on reactions' stoichiometry and directionality , the prediction of biomass yield works by searching for a feasible flux distribution with maximal flux through the biomass production reaction , considering an arbitrary upper bound on the uptake rate of the carbon nutrient . The maximal biomass production rate predicted by FBA reflects optimal yield metabolism and is equal to the assumed uptake rate multiplied by maximal biomass yield . The prediction of actual growth rate by FBA is theoretically possible when experimental measurement of nutrient uptake rates is available and is used to constrain the uptake flux in the model ( or alternatively , by multiplying FBA-predicted biomass yield with the measured uptake rates ) . However , experimental studies have shown that microorganisms exhibit non optimal-yield metabolism under various conditions , for example , in the case of over-flow metabolism where excess nutrient uptake is metabolized inefficiently [5] , [6] . In fact , growth rate was found to be inversely correlated with biomass yield in some microorganisms under different growth environments ( see Section 4 in Supp . Material of [7] ) . Hence , growth rate prediction obtained by FBA ( reflecting optimal yield metabolism ) are likely to be unrealistically high in many cases . Predicting the correct growth rate even when nutrient uptake rates are known is a challenging task . A more ambitious conceptual challenge is the prediction of growth rate without measurements of nutrient uptake rates under a variety of environmental and genetic conditions . FBA with Molecular Crowding ( FBAwMC ) is a recently developed extension of FBA which was shown to enable the prediction of growth rates of E . coli across a small set of growth media ( without given measurements of nutrients uptake rates ) , as well as under conditions of over-flow metabolism [8] , [9] . This was achieved by accounting for the enzyme concentrations required for catalyzed metabolic flux ( utilizing data on enzyme kinetic constants ) , considering a physiological upper bound on the total cellular volume used by metabolic enzymes . Other recent modeling approaches aim to predict cellular metabolism by integrating molecular crowding constraint with kinetic parameters: ( i ) A recent study has utilized a variant FBAwMC to predict inefficient metabolism in cancer cells , in accordance with the Warburg effect [10] . ( ii ) A method by Zhuang et al [11] accounts for a constraint relating to the competition for membrane space between nutrient transporters and respiratory chain proteins was shown to improve metabolic prediction , without requiring explicit data on nutrient uptake rates . ( iii ) Goelzer et al [12] models cellular metabolism by accounting for both solvent capacity constraints and translation apparatus . Another method recently shown to utilize enzyme turnover numbers to improve metabolic flux prediction is Integrative Omics Metabolic Analysis ( IOMA ) , requiring further quantitative proteomic and metabolomics data as input [13] . Another method that aims to predict cellular metabolism without requiring nutrient uptake rates is E-flux [14] , which relies on high-throughput gene expression data ( shown to predict growth rates in a qualitative manner ) . Still , none of these approaches were shown to successfully predict in a quantitative manner the growth rate of microbes across conditions , without utilizing a-priori data on nutrient uptake rates . In this paper , we present a method , MetabOlic Modeling with ENzyme kineTics ( MOMENT ) , for predicting metabolic fluxes and growth rates by accounting for the maximal cellular capacity for metabolic enzymes without the requirement of experimentally determined uptake rates . Extending upon FBAwMC , MOMENT accurately quantifies the enzyme concentrations required for catalyzing each metabolic reaction based on known kinetic constants , accounting for isozymes , protein complexes and multi-functional enzymes . MOMENT is shown to predict growth rates for E . coli under a diverse set of growth media that are significantly correlated with experimental measuements , without requiring measured nutrient uptake rates , significantly outperforming the prediction accuracy of FBAwMC . Furthermore , MOMENT is shown to markedly improve the prediction performance of various metabolic phenotypes , including metabolic fluxes and expression level of metabolic genes . We begin our analysis by exploring the relation between enzyme kinetic parameters and measured metabolic flux , showing a design principle in which enzymes catalyzing high flux reactions across different media tend to be more efficient in terms of having higher turnover numbers ( hence requiring lower concentration to achieve a certain flux rate ) . This suggests that a a physiological constraint on total cellular enzyme concentration , which underlies MOMENT , significantly affects cellular metabolism and the evolution of enzyme kinetic parameters .
An enzyme turnover number is defined as the maximal number of molecules of substrate that the enzyme can convert to product per catalytic site per unit of time . We extracted enzyme turnover numbers for 251 reactions from BRENDA [15] and SABIO-RK [16] databases . To infer genome-scale metabolic flux rates , we utilized several dozen metabolic fluxes under various growth rates in glucose minimal media ( obtained from Ishii et al . [17] and Schuetz et al . [18] ) , and integrated them with a genome-scale metabolic network model of E . coli [19] to infer the most likely rates through the entire network . Specifically , this was done based on standard quadratic programming optimization by minimizing the Euclidian distance between the predicted and the measured fluxes to fit the predicted fluxes to measured ones [20] . Notably , this analysis does not make usage of kinetic data as input . As an alternative approach for inferring global flux distributions , we employed Flux Balance Analysis , followed by Flux Variability Analysis [21] , to identify metabolic reactions whose flux can be uniquely determined based on stoichiometric mass-balance constraints and maximal biomass yield assumption ( obtaining overall similar results in the analysis described below for the flux distributions obtained by the two approaches; Table S1; Figure S1 ) . When comparing the enzyme kinetic parameters in E . coli and measured flux rates , we found that enzymes catalyzing high flux reactions have high turnover number , with statistically significant Pearson correlations of 0 . 45 ( p-value = 7 . 8e-5 ) and 0 . 46 ( p-value = 3 . 6e-5 ) between turnover rates and fluxes under conditions of low and high growth rates , respectively ( Figure 1; considering base 10 log of both fluxes and turnover numbers ) . These correlations suggest that higher selection pressure for enzymatic efficiency ( i . e . higher turnover rates ) acts on enzymes carrying high flux reactions . Notably , our results extend upon a recent finding that central metabolic enzymes have higher turnover rates than secondary metabolic enzymes [22] , by considering actual flux rates instead of relying on rough categorization of enzymes to primary and secondary metabolism . Enzyme molecular weights were computed based on genomic sequences , extracted from KEGG [23] . A statistically significant Pearson correlation of 0 . 22 ( p-value = 3 . 2e-5 ) was also found between metabolic flux rates and enzyme molecular weights , indicating higher flux rates for enzymes with high molecular weights ( Figure 1A ) . Interestingly , a simple linear regression model that aims to explain metabolic flux rates based on both enzyme turnover numbers and molecular weights provided a Pearson correlation of 0 . 55 ( p-value = 8 . 1e-7 ) with metabolic flux rates , suggesting that each parameter contributes independently to explaining flux rates ( Figure 1A ) . While enzyme kinetic parameters are scarcely used in genome-scale metabolic modeling approaches , gene expression data is commonly utilized as the basis for metabolic flux prediction [24] , [25] . However , computing the correlation between the above described flux rates and gene expression [26] measured also under glucose minimal media , resulted in Pearson correlations of only 0 . 26 and 0 . 265 under low and high growth rates , respectively . The latter correlations are markedly lower than those obtained between flux rates and enzyme turnover numbers . This is a remarkable result considering that both the gene expression and metabolic fluxes were measured under the very same growth media , while the kinetic parameters are constant characteristics of the enzymes across different growth conditions . Adding the gene expression data to the above described regression model provided an insignificant contribution to metabolic flux predictions ( Figure 1A ) . Futher utilizing proteomic data for 67 enzyme-coding genes in E . coli measured under the same growth media [26] , we did not find a significant correlation between protein concentrations and the metabolic flux rates . These findings further highlight the importance of utilizing enzyme kinetic data as a prime data source for metabolic flux prediction . Having shown that enzyme turnover numbers are significantly correlated with measured flux rates under glucose minimal media , we set to examine the correlation between enzyme turnover numbers and flux rates under a diverse set of growth media . Towards this end , we applied FBA to predict likely flux distributions under a set of media ( listed in the Methods ) , setting the growth rate to experimental measurements and optimizing for maximal yield . We find that the average Pearson correlation between the enzyme turnover numbers and the predicted fluxes across these media is 0 . 46 ( Figure 2A ) . Next , we computed the correlation between the mean flux rate per enzyme across the growth media and enzyme turnover numbers , finding a Pearson correlation of 0 . 52 , which is higher than the correlations obtained under any single medium . This result suggests that enzyme turnover rates may potentially evolve to support efficient metabolism across multiple media . To explore whether metabolism is better tuned for a specific growth medium , we compared the correlation between predicted fluxes and enzyme turnover numbers achieved for aerobic versus anaerobic conditions ( Figure 2B ) . We found that the correlation between predicted fluxes and enzyme turnover numbers is significantly higher in aerobic conditions ( paired Wilcoxon test p-value = 3e-15 ) , suggesting a potentially stronger selection pressure for efficient metabolism under aerobic conditions . These results suggest that data on enzyme kinetics and metabolic flux may provide valuable insight into organisms' natural environment , in line with previous attempts to do so via other molecular data sources such as codon usage and gene expression [27] . The fact that enzyme turnover numbers and the molecular weights of enzymes are significantly correlated with metabolic flux rates suggests that the utilization of the latter data sources within metabolic modeling approaches may provide improved prediction accuracy of metabolic phenotypes ( as also shown in [9] , [10] , [11] , [12] ) . Towards this end , we developed a method called MetabOlic Modeling with ENzyme kineTics ( MOMENT ) , which utilizes the kinetic parameters under the limitation of the total enzymatic pool available . Given a growth condition of interest , MOMENT predicts a flux distribution that satisfies stoichiometric mass-balance and reaction directionality constraints , such that the total mass of enzymes required to catalyze the predicted flux is bounded by the total enzymatic mass , considering a similar constraint to that used by FBAwMC and by [12] , [21] ( Methods ) . Enzyme turnover numbers are used to compute an upper bound on enzyme concentrations required to catalyze the corresponding flux rates , and enzyme molecular weights to transform concentrations to units of mass . However , unlike FBAwMC , MOMENT jointly searches for a feasible flux distribution and for the corresponding enzyme concentrations required , considering isozymes , enzymatic complexes , and multi-functional enzymes . This is achieved by making usage of detailed gene-to-reaction mapping that is commonly represented in CBM models via Boolean equations ( Methods ) . For isozymes , the gene-to-reaction mapping denotes that the expression of one of several genes is required to catalyze a certain reaction , while for enzyme complexes , that the expression of several genes is jointly required . Notably , the entire set of gene-to-reaction mapping is formulated as part of the linear programming in a recursive manner , without requiring a more complex optimization such as mixed-integer linear programming that is commonly used to model this mapping [28] , [29] . For reactions in E . coli for which no enzyme turnover numbers were extracted from the above described databases , mean turnover numbers from other species were considered , yielding a total set of turnover numbers for 513 enzymes . Reactions for which no turnover number was available in any species were assigned with the median turnover number across all reactions in E . coli , as in [10] . Using only enzyme turnover numbers measured for E . coli provided lower prediction accuracy for MOMENT as well as for the other computational approaches , still showing a marked advantage in prediction accuracy to MOMENT ( Table S2 ) . An implementation of MOMENT is available via http://www . cs . technion . ac . il/~tomersh/tools/ . To evaluate MOMENT's ability to predict microbial growth rates , we experimentally measured E . coli's growth rates on 24 single carbon and energy source media ( Methods , Table S3 ) and compared the predicted and measured rates . The predictions were obtained by applying MOMENT on the genome-scale metabolic network model of E . coli iAF1260 [19] . We found that growth rate predictions obtained by MOMENT were significantly correlated with the measured ones , with a Pearson correlation of 0 . 468 ( p-value = 0 . 02; Figure 3 and 4 ) , and a Spearman correlation of 0 . 473 ( p-value = 0 . 0196 ) . Notably , varying the threshold on the total enzyme mass linearly scales the predicted growth rates ( and hence , by definition , does not change the above correlations between predicted and measured growth rates; Text S1 ) . Protein mass was previously shown to account for 56% ( g enzymes/gDW ) of cellular mass ( based on experimental measurements [30] ) . Assuming that the entire protein mass is allocated to metabolic enzymes , we initially predicted a mean growth rate of 1 . 02 ( 1/h ) across growth media , which is markedly higher than the mean measured growth rate of 0 . 47 ( 1/h ) . Searching for a threshold on the total enzymatic mass that minimizes the deviation between measured and predicted growth rates ( in terms of square differences ) resulted in a threshold of 27% ( g enzymes/gDW ) ( suggesting that only 48% of protein mass is taken by metabolic enzymes ) . Hence , practically , in order to scale the predicted growth rates to the correct range , prior knowledge on the total mass of metabolic enzymes should be used . In this case , the identified fraction of proteins mass devoted to metabolic enzymes in E . coli is further supported by analyzing gene expression data [26] , which show that the sum of expression level of enzyme-coding genes is 35% of the total sum of expression level of all genes . While the growth rate predictions obtained by MOMENT are significantly correlated with the measured ones , the standard deviation of the predicted rates ( across the different media ) is markedly lower than that of the measured growth rates ( 0 . 054 for predicted rates versus 0 . 14 for measured rates ) . A potential explanation for the differences between the standard deviations in the observed and predicted growth rates could be that the fraction of protein mass devoted to metabolic enzymes increases under high growth rates . Notably , checking this hypothesis would require high-throughput protein concentration data measured under various growth rates , though this kind of data is currently unavailable . Overall , utilizing additional experimental data on the total enzyme concentration in a specific growth condition of interest is expected to further improve MOMENT's predictive performance . To evaluate the importance of the utilized enzyme turnover numbers , we repeated MOMENT's growth rate predictions with randomly shuffled turnover numbers , which were found to provide significantly lower prediction accuracy ( Figure 3; p-value = 0 . 026 , representing the fraction of random samplings , which have led to a higher correlation with the measured growth rates than that achieved with the known turnover numbers ) . To benchmark our new method , we tested the prediction performance of the previously developed FBAwMC . Here , FBAwMC was provided with the very same enzyme turnover rates given to MOMENT , while performing a sampling procedure for missing parameters as described in Beg et al did not improve the predicted performance ( data not shown ) . We found that the growth rate prediction achieved by FBAwMC are not significantly correlated with the measured ones ( Figure 3; Figure S2; p-value = 0 . 17 ) , although a significant correlation between measured and predicted growth rates was reported for a smaller set of 10 media by Beg et al . [8] . Notably , the scope of the 24 media considered here is significantly wider as it includes also nucleotides and amino-acids which were not considered in the set of 10 media studied by Beg et al . When focusing on the growth rate measurements for the limited set of 10 media made by Beg et al , both FBAwMC and MOMENT achieve significant Pearson correlations , though insignificant Spearman correlations ( see Table S4 ) . A previous study by Wong et al . [31] suggests that growth rate is proportional to the square root of growth yield . We find that MOMENT's predictions satisfy this relation , with a significant correlation ( Pearson R = 0 . 3953 , p-value = 0 . 05; Spearman R = 0 . 5046 , p-value = 0 . 01 ) between the predicted growth rate and the square root of the predicted biomass yield . To evaluate the performance of MOMENT in predicting intracellular fluxes , we compared experimental flux measurements for 28 reactions in E . coli measured under exponential growth phase by Schuetz et al . [18] with the predicted fluxes . We found that flux predictions obtained by MOMENT achieve a Pearson correlation of 0 . 76 with the measured fluxes , significantly outperforming FBAwMC and FBA , which achieve correlations of 0 . 64 and 0 . 51 , respectively . As a further control , we tested a variant of FBA , which maximizes ATP yield per sum of flux square , previously shown by Schuetz et al . to improve flux prediction accuracy [18] . We found that predictions obtained by the latter approach achieve a Pearson correlation of 0 . 68 with the measured fluxes , which is still markedly lower than MOMENT's prediction accuracy ( Figure 5A; Figure S3A; Table S5 ) . Also here , the utilization of the randomly sampled enzyme turnover numbers led to worse predictions ( Figure 5A; Table S5 ) . To further evaluate the predictive performance of MOMENT , we extracted data from [32] on gene expression changes in E . coli under glucose minimal media , between low and high growth rate conditions , the latter involving over-flow metabolism , and compared it to predicted changes in enzyme concentrations . Applying MOMENT to predict changes in protein concentrations between these low and high growth rate conditions , we predicted 28 enzymes with a significant change in concentration ( deviating from the expected increase in enzyme levels due to the fold change increase in growth rate ) . We found that changes in these enzyme concentrations predicted by MOMENT between the low and high growth rates achieve a Pearson correlation of 0 . 84 ( p-value = 2 . 4e-8 ) with the measured changes in gene expression ( Figure 5B; Figure S3B; Table S6 ) , even though gene and protein expression levels tend to be only moderately correlated [33] . Also here , the performance of FBAwMC is significantly lower in this case ( Figure 5B; Table S6 ) , with a Pearson correlation of 0 . 34 ( p-value = 2 . 9e-3 ) . Notably , naïve FBA was not evaluated here as it cannot be applied to predict differential metabolism across different growth rates . As a further benchmark , we applied a recently developed method called Parsimonious enzyme usage FBA ( pFBA ) , to classify genes in E . coli according to whether they are used in the optimal growth solutions ( as this classification was previously shown to correlate with changes in gene expression following laboratory-evolved E . coli straints that increased their growth rates ) [34] . We found only a weak correlation between this gene classification and the changes in gene expresion between the low and high growth rates conditions [32] ( Pearson R = 0 . 092 , p-value = 0 . 02; Spearman R = 0 . 074 , p-value = 0 . 06 ) .
Computational prediction of microbial growth rates represents a major challenge . Here , we present a novel computational approach , MOMENT , that addresses this challenge by integrating genome-scale metabolic modeling with enzyme kinetic parameters . MOMENT is shown to predict growth rates for E . coli under various growth media that are significantly correlatred with experimental measurements , and to improve the prediction accuracy of several metabolic phenotypes including intracellular fluxes , and gene expression of enzyme-coding genes . The method is based on an identified design principle of metabolism , in which enzymes catalyzing high flux reactions across different media tend to have higher turnover numbers . While MOMENT enables genome-scale prediction of metabolic phenotypes it is bound to make simplifying assumptions that in some cases may lead to false predictions: ( i ) MOMENT requires as input information on the fraction of total protein concentrations that is devoted to metabolic enzymes . Since this information is difficult to obtain for each modeled condition , here we assumed that this fraction remains constant across a variety of growth media , which is expected to bias the predictions . ( ii ) MOMENT does not take into account several important factors that affect growth rate such as the cost of protein synthesis by ribosomes and local substrate turnover numbers , etc [35] . ( iii ) MOMENT requires data on enzyme kinetic constants , which is still unavailable for hundreds of enzymes in E . coli . Specifically , kinetic data on various membrane transporters is missing from both BRENDA [15] and SABIO-RK [16] , which may lead to false prediction regarding the cost of activating specific transporters and regarding the effect of knocking out transporters . ( iv ) MOMENT does not predict metabolite concentrations and hence does not take into account thermodynamic considerations ( regarding flux directionality ) or enzyme saturation considerations in computing required enzyme levels ( implicitly assuming that the majority of enzymes in E . coli are fully saturated , following [36] ) . Future studies may extend MOMENT to also predict metabolite concentrations , satisfying the 2nd law of thermodynamics as done in [37] , while considering enzyme saturation effects via known enzyme binding affinity constants ( Km ) . From an applicative standpoint , the improved metabolic modeling performance achieved by MOMENT is expected to significantly contribute to metabolic engineering applications and specifically to optimal strain design . Specifically , the additional constraints employed by MOMENT on the requirement for specific enzyme concentrations for catalyzing predicted metabolic flux rates , can be integrated and potentially improve the accuracy of computational metabolic engineering methods such as OptKnock , RobustKnock , OptStrain , etc [38] , [39] , [40] . MOMENT's ability to correctly predict microbial growth rates supports the underlying assumption that a physiological bound on cellular enzyme mass is a key factor that determines growth rate .
Enzyme turnover rates were extracted from BRENDA [15] and SABIO-RK [16] . based on Enzyme Commission ( EC ) numbers and reactant names in the E . coli metabolic model by Feist et al . [19] ( Table S7 ) . Measured turnover rates for mutated enzymes were filtered out . When multiple turnover numbers were available for a certain enzyme , the median value was chosen . Similar to FBA , MOMENT searches for a feasible flux distribution vector v ( mmol/gDW/h ) with maximal growth rate ( i . e . flux through the biomass production reaction ) , satisfying mass-balance and reaction directionality constraints based on the following linear constraints: Where S denotes a stiochiometric matrix S ( NxM ) composed of N metabolites and M reactions ( Sij corresponds to the stoichiometric coefficient of metabolite i in reaction j ) and vlb and vub represent known lower and upper bounds , respectively , on flux rates . Here , vlb is set to either −inf for reversible reactions or 0 for irreversible reactions , and vub is set to +inf for all reactions . In addition to searching for a flux distribution , MOMENT searches for a vector of enzyme concentrations , denoted g ( mmol/gDW ) , such that each flux rate in v has a sufficiently high enzyme concentration to catalyze it . To associate flux rates with enzyme concentrations , we utilize the Boolean gene-to-reaction mapping that is included in the E . coli model of Feist et al . [19] , as follows: This can be formulated in a linear equation by defining an auxiliary variable ga&b that is constrained to be smaller than both ga and gb . To account for more complex gene-to-reaction mappings , where multiple alternative enzyme complexes can catalyze a certain reaction , we applied the above rules recursively by adding auxiliary variables for AND and OR operators . For enzymes whose turnover number is unknown , we use the me median turnover number across all reactions in E . coli ( see Results ) . The enzymes solvent capacity constraint is formulated as;where , MWi denotes the molecular weight of protein coded by gene i , and C denotes the total weight of proteins , which was assumed to be 56% out of the E . coli dry weight mass [30] . Notably , the latter constraint resembles the molecular crowding constraint employed by [9] , [12] , [21] , though here , the gene-to-reactions mapping is taken into account . To maximize ATP yield per sum of flux square , , as performed by Schuetz et al . [18] , requies non-convex optimization . To overcome that , we utilized the same approach suggested by Schuetz at el . [18] , and solved a series of quadratic programming optimization problems of the form:where ε represents a trade-off between ATP maximization and minimization of flux norm . Specifically , we iterate over 10000 values of ε between 0 . 5 to 1 . 5 to identify the optimal flux distribution that maximizes ATP yield per flux unit . | While current genome-scale metabolic modeling approaches enable us to successfully predict a variety of metabolic phenotypes , identifying the factors that determine microbial growth rate and the prediction of growth rates under various conditions is still an open challenge . Here we present a metabolic network-based approach , MetabOlic Modeling with ENzyme kineTics ( MOMENT ) , which predicts growth rates by integrating standard stoichiometric modeling with prior data on enzyme turnover rates and enzyme molecular weights , considering a physiological bound on total enzymes' concentration . The method is based on a finding that enzymes catalyzing high flux reactions tend to be more efficient in terms of having higher turnover numbers . MOMENT predicts growth rates of E . coli across a set of 24 different media that are significantly correlated with experimental measurements , while existing state-of-the art stoichiometric modeling approaches fail to do so . These results suggest that a bound on cellular enzyme concentrations is a key factor that determines microbial growth rate . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biochemistry",
"biology",
"computational",
"biology",
"metabolic",
"networks",
"metabolism"
] | 2012 | Prediction of Microbial Growth Rate versus Biomass Yield by a Metabolic Network with Kinetic Parameters |
The success of Mycobacterium tuberculosis ( Mtb ) as a pathogen rests upon its ability to grow intracellularly in macrophages . Interferon-gamma ( IFN-γ ) is critical in host defense against Mtb and stimulates macrophage clearance of Mtb through an autophagy pathway . Here we show that the host protein ubiquilin 1 ( UBQLN1 ) promotes IFN-γ-mediated autophagic clearance of Mtb . Ubiquilin family members have previously been shown to recognize proteins that aggregate in neurodegenerative disorders . We find that UBQLN1 can interact with Mtb surface proteins and associates with the bacilli in vitro . In IFN-γ activated macrophages , UBQLN1 co-localizes with Mtb and promotes the anti-mycobacterial activity of IFN-γ . The association of UBQLN1 with Mtb depends upon the secreted bacterial protein , EsxA , which is involved in permeabilizing host phagosomes . In autophagy-deficient macrophages , UBQLN1 accumulates around Mtb , consistent with the idea that it marks bacilli that traffic through the autophagy pathway . Moreover , UBQLN1 promotes ubiquitin , p62 , and LC3 accumulation around Mtb , acting independently of the E3 ligase parkin . In summary , we propose a model in which UBQLN1 recognizes Mtb and in turn recruits the autophagy machinery thereby promoting intracellular control of Mtb . Thus , polymorphisms in ubiquilins , which are known to influence susceptibility to neurodegenerative illnesses , might also play a role in host defense against Mtb .
Mycobacterium tuberculosis ( Mtb ) infects one-third of the world’s population . It can remain dormant in its host for decades and ultimately kills more people than any other bacteria . Mtb survives within macrophages by preventing its own delivery to the degradative , phagolysosomal compartment [1] . Macrophages that are activated by IFN-γ partially overcome the arrest in phagosome maturation imposed by Mtb [2 , 3] . IFN-γ stimulates macroautophagy [4–6] ( hereafter autophagy ) , a process by which double-membrane organelles termed autophagosomes capture and degrade cytoplasmic components . In non-selective autophagy , which occurs in response to nutrient limitation , a portion of the cytoplasm is engulfed . In a form of autophagy that is called xenophagy , invading microorganisms are targeted . Autophagosomes that sequester Mtb fuse with lysosomes and impair mycobacterial replication . Autophagy partially restricts Mtb replication , and conditions that activate autophagy , including exposure to IFN-γ , promote mycobacterial clearance [4 , 6–8] . A prevailing model for how autophagy contributes to antimicrobial host defense begins with bacteria damaging or escaping phagosomes [9 , 10] . In the case of Mtb , damage and/or escape depends upon the mycobacterial ESX-1 Type VII secretion system and the secreted effector EsxA ( also known as ESAT-6 ) [11–15] . Phagosomal damage allows mycobacterial DNA and peptidoglycan to activate host cytosolic sensors [14 , 16] . Ubiquitinated proteins accumulate around the bacteria , which partially depends upon the E3 ligase parkin ( encoded by the PARK2 gene ) [17] . The host proteins p62 ( also known as SQSTM1 ) , NDP52 ( nuclear dot protein 52 kD , also known as CALCOCO2 ) , and NBR1 ( next to BRCA1 gene 1 ) can all bind ubiquitin as well as the autophagy protein LC3; they are thought to serve as cargo adaptors that link ubiquitin-conjugated Mtb or phagosomal remnants to LC3 [8 , 9 , 17] . A major unanswered question in the field is how Mtb , or any other bacteria , are physically linked to the autophagy machinery . In mitophagy , an analogous process in which mitochondria are selectively cleared from the cytoplasm , certain outer mitochondrial membrane proteins directly bind LC3 or recruit parkin to damaged mitochondria [18] . How parkin and other E3 ligases are recruited to the invading tubercle bacilli is uncertain; no mechanism comparable to that described for mitophagy has been shown to link the bacterial surface to the autophagy machinery . We found that host UBQLN1 ( also known as PLIC-1 ) binds a subset of Mtb secreted proteins and recognizes Mtb during xenophagy . UBQLN1 is a member of a family of highly related proteins that contain a ubiquitin-like ( UBL ) domain , a ubiquitin-associated domain ( UBA ) , and STI1 motifs that are found in the co-chaperone Sti-1 ( also known as HOP ) ( Fig 1A ) . UBQLN1 and UBQLN2 are thought to facilitate degradation of ubiquitinated targets by the proteasome [19 , 20] . More recently , they have also been shown to play a role in autophagy . Ubiquilins associate with autophagosomes , participate in autophagosome formation , and protect against starvation-induced cell death [21 , 22] . They are implicated in clearing protein aggregates in neurodegenerative disorders , including Alzheimer’s disease , amyotrophic lateral sclerosis ( ALS ) , and Huntington’s disease [23–25] . Here , we show that UBQLN1 recognizes Mtb , acts upstream of ubiquitination , and promotes autophagy-mediated clearance of Mtb . Therefore , we provide evidence that UBQLN1 serves as a link between the bacterial surface and the host autophagy pathway .
Previously , we used a stringent yeast two-hybrid ( Y2H ) system to map Mtb-human protein-protein interactions . UBQLN1 interacted with 12 Mtb proteins , which we call MUPs for mycobacterial ubiquilin-interacting proteins [26] ( S1 Text , S1 Table ) . UBQLN1 exhibited selectivity in its interactions , as we identified it only 12 times in a screen of 339 secreted Mtb proteins . It did not interact with any of the 60 non-secreted Mtb proteins that were also screened or Antigen 85b , EsxA , and EsxB when directly tested [26] . Specificity was also indicated by the finding that MUPs interacted weakly or not at all with NDP52 , an autophagy receptor that , like UBQLN1 , contains a ubiquitin-binding domain ( S1 Fig ) . To evaluate whether MUPs interact with other ubiquilin family members , we examined murine UBQLN1 , UBQLN2 , and UBQLN4 , which are 88% , 67% , and 60% identical to human UBQLN1 , respectively . We did not test UBQLN3 because its expression is restricted to the testes [27] . Most MUPs interacted with murine UBQLN1 ( both splice isoforms ) , UBQLN2 , and UBQLN4 ( Fig 1B ) . The MUPs are largely uncharacterized; there is not a domain common to all of them , although two MUPs contain p60 domains ( S1 Table ) . Most have a predicted signal peptide that targets them for secretion , and the majority have been found in culture filtrate in at least one study , suggesting they are accessible to host interactions . Few MUPs are found exclusively in the culture filtrate; most are also present in the cell membrane or whole cell lysate ( S2 Fig ) [28] . To conclude , we found that ubiquilin family members can interact with numerous Mtb secreted and surface proteins . To determine whether MUPs interact with UBQLN1 in mammalian cells , we expressed V5-tagged MUPs in HEK293 cells . We could detect expression of eight MUPs in HEK293 cells , and we tested four in co-immunoprecipitation assays with human UBQLN1 . We could not detect endogenous UBQLN1 in HEK293 cells , so we co-transfected myc-UBQLN1 along with V5-tagged MUPs . myc-UBQLN1 immunoprecipitated MUPs Rv2911 and Rv3584 , but not Rv1926 . MUPs were not detected in immunoprecipitates from cells lacking myc-UBQLN1 ( Fig 1C ) . Although myc-UBQLN1 did not co-immunoprecipitate Rv1926c-V5 , Rv1926c-V5 did co-immunoprecipitate UBQLN1 when using an antibody recognizing V5 , whereas minimal myc-UBQLN1 was detected when an isotype control antibody was used ( Fig 1D ) . The amount of UBQLN1 that co-immunoprecipitate with Rv1926 was 27% of that which could be precipitated directly using the myc antibody . In the case of Rv1566-V5 and myc-UBQLN1 , we could co-immunoprecipitate the proteins in both directions ( Fig 1E ) , and in both cases 7% of the amount directly precipitated was co-immunoprecipitated . Thus , all of the MUPs tested co-immunoprecipitated in at least one direction with UBQLN1 . Since UBQLN1 can interact with Mtb surface proteins , we reasoned that it might associate with Mtb in a cell free system . We incubated Mtb with cytosol from HEK293 cells that had been transfected with plasmid encoding UBQLN1 , truncated versions of UBQLN1 , or vector control ( Fig 2A , 2B and 2C ) . After four hours of incubation , we washed the bacteria to remove unbound host proteins . Actin was removed after the first wash , whereas UBQLN1 remained associated with Mtb after five washes . As additional controls for specificity , we tested cytosol from HEK293 cells transfected with GFP or the E3 ligase parkin . In contrast to UBQLN1 , neither GFP nor parkin associated with Mtb ( Fig 2B ) . To test if Ubqln1 binds Mtb directly , we purified GST-UBLQN1 and added it to Mtb in vitro . As we found with HEK293 cell lysate , UBQLN1 remained associated with Mtb after five washes ( Fig 2D ) . To further understand how UBQLN1 interacts with Mtb , we examined the roles of the UBL and UBA domains . A recombinant , truncated version of UBQLN1 lacking the UBL domain ( UBQLN1-ΔUBL ) failed to bind Mtb , whereas the UBA domain mutant had preserved binding ( Fig 2C ) . Although recombinant UBQLN1-ΔUBL did not bind Mtb in vitro , when UBQLN1-ΔUBL was expressed in HEK293 cells or yeast , it interacted with intact Mtb as well as MUPs ( Fig 2D and S3 Fig ) . This suggests that in co-immunoprecipitation and Y2H experiments , additional proteins , including UBQLN family members , may bridge or stabilize the UBQLN1-Mtb and MUP interactions . When we expressed UBQLN1 lacking the UBA domain ( UBQLN1-ΔUBA ) in HEK293 cells , there were numerous products both smaller and larger than the predicted protein , suggesting that UBQLN1- ΔUBA was degraded and post-translationally modified . Degradation products associated with Mtb in the cell free system ( Fig 2C ) and also co-immunoprecipitated with MUPs ( Rv1566 and Rv2911; S3 Fig ) . In the Y2H , UBQLN1-ΔUBA failed to interact with MUPs , perhaps related to its propensity for degradation ( S3 Fig ) . The UBA domain alone failed to associate with Mtb ( Fig 2C ) . In conclusion , Mtb associates with recombinant UBQLN1 in vitro and in a cell free system . The UBL domain appears important in vitro , although in the context of cytoplasm , it is dispensable , perhaps related to the ability of endogenous UBQLN family members or other adaptor proteins to multimerize and recruit the truncated protein . Determining the contribution of the UBA domain is confounded by the propensity for the truncated protein to be degraded , but the combined data suggest the UBA domain is not required . UBQLN1 interacted with Mtb proteins and was recruited to the bacteria from host cytosol . Therefore , it should co-localize with the bacteria during an infection . We used a UBQLN1-specific antibody to examine its localization by immunofluorescence microscopy . UBQLN1 was predominantly found in small cytoplasmic punctae ( S4 Fig ) . When we examined the relationship of UBQLN1 to Mtb in unactivated macrophages , we found a low level of co-localization . The association was enhanced in IFN-γ activated macrophages , in which we found UBQLN1 co-localized with 13% of Mtb ( Fig 3A and 3B ) . Although the association of UBQLN1 with Mtb was more prominent in IFN-γ activated macrophages , the protein was present at equivalent levels in untreated macrophages ( Fig 3C ) and the overall cellular distribution of UBQLN1 looked similar in activated and naïve cells ( S4 Fig ) . In IFN-γ activated macrophages , there was little co-localization of UBQLN1 with an Mtb ΔesxA mutant , which does not permeabilize the phagosome ( Fig 3B ) [12 , 14] , suggesting that the association of UBQLN1 with Mtb requires phagosomal damage to provide access to the bacteria . To determine whether UBQLN1 plays a role during infection , we examined macrophages in which UBQLN1 was depleted using RNAi . We used two different siRNAs which reduced UBQLN1 levels by 88% ( siRNA #1 ) and 90% ( siRNA #2; Fig 3D ) based upon quantification using ImageJ and normalizing to actin levels . There was no significant effect on UBQLN2 levels . Following UBQLN1-depletion , bone marrow-derived macrophages ( BMDMs ) were infected with a live/dead Mtb reporter strain that expresses mCherry constitutively and GFP under control of a tetracycline-inducible promoter [29] . After treatment with anhydrotetracycline ( AnTc ) , metabolically active bacteria express both GFP and mCherry , whereas dead bacteria only express mCherry ( Fig 3E ) . As expected , control macrophages that were activated with IFN-γ restrained growth of Mtb . In contrast , UBQLN1-depleted macrophages were impaired in restricting Mtb growth ( Fig 3F ) . We corroborated these results by plating for colony forming units ( CFU ) . UBQLN1 silencing had no effect on bacterial uptake 4 hours post infection ( hpi ) , but it rendered IFN-γ activated macrophages defective in their ability to control Mtb at 96 hpi ( Fig 3G ) . UBQLN1 was also required to control Mtb in IFN-γ activated RAW264 . 7 ( RAW ) cells , a murine macrophage cell line ( S5 Fig ) . In accordance with the localization data , there was no effect of UBQLN1 silencing in unactivated macrophages ( S5 Fig ) or on infections with the ΔesxA mutant ( Fig 3G and S5 Fig ) . To determine whether UBQLN1 controls the general antimicrobial capacity of macrophages , we examined its effect on the intracellular growth of Staphylococcus aureus and Mycobacterium smegmatis . There was no effect of depleting UBQLN1 on S . aureus or M . smegmatis survival ( S6 Fig ) . Thus , although UBQLN1 restricts growth of Mtb , it is not universally required for the antimicrobial capacity of macrophages towards all bacteria . Because MUPs also bound UBQLN2 and UBQLN4 , we examined whether these family members play a role in controlling Mtb replication in macrophages . We found that UBQLN2 , but not UBQLN4 , was present in macrophage lysate ( S4 Fig ) . We attempted to silence UBQLN2 using siRNA , but we only achieved limited depletion , which did not result in any change in mycobacterial CFU ( S4 Fig ) . Thus , UBQLN1 restricts Mtb growth in activated macrophages , and we were unable to draw definitive conclusions about UBQLN2 . Another important function of macrophages is to present antigen presentation to and activate T cells . We examined the ability of UBQLN1-silenced macrophages to activate Th1 polarized CD4+ T cells using P25TCR-Tg T cells which recognize the peptide 25 epitope of Mtb Antigen 85B . T cell production of IFN-γ was antigen specific , as IFN-γ was not detected when macrophages were infected with a strain lacking Antigen 85b ( ΔfbpB ) . Notably , IFN-γ secretion was lower when P25TCR-Tg T cells were co-cultured with UBQLN1-depleted macrophages compared to co-culture with controls ( Fig 3H ) . There was no effect of UBQLN1 silencing on MHCII surface expression ( S7 Fig ) . UBQLN1 silencing did not impair the ability of macrophages infected with ΔesxA to activate T cells ( Fig 3I ) , consistent with the idea that UBQLN1 acts on bacilli that access the cytosol . This also argues that impaired T cell activation is not caused by a non-specific effect such as decreased macrophage viability . In summary , UBQLN1 limits Mtb replication and enhances the ability of macrophages to activate effector T cells . The above findings demonstrate that UBQLN1 recognition of Mtb and its role in controlling Mtb replication depends upon the bacterial ESX-1 system and host IFN-γ activation , both of which promote Mtb xenophagy [4 , 14] . We examined the accumulation of LC3-II by western blotting in uninfected macrophages , and found that UBQLN1 was required for basal autophagy in uninfected , IFN-γ activated BMDMs ( S5 Fig ) . To evaluate whether UBQLN1 is involved in xenophagy , we examined the localization of UBQLN1 positive Mtb relative to autophagy components ( Fig 4A ) . In activated macrophages , we found that more than 60% of the UBQLN1 positive Mtb co-localized with ubiquitinated proteins , which surround mycobacteria that damage or escape the phagosome [8 , 12 , 30] . 45% of the UBQLN1 positive bacteria also co-localized with p62 , an autophagy adaptor that is recruited to ubiquitinated bacteria and binds LC3 . Finally , more than 60% of the UBQLN1 positive bacteria co-localized with LC3 . Likewise , a prominent fraction of the bacteria that associated with FK2 , p62 , or LC3 were UBQLN1 positive ( Fig 4B ) , all of which is consistent with the idea that UBQLN1 functions in autophagy-mediated clearance of Mtb . We next asked whether UBQLN1 positive Mtb are trafficked through the autophagy pathway . Initially , we examined whether IFN-γ-mediated control of Mtb depends upon Atg16L1 , a component of the autophagy elongation complex that conjugates LC3 to phosphatidylethanolamine on the incipient autophagosome . IFN-γ restricted bacterial growth in wild type ( wt ) macrophages ( Atg16L1flox/flox Cre- ) , but not in autophagy-deficient macrophages ( Atg16L1flox/flox Lyz-Cre+ ) ( Fig 4C ) , demonstrating that autophagy is required for the antibacterial properties of IFN-γ . In IFN-γ activated macrophages , there were more p62-positive Mtb in autophagy-deficient cells than controls ( 22 . 8 +/- 2 . 2% versus 11 . 8 +/- 5 . 3%; p<0 . 03 ) , consistent with the idea that p62 associates with Mtb that are directed to the autolysosome where p62 is subsequently degraded . Similarly , IFN-γ-activated , Atg16L1-deficient macrophages contained more UBQLN1 positive Mtb compared to autophagy-competent cells ( 20 . 4 +/- 4 . 7% vs 10 . 5 +/- 2 . 7% , p<0 . 04 ) , suggesting that UBQLN1-decorated Mtb are also targeted for xenophagy . Moreover , the fraction of UBQLN1 positive bacteria that were also positive for p62 increased from ~50% in wt macrophages to ~80% in the autophagy-deficient cells ( Fig 4D and 4E ) , implying that double positive bacteria are eliminated by autophagy . Combined , these data support the idea that UBQLN1 associates with bacteria that are cleared through autophagy . UBQLN1 binds mono and polyubiquitin [31] , so we thought that it might be recruited to ubiquitinated Mtb , like the autophagy adaptors p62 and human NDP52 . Alternatively , since UBQLN1 can bind MUPs , we also thought that UBQLN1 might directly recognize Mtb and act upstream of ubiquitination . To distinguish these possibilities , we examined ubiquitination of Mtb in UBQLN1-silenced BMDMs . As expected , IFN-γ promoted the ubiquitination of Mtb . Notably , the recruitment of ubiquitin to Mtb in activated cells depended upon UBQLN1 . Similarly , the enhanced co-localization of p62 and LC3 with Mtb seen in IFN-γ treated macrophages was blunted in cells lacking UBQLN1 ( Fig 5A , 5B and 5C ) . Thus , we conclude that UBQLN1 promotes Mtb-associated ubiquitination and subsequent recruitment of adaptor proteins and the autophagy machinery during IFN-γ promoted xenophagy . The E3 ligase parkin is required for ubiquitin recruitment to Mtb in naïve macrophages [17] , and we found that it also played a role in IFN-γ activated macrophages . In activated macrophages , the parkin knockout macrophages had half as many ubiquitin positive Mtb as wt macrophages ( Fig 6A ) . We asked whether parkin is also required for UBQLN1 recruitment to Mtb . While there was a slight trend towards decreased UBQLN1 positive Mtb in the parkin mutant , it was not statistically significant ( Fig 6B ) . This suggests that UBQLN1 largely localizes independently of parkin and ubiquitination . In addition , in macrophages lacking parkin we found that UBQLN1 silencing diminished ubiquitin association with Mtb , much as it did in wt macrophages ( Fig 6C ) . Combined , these results suggest that in activated macrophages , UBQLN1 and parkin act independently to recruit ubiquitinated proteins to Mtb .
Although in vitro autophagy makes only a limited contribution to macrophage control of Mtb , in vivo mice are profoundly susceptible to Mtb infection if they lack parkin or autophagy proteins in myeloid cells [8 , 17 , 32] . In addition to controlling bacterial replication , xenophagy modulates cytokine responses and promotes MHCII antigen presentation [32 , 33] . Here , we provide multiple lines of evidence that UBQLN1 associates with Mtb and links them to the autophagy machinery: UBQLN1 binds MUPs , binds Mtb in vitro , and localizes to Mtb during infection . The association of UBQLN1 with Mtb depends upon EsxA , which is likely related to the role of EsxA in damaging the phagosomal membrane; however , it is also possible that EsxA plays some other role in recruiting UBQLN1 to the bacteria . Consistent with a role for UBQNL1 in recognizing Mtb , in the absence of UBQLN1 there is less ubiquitin , p62 , and LC3 recruited to Mtb , which correlates with impaired control of bacterial replication and diminished CD4+ T cell activation . Overall , our data are consistent with a model in which UBQLN1 recognizes Mtb that become accessible to the cytosol upon phagosomal damage; UBQLN1 then assists in recruiting the autophagy machinery . Why did we only detect a role for UBQLN1 in IFN-γ activated macrophages when the protein is also present in unactivated macrophages ? UBQLN1 probably does not function in the IFN-γ signaling pathway , as there was no difference in surface MHCII , an IFN-γ induced gene , between control and UBQLN1-silenced BMDMs ( S7 Fig ) . The apparent selectivity for activated macrophages may simply reflect that we see very low levels xenophagy in unactivated cells , as evidenced by little FK2 , p62 , and LC3 association with Mtb in naïve macrophages ( Fig 5 ) . It may be difficult to detect significant differences when we alter the trafficking of this minor bacterial population , particularly using RNAi , which generates hypomorphic effects rather than complete loss of function . In addition , autolysosomes may be less antimicrobial in naïve cells compared with activated macrophages . Other investigators have reported higher levels of ubiquitin and LC3 association in unactivated macrophages ( for example see [8 , 17] ) , which might be due to strain or other experimental differences . When we activated autophagy chemically with rapamycin , it resulted in considerable toxicity in siRNA-transfected , Mtb-infected cells making it difficult to draw any conclusions about the role of UBQLN1 in this context . However , we suspect UBQLN1 would play a detectable role in naive macrophages if they had robust xenophagy . It is surprising that UBQLN1 recognizes so many Mtb proteins that appear to have little sequence conservation . In addition , there are likely to be additional MUPs as our Y2H screen only examined 399 Mtb proteins [26] . The MUPs do not appear to share much in common ( S1 Table ) , although one thing almost all of the MUPs have is a predicted signal peptide . However , not all proteins containing signal peptides interact with UBQLN1 , since UBQLN1 did not interact with Ag85B or the other proteins with signal sequences in our original Y2H screen [26] . Domain mutants of the MUPs have so far not been revealing , as all fragments have failed to interact . One possibility is that MUPs are prone to misfolding and aggregation . UBQLN1 is proposed to have chaperone activity [34] , which may involve the STI1 motifs . Our findings that UBQLN1 promotes ubiquitination and adaptor recruitment is consistent with it playing an early role in xenophagy; however , it is possible that we identified the MUP-UBQLN1 interaction fortuitously , and MUPs are not what is responsible for recruiting UBQLN1 to Mtb . To verify the UBQLN1-MUP interactions or identify the bona fide UBQLN1 interacting proteins during an infection is technically challenging and an area of ongoing effort . How do we envision UBQLN1 might work to promote xenophagy ? One possibility is that it recruits or activates an E3 ligase , which ubiquitinates bacterial or host phagosomal components . In several studies , UBQLN1 has been shown to promote ubiquitination of target proteins [35–37] , although the E3 ligases were not identified . For Mtb xenophagy , parkin was a strong candidate , since it was recently shown to be required for ubiquitination around Mtb [17] . However , UBQLN1 localized to Mtb and promoted ubiquitination even in the absence of parkin ( Fig 6 ) . In addition , UBQLN1 and parkin did not interact in Y2H or co-immunoprecipitation experiments . Hence , a different E3 ligase likely acts downstream of UBQLN1 and parallel to parkin . Another possibility is that UBQLN1 is part of an amplification loop that fosters the association of ubiquitinated proteins with Mtb; UBQLN1 might localize to Mtb by virtue of binding MUPs , misfolded or aggregated proteins , or ubiquitinated proteins and then recruit additional ubiquitinated proteins . Whether UBQLN2 also plays a role in Mtb xenophagy warrants additional investigation . Although we did not detect a role for UBQLN2 , this may have been due to insufficient silencing . In conclusion , we show that UBQLN1 is required for autophagy-mediated clearance of Mtb in response to IFN-γ . UBQLN1 associates with Mtb and promotes recruitment of ubiquitinated proteins , autophagy adaptors , and the autophagy machinery . We speculate that UBQLN1 recognizes MUPs or other aggregation prone proteins generated by Mtb or present in the phagosome . In doing so , UBQNL1 promotes innate resistance to Mtb in the same way that it protects cells from cytotoxicity due to aggregation-prone cellular proteins , such as APP , TDP-43 , and polyQ-expanded Huntington’s disease protein [24] . Thus , in addition to their role in Alzheimer’s disease , polymorphisms in UBQLN1 may influence susceptibility to tuberculosis , analogous to the dual role of PARK2 ( which encodes parkin ) in Parkinson’s and leprosy [17 , 38] . UBQLN2 mutations , which confer risk of ALS [23] , are also worthy of further investigation . Overexpression of UBQLN1 ameliorates damage in murine models of stroke and Huntington’s disease [39 , 40] . Therefore , therapeutics that promote the activity of ubiquilins might have efficacy in neurodegenerative disorders and tuberculosis .
For WB , cellular lysates were prepared in phosphate buffered saline ( PBS ) with 1% NP-40 and Halt Protease Inhibitor Cocktail ( Thermo Scientific ) . For immunoprecipitations ( IPs ) , HEK293 cells were lysed in PBS with 0 . 1% NP-40 and passed 25 times through a 25 gauge needle . Lysates were incubated with Sepharose G agarose beads ( GE Healthcare ) pre-bound with anti-Ubqln1 ( Abcam ) or anti-myc antibody ( Fig 1C ) . Alternatively , Sepharose G Dynabeads coated with anti-myc , anti-V5 , or control IgG antibodies were used ( Fig 1D and 1E ) . Bound proteins were analyzed by WB . RAW cells or BMDMs were transfected with siRNAs for 2d ( RAW ) or 3d ( BMDM ) prior to infection . 200 U/ml murine IFN-γ ( Gibco ) was added 24h before infection as indicated . For Mtb and M . smegmatis , macrophages were infected with a single cell suspension at an MOI of 3 with at least three replicates per experiment as previously described [26] . 4 hpi macrophages were extensively washed , lysed with 0 . 1% Triton X-100 at indicated time points , and serial dilutions were plated on 7H11 . CFU were counted 15–21 days later for Mtb and 2–3 days later for M . smegmatis . For S . aureus infection , bacteria were opsonized with human serum for 1h prior to infection . Macrophages were infected at an MOI of 1 , washed extensively 30 min post-infection , and lysed in 0 . 1% Triton-X-100 at indicated time points . S . aureus were plated on Tryptic Soy Agar , and CFU were quantified the following day . RAW cells or BMDMs from C57BL/6 , parkin KO , or LC3-GFP-expressing mice were plated in 8 well chamber slides . They were infected with a single cell suspension of Mtb expressing GFP , DsRed , or the live/dead plasmid at MOI of 5 followed by washing 4 hpi . At indicated time points , they were fixed in 1% paraformaldehyde ( PFA ) /PBS overnight . For live/dead analysis , 200 nM anhydrotetracycline ( AnTc ) was added 20–24 hours prior to fixation . % viable Mtb was calculated using the live/dead strain as a ratio of GFP-bright , metabolically active bacteria to total mCherry-positive bacteria . For immunofluorescence microscopy , macrophages were permeabilized with 0 . 1% Tween-20 prior to immunostaining with primary and corresponding secondary antibody . Images were acquired using the Nikon Eclipse TiE/B fluorescent microscope at 60x magnification and deconvoluted as previously described [26] . At least 100 bacteria from a minimum of three independent fields were examined per experiment . For reproduced images , in some cases background was subtracted by selecting an ROI ( region of interest ) where there were no cells . For reproduced images , contrast was altered equally for a given single channel image for all samples in an experiment . For example , the signal corresponding to UBLQN1 was contrast adjusted equally for all panels in Fig 3A and likewise for all panels in Fig 4A . In Fig 5 , the Ub , p62 , and LC3 channels are adjusted equivalently for the siCON panel and the siUBQNL1 panel . Similarly , the UBQLN1 and p62 channels were equally adjusted for the Cre+ and Cre- samples . 1 x 106 HEK293 were plated , transfected the following day , and lysed 2 days later in PBS with 0 . 1% NP-40 ( lysis buffer ) by passage through a 25 gauge needle 25 times in the same manner as co-immunoprecipitation experiments . Mtb were grown to between O . D . = 0 . 5 and 1 , washed twice with PBS , and one O . D . of bacteria was mixed with HEK293 lysate and incubated at 4°C for 4 h . Bacteria were then pelleted , resuspended in 1 mL lysis buffer , transferred to a new tube , and washed 5 times with 1 mL lysis buffer . The resulting bacterial pellet was suspended in 100 μl lysis buffer , transferred to a new tube with SDS loading buffer , boiled , and analyzed by WB . This study was conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The New York University School of Medicine Institutional Animal Care and Use Committee approved all work with mice ( protocol #130707–02 ) . Euthanasia was performed prior to bone marrow harvest in accordance with the 2013 AVMA Guidelines for the Euthanasia of Animals . | More people die from Mycobacterium tuberculosis ( Mtb ) , the causative agent of tuberculosis ( TB ) , than any other bacterial pathogen . It has long been appreciated that Mtb can survive and divide within macrophages , white blood cells that normally kill bacteria . Macrophages are able to partially control Mtb through a degradative process called autophagy . Autophagy is activated by the cytokine interferon-gamma ( IFN-γ ) , which promotes control of Mtb infection . How the tubercle bacilli are targeted to the autophagy pathway remains unclear . Here we show that the human protein ubiquilin 1 can interact with Mtb surface proteins and associate with Mtb that are present in the host cell cytosol . We propose a model in which activating autophagy with IFN-γ promotes UBQLN1 recruitment to Mtb , which in turn leads to recruitment of the autophagy machinery , autophagy-mediated degradation of the bacteria , and activation of effector T cells . Since IFN-γ is critical in human control of Mtb , our study suggests that polymorphisms in ubiquilins , known to influence susceptibility to neurodegenerative illnesses , might also play a role in host defense against Mtb . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Ubiquilin 1 Promotes IFN-γ-Induced Xenophagy of Mycobacterium tuberculosis |
The burden of scrub typhus in endemic areas is poorly understood . This study aimed at estimating the proportion of hospitalisations and outpatient visits for undifferentiated fever in the community that may be attributable to scrub typhus . The study was a retrospective cohort with a nested case-control study conducted in the South Indian state of Tamil Nadu . We conducted house-to-house screening in 48 villages ( 42965 people , 11964 households ) to identify hospitalised or outpatient cases due to undifferentiated fever during the preceding scrub typhus season . We used scrub typhus IgG to determine past infection . We calculated adjusted odds ratios for the association between IgG positivity and case status . Odds ratios were used to estimate population attributable fractions ( PAF ) indicating the proportion of hospitalised and outpatient fever cases attributable to scrub typhus . We identified 58 cases of hospitalisation and 236 outpatient treatments . 562 people were enrolled as control group to estimate the background IgG sero-prevalence . IgG prevalence was 20 . 3% in controls , 26 . 3% in outpatient cases and 43 . 1% in hospitalised cases . The PAFs suggested that 29 . 5% of hospitalisations and 6 . 1% of outpatient cases may have been due to scrub typhus . In villages with a high IgG prevalence ( defined as ≥15% among controls ) , the corresponding PAFs were 43 . 4% for hospitalisations and 5 . 6% for outpatients . The estimated annual incidence of scrub typhus was 0 . 8/1000 people ( 0 . 3/1000 in low , and 1 . 3/1000 in high prevalence villages ) . Evidence for recall error suggested that the true incidences may be about twice as high as these figures . The study suggests scrub typhus as an important cause for febrile hospitalisations in the community . The results confirm the adequacy of empirical treatment for scrub typhus in hospitalised cases with undifferentiated fever . Since scrub typhus may be rare among stable outpatients , the use of empirical treatment remains doubtful in these .
Scrub typhus is a febrile illness caused by Orientia tsutsugamushi , a bacterial species belonging to the genus Orientia ( family Rickettsiaceae ) [1] . The infection is transmitted by the larvae ( chiggers ) of trombiculid mites which infect mammals as incidental hosts [2] . Scrub typhus occurs over much of tropical and subtropical East Asia , South Asia and South-east Asia . Populous countries like India , China , Bangladesh , Pakistan , Indonesia , Vietnam and Japan are endemic [3] . The disease has recently been identified in Chile [4] and possibly East Africa [5] . One million annual symptomatic infections have been estimated to occur globally but this figure is uncertain [6] . In many endemic areas , scrub typhus accounts for 15% to 30% of febrile illness leading to health care utilisation [7 , 8 , 9] . Scrub typhus is associated with significant mortality , estimated at 6% to 10% of untreated cases [2 , 10] . Mortality in complicated cases remains substantial despite treatment [11] . Mortality for patients with acute respiratory distress syndrome ( ARDS ) , the most common complication , may be up to 25% [11] . Other common complications include meningo-encephalitis , shock and renal failure [9 , 11] . Complications may be avoided by early administration of antibiotics such as doxycycline , chloramphenicol or azithromycin [12] . Adverse pregnancy outcomes resulting in stillbirth , prematurity and low birthweight may occur in 40% of pregnant women with scrub typhus [13 , 14] . The global burden of scrub typhus has been explored based on cross sectional serological surveys with no clear link to clinical disease [15] , or studies relying on passive case detection [16] . The vast majority of studies on scrub typhus are hospital-based , single institution studies , and there is a paucity of population-based epidemiological data . The study was conducted as a pilot study to estimate the proportion of undifferentiated fever cases leading to health care use that may be attributable to scrub typhus , and to obtain an approximate estimate of scrub typhus incidence in the community . The intention was to explore the feasibility of a large prospective cohort study .
The study was conducted in rural villages in Vellore District , Tamil Nadu ( India ) . The district is characterised by mainly agricultural villages , with rice , sugarcane , sorghum , coconut , pulses and turmeric as major crops . Animal husbandry mainly includes cattle , goats and poultry . The climate is tropical savannah . Monsoon rains occur between June and December . The scrub typhus season approximately lasts from July to February with a peak from November to January . Scrub typhus occurs sporadically between seasons [17] . The study was conceived as a retrospective cohort with a nested case-control study . We conducted house-to-house screening for unspecified febrile illnesses occurring during the preceding scrub typhus season ( 2017 / 2018 ) and used serology ( scrub typhus IgM and IgG ) to determine past infection with scrub typhus . A control group was enrolled through systematic sampling to estimate the background IgM and IgG seroprevalence . The study was conducted between 23rd of March and 30th of June 2018 . For this pilot study we wished to target villages likely to be endemic for scrub typhus . We used two different approaches to select villages . We first enrolled villages in Vellore district based on hospital admission records at the Christian Medical College Vellore ( CMC ) . Villages were enrolled if at least two scrub typhus cases from that village were admitted to CMC during the three scrub typhus seasons from June 2014 to February 2017 , excluding the current season under study ( June 2017 to February 2018 ) . We restricted enrolment to villages reachable within one hour from the hospital , resulting in 29 villages . However , because field workers were unable to identify sufficiently many households and cases in these villages , and due to logistical constraints , we then switched to the second approach . We identified a rural area close to Vellore city defined by the Palar River in the south , the limits of Vellore city in the east and the border to Andhra Pradesh in the North , where several villages had met the original eligibility criteria . Assuming this area to be endemic for scrub typhus as a whole we systematically enrolled villages not previously enrolled moving from east to west until the intended number of over 11000 households were screened . This resulted in a further 19 villages . Field staff identified the approximate boundaries of a village using satellite images , and then attempt to cover the whole village through house-to-house enquiry . Houses without anyone present were left out , and not revisited . We used the following eligibility criteria for enrolment of a case of undifferentiated febrile illness: 1 ) aged 12 years or older , 2 ) hospitalised for febrile illness , or visited an outpatient department , local clinic or pharmacy due to febrile illness at any time between June 2017 and March 2018 , 3 ) cause for febrile illness not known , or described ( by respondent or in available health records ) as scrub typhus , malaria , dengue , typhoid , meningitis or pneumonia , 4 ) absence of leg infection , 5 ) no operation was done at the hospital , 6 ) no other surgical cause for fever identified from patients memory or available health records , 7 ) absence of urinary tract infection ( only used to exclude cases if urine culture positive ) , 8 ) duration of fever of at least 2 days or duration of fever not known , 9 ) the fever occurred while residing in the study village and health care was sought at a health centre in the district . Hospitalisation was defined as staying at least for one night . All other health care uses were treated as outpatient / pharmacy visit . If a hospitalised case was not present at the time of the interview , we made an appointment with the participants for blood sampling and questionnaire administration . Absent cases meeting the enrolment criteria for outpatients were not revisited due to logistical constraints . Case and control households were geo-referenced using hand-held GPS receivers . We enrolled controls through systematic sampling during house-to-house screening , by contacting household members of every 20th house during the walk . Controls were eligible if they had not sought health care due to febrile illness between June 2017 and March 2018 , and were living in the study area during that time . Because of concerns that field workers would predominantly enrol older people and females who were deemed more likely to be present , we used a stratified enrolment procedure , using four strata: females ≥50 years old , females <50 years old , males ≥50 years old , males <50 years old . Field workers enrolled controls in blocks of four , with each stratum being represented once . The aim of this procedure was to obtain a reasonably age and sex balanced control group without requiring a formal sampling frame . Controls were asked to give a blood sample and were asked whether they had had any high grade fever not leading to health care use between June 2017 and March 2018 . After collection , blood samples were brought to CMC . Serum was separated from blood cells , divided into 3 aliquots and stored at -70°C until testing . We used enzyme-linked immunosorbent assays ( ELISA ) to detect IgG and IgM antibodies to scrub typhus ( Scrub Typhus Detect , InBios International , Inc . , Seattle , WA , USA ) following the manufacturer’s specifications . This ELISA uses Karp , Kato , Gilliam and TA716 recombinant proteins of the 56-kD outer membrane protein . Commercially available ELISA IgM assays , such as the ones used in this study , have been shown to have a sensitivity and specificity of over 90% in a study from Thailand [18] , and 80% sensitivity and 96% specificity in a study from South India [19] . The sensitivity and specificity of IgG ELISA to detect past infection remains to be determined . The distributions of the optical densities for IgG and IgM are shown in Fig 1 . Given the pronounced bimodal shape of the distribution for IgG , we chose the apparent midpoint of 1 . 5 as the OD cut-off point for this study . For IgM , the choice for the cut-off point was made difficult by the unusual application of the test , which has been developed to diagnose acute infection , not past infection . IgM is known to wane during the months following scrub typhus infection , although precise data on the timing of the decline have not been published . The OD cut-off point of 1 . 0 currently used at CMC for acute scrub typhus infection would have been inappropriately conservative . To avoid lack of specificity , we determined a suitable OD cut-off post-hoc so that IgM prevalence in controls was below 5% . Given this ad hoc approach to determine a cut-off we decided to use the IgM results only in the sensitivity analysis , not for the primary analysis . The sample size calculation was based on a preparatory study done by medical interns ( CMC , Vellore ) to test the questionnaire in three villages not included in the present study . In this study , 11 hospitalisations of unspecified febrile illness over the approximate length of a scrub typhus season were identified with a denominator of 2271 people ( cumulative incidence of 4 . 8 per 1000 people aged 12 years or older ) . We assumed treatment as outpatient or at a local pharmacy to be 3 times more common ( about 15 cases per 1000 ) . We anticipated a population attributable fraction ( PAF ) of scrub typhus of 25% of hospitalisations , i . e . an incidence of hospitalisation due to scrub typhus of 1 . 2/1000 people and aimed at identifying at least 25 cases of scrub typhus hospitalisation ( out of 100 total cases of hospitalisation ) . Assuming 30% of households refusing to participate , 30% of cases not being remembered , and a household size of 3 . 6 ( aged 12 years or older ) , resulted in 11800 households to be screened for active case detection . The 95% confidence interval for the sero-prevalence of scrub typhus IgG among controls was calculated using linearized standard errors ( svy: proportion command in STATA 12 . 0 , Statacorp ) . Continuous variables were compared across groups using the t-test . Differences in IgG sero-prevalence by sex were tested using the chi-square test . Adjusted differences in IgG sero-prevalence by sex were calculated using binomial regression ( identity link function ) . The intra-class correlation coefficient for village clustering of IgG prevalence was calculated using the loneway command in STATA 12 . 0 . The spatial correlogram for IgG prevalence in controls was estimated using the ncf package in R ( R project ) . We used 100m increments . P values for Moran’s I were estimated based on 500 resampling rounds . The sero-prevalence of IgM and IgG was compared between cases and controls using logistic regression . Odds ratios were adjusted for age and sex . We used robust standard errors to estimate 95% confidence intervals accounting for the village-level sampling approach . Interactions between sex and IgG sero-positivity , and age and IgG sero-positivity were explored using likelihood ratio tests . The PAF is commonly defined as the proportion of cases ( here: hospitalisation or outpatient / pharmacy treatment ) that could be prevented if an exposure ( here: scrub typhus infection ) were removed from the population . Missing data were ignored in the analysis . We calculated the PAF based on the following formula [20]: PAF=p×AOR-1AOR where p is the prevalence of scrub typhus IgG or IgM positivity in cases , and AOR the age/sex adjusted odds ratio of IgG or IgM sero-prevalence between cases and controls . Note that the formula is only meaningful for odds ratios of 1 or greater . The incidence of scrub typhus was estimated using the number of cases as nominator and the number of screened individuals as denominator , multiplied by the PAF . The study was approved by CMC’s Institutional Review Board and LSHTM’s Research Ethics Committee . Written consent was obtained from all adult participants . Written or verbal assent was obtained from minors , alongside written consent from their parents/guardians .
The overall IgG sero-prevalence among controls was 20 . 3% ( 95% CI 14 . 0% to 26 . 5% ) . Among controls there was a near-linear relationship between age and IgG sero-prevalence ( Fig 2 ) . Female controls were younger than male controls ( 43 . 9 years vs 46 . 7 years , p = 0 . 03 ) . Sero-prevalence was slightly higher among control females than males ( 21 . 5% vs 18 . 6% , p = 0 . 411 ) . This difference of 2 . 9% increased to 3 . 6% after adjusting for age ( binomial regression , identity link ) . The IgG sero-prevalence among controls varied between villages . Seventeen villages had 0% sero-prevalence while 10 villages had a sero-prevalence of 40% or higher . The variation was higher than expected by chance , indicated by an intra-class correlation coefficient of 0 . 16 ( 95% CI 0 . 07 to 0 . 25 ) . The spatial correlogram ( Fig 3 ) suggests spatial auto-correlation of IgG sero-positivity largely occurring within 300m . The prevalence of scrub typhus IgG sero-positivity was 20% in controls , 26% in outpatients and 43% in hospital cases . Among IgG sero-positive individuals ( OD≥ 1 . 5 ) the mean OD was 2 . 8 for controls ( range 1 . 6–3 . 4 , SD 0 . 5 ) , 2 . 7 for outpatients ( range 1 . 5–3 . 2 , SD 0 . 5 ) and 2 . 8 for hospital cases ( range 1 . 6–3 . 2 , SD 0 . 5 ) . Similar to the total study population ( Fig 1 ) , all three groups displayed a pronounced bimodal pattern of OD values ( S1 Fig ) . The proportion of cases attributable to scrub typhus ( calculated based on IgG ) was generally much larger for hospitalised cases than for outpatient / pharmacy cases ( Table 2 ) . The PAFs for hospitalisation exceeded 40% in women , participants aged 36 to 50 years and those living in villages with an IgG sero-prevalence of 15% or higher . The PAF for hospitalisation was only about 10% in men and negligible for participants aged 35 years or younger . In logistic regression , the test for interaction between sex and IgG positivity as predictors of hospitalisation showed a p value of p = 0 . 046 , indicating support for women having higher PAFs for scrub typhus than men . The p value for the test for interaction between age 35 years or under/above 35 and IgG positivity was p = 0 . 024 , suggesting hospitalisations in older ages may be more likely to be due to scrub typhus than in younger ages . There were trends for IgG positive hospital cases having a shorter average duration of fever ( 8 . 0 vs 9 . 3 days , p = 0 . 41 ) and a shorter duration of hospital stay ( 4 . 5 vs 6 . 1 days , p = 0 . 35 ) than IgG negative hospital cases . However , due to small case numbers and large variations in fever duration and hospital stay , the p values indicated low statistical support for these differences . The cumulative incidence of hospitalisation and outpatient / pharmacy treatment over the 2017 / 2018 season is shown in Table 3 . Because of the high potential for under-reporting especially of outpatient / pharmacy visits , these data should be interpreted with caution . However , they suggest a higher incidence of scrub typhus in villages with an IgG sero-prevalence of 15% or higher . The higher crude incidence of hospital admission in high prevalence villages ( 1 . 7 vs 1 . 2 per 1000 ) may be explained by scrub typhus leading to an excess of hospitalisations compared to low prevalence villages . The large difference in outpatient / pharmacy visits between high and low prevalence villages ( 9 . 5 vs 3 . 7 per 1000 ) is unlikely to be due to scrub typhus given the small PAFs for scrub typhus in outpatients ( Table 3 ) . Focussing on the overall , unstratified analysis ( Table 2 , top ) , we tested different OD values to explore how different cut-offs for IgG OD affected the PAF estimates . As shown in Table 4 , the IgG sero-prevalence values and the PAFs for outpatient/pharmacy and hospital cases were fairly robust to varying the OD cut-off between 1 . 2 and 2 . 0 . By contrast , changing the OD cut-off for IgM substantially affected the IgM sero-prevalence estimates . An OD cut-off of 0 . 2 maximised the difference between hospitalised cases and controls , probably at the cost of low specificity . We used an OD of 0 . 4 as the default cut-off for the IgM to explore the robustness of the PAFs calculated in Table 2 for different strata to using IgM instead of IgG prevalence . As shown in Table 5 , the PAFs are lower compared to the PAFs estimated based on IgG , but reflect the higher risk in women , in those aged 35 years or older and those living in high prevalence villages .
The study suggests that scrub typhus accounts for a substantial proportion of hospitalisations due to undifferentiated fever in this South Indian rural setting . By contrast , scrub typhus may not be a common cause of fever among cases managed as outpatients , even in areas with a high IgG sero-prevalence . The overall estimate of about 30% of undifferentiated fever hospitalisations being attributable to scrub typhus may be realistic: first , we found a strong relationship between village-level sero-prevalence and PAF estimates , which lends some support to the assumption that scrub typhus may indeed be the cause of many of these hospitalisations . Second , the proportion is similar to the proportion of fever cases admitted to CMC that are due to scrub typhus ( 35 . 9% ) [9] , even though this comparison is made difficult by CMC being a tertiary care centre receiving patients from throughout the region , often with complicated infection . Third , the results confirm the observation by local medical staff that the proportion of scrub typhus cases requiring admission appears to be high compared to other undifferentiated fevers such as dengue or respiratory viral infections . The results further suggested a higher PAF of hospitalisations due to scrub typhus in people aged 35 and older , and in women . Due to small case numbers , the stratified analyses need to be treated with caution . Given the substantial PAFs of hospitalisations due to scrub typhus , the findings support the presumptive treatment with doxycycline of hospitalised fever cases . The scope for presumptive treatment of stable fever patients managed as outpatients seems more doubtful , as only a small proportion of these cases may be due to scrub typhus , even in settings where the sero-prevalence exceeds 15% . This confirms the role for rapid diagnostic tests or risk scoring based on clinical findings [21] particularly in outpatients . Our estimate of the annual scrub typhus incidence in the study communities ( between 0 . 3/1000 in low prevalence areas and 1 . 3/1000 in high prevalence areas ) is difficult to compare with other studies as these ( being hospital-based ) usually lacked a clearly defined denominator . A Malaysian study from the 1970s which relied on passive case finding but appears to include many mild cases while lacking a well-defined denominator , suggested an annual incidence of 12 per 1000 . This is substantially higher than our estimate , even when accounting for under-reporting in our study . It could be due to the particularly high risk in the selected study area ( communities working at palm oil plantations ) . The study is limited primarily by the indirect way of estimating case numbers which , not being a prospective design , could not be based on individual case confirmation . Because of the high background IgG positivity in controls , the PAF approach does not allow classifying an individual as a scrub typhus case or not . Estimation of case numbers only refers to the population under study as a whole . The concept of the PAF has been developed mainly for non-infectious diseases to determine the proportion of cases that could be prevented if a given risk factor is completely eliminated from the population . The PAF is not ideally suited to study the aetiology of common conditions of presumed infectious origin such as diarrhoea , respiratory illness or fever , although it is often used in this context , ( e . g . Kotloff and colleagues [22] and Smith and colleagues [23] ) . The PAFs calculated here assume that pre-existing antibodies do not affect the risk of subsequent scrub typhus infection . They would overestimate the proportion of cases due to scrub typhus if IgG was protective . While it is known that one individual can be infected with scrub typhus repeatedly even within two seasons [24] , partial protection is a possibility . On the other hand , cases with positive IgG antibodies could represent a subgroup of the population at particular risk of scrub typhus . In this subgroup , more cases could be due to scrub typhus in the current season than suggested by the PAF , which does not account for the possibility of repeat infection . In this scenario , the PAF may underestimate the proportion of cases due to scrub typhus . The study is further limited by recall error , the possibility of cross reactivity ( see below ) , uncertainty in the choice of cut-off points for the diagnostic tests , and the non-random enrolment of villages into the study . We found evidence that recall error reduced the reported incidence of hospitalisation and outpatient care use . The pilot study suggested a three times higher incidence of hospitalisation due to undifferentiated fever , which may have been due to medical interns being more able to elicit a history of febrile illnesses occurring several months ago than the nurses employed for the main study . It could therefore be argued that the true incidence of scrub typhus in the community may be twice or three times as high as estimated in this study . Following this logic , plausible figures for the annual incidence could be 1 . 5 to 2 . 5 per 1000 people overall , and between 2 . 5 and 4 per 1000 people in high prevalence villages . It seems unlikely that the true incidence of scrub typhus in this setting will be much higher than that . Most respondents were able to remember past episodes of health care use after engaging them in a conversation to build trust and explain the study purpose , but nurses found it difficult to do this while screening a large number of households per day . The potential for recall error to affect the PAF estimates , which do not rely on all cases in the community being identified , is smaller than for incidence estimates . Nevertheless , it is possible that the cases of hospitalisation and outpatient treatment identified in this study are not a random subset of the total cases . Fever duration and hospital stay were somewhat shorter for IgG positive compared to IgG negative cases ( p-values indicated that these differences could have been by chance ) . It seems unlikely for recall error to be strongly associated with IgG positivity , especially not in a way overestimating the PAF for hospitalisations due to scrub typhus . ELISA tests for scrub typhus antibodies are thought to cross-react with a range of infections such as leptospirosis , tick-borne spotted fever , murine typhus and others . Including a control group largely controls for antibodies circulating in the general population independent of febrile illness but may be subject to misclassification of cases of febrile illness . For example , if a substantial proportion of febrile illness cases enrolled in this study were due to leptospirosis or spotted fever , and cross-reactivity was substantial , then the PAFs calculated here would overestimate the proportion of fevers due to scrub typhus . Leptospirosis and spotted fevers are probably too rare in the study area to affect the findings [9] . Murine typhus and spotted fever group rickettsiosis antibodies may be particularly prone to cross-react with scrub typhus antigens as these are caused by related organisms . A study from an urban setting in Laos jointly examined the prevalence of positive scrub typhus and murine typhus IgG antibodies ( based on ELISA tests ) in the general population demonstrating different geographic risk factors for the two infections [25] . In this study of 2002 people , 314 were found to be positive only for scrub typhus antibodies , 360 positive only for murine typhus antibodies , while 80 were positive for both antibodies . The expected number of dual positives assuming independence between the two infections would be 86 cases—very similar to the observed 80 . If cross reactivity were substantial , one would expect a higher proportion of participants to be positive for both infections . This finding suggests that cross-reactivity , while of clinical relevance especially in acute cases , may be of lesser importance in serological surveys using IgG . However , co-infection and cross-reactivity pattern may differ between Laos and India . Large prospective studies in India are clearly needed to further explore the issue of cross-reactivity . In the absence of a generally agreed method to determine suitable cut-off points for OD values from ELISA tests , the cut-off points in this study were chosen based on the data collected . However , the sensitivity analysis of different IgG OD cut-off points broadly confirmed the main findings . The results were also broadly confirmed when using IgM even at the chosen cut-off point of an OD of 0 . 4 . The PAFs for using IgM were lower than for IgG probably because IgM declines faster than IgG . When using an IgM OD cut-off of 1 . 0 currently applied at CMC to confirm active cases , then only 0 . 5% of controls and 3 . 5% of hospitalised cases would have been IgM positive ( Table 4 ) . The low IgM prevalence in cases and controls when using a conventional cut-off in this study suggest a rapid decline within a few weeks or months after infection . If a blood sample can be collected within weeks of a febrile illness , then IgM ELISAs should be a suitable tool for retrospective case identification in large cohort studies . The longevity of IgG and IgM antibodies is still under debate [26 , 27] . The simplest explanation for the strong increase of IgG prevalence with age may be that IgG antibodies remain positive for many years after infection . This is supported by a similar mean , standard deviation and range of OD values between sero-positive hospital cases , outpatient cases and controls . Alternatively , older people may be at particular risk of scrub typhus . Older people may be more exposed to infectious mite larvae due to behavioural factors , which in our view is not very likely . They may also be at higher risk of infection compared to younger people when exposed to infectious mite larvae because of age-related changes in skin anatomy , physiology and immunology . The ELISA tests used in this study contain the recombinant antigens of Karp , Kato , Gilliam and TA716 . Due to the great antigenic diversity of Orientia tsutsugamushi strains [28] these diagnostic assays may not detect scrub typhus-specific antibodies of all strains . In a study from South India , 16 out of 240 nested PCR confirmed scrub typhus cases were negative for serological scrub typhus tests [19] . These 16 samples were taken predominantly in the first seven days of illness during which serological tests may often be negative . These findings suggest that non-recognition of antibodies by the IgG ELISA tests applied to convalescent samples in the present study is unlikely to strongly affect estimates . While the enrolment of cases and controls in villages may approximately have represented a near-random selection process , the enrolment of villages did not follow a random or equivalent systematic sampling procedure . The study aim was to identify endemic villages and approximately measure the incidence in these for the purposes of informing the design of a larger study . It may be inappropriate to refer to this study as population-based in the strict sense . But even if villages had been selected at random , the study highlights the difficulties in defining a suitable target population for inference when studying infections occurring in distinct foci . As done in this analysis , the most suitable approach may be to define the target population based on quite easily obtainable sero-prevalence data , an approach that has been found useful to classify areas with respect to transmission intensity of dengue fever [29] . We therefore believe that our results , when stratified by sero-prevalence , can be generalised to other villages in South India with similar sero-prevalence levels . Whether or not the association between sero-prevalence and scrub typhus incidence applies to other scrub typhus endemic areas in Asia needs to be confirmed . The high burden of scrub typhus generally observed in Vellore district may indicate a particularly high risk in this area or greater awareness of the infection and better availability of scrub typhus diagnostics . In hospital-based studies conducted in other parts of India such as Puducherry , Goa and Chandigarh , scrub typhus was responsible for 24% to 41% of cases with undifferentiated fever [30–32] . These findings suggest that scrub typhus risk our study area may not be unusually high . This calls for sero-prevalence studies and perhaps similar low-cost nested case-control studies as this one in the wider region to understand the burden of scrub typhus . Being one of the first attempts to measure the incidence of scrub typhus in the community not relying on passive case finding , we believe the results despite their limitations give an approximate estimate of the true burden of scrub typhus in endemic settings . Future studies using a similar retrospective design should employ a shorter recall period ( e . g . a second survey round in the middle of the season ) and thorough training and supervision of field staff to reduce recall error . Our findings can more reliably be confirmed by conducting an adequately sized prospective cohort study , which we think is feasible . There seems little doubt that scrub typhus is an important infection from the public health perspective , deserving the allocation of larger research funds than are currently made available . | Scrub typhus is an important cause of fever in many Asian countries , including China , India , Vietnam and Japan . It is caused by the intra-cellular bacteria Orientia tsutsugamushi and is transmitted to humans by mite larvae ( chiggers ) which attach to the skin of the host . Scrub typhus is potentially life-threatening but treatable with relatively cheap antibiotics such as doxycycline . The incidence of scrub typhus in the community is not known as most studies were done at hospitals and health posts lacking a clearly defined source-population for patients seeking health care for fever . In this study , the proportion of fever cases due to scrub typhus and the incidence of the infection were estimated by retrospective case identification in a population of 42000 people living in 48 villages in the South Indian state of Tamil Nadu . We found that about 30% to 40% of hospitalisations for undifferentiated fever may be due to scrub typhus . In contrast , scrub typhus accounted for only about 5% of outpatient fever cases . Our findings confirm the need for large population-based cohort studies to better understand the epidemiology of scrub typhus in endemic settings . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"typhus",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"immunology",
"social",
"sciences",
"pulmonology",
"social",
"geography",
"health",
"care",
"bacterial",
"diseases",
"signs",
"and",
"symptoms",
"patients",
"immunologic",
"techniques",
"antibodies",
"human",
"geography",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"geography",
"proteins",
"immunoassays",
"hospitals",
"scrub",
"typhus",
"biochemistry",
"outpatients",
"diagnostic",
"medicine",
"health",
"care",
"facilities",
"fevers",
"physiology",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences"
] | 2019 | Hospitalisations and outpatient visits for undifferentiated fever attributable to scrub typhus in rural South India: Retrospective cohort and nested case-control study |
Studies in diverse organisms have revealed a surprising depth to the evolutionary conservation of genetic modules . For example , a systematic analysis of such conserved modules has recently shown that genes in yeast that maintain cell walls have been repurposed in vertebrates to regulate vein and artery growth . We reasoned that by analyzing this particular module , we might identify small molecules targeting the yeast pathway that also act as angiogenesis inhibitors suitable for chemotherapy . This insight led to the finding that thiabendazole , an orally available antifungal drug in clinical use for 40 years , also potently inhibits angiogenesis in animal models and in human cells . Moreover , in vivo time-lapse imaging revealed that thiabendazole reversibly disassembles newly established blood vessels , marking it as vascular disrupting agent ( VDA ) and thus as a potential complementary therapeutic for use in combination with current anti-angiogenic therapies . Importantly , we also show that thiabendazole slows tumor growth and decreases vascular density in preclinical fibrosarcoma xenografts . Thus , an exploration of the evolutionary repurposing of gene networks has led directly to the identification of a potential new therapeutic application for an inexpensive drug that is already approved for clinical use in humans .
Systems biology has shown great promise in providing a better understanding of human disease and in identifying new disease targets . These methods typically leave off once the target is identified , and further research transitions to established paradigms for drug discovery . However , the vast majority of molecular pathways that function in human disease are not specific to humans , but rather are conserved across vertebrates and even to very distantly related organisms . The remarkable growth of genetic data from tractable model organisms implies that most genetic modules relevant to human biology are currently best characterized in non-human species . Such evolutionary conservation , even when the homology of the systems to the human case is distant or perhaps non-obvious , should enable new drug design strategies . Clearly , identification of deeply conserved gene networks in distant organisms opens the possibility of pursuing drug discovery in those organisms . While traditional methods of drug discovery focus on gene-by-gene rather than network- or system-level similarities , we suggest that phenologs—gene networks that while orthologous may nonetheless produce different phenotypes due to altered usage or organismal contexts [1]—can provide a basis not just for screening against a single protein , but also for simultaneous drug discovery efforts against multiple targets in parallel . Given the key roles that model organisms already play in biomedical research , identification of such deep homologies should also allow us to better leverage the particular strengths of the wide variety of animal models in order to rapidly test candidate drugs found from such an approach . We recently developed a method for systematically discovering phenologs , and this approach identified a conserved module that is relevant to lovastatin sensitivity in yeast and is also responsible for regulating angiogenesis in vertebrates [1] . Angiogenesis , the process of forming new blood vessels , plays an essential role in development , reproduction , and tissue repair [2] . Because the vascular network supplies oxygen and nutrients to cancer cells as well as to normal cells , angiogenesis also governs the growth of many types of tumors , and is central to malignancy [2]–[5] . The vasculature is thus considered to be a major therapeutic target for drug development . Some cancers , such as the most common and deadly brain neoplasm , glioblastoma multiformae [6] , are heavily vascularized , but have not responded to current angiogenesis inhibitors [7] , [8] . Because new agents that target the vasculature would increase our arsenal for battling cancers resistant to current therapies [2]–[5] , there is a clear clinical need for novel approaches to their identification . Here , we have exploited data mining of genetic interactions in yeast , in vivo time-lapse imaging in a non-mammalian vertebrate , loss-of-function analysis in cultured human cells , and preclinical xenografts in mice to identify and characterize a novel anti-angiogenic small molecule ( Figure 1 ) . Excitingly , this compound is already FDA approved for use in treating certain infections in humans , making it an excellent candidate for rapid translation to the clinic . This research exemplifies a general strategy for exploiting deeply conserved genetic modules for drug screening , characterized by screens focused not on single genes but rather on conserved genetic modules and by a strong reliance on tractable model organisms in order to speed the discovery of therapeutics .
The remarkable conservation of a genetic module that controls lovastatin sensitivity in yeast and angiogenesis in vertebrates ( [1]; Figures 2AB and S1; Table S1 ) led us to test the possibility that small-molecule inhibitors modulating the yeast pathway might also act as angiogenesis inhibitors . Indeed , preliminary evidence suggests that lovastatin itself at least partly inhibits angiogenesis [9] , [10] and may even reduce the incidence of melanoma [11] , [12] . We therefore devised a strategy to exploit the evolutionary repurposing of this module in order to direct our search ( Figure 1 ) . Specifically , we desired to identify compounds in a manner that did not require their mechanism of action or even their biochemical target to match that of lovastatin; we thus employed a genetic strategy in yeast in order to select compounds that genetically interacted with this module . By computationally mining available large-scale chemical sensitivity datasets [13] , candidate compounds were prioritized based upon their measured synthetic genetic interactions with yeast genes , using clustering algorithms to identify those compounds with genetic interaction profiles most similar to that of lovastatin ( Figure 2C; Table S2; Figure S2 ) . Notably , four out of eight prioritized chemicals were already known to modulate angiogenesis , indicating strong enrichment for angiogenesis effectors ( Table S2 ) . One compound—thiabendazole ( TBZ; 4- ( 1H-1 , 3-benzodiazol-2-yl ) -1 , 3-thiazole ) —stood out because it has already been approved by the U . S . Food and Drug Administration ( FDA ) for systemic oral use in humans ( as an anti-fungal and anti-helminthic treatment ) . TBZ was initially marketed by Merck as Mintezol , and is now off-patent and issued as a generic under the trade names Apl-Luster , Mertect , Mycozol , Tecto , Tresaderm , and Arbotect . TBZ has been used by humans since its FDA approval in 1967 , so its safety has been well-established . In animals , TBZ has no carcinogenic effects in either short- or long-term studies at doses up to 15 times the usual human dose [14] , [15] . Moreover , TBZ does not appear to affect fertility in mice or rats , and it is not a mutagen in standard in vitro microbial mutagen tests , micronucleus tests , or host-mediated assays in vivo [14] , [15] . Thus , TBZ was an outstanding candidate for further study . We first tested the effect of TBZ on the expression of vascular-specific genes in developing Xenopus embryos , which provide a rapid , tractable , and accurate model for in vivo studies of angiogenesis [16]–[19] . Using in situ hybridization to either the apelin-receptor ( aplnr ) or the vascular ETS factor ( erg ) , we found that TBZ treatment severely impaired angiogenesis ( Figure 3A–D ) . This result was confirmed in living embryos in which vasculature was visualized by expression of GFP under control of a kdr enhancer/promoter fragment ( Figure 3E–F ) [19] . Notably , TBZ also inhibited angiogenesis in a dose-dependent manner in cultured human endothelial cells ( HUVECs ) , suggesting that the activity of TBZ is conserved in vertebrates ( Figure 4 ) . We then sought to position the site of TBZ action relative to that of VEGF , as this growth factor is central to both normal and pathogenic angiogenesis [3] , [4] . In frog embryos , ectopic VEGF potently induces ectopic angiogenesis [16] , and this effect was blocked by TBZ , suggesting that the drug acts downstream of this key regulatory node ( Figure S3 ) . These data implicate TBZ as an effective inhibitor of angiogenesis . Importantly , we observed angiogenesis inhibition in both human cells in vitro and in Xenopus embryos in vivo at a concentration of 100–250 µM . This dose corresponds to 20–50 mg/kg ( Figures 3 and 4 ) , which is notable because the oral LD50 of MINTEZOL is 1 . 3–3 . 6 g/kg , 3 . 1 g/kg , and 3 . 8 g/kg in the mouse , rat , and rabbit , respectively , and the human approved recommended maximum daily dose is 3 grams , corresponding to 50 mg/kg for 60 kg patients . Finally , we note that the overall morphology and patterning of TBZ-treated Xenopus embryos was grossly normal at the stages when the vasculature was severely disrupted ( Figure S4 ) . Consistent with this , TBZ has good safety data in humans and model animals at the doses for which we observe a specific inhibition of angiogenesis [14] , [15] . We next asked if angiogenesis inhibition may be a general property of benzimidazoles . Examination of commercially available TBZ derivatives showed that this is not the case , with benzimidazole itself inactive at doses up to 1 mM and administration of other benzimidazoles causing diverse developmental defects but not angiogenesis inhibition ( Figure S5 ) . These findings are thus significant for demonstrating a high level of precision for this evolutionary approach to drug discovery . We next sought to better understand the cellular basis for angiogenesis inhibition by TBZ . In the course of our studies , we noted an interesting feature of the vasculature in TBZ-treated embryos: disconnected and scattered arrays of cells in which vascular gene expression persisted ( Figures 3B , D and S3 ) . Hypothesizing that such morphological defects in the absence of changes to vascular gene expression may stem from direct impairment of vessel integrity , we tested the ability of TBZ to disrupt pre-existing vasculature by treatments at later stages , when blood vessels were already well formed and patent [20] . TBZ treatment elicited overt breakdown of established vasculature at these stages ( Figure S6 ) . The ability of TBZ to disassemble extant blood vessels was especially significant because such an activity has recently drawn the attention of cancer biologists [4] , [21] , [22] . A new class of drugs called Vascular Disrupting Agents ( VDAs ) break down existing vascular structures , thereby disrupting blood flow , particularly within solid tumors [4] , [21] , [22] . No VDAs have as yet been approved for use in humans , although several such agents are therapeutically promising and are in phase II and III trials [4] . As a direct test of the vascular disrupting activity of TBZ , we performed time-lapse imaging of developing vasculature . Using Kdr-GFP transgenic embryos [19] and time-lapse confocal microscopy [23] , we could effectively image developing vasculature in vivo for periods of up to 20 h . During this time , the growth of existing vessels and the sprouting of new vasculature could be easily followed ( Figure S7; Movie S1 ) . Treatment with TBZ completely prevented growth and sprouting of vessels , and moreover elicited a striking disintegration of established vessels after ∼90 min of exposure ( Figures 5 and S8; Movie S2 ) . Upon longer exposures , endothelial cells scattered and many underwent dramatic rounding ( Figures 5A and S8; Movie S2 ) . These data demonstrate the efficacy of TBZ as a vascular disrupting agent . Previously defined VDAs can act either by targeting endothelial cells for selective cell death ( e . g . , ASA404 [24] ) or by disrupting endothelial cell behaviors ( e . g . , combrestatin A4 [25] ) , and so we sought to distinguish between these two possible mechanisms for TBZ action . We noted that treatment with TBZ doses sufficient to severely perturb the vasculature elicited only modest increases in apoptosis in cultured HUVECs ( Figure S9 ) . Moreover , vascular gene expression in dispersed , rounded kdr-GFP+ endothelial cells in vivo reliably persisted for up to 17 h after TBZ treatment ( Figures 3F , 5 , and S8 ) . These data argue against a role for apoptosis in vascular disruption by TBZ . To test this idea more directly , we performed washout experiments . Compellingly , washout of the drug after overt TBZ-induced endothelial cell dispersal and rounding resulted in significant re-spreading of endothelial cells and re-formation of vessels in living Xenopus embryos assessed by time-lapse imaging ( Figure 6; Movie S3 ) . In several cases , widely separated kdr-GFP-positive endothelial cells reconnected into nascent vessels after washout of TBZ ( Figure 6 ) . Finally , we found that treatment with TBZ significantly slowed endothelial cell migration in a scratch wound assay using cultured HUVECs ( Figure 7AB ) . This quantitative in vitro assay with mammalian cells , combined with our in vivo data from Xenopus , demonstrate that TBZ disrupts established vasculature not by eliciting cell death but rather by perturbing endothelial cell behavior . The effect of TBZ on endothelial cells is striking and rapid . Our in vivo imaging of the vasculature revealed that endothelial cells retract from one another and round up within 2 h of TBZ treatment ( Figures 5 and 6 ) . Moreover , we observed that this effect is reversible by washout within a similarly rapid time frame ( Figure 6 ) . The rapid time-frames observed here argue that TBZ may act at the level of the cytoskeleton to influence endothelial cell behavior . We first considered that , while not an assumption of the phenolog approach ( see above ) , TBZ may nonetheless impact the vasculature by the same mechanism as lovastatin . Lovastatin disrupts angiogenesis at least in part by perturbing the geranyl-geranylation of the RhoA GTPase , thereby abrogating its activity [10] . RhoA is a critical regulator of actin-based behaviors in all animal cells [26] , and the loss of RhoA signaling in endothelial cells treated with lovastatin is directly linked to cytoskeletal changes and inhibition of angiogenesis [10] . Indeed , inhibition of angiogenesis by lovastatin can be overcome by addition of geranyl-geranyl pyrophosphate ( GGPP; [9] , [10] ) . We therefore used the HUVEC scratch-wound closure model to quantitatively assess the effects of GGPP addition on TBZ action . However , we found that addition of GGPP did not reverse the action of TBZ on HUVEC cell motility in this assay ( Figure S10 ) . Similarly , while TBZ has been observed to affect the activity of porcine heart mitochondria [27] , we detected no differences in mitochondrial mass ( measured by MitoTracker Green signal ) or mitochondrial membrane potential ( measured as the ratio of MitoTracker Red signal to Mitotracker Green signal ) ( unpublished data ) , thus ruling out this potential activity as being relevant . We next considered the possibility that TBZ acted on the vasculature at the level of the microtubule ( MT ) cytoskeleton , because TBZ has been found to disrupt microtubule assembly and dynamics in a number of cell types ( e . g . , [28]–[31] ) , and because several currently-studied VDAs act as MT-disrupting agents [21] , [32] . Curiously , TBZ had only a very slight effect on the gross organization of the MT cytoskeleton in HUVEC cells in culture ( Figure S11A ) , but a quantitative analysis using mass-spectrometry revealed a significant reduction in the abundance of several tubulin proteins following treatment of HUVECs with TBZ ( Figure S11B ) . Many MT-targeting VDAs act via hyper-activation of Rho signaling [25] , [33] , [34] , likely reflecting the key role of MT-binding RhoGEFs [35] . We reasoned , therefore , that TBZ may also act via increased Rho signaling , as the drug elicited several phenotypes known to be associated with dysregulated Rho signaling ( e . g . , cell rounding , re-distribution of actin filaments , and defects in cell motility; Figure 7 ) . To test this model directly , we asked if disruption of Rho signaling might counteract the effects of TBZ . Indeed , pharmacological disruption of Rho kinase function using the small molecule Y27632 elicited a significant and dose-dependent rescue of the TBZ-induced HUVEC cell motility defect ( Figure 7A , B ) . Together , these data suggest that vascular disruption by TBZ results from reduced tubulin levels and hyper-active Rho signaling . It is hoped that VDAs may open new therapeutic avenues by complementing the action of currently used angiogenesis inhibitors ( e . g . , [4] ) . Moreover , the data above suggest that the mechanism of TBZ action distinguishes it from VDAs such as ASA404 , which act by inducing endothelial cell apoptosis [24] , but which failed to show efficacy in a recent Phase III clinical trial for treatment of lung cancer [36] . To begin to ask if TBZ may be useful in the arena of cancer therapy , we tested the ability of TBZ to slow the growth of solid vascularized tumors in a mammal . We therefore employed a mouse xenograft model typical of those proven valuable in indicating the effectiveness of anti-angiogenesis therapy [37] , [38] . We found that TBZ treatment significantly slowed HT1080 human fibrosarcoma xenograft growth in athymic Cre nu/nu mice [39] , as assessed by a time course of tumor size and also by final tumor mass ( Figure 8 ) . Our in vivo data from Xenopus , as well as our human in vitro data , suggest that TBZ likely slows tumor growth by acting at the level of the vasculature ( Figures 3 and 4 ) . Consistent with this model , TBZ treatment did not alter the rate of proliferation in HT1080 cells when cultured in vitro but did significantly impair tumor microvessel density in xenografts ( Figures 9 and S12 ) . In addition , we noted that treatment with TBZ did not alter the levels of VEGF expressed or secreted by HT1080 cells , consistent with it acting downstream of VEGF in tumor xenografts ( Figure S13 ) , as it does in developing Xenopus embryos in vivo ( Figure S3 ) . Notably , we employed a TBZ dose of 50 mg/kg for these experiments , which is concordant with the FDA-approved maximum recommended daily dose of TBZ in humans , suggesting the possibility of chemotherapeutic use in humans . In sum , an analysis of evolutionary repurposing of a genetic module shared from yeast to humans has led directly to the discovery that an orally available drug , thiabendazole , already FDA approved for clinical use in humans , also acts as an angiogenesis inhibitor and vascular disrupting agent . Moreover , these data establish TBZ as the only VDA currently approved for human use ( albeit for a different purpose ) . Our data suggest that , even for antifungal or antihelmintic use , the possibility of side effects related to vascularization should be considered , for example , in patients with cardiovascular disease or to the fetus if administered to pregnant women , for whom TBZ has not been broadly tested . Significantly , while research on VDAs has largely centered on cancer therapy , their use may also provide new therapeutic avenues for non-malignant diseases , such as diabetic retinopathy and macular degeneration [40] , [41] . With more than 40 years of human use , the low cost and generic availability of TBZ make it a compelling candidate for translation into the clinic as a VDA . Finally , this research emphasizes the advantages of an evolutionary approach to drug discovery , in which the natural experimental strengths of various organisms can be exploited to accelerate our understanding of a conserved genetic module . Importantly , this approach proceeded from a gene module-based discovery strategy and proved effective even though the associated organismal phenotypes were entirely unrelated . Curiously , at least two other known antifungal drugs can also act as angiogenesis inhibitors . Itraconazole , an azole antifungal drug otherwise structurally unrelated to TBZ and acting via different mechanisms , was identified as an angiogenesis inhibitor via high-throughput screening [42] . This observation , in addition to the dual anti-fungal and anti-angiogenic properties of lovastatin [9] , [10] , [43] , suggests additional interesting evolutionary connections between the processes of yeast cell wall metabolism and vertebrate angiogenesis . Such evolutionary connections further support the yeast cell wall-relevant activity of TBZ , from among its multiple pharmacological targets , as being the relevant activity for angiogenesis . Thus there is no reason to suspect that a more highly targeted agent could not be repurposed in a similar fashion . Overall , these results suggest that a fundamental understanding of systems biology will prove to be directly relevant to drug discovery , complementing traditional screening approaches to pharmacophore discovery and accelerating both basic and clinical biomedical research .
Compound genetic interaction profiles were downloaded from http://chemogenomics . stanford . edu:16080/supplements/global/download . html . We employed the p values reported for fitness defects in the yeast homozygous deletion collection for all analyses [13] . Candidate angiogenesis inhibitors were prioritized that consistently clustered with lovastatin across different choices of similarity measures and hierarchical clustering algorithms , specifically centered and uncentered correlation , Spearman rank correlation , absolute correlation ( centered and uncentered ) , Euclidean distance , and City-block distance , employing centroid linkage , complete linkage , single linkage , or average linkage clustering . Clustering results were visualized with Cluster 3 . 0 ( http://bonsai . hgc . jp/~mdehoon/software/cluster/software . htm ) and Java TreeView ( http://jtreeview . sourceforge . net/ ) . Female adult Xenopus were ovulated by injections of human chorionic gonadotropin , and eggs were fertilized in vitro and dejellied in 3% cysteine ( pH 7 . 9 ) and subsequently reared in 1/3× Marc's modified Ringer's ( MMR ) solution . For microinjections , embryos were placed in a solution of 2% Ficoll in 1/3× MMR solution , injected using forceps and an Oxford universal manipulator , reared in 2% Ficoll in 1/3× MMR to stage 9 , then washed and reared in 1/3× MMR solution alone . For bilateral rab11b knock-down experiments , the posterior cardinal vein and intersomitic veins were targeted by injecting Morpholino antisense oligonucleotides ( MOs ) into the two ventral cells equatorially at the four-cell stage . For unilateral knockdown , only one ventral cell was injected . MOs were injected at 40 ng per blastomere . For the experiments to see the drug effects , embryos were placed in a solution of each chemical dissolved in 1% DMSO diluted in 1/3× MMR during indicated stages . For bead micro-surgery implantation , Affi-Gel Blue Gel beads ( Bio-Rad ) were soaked with 0 . 7 mg/ml recombinant mouse VEGF 164 aa ( R&D systems ) or BSA as a control . Whole-mount in situ hybridization for erg and aplnr was performed as described [44] . Erg and aplnr cDNAs were obtained from Open BioSystems ( erg: IMAGE:5512670 , aplnr: IMAGE:8321886 ) . Translation-blocking antisense morpholinos for rab11b were designed based on the sequences from the National Center for Biotechnology Information database ( accession number: BC082421 . 1 ) . MOs were obtained from Gene Tools with the following sequence: 5′-CGTATTCGTCATCTCTGGCTCCCAT-3′ . Human umbilical vein endothelial cells ( HUVECs ) were purchased from Clonetics , and were used between passages 4 and 9 . HUVECs were cultured on 0 . 1% gelatin-coated ( Sigma ) plates in endothelial growth medium-2 ( EGM-2; Clonetics ) in tissue culture flasks at 37°C in a humidified atmosphere of 5% CO2 . HUVECs ( 104 cells ) were seeded in a 96-well plate coated with 50 µl of ECMatrix ( Chemicon ) or Matrigel ( BD Bioscience ) according to the manufacturer's instructions . Cells were incubated for 16 h on EGM-2 containing thiabendazole , dissolved in 1% DMSO . Negative control cells were treated with 1% DMSO in the same manner . As a positive control , siRNA versus the human HoxA9 sequence [45] was transfected into HUVECs using Lipofectamine RNAiMAX ( Invitrogen ) according to the manufacturer's instructions . Tube formation was observed using an inverted microscope ( Nikon , eclipse TS100 ) , and branch points were measured using ImageJ software ( http://rsb . info . nih . gov/ij ) . HUVECs ( 1 . 2×105 cells ) were seeded into 24-well plates for 24 h , and the monolayers were wounded identically . Then , cells were washed with PBS and treated with EBM-2 containing 1% DMSO or 250 µM TBZ dissolved in 1% DMSO with a combination of Y27632 or GGPP ( Sigma ) . In the case of Y27632 treatment , cells were preincubated for 2 h before wounding . Cells were photographed at time zero and after 15 h , and the ratios of cell free area [ ( 0 h–15 h ) /0 h] were calculated . Specific pathogen-free athymic Cre nu/nu mice were purchased from Charles River Laboratories . The HT1080 human fibrosarcoma cell line was obtained from the American Type Culture Collection ( ATCC ) . HT1080 cells were cultured in DMEM ( Gibco ) containing 10% fetal bovine serum ( FBS , Gibco ) in tissue culture flasks at 37°C in a humidified atmosphere of 5% CO2 . In order to generate a mouse xenograft model , a suspension of the HT1080 cells ( 3×106 in 50 µl PBS ) mixed with an equal volume of Matrigel ( BD Bioscience ) was subcutaneously implanted into the flank region of 7–8-wk-old female mice . Upon establishment of tumors ( approx . 40 mm3 ) , mice were given daily intraperitoneal injections of 1 mg thiabendazole ( Sigma-Aldrich ) , suspended in 20 µl DMSO . Mice weighed on average 20 grams; this dose thus corresponded to 250 µM TBZ . As a control , an equal volume of DMSO was injected in the same manner . Tumor growth was monitored by measuring the length and width of each tumor using digital calipers , and the tumor volume in mm3 calculated by the formula: Volume = ( width ) 2×length/2 . Upon a tumor reaching the maximum size permitted by the Institutional Animal Care and Use Committee ( 1 . 5 cm in diameter ) , the mouse was sacrificed , and the tumor excised . Each tumor was fixed with 4% paraformaldehyde in PBS , and cryostat sections were processed . After blocking with 5% goat serum in PBST ( 0 . 3% Triton X-100 in PBS ) for 1 h at room temperature , sectioned tissues were incubated with anti-mouse CD31 antibody , hamster clone 2H8 , 1∶100 ( Millipore ) . After several PBST washes , samples were incubated for 2 h at room temperature with FITC-conjugated anti-hamster IgG antibody , 1∶1 , 000 ( Jackson ImmunoResearch ) . In order to determine the effect of thiabendazole on proliferation and apoptosis , 2×105 HUVECs or HT1080 were cultured in 6-well plates and treated with thiabendazole dissolved in 1% DMSO . Control cells received 1% DMSO . For actin and tubulin cytoskeleton analysis , 7×104 HUVECs were seeded . After 24 h , cells were fixed using 4% paraformaldehyde in PBS . Cell membranes were permeabilized with 0 . 2% Triton X-100 in PBS , and nonspecific immunobinding sites were blocked with 5% goat serum for 1 h at room temperature . Cells were incubated with primary antibodies to Caspase-3 ( Abcam ) , Phospho-histone H3 ( Ser10; Millipore ) , or β-tubulin ( Sigma ) at 4°C overnight . After washing with PBST , primary antibodies were detected by Alexa Fluor-488 or 555 goat anti-rabbit immunoglobulin ( IgG ) . Alexa Fluor 488 phalloidin ( Invitrogen ) and/or 4′ , 6-Diamidino-2-phenylindole ( Sigma ) were added as needed . Immunostaining for Xenopus was performed as previously described [46] . Embryos at stage 35–36 were fixed in 1× MEMFA . 12/101 ( 1∶500; DSHB ) and primary antibodies were detected with Alexa Fluor-488 or 555 goat anti-mouse Immunoglobulin ( IgG ) . Immunohistochemistry experiments and kdr:GFP transgenic Xenopus laevis were imaged on an inverted Zeiss LSM5 Pascal confocal microscope and Zeiss 5-LIVE Fast Scanning confocal microscope . Confocal images were processed and cropped in Imaris software ( BITPLANE ) and Adobe Illustrator and Adobe Photoshop for compilation of figures . HUVECs were treated with 1% DMSO or 1% DMSO , 250 µM TBZ for 24 h , and lysed by Dounce homogenization in low salt buffer ( 10 mM Tris-HCl , pH 8 . 8 , 10 mM KCl , 1 . 5 mM MgCl2 ) with 0 . 5 mM DTT and protease inhibitor mixture ( Calbiochem ) . 2 , 2 , 2-trifluoroethanol was added to 50% ( v/v ) for each sample , and samples were reduced with 15 mM DTT at 55°C for 45 min and then alkylated with 55 mM iodoacetamide at room temperature for 30 min . Following alkylation , samples were diluted in digestion buffer ( 50 mM Tris-HCl , pH 8 . 0 , 2 mM CaCl2 ) to a final 2 , 2 , 2-trifluoroethanol concentration of 5% ( v/v ) and digested using proteomics grade trypsin ( Sigma ) at 1∶50 ( enzyme/protein ) concentration and incubated at 37°C for 4–5 h . Digestion was halted with the addition of 1% formic acid ( v/v ) , and sample volume was reduced to 200 µl by SpeedVac centrifugation prior to loading on HyperSep C-18 SpinTips ( Thermo ) . Samples were eluted ( 60% acetonitrile , 0 . 1% formic acid ) , reduced to 10 µl by SpeedVac centrifugation , and resuspended in sample buffer ( 5% acetonitrile , 0 . 1% formic acid ) . Tryptic peptides were then filtered through Microcon 10-kDa centrifugal filters ( Millipore ) , and collected as flow-through . Peptides were chromatographically separated on a Zorbax reverse-phase C-18 column ( Agilent ) via a 230 min 5%–38% acetonitrile gradient , then analyzed by on-line nanoelectrospray-ionization tandem mass spectrometry on an LTQ-Orbitrap ( Thermo Scientific ) . Data-dependent ion selection was performed , collecting parent ion ( MS1 ) scans at high resolution ( 60 , 000 ) and selecting ions with charge >+1 for collision-induced dissociation fragmentation spectrum acquisition ( MS2 ) in the LTQ , with a maximum of 12 MS2 scans per MS1 . Ions selected more than twice in a 30 s window were dynamically excluded for 45 s . MS2 spectra were interpreted using SEQUEST ( Proteome Discoverer 1 . 3 , Thermo Scientific ) , searching against human protein-coding sequences from Ensembl release 64 [47] . Search results were then processed by Percolator [48] at a 1% false discovery rate . Protein groups were generated comprising proteins with identical peptide evidence , omitting those proteins whose observed peptides could be entirely accounted for by other proteins with additional unique observations . Differential expression of proteins across TBZ-treated and control samples was quantified from the MS2 spectral count data using the APEX method of relative quantification [49] . HT1080 ( 2×105 or 4×104cells ) were cultured in 6-well plates and treated with 1% DMSO or 1% DMSO , 250 µM TBZ for 24 h . Cells were lysed in cell lysis buffer ( Cell Signaling Technology ) containing 1 mM PMSF , and analyzed by SDS-PAGE and Western blotting using anti-VEGF ( Santa Cruz , A-20 ) or anti-GAPDH ( Cell Signaling Technology ) antibodies . The secreted VEGF level in culture medium was determined by enzyme-linked immunosorbent assay ( ELISA; R&D ) according to the manufacturer's instructions . | Yeast cells and vertebrate blood vessels would not seem to have much in common . However , we have discovered that during the course of evolution , a group of proteins whose function in yeast is to maintain cell walls has found an alternative use in vertebrates regulating angiogenesis . This remarkable repurposing of the proteins during evolution led us to hypothesize that , despite the different functions of the proteins in humans compared to yeast , drugs that modulated the yeast pathway might also modulate angiogenesis in humans and in animal models . One compound seemed a particularly promising candidate for this sort of approach: thiabendazole ( TBZ ) , which has been in clinical use as a systemic antifungal and deworming treatment for 40 years . Gratifyingly , our study shows that TBZ is indeed able to act as a vascular disrupting agent and an angiogenesis inhibitor . Notably , TBZ also slowed tumor growth and decreased vascular density in human tumors grafted into mice . TBZ’s historical safety data and low cost make it an outstanding candidate for translation to clinical use as a complement to current anti-angiogenic strategies for the treatment of cancer . Our work demonstrates how model organisms from distant branches of the evolutionary tree can be exploited to arrive at a promising new drug . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"xenopus",
"laevis",
"microbiology",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"morphogenesis",
"biology",
"mouse",
"evolutionary",
"genetics",
"drug",
"discovery",
"systems",
"biology",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"evolutionary",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Evolutionarily Repurposed Networks Reveal the Well-Known Antifungal Drug Thiabendazole to Be a Novel Vascular Disrupting Agent |
The pseudohyphal growth response is a dramatic morphological transition and presumed foraging mechanism wherein yeast cells form invasive and surface-spread multicellular filaments . Pseudohyphal growth has been studied extensively as a model of conserved signaling pathways controlling stress responses , cell morphogenesis , and fungal virulence in pathogenic fungi . The genetic contribution to pseudohyphal growth is extensive , with at least 500 genes required for filamentation; as such , pseudohyphal growth is a complex trait , and linkage analysis is a classical means to dissect the genetic basis of a complex phenotype . Here , we implemented linkage analysis by crossing each of two filamentous strains of Saccharomyces cerevisiae ( Σ1278b and SK1 ) with an S288C-derived non-filamentous strain . We then assayed meiotic progeny for filamentation and mapped allelic linkage in pooled segregants by whole-genome sequencing . This analysis identified linkage in a cohort of genes , including the negative regulator SFL1 , which we find contains a premature stop codon in the invasive SK1 background . The S288C allele of the polarity gene PEA2 , encoding Leu409 rather than Met , is linked with non-invasion . In Σ1278b , the pea2-M409L mutation results in decreased invasive filamentation and elongation , diminished activity of a Kss1p MAPK pathway reporter , decreased unipolar budding , and diminished binding of the polarisome protein Spa2p . Variation between SK1 and S288C in the mitochondrial inner membrane protein Mdm32p at residues 182 and 262 impacts invasive growth and mitochondrial network structure . Collectively , this work identifies new determinants of pseudohyphal growth , while highlighting the coevolution of protein complexes and organelle structures within a given genome in specifying complex phenotypes .
The budding yeast Saccharomyces cerevisiae undergoes a pronounced growth transition in response to nitrogen limitation or glucose limitation , forming multicellular pseudohyphal filaments that can spread outward from a colony and/or invade the surface of a solid growth substrate [1] , [2] . Yeast pseudohyphal filament formation is a presumed foraging mechanism , accomplished through underlying changes in cell adhesion , cell cycle progression , and budding [1] , [3] , [4] . During pseudohyphal growth , yeast cells remain physically connected after cytokinesis via mechanisms encompassing the regulated expression and shedding of the flocculin Flo11p [5]–[7] . Cells undergoing pseudohyphal growth exhibit increased apical growth through reorganization of the actin cytoskeleton , regulation of polarity proteins , and delayed G2/M progression [8]–[12] . The molecular basis of yeast pseudohyphal growth has been studied extensively as a model of conserved signaling pathways controlling cell morphogenesis and polarity . Furthermore , related processes of filamentous development in the principal opportunistic human fungal pathogen Candida albicans are required for virulence , and signaling pathways between the related yeasts are conserved [13] . Classic studies of pseudohyphal growth in S . cerevisiae have resulted most prominently in the identification of core pseudohyphal growth signaling modules encompassing the Kss1p mitogen-activated protein kinase ( MAPK ) cascade , the cAMP-dependent protein kinase A ( PKA ) pathway , and the AMP-activated protein kinase ortholog Snf1p [14]–[20] . The pseudohyphal growth MAPK cascade encompasses Ste11p , Ste7p , and the MAPK Kss1p [10] , [14] . Kss1p phosphorylates the Ste12p transcription factor , resulting in dissociation of the negative regulatory Dig1p and Dig2p interactors and binding of a Ste12p-Tec1p heterodimer to target promoters , such as the FLO11 promoter [21]–[23] . Tpk2p , a catalytic subunit of PKA , phosphorylates the Flo8p transcription factor , promoting Flo8p binding and transcriptional activation at the FLO11 promoter and other regulatory sites [17] , [24]–[26] . In response to glucose limitation , FLO11 transcription is regulated by Snf1p; the Snf1p-Gal83p isoform promotes cell adhesion during invasive filamentation by antagonizing Nrg1p- and Nrg2p-mediated repression of FLO11 [19] , [27] . While the central components of these signaling pathways have been identified , the scope of the yeast pseudohyphal stress response is broad [28]–[33] , and the mechanisms enabling these genes and gene products to drive pseudohyphal filamentation are incompletely defined , as are the genetic determinants within this gene set that underlie filamentation . To further dissect pseudohyphal growth pathways , we implemented a linkage study , coupling whole genome sequencing with pooled segregant analysis . The results present previously unidentified genetic determinants of yeast invasive growth and indicate the coevolution of proteins within complexes in driving phenotype .
For linkage analysis , we selected as parents the non-filamentous S288C-derived strain BY4741 and the filamentation-competent strains Σ1278b and SK1 [34] , [35] . Filamentous-form growth in haploid strains is classically assessed using the plate-washing assay of Gimeno et al . [1] to identify pseudohyphal cells that have invaded the agar substrate . The invasive phenotype of each parent strain in this assay is indicated in Fig . 1A . The experimental design of the linkage study is presented in Fig . 1B . The non-invasive S288C-derived strain was mated with each of the filamentous Σ1278b and SK1 strains , and the resulting diploid strain from each cross was sporulated . Meiotic progeny from dissected tetrads were assayed for agar invasion by plate-washing , and spores indicating strongly non-invasive or invasive phenotypes were pooled for subsequent linkage analysis . Only spores resulting from complete meiosis were included in these phenotypic pools , and intermediate filamentation phenotypes were excluded from subsequent analysis to provide the greatest likelihood of identifying allelic variation with a strong effect on filament formation . Genomic DNA was extracted from each segregant pool and subjected to high-throughput sequencing that yielded greater than 100-fold coverage per pool . From the BY4741-by-Σ1278b cross , 31 complete tetrads ( 124 spores ) were screened for agar invasion , identifying 37 strongly invasive spores and 63 non-invasive spores ( Fig . 2A ) . The segregant pools were sequenced , and candidate determinants of the invasive phenotype were identified using a linkage LOD score of greater than 3 as an arbitrarily defined cut-off . Table S1 provides a listing of these alleles , encompassing only variants that are in protein-coding sequence and that are non-synonymous with respect to the encoded amino acid sequence . This allele set affects 50 genes in eleven linkage blocks physically located on seven yeast chromosomes . Figure S1 summarizes the available functional information for this gene set . Representative plots of non-synonymous allelic variation with respective LOD scores are graphed in Fig . 2B for chromosomes V and IX , highlighting the pseudohyphal growth transcription factor gene FLO8 and the flocculin effector gene FLO11 . FLO8 is a pseudogene in S288C-derived strains [36] , and in this analysis , the BY4741 allelic variant containing a premature translational stop at codon 142 of the FLO8 sequence yielded a LOD score greater than 17 ( Fig . 2B and C ) . The FLO11 locus exhibits fifteen allelic changes linked with invasive growth phenotypes ( Fig . 2B and C ) . Previous studies identified allelic variation in FLO11 sequence encoding amino- and carboxy-terminal regions linked with the ability to form biofilms on the surface of wine [37] . We recovered these as well as additional sites of DNA sequence variation in FLO11 , with the Σ1278b-encoded alleles indicating linkage with strong invasive growth . The FLO11 sequence contains an internal repeat region that is a source of allelic variation between some strains and colonies [38] , [39]; however , we did not observe a change in the number of these repeats between BY4741 and Σ1278b . Collectively , the identification of these known pseudohyphal growth genes demonstrates the relevance of results obtained from our pooled segregant analysis . To further identify important determinants of invasion , we screened candidates from Table S1 as follows: 1 ) we generated gene deletions and assayed for invasive growth phenotypes ( Table S2 ) , and 2 ) for genes yielding deletion phenotypes , we generated mutants with swapped alleles to identify genetic variants required for invasive growth in Σ1278b . In particular , we focused on alleles of genes that contributed to cell polarity , cell cycle progression , cell morphology , and cell responses to nitrogen/carbon limitation , as these are hallmark characteristics of filamentation . By this approach , we identified variation in PEA2 as an important part of the genetics underpinning invasive growth . Pea2p localizes to sites of polarized growth as a component of a protein complex , termed the polarisome [40] , [41] . PEA2 is required for wild-type invasive growth , mating projection formation , and bipolar bud site selection in diploids [42] , [43] . In the filamentous Σ1278b strain , PEA2 codon 409 specifies methionine rather than the leucine residue encoded in the S288C-derived reference genome . The pea2-M409 allele was linked with invasive growth , and generation of an integrated site-specific mutation ( pea2-M409L ) reconstituting the S288C-encoded PEA2 allele in Σ1278b resulted in decreased invasive growth ( Fig . 3A ) . Relative to wild type Σ1278b , the cell morphology of the pea2-M409L mutant is altered , exhibiting decreased elongation ( Fig . 3B ) ; over a population of 200 cells , the percentage of pea2-M409L cells with a length:width ratio of less than 1 . 5 was nearly four-fold the corresponding percentage in a wild type strain . In addition , the pea2-M409L mutant is impaired in Kss1p MAPK signaling activity . The Kss1p kinase activates the Ste12p/Tec1p transcription factor complex , which recognizes a regulatory element ( FRE ) in the FLO11 promoter . The plasmid-based Pflo11-9/10-lacZ construct contains the Ste12p/Tec1p-responsive region of the FLO11 promoter fused to lacZ [6] , and , by this reporter , the pea2-M409L mutant yields significantly decreased Ste12p/Tec1p-dependent transcriptional activation of FLO11 relative to wild-type Σ1278b ( Fig . 3C ) . In contrast , the pea2-M409L mutation results in wild-type levels of a similarly designed FLO11 promoter fusion responsive to the PKA pathway effector Flo8p ( Fig . 3C ) [6] . Under conditions of vegetative growth haploid yeast cells bud in an axial pattern , with new buds emerging adjacent to the preceding bud site [44] . Haploid cells undergoing pseudohyphal growth , however , adopt a predominantly unipolar budding pattern wherein the first bud forms distal to the original cell division site , and subsequent buds cluster in the distal pole [1] , [10] . Here , we find that in the Σ1278b background the pea2-M409L mutant , corresponding to the S288C-encoded PEA2 allele , exhibited a decrease in unipolar budding and an increase in axial budding relative to wild type ( p<0 . 001 ) , with levels intermediate between an otherwise isogenic wild-type strain and a pea2Δ mutant ( Fig . 3D ) . For this analysis , we examined a population of invasive cells exhibiting three or more bud scars , such that patterns of axial , unipolar , bipolar , and random budding could be reliably distinguished [44] , [45] . This budding phenotype was evident in invasive cells , but not in an equally sized population of cells scraped from the surface of an agar plate . Previous studies have indicated that the majority of bud sites are distal in a pea2Δ mutant [12]; results here also indicate that the majority of bud sites are distal in pea2 mutants , but budding pattern analysis does indicate that Pea2p residue 409 impacts unipolar budding in invasive haploid cells . In the polarisome complex , Pea2p binds the scaffolding protein Spa2p , a large coiled-coil domain-containing protein required for polarisome function [41] , [46] . Here , we assessed the possibility that allelic variation at the PEA2 locus impacts Spa2p binding , using Protein A ( ProA ) -tagged Pea2p variants to recover by co-immunoprecipitation Spa2p tagged at its amino terminus with the hemagglutinin ( HA ) epitope . In the Σ1278b strain , the Pea2p-M409-ProA variant recovered more HA-Spa2p than the Pea2p-L409-ProA variant , representing the S288C-encoded PEA2 allele ( Fig . 3E ) . The BY4741-by-SK1 cross was implemented as described in Experimental Procedures , and phenotypic analysis of meiotic progeny identified 51 and 24 strongly invasive and non-invasive spores , respectively ( Fig . 4A ) . Subsequent deep-sequencing of the phenotypic pools identified allelic variation linked with invasive growth phenotypes within eleven separated locus blocks encompassing 88 genes exhibiting non-synonymous changes and a LOD score of greater than 4 ( Table S3 ) . A functional breakdown of these genes is indicated in Figure S2 . In this analysis , we used a higher LOD score relative to the Σ1278b-by-BY4741 cross in order to limit the number of selected allelic variants to a manageable size for further study , as the SK1 and BY4741 genomes are more divergent ( 99 . 5% sequence identity ) than the Σ1278b and BY4741 genomes ( 99 . 7% identity ) [47] . Very few allelic variants linked with invasive growth in Σ1278b were also identified in SK1 , aside from a few sequences near FLO8 that are unlikely to be causative . The set of identified alleles was primarily distinct between the two linkage studies , and deletion phenotypes for tested genes in SK1 are indicated in Table S4 . In haploid spores from this cross , genetic variation is most strongly linked with the invasive growth phenotype over a region of roughly 80 , 000 bp on chromosome XV , encompassing SFL1 ( Fig . 4B and C ) . SFL1 encodes a transcriptional repressor of pseudohyphal growth that functions by binding to the FLO11 promoter , thereby blocking transcriptional activation [24] , [25] , [48]–[51] . Consistent with its function in repressing FLO gene expression , deletion of SFL1 results in exaggerated invasive growth [28] , [32] , [48] . Interestingly , the SK1 strain contains an allelic variant of SFL1 with respect to S288C-derived strains , resulting in the conversion of codon 477 ( CAA encoding glutamine ) to a TAA stop codon ( Fig . 4D ) . This premature stop codon truncates SFL1 prior to the sequence encoding a domain ( AA 571–658 ) that is strongly similar to a conserved region in Myc oncoproteins [48] . Previous studies have identified hyperactive filamentation in a mutant of the CEN . PK 113-7D background upon introduction of a premature translational stop at SFL1 codon 320 ( Q320-stop ) [51] . Here , we found that allelic variation in SFL1 , encompassing a premature stop codon ( C1430T , Q477-stop ) in the SK1 background , is linked to the aggressively invasive phenotype of SK1 relative to BY4741 . In the BY4741-by-SK1 cross , allelic variation in MDM32 was linked to invasive growth more strongly than any other identified locus , with a LOD score of 9 . As indicated in Fig . 4B–D , MDM32 is found on chromosome XV , and relative to BY4741 , the SK1 allele of MDM32 encodes Ser182 and Phe262 rather than Cys and Leu , respectively . MDM32 encodes a protein complex subunit of the mitochondrial inner membrane required for membrane organization , the maintenance of elongated mitochondrial morphology , and mitochondrial DNA nucleoid stabilization [52] . Mitochondrial function is required for pseudohyphal growth , as filamentation-competent strains of S . cerevisiae containing a deleted version of the mitochondrial genome are unable to form pseudohyphae [53] , [54]; however , a role for Mdm32p in enabling invasive growth remains to be identified . To determine the effect of MDM32 allelic variation on pseudohyphal growth , we replaced the SK1-encoded MDM32-C546/T787 allele , specifying Mdm32p-S182/F262 , with BY4741-encoded MDM32-G546/A787 , specifying Mdm32p-C182/L262 , in the SK1 genetic background . This allelic swap decreased invasive growth in SK1 , and agar invasion was rescued upon reintroduction of the native SK1-encoded MDM32 allele ( Fig . 5A ) . The SK1 mutant containing the BY4741-encoded allele of MDM32 exhibited a more rounded cell morphology , with the percentage of cells displaying a cell length:width ratio of less than 1 . 5 increasing from 17% in wild-type SK1 cells to 68% in the SK1 mutant ( Fig . 5B ) . Reintroduction of SK1-encoded MDM32 recovered levels of cell elongation similar to wild type . To assess the impact of this allelic variation on mitochondrial function , we grew the SK1 mutant with the BY4741 allele of MDM32 on medium containing non-fermentable glycerol as the sole carbon source . As shown in Fig . 5C , the allele-swapped SK1 mutant grows poorly in glycerol-containing media , indicating that oxidative phosphorylation is impaired . The structure of the mitochondrial network is also perturbed upon introduction of the BY4741 allele of MDM32 in the SK1 background . Using the mitochondrion-specific MitoTracker fluorescent dye , which diffuses passively across the plasma membrane and concentrates in active mitochondria by membrane potential , we can visualize a compact and collapsed mitochondrial network in SK1 cells containing the BY4741-encoded Mdm32p-C182/L262 variant , similar to that observed in mdm32Δ ( Fig . 5D ) . MDM32 is a paralog of MDM31 , and the encoded proteins have been found to interact , albeit transiently and weakly , as components of protein complexes at the mitochondrial inner membrane [52] . We , therefore , assessed the effect of allelic variation at MDM32 on Mdm31p binding; however , we observed no difference in the recovery of Mdm31p by co-immunoprecipitation between the respective Mdm32p variants ( Figure S3 ) . In sum , MDM32 is a determinant of invasive growth , and replacement of the native SK1 allele of MDM32 with the BY4741-encoded allele yields a mutant filamentous growth phenotype .
The linkage analysis presented here identifies a broad gene set contributing to pseudohyphal growth ( Fig . 6 ) . The pattern of allelic linkage indicates a large number of determinant loci underlying invasive growth , consistent with results from systematic single-gene deletion and overexpression studies [28] , [31] , [32] . Within this gene set , components of the polarisome and mitochondria play important roles in enabling invasive growth . We report here that Pea2p residue 409 impacts bud site selection in haploid cells undergoing invasive growth , although the effect is less pronounced in determining initial distal-versus-proximal budding in virgin mother cells . Pea2p residue 409 lies distinct from the Pea2p coiled-coil region between residues 236 and 327 and is important for Spa2p binding . Spa2p interacts with Ste11p and Ste7p from the Kss1p MAPK pathway , providing a mechanism for polarisome-mediated regulation of Kss1p MAPK activity [41] . Our results further indicate that nuclear-encoded Mdm32p is required for invasive growth in SK1 , and that residues 182 and 262 , located outside of the mitochondrial pre-sequence ( AA 1–102 ) and at the boundary or outside of a transmembrane domain ( AA 161–184 and 636–653 ) , are important in enabling invasion , as well as in determining aerobic respiratory function and mitochondrial morphology . Mdm32p is proposed to function cooperatively with other inner membrane proteins and components of the outer mitochondrial membrane in the maintenance of mitochondrial morphology [52] , potentially through cytoskeletal interactions that may be affected by variation at these sites . This analysis highlights two additional points . First , the non-filamentous BY4741 background is not uniformly repressive with respect to pseudohyphal growth . From the Σ1278b and SK1 crosses , we identified three and five blocks of allelic variation , respectively , in BY4741 linked with the invasive growth phenotype; full listings of the encompassed alleles with respect to each cross are presented in Tables S5 and S6 . Alleles in S288C-derived strains that promote pseudohyphal growth are antagonized by alleles that repress filament formation , such as the pseudogene form of FLO8; similarly , alleles in SK1 linked with the non-invasive phenotype may be offset by alleles that promote invasion , such as the SFL1 allele containing a premature stop codon . Second , the identified allelic variation in PEA2 and MDM32 and the allele-swapping experiments performed here indicate that within a given genome , functionally interacting genes coevolve to impact phenotype . The majority of genetic variation linked with invasive phenotype in this study involves site-specific changes that do not create pseudogenes . Alleles of these genes yield functional proteins within the respective genomic contexts; however , a given allele results in a hypomorphic phenotype when introduced into a non-native strain . It should be noted that the BY4741-encoded allele of PEA2 may be viewed as being pseudohyphal competent , as Liu et al . [36] reported that the introduction of Σ1278b-encoded FLO8 in a S288C-derived strain is sufficient to enable at least some degree of pseudohyphal growth . The data here suggest that partner genes have likely co-evolved with genes such as PEA2 and MDM32 , and the resulting protein complexes are , thus , an important determinant of cell phenotype . These findings highlight the utility in studying these complexes as a whole , in supplement to individual proteins , in order to accurately identify the functions and properties that specify phenotype . It is interesting that the studies here indicated very little overlap between alleles linked with invasive growth in the Σ1278b and SK1 strains with respect to BY4741 ( Fig . 6 ) . Studies mapping quantitative trait loci ( QTL ) in a cross of the laboratory strain BY4716 and the vineyard strain RM-11 identified hotspots impacting gene expression , protein abundance , and small molecule-dependence [55]–[58] . These hotspots were principally due to alleles in the BY4716 background , leading Ronald and Akey [59] to suggest that the causative polymorphisms may occur at low frequency . The non-overlapping allele sets identified in our analysis are not suggestive of hotspots , but rather highlights the substantial importance of epistatic interactions in determining the sum filamentous phenotype resulting from variant alleles in the haploid segregants . These epistatic interactions likely represent instances of gene coevolution , which has been suggested to occur at an elevated rate for genes encoding proteins of shared biological functions and/or for proteins that have coevolved between species [60] , [61] . Clark et al . [60] have analyzed the rate of covariation for pairs of proteins over evolutionary time , and by this analysis , polarisome components as a whole do not exhibit statistically significant evidence of covariation , although many mitochondrial complexes do yield signature indicating evolutionary rate covariation . Further analyses of individual protein pairs from the strains used in our study will be necessary to identify a set of coevolved proteins that drive the filamentous growth phenotype . In summary , we used pooled segregant whole-genome sequencing to dissect gene networks that determine yeast pseudohyphal growth . This analysis identified allelic variation in the known pseudohyphal growth genes FLO8 and FLO11 , while also revealing variation in the negative regulator SFL1 , the coding sequence of which contains a premature stop codon in the invasive SK1 background . We further found that amino acid 409 in the polarisome protein Pea2p is a site of allelic variation critical for the protein's ability to signal through the Kss1p MAPK pathway , establish unipolar budding during pseudohyphal growth , and bind the Spa2p polarisome scaffold . Linkage analysis identifies variation in MDM32 as a determinant of invasive growth between S288C derivatives and the SK1 strain; the 182 and 262 residues are sites of variation and contribute to Mdm32 function in aerobic respiration and invasive growth .
A listing of yeast strains and plasmids used in this study is provided in Tables S7 and S8 . Haploid deletion mutants were constructed by PCR-mediated gene disruption using pFA6a-KanMX6 or pUG72 [62] , [63] . Yeast strains were propagated on rich YPD medium ( 1% yeast extract , 2% polypeptone and 2% glucose ) medium or synthetic medium as described [64] . Yeast invasive growth was assayed on YPD medium . The statistical modeling used to derive the probabilities of identifying linkage are described by Birkeland et al . [65] . Following mating as indicated in Fig . 1 , resulting strains were sporulated and asci were dissected . The dissected spores were grown overnight at 30°C and were individually tested for mating type . Spores resulting from complete meiosis ( four viable spores with two each of the α and a mating types ) were then used for whole genome sequencing . Each spore was assigned to invasive or non-invasive pools based on its invasive growth phenotype . To ensure equal representation of all segregants in a pooled population , each haploid strain was grown overnight at 30°C in individual 4 ml YPD cultures . The OD600 of the cultures was determined and used to calculate the appropriate volume of each strain so that upon mixing , we would achieve equal numbers of cells . A mate-pair library with 300-bp fragments was prepared for each of the phenotypic pools , and each library sequenced as paired-end reads using the Illumina Genome Analyzer ( University of Michigan DNA Sequencing Core ) . Sequence analysis was performed as described [65] . To obtain an estimate of the recombinant and non-recombinant spore counts in each phenotypic pool for a given observed sequence variant , we multiplied the number of spores in the pool by the fraction of sequence reads from that pool that matched the corresponding allele variant . These values were then used in standard LOD score calculations . Budding patterns of invasive cells were determined as previously described [44] , [45] . In brief , equal concentrations of mid-log phase cultures were spotted onto YPD plates and incubated for 7 days at 30°C; surface cells were subsequently washed off under a gentle stream of water . Residual invaded cells were recovered from the agar using a sterile toothpick . Cells were washed twice in sterile water and were stained with 2 µg/ml calcoflour white . Bud scars were visualized by fluorescence microscopy . Cells with more than three bud scars were examined . Budding patterns were determined by criteria previously described [45] . Budding patterns were divided into four sub-groups: axial , bipolar , unipolar and random . The axial pattern was defined as a long chain of bud scars on the proximal cell pole . Cells with a cluster of scars exclusively at the distal pole were classified as exhibiting unipolar budding . A pattern of medial bud scars was scored as random budding , whereas cells with bud scars equally distributed on both proximal and distal poles were classified as undergoing bipolar budding . For these analyses , 200–250 cells from each strain were scored . | Cellular processes in eukaryotes are brought about through the contributions of large gene sets , and a continuing obstacle in studying these processes lies in the identification of critical constituent genes . The yeast pseudohyphal growth transition is an important example of a complex cellular growth transition . During pseudohyphal growth , yeast cells form connected chains or filaments , constituting a means of foraging for nutrients under conditions of nitrogen and/or glucose limitation . Yeast pseudohyphal growth has been studied for over two decades as a model of signaling systems controlling stress responses , cell shape , and fungal virulence . Hundreds of genes are required for pseudohyphal growth , however , and the critical genes that determine the filamentous phenotype have not been elucidated . Towards this goal , we implemented a genetic approach to identify alleles linked with the pseudohyphal growth phenotype . These studies identified previously unstudied variation in proteins functioning in a complex that controls cell polarity and in a protein of the mitochondrial inner membrane . This work indicates that proteins in complexes and organelles have coevolved within a given genome to yield distinct outputs and phenotype , while highlighting the application of an approach that is useful for the analysis of complex phenotypes in many eukaryotes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systems",
"biology",
"fungal",
"genomics",
"functional",
"genomics",
"fungal",
"genetics",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics"
] | 2014 | Pooled Segregant Sequencing Reveals Genetic Determinants of Yeast Pseudohyphal Growth |
Sustained molecular oscillations are ubiquitous in biology . The obtained oscillatory patterns provide vital functions as timekeepers , pacemakers and spacemarkers . Models based on control theory have been introduced to explain how specific oscillatory behaviors stem from protein interaction feedbacks , whereas the energy dissipation through the oscillating processes and its role in the regulatory function remain unexplored . Here we developed a general framework to assess an oscillator’s regulation performance at different dissipation levels . Using the Escherichia coli MinCDE oscillator as a model system , we showed that a sufficient amount of energy dissipation is needed to switch on the oscillation , which is tightly coupled to the system’s regulatory performance . Once the dissipation level is beyond this threshold , unlike stationary regulators’ monotonic performance-to-cost relation , excess dissipation at certain steps in the oscillating process damages the oscillator’s regulatory performance . We further discovered that the chemical free energy from ATP hydrolysis has to be strategically assigned to the MinE-aided MinD release and the MinD immobilization steps for optimal performance , and a higher energy budget improves the robustness of the oscillator . These results unfold a novel mode by which living systems trade energy for regulatory function .
Similar to man-made systems that commonly employ sustained oscillations to measure time and length , living organisms use molecular oscillators to process spatiotemporal information for regulation . For example , the periodic pole-to-pole oscillation of Min proteins in Escherichia coli designates the mid-cell position for symmetric cell division [1 , 2]; the oscillatory spindle dynamics in Caenorhabditis elegans [3 , 4] and human cells [5] help position and orient the spindle at proper division site along the cell body; the genetic [6 , 7] and non-genetic [8 , 9] circadian rhythm networks repeatedly reset the intracellular environment every 24 hours; the RhoA and stress-fiber mediated oscillation synchronizes and coordinates the development of cells in Drosophila embryo [10 , 11]; the traveling and standing waves set the differentiation markers in the zebrafish segmentation process [12 , 13] . These different types of oscillators all emerge from various promotive and inhibitive interactions between the involved protein molecules , and their vital functions have been investigated extensively . However , the costs to sustain those functions have been overlooked almost completely . In particular , the free energy costs to drive the highly dissipative oscillating process , and “exchange rate” at which living organisms trade free energy for the above oscillatory regulation functions , remain largely unexplored . We address these questions by investigating the E . coli MinCDE oscillatory network for regulation of the cells’ symmetric division . This Min oscillator comprises three proteins: the division inhibitor MinC , the ATPase MinD , and the catalytic enzyme MinE . Experiments have shown that these Min molecules interact with each other under the mediation of ATP and the phospholipid membrane [14 , 15]: the ATP-bound MinD cooperatively associates with the cell membrane [16 , 17] , where MinE and MinC are recruited [18 , 19]; MinE enhances MinD’s ATPase activity , which releases MinD from membrane after ATP is hydrolyzed [15 , 20]; meanwhile MinC reduces the stability of FtsZ polymers that construct the scaffold of the division ring [21] , which in turn inhibits cell division at MinC-rich locations . Based on these protein-protein interaction logics , a molecular reaction-diffusion mechanism emerges: cytoplasmic MinD molecules associate with ATP and aggregate on the cell membrane at one of the two poles; MinE molecules chase and bind to this MinD colony , catalyze the ATP hydrolysis , and eventually destroy the MinD aggregation; the released MinD molecules will then diffuse to the other pole of the cell to start the aggregation process again . [22–27] . Such reaction-diffusion process gives rise to the spatiotemporal oscillation of Min molecules between two cell poles ( Fig 1A , 1B and 1C ) . An emerged oscillatory pattern ( S1 Movie ) will then result , on time average , in a “V”-shaped concentration profile of MinC along the cell’s long axis with the minimum at the mid-cell [28] ( Fig 1D ) . It is known that any sustained biochemical oscillation is dissipative and requires continuous free energy input [29 , 30] . In line with this general concept , the energy-bearing ATP is an essential element of the Min oscillator [14 , 15] . Different from the existing theoretical models that emphasize the topological and dynamic properties of the Min protein network [22–26 , 31] , we construct a general analysis framework to quantify the dissipative nature of the oscillator as well as the biophysical role of ATP . Our results explicitly indicate that sufficient free energy dissipation is required to switch on the oscillator , but counterintuitively , we found that free energy input does not always promote the differentiation of the mid-cell from the cell poles ( defined as the “performance” [31]; Fig 1D , also see Methods ) . Through a global optimization analysis , we further discovered that the best performance can only be achieved when most of the ATP hydrolysis energy is dissipated in the steps of the MinE-aided MinD release and the MinD immobilization , suggesting an optimal free energy dissipation strategy for E . coli under the pressure of natural selection . These discoveries set the Min oscillator apart from stationary regulators , such as the sensory adaption systems [32 , 33] and the kinetic proofreading system [34–36] , whose performance is monotonically improved by higher free energy dissipation . Thus , our results suggest different free energy conversion modes for stationary and oscillatory bio-regulators .
To analyze the thermodynamic properties of the MinCDE oscillator , we constructed a detailed biochemical model ( Fig 1A and 1B ) based on the protein-protein interaction logics suggested in previous studies [25] . In this model , two regulatory motifs are involved for oscillation: positive auto-regulation ( immobilized MinD cooperatively recruits more of itself ) and negative feedback ( the recruited MinE inhibits immobilized MinD via enhancing its ATPase activity ) . This category of interlinked positive and negative feedback loops are known to give rise to robust and tunable biological oscillations [37 , 38] . Since MinC only contributes to ring-inhibition but not oscillation [2 , 19] , it is not explicitly included in this study of the Min oscillator . A critical difference between our model and existing models is that we consider all reaction steps as “microscopically reversible” processes , which allows us to assess the free energy dissipation of individual steps through the forward and backward reaction fluxes [39] . These fluxes ( j±i ) can be expressed as: j+1=k+1ρD:D , j−1=k−1ρD:T , j+2=k+2ρD:T , j−2=k−2ρd , j+2′=k+2′ρD:T ( ρd+ρde ) , j−2′=k−2′ρd ( ρd+ρde ) , j+3=k+3ρEρd , j−3=k−3ρde , j+4=k+4ρde , j−4=k−4ρD:DρE , where ρD:D , ρD:T & ρE are the concentrations of cytoplasmic MinD:ADP , MinD:ATP complexes and MinE dimers , respectively; ρd & ρde are the concentrations of MinD:ATP and MinE:MinD:ATP complexes on the membrane , respectively . The reaction index i ( = 1 , 2 , 3 & 4 ) represents the reactions labeled in Fig 1B accordingly . It is worth noting that j ± 2 ′ represent the cooperative recruitment of MinD:ATP to the cell membrane by existing membrane-associated MinDs ( i . e . d & de ) . For each reaction step , the net flux is ji = j+i−j−i ( i = 1 , 2 , 3 & 4 ) ( and j 2 ′ = j + 2 ′ − j − 2 ′ for the cooperative recruitment ) . In consideration of the diffusion of protein molecules in the cell volume , the dynamics of the system can therefore be described by a set of reaction-diffusion equations ( see Methods for details ) , and if detailed balance is satisfied , the parameter γ ≡ k + 1 k + 2 k + 3 k + 4 k - 1 k - 2 k - 3 k - 4 shall be unity; otherwise , the system is out of equilibrium and the chemical free energy from ATP hydrolysis is constantly dissipated . The developed microscopically reversible reaction network allows us to assess the dissipation level of the oscillator . However , although the physics and methodology for evaluating free energy dissipation ( or entropy production ) of chemical reaction systems with stationary solutions has been nicely reviewed [29 , 39] , to our knowledge , dissipation of reaction-diffusion systems has yet to be broadly investigated [40] . A seemingly straightforward way to analyze such spatiotemporally varying systems is to compute and integrate the unbalanced fluxes of all reaction steps , as well as the diffusion processes ( see Methods for details ) . Here , we obtain the total free energy dissipation rate of the reaction-diffusion system in a more elucidating way . For any open biochemical reaction system in a steady or sustained oscillatory state , its average free energy dissipation rate is equal to the average consumption rate of the chemical free energy embedded in the environmental fuel molecules [41] . In our case of the Min oscillator , ATP is the energy source continuously supplied from the cytoplasm . The oscillator uptakes ATP molecules and excretes the hydrolysis products ( ADP and inorganic phosphate , Pi ) in the nucleotide exchange step ( k±1 ) and the MinE-aided MinD release step ( k±4 , which is also the ATP hydrolysis step ) , respectively . Therefore , the chemical free energy supplied from ATP should directly be the free energy consumed by the oscillator , which is ln γ per ATP ( in unit of kB T ) [41] . It is worth noting that this free energy depends on the cytoplasmic concentrations of ATP , ADP and Pi , and is different from the standard free energy change of ATP hydrolysis [42] . Hence , the continuous free energy dissipation rate is: σ ATP ( t ) = ∫ V d x j 1 ( x , t ) ln γ ≡ J 1 ( t ) ln γ , ( 1 ) where the integral is taken over the cell volume , and the integrated flux J1 ( t ) is the number of MinDs transiting from MinD:ADP to MinD:ATP per unit time . For a stable stationary system , J 1 = J 2 + J 2 ′ = J 3 = J 4 ≡ J is invariant over time; whereas for a sustained oscillatory system , the flux-balance relation holds only for time average over the oscillation period ( P ) : ⟨ J 1 ⟩ ≡ P − 1 ∫ t 0 t 0 + P d t J 1 ( t ) = ⟨ J 2 + J 2 ′ ⟩ = ⟨ J 3 ⟩ = ⟨ J 4 ⟩ ≡ ⟨ J ⟩ . From an energetic point of view , σATP represents the chemical free energy “deposit” rate through uptaking ATP molecules from the cell’s cytoplasm . One should be aware that under sustained oscillating condition , the time averaged σATP over each period is equal to the time average of the instantaneous dissipation rate σtot ( Fig 1C ) , i . e . ⟨ σ ATP ⟩ = ⟨ σ tot ⟩ = ⟨ J ⟩ ln γ ≡ ⟨ σ ⟩ . ( 2 ) We use this average dissipation rate for sustained oscillation in our analysis . Also , to analyze the system’s behavior within the same framework , we use ⟨σ⟩ to denote the dissipation rate for non-oscillatory states whose average value is equal to the instantaneous value . It is straightforward from Eq ( 2 ) that the equilibrium situation ( γ = 1 ) is non-dissipative . Eq ( 2 ) provides a convenient way to compute free energy dissipation . We first investigate how the dissipation rate ⟨σ⟩ depends on the non-equilibrium parameter γ by tuning the backward nucleotide exchange step ( MinD:ATP → MinD:ADP , k−1 ) , while keeping other reaction rates fixed ( see Methods ) . The system exhibits two distinct regimes for the ⟨σ⟩ − γ dependence: for small γ , ⟨σ⟩ initially increases slowly with ln γ; when γ is further increased , ⟨σ⟩ increases dramatically with ln γ and then settles to a logarithmic regime where ⟨σ⟩ ∝ ln γ ( Fig 2A; S1 and S2 Figs ) . The threshold between the two regimes is the same transition point for oscillation to occur ( circle in Fig 2A ) , and the Min system exhibits only the regulatory function in the oscillatory regime ( in the non-oscillatory stable regime , MinD is distributed uniformly on the membrane or slightly higher in the middle due to the cell-end effect as shown in S6 Fig ) . It is worth noting that the logarithmic dependence has been also observed in the stationary sensory adaptation systems when the net fluxes ⟨J⟩ reach their saturated levels [32 , 33] . Our results suggest that this category of logarithmic relation between the non-equilibrium parameter and the associated free energy dissipation may be a universal qualitative indicator of a biochemical network providing its regulatory function , regardless of it being intrinsically stationary or oscillatory . To investigate further how this logarithmic regime coincides with the oscillatory behavior of the system , we analyze the spatiotemporal patterns of Min protein molecules while the system becomes more dissipative . We find that the MinE dimers remain mostly in cytoplasm until the same dissipative “threshold” is reached; beyond this “threshold” , most MinEs are confined onto the membrane ( Fig 2B ) . Due to the effective inhibitive role of MinE on MinD:ATP being associated with the membrane , oscillation can occur only at low cytoplasmic MinE molecule numbers . Therefore , when oscillation is switched on , the average flux is rate-limited by the MinE number: ⟨J⟩osc ≈ ⟨J⟩max = k+4 NE , and the free energy dissipation is then ( Fig 2A , dashed line ) : ⟨ σ ⟩ osc ≈ k + 4 N E ln γ . ( 3 ) Similar threshold and logarithmic dependence of free energy dissipation on γ can also be observed by constraining other reverse reaction rates from their equilibrium values ( S3 and S4 Figs ) , and the dissipation rate can still be approximated by Eq ( 3 ) in the oscillatory/regulatory regime ( see S1 Text for details ) . The threshold to switch on the oscillator ( Fig 2 ) indicates that oscillation occurs only when the dissipated free energy is large enough to sequestrate MinE from cytoplasm . Because the nucleotide exchange and MinE-aided MinD release steps are the two material exchange steps between the oscillator and the intracellular environment , we systematically investigate how the oscillator’s regulatory performance depends on these two interfaces . Fig 3A and 3C show the performance analysis in the ( k+1 , k−1 ) and ( k+4 , k−4 ) spaces , respectively . The sharp bifurcation boundaries separate the oscillatory ( colored ) and stationary ( white ) regions: for the nucleotide exchange step , our result indicates that a fast ATP replacement rate and a large γ are both necessary for the oscillator to operate . In contrast , for the MinE-aided MinD release step , the system oscillates only when the releasing rate ( k+4 ) remains within a moderate range while the rebinding rate ( k−4 ) stays low . Despite the quantitative difference in the parameter scales , the regulatory performance is positive only in the oscillatory regions , and a minimum free energy dissipation needs to be satisfied for sustained Min oscillation for both material exchange interfaces . Once the energetic threshold is met , the system reacts differently to the tuning of these two interfaces . When its dissipative behavior is tuned through the nucleotide exchange step ( k±1 ) , the system achieves its best performance in the region adjacent to the bifurcation boundary ( red region in Fig 3A ) . A more explicit representation is given in Fig 3B: the solid and dashed lines correspond to the solid and dashed lines in Fig 3A , along which the regulatory performance and energetic costs are quantified . In both cases , the system’s performance first increases from zero to a finite peak value , and then decreases to a lower level ( see panels A & B in S5 Fig for the membrane-associated MinD profiles ) . The negative performance before the bifurcation is due to the cell-end effect ( S6 Fig ) . On the other hand , when the system is tuned through the MinE-aided MinD releasing step ( k±4 ) , the high performance region is buried deeply in the oscillatory phase , and the performance exhibits simple monotonic dependence on free energy dissipation ( Fig 3C and 3D and panels C & D in S5 Fig ) : at any given k+4 , a higher free energy dissipation ( i . e . smaller k−4 ) always leads to a better performance in mid-cell recognition . Interestingly , if we hold k−4 constant and vary k+4 ( along a horizontal line in the performance map ) , the performance to dissipation relation would be non-monotonic with the best performance being achieved at moderate k+4 values . These qualitatively distinct performance-to-cost relations indicate that not only different transition steps have different roles in the regulation process , but also the forward and backward reactions in the same step have different influences on the regulatory function . Moreover , the existence of an “optimal” operating region is distinct from other known stationary regulatory systems ( e . g . sensory adaptation and kinetic proofreading systems ) whose performance always improves monotonically with increasing free energy dissipation [32–36] , suggesting that , in addition to direct free energy consumption , more complex dynamic requirements have to be satisfied for the Min oscillator to assume the role as an efficient mid-cell marker . The results shown in Fig 3 suggest that the regulatory performance of the Min oscillator does not depend simply on how much free energy is dissipated , but indeed on how exactly the free energy is dissipated through the individual steps in the reaction-diffusion process . To investigate the E . coli cell’s dissipation strategy in mid-cell determination , we performed a systematic analysis of the Min oscillator’s performance by changing all the reverse reaction rates ( k−i , i = 1 ∼ 4 ) while keeping γ fixed at the physiological level ( i . e . ln γ is maintained at around 18 kB T to match the chemical free energy released from hydrolyzing each ATP molecule in E . coli[42] ) . This is equivalent to assigning different free energy consumption to different reaction steps while keeping the total free energy budget constant ( given in Eq ( 3 ) ; see Fig 4A for illustration ) . Fig 4B summarizes the results of our in silico experiments . In this figure , we use ΔGi = ln k−i/k+i to denote the standard free energy difference through the ith reaction step . It is worth special attention that the “standard” condition used here is slightly different from the conventional standard condition: since ATP , ADP and Pi are not explicitly included as reactants/products in our model but are implicitly embedded in the reaction rates k±i , we set our “standard” condition to be at unit concentration of Min molecules and at physiological levels of ATP , ADP and Pi . Since the total Δ G = ∑ i = 1 4 Δ G i = − ln γ per ATP is constant , only three out of four of the reactions are independent . In each elementary panel of Fig 4B , we present the performance contour plots in the ( ΔG1 , ΔG4 ) space at different ΔG3 values . The best performance point in each contour plot is identified and plotted in Fig 4C , where the red bars show the best performance scores at given ΔG3 levels , and the partitioned color bars indicate the calculated free energy dissipation rates for each of the individual reactions ( magenta , green , yellow & blue for nucleotide exchange , MinD immobilization , MinE recruitment & MinE-aided MinD release , respectively ) and the diffusion process ( cyan ) to achieve the best performance . The total heights of the partitioned color bars directly represent the total free energy dissipation rates , which are nearly constant ( close to k+4 NE ln γ = 12 , 600 kB Ts−1 ∼ 700 ATPs per second ) . These results provide a full spectrum picture of how energy is used to promote the Min oscillator’s performance . Firstly , oscillation occurs only within certain dissipation-strategy regions ( colored in Fig 4B ) . A “bad” strategy can eliminate the oscillations , therefore having no regulatory function even with abundant energy source . Secondly , under the global optimal scenario ( ΔG3 = −3 kB T in Fig 4C , where the performance score is globally the highest ) , the largest amount of free energy ( 4 , 440 . 9 kB Ts−1 , 36 . 5% of ⟨σ⟩osc ) is dissipated in the MinE-aided MinD release step , where ATP is hydrolyzed and Pi is released; a similar amount of free energy ( 4 , 332 . 5 kB Ts−1 , 35 . 6% of ⟨σ⟩osc ) is dissipated to ensure efficient MinD immobilization onto the cell membrane; a significant amount of free energy ( 2 , 681 . 4 kB Ts−1 , 22% of ⟨σ⟩osc ) is dissipated to recruit MinE to the membrane; and the least free energy ( 460 . 4 kB Ts−1 , 3 . 8% of ⟨σ⟩osc ) is used for nucleotide exchange ( the rest 2 . 1% is dissipated in diffusion ) . From a structural point of view , it is known that binding of ATP helps the MinD molecule to rearrange its structure for a higher affinity to phospholipid and to MinE [43–45] . Therefore , the large amount of free energy dissipated in the membrane and the MinE involved steps might suggest that , under the pressure of natural selection , E . coli cells may have evolved to use the energy bearing ATP molecules fully to coordinate with the necessary structural changes for optimal operation . This theoretically-obtained optimal partition also clearly indicates that different reaction steps play different roles in converting free energy for the oscillatory functions . We want to point out that the above dissipation partitions , as well as the color bars in Fig 4C , are calculated using the time averaged dissipation rates in individual reaction steps ⟨σi⟩: ⟨ σ i ⟩ = 1 P ∫ P d t σ i ( t ) = 1 P ∫ P d t ∫ x d x j i ( x , t ) ln j + i ( x , t ) j - i ( x , t ) . Taking reaction 1 for example , ⟨ σ 1 ⟩ = - ⟨ J ⟩ Δ G 1 + 1 P ∫ P d t ∫ V d x j 1 ( x , t ) ln ρ D : D ( x , t ) ρ D : T ( x , t ) , which is different from ΔG1 = ln ( k−1/k+1 ) , which is the standard free energy difference between reactants and products under cytoplasmic ATP , ADP and Pi concentrations . ΔGi is useful for understanding the reaction landscape of the system as illustrated in Fig 4A , but ⟨σi⟩ represents the true dissipation at particular reaction steps ( colored bars in Fig 4C ) . It is also worth noting that Eq ( 2 ) leads to ⟨σ⟩ = −⟨J⟩ΔG ( i . e . the dissipated free energy is equal to the chemical free energy consumed through ATP hydrolysis ) . Maintaining the above two dynamic conditions requires sufficient energy input to break symmetry in the proteins’ concentration distribution between the two poles . Fig 2A shows a minimum γ value obtained from tuning the nucleotide exchange step . We extended our analysis to a broader biochemical space to include all reverse reaction steps , and we further applied the global optimization method to different γ values ( i . e . total energy budgets for dissipation ) . Our results show that , the Min oscillation can never be switched on if the total energy budget is too low , no matter how the system is optimized ( Figs 5 and 6 ) . This minimum value of ln γ is around 6 kB T . By comparing the operation phase plots at different γ values ( Figs 4B and 5 ) , we clearly show that the higher the total energy budget , the bigger the area in which oscillation could occur , implying higher robustness of the oscillator . Thus with an adequate energy budget , the organism has more freedom to arrange its internal environment while maintaining the vital oscillatory function . Fig 6 summarizes the global optimal performance with increasing γ . By carefully tuning the Min pathway for optimization , a higher energy budget can eventually lead to a better global optimal performance until a saturation level is reached . This result indicates that the total energy budget is important for the highest achievable differentiation between the mid-cell and the pole regions , but the detailed dissipative strategy plays a more direct role in controlling the actual performance that the system can deliver .
Living organisms are all dissipative and the dissipated free energy is used to perform mechanical work , facilitate bio-synthesis/degradation processes , and process regulatory information . However , although it is intuitive to connect energy to mechanical work and to bio-mass turnover , a bigger challenge is the quantitative understanding of the free energy conversion mechanism through biological regulation processes . This difficulty is mainly because the free energy consumption in biological regulation is hardly accessible at the molecular level , and meanwhile the embedded information processing is difficult to quantify and evaluate . To date , only two energy-regulation conversion schemes have been quantitatively identified: the KP scheme and the ESA tradeoff scheme . The KP scheme was proposed by Hopfield for the Kinetic-Proofreading processes whose regulatory accuracy can be enhanced through every cycle of enzymatic checking [34–36] . In this scheme , the dissipated free energy effectively lowers the energy of an already stable state . On the other hand , it was recently discovered the ESA scheme in the sensory adaptation systems that higher free energy consumption , which is the product of the Energy dissipation rate and the inverse of the adaptation Speed , exponentially improves adaption Accuracy . In those processes , energy is used to stabilize an originally unstable state [32 , 33] . Both schemes are stationary and exhibit monotonic performance-to-cost relation . In this work , we present a third energy-regulation conversion scheme that is unique for oscillatory regulators . Our results show that , although energy is necessary to sustain biochemical oscillation , the regulatory performance does not monotonically depend on the total free energy dissipated over the full reaction cycle . Instead , the oscillator’s performance largely depends on how the energy is partitioned and dissipated through individual catalytic steps . In particular , for the MinCDE system , an optimal dissipation strategy is to allocate most of the available free energy to the MinE-aided MinD release and the MinD immobilization steps ( i . e . large σ2 & σ4 ) , whereas the nucleotide exchange step has to be kept less dissipative ( i . e . small σ1 ) . This non-monotonic optimizable feature is distinct from the KP and ESA schemes . Our results provide energetic insights into the question of what drives the MinCDE oscillation . As predicted from the reaction-diffusion mechanism , the sustained MinCDE oscillation would require an uninterrupted pole-to-pole passage of MinD molecules together with a strong confinement of MinE molecules to the cell membrane [31] . To guarantee the uninterrupted MinD passage , a MinD concentration gradient between the two cell poles has to be established and maintained during the fast diffusion process ( ∼ 0 . 5 seconds ) with the following perspectives: from a mass transfer point of view , such a gradient requires a “source” at the previously MinD-occupied pole and a “sink” at the far end of the cell; from an energetic standpoint , the “source” and “sink” are maintained by the highly “dissipative” MinE-aided MinD release and the MinD immobilization steps , respectively . In particular , a large σ4 keeps the rebinding of MinD at low levels , providing a net inward flux of MinD from the membrane; whereas a large σ2 guarantees near-perfect absorption of MinD onto the membrane at the far end . Once a gradient is established , a remaining task is to avoid MinD binding to other parts of the cell membrane other than the far end cell pole . To this end , two additional requirements have to be satisfied: 1 ) the nucleotide exchange step should have a relatively high reverse rate with low dissipation level ( i . e . small k+1/k−1 ratio and small σ1 ) so that once MinD associates with ATP ( high membrane affinity ) , it can easily exchange back to the ADP state ( low membrane affinity ) ; 2 ) the nucleotide exchange rates ( k±1 ) have to be fast so that diffusion cannot flatten out the gradient profile before MinD is ready for membrane binding . The first condition reduces the chance of MinD binding to the membrane during its first passage down the gradient; whereas the second condition reduces the chance of “backfire” in which MinD diffuses back . These two requirements are evident in Fig 3A: the performance is higher near the bifurcation boundary than deep inside the oscillatory phase region; along the γ contour line ( in this case , also the k+1/k−1 contour line ) passing the bifurcation boundary , the larger the rates , the better the performance will be . Sustained MinCDE oscillation also requires MinE dimers to be confined at the old pole and meanwhile the cytoplasmic MinE has to be maintained at a low level to allow MinD accumulation at the new pole . To satisfy this condition , the cooperative recruitment step , where MinE are recruited , acquires a significant amount of free energy ( σ3 ) to create a strong recruiting trap for MinE . Therefore , the MinCDE oscillator uses the free energy in ATP to establish alternating gradients across the long axis of the cell , so that the MinD can diffuse to and cooperatively build up a colony at the far end of the cell before the MinE can chase and destroy it . Furthermore , the limited intracellular energy budget has to be partitioned wisely to balance the requirements from individual reaction steps . And our predicted optimal free energy dissipation strategy captures these dynamic logics , suggesting that energetic and dynamic requirements are deeply connected . It is worth noting that our study is based on a deterministic picture of the problem . In a previous study by Rex et al . , the stochastic effects from finite numbers of molecules have been investigated numerically using microscopically irreversible reactions [46] . They found that , at physiological levels of molecular numbers ( > 2 , 000 molecules ) , the oscillatory behavior of the stochastic MinCDE system is close to those derived from the deterministic equations , and the noise of the oscillatory period and the averaged MinD concentration profile are at a relatively low level . These results encourage us to believe that our discoveries of how free energy is traded for better identifying the mid-cell position would still be valid in the context of a stochastic cellular environment . But it would require more detailed investigation to identify whether free energy dissipation could also contribute to reducing the noise arising from the finite numbers of molecules . All existing theoretical models , for simplification , consider biological reactions as microscopically irreversible processes [22–27] . Although these models have successfully captured many qualitative and quantitative features of the studied biological systems , we show in this paper that the predicted behaviors of the MinCDE system from irreversible models do not converge to the optimal operating mode obtained from reversible analysis . In particular , our analysis ( Fig 3A ) explicitly shows that the MinCDE system can benefit from its reversibility and the reversible system is capable of achieving a much higher performance compared to the irreversible counterpart ( note the maximum performance near the bifurcation boundary ) . These results indicate that the backward reaction does not just reduce the “net” forward reaction rate , but essentially enlarges the “volume” for the system’s dynamic trajectories to occupy , which in turn leads to richer dynamic behaviors . Furthermore , the obtained performance-to-cost relation via reversible analysis implies that “energy efficiency” as well as the “operational robustness” might be important factors for living organisms to survive the pressure of natural selection , shedding light on the evolutionary principles of regulatory networks . Therefore , we believe that it is worthwhile to reexamine other biochemical systems using the reversible modeling framework introduced here for more comprehensive thermodynamic understanding .
Given the net flux for each biochemical reaction step ( ji = 1 , 2 , 3 & 4 and j 2 ′ ) , the dynamics of the reaction-diffusion system can be described as: ∂ρD:D∂t=DD∇2ρD:D−j1 , ( 4 ) ∂ρD:T∂t=DD∇2ρD:T+j1 , ( 5 ) ∂ρE∂t=DE∇2ρE , ( 6 ) where DD & DE are the diffusion constants of MinD:ADP and MinE in cytoplasm , respectively . The system’s boundary conditions are defined by the membrane-associated reactions on the cell membrane: ∂ρd∂t=j2+j2′−j3 , ( 7 ) ∂ρde∂t=j3−j4 , ( 8 ) D D n · ∇ ρ D : D = j 4 , ( 9 ) D D n · ∇ ρ D : T = - j 2 - j 2 ′ , ( 10 ) D E n · ∇ ρ E = - j 3 + j 4 , ( 11 ) where n is the unit normal vector of the membrane pointing outward . We want to point out that we adopt here the cooperative recruitment type of MinCDE oscillatory mechanism in which the diffusion on the cell membrane is considered unessential [25 , 31 , 47] . Similar analyses can be carried out for the other non-linear aggregation type of mechanism where 2-dimensional diffusion on the membrane plays an important role [24 , 26] , or for the recent comprehensive study where the two mechanisms and the direct binding of MinE to the membrane [18 , 48 , 49] are taken into account [27] . These would be beyond the scope of this paper , but we expect that qualitatively similar conclusions would be reached . Using COMSOL Multiphysics 4 . 3 , we simulate the E . coli cell as a cylinder with radius R = 0 . 5 μm and length L = 4 μm . The diffusion constants are set to be DD = 16 μm2 s−1 and DE = 10 μm2 s−1[50] . The total numbers of MinD particles and MinE dimers are fixed at ND = 2 , 000 and NE = 700 , respectively [51] . Unless stated otherwise , the forward reaction rates are chosen from experimental measurements and previously theoretical studies: k+1 = 6 s−1 , k+2 = 0 . 1 μms−1 , k + 2 ′ = 0 . 01 μ m 3 s − 1 , k+3 = 0 . 4 μm3 s−1 and k+4 = 1 s−1[17 , 18 , 25 , 27 , 31] . The backward rates , on the other hand , are tuned to alter the system from equilibrium to non-equilibrium and are specified in the corresponding sections of the paper . In addition , although many experiments have confirmed that the binding process of MinD to the membrane is cooperative , yet lipid specific [17 , 18] , the detailed mechanism is unclear . In this paper , we adopt a simple catalytic view that treats the spontaneous ( j±2 ) and the cooperative attachment ( j ± 2 ′ ) with the same free energy change . Therefore , k + 2 / k − 2 = k + 2 ′ / k − 2 ′ is satisfied in our minimal simulation setup . Using these parameters , the oscillation period is around 40s as long as k−4 stays small as shown in S7 Fig , which is in agreement with experimental observations [2 , 52] . S1 Movie shows an example of the oscillatory dynamics . Pole-to-pole oscillation is tightly coupled with the inequality of Min protein concentrations along the cell’s long axis . In our performance studies , we use an automated method to first detect the oscillatory dynamics: we collect the species’ concentrations over long time ( for example , MinD concentration at one of the cell poles ) , and evaluate the variances of the time course data in steady state . The system is regarded to undergo oscillatory dynamics if the variance is non-zero . Based on the time-averaged concentration profile along a cell’s long axis , we apply Halatek and Frey’s definition for the regulatory performance [31]: let h and w , respectively , be the normalized depth and width of the valley of the concentration profile of the membrane-bound MinD and MinD:MinE complex; then the performance is defined as h/w . ( Fig 1D shows a particular example which has one extremum . A more detailed and general definition/illustration can be found in S8 Fig ) During oscillation , due to the canalized MinD transfer , MinD molecules periodically switch their occupancy at two cell poles . This results in a profile with lower MinD concentration at the mid-cell region . Therefore , the oscillation and the system’s positive performance are tightly coupled . A narrower and deeper valley at the mid-cell is considered superior for correct symmetric cell division , and is quantified by a higher performance score . We use such scores to demonstrate the relation between the regulatory performance and the associated energetic cost . We use the imbalanced fluxes in the reaction-diffusion process to calculate the free energy dissipation rate . Free energy dissipations in individual reaction steps . For a particular chemical reaction i , if the forward and backward flux densities at a spatial position x are j+i ( x , t ) and j−i ( x , t ) respectively , the dissipation rate density is [39]: σi ( x , t ) =[j+i ( x , t ) −j−i ( x , t ) ]lnj+i ( x , t ) j−i ( x , t ) =ji ( x , t ) lnj+i ( x , t ) j−i ( x , t ) . ( 12 ) For notational simplicity , the spontaneous and cooperative MinD attachments are denoted as one reaction here , i . e . , σ 2 = ( j 2 + j 2 ′ ) ln ( j + 2 / j − 2 ) . The total dissipation rate of all the reactions is therefore: σ tot react ( t ) = ∫ V d x σ 1 ( x , t ) + ∫ S d x ∑ i = 2 4 σ i ( x , t ) , where the subscripts V and S denote the cell volume and cell surface integrals , respectively . Free energy dissipations in diffusion . The dissipation of the diffusion process is less obvious . We model the diffusion process as a number of particles performing random walks on a uniform 3-dimensional lattice space with grid size dxdydz . Each molecule on one node ( x ≡ ( x , y , z ) ) can jump to its six neighboring nodes with rate a . Taking two neighbors along x direction for example , the forward and backward fluxes are j + x ( x ) = j ( x → x + d x , y , z ) = a ρ ( x , y , z ) d x j - x ( x ) = j ( x + d x → x , y , z ) = a ρ ( x + d x , y , z ) d x This leads to the well-known equality for the net flux: j x ( x ) = lim d x → 0 a [ ρ ( x , y , z ) - ρ ( x + d x , y , z ) ] d x = - D ∂ ρ ( x ) ∂ x and the diffusion constant is defined as D = limdx → 0 a ( dx ) 2 . The dissipation rate between these two neighbors is: lim d x → 0 d y d z j x ( x ) ln j + x ( x ) j - x ( x ) = d x d y d z j x 2 ( x ) D ρ ( x ) , ( 13 ) where jx ( x ) is the x component of the net diffusion flux j ( x ) at location x . Hence the dissipation rate density of diffusion is σ diff ( x , t ) = j x 2 + j y 2 + j z 2 D ρ ( x , t ) = | j ( x , t ) | 2 D ρ ( x , t ) . ( 14 ) This final result in terms of the flux vector is independent of coordinates . Because we disregard the diffusion on cell membrane , the total dissipation rate of the diffusion processes of the three diffusive cytoplasmic molecules can be written as: σ tot diff ( t ) = ∫ V d x [ σ D : T diff ( x , t ) + σ D : D diff ( x , t ) + σ E diff ( x , t ) ] , ( 15 ) where the integral is taken over the cell volume . We find that the dissipation of the diffusion process is small compared to the reactions ( see Fig 4C ) . Combining reaction and diffusion , the total dissipation rate of the system can be calculated by summing these two parts: σ tot ( t ) = σ tot react ( t ) + σ tot diff ( t ) . ( 16 ) Please note that the instantaneous dissipation rates calculated from Eqs ( 1 ) and ( 16 ) are different for oscillatory systems . However , the total dissipated free energy over each period is the same from these two analysis methods ( as shown in Eq ( 2 ) and Fig 1C ) . We use COMSOL Multiphysics 4 . 3 to solve the partial differential equations . The cell is set up as a cylindrical volume with axial symmetry , which leaves us with a mathematical 2-dimensional model . The maximum grid size is set to be 0 . 1 μm . The entire system is then simulated using two physics modules provided by COMSOL: the “Transport of Diluted Species” module is used to simulate the reaction-diffusion process in cytoplasm , and the “Boundary ODEs and DAEs” module is used to account for the reactions on the cell membrane . These two modules are coupled: the membrane reactions serve as the boundary condition for the cytoplasmic reaction-diffusion process . The deterministic equations have an unstable solution with a uniform distribution of all species , so to break this unstable symmetry solution we used a step function as our initial condition for the MinD:ADP concentration in cytoplasm , while making MinD:ADP and MinE homogeneously distributed in the cytoplasm . All the simulations are run for a time long enough to cover at least 10 full periods of sustained oscillation with stable amplitude , and data are collected for these stable oscillations . | This paper presents a unique dissipation mode of converting biochemical free energy in ATP to regulatory function through the MinCDE bio-oscillator that marks the mid-cell position for symmetric bacterial cell division . Through assessing the oscillator’s performance-to-cost relation , we demonstrate that some dissipation threshold needs to be satisfied to switch on the oscillation , but the oscillator’s performance can be damaged by excess free energy dissipation , which is distinct from the known monotonic tradeoff relation of stationary regulators . An optimal dissipation strategy has been unveiled: the ATP free energy must be precisely allocated to specific reaction steps for accurate mid-cell recognition , which also coincides with the dynamic requirements for robust oscillation to occur . These discoveries identify an optimizable operation scheme of free energy consumption in biological systems and provide deep insights into the evolution of dynamic regulatory networks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | An Optimal Free Energy Dissipation Strategy of the MinCDE Oscillator in Regulating Symmetric Bacterial Cell Division |
When searching sequence databases for RNAs , it is desirable to score both primary sequence and RNA secondary structure similarity . Covariance models ( CMs ) are probabilistic models well-suited for RNA similarity search applications . However , the computational complexity of CM dynamic programming alignment algorithms has limited their practical application . Here we describe an acceleration method called query-dependent banding ( QDB ) , which uses the probabilistic query CM to precalculate regions of the dynamic programming lattice that have negligible probability , independently of the target database . We have implemented QDB in the freely available Infernal software package . QDB reduces the average case time complexity of CM alignment from LN2 . 4 to LN1 . 3 for a query RNA of N residues and a target database of L residues , resulting in a 4-fold speedup for typical RNA queries . Combined with other improvements to Infernal , including informative mixture Dirichlet priors on model parameters , benchmarks also show increased sensitivity and specificity resulting from improved parameterization .
Many functional RNAs conserve a base-paired secondary structure . Conserved RNA secondary structure induces long-distance pairwise correlations in homologous RNA sequences . When performing database searches to identify homologous structural RNAs , it is desirable for RNA similarity search programs to score a combination of secondary structure and primary sequence conservation . A variety of approaches for RNA similarity searching have been described . There are specialized programs for identifying one particular RNA family or motif , such as programs that identify transfer RNAs [1 , 2] , small nucleolar RNAs [3 , 4] , microRNAs [5 , 6] , signal recognition particle ( SRP ) RNAs [7] , and rho-independent transcription terminators [8] . There are also pattern-matching algorithms that rely on expertly designed query patterns [9] . However , the most generally useful approaches are those that take any RNA ( or any multiple RNA alignment ) as a query and use an appropriate scoring system to search a sequence database and rank high-scoring similarities [10 , 11] , just as programs like Blast ( http://www . ncbi . nlm . nih . gov/BLAST/ ) do for linear sequence comparison [12] . In a general search program , one wants to score a combination of RNA sequence and structural conservation in a principled rather than an ad hoc manner . A satisfactory solution to this problem is known , using probabilistic models called stochastic context-free grammars ( SCFGs ) . SCFGs readily capture both primary sequence and ( non–pseudo-knotted ) RNA secondary structure conservation [13 , 14] . Just as hidden Markov models ( HMMs ) are useful for many different linear sequence modeling applications , including gene finding , multiple alignment , motif finding , and similarity search [14] , SCFGs are a generally useful paradigm for probabilistic RNA sequence/structure analysis , with applications including secondary structure prediction and gene finding . A particular SCFG architecture called covariance models ( CMs ) was developed specifically for the RNA similarity search problem [15] . CMs are profile SCFGs , analogous to the use of profile HMMs in sequence analysis [15 , 16] . The Rfam database of RNA families [17] is based on CM software ( Infernal [inference of RNA alignment]; http://infernal . janelia . org ) in much the same way that the Pfam database of protein families is based on profile HMM software ( HMMER; http://hmmer . janelia . org ) [18 , 19] . The most serious problem with using CMs has been their computational complexity . Applying standard SCFG dynamic programming ( DP ) alignment algorithms to the particular case of CMs results in algorithms that require O ( N3 ) memory and O ( LN3 ) time for a query of length N residues ( or consensus alignment columns ) and a target database sequence of length L . The memory complexity problem has essentially been solved , by extending divide-and-conquer DP methods ( the Hirshberg or Myers/Miller algorithm ) to the case of CMs [16] , but the time complexity problem still stands . Weinberg and Ruzzo [20–22] have described several filtering methods for accelerating CM searches . The original idea ( “rigorous filters” ) was to score a target sequence first by a linear sequence comparison method , using a profile HMM specially constructed from the query CM such that the profile score was provably an upper bound on the CM score; the subset of hits above threshold would then be passed for rescoring with the more expensive CM alignment algorithm [21] . Subsequently a “maximum likelihood heuristic” filter profile was developed that gives up the guarantee of recovering the same hits as the unfiltered search but offers greater speedups [22] . For most current Rfam models , Weinberg–Ruzzo filters give about a 100-fold speedup relative to a full CM-based search at little or no cost to sensitivity and specificity . However , because these filters depend on primary sequence conservation alone , they can be relatively ineffective for RNA families that exhibit poor sequence conservation—unfortunately , precisely the RNAs that benefit the most from SCFG-based search methods . Indeed , in this respect , we are concerned that the overall performance of rigorous filters on the current Rfam database may be somewhat misleading . Rfam currently uses a crude Blast-based filtering method to accelerate the CM searches used in curating the database . This step introduces a bias toward high primary sequence similarity in current Rfam alignments . As Rfam improves and incorporates more diverse structural homologs , the effectiveness of sequence-based filters will decrease . To address this worry , Weinberg and Ruzzo [20] have also described additional heuristics ( “sub-CMs” and the “store-pair” technique ) that should capture more secondary structure information in the filtering process . Bafna and coworkers [23] have described further improvements to sequence filtering methods . Currently , the Infernal codebase includes Weinberg's C++ implementation of rigorous filters but not , as yet , the ML heuristic , sub-CM , or store-pair methods . All these methods are important , but it also remains important to us to identify yet more methods for accelerating CMs . Here , we describe a method for accelerating CM searches using a banded DP strategy . In banded DP , one uses a fast method to identify a band through the DP matrix where the optimal alignment is likely to lie and then calculates computationally expensive DP recursions only within that band . In most cases , including our approach , banded DP is a heuristic that sacrifices guaranteed alignment optimality . Banding is a standard approach in many areas of sequence analysis . Gapped Blast uses banded DP to convert ungapped high-scoring pairs ( HSPs ) to full gapped alignments [12] . LAGAN and Multi-LAGAN ( http://lagan . stanford . edu ) use banded DP ( referred to as limited-area DP ) to stitch together alignments between anchored sequences when aligning long genomic sequences [24] . Banding has also been applied to profile SCFGs by Michael Brown in his RNACAD program by using information from a profile HMM alignment to define bands for the expensive SCFG alignment [25] . The key to developing a banded DP strategy is in deciding how the bands are identified . Usually , including all the examples just mentioned , banded DP involves performing some sort of rapid approximate sequence alignment between the query and the target . In contrast , the method we describe here , called query-dependent banding ( QDB ) , takes advantage of specific properties of CMs in order to predefine bands that are independent of any target sequence . QDB depends on the consensus secondary structure of the query , so it is complementary to acceleration methods such as the Weinberg–Ruzzo filters that rely on sequence but not structure .
CMs are a convention for mapping an RNA secondary structure into a treelike , directed graph of SCFG states and state transitions ( or , equivalently , SCFG nonterminals and production rules ) . The CM is organized by a binary tree of nodes representing base pairs and single-stranded residues in the query's structure . Each node contains a number of states , where one state represents the consensus alignment to the query , and the others represent insertions and deletions relative to the query . Figure 1 shows an example of converting a consensus structure to the guide tree of nodes and part of the expansion of those guide tree nodes into the CM's state graph . Here , we will only concentrate on the aspects of CMs necessary to understand QDB , and a subset of our usual notation . For full details on CM construction , see [16 , 26] . A guide tree consists of eight types of nodes . MATP nodes represent consensus base pairs . MATL and MATR nodes represent consensus single-stranded residues ( emitted to the left or right with respect to a stem ) . BIF nodes represent bifurcations in the secondary structure of the family , to deal with multiple stem-loops . A ROOT node represents the start of the model . BEGL and BEGR nodes represent the beginnings of a branch on the left and right side of a bifurcation , respectively . END nodes end each branch . The CM is composed of seven different types of states , each with a corresponding form of production rule , with notation defined as follows: That is , for instance , if state v is a pair state , it produces ( aligns to and scores ) two correlated residues , a and b , and moves to some new state , Y . The probability that it produces a residue pair a , b is given by an emission probability ev ( a , b ) . The probability that it moves to a particular state Y is given by a transition probability tv ( Y ) . The set of possible states Y that v may transit to is limited to the states in the next ( lower ) node in the guide tree ( and insert states in the current node ) ; the set of possible children states Y is called Cv , for “children of v . ” The indicators and are used to simplify notation in CM DP algorithms . They are the number of residues emitted to the left and right of state v , respectively . Bifurcation rules are special , in that they always transition to two particular start ( S ) states , at the root of subtrees in the guide tree , with probability 1 . 0 . These state types essentially define a “normal form” for SCFG models of RNA , akin to SCFGs in Chomsky normal form where all productions are in one of two forms , Y → a or Y → YY . We describe CM algorithms ( including QDB ) in terms of this normal form . CMs define a specific way that nodes in the guide tree are expanded into states and how those states are connected within each node and to states in the next node in the guide tree . For example , a MATP node that deals with a consensus base pair contains six states called MATP_MP ( a P state for matching the base pair ) , MATP_ML and MATP_MR ( an L and an R state for matching only the leftmost or rightmost base and deleting the right or left one , respectively ) , MATP_D ( a D state for deleting the base pair ) , and MATP_IL and MATP_IR ( L and R states with self-transitions , for inserting one or more residues to the left and/or right , respectively , before going to the next node ) . Thus , a CM is a generative probabilistic model of homologous RNAs . A sequence is emitted starting at the root , moving downward from state to state according to state transition probabilities , emitting residues and residue pairs according to emission probabilities , and bifurcating into substructures at bifurcation states . An important property of a CM is the states can be numbered from 0 . . . M − 1 ( from root to leaves ) such that for any state v , the states y that it can transit to must have indices y ≥ v . There are no cycles in a CM , other than self-transitions on insert states . This is the property that enables the recursive calculations that both CM DP alignment algorithms and QDB rely on . Without any change in the above description , CMs apply to either global or local alignment , and to either pairwise alignment to single RNA queries or profile alignment to a consensus query structure of a multiple RNA sequence alignment . CMs for single RNA queries are derived identically to profiles of a consensus structure , differing only in the parameterization method [27] . Local structural alignment to substructures and truncated structures ( as opposed to requiring a global alignment to the whole RNA structural model ) is achieved by adding state transitions from the ROOT that permit entering the model at any internal consensus state with some probability , and state transitions from any internal consensus state to an END with some probability [26 , 27] . Observe that for any state v , we could enumerate all possible paths down the model from v to the END ( s ) . Each path has a certain probability ( the product of the transition probabilities used by the path ) , and it will emit a certain number d of residues ( two per P state , one per L or R state in the path ) . The sum of these path probabilities for each d defines a probability distribution γv ( d ) , the probability that the CM subgraph rooted at v will generate a subsequence of length d . Given a finite limit Z on maximum subsequence length ( defined later ) , we can calculate γv ( d ) by an efficient recursive algorithm , working from the leaves of the CM toward the root and from smallest subsequences to largest: for v = M − 1 down to 0: v = end state ( E ) : v = bifurcation ( B ) : else ( v = S , P , L , R ) : For example , if we are calculating γv ( d ) where v is a pair state , we know that v must emit a pair of residues and then transit to a new state y ( one of its possible transitions Cv ) , and then a subgraph rooted at y will have to account for the rest of the subsequence of length d − 2 . Therefore , γv ( d ) must be the sum , over all possible states y in Cv , of the transition probability tv ( y ) times the probability that the subtree rooted at y generates a subsequence of length d − 2 , which is γv ( d − 2 ) , guaranteed to have already been calculated by the recursion . For the B state ( bifurcation ) calculation , indices y and z indicate the left and right S ( start ) state that bifurcation state v must connect to . A band dmin ( v ) …dmax ( v ) of subsequence lengths that will be allowed for each state v is then defined as follows . A parameter β defines the threshold for the negligible probability mass that we are willing to allow outside the band . ( The default value of β is set to 10−7 , as described later . ) We define dmin ( v ) and dmax ( v ) such that the cumulative left and right tails of γv ( d ) contain less than a probability : Larger values of β produce tighter bands and faster alignments , but at a cost of increased risk of missing the optimal alignment . β is the only free parameter that must be specified to QDB . Because CMs have emitting self-loops ( i . e . , insert states ) , there is no finite limit on subsequence lengths . However , we must impose a finite limit Z to obtain a finite calculation . Z can be chosen to be sufficiently large that it does not affect dmax ( v ) for any state v . On a digital computer with floating point precision ɛ ( the largest value for which 1 + ɛ = 1 ) , it suffices to guarantee that , for all v: Empirically , we observe that the tails of the γv ( d ) densities decrease approximately geometrically . We can estimate the mass remaining in the unseen tail by fitting a geometric tail to the observed density γv ( d ) . Our implementation starts with a reasonable guess at Z and verifies that the above condition is true for each v , assuming these geometrically decreasing tails; if it is not , Z is increased and bands are recalculated until it is . A QDB calculation needs to be performed only once per query CM to set the bands . Overall , a QDB calculation requires Θ ( MZ ) in time and space , or , equivalently , because both M and Z scale roughly linearly with the length L in residues of the query RNA , Θ ( L2 ) . The time and space requirement is negligible compared with the requirements of a typical CM search . A standard algorithm for obtaining the maximum likelihood alignment ( parse tree ) of an SCFG to a target sequence is the Cocke–Younger–Kasami ( CYK ) DP algorithm [28−30] . Formally , CYK applies to SCFGs reduced to Chomsky normal form , and it aligns to the complete sequence . The CM database search algorithm is a CYK variant , specialized for the “normal form” of our seven types of RNA production rules and for scanning long genomic sequences for high-scoring subsequences ( hits ) [14] . The CM search algorithm recursively calculates αv ( j , d ) , the log probability of the most likely CM parse subtree rooted at state v that generates ( aligns to ) the length d subsequence xj−d+1… xj that ends at position j of target sequence x [14 , 15] . This calculation initializes at the smallest subgraphs ( E states ) and shortest subsequences ( d = 0 ) and iterates upward and outward to progressively larger subtrees and longer subsequences up to a preset window size W . The outermost loop iterates over the end position j on the target sequence , enabling an efficient scan across a long target like a chromosome sequence . Banding is achieved simply by limiting all loops over possible subsequence lengths d to the bounds dmin ( v ) …dmax ( v ) derived in the band calculation algorithm , rather than all possible lengths 0…W . The banded version of the algorithm is as follows: For example , if we are calculating αv ( j , d ) and v is a pair state ( P ) , v will generate the base pair xj−d+1 , xj and transit to a new state y ( one of its possible transitions Cv ) , which then will have to account for the smaller subsequence xj−d+2… xj−1 . The log probability for a particular choice of next state y is the sum of three terms: an emission term log ev ( xj−d+1 , xj ) , a transition term log tv ( y ) , and an already calculated solution for the smaller optimal parse tree rooted at y , αy ( j – 1 , d – 2 ) . The value assigned to αv ( j , d ) is the maximum over all possible choices of child states y that v can transit to . The W parameter defines the maximum size of a potential hit to a model . Previous Infernal implementations required an ad hoc guess at a reasonable W . The band calculation algorithm delivers a probabilistically derived W for database search in dmax ( 0 ) , the upper bound on the length of the entire sequence ( the sequence generated from the root state of the CM ) . In our implementation , this algorithim is encoded in a more memory-efficient form that allocates space for only two sequence positions in j ( current and previous ) for most states rather than for all j = 0…L , using essentially the same techniques described for CYK search in [14] . We have omitted the necessary details here for clarity . QDB does not reduce the asymptotic computational complexity of the CM search algorithm . Both the banded algorithm and the original algorithm are O ( MW + BW2 ) memory and O ( L ( MW + BW2 ) ) time , for a model of M states containing B bifurcation states , window size W of residues , and target database length L . M , B , and W all scale with the query RNA length N , so roughly speaking , worst-case asymptotic time complexity is O ( LN3 ) . The subsequence length distributions calculated by QDB depend on the CM's transition probabilities . Transition probability parameter estimation is therefore crucial for obtaining predicted subsequence length bands that reflect real subsequence lengths in homologous RNA targets . Transition parameters in Infernal are mean posterior estimates , combining ( ad hoc weighted ) observed counts from an input RNA alignment with a Dirichlet prior [26] . Previous to this work , Infernal used an uninformative uniform Dirichlet transition prior , equivalent to the use of Laplace “plus-1” pseudo-counts . However , we found that transition parameters derived under a uniform prior inaccurately predict target subsequence lengths , as shown in an example in Figure 2 . The problem is exacerbated when there are few sequences in the query alignment , when the choice of prior has more impact on mean posterior estimation . To alleviate this problem , we estimated informative single component Dirichlet prior densities for CM transition parameters , as follows . The training data for transition priors consisted of the 381 seed alignments in the Rfam database , version 6 . 1 [17] . For each alignment , we built CM structures by Infernal's default procedure and collected weighted counts of observed transitions in the implied parse trees of the training sequences . Considering all possible combinations of pairs of adjacent node types , there are 73 possible distinct types of transition probability distributions in CMs . To reduce this parameter space , we tied these 73 distributions into 36 groups by assuming that certain distributions were effectively equivalent . Thirty-six Dirichlet densities were then estimated from these pooled counts by maximum likelihood as described in [31] , with the exception that we optimize by conjugate gradient descent [32] rather than by expectation–maximization ( EM ) . The results , including the Dirichlet parameters , are given in Table 1 . Using these priors for transition probability parameter estimation results in an improvement in the utility of QDB calculations , often yielding tighter , yet accurate subsequence length distributions , as illustrated by anecdotal example in Figure 2 . We also estimated informative mixture Dirichlet density priors for emission probabilities . Emission probabilities have no effect on QDB , but informative emission priors should improve sensitivity and specificity of CM searches , as they do for profile HMMs [31 , 33] . We collected filtered counts of aligned single-stranded residues and base pairs from annotated ribosomal RNA alignments from four alignments in the 2002 version of the European Ribosomal RNA Database [34 , 35]: large subunit rRNA ( LSU ) , bacterial/archaeal/plastid small subunit rRNA ( SSU-bap ) , eukaryotic SSU rRNA ( SSU-euk ) , and mitochondrial SSU rRNA ( SSU-mito ) . These alignments were filtered , removing sequences in which either less than 40% of the base-paired positions are present or more than 5% of the nucleotides are ambiguous , and removing selected sequences based on single-linkage clustering such that no two sequences in a filtered alignment were greater than 80% identical ( in order to remove closely related sequences ) . Summary statistics for the filtered alignments and collected counts in the training data set are given in Table 2 . These data were used to estimate a nine-component Dirichlet mixture prior for base pairs and an eight-component Dirichlet mixture prior for single-stranded residues . The base pair prior is given in Table 3 , and the singlet residue prior is given in Table 4 . The reason to use two different data sets to estimate transition versus emission priors is the following . Rfam provides many different structural RNA alignments but of uneven quality and varying depth ( number of sequences ) . The European rRNA database provides a small number of different RNA alignments but of high quality and great depth . A transition prior training set should be maximally diverse , so as not to bias any transition types toward any particular RNA structure , so we used the 381 different Rfam alignments for transitions . Emission prior estimation , in contrast , improves with alignment depth and accuracy but does not require broad structural diversity per se , so we used rRNA data for emissions . Inspection of the Dirichlet α parameters shows sensible trends . In the transition priors , transitions between main ( consensus ) states are now favored ( higher α values ) relative to insertions and deletions . In the base pair emission mixture prior , all components favor Watson–Crick and G-U pairs , with different components preferring different proportions of pairs in a particular covarying aligned column ( for instance , component 1 likes all four Watson–Crick pairs , component 2 describes covarying conservation of CG , UA , UG pairs , and component 3 specifically likes conserved CG pairs ) , and the mean α parameters prefer GC/CG pairs over AU/UA pairs . In the singlet emission mixture prior , some components are capturing strongly conserved residues ( component 1 favors conserved U's , for example ) while other components favor more variation ( components 4 and 5 , for example ) , and the marginal α parameters show a strong A bias , reflecting the known bias for adenine in single-stranded positions of structural RNAs ( especially ribosomal RNAs ) . There is redundancy between some components ( notably 5 and 8 in the base pair mixture and 2 , 3 and 8 in the singlet mixture ) . This is typical for statistical mixture estimation , which ( unlike , say , principal components analysis ) does not guarantee independence between components . The decision to use nine pair and eight singlet components was empirical , as these priors performed better than priors with fewer components on the benchmark we describe below ( unpublished data ) . Note that all singlet positions are modeled with one singlet mixture prior distribution , and all base pairs are modeled with one base pair mixture prior . These priors do not distinguish between singlet residues in different types of loops , for example , or between a stem-closing base pair versus other base pairs . In the future , it may prove advantageous to adopt more complex priors to capture effects of structural context on base pair and singlet residue preference . In another step to increase sensitivity and specificity of the program , we adopted the “entropy weighting” technique described for profile HMMs [36] for estimating the total effective sequence number for an input query alignment . This is an ad hoc method for reducing the information content per position of a model , which helps a model that has been trained on closely related sequences to recognize distantly related homologs [37] . In entropy weighting , one reduces the total effective sequence number ( which would normally be the actual number of sequences in the input alignment ) , thereby increasing the influence of the Dirichlet priors , flattening the transition and emission distributions , and reducing the overall information content . We approximate a model's entropy as the mean entropy per consensus residue , as follows . Let C be the set of all MATP_MP states emitting consensus base pairs ( a , b ) , and let D be the set of all MATL_ML and MATR_MR states emitting consensus singlets ( a ) ; the entropy is then calculated as: For each input multiple alignment , the effective sequence number is set ( by bracketing and binary search ) so as to obtain a specified target entropy . The target entropy for Infernal is a free parameter , which we optimized on the benchmark described below to identify our default value of 1 . 46 bits . To assess the effect of QDB , informative priors , and entropy weighting on the speed , sensitivity , and specificity of RNA similarity searches , we designed a benchmark based on the Rfam database [17] . The benchmark was designed so that we would test many RNA query/target pairs , with each query consisting of a given RNA sequence alignment , and each target consisting of a distantly related RNA homolog buried in a context of a random genome-like background sequence . We started with seed alignments from Rfam version 7 . 0 . In each alignment , sequences shorter than 70% of the median length were removed . We clustered the sequences in each family by single-linkage clustering by percent identity ( as calculated from the given Rfam alignment ) and then split the clusters such that the training set and test sequences satisfied three conditions: ( 1 ) no training/test sequence pair is more than 60% identical; ( 2 ) no test sequence pair is greater than 70% identical; and ( 3 ) at least five sequences are in the training set . Fifty-one families satisfy these criteria ( listed in Table 5 ) , giving us 51 different query alignments ( containing 5 to 1 , 080 sequences each ) and 450 total test sequences ( from 1 to 66 per query ) . We embedded the test sequences in a 1-Mb “pseudo-genome” consisting of twenty 50-kb “chromosomes , ” generated as independent , identically distributed ( iid ) random sequences with uniform base frequencies . The 450 test sequences were embedded into this sequence by replacement , by randomly choosing a chromosome , orientation , and start position , and disallowing overlaps between test sequences . The total length of the 450 test sequences is 101 , 855 nucleotides , leaving 898 , 145 nucleotides of random background sequence . The benchmark proceeds by first building a CM for each query alignment and then searching the pseudo-genome with each CM in local alignment mode . All hits above a threshold of 8 . 0 in raw bit score for each of the 51 queries were sorted by score into 51 ranked family-specific lists , as well as one ranked master list of all 51 sets of scores . Each hit is classified into one of three categories: “positive , ” “ignore , ” or “negative . ” A “positive” is a hit that significantly overlaps with a true test sequence from the same family as the query . An “ignore” is a hit that significantly overlaps with a test sequence from a different family , where “significantly overlap” means that the length of overlap between two sequences ( either two hits , or one hit and one test sequence embedded in the pseudo-genome ) is more than 50% of the length of the shorter sequence . ( Although it would be desirable to measure the false-positive rate on nonhomologous structural RNAs , we cannot be sure that any given pair of Rfam families is truly nonhomologous . Like most sequence family databases , Rfam is clustered computationally , and more sensitive methods will reveal previously unsuspected relationships that should not be benchmarked as “false positives . ” ) A “negative” is a hit that is not a positive or an ignore . For any two negatives that significantly overlap , only the one with the better score is counted . The minimum error rate ( MER ) ( “equivalence score” ) [38] was used as a measure of benchmark performance . The MER score is defined as the minimum sum of the false positives ( negative hits above the threshold ) and false negatives ( true test sequences that have no positive hit above the threshold ) , at all possible choices of score threshold . The MER score is a combined measure of sensitivity and specificity , where a lower MER score is better . We calculate two kinds of MER scores . For a family-specific MER score , we choose a different optimal threshold in each of the 51 ranked lists , and for a summary MER score , we choose a single optimal threshold in the master list of all hits . The summary MER score is the more relevant measure of our current performance , because it demands a single query-independent bit score threshold for significance . A family-specific MER score reflects the performance that could be achieved if Infernal provided E-values ( currently , it reports only raw bit scores ) . For comparison , BlastN was also benchmarked on these data using a family-pairwise search ( FPS ) procedure [39] . For each query alignment , each training sequence is used as a query sequence to search the pseudo-genome , all hits with an E-value of less than 1 . 0 were sorted by increasing E-value , and the lowest E-value positive hit to a given test sequence is counted . Using this benchmark , we addressed several questions about QDB's performance . What is the best setting of the single QDB free parameter , β , which specifies how much probability mass to sacrifice ? Figure 3 shows the average speedup per family and summary MER score as a function of varying β . There is no clear choice . The choice of β is a tradeoff of accuracy for speed . We chose a default of β = 10−7 as a reasonable value that obtains a modest speedup with minimal loss of accuracy . How well does QDB accelerate CM searches ? Figure 4 shows the time required for searching the 1-Mb benchmark target sequence with each of the 51 models , as a function of the average query RNA length . QDB reduces the average-case running time complexity of the CM search algorithm from LN2 . 36 to LN1 . 32 . Observed accelerations relative to the standard algorithm range from 1 . 4-fold ( for the IRE , iron response element ) to 12 . 7-fold ( for the 5′ domain of SSU rRNA ) , with an average speedup per family of 4 . 2-fold . In total search time for the benchmark ( sum of all 51 searches ) , the acceleration is 6-fold , because large queries have disproportionate effect on the total time . How much does QDB impact sensitivity and specificity ? Optimal alignments are not guaranteed to lie within QDB's high-probability bands . This is expected to compromise sensitivity . The hope is that QDB's bands are sufficiently wide and accurate that the loss is negligible . Figure 5 shows ROC plots ( sensitivity versus false-positive rate ) on the benchmark for the new version of Infernal ( version 0 . 72 ) in standard versus QDB mode . These plots are nearly superposed , showing that the loss in accuracy is small at the default QDB setting of β = 10−7 . How much do our changes in parameterization ( the addition of informative Dirichlet priors and entropy weighting ) improve sensitivity and specificity ? Figure 5 shows that the new Infernal 0 . 72 is a large improvement over the previous Infernal version 0 . 55 , independent of QDB . ( On average , in this benchmark , Infernal 0 . 55 is no better than a family-pairwise search with BlastN . ) Table 6 breaks this result down in more detail , showing summary and family-specific MER scores for a variety of combinations of prior , entropy weighting , and QDB . These results show that both informative priors and entropy weighting individually contributed large improvements in sensitivity and specificity .
CM searches take a long time , and this is the most limiting factor in using the Infernal software to identify RNA similarities . Prior to this work , Infernal 0 . 55 required 508 CPU-hours to search 51 models against just 1 Mb of sequence in our benchmarks ( Table 5 ) . Using QDB with β banding cutoffs that do not appreciably compromise sensitivity and specificity , Infernal 0 . 72 offers a 6-fold speedup , performing the benchmark in 85 hours . Our eventual goal is to enable routine genome annotation of structural RNAs: to be able to search thousands of RNA models against complete genome sequences . A search of all 503 Rfam 7 . 0 models against the 3-GB human genome with Infernal 0 . 72 in QDB mode would take on the order of 300 CPU-years ( down from 1 , 800 CPU years with Infernal 0 . 55 ) . We need to be able to do it in , at the most , a few days , so we still need to increase CM search speed by five orders of magnitude . Thus , the QDB algorithm is a partial but certainly not complete solution to the problem . However , QDB combines synergistically with other acceleration techniques . Parallelization , on large clusters ( although prohibitively expensive for all but a few centers ) , could give us further acceleration of three orders of magnitude . Software improvement ( code optimization ) will also contribute but probably only about 2-fold . Hardware improvements will contribute about 2-fold per year or so as long as Moore's law continues . Finally , QDB is complementary to the filtering methods recently described by Weinberg and Ruzzo [20−22] . We view QDB as part of a growing suite of approaches that we can combine to accelerate Infernal . Is it really worth burning all this CPU time in the first place ? Do CM searches identify structural RNA homologies that other methods miss ? Obviously we think so , but one would like to see convincing results . For large , diverse RNA families like transfer RNA , where a CM can be trained on more than 1 , 000 well-aligned sequences with a well-conserved consensus secondary structure , CM approaches have been quite powerful . The state of the art in large-scale transfer RNA gene identification remains the CM-based program tRNAscan-SE [1] , and CMs were also used , for example , to discover the divergent tRNA for pyrrolysine , the “22nd amino acid” [40] . But Figure 5 shows that on average , in more than 51 more or less “typical” RNA families of various sizes and alignment quality , Infernal 0 . 55 was actually no better than doing a family-pairwise search with BlastN . Until recently , we have spent relatively little effort on how Infernal parameterizes its models and relatively more on reducing its computational requirements [16] , so previous versions of Infernal have performed best where naive parameterization works best: on very large , high-quality alignments of hundreds of sequences , which are atypical of many interesting homology search problems . In this work , partly because the level of acceleration achieved by QDB is sensitive to transition parameterization , we have brought Infernal parameterization close to the state of the art in profile HMMs , by introducing mixture Dirichlet priors [31] and entropy weighting [36] . This resulted in a large improvement in the sensitivity and specificity of searches , as judged by our benchmark ( Figure 5 ) . The difference between Infernal and family-pairwise BlastN now appears pronounced for average-case behavior , not just best-case behavior . However , while we trust our benchmarking to tell us we have greatly improved Infernal relative to previous versions of itself , our benchmarking does not allow us to draw firm conclusions about our performance relative to other software . For that , we prefer to see independent benchmarks . Benchmarks by tool developers are notoriously biased , and however honest we may try to be , some biases are essentially unavoidable . For one thing , establishing an internal benchmark for ongoing code development creates an insidious form of training on the test set , because we accept code changes that improve benchmark performance . In particular , we set the entropy weighting target of 1 . 46 bits and the numbers of mixture prior components by optimizing against our benchmark . Further , our benchmark does not use a realistic model for the background sequence of the “pseudo-genome , ” because we construct the background as a homogeneous independent , identically distributed ( iid ) sequence , and this poorly reflects the heterogeneous and repetitive nature of genomic sequence . This benchmark should be sufficient for an internal comparison of versions 0 . 55 and 0 . 72 of Infernal , because we have not altered how Infernal deals with heterogeneous compositional bias . But we cannot safely draw conclusions from our simple benchmark about the relative performance of Infernal and Blast on real searches , for example , because Blast may ( and in fact does ) treat sequence heterogeneity better than Infernal does . In this regard , currently we are aware of only one independent benchmark BRaliBase III [41] . BRaliBase III consists of many different query alignments of five or 20 RNA sequences , drawn from three different RNA families ( U5 , 5S rRNA , and transfer RNA ) . These authors' results broadly confirm our internal observations: while Infernal 0 . 55 showed mediocre performance compared with BlastN and several other tools , a recent version of Infernal stood out as a superior method for RNA similarity search . Nonetheless , although Infernal 0 . 72 shows large improvements in speed , sensitivity , and specificity over previous versions , there are numerous areas where we need to improve further . A significant gap in our current implementation is that Infernal reports only raw bit scores and does not yet report expectation values ( E-values ) . CM local alignment scores empirically follow a Gumbel ( extreme value ) distribution [27] , just as local sequence alignment scores do [42] , so there are no technical hurdles in implementing E-values . This will be an immediate focus for the next version of Infernal . E-value calculations not only have the effect of reporting statistical significance ( more meaningful to a user than a raw bit score ) but also normalize each family's score distribution into a more consistent overall rank order , because different query models exhibit different null distributions ( particularly in the location parameter of the Gumbel distribution ) . We therefore expect E-values to contribute a large increase in performance whenever a single family-independent threshold is set . Table 6 roughly illustrates the expected gain , by showing the large difference between summary MER scores and family-specific MER scores . Parameterization of both CMs and profile HMMs remains problematic , because these methods continue to assume that training sequences are statistically independent , when in fact they are related ( often strongly so ) by phylogeny . Methods like sequence weighting and entropy weighting do help , but they are ad hoc hacks: unsatisfying and unlikely to be optimal . Even mixture Dirichlet priors , although they appear to be mathematically sophisticated , fundamentally assume that observed counts are drawn as independent multinomial samples , and therefore the use of Dirichlet priors is fundamentally flawed . Probabilistic phylogenetic inference methodology needs to be integrated with profile search methods . This is an area of active research [43−45] in which important challenges remain , particularly in the treatment of insertions and deletions . Finally , QDB is not the only algorithmic acceleration method we can envision . Michael Brown described a complementary banding method to accelerate his SCFG-based RNACAD ribosomal RNA alignment software [25] , in which he uses profile HMM-based sequence alignment to the target to determine bands where the more rigorous SCFG-based alignment should fall ( because some regions of the alignment are well-determined based solely on sequence alignment ) . The gapped Blast algorithm ( seed word hits , ungapped hit extension , and banded DP ) can conceivably be extended from two-dimensional sequence alignment to three-dimensional CM DP lattices . Developing such algorithms , and incorporating them into a widely useful , freely available codebase , are priorities for us .
The version and options used for Blast in our benchmark are WU-BLASTN-2 . 0MP -kap -W = 7 . For Infernal , versions 0 . 55 and 0 . 72 were used as indicated . The complete Infernal software package , including documentation and the Rfam-based benchmark described here , may be downloaded from http://infernal . janelia . org . It is developed on GNU/Linux operating systems but should be portable to any POSIX-compliant operating system , including Mac OS/X . It is freely licensed under the GNU General Public License . The ANSI C code we used for estimating maximum likelihood mixture Dirichlet priors depends on a copyrighted and nonredistributable implementation of the conjugate gradient descent algorithm from Numerical Recipes in C [32] . Our code , less the Numerical Recipes routine , is freely available upon request . | Database similarity searching is the sine qua non of computational molecular biology . Well-known and powerful methods exist for primary sequence searches , such as Blast and profile hidden Markov models . However , for RNA analysis , biologists rely not only on primary sequence but also on conserved RNA secondary structure to manually align and compare RNAs , and most computational tools for identifying RNA structural homologs remain too slow for large-scale use . We describe a new algorithm for accelerating one of the most general and powerful classes of methods for RNA sequence and structure analysis , so-called profile SCFG ( stochastic context-free grammar ) RNA similarity search methods . We describe this approach , called query-dependent banding , in the context of this and other improvements in a practical implementation , the freely available Infernal software package , the basis of the Rfam RNA family database for genome annotation . Infernal is now a faster , more sensitive , and more specific software tool for identifying homologs of structural RNAs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"none",
"computational",
"biology"
] | 2007 | Query-Dependent Banding (QDB) for Faster RNA Similarity Searches |
Serratia marcescens is an entomopathogenic bacterium that opportunistically infects a wide range of hosts , including humans . In a model of septic injury , if directly introduced into the body cavity of Drosophila , this pathogen is insensitive to the host's systemic immune response and kills flies in a day . We find that S . marcescens resistance to the Drosophila immune deficiency ( imd ) -mediated humoral response requires the bacterial lipopolysaccharide O-antigen . If ingested by Drosophila , bacteria cross the gut and penetrate the body cavity . During this passage , the bacteria can be observed within the cells of the intestinal epithelium . In such an oral infection model , the flies succumb to infection only after 6 days . We demonstrate that two complementary host defense mechanisms act together against such food-borne infection: an antimicrobial response in the intestine that is regulated by the imd pathway and phagocytosis by hemocytes of bacteria that have escaped into the hemolymph . Interestingly , bacteria present in the hemolymph elicit a systemic immune response only when phagocytosis is blocked . Our observations support a model wherein peptidoglycan fragments released during bacterial growth activate the imd pathway and do not back a proposed role for phagocytosis in the immune activation of the fat body . Thanks to the genetic tools available in both host and pathogen , the molecular dissection of the interactions between S . marcescens and Drosophila will provide a useful paradigm for deciphering intestinal pathogenesis .
A major arm of the Drosophila host defense against microbial infections is the systemic humoral response that consists primarily of the massive synthesis and release of potent antimicrobial peptides ( AMPs ) by cells of the fat body ( reviewed in [1 , 2] ) . The role of these AMPs in fighting infections has been previously demonstrated by biochemical and genetic assays [3 , 4] . The detection of invading microorganisms by host receptors of the Peptidoglycan recognition protein ( PGRP ) or Gram-negative binding protein ( GNBP ) families triggers the activation of signal transduction pathways in the fat body via the Toll and PGRP-LC receptors [5–12] . Genetic analysis has led to the delineation of two NF-κB-like signaling pathways , the immune deficiency ( imd ) and the Toll pathways , which control the expression of genes encoding the AMPs via activation of distinct NF-κB transcription factors . The imd pathway principally leads to the activation of Relish and transcription of genes mediating the response against Gram-negative bacteria . The Toll pathway , via activation of Dorsal-related immunity factor ( DIF ) , controls the expression of effector genes that mainly target fungi and Gram-positive bacteria . Barrier epithelia also produce AMPs . Their overall contribution to the Drosophila host defense had not been assessed , however , for lack of relevant infection models [13–15] . This situation has now changed thanks to studies on intestinal infection models . The intestine's function is to assimilate microbe-rich food such as decaying fruits while preserving the fly from infections . The midgut is protected by a chitinous physical barrier , the peritrophic matrix that lines the intestinal epithelium and contains microbes within the lumen of the digestive tract . This matrix is continuously synthesized by the cardia ( also known as the proventriculus ) , an elaborate structure at the entrance of the midgut that can also express AMPs such as Diptericin [14] . This expression of AMPs is enhanced by activation of the imd pathway and provides a partial protection against entomopathogenic bacteria such as Pseudomonas entomophila [16] . It complements another arm of the immune response to intestinal infections , namely the production of reactive oxygen species ( ROS ) , and is important since some yeasts and bacteria are resistant to host ROS [16 , 17] . To date , all studies on the intestinal immune response have been performed with bacteria that consistently remain confined to the digestive tract , whether entomopathogenic ( P . entomophila ) or phytopathogenic ( Erwinia carotovara carotovora ) [18 , 19] . Bacteria present in the digestive tract induce the imd pathway locally in the cardia and systemically in the fat body of larvae ( P . entomophila , E . carotovora ) and adults ( P . entomophila ) [16 , 18 , 19] . These bacteria are thought to release small peptidoglycan ( PGN ) fragments that pass the intestinal epithelium and bind to the membrane-bound PGRP-LC and hemolymphatic PGRP-LE microbial receptors that in turn activate the imd pathway in fat body cells [20] . At the same time , the system appears to be negatively regulated by the amidase PGRP-LB , which can break down PGN fragments into nonstimulatory molecules . Thus , the ingestion of a small PGN fragment , tracheal cytotoxin , triggers the imd pathway and the systemic immune response only in PGRP-LB mutant flies . Further , the response to ingested tracheal cytotoxin is blocked when recombinant PGRP-LB is injected directly into the hemolymph of the PGRP-LB mutants [20] . Some pathogens have developed strategies that allow them to gain access to target host tissues and that help them survive the host immune response . We have chosen to investigate the interactions between Drosophila and a potent entomopathogen , Serratia marcescens , as a model for understanding how flies handle invasive pathogens . This enterobacterium is a pathogen for many other host organisms , including plants and nematodes and is also an opportunistic pathogen of mammals ( [21 , 22] and references therein ) . As regards human health , S . marcescens is increasingly responsible for nosocomial infections in intensive care and neonatal units and it commonly infects people suffering from chronic granulomatous disease [23] . It was also the contaminant that led to the recent withdrawal of an influenza vaccine [24] . In the present study we have used S . marcescens Db11 , a streptomycin-resistant derivative of a strain originally isolated by Flyg and Boman from moribund flies [25 , 26] . Db11 is virulent when inoculated ( septic injury model ) , but is much less virulent when fed to the flies ( ingestion model ) . Genetic studies with S . marcescens have led to the isolation and identification of two bacterial strains that are almost avirulent in Drosophila after septic injury , Db1140 and 20C2 . Db1140 has pleiotropic defects compared to its Db11 parent: the activity of secreted proteases is strongly decreased [27]; it produces a truncated lipopolysaccharide ( LPS ) lacking the O-antigen [22] , and is nonmotile [28] . The 20C2 transposon insertion mutant also lacks the LPS O-antigen , but secretes proteases and is motile [22] . Both strains are highly attenuated in a Caenorhabditis elegans model of infection [22] . We have examined here the distinct pathogenic properties that S . marcescens displays upon ingestion as opposed to direct injection into Drosophila . We show that injected S . marcescens resists the systemic immune response essentially because of the O-antigen of its LPS . We find that ingested S . marcescens escapes from the digestive tract into the hemocoele and document its passage through the intestinal epithelium . Although the imd pathway does not protect the fly effectively against Db11 in the septic injury model , we found that S . marcescens is sensitive to the imd pathway–mediated local response in the gut . We demonstrate that phagocytosis is an effective defense against ingested S . marcescens that have escaped into the hemolymph but not against Db11 introduced directly into the hemocoele . Finally , we investigate why ingested Db11 does not induce systemically the imd pathway when present in the hemolymph .
Previous studies have indicated that S . marcescens Db11 is highly pathogenic in a Drosophila septic injury model [22 , 27] . We address here the molecular basis of S . marcescens virulence in this system and determine whether the classical arms of Drosophila host defense , namely the humoral and cellular responses , are efficient in fighting off this bacterium . As shown in Figure 1A , wild-type , Toll ( Dif ) , or imd ( kenny [key] ) pathway mutant flies died at the same rate , in less than a day , after a challenge with about 100 Db11 bacteria . We also did not observe a difference between key and wild-type flies after a challenge with a lower dose of about five bacteria ( unpublished data ) . The similar sensitivity of wild-type and mutant flies to Db11 was mirrored by the rate of bacterial growth in the hemolymph of infected flies ( Figure 1D ) . These data suggest that strain Db11 is resistant to the insect systemic immune response since immunodeficient flies do not display an enhanced susceptibility to Db11 . We next saturated phagocytosis in wild-type flies by injecting latex beads that are taken up , but not degraded , by hemocytes , thus effectively blocking this cellular defense mechanism [29 , 30] . We observed a similar death rate between latex bead–injected and noninjected flies after a Db11 challenge , indicating that the cellular immune response does not contribute significantly to host defense against this bacterium ( Figure 1E ) . One possible explanation for the virulence of S . marcescens is that this pathogen inhibits the activation of the Drosophila systemic immune response . This is not the case , however , because we detected the induction of the Diptericin gene , a classical readout of imd pathway activation , upon a challenge with the wild-type or mutant S . marcescens strains Db11 , 20C2 ( Figure 1F ) , and Db1140 ( unpublished data ) even though only about 100 bacteria were introduced into the insect body cavity . The expression of another imd-dependent AMP gene , CecropinA , was similarly induced ( unpublished data ) . Thus , the virulence of S . marcescens in the septic injury model is likely due to its ability to withstand the systemic immune response mediated by the imd pathway . Indeed , the stimulation of the systemic immune response either by a prior challenge with Escherichia coli or Enterobacter cloacae , or the ubiquitous overexpression of Diptericin , did not confer any protection to the flies against a subsequent challenge with Db11 ( unpublished data and Figure S1A ) . In keeping with these observations , S . marcescens Db11 was also resistant to the action of 200 μM Drosocin , Cecropin , and Defensin in an in vitro assay ( D . Rabel , personal communication ) , even though at such a high concentration these AMPs are highly effective against several other bacteria [3] . We next analyzed the pathogenicity of S . marcescens using bacterial mutants with an attenuated virulence . We have recovered several mutants that disrupt an operon involved in O-antigen biosynthesis from screens performed in C . elegans [22 , 31] . The 20C2 mutant was susceptible to the imd-dependent systemic immune response: key but not Dif or wild-type flies succumbed to 20C2 infection within 24 h ( Figure 1B ) . Accordingly , 20C2 bacteria grew rapidly only in the key mutant background ( Figure 1D ) . Nevertheless , this bacterial mutant ultimately killed 50% of the wild-type flies in approximately 5 d . Similar results were obtained with other O-antigen-deficient mutants bearing different disruptions of the same operon ( unpublished data ) . We also tested the Db1140 strain , which also lacks a LPS O-antigen , and found that it behaves like 20C2 in the septic injury model both in wild-type and imd pathway mutant flies . Serratia is commonly found in the environment and contaminates insects in the absence of injury in insectaries [21 , 32] . We reestablished a model of infection through an oral route [25] . We observed that wild-type flies feeding continuously on Db11 diluted in a sucrose solution were killed within 6 d ( Figure 2A ) . We found that the death rate of these infected flies varied with the bacterial load in the food solution and with temperature ( Figure S2 and unpublished data ) , whereas the death rate was negligible in flies feeding on sugar solution alone ( Figure 2C ) . Flies do not succumb to a secondary infection by bacterial commensals of the midgut , because flies feeding on Db11GFP ( a derivative of Db11 that expresses GFP and is resistant to ampicillin and streptomycin ) in the presence of ampicillin and streptomycin died at the same rate as control flies fed on Db11GFP alone ( unpublished data ) . The slow killing rate of Db11 , as compared to that observed in the septic injury model , is not due to the containment of Db11 within the digestive tract , since we were able to recover bacteria from the hemolymph of infected flies ( Figure 2D ) . The bacterial titer in the hemolymph increased slowly , as did the bacterial titer in the gut ( Figure 2E ) . These data suggest that Db11 can rapidly escape from the digestive tract into the internal body cavity of the host , yet fails to kill it rapidly . We then examined the infectious process by microscopy using GFP- or DsRed- labeled bacteria , concentrating initially on the first 2 d of the infection . We established the presence of S . marcescens along the whole length of the digestive tract ( Figure 3A ) . During the initial stage of the infection , midgut morphology appeared normal when observed at low magnification , with bacteria mainly confined to the lumen . We could observe bacteria penetrating the deep invaginations of the acid-secreting copper cells [33] after 48 h of infection ( Figure S3 ) . Even though the bacteria remain topologically outside of the epithelial layer , they have nevertheless crossed the peritrophic matrix during this period . Although the integrity of the peritrophic matrix appeared preserved at the ultrastructural level in our ultrathin sections ( Figure 4A and 4B ) , we cannot exclude that it is locally ruptured . Once they had crossed the intestinal epithelium , bacteria were sometimes observed in the midgut muscles and surrounding tracheoles ( Figure S4 ) . To determine whether Db11 pass through or in between intestinal epithelial cells , we used higher concentrations of bacteria and observed the location of Db11 in fixed tissues either as whole mount preparations or on frozen sections . We occasionally observed bacteria in the anterior midgut and cardia that appeared to be intracellular in whole mount preparations ( Figure 3G ) . More often , we observed bacteria localized close to the apical parts of the epithelial cells ( Figure S5 ) or even entering these cells ( Figure S5D ) . In frozen sections , we relied on DAPI staining , as DsRed fluorescence was often strongly reduced . Figure 3H shows the intracellular distribution of bacteria in wild-type ( 1 . 83 bacteria on average per section ) and key mutants ( 4 . 29 bacteria on average per section ) . We seldom observed bacteria crossing the intestinal barrier in wild-type flies by electron microscopy . In the rare cases where we did detect bacteria within the epithelium during early infection , the bacteria were detected at an intracellular location , within a vacuole ( Figure 4D ) . To increase the number of observed events , we analyzed immunocompromised imd pathway mutant flies ( key ) , since they display a higher number of intracellular Db11 ( Figure 3H ) . We detected many vacuoles that evoked autophagic bodies ( Figure 4E ) in intestinal cells . Other vacuoles enclosed bacteria . These data establish that S . marcescens can invade intestinal epithelial cells . By 72 h of infection , the digestive tract was distorted with more bacteria concentrated in the lumen , which was dilated in some places ( Figure 3A and 3B ) . At this stage of the infection , the structure of the midgut began to be affected , and the epithelial lining appeared very thin ( Figure 3B; compare Figure 3C and 3E; 3D and 3F ) . The alteration of the midgut epithelium was obvious at the ultrastructural level: the cytoplasm of intestinal cells appeared very different from that of control epithelial cells ( Figure 4C ) . It was characterized by the presence of numerous small electron-translucent vacuoles , as noted previously in Shigella-infected Henle cells , indicating an important cellular stress [34] . The cells did not usually exhibit any of the hallmarks of impending apoptosis , such as nuclear fragmentation or a homogenous cytosol with a low number of organites , nor of necrosis . We reasoned that bacterial proteases might be responsible for the progressive degradation of gut structure during infection . We therefore tested the Db1140 strain , which is partially deficient in protease activity , and possibly for other functions [27] . This strain is sensitive to the imd-dependent systemic response because , like 20C2 , it lacks the LPS O-antigen ( Figures 1C and 2F ) . In the ingestion model , Db1140 failed to kill wild-type flies ( Figure 7D ) , unlike Db11 and 20C2 ( Figure 2 ) . We observed that gut integrity was preserved in flies that were fed on Db1140 ( Figure S6 ) . Yet , we found in rare instances that these bacteria still appeared to be taken up by intestinal cells ( Figure S6B , arrow ) . In addition , Db1140 was transiently recovered from the hemolymph in immunodeficient flies , which slowly succumbed to the infection ( Figure 7G ) . Taken together , these data suggest that the degradation of the gut contributes to the lethal outcome of the oral infection . To delineate the host response to intestinal infections , we first determined whether the Drosophila signal transduction pathways regulating the systemic immune response to septic injury are also involved in the host defense in our ingestion model . We found that Dif mutant flies [35] , in which induction of Toll pathway target genes is impaired , displayed an increased susceptibility to oral infections with S . marcescens Db11 ( median survival time to death [LT50] 5 . 0 d ) when compared to wild-type flies ( LT50: 5 . 7 d ) ( Figure 2A and 2B ) . However , the susceptibility to the ingestion of Db11 was more pronounced with key ( kenny ) mutants ( LT50: 4 . 0 d ) in which the imd pathway is defective [36] , even though Db11 is resistant to the key-dependent systemic immune response in the septic injury model . Similar results were found with other mutants of the imd pathway , including imd , Relish , DREDD , and FADD , while PGRP-LC and PGRP-LE; PGRP-LC mutants exhibited a mild phenotype ( Figure S7 and unpublished data ) . The sensitivity of key mutants was correlated at 72 h post-infection with a reproducible 10-fold increase in the number of bacteria retrieved from the hemolymph , as compared to wild-type ( Figure 2D ) . key mutants displayed an altered epithelial midgut morphology 24 h earlier than wild-type flies ( Figure 3A ) , suggesting that the imd pathway helps control the deleterious effects of Db11 in the midgut , even though there was no difference between the number of Db11 bacteria recovered from key mutant and wild-type midguts ( Figure 2E ) , possibly because the flies constantly feed on fresh bacteria . To determine whether the imd pathway is involved in the host response to intestinal infections , we next analyzed the behavior of mutant bacteria that are sensitive to the action of the imd pathway in the septic injury model ( Figure 1 ) . Wild-type flies feeding on 20C2 bacteria died with a 1-d delay as compared to Db11 ( LT50: 6 . 7 d; Figure 2B ) . 20C2 bacteria were detected only transiently in the hemolymph at 24 h ( before increasing again from 72 h onwards ) , suggesting that they are also susceptible to the host response to intestinal infections ( Figure 2D ) . As in the septic injury model , 20C2 was , however , retrieved in large amounts in the hemolymph of key mutant flies ( Figures 1D and 2D ) , indicating that the imd pathway either controls the growth of the LPS-defective 20C2 in the hemocoele or hinders its escape from the gut . Because a systemic response in the fat body is triggered by some intestinal pathogens [16 , 37] , we suspected that that it might account for the enhanced susceptibility of imd pathway mutants to Db11 oral infections . We therefore measured this response in flies that have ingested Db11 or 20C2 and detected no induction of the imd-dependent AMP genes Diptericin and Cecropin in whole flies ( Figure 5A and unpublished data ) . We also observed no induction of AMP genes in the fat body using a set of GFP reporter genes or a Diptericin-lacZ transgene ( Figure 5B and unpublished data ) . This lack of a systemic response is not due to an active inhibition by ingested S . marcescens since an additional septic injury by E . coli led to a normal induction of Diptericin expression ( Figure 5C and unpublished data ) . Flies that had ingested Db11 also resisted a septic wound with E . coli as robustly as flies feeding on sugar solution alone ( Figure 5D ) . Altogether , our experiments revealed that ingested Db11 ( and 20C2 ) fail to elicit a response when they gain access into the hemolymph from the gut . In the absence of a systemic response , we searched for a local immune response in the midgut to account for the role of the imd pathway in host defense . We found that Diptericin transcription was induced in large segments of the midgut by uptake of S . marcescens , as judged by the expression of lacZ and GFP reporter genes ( Figure 6A , 6B , and 6E ) . Importantly , Diptericin expression was reduced in PGRP-LC mutants and absent in key mutants , indicating that the imd pathway is involved in the inducible expression of this AMP in the gut ( Figure 6C and 6D ) . Among several AMP reporter genes tested ( Table S1 ) , Diptericin-GFP was the only one that displayed a Db11-induced expression in the digestive tract . We observed , however , the induction of other AMP reporter genes in various epithelial tissues ( Table S1 ) . We next set out to determine whether this local activation of the imd pathway in the cardia and the midgut is sufficient to confer some protection against ingested Db11 . As shown in Figure 6F , key flies that expressed in the midgut either the Diptericin gene or the wild-type copy of key , displayed respectively a partially or totally rescued phenotype of survival to ingested Db11 , as compared to key and wild-type flies . Note that the midgut Gal4 drivers we used for these experiments ( NP1 , NP3084 ) led to a level of Diptericin expression that was ten times higher than that observed during a Db11 oral infection in wild-type flies ( Figure 6G ) . In keeping with a role for the imd pathway in the midgut , overexpression of Diptericin alone in wild-type flies increased resistance to intestinal Db11 infection ( Figure S1B ) . Together , these experiments confirmed the antimicrobial capacity of Drosophila Diptericin in vivo , which had not been revealed in the context of the systemic immune response ( Figure S1A ) [4 , 16] . Db11 has the ability to traverse the gut barrier and yet does not induce a systemic immune response . Since these bacteria were observed to be either free in the hemolymph , or attached to or engulfed by hemocytes ( Figure 7A ) , we evaluated the contribution of phagocytosis to the host defense in this model of infection using the latex bead saturation technique . Latex bead–injected flies showed a markedly enhanced susceptibility when fed Db11 , 20C2 , or Db1140 ( Figure 7B–7D ) . This is unlikely to be due to a nonspecific effect of the latex beads since injection of cytochalasin D , which blocks phagocytosis , also leads to enhanced sensitivity to Db11 oral infection ( Figure S8 ) . The premature death of infected latex bead–injected flies correlated with the presence of a number of Db11 and 20C2 bacteria in the hemolymph that was almost two orders of magnitude higher than that in nontreated flies ( Compare Figure 7E and 7F to Figure 2D ) . In contrast to Db11 and 20C2 , Db1140 bacteria fail to kill their host , and few Db1140 bacteria were retrieved from the hemolymph of nontreated wild-type flies ( Figure 7G ) . In contrast , key mutants in which phagocytosis had been blocked beforehand were killed by Db1140 and many bacteria where detected in the hemolymph ( Figure 7G ) . Taken together , these data indicate that phagocytosis plays a vital role in the control of ingested S . marcescens that have gained access to the hemolymph . In the absence of this cellular immune response , the apparent proliferation of bacteria correlated with the advent of a strong systemic immune response as observed in latex bead–injected flies ( Figure 7H ) . At late stages of the infection , more than 1 , 500 bacteria could be retrieved from the hemolymph of a single fly , yet no systemic induction of the imd pathway could be detected . This lack of a response is not due to a number of bacteria that would be below a detection threshold since the injection of about 1 , 500 E . coli bacteria is sufficient to trigger a strong expression of Diptericin ( unpublished data ) . Because the PGRP-LC/PGRP-LE sensing system is able to detect small PGN fragments released during bacterial growth and division [11 , 38] , we wondered whether the bacteria that have traversed the intestine proliferate in the hemocoele . Indeed , the slow accumulation of S . marcescens in the hemolymph of flies feeding on Db11 could be due to bacterial proliferation . A second hypothesis is that the increase might be due to the continuous passage of bacteria from the intestine to the hemolymph . To discriminate between these possibilities , we first fed flies on Db11-GFP bacteria for 24 h . We then switched the flies to a food source that contained only Db11-DsRed . We monitored the number of green and red bacteria recovered both from the gut and from the hemolymph of infected flies ( Figure 8A and 8B ) . As expected , the number of green Db11 in the digestive tract was overtaken by that of red Db11 bacteria ( Figure 8A ) . Yet , the number of green bacteria did not decrease but remained stable . Strikingly , the count of green bacteria present in the hemolymph remained stable , whereas there was a steady increase in the number of Db11-DsRed ( Figure 8B ) , thus indicating that the net increase of the number of bacteria in the hemolymph during intestinal infection is mostly due to bacteria that have traversed the digestive tract . These data , however , do not exclude the possibility of an equilibrium in the hemolymphatic compartment between bacterial proliferation on the one hand and phagocytosis on the other . Consistent with this hypothesis , we found that both GFP and DsRed-labelled bacteria numbers increased by several orders of magnitude within 48 h when we repeated this experiment in flies that had been previously injected with latex beads to saturate their phagocytic apparatus ( Figure 8B ) . Two hypotheses can account for this result . One possibility is that the observed number of bacteria reflects a high rate of passage from the gut that is effectively counterbalanced by the cellular response when not inhibited by latex beads . The alternative is that this high number of bacteria results from bacterial division within the hemocoele that is not controlled when phagocytosis is blocked . To test these possibilities , we blocked the intestinal supply by feeding flies on a sucrose solution containing gentamicin following 1 d ( or 4 h ) of feeding on Db11 . The treatment was effective in killing Db11 in the gut ( Table S2 and unpublished data ) . In the treated flies , hardly any bacteria were recovered from the hemolymph . The decreased bacterial concentration in the hemolymph of gentamicin-treated flies is not the result of antibiotics treatment but of elimination of these bacteria by phagocytes , because we detected a large hemolymphatic bacterial count in gentamicin-treated flies in which phagocytosis had been blocked by latex beads injection ( Table S2 and unpublished data ) . Importantly , this latter experiment in latex bead–injected flies shows that bacterial proliferation occurs in the hemolymph when phagocytosis is blocked and the intestinal reservoir depleted by antibiotics treament . Taken together , these data indicate that the high number of bacteria retrieved from the hemolymph of flies in which phagocytosis has been blocked results from bacterial division , which thus correlates with the elicitation of the imd pathway in the fat body .
In the septic injury model , Db11 kill wild-type and imd pathway mutant flies at the same speed . In contrast , LPS-defective strains are less virulent in wild-type flies but regain their full virulence when introduced in imd immunocompromised flies ( Figure 1B and 1C ) . These data demonstrate that a major determinant of Db11′s virulence is its ability to resist the systemic immune response . The resistance of Db11 could be caused by an ability to degrade AMPs or by a capacity to resist the attack of these antimicrobial peptides [16 , 27] . Since both S . marcescens Db1140 and 20C2 LPS-deficient mutants are sensitive to the systemic immune response , and since 20C2 , in contrast to Db1140 , is not impaired in protease secretion , it follows that resistance to AMP action is the major factor that determines virulence of Db11 in Drosophila after a septic injury . Previous experiments by Flyg and Xanthopoulos also support this conclusion . These authors determined that only Db1140 has a strongly attenuated virulence in comparison to its parent strain Db1121 , which is also impaired in protease secretion but presumably has a normal O-antigen [27] . Taken together , these data indicate that the ability of S . marcescens Db11 to withstand AMP attack in vivo depends primarily on the presence of its O-antigen , in keeping with a similar phenomenon described for Shigella [39] . This may represent a novel mechanism of resistance to the action of AMPs [40] . It remains to be determined whether this effect is due to the structure of the LPS-O-antigen or whether the O-antigen is required to anchor or stabilize a putative microbial effector that would neutralize AMPs . This resistance mechanism , however , is not effective in the midgut environment . One important question is that of the passage through the intestinal tract . We have been able to detect Db11 bacteria within midgut cells at a low frequency in wild-type flies during early infection . This rate increased significantly in key mutants . Interestingly , these bacteria were always observed inside vacuoles , in keeping with observations in human bladder cells that have internalized S . marcescens [41] . Some , or all , of these bacteria may be destroyed during this intracellular stage , as we could observe in key mutants many vacuoles containing debris of unknown origin . One possibility to account for the increased number of bacteria observed intracellularly in the key midgut epithelium is that many bacteria are prevented from reaching the midgut epithelium by the imd-dependent local immune response in wild-type flies . Alternatively , bacteria may access the wild-type or key gut epithelial cells at the same rate and be eliminated more efficiently in wild-type flies within the epithelial cells by the joint action of the imd response and a putative intracellular defense mechanism . We have not yet observed by electron microscopy bacteria entering or exiting the intestinal epithelium . Thus , we cannot formally rule out that all intracellular bacteria are killed and that the actual passage occurs in between cells as opposed to an intracellular route . Presumably , the passage between cells would involve proteases to disrupt the junctions between adjacent cells . In this respect , we note that Db1140 bacteria that secrete a greatly reduced level of proteases are still able to traverse the intestinal barrier ( Figure 7G ) and can be detected intracellularly ( Figure S6 ) . This observation suggests that S . marcescens has an inherent ability to cross intact epithelia through the cells . We have failed to detect bacteria crossing the junctions that seals the epithelium at early stages of the infection . This mode of crossing may , however , be used at later times of the infection , when the integrity of the midgut appears to be severely affected . Consistent with this idea , the imd pathway–sensitive 20C2 bacteria are present in the hemolymph at a low level for 72 h , and then their number increases strongly in this compartment ( Figure 2D ) . We propose that this increased passage results from an augmented “permeability” to bacteria of the intestinal epithelium that becomes compromised , possibly by secreted bacterial proteases . Our experiments with Db11 , and especially the imd pathway–sensitive strain 20C2 that regains wild-type virulence in key mutant flies , confirm and further document the importance of the local immune response in the midgut in the defense against intestinal infections ( Figure 1 ) [16 , 17] . First , the imd pathway is activated in the midgut and , in contrast to previous studies [16 , 18] , we find that the induction of the imd pathway is not limited to the cardia but extends to large portions of the midgut . Second , the bacterial load of imd pathway–sensitive 20C2 is lower than that of Db11 at 96 h of infection of the wild-type , whereas both bacterial titers are similar in a key mutant background ( Figure 2E ) . Third , we could not detect any induction of the systemic immune response in the fat body , in contrast to other infection models [18 , 19] . Finally , the key mutant phenotype could be fully rescued by expressing a transgene only in the midgut . While Db11 is resistant to the strong ubiquitous expression of Diptericin in the septic injury model ( Figure S1A ) , it is sensitive to some degree to a strong expression of this AMP when present in the midgut ( Figure 6F ) . However , since even under these nonphysiological conditions Diptericin is not able to provide full protection , the imd pathway must control the expression of other effectors of the gut local response . These additional defenses might include proteases , lysozymes , and nitrogen oxide production , as well as unidentified molecules . Even though we could not detect the induction of other AMP genes with our set of GFP reporter genes , we cannot exclude the possibility that Cecropin and Attacin are induced to some extent in our system , as described for other oral infection models [16 , 17] . The imd response may be potentiated or act in conjunction with the oxidative burst induced independently of the imd pathway by bacterial feeding [42 , 43] . Indeed , recent data indicate a partial sensitivity of flies in which the expression of the Duox gene , which mediates the oxidative burst , is decreased by transgenic RNAi ( B . Kele , unpublished data ) . In this context , one can also note the presence of a catalase and of three SOD genes in the Db11 genome , which may mediate the resistance to this host defense ( P . Giammarinaro and J . Ewbank , unpublished data ) . Our studies extend to Drosophila the concept that the epithelial response is an important and ancient aspect of host defense against infection [44–47] . Phagocytosis has often been described as playing an ancillary function in host defense against systemic bacterial infections in Drosophila [30 , 48] . This cellular process , however , plays a primordial role in our oral infection model , as it controls the proliferation of S . marcescens bacteria that have escaped from the alimentary canal . When phagocytosis is inhibited with latex beads , we observe septicemia , which is likely to cause the demise of the infected fly given the high bacterial load at the time of death . Indeed , a similarly high titer of bacteria is observed in the septic injury model . The importance of phagocytosis in the host defense against Db11 oral infections has been established in a separate study that used a mutant line defective for a novel phagocytic receptor gene , eater , where essentially the same results were obtained in terms of survival and Db11 bacterial growth [49] . The cellular response and the local response complement each other . Indeed , we observed that the effects of imd pathway and phagocytosis inactivation are additive ( Figure 7B and 7C ) . This effect has been confirmed in mutants doubly defective for the IKKγ homolog , KEY , and for Eater ( N . Nehme , unpublished data ) . Similarly , the protease-deficient Db1140 strain , which is also sensitive to the action of the imd pathway , could only proliferate in the hemolymph of key flies in which phagocytosis had been inactivated ( Figure 7G ) . A striking finding from the present study is that bacteria that have escaped into the body cavity of orally infected flies do not kill their host rapidly , in contrast to the septic injury model wherein flies succumb to infection by 100 bacteria in less than a day . S . marcescens is sensitive to phagocytosis in the ingestion model , while it is apparently not susceptible to it in the septic injury model . These findings suggest that S . marcescens does not express the same virulence program in both models , possibly as a result of its exposition to midgut defenses . Consequently , bacteremia is unlikely to be the cause of lethality of wild-type flies fed on Db11 , as the bacterial titer in the hemolymph is lower by two orders of magnitude than that measured in the septic injury model or when phagocytosis is blocked ( compare Figure 2D to Figures 1D and 7E ) . Rather , we surmise that the death of flies is due to the severe degradation of the midgut epithelium . The progressive thinning of the midgut epithelium is reminiscent of that observed in C . elegans infections; however , in that model , the bacteria are unable to escape from the nematode intestine [22] . Flies or nematodes infected with Db1140 , a strain displaying a reduced production of proteases , display an apparently normal structure of the intestinal epithelium and do not succumb to the infection , suggesting that bacterial proteases are involved in the attack of the intestinal epithelium . The presence of S . marcescens in the digestive tract following oral infection does not induce a systemic immune response . One possibility is that S . marcescens does not proliferate enough in the gut to release small PGN fragments in quantities sufficient to overcome the immune-suppressive action of PGRP amidases in the gut and hemocoele that thus prevent the systemic activation of the imd pathway [20 , 50] . An unexpected finding is that S . marcescens bacteria present in the hemocoele after passage through the gut fail to elicit the systemic response , whereas they do when introduced by septic injury . This situation may be similar to that observed in domino mutant larvae , which lack hemocytes and harbor microorganisms in their hemolymph that also do not elicit the systemic immune response [48] . The lack of a systemic response to the presence of bacteria in the hemolymph is not due to the absence of a wounding response in this model of infection , because a clean injury performed on flies that have ingested Db11 did not induce the imd pathway any more than in flies fed on a sugar solution ( unpublished data ) . In contrast , a marked systemic immune response was observed following a block of phagocytosis ( Figure 8C ) . One supposition is that bacteria in the hemolymph fail to grow and divide actively and thus do not release PGN fragments . This hypothesis would explain why the net increase in bacterial number in the hemolymph is mostly due to bacterial passage from the gut ( Figure 8B ) . However , to account for the stimulation of the systemic response when phagocytosis is blocked , one would have to hypothesize that phagocytosis somehow inhibits bacterial growth and division; this inhibition would be relieved upon ingestion of latex beads . We have so far failed to obtain any direct evidence in support of Db11 being in a dormancy state [51 , 52] . Alternatively , Db11 bacteria divide actively in the hemolymph and are phagocytosed at a rate that approximately equals that of divisions , since the net increase of the bacterial number in this compartment appears to be mostly due to bacteria transferred from the gut ( Figure 8 ) . The continuous passage of bacteria from the intestine would compensate the bacterial loss due to phagocytosis . In the absence of a supply of fresh bacteria from the intestine , the bacteria are cleared from the hemolymph , as observed in the gentamicin experiments . We therefore propose that phagocytosis of growing and dividing bacteria keeps the levels of PGN fragments below the threshold of detection . In larvae , it has been proposed that hemocytes are required to signal the presence of bacteria to the fat body to trigger the systemic immune response , either by emitting a cytokine or by releasing PGN fragments after phagocytosis [18 , 53] . This model is appealing because the PGN layer of the Gram-negative cell wall is not directly accessible and is buried under the outer membrane and a LPS shell , and thus some sort of cellular processing might be required to uncover the PGN polymers . Our data , however , argue against such a model in adults , since we observe a systemic response only when phagocytosis is blocked . Rather , our results are best accounted for by the release of short PGN fragments during bacterial growth and proliferation . Furthermore , in a septic injury model , we observed a sustained activation of the imd pathway in flies in which phagocytosis was impaired by the prior injection of latex beads ( N . Nehme , unpublished data ) . The immunity of mucosal surfaces , especially that of the intestinal epithelium , is the focus of intense scrutiny [54] . The human digestive tract , however , is complex since it harbors more than 400 distinct microbial species and is protected by both innate and adaptive immune responses [55] . In contrast , Drosophila provides a simple and powerful model that allows the dissection of the innate immune responses in the digestive tract . In addition , these studies can be performed at the whole organism level , as exemplified by the possibility of investigating phagocytosis [49] as a complement to intestinal defenses . Because it is able to cross the fly intestinal barrier , and because it is resistant to some of the host immune responses , S . marcescens Db11 constitutes an attractive model for the in vivo study of enteric pathogenesis . Interestingly , the treatment of cancer patients by chemotherapy leads to neutropenia and associated bacterial translocation and bacteremia , a striking parallel to the mechanism we describe in phagocytosis-deficient flies infected orally with S . marcescens . Further , this microorganism is amenable to genetic analysis and manipulation [56] and its genome has recently been sequenced ( http://www . sanger . ac . uk/Projects/S_marcescens/ ) . The stage is now set for a thorough investigation of the host–pathogen relationships between Drosophila and S . marcescens from the vantage of both the fly and that of the bacterium .
The nonpigmented S . marcescens strain Db10 was isolated originally from a moribund fly ( Db is for Drosophila bacterium ) ; Db11 is a spontaneous streptomycin-resistant mutant of Db10 [25] . Db1121 was derived from Db11 following two rounds of chemical mutagenesis and selection for decreased secretion of proteases . Db1140 is a spontaneous mutant derived from Db1121 that is resistant to phage ΦJ [27] . The Db11 miniTn5Cm insertion mutant 20C2 was described in Kurz et al . [22] . Since its original isolation , the locus affected in 20C2 has been better characterized thanks to whole genome sequencing . The insertion site is at genomic position 914398 and disrupts SMA0873 involved in O-antigen biosynthesis ( http://www . sanger . ac . uk/Projects/S_marcescens/ ) . The O-antigen biosynthesis operon spans genes SMA0868 to SMA0879 . Other mutants in this operon have been isolated from a library of Tn5 insertion mutants generated in Db10 bacteria ( Db11 is a streptomycin-resistant strain derived from Db10; Db10 behaves as Db11 in oral and septic injury infection assays ( N . Nehme , unpublished data; [25] ) ) . These mutants correspond to transposon insertions into SMA082 , SMA0873 , and SMA0876 . GFP or DsRed derivatives of Db11were obtained by transformation with plasmids pUFR-GFP ( ampicillin and gentamicin resistance ) or pEP933 ( tetracyclin and gentamicin resistance ) , respectively . The DsRed- and GFP-labelled S . marcescens transformants behaved as their cognate strain in both Drosophila infection models . Strains were grown in LB ( Luria Bertani medium ) at 37 °C . When required , antibiotics were added at the following concentrations: ampicillin , 100 μg/ml ( Db11GFP ) ; streptomycin , 100 μg/ml; chloramphenicol , 30 μg/ml; and gentamicin , 10 μg/ml . Stocks were raised on standard corneal-agar medium at 25 °C . cn bw flies were used as wild-type for most of the experiments since key1 and Dif1 mutants were generated in this background [57] . Dredd , FADD , Relish , and PGRP-LE mutant strains were respectively described in [58–61] . The A5001 strain is the wild-type strain that was used to generate the PGRP-LC mutant strain [6] . Where indicated , we used an Oregon-R stock as wild-type control . The UAS-Diptericin line has been described previously [4] . The UAS-key and pTEP-GFP lines were kind gifts of Sophie Rutschmann and Daniel Doucet , respectively . pTEP-GFP line was generated using the promoter of the TEP1 gene fused to GFP; this construct is specifically expressed in hemocytes ( unpublished data ) . All the mutant lines are described on the Flybase Web site ( http://flybase . bio . indiana . edu/ ) . The NP1 and NP3084 lines were selected in a screen of enhancer trap Gal4 lines expressed in embryonic and/or larval gut tissues and available from the Drosophila Genetic Resource at the National Institute of Genetics ( Shizuoka , Japan; http://www . shigen . nig . ac . jp/fly/nigfly/ ) . These strains were selected for their strong and specific expression in the midgut of adult flies . No expression has been observed in the malpighian tubules or other surrounding tissues . The key rescue lines were constructed by standard genetic crosses . Survival experiments were performed as previously described [62] . Briefly , batches of 20–25 wild-type and mutant flies were challenged by septic injury using a needle previously dipped in a concentrated solution of E . coli . As regards S . marcescens , an overnight culture was diluted to an optical density at 600 nm ( OD600 ) of 0 . 1 in LB . This leads to the inoculation of 50–100 bacteria per fly , as checked by performing colony counts on flies crushed right after septic injury . The vials containing the challenged flies were then put in an incubator at the desired temperature and the surviving flies counted every few hours . Each experiment shown is representative of at least three independent experiments . Statistical tests were performed using the Log Rank test within Prism software . Batches of 20–25 adult wild-type and mutant female flies were used in these experiments . The food solution containing bacteria was prepared from a culture grown exponentially at 37 °C to OD600 = 1 . This culture was diluted with a sterile 50-mM sucrose solution to a final OD600 = 0 . 1 . For Db1140 infections , the bacteria were collected by centrifugation and resuspended directly in the 50-mM sucrose solution . A pile of folded papers ( Tork ) was placed in the bottom of medium-sized fly culture tubes and soaked with about 2 . 5 ml of the contaminated sucrose solution . The flies were then transferred to these vials and fed continuously on this solution . Surviving flies were usually counted twice a day . Most experiments were performed at 25 °C , except for experiments involving Gal4 drivers , which were conducted at 29 °C . Sixty-nine nanoliters of 4-fold concentrated Surfactant-Free Red CML Latex beads ( 0 . 30 μm-diameter polystyrene beads; Interfacial Dynamics Corp ) were injected into recipient flies to block phagocytosis , as previously described [63] . The effectiveness of the procedure was checked by testing the phagocytosis of FITC-labeled E . coli [30] . Whole flies count . Flies were infected in batches of 20 with S . marcescens Db11-GFP or 20C2-GFP . Flies were crushed in 0 . 5 ml of LB medium at various times after infection using a micropestle , and the homogenate was serially diluted in LB medium . The number of colony-forming units ( CFU ) was determined through growth overnight at 37 °C on LB agar with the appropriate antibiotics . Hemolymph count . Hemolymph was collected from batches of 20 flies by pricking with an empty capillary mounted on a Nanoject II ( Drumond Scientific ) . The hemolymph was collected in sterile PBS on ice , serially diluted , and plated with the appropriate antibiotics . Intestinal count . The experiment was done as described above for whole flies except that the intestines of 20 infected flies were dissected in sterile PBS and collected in PBS on ice . This analysis was done as previously described [9] . For the images in Figure 2A , intestines were dissected in PBS and immediately observed using a Zeiss SteREO Lumar . V12 dissection microscope equipped with an AxioCam camera and AxioVision 4 . 1 software . For Apotome microscopy , intestines were dissected in PBS , mounted in Vectashield , and observed immediately using a Zeiss Axiovert 200 inverted microscope equipped with an AxioCam camera and AxioVision 4 . 1 software . Optical sections through the fluorescent sample were taken using the Apotome fringe projection system . To visualize GFP , a FITC filter set was used , whereas the rhodamine filter set was employed for DsRed . For confocal microscopy , dissected guts were fixed 30 min in 4% paraformaldehyde ( PFA ) and stained 1 h by 10 μM FITC-labeled phalloidin ( Fluka ) in PBS + 0 . 1% Triton X100 for 2 h . Guts were observed under an inverted Zeiss Axiovert 100 M microscope equipped with the LSM510 laser scanning confocal module Images were processed with LSM510 ( version 2 . 5 ) and ImageJ ( version 1 . 37h ) software . Fly midguts were dissected in phosphate buffer 0 . 1 M ( pH 7 . 2 ) and fixed with 4% glutaraldehyde for 30 min at room temperature . Samples were postfixed for 4 h with 1% osmium tetroxide in the same buffer at 4 °C , rinsed , dehydrated through a graded ethanol series , and embedded in Epon/araldite resin . Ultra-thin sections were contrasted with uranyl acetate and lead citrate . Sections were observed at 60 kV on a Hitachi 7500 transmission electron microscope . Midguts of adult flies were dissected , fixed for 30 min at room temperature in 4% PFA in phosphate buffer ( pH 7 . 2 ) , and infiltrated overnight with 0 . 44 M phosphate-buffered sucrose . After being rapidly frozen in 0 . 22 M phosphate-buffered sucrose with 7 . 5% gelatine , 8-μm-thick sections were cut on a Leica CM3050S cryostat . Sections were then blocked for 30 min in PBS + 2% BSA and incubated overnight with the anti-α-spectrin monoclonal antibody ( antibody 3A9 , obtained from the Developmental Studies Hybridoma Bank ) at a 1:10 dilution in PBS + 0 . 2% BSA . Sections were then incubated for 1 h with goat anti-mouse secondary antibodies conjugated to Alexa488 ( Molecular Probes ) . Finally , sections were incubated for 15 min at room temperature in a 5 μg/ml DAPI solution in PBS . Tissues were dissected and fixed in a glutaraldehyde 1% solution for 10–15 min , and stained in a 30 μl/ml X-gal stock solution ( 5% in DMF ) . Cytochalasin D was dissolved in DMSO to make a 1 mg/ml concentrated solution , which was diluted in PBS to a 20 μg/ml solution . One hundred nanoliters of this solution was injected in each treated fly . Mock-injected flies were injected with a PBS solution containing 2% DMSO . The efficiency of the treatment on phagocytosis inhibition was checked as described [30] . | The gut is a crucial interface of the host with its environment and represents an important portal of entry for pathogens . Here , we have developed a novel model of intestinal infections in the genetic model organism Drosophila melanogaster using the potent entomopathogen bacterium Serratia marcescens . In contrast to other enteropathogens , this bacterium traverses the intestinal epithelium despite a local immune response and gains access to the body cavity of the fruit fly . The cellular arm of innate immunity controls its proliferation in the hemocoele . Interestingly , ingested bacteria that have moved to the hemolymph compartment are not detected by the humoral immune system of the fly unless phagocytosis is blocked . In a septic injury model , S . marcescens kills its host in a day . In contrast , the flies succumb slowly to an intestinal infection , even though the bacterium is present in the hemolymph . We surmise that the bacterium expresses distinct virulence programs according to the mode of infection . Thanks to the genetic tools available in both host and pathogen , the molecular dissection of the interactions between S . marcescens and Drosophila will provide a useful paradigm to decipher intestinal pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"gastroenterology",
"and",
"hepatology",
"immunology",
"microbiology",
"drosophila",
"eubacteria"
] | 2007 | A Model of Bacterial Intestinal Infections in Drosophila melanogaster |
Macrobrachium rosenbergii nodavirus ( MrNV ) is a pathogen of freshwater prawns that poses a threat to food security and causes significant economic losses in the aquaculture industries of many developing nations . A detailed understanding of the MrNV virion structure will inform the development of strategies to control outbreaks . The MrNV capsid has also been engineered to display heterologous antigens , and thus knowledge of its atomic resolution structure will benefit efforts to develop tools based on this platform . Here , we present an atomic-resolution model of the MrNV capsid protein ( CP ) , calculated by cryogenic electron microscopy ( cryoEM ) of MrNV virus-like particles ( VLPs ) produced in insect cells , and three-dimensional ( 3D ) image reconstruction at 3 . 3 Å resolution . CryoEM of MrNV virions purified from infected freshwater prawn post-larvae yielded a 6 . 6 Å resolution structure , confirming the biological relevance of the VLP structure . Our data revealed that unlike other known nodavirus structures , which have been shown to assemble capsids having trimeric spikes , MrNV assembles a T = 3 capsid with dimeric spikes . We also found a number of surprising similarities between the MrNV capsid structure and that of the Tombusviridae: 1 ) an extensive network of N-terminal arms ( NTAs ) lines the capsid interior , forming long-range interactions to lace together asymmetric units; 2 ) the capsid shell is stabilised by 3 pairs of Ca2+ ions in each asymmetric unit; 3 ) the protruding spike domain exhibits a very similar fold to that seen in the spikes of the tombusviruses . These structural similarities raise questions concerning the taxonomic classification of MrNV .
The giant freshwater prawn M . rosenbergii is widely cultivated in tropical and subtropical areas for food . The global production of this species has increased dramatically from about 3 , 000 tons in 1980 to more than 220 , 000 tons in 2014 [1] . However , productivity is threatened by white-tail disease ( WTD ) , which is caused by M . rosenbergii nodavirus ( MrNV ) . This often leads to 100% mortality rates in larvae and post-larvae of M . rosenbergii [2] . The first MrNV outbreak was reported in Pointe Noire , Guadeloupe in 1997 , followed by China [3] , India [4] , Taiwan [5] , Thailand [6] , Malaysia [7] , Australia [8] , and recently Indonesia [9] . To date , neither a vaccine nor effective treatment is available to prevent or manage MrNV outbreaks . MrNV has been classified within the Nodaviridae family of viruses . These nonenveloped viruses have bipartite , positive-sense , single-stranded RNA genomes that are packaged within T = 3 icosahedral capsids . Presently , this family has 2 established genera: Alphanodavirus and Betanodavirus . The former consists of insect-infecting nodaviruses such as Flock House virus ( FHV ) , Pariacoto virus ( PaV ) , black beetle virus ( BBV ) , Nodamura virus ( NoV ) , and Boolarra virus ( BoV ) , while the latter contains fish-infecting nodaviruses such as Malabaricus grouper nervous necrosis virus ( MGNNV ) , grouper nervous necrosis virus ( GNNV ) , and striped jack nervous necrosis virus ( SJNNV ) . Although MrNV is classified within the Nodaviridae , amino acid sequence comparison revealed that its capsid protein ( CP ) shares low similarity ( less than 20% ) with other nodaviruses in the 2 established genera . Thus , it is ambiguous whether MrNV should be grouped in either one of these genera . Conversely , the amino acid sequence of MrNV CP shares approximately 80% similarity with that of Penaeus vannamei nodavirus ( PvNV ) . These 2 crustacean nodaviruses have therefore been proposed to be grouped into a new genus: Gammanodavirus [10] . Nodaviruses are very simple , with their 2 short-genomic RNAs encoding 3 gene products . RNA 1 ( 3 . 2 kb ) encodes the RNA-dependent RNA polymerase ( RdRp ) and the nonstructural B2-like protein , while RNA 2 ( 1 . 2 kb ) encodes the viral CP . The full-length MrNV CP is a polypeptide of 371 amino acids . The N-terminal arginine-rich region interacts with the RNA genome [11] , while the C-terminal domain plays crucial roles in host cell attachment and internalisation [12] . A nuclear localisation signal ( NLS ) has also been identified at the N-terminus ( N-ter ) ( amino acids 20–29 ) of CP . This has been shown to target the viral capsid to the nucleus of insect cells [13] . Further functional regions of CP have yet to be defined . Several three-dimensional ( 3D ) structures of alpha- and betanodaviruses have been determined using both X-ray crystallography and cryogenic electron microscopy ( cryoEM ) [14–18] . These analyses have revealed several common features . Nodaviruses assemble T = 3 icosahedral capsids; 180 CP protomers assemble such that the asymmetric unit comprises 3 identical capsid subunits in 3 quasiequivalent positions termed A , B , and C ( here termed CPA , CPB , and CPC ) . To date , all alpha- and betanodaviruses have been found to have capsomeres that present a trimeric spike . The arginine-rich N-terminal region of the viral CP interacts with the viral RNA segments ( exemplified by PaV , Protein Data Bank [PDB]: 1F8V [15] ) , leading to the formation of a dodecahedral RNA cage at the virion interior . We have previously shown that recombinant CP of MrNV produced using baculovirus expression in Spodoptera frugiperda ( Sf9 ) cells assembles into virus-like particles ( VLPs ) with a diameter of approximately 40 nm [19] . We determined the intermediate resolution structure of the MrNV capsid using cryoEM and image reconstruction [20] . At this resolution , our reconstruction revealed distinct dimer-clustering of capsomeres in the T = 3 MrNV icosahedral capsid . Capsomeres were seen to form square , thin , and blade-like spikes on the virion surface . All other nodaviruses have been shown to assemble with trimeric capsomers . Our structure therefore revealed a strikingly divergent morphology for MrNV , lending weight to the proposed classification of MrNV within a new genus of nodaviruses [20] . Here , we present a high-resolution 3D reconstruction of the MrNV VLP , solved at 3 . 3 Å resolution . From these data , we have constructed an atomic model of the MrNV CP . We show that the shell ( S ) domain of MrNV CP possesses the canonical 8-stranded β-barrel structure common to all nodaviruses . There are , however , striking structural similarities between the MrNV capsid and those of members of the Tombusviridae . The protruding ( P ) domain exhibits a similar fold to that which has been previously shown for tomato bushy stunt virus ( TBSV ) , although the spikes are narrower and are oriented quite differently in the 2 dimeric forms ( AB and CC ) . CP–CP interactions in the S domain are stabilised by coordinated Ca2+ ions . Protomers forming CC dimers , located at the icosahedral 2-fold symmetry axes of the capsid , possess an ordered N-terminal arm ( NTA ) . This passes along the capsid interior , forming intermolecular interactions with neighbouring protomers to stabilise the capsid . Unlike the NTA of TBSV , however , which folds back to form an additional β-strand in CPC before going on to form a structure known as a β-annulus at the adjacent icosahedral 3-fold axis , the NTA of the MrNV CPC crosses the icosahedral 2-fold symmetry axis and inserts into a symmetry-related CPB . The NTA then forms a β-annulus at the next nearest 3-fold symmetry axis before continuing to pass along the capsid interior , donating a strand to a β-sheet in a second neighbouring CPB molecule . We also present an intermediate-resolution structure of the authentic MrNV virion , purified from infected larvae of M . rosenbergii , suggesting that the infectious virus exhibits an identical capsid structure to the one that we have determined for the VLP . Our data present a detailed structural view of this economically important pathogen and raise questions concerning the taxonomic classification of both MrNV and the related PvNV .
To calculate an atomic model of the MrNV CP , we sought to determine a high-resolution 3D reconstruction of the capsid . Frozen hydrated preparations of MrNV VLPs were imaged in a Thermo-Fisher Titan Krios at the United Kingdom Electron Bio-Imaging Centre ( eBIC ) ( Fig 1A ) . A total of 40 , 883 particle images were used to calculate a reconstruction with an overall resolution of 3 . 3 Å ( Fig 1B and 1C and S1 Fig , S1 Movie ) . The reconstructed density map closely matched our previously published 7 Å structure of the MrNV VLP produced in Sf9 cells , showing pronounced blade-shaped dimeric spikes on the capsid exterior and a dodecahedral cage of RNA density within the particle . A cross section through the reconstructed density revealed that the S domains of the VLP were sharply resolved , while the P domains were less well defined , having weaker , fuzzier density ( Fig 1B ) . Local resolution analysis confirmed this , revealing that much of the S domain was solved to 3 . 2 Å resolution , while the tips of the dimeric capsomeres were poorer than 4 Å resolution ( Fig 1D ) . Local resolution filtering and sharpening was applied with a B-factor of −140 Å2 to generate a density map that was suited for high-resolution model building . The asymmetric unit of the T = 3 MrNV capsid comprises 3 copies of MrNV CP; CPA , CPB , and CPC . We have previously shown that the P domains assemble to form dimeric spikes , with AB dimers arranged about the 5-fold symmetry axes and the CC dimers located at the 2-fold symmetry axes . We set out to build the sequence of the MrNV CP ( S2 Fig ) into our density map to produce an atomic model of the MrNV CP for each quasiequivalent position . As a starting point , we docked a homology model into our density map [12 , 20] . Overall , this model fitted poorly within the reconstructed density map , with the exception of 2 regions between amino acid residues 104–135 and 232–243 . A model for the S domain of CPA was therefore manually built and refined from this starting point . This partial model was then docked to CPB and CPC and further edited and refined , leading to reliable models for the S domains at each quasiequivalent position . Interestingly , our density map presented density consistent with the presence of 2 metal ions per CP . Based on the surrounding residues and distances to coordinating atoms , we have modelled these as calcium ions ( discussed below ) . In our 3D reconstruction , density for the S domain was very well resolved . This allowed us to model this region to a high degree of confidence and with relative ease . The P domains were , however , rather less well defined and presented a more challenging task , particularly at the distal tips of the dimeric spikes . CPB was found to be the best resolved as judged by continuity of density , while CPC appeared the least well resolved . Throughout the amino acid sequence of the P domain , there are bulky amino acid side chains that gave confidence in our interpretation of the map . Following several rounds of manual editing and refinement , a model was achieved for the full asymmetric unit that matched our density map and had reasonable geometry ( Fig 2 , S2 Movie , S1 Table ) . The N-terminal regions of CP that include the RNA binding sites ( amino acid residues 21–29 ) were not resolved for any of the chains ( CPA , CPB , or CPC ) in our density map . For CPA , we have successfully modelled amino acid residues 56–371 , while for CPB , we were able to build amino acid residues 55–371 . CPC has a well-resolved NTA that allowed modelling from amino acid 31 . Interestingly , the CPC NTA was found to form extensive contacts with symmetry-related CPB and CPC molecules , forming a network that crosses the capsid interior and is reminiscent of the NTAs previously described for several tombusviruses ( Fig 3 , S3 Movie ) . The CC dimer interface lies at the icosahedral 2-fold axis . Each CPC NTA emerges from the S domain close to this symmetry axis and interacts with 2 CPB protomers , donating β-strands to β-sheets within the CPB S domains . The CPC NTA extends across the CC dimer ( and icosahedral ) 2-fold symmetry axis and inserts into the first CPB subunit , which lies adjacent to the symmetry-related CPC subunit ( Fig 3C and 3D ) . Moving from the C-terminus ( C-ter ) to the N-ter , the NTA then crosses the adjacent icosahedral 3-fold axis and inserts into the next nearest CPB subunit , donating a second β-strand to the β-sheet comprising that molecule and a symmetry-related CPC NTA ( Fig 3E and 3F ) . This interdigitated arrangement of NTAs extending from CPs at the C-position was first described for TBSV [21] and termed a β-annulus owing to the manner in which CPC NTAs wrap around each other at the icosahedral 3-fold axes . Nevertheless , in the case of MrNV , the lacing together of CPC molecules is more extensive because the NTA does not fold back on the CPC to emerge from the S domain at the nearest 3-fold axis and there form the β-annulus structure ( as it does in TBSV ) . Rather , it crosses the 2-fold axis of the CC dimer and then inserts into 2 CPB molecules arranged about the opposite 3-fold axis , where the β-annulus is formed . The last resolved N-terminal residue ( Pro31 ) lies under the next neighbouring 2-fold axis . This is related to the originating 2-fold by a counterclockwise rotation of 120o about the 3-fold axis of the β-annulus ( viewed from the capsid exterior ) . Although it is not resolved , the arginine-rich putative RNA binding site ( amino acids 21–29 ) must therefore be located proximal to the 2-fold symmetry axes for the CPC chains . The MrNV CP S domain comprises residues 62–242 and forms the contiguous shell of the capsid . The T = 3 assembly is made up of 180 copies of the canonical 8-stranded antiparallel β-barrel fold , known as the β-jelly roll . This is commonly seen in positive-sense RNA-containing viruses , including both the nodaviruses and tombusviruses . Another interesting parallel between the structure of MrNV and the tombusviruses is the presence of coordinated metal ions at the interface between CP subunits ( Fig 4 , S4 Movie ) . X-ray crystallographic difference mapping of TBSV following EDTA treatment identified 2 divalent-cation–binding sites at the interface between CPs within the asymmetric unit , which were modelled as Ca2+ [22] . Chelation followed by a rise in pH ( >7 . 0 ) has been shown to cause a structural transition to a ‘swollen’ state in these virions , indicating that these divalent cations play a role in virion stabilisation or possibly control of uncoating . Based on the striking similarity in the locations of these putative metal ions in our data compared to those previously published for TBSV ( PDB: 2TBV ) , and the surrounding residues , we have modelled these putative metal ions as calcium ( Fig 4B ) . We have previously noted the striking differences in the orientations of the P-domain dimer spikes , relative to the underlying capsid shell , between AB and CC dimers . The CC-dimer spike is rotated approximately 85° counterclockwise relative to that of the AB dimer ( viewed from the capsid exterior ) . Moreover , CC-dimer P domains are raised from the capsid surface upon legs of density , whereas the AB P domains sit closer to the capsid shell and are tilted towards their nearest 2-fold symmetry axis . Our atomic model of the MrNV VLP reveals the reason for the substantial differences in pose of these 2 capsomere forms . There is a large linker region between the S and P domains at amino acid residues 241–258 . In the CC-dimer , this linker emerges from the S-domain β-jelly roll and forms a straight leg that is normal to the capsid surface ( Fig 2E ) . The interdomain linker in CPA and CPB , on the other hand , has 2 bends: one between residues Pro247 and Pro249 , which causes the linker to make a right-angled turn , and another at Ile252–Gln254 , which likewise causes a right-angled turn , restoring the path of the linker to its original radial orientation ( Fig 2C ) . The twist in the linkers at the AB dimer induced by these turns therefore accounts for the major differences in the orientations of the 2 types of spike . Interestingly , although we previously noted that the CPB P domain was more closely apposed to the S domain than CPA , our model does not show any contacts . The AB P domain’s orientation is defined by interactions within the AB linker region and with the CPC P domain ( S5 Movie ) . CC-dimer spikes are less well resolved in our cryoEM map than those of the AB dimers . This is to be expected given the manner in which the AB-dimer spike is stabilised through interactions in the AB linker . In contrast , the P domains of CPC stand on extended polypeptide legs that may not offer the same support . The CC spike is instead stabilised by contacts between the P domains of CPB and CPC . AB spikes act as buttresses to the CC capsomere through polar interactions between amino acid residues 270–276 of CPB and 307–317 of CPC ( Fig 5 , S5 Movie ) . This gives rise to the formation of a blade-like superstructure that lays across the 2-fold symmetry axis and comprises 1 CC dimer and 2 AB dimers . It is noteworthy that the previously described homology model for the MrNV CP structure [12] was based on the CP of a tombusvirus , cucumber necrosis virus ( CNV; PDB: 4LLF [23] ) , rather than other known nodavirus structures . Although the homology model was a poor match for our cryoEM density map , our analysis has confirmed the hypothesised dimer-clustered T = 3 icosahedral capsid structure . Close inspection of the fold of the MrNV P domain also reveals an unexpected structural homology between this nodavirus and the tombusviruses ( Fig 6 ) . Tombusvirus P domains have been shown to comprise a 10-stranded antiparallel β-barrel made up of 2 β-sheets annotated as BAJEHG and CDIF ( Fig 6A and 6C ) . Secondary structure assignment in the P domains of the MrNV structure was challenging owing to the poorer resolution in this region . Density in the P domain of CPB was found to be the most clearly resolved , allowing us to build a model in which we have identified a similar β-barrel motif composed of 9 strands arranged into 2 β-sheets ( S3 Fig ) . Based on a 3D pairwise alignment of the P domains for CPB of CNV and MrNV , we have annotated this fold as AJEH2 and DIFGH1 ( Fig 6B and 6D , S4 Fig ) . To ensure that our atomic resolution model of the MrNV capsid is an accurate description of the authentic virion , we determined the structure of purified virions at intermediate resolution . Virions were purified from homogenised , MrNV-infected post-larvae and prepared for cryoEM . 3 , 931 particle images of frozen hydrated MrNV virions were processed to produce a reconstruction at 6 . 6 Å resolution ( Fig 7 , S5 Fig ) . At this resolution , the map appears identical to that of the MrNV VLP in all respects ( compare Figs 7A and 1C ) . Furthermore , the packaged RNA shows a very similar , albeit noisier , structure to the previously described dodecahedral cage . Thus , we conclude that our high-resolution model is an accurate representation of the structure of the authentic MrNV virion .
Beyond providing a detailed description of the structure of MrNV , our data revealed startling similarities between the MrNV capsid structure and those of tombusviruses . We found that the MrNV capsid’s asymmetric unit is stabilised by 6 Ca2+ ions in a manner highly reminiscent of that seen in TBSV . Furthermore , the CPC NTA was found to form an extensive network at the capsid interior that involved an interdigitated structure known as a β-annulus . This motif is also a feature of the tombusviruses . Finally , we found that the fold of the P domain consisted of a β-barrel that also bore a close resemblance to the P-domain structure of the tombusviruses . Capsid stabilisation by binding of divalent cations is well documented in both nodaviruses and tombusviruses ( Fig 8 ) . The alphanodavirus FHV and the betanodavirus GNNV have both been shown to bind a single Ca2+ ion at the interface between each CP subunit in the asymmetric unit ( Fig 8E–8H ) [17 , 24] . FHV also binds a single Ca2+ ion at the quasi-3–fold ( Q3 ) symmetry axis lying at the centre of the asymmetric unit ( Fig 8E ) . Metal binding in MrNV more closely resembles that seen in TBSV , however , in which 2 Ca2+ ions are bound by a DxDxxD motif [25] . There is no significant sequence homology between the CPs of alpha- and betanodaviruses , and comparison of known structures for these genera reveals substantial differences [17] . While both genera assemble capsids that have trimeric spikes , the FHV spike is small , comprising a single β-hairpin motif contributed by each CP subunit , whereas the GNNV CP has a distinct P domain that forms a more substantial capsomere . This feature was previously described as ‘TBSV like’ [26] . Alphanodavirus CPs fold such that both termini are located at the capsid interior; they also undergo proteolytic maturation to produce the γ-peptide , which is required for entry . Betanodaviruses and MrNV both have their C-termini at the capsid exterior and do not encode a γ-peptide . Like MrNV , GNNV exhibits an extended CPC NTA that crosses the icosahedral 2-fold symmetry axis [17] . The CPC NTA also interacts with symmetry-related CPC NTAs at the adjacent 3-fold axis and was described as forming a β-annulus; however , the CPC NTA is not as extensive as that of MrNV and importantly does not form the interdigitated β-strands that are characteristic of the β-annuli of the tombusviruses ( and MrNV ) . Rather , the network is stabilised by limited hydrogen bonding between CPC NTAs at the icosahedral 3-fold axes . Close comparison of the structures of the MrNV capsid with those encoded by both genera of the Nodaviridae and by the Tombusviridae ( Fig 8 ) led us to conclude that the structure of MrNV more closely resembles that of the tombusviruses than either the alpha- or betanodaviruses . The divergence of amino acid sequence and distinct structural features of MrNV compared with other nodaviruses supports the assertion that MrNV , along with the related PvNV , might be classified into a new genus , Gammanodavirus [10 , 20] . Nodaviruses are characterised as small positive-sense RNA-containing viruses having bipartite genomes that infect fish and invertebrates . Tombusviruses are plant viruses that are classified on the nature of their RNA polymerases but are also seen to have consistent capsid structures . Thus , MrNV having features of both virus families poses a conundrum with respect to its taxonomic status . It has been demonstrated that the C-terminal domain of the MrNV CP is important for virus attachment and entry . In particular , deletion of the last 26 amino acid residues substantially reduced infectivity . Inspection of the P-domain structure , however , strongly suggests that deletion of this region is liable to significantly disrupt the β-barrel , as it would remove 2 β-strands from the structure , one from the centre of each β-sheet ( S6 Fig ) . Thus , while it seems likely that the receptor binding site is within the P domain , it may not be limited to the last 26 amino acid residues ( 345–371 ) . We have previously demonstrated insertion of heterologous epitopes into the MrNV capsid structure at the C-ters , such as the ‘a’ determinant of hepatitis B virus surface antigen ( HBsAg ) [27] and the ectodomain of matrix 2 protein ( M2e ) of influenza A virus [28] . Both epitopes were shown to be displayed on the surface of VLPs . Thus , MrNV VLPs present an attractive platform for antigen display . Our structure of MrNV CP now allows us to refine the placement of foreign epitopes . Indeed , each of the 4 loops on the outer surface of the P domain ( amino acid residues 268–275 , 296–303 , 322–326 , and 350–355 ) represents potentially improved targets for further insertions . Combined with the capacity to package nucleic acids , MrNV VLPS may therefore prove to be a useful tool for both vaccine and DNA/RNA delivery . MrNV threatens livelihoods and food security in developing nations . Our atomic-resolution model of the MrNV capsid provides insights into the fundamental biology of this important pathogen , highlighting features that may prove important in our understanding of virus assembly or entry , such as the presence of metal ions that stabilise the asymmetric unit and the structure of the receptor-binding P domain . Such detailed understanding of the capsid structure provides a platform for the development of interventions to control or prevent disease outbreaks in the future .
The gene encoding MrNV CP was amplified from plasmid pTrcHis2-TARNA2 [29] . The forward and reverse primers used to amplify the coding region were 5′-ATG GCC CTT AAC ATC ACC ATG GCT AGA GGT AAA CA-3′ ( NcoI restriction site is underlined ) and 5′-CTA TCG TCG GCA ATA ATT AAG GCG AAT TCG AAG CTT ACG T-3′ ( EcoRI restriction site is underlined ) , respectively . PCR profile was denaturation at 95°C for 3 min , followed by 35 cycles of i ) denaturation at 95°C for 30 s , ii ) annealing at 59°C for 30 s , and iii ) extension at 72°C for 1 min . The final extension was performed at 72°C for 10 min . The PCR products were excised and purified using the QIAquick Gel Extraction Kit ( Qiagen , Hilden , Germany ) . The purified DNA was ligated with the linearised pGEM-T vector ( Promega , Madison , WI , United States of America ) and introduced into competent Escherichia coli DH5α cells . The transformants were plated on Luria Bertani ( LB ) agar plates containing ampicillin ( 100 μg/ml ) . Following an overnight incubation at 37°C , single bacterial colonies were picked and cultured in LB broth . The orientation and nucleotide sequence of the DNA insert were confirmed by DNA sequencing . To produce the MrNV-CP without a His tag , the QuickChange II site-directed mutagenesis kit ( Agilent Technologies , Santa Clara , CA , USA ) was used to create an NcoI restriction-nuclease–cutting site in the pFastBac-HTC plasmid ( Invitrogen , Carlsbad , CA , USA ) . The primers used for mutagenesis were 5′-CGG GCG CGG ATC TCG GTC CGA AAC CAT GGC GTA CTA CCA TCA CC-3′ and 5′-GGT GAT GGT AGT ACG CCA TGG TTT CGG ACC GAG ATC CGC GCC CG-3′ , where the NcoI restriction-cutting site is underlined . The pGEM-T TARNA2 plasmid and the mutated pFastBacHT-C plasmid were digested with EcoRI and NcoI , respectively . The digested products were purified using the QIAquick Gel Extraction Kit ( Qiagen , Hilden , Germany ) and ligated together to produce the pFastBacHTC-TARNA2 . This was introduced into competent E . coli DH10Bac cells ( Invitrogen , Carlsbad , CA , USA ) and plated on LB agar plates containing kanamycin ( 50 μg/ml ) , gentamicin ( 7 μg/ml ) , and tetracycline ( 10 μg/ml ) . White bacterial colonies containing the recombinant plasmid were selected and cultured in LB broth . The recombinant bacmid DNA was extracted , and the presence of DNA insert was confirmed by PCR . The primers used in the PCR were pUC/M13 forward 5′-CCC AGT CAC GAC GTT GTA AAA CG-3′ and pUC/M13 reverse 5′-AGC GGA TAA CAA TTT CAC ACA GG-3′ . Sf9 cells ( 8 × 105 cells/well ) in a 6-well plate were transfected with the recombinant bacmid pFastBacHTC-TARNA2 using Cellfectin II reagent . The transfected cells were incubated at 27°C for 72 h . The cell culture medium was harvested by centrifugation at 500 × g for 5 min at 4°C . The P1 baculovirus stock was amplified by infecting the Sf9 cells ( 2 × 106 cells/mL ) in serum-free Sf-900 III SFM medium ( Gibco , Gaithersburg , MD , USA ) . The infected cells were incubated at 27°C for 72 h . The P2 baculovirus stock was harvested by centrifugation at 500 × g for 5 min at 4°C and stored at 4°C for subsequent experiments . Sf9 cells were cultured as suspension cells at 27°C in a serum-free Sf-900 III SFM medium to reach a cell density of 2 × 106 cells/ml . Recombinant baculovirus stock ( 10% [v/v] ) was added into the culture , which was further incubated for 4 d at 27°C . The MrNV capsid and the Sf9 cells were separated by centrifugation at 500 × g for 5 min at 4°C . The MrNV capsid was precipitated at 60% ( w/v ) ammonium sulphate saturation for 2 h at 4°C . The proteins were pelleted by centrifugation at 18 , 000 × g for 20 min at 4°C . The pellet was resuspended in HEPES buffer A ( 20 mM HEPES , 100 mM NaCl; pH 7 . 4 ) and dialysed in the same buffer overnight . The dialysed sample was purified by size-exclusion chromatography ( SEC ) using a HiPrep 16/60 Sephacryl S-500 HR column ( GE Healthcare , Chicago , IL , USA ) , which was attached to a fast protein liquid chromatography ( FPLC ) system ( Akta Purifier; GE Healthcare , Chicago , IL , USA ) . The purified protein was concentrated with a 100 kDa molecular cutoff centrifugal concentrator ( Pall , USA ) , and the protein concentration was determined with the Bradford assay [30] . Lysate of MrNV-infected post-larvae was prepared according to published methods [31] with some modifications . Briefly , the infected post-larvae were homogenised in HEPES buffer B ( 25 mM HEPES , 150 mM NaCl; pH 7 . 4 ) , and the homogenate was centrifuged at 6 , 000 × g for 10 min at 4°C to remove large debris . The supernatant was further clarified by centrifugation at 12 , 100 × g for 30 min at 4°C . The clarified supernatant was loaded onto a sucrose gradient ( 8–50% [w/v] ) and centrifuged at 210 , 000 × g for 4 . 5 h at 4°C . Fractions ( 500 μl ) were collected and analysed by SDS-PAGE and Western blotting . Fractions containing MrNV were pooled and dialysed in HEPES buffer B . The purified MrNV was concentrated by centrifugation using a centrifugal concentrator ( molecular weight cutoff 10 kDa , Vivaspin Turbo 15 , Sartorius , Göttingen , Germany ) . The final concentration of purified MrNV was determined using the Bradford assay [30] . Purified MrNV VLPs ( at approximately 0 . 2 mg/ml ) or virions ( at approximately 0 . 1 mg/ml ) were prepared for cryogenic transmission electron microscopy using a Thermo-Fisher Vitrobot Mk IV ( Thermo Fisher Scientific , Waltham , MA , USA ) . Particles were imaged on thin-continuous carbon films that had been applied to C-flat holey carbon support films ( R1 . 2/1 . 3; Protochips , Morrisville , NC , USA ) . Four μl of VLP or virion preparation was loaded onto a grid for 1 min , blotted for 4 s , and plunged into liquid ethane . Vitrified samples were imaged at low-temperature ( around 95 K ) and under low-electron–dose conditions . To collect high-resolution data on MrNV VLPs , grids were imaged at the eBIC , Diamond Light Source ( UK ) using a Thermo-Fisher Titan Krios ( Thermo Fisher Scientific , Waltham , MA , USA ) operated at 47 , 170× magnification . A total of 2 , 459 cryomicrograph movies were recorded on a Gatan K2 BioQuantum energy-filtered direct detector camera ( Gatan , Pleasanton , CA , USA ) operated in zero-loss imaging mode with a slit width of 20 eV . Five-s exposures were recorded in electron counting mode at a frame-rate of 4 frames per s and a dose rate of 1 . 8 electrons/pixel/frame . The pixel size was 1 . 06 Å/pixel . MrNV virions were imaged using a JEOL 2200 FS cryo-microscope ( JEOL , Tokyo , Japan ) operated at a nominal magnification of 50 , 000× and an accelerating voltage of 200 kV . Frozen grids were held in a Gatan 626 cryo-stage ( Gatan , Pleasanton , CA , USA ) . 263 cryomicrograph movies were recorded on a Direct Electron DE20 camera ( San Diego , CA , USA ) as 2-s exposures at 20 frames per s and approximately 1 . 5 electrons/pixel/frame . The pixel size was 1 . 11 Å/pixel . All image processing was performed using Relion 2 . 1 [32] . Image stacks of movie frames were motion-corrected using motioncor2 [33] . Defocus estimation was performed using GCTF [34] . For each dataset , a small subset of particle images was manually picked and subjected to 2D classification to prepare a template for automated particle picking . Thereafter , particles were automatically picked for all motion-corrected micrographs . Individual particle images were extracted in 5122-pixel boxes . For MrNV VLPs , a total of 60 , 939 putative particles were extracted from motion-corrected micrographs and subjected to 2D classification . Class averages showing particle images with well-resolved structure were selected , reducing the dataset to 56 , 762 particles . 3D classification was then used to select the best particles for inclusion in the final reconstruction , further reducing the dataset to 40 , 883 particles . This dataset was then refined , leading to the calculation of a reconstruction with an overall resolution of 3 . 3 Å . The MrNV virion reconstruction was calculated following an identical workflow in which 7 , 236 putative particle images were analysed , leading to the definition of a final dataset comprising 3 , 931 particle images . These data were reconstructed at a resolution of 6 . 6 Å . Reconstructions were evaluated to determine global and local resolution as well as estimated B-factors by postprocessing of maps calculated from randomised half sets of data , using the Relion postprocessing routine ( S2 Table ) . In our study of MrNV VLPs , we used our previously calculated 3D reconstruction as a template for starting the classification . For authentic virions , we performed the first 3D classification analysis using a Gaussian sphere as the starting model to prevent model bias , as previously described [20] . Atomic models were built from the high-resolution density maps using the CCP-EM suite of programmes [35] , in particular COOT [36] . The model was refined using REFMAC [37] and PHENIX [38] . Validation was performed using MOLPROBITY [39] . Secondary structure assignment was performed using STRIDE ( http://webclu . bio . wzw . tum . de/cgi-bin/stride/stridecgi . py ) [40] . Density maps and atomic resolution models were visualised using UCSF Chimera [41] . Validation of metal ion assignment was performed using the ‘checkmymetal’ server ( https://csgid . org/metal_sites ) [42] . Contact interface analysis was performed using the PISA server ( http://www . ebi . ac . uk/msd-srv/prot_int/cgi-bin/piserver ) [43] . A 3D pairwise alignment of the MrNV CP P-domain structure and that of CNV ( PDB: 4LLF ) was performed using the FatCat server ( http://fatcat . burnham . org ) [44] . Protein topology diagrams were generated using Pro-Origami ( http://munk . csse . unimelb . edu . au/pro-origami/porun . shtml ) [45] and edited using Inkscape ( https://inkscape . org/en/ ) . The cryoEM map of the MrNV VLP was deposited in the Electron Microscopy Data Bank with accession number EMD-0129 . The cryoEM map of the MrNV virion was deposited in the Electron Microscopy Data Bank with accession number EMD-0130 . The atomic coordinates for the asymmetric unit of the MrNV VLP were deposited in the PDB with accession number PDB: 6H2B . The cryoEM image data for EMD-0129 are deposited in EMPIAR as motion-corrected single-frame micrographs with accession number EMPIAR-10203 . | The freshwater prawn Macrobrachium rosenbergii is widely cultivated for food . Production is threatened by Macrobrachium rosenbergii nodavirus ( MrNV ) , the causative agent of white-tail disease . Outbreaks in hatcheries often result in mortality rates of up to 100% in larvae and post-larvae , leading to devastating economic losses and threatening food security . We describe the atomic structure of the MrNV capsid , solved by cryogenic electron microscopy and three-dimensional image reconstruction . Our analysis revealed surprising differences between the structure of MrNV and that of other known nodaviruses . Moreover , we observed several features in the MrNV capsid that have been previously described in virion structures of the plant-infecting tombusvirus family . Most notably , the MrNV capsid exhibits pronounced dimeric spikes on its surface , the topology of this region closely resembling tombusvirus capsid spikes . Known nodavirus structures have trimeric spikes and do not display the same protein fold . The MrNV capsid is stabilised by divalent cations and laced together by a network of N-terminal arms that line the interior of the virion . Our analysis raises questions about the taxonomic classification of MrNV as well as revealing the structure of the capsid of this important pathogen . These data have the potential to inform the development of future interventions to prevent white-tail disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"microbiology",
"viral",
"structure",
"signs",
"and",
"symptoms",
"aquatic",
"environments",
"microscopy",
"fresh",
"water",
"centrifugation",
"research",
"and",
"analysis",
"methods",
"separation",
"processes",
"marine",
"and",
"aquatic",
"sciences",
"biological",
"databases",
"viral",
"packaging",
"proteomics",
"viral",
"replication",
"virions",
"necrosis",
"biochemistry",
"freshwater",
"environments",
"diagnostic",
"medicine",
"proteomic",
"databases",
"virology",
"earth",
"sciences",
"database",
"and",
"informatics",
"methods",
"electron",
"microscopy",
"biology",
"and",
"life",
"sciences"
] | 2018 | Structure of the Macrobrachium rosenbergii nodavirus: A new genus within the Nodaviridae? |
To validate and update a prediction rule for estimating the risk of leprosy-related nerve function impairment ( NFI ) . Prospective cohort using routinely collected data , in which we determined the discriminative ability of a previously published rule and an updated rule with a concordance statistic ( c ) . Additional risk factors were analyzed with a Cox proportional hazards regression model . The population consisted of 1 , 037 leprosy patients newly diagnosed between 2002 and 2003 in the health care facilities of the Rural Health Program in Nilphamari and Rangpur districts in northwest Bangladesh . The primary outcome was the time until the start of treatment . An NFI event was defined as the decision to treat NFI with corticosteroids after diagnosis . NFI occurred in 115 patients ( 13%; 95% confidence interval 11%–16% ) . The original prediction rule had adequate discriminative ability ( c = 0 . 79 ) , but could be improved by substituting one predicting variable: ‘long-standing nerve function impairment at diagnosis’ by ‘anti-PGL-I antibodies’ . The adjusted prediction rule was slightly better ( c = 0 . 81 ) and identified more patients with NFI ( 80% ) than the original prediction rule ( 72% ) . NFI can well be predicted by using the risk variables ‘leprosy classification’ and ‘anti-PGL-I antibodies’ . The use of these two variables that do not include NFI offer the possibility of predicting NFI , even before it occurs for the first time . Surveillance beyond the treatment period can be targeted to those most likely to benefit from preventing permanent disabilities .
Preventing permanent disabilities due to nerve function impairment ( NFI ) [1] remains a major concern in leprosy control . Mycobacterium leprae , the causative agent of leprosy , infiltrates Schwann cells of peripheral nerve fibers [2] . Subsequently , the nerve fibers can be damaged by accumulation of bacteria and hypersensitivity reactions of the immune system . The decline of nerve function can take place before , during and/or after leprosy treatment . Early detection ( within 6 months ) and corticosteroid treatment may prevent further decline [3] . With leprosy control becoming less specialized and increasingly integrated into general health care services , there is a need for simplified procedures at the field level for timely identification and treatment of NFI in leprosy patients . The chances of preventing disability increase when health care workers pay special attention to patients who have a high risk of developing NFI . To date , several risk factors for NFI have been identified [4]–[6] , and an NFI prediction rule was formulated based on data from the Bangladesh acute nerve damage study ( Bands ) [4] . The Bands prediction rule categorizes patients into NFI risk groups based on their World Health Organization ( WHO ) classification ( ie , paucibacillary [PB] or multibacillary [MB] leprosy ) and the presence of long-standing NFI at diagnosis . However , validation of the Bands prediction rule is needed because i ) the definition of NFI has since changed; ii ) shorter detection delays have led to a smaller percentage of patients with NFI at diagnosis [7] which may change the contribution of this variable to the prediction rule; iii ) a new and simple serological test for anti-phenolic glycolipid I ( PGL-I ) antibody detection [8] , [9] has made routine screening feasible; and iv ) no study has simultaneously assessed all known potential NFI risk factors , namely sex , age , WHO leprosy classification , long-standing NFI at diagnosis , bacterial load and anti-PGL-I antibodies [4]–[6] . We first validated the Bands NFI prediction rule . Next , we compared the performance of an adjusted NFI prediction rule , taking presence of anti-PGL-I antibodies into account .
Patients were previously untreated leprosy patients , newly diagnosed at the Rural Health Program ( RHP ) in northwest Bangladesh in 2002 and 2003 . All patients participated in the Colep trial ( ISRCTN 61223447 ) [10] , which studied the effect of chemoprophylaxis in persons who had contact with leprosy patients ( n patients = 1 , 037 ) . Patients were classified as PB or MB according to the 1998 WHO classification [11] for treatment purposes . For the current analysis , patients who had experienced NFI for <6 months at the time of diagnosis ( n = 162 ) were excluded as they required immediate treatment with corticosteroids , which influences the future occurrence of NFI , the primary outcome event of the study . Eleven patients were excluded because essential data were missing . This leaves a study population of 864 patients ( 538 males , 326 females; median age 34 years , range 5–84 years ) . Follow-up ended in September 2006 ( median follow-up time 46 months ) . Patient information was prospectively recorded on standardised forms by the RHP staff . The primary outcome was the time until the start of treatment . An NFI event was defined as the decision to treat NFI with corticosteroids after diagnosis . The decision was based on the guidelines described in the Rural Health Program ( RHP , formerly DBLM ) treatment protocol , [12] which states that a full dose course of prednisolone ( starting with 40 mg/day and tapering off over 16 weeks for adults ) should be given in case of i ) nerve function reduction by ≥2 points in sensory and/or motor function tests of the ulnar , median , and/or posterior tibial nerves; ii ) corneal anaesthesia; iii ) a nerve tenderness score of 2; or iv ) mixed mild symptoms of neuritis ( ie , tenderness , sensory , and motor function scores of 1 ) . The level of tenderness was defined as mild ( score = 1 ) if palpation of the nerve causes some pain , but does not cause the patient to jump or cry out and defined as severe ( score = 2 ) if touching the nerve causes the patient to jump or cry out . A low dose course of prednisolone ( starting with 20 mg/day and tapering off over eight weeks for adults ) is given for i ) cutaneous neuritis; or ii ) a mild skin reaction in a patch near or overlying a facial nerve . Thus , the criteria to treat NFI with prednisolone include all leprosy reactional and silent neuritis events . In both the Bands and the current study , sensory testing was performed with the Watson ball-point pen test , [13] motor function was assessed according to Medical Research Council grading [14] , and changes in nerve function were evaluated by a physiotechnician trained in nerve function assessment . Patients were under monthly surveillance during standard multidrug treatment ( MDT ) : 6 months for PB patients , 12 months for MB patients . In the original Bands study [4] recommendations for extended surveillance were formulated , stating that for the low-risk group—PB patients without long-standing NFI at diagnosis—routinely performed surveillance for NFI during MDT is sufficient and health education should be provided so that patients are able to recognise and report NFI after completion of MDT . Medium-risk group patients—PB patients with and MB patients without long-standing NFI at diagnosis—require 12 months of surveillance and health education , implying that extended surveillance is only necessary for PB patients , who receive 6 months of MDT . For the high-risk group—MB patients with long-standing NFI at diagnosis—24 months of surveillance is recommended , resulting in 12 months of surveillance in addition to the routine follow-up during MB MDT . The bacterial load was determined by microscopy on Ziehl-Neelsen stained slit skin smears [15] taken from the earlobe , forehead and a skin lesion . The bacterial load was positive if any bacteria were detected in one of the smears . The presence of IgM antibodies against M . leprae was determined at diagnosis with a previously described enzyme-linked immunosorbent assay ( ELISA ) , [16] using dried blood on filter paper . Briefly , the terminal trisaccharide of phenolic glycolipid I ( PGL-I ) linked to bovine serum albumin via a phenolic ring ( NT-P-BSA , kindly provided by Prof . T . Fujiwara , University of Nara , Japan ) was used as a semisynthetic analogue [17] . The titer of IgM antibodies against M . leprae was expressed as net optical density ( OD ) : the absorbance of NT-P-BSA minus that of BSA-coated wells at 450 nm . The status “seropositive” was assigned if the net OD was ≥0 . 20 . This study uses data and samples that are routinely collected by the Rural Health Program from all leprosy patients before , during and after treatment and when patients undergo leprosy reactions . All patients included here gave written informed consent to participate in the Colep trial ( ISRCTN 61223447 ) [10] , a study approved by the Bangladesh Medical Research Council ( BMRC/ERC/2001-2004/799 ) . By giving written informed consent to participate in Colep and accepting treatment they agreed that their data could be used anonymously for research . Kaplan-Meier survival curves were used to determine the cumulative incidence of NFI for the risk groups defined by the prediction rules . Discriminative ability was expressed as a concordance ( c ) statistic [18] ( range 0 . 5–1 . 0 ) . Cox proportional hazards regression was used to identify independent variables that influenced the hazard ratio for NFI . The results are expressed as rate ratios or hazard ratios . Variables associated with NFI in univariate analyses ( p<0 . 10 ) were selected for multivariable analysis in which stepwise backward selection was used to lessen the number of predictors , inclusion at p<0 . 05 . Interactions between variables were tested but not included because they had limited predictive effects . The total number of monthly surveillances was calculated by multiplying the number in a risk group with the recommended surveillance period . The formula for routine surveillances was [ ( n PB*6 ) + ( n MB*12 ) ] , and for surveillance based on the prediction rule the formula was ( n low risk*6 ) + ( n medium risk*12 ) + ( n high risk*24 ) ] . The number of surveillances needed to detect 1 case is the total number of surveillances/NFI cases found . Data analyses were performed with SPSS for Windows ( version 14 . 0 SPSS Inc . , Chicago , IL ) and R software ( version 2 . 3 . 1 www . r-project . org ) .
NFI occurred in 115 of 864 patients ( 13%; 95% confidence interval [CI] 11–16% ) . The Bands prediction rule defines NFI risk groups according to the WHO leprosy classification ( PB/MB ) and longstanding NFI at diagnosis . The low-risk group , comprised of PB patients without longstanding NFI at diagnosis , had a cumulative NFI incidence of 4 . 0% ( 95% CI 2 . 8–5 . 9% [Figure 1] ) , the medium-risk group—PB patients with and MB patients without longstanding NFI at diagnosis—of 37% ( 95% CI 30–45% ) and the high-risk group—MB patients with longstanding NFI at diagnosis—of 53% ( 95% CI 40–68% ) . The cumulative incidences of NFI between the medium- and high-risk groups did not differ significantly . Substituting ‘long-standing NFI at diagnosis’ with ‘anti-PGL-I antibodies’ resulted in risk groups with cumulative incidences similar to those observed in the original Bands study ( Figure 2 ) [4] . With the adjusted prediction rule the low-risk group—seronegative PB patients—had a cumulative incidence of NFI of 3 . 5% ( 95% CI 2 . 2–5 . 4% ) , the medium-risk group—seropositive PB patients and seronegative MB patients—of 13% ( 95% CI 8 . 5–19% ) , and the high-risk group—seropositive MB patients—of 53% ( 95% CI 45–62% ) . The cumulative incidences of NFI differed significantly between low- , medium- , and high-risk groups . The cumulative incidence of this medium-risk group is much lower than the 37% in the medium-risk group defined by the Bands prediction rule . Statistical analyses ( Table 1 ) evaluated the association of NFI with sex , age , WHO classification , long-standing NFI at diagnosis , bacterial load , and anti-PGL-I antibodies . All variables but age were univariately associated with NFI ( p<0 . 05 ) . A multivariable analysis indicated that ‘WHO classification’ and ‘anti-PGL-I antibodies’ were significantly associated with NFI ( p<0 . 0001 ) . MB patients were at an increased risk of NFI ( HR 8 . 0 , 95% CI 5 . 0–13 . 0 ) compared to PB patients , and seropositive patients had an increased hazard risk of 2 . 9 ( 95% CI 1 . 8–4 . 6 ) compared to seronegative patients . When adjusted for WHO classification , the variables sex , age , bacterial load , and longstanding NFI at diagnosis were not significantly associated with NFI anymore . The observed c statistic for the Bands prediction rule in our study was 0 . 79 . The c statistic for the adjusted prediction rule was 0 . 81 , showing a better discriminative ability . Table 2 shows the number of patients that would be classified differently with the adjusted prediction rule compared to the Bands prediction rule . The adjusted prediction rule would place 115 of the low-risk group patients in the medium-risk group and 97 of the medium-risk group patients in the high-risk group; only 18 patients from the medium-risk group and seven patients from the high-risk group would be placed in a lower risk group . Seventy-six ( 76/115 , 66% ) NFI events occurred while patients were undergoing routine surveillance . For the remaining 39 NFI events , additional surveillance would have been necessary for early detection . Extended surveillance using the Bands prediction rule [4] led to the detection of an additional seven patients with NFI for a total of 83 ( 83/115 , 72%: 726 extra visits needed ) . Using the adjusted prediction rule , the number of additional detected patients with NFI increased to 16 , for a total of 92 ( 92/115 , 80%: 2388 extra visits needed ) . With routine surveillance , 83 . 6 visits led to the detection of 1 case , for the Bands prediction rule this was 85 . 3 , and for the adjusted prediction rule 95 . 0 .
Predicting NFI is important for identifying new leprosy patients that are at risk for nerve damage and , consequently , permanent disability . We describe an adjusted NFI prediction rule that replaces the variable ‘longstanding NFI at diagnosis’ with ‘anti-PGL-I antibodies’ . The adjusted prediction rule was better able to identify patients at risk of developing NFI after diagnosis . The original Bands prediction rule for NFI is based on WHO leprosy classification and long-standing NFI at diagnosis [4] . A Kaplan-Meier survival analysis showed that the medium- and high-risk groups had similar survival curves ( Figure 1 ) , indicating that the Bands prediction rule could not differentiate between these two groups . One explanation may be that the definition of NFI has changed since the Bands study: a new NFI category , with less serious events that require a low dose course of prednisolone , was added to original NFI events that require a full dose course [12] . This leads to more patients being identified at an early stage of NFI . In addition , a smaller percentage of long-standing NFI ( >6 months ) and a higher percentage of recent NFI ( <6 months ) , due to shorter detection delays , may have changed the contribution of longstanding NFI at diagnosis . Presence of anti-PGL-I antibodies against M . leprae are a well-known risk factor for NFI [5] . In-depth analysis of all known risk factors for NFI in the current patient cohort showed that NFI is best predicted by ‘WHO classification’ and ‘anti-PGL-I antibodies’ ( Table 1 ) . We adjusted the Bands prediction rule by replacing ‘long-standing NFI at diagnosis’ by ‘anti-PGL-I antibodies’ . The adjusted rule was able to differentiate between three risk groups with significantly different cumulative incidences of NFI ( Figure 2 ) ; the c statistic increased from 0 . 79 to 0 . 81 . Unfortunately , we could not validate the adjusted prediction rule on the original Bands cohort because no serology data were available . The adjusted prediction rule distinguished three risk groups comparable to those in the Bands study [4] ( Figure 2 ) . Therefore , the surveillance recommendations that were based on the Bands study [4] can be maintained ( see Methods ) . When replacing the Bands prediction rule with the adjusted prediction rule 212 patients were reassigned to a higher risk group and 25 patients to a lower risk group ( Table 2 ) , suggesting that the adjusted prediction rule has considerable implications for patient care . The reassignment of these patients to a higher risk group is warranted because they have a higher-than-average risk to develop NFI: 7% for patients moving from the low to the medium risk group and 51% for patients moving from the medium to the high risk group . The adjusted prediction rule can thus be used to identify a substantially higher number of new NFI cases than either routine or Bands rule-based surveillance and offers increased opportunity to prevent nerve damage in leprosy . However , the number of visits needed to detect one case is higher than with alternative strategies . We consider this operationally feasible and medically justifiable in view of the serious consequences of NFI , including life-long disability . WHO classification is a good predictor of future NFI [6] but it rather crudely divides leprosy patients into two groups ( PB and MB ) . The presence of anti-PGL-I antibodies is known to correlate with the bacterial load [16] , and thus offers a further refinement of the WHO classification into patients with high and low bacterial loads . This may explain the added predictive value of the presence of antibodies . In contrast to the Bands rule the adjusted rule uses two variables that do not include NFI . This offers the possibility of predicting NFI before it actually occurs . We expect that the adjusted NFI prediction rule will be relevant in other settings , since the predicting variables are well defined and easily determined , but it should be validated externally . We believe that the adjusted prediction rule can be applied in current health services , since it fulfils the need for simplified guidelines and diagnostic protocols . Contrary to the Bands prediction rule , the adjusted rule does not rely on a specialist physiotechnician for the prediction . However , this person is needed to document the baseline nerve status and for surveillance during follow up examinations . Recently , a simple anti-PGL-I field test was developed that gives results within ten minutes , [8] , [9] making routine testing feasible . Thus , leprosy diagnosis and NFI prediction can be accomplished during a single consultation . Additional benefits of the anti-PGL-I test are that it assists with the classification and aids diagnosis of leprosy patients with doubtful clinical signs [8] , [9] , [16] . With the adjusted prediction rule , the necessity to continue surveillance beyond the treatment period can be determined . New leprosy patients can be assigned to an NFI risk group , and appropriate surveillance can be planned . Nerve damage can thus be successfully prevented despite the fact that leprosy control has been integrated into general health services . | Leprosy is caused by a bacterium that attacks the peripheral nerves . This may cause nerve function impairment ( NFI ) , resulting in handicaps and disabilities . Therefore , prediction and prevention of NFI is extremely important in the management of leprosy . In 2000 , a prediction rule for NFI was published , but circumstances have changed since the study was performed in the 1990s: the leprosy detection delay has shortened and the definition of NFI has changed . The original rule used ‘leprosy classification’ and ‘NFI present at diagnosis’ to predict future NFI . In the current patient population we studied an adjusted rule based on ‘leprosy classification’ and ‘presence of antibodies’ . This adjusted rule predicted NFI more often than the original rule . With the adjusted rule it is now also possible to assess NFI risk before the first nerve damage event takes place . This may help doctors and health workers to improve surveillance for people at high risk . Early detection and treatment can then prevent permanent disabilities . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"evidence-based",
"healthcare/clinical",
"decision-making",
"pathology/neuropathology",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"neurological",
"disorders/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"immunology/immune",
"response",
"dermatology/skin",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"immunology",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2008 | Preventing Nerve Function Impairment in Leprosy: Validation and Updating of a Prediction Rule |
Aging is a complex phenotype responsive to a plethora of environmental inputs; yet only a limited number of transcriptional regulators are known to influence life span . How the downstream expression programs mediated by these factors ( or others ) are coordinated into common or distinct set of aging effectors is an addressable question in model organisms , such as C . elegans . Here , we establish the transcription factor ETS-4 , an ortholog of vertebrate SPDEF , as a longevity determinant . Adult worms with ets-4 mutations had a significant extension of mean life span . Restoring ETS-4 activity in the intestine , but not neurons , of ets-4 mutant worms rescued life span to wild-type levels . Using RNAi , we demonstrated that ets-4 is required post-developmentally to regulate adult life span; thus uncoupling the role of ETS-4 in aging from potential functions in worm intestinal development . Seventy ETS-4-regulated genes , identified by gene expression profiling of two distinct ets-4 alleles and analyzed by bioinformatics , were enriched for known longevity effectors that function in lipid transport , lipid metabolism , and innate immunity . Putative target genes were enriched for ones that change expression during normal aging , the majority of which are controlled by the GATA factors . Also , some ETS-4-regulated genes function downstream of the FOXO factor , DAF-16 and the insulin/IGF-1 signaling pathway . However , epistasis and phenotypic analyses indicate that ets-4 functioned in parallel to the insulin/IGF-1 receptor , daf-2 and akt-1/2 kinases . Furthermore , ets-4 required daf-16 to modulate aging , suggesting overlap in function at the level of common targets that affect life span . In conclusion , ETS-4 is a new transcriptional regulator of aging , which shares transcriptional targets with GATA and FOXO factors , suggesting that overlapping pathways direct common sets of lifespan-related genes .
The emerging picture from studies with model organisms is that animal life span is regulated by coordination of gene regulatory networks in response to environmental inputs . In C . elegans a number of transcription factors function as genetic modifiers of aging , including Forkhead ( DAF-16 and PHA-4 ) and GATA ( ELT-3 , ELT-5 , and ELT-6 ) factors [1]–[3] . The transcription factors that function in life span determination often respond to evolutionarily conserved pathways and cellular processes , including insulin/IGF-1 signaling , c-Jun N-terminal kinase signaling ( JNK ) , the Target of rapamycin pathway ( TOR ) , caloric intake , mitochondrial respiration and signaling from the germ line [1] , [2] , [4]–[8] . Gene expression profiling has identified overlapping downstream targets for these transcription factors that regulate development , metabolism , reproduction , stress response and innate immunity [3] , [9]–[11] . Consistent with this complexity , genetic tests show that disruption of individual downstream target genes has a modest impact on longevity . Therefore , the downstream effectors of each transcription factor are proposed to act collectively to mediate the significant impact of signaling pathways on life span [9] . In this study , we identify the ETS transcription factor , ETS-4 , as a longevity determinant in C . elegans . Gene knock-out studies performed for 20 of the 26 ETS genes in mice have implicated ETS factors in diverse cellular processes such as proliferation , differentiation , migration , apoptosis , and cell-cell interactions [12]–[15] . However , discerning molecular mechanisms of ETS protein function in mice has been complicated by the large number of ETS paralogs expressed in any particular cell type [15] , [16] . C . elegans , with only ten ets genes , provides a simpler and more genetically tractable model to investigate ETS factor function in vivo . C . elegans ets-4 is the apparent ortholog of vertebrate SPDEF/SAM pointed domain containing ETS transcription factor ( also known as PDEF/Prostate derived ETS factor ) ( Figure S1 ) . The mRNA levels of SPDEF are altered in breast and prostate tumors [17] , [18] . Studies using tumor cell lines show that SPDEF affects cell migration and invasion pathways [18]–[23] . A deletion allele in mice suggests a role in specialized intestinal epithelial cell differentiation [24] . Thus , SPDEF may have significant disease relevance , yet its physiological role in normal animal development and homeostasis is not completely understood . To decipher ets-4 function , we undertook a reverse genetics approach in C . elegans and established ets-4 as a transcriptional regulator of longevity . Expression of ets-4 was observed in a number of tissues , however , transgenic rescue experiments implicated the intestine in the aging phenotype . Gene expression profiling identified ETS-4-regulated genes that function in life span determination . Strikingly , a significant number of these life span effectors have been previously shown to function downstream of the insulin/IGF-1 signaling pathway as well as the GATA factor , ELT-3 . Genetic tests reveal that ets-4 functions in parallel to the insulin/IGF-1 signaling pathway , yet requires the FOXO transcription factor , daf-16 , to modulate life span . Taken together , our findings identify the physiological role of ETS-4 in C . elegans and indicate transcriptional control of aging effectors by ETS-4 .
To determine the function of ETS-4 in vivo , ets-4 ( ok165 ) worms carrying a deletion at the ets-4 locus were analyzed . Coding sequences for both the ETS DNA binding domain , and the PNT domain , which is a protein-protein interaction domain conserved in a subset of ETS proteins , were lacking in the ets-4 ( ok165 ) worms ( Figure 1A and 1C ) . Animals carrying the ok165 allele lacked full-length ets-4 mRNA as confirmed by RT-PCR analysis ( Figure 1B ) . Therefore , we propose that ets-4 ( ok165 ) is a null allele for ETS-4 function . After outcrossing six times to the wild-type ( N2 ) strain , the ets-4 ( ok165 ) worms were examined for phenotypes . The larval developmental time , measured as the time taken for L1 larvae to reach the young adult stage at 20°C , was 6–8 hr longer in ets-4 ( ok165 ) than wild-type worms ( Figure 2A ) . Despite this 10–13% delay , ets-4 ( ok165 ) larvae exhibited apparently wild-type development and morphology . Neither arrests at particular larval stages nor molting defects were observed . The self brood size of ets-4 ( ok165 ) worms was similar to that of wild-type worms , suggesting normal fecundity ( Figure 2B ) . However , a difference in the rate of egg-laying was observed ( Figure 2C ) . During the peak egg-laying period ( day 2 of egg-laying ) the ets-4 ( ok165 ) worms laid significantly fewer eggs ( 109±5 ) than wild-type worms ( 142±6 ) ( Figure 2C ) . In addition , ets-4 ( ok165 ) hermaphrodites produced significantly more progeny ( 39±6 ) later in life ( day 4 of egg-laying ) than wild-type worms ( 14±2 ) ( Figure 2C ) . The ets-4 ( ok165 ) males exhibited mating efficiency comparable to wild-type males in crosses with temperature-sensitive fem-3 ( e2006 ) hermaphrodites [25] that are incapable of producing self-progeny at the restrictive temperature ( data not shown ) . In summary , ets-4 ( ok165 ) worms exhibited a 10–13% delay in larval development and an altered egg-laying rate , but were wild-type in morphology , development and fecundity suggesting normal fitness . Considering a post-developmental role for ets-4 , we monitored the adult life span of ets-4 ( ok165 ) worms . Because different laboratory strains of C . elegans can have significantly different life spans [26] , we compared the life spans of ets-4 ( ok165 ) animals with those of isogenic ets-4 ( + ) controls obtained by outcrossing ets-4 ( ok165 ) to the wild-type ( N2 ) strain . Age-matched wild-type or ets-4 ( ok165 ) L4 stage larvae were picked and the first day of adulthood counted as day one . At 25°C , the mean adult life span of ets-4 ( ok165 ) worms ( 18 . 0±0 . 4 days ) was significantly longer than that of isogenic ets-4 ( + ) wild-type worms ( 13 . 3±0 . 6 days ) ( Figure 3A and Table 1 ) . Thus , loss of ets-4 led to a significantly longer mean adult life span compared to wild-type worms at 25°C ( Figure 3A , Table 1 and Table S1 ) . Since growth temperature strongly influences C . elegans longevity [27] , the life span of ets-4 ( ok165 ) animals was also monitored at 20°C . The mean adult life span of ets-4 ( ok165 ) animals at 20°C ( 27 . 1±0 . 8 days ) was significantly longer than that of isogenic ets-4 ( + ) wild-type worms ( 15 . 4±0 . 6 days ) , confirming the extended life span phenotype of ets-4 null mutant animals ( Figure 3B and Table 1 ) . Because feeding defective ( eat ) mutant animals are also slow-growing and long-lived [6] , [28] , the feeding behavior of ets-4 ( ok165 ) was examined by recording the pharyngeal pumping and defecation rates . Under well-fed conditions , the feeding behavior of ets-4 ( ok165 ) worms was indistinguishable from that of wild-type animals ( Figure S2 ) . Thus , ETS-4 regulates life span without modifying feeding behavior . To further characterize the role of ets-4 in life span regulation , we initiated ets-4 inactivation by RNAi at the L4 larval stage to bypass potential earlier developmental roles [29] . Similar to the long-lived phenotype of ets-4 ( ok165 ) mutant animals , ets-4 ( RNAi ) on wild-type worms resulted in significant extension of mean adult life span ( Figure 3C and Table 1 ) . These data suggest that ets-4 functions post-developmentally in regulation of adult life span . We generated a second deletion allele of ets-4 , uz1 , which removes sequences coding for most of the ETS domain and the 3′UTR , but retains the PNT domain coding sequence ( Figure 1A and 1C ) . An abridged ets-4 mRNA that could encode a truncated protein was detected in ets-4 ( uz1 ) worms ( Figure 1B ) . Similar to the ets-4 ( ok165 ) worms , we observed slow larval growth , altered progeny production and extended life span phenotypes in ets-4 ( uz1 ) mutant animals ( Figure 2 , Figure 3A and 3B ) . Thus , examination of this second ets-4 deletion allele corroborated the loss-of-function phenotypes observed in ets-4 ( ok165 ) mutant animals and enabled further use of this allele . However , ets-4 ( uz1 ) worms exhibited several unique phenotypes not observed in ets-4 ( ok165 ) worms including ruptured vulva , distorted seam cell syncytia and broken alae ( data not shown ) . We speculate that these added effects are likely due to the interfering activity of the truncated ETS-4 encoded by the uz1 allele that would lack the ability to bind to ETS-4 target genes and yet , may interact with various protein partners through the PNT domain . Therefore , the additional phenotypes associated with the ets-4 ( uz1 ) allele may be involved with , but not necessarily restricted to , normal ETS-4 function and were not studied further . Nevertheless , analyses of two distinct deletion alleles of ets-4 enabled the identification of ets-4 loss-of-function phenotypes . To better understand how ETS-4 regulates life span , we sought to identify the tissues in which ETS-4 functions . GFP reporter constructs containing 5 kb of the ets-4 promoter alone ( Pets-4::gfp ) or including the genomic DNA coding for ETS-4 ( ets-4::gfp ) were used to generate several independent transgenic lines . Robust GFP expression for both constructs was observed in the intestinal cells of transgenic worms starting at the 3-fold embryonic stage and was maintained through larval development and in the adult ( Figure 4A and 4B ) . GFP expression was also seen in several cells of the anterior and posterior bulbs of the pharynx and in seam cells ( Figure 4B , 4C and 4D ) . Lastly , ets-4 expression was observed in a few unidentified cells of the vulva , hypodermal nuclei , several unidentified neurons , labial socket cells of the head , and a few cells of the rectum ( data not shown ) . Thus , these results expanded the observations made in previous studies using shorter regions of the ets-4 promoter in GFP constructs [30]–[32] and demonstrated ets-4 expression in the intestine and neurons , key tissues known to regulate longevity in worms [33]–[35] . Studies in mammals , flies and worms have identified subsets of cells , including adipose , intestinal and neuronal tissues , that affect the rate of aging of the entire organism [33]–[40] . To identify cell types in which ETS-4 functions to modulate life span , a transgene encoding a YFP::ETS-4 fusion was expressed in intestinal cells or neurons of ets-4 ( ok165 ) worms by use of the gly-19 or rab-3 promoters , respectively [35] , [41] , [42] . Analysis of at least two independent transgenic lines showed that YFP::ETS-4 expressed in intestinal cells restored the life span of ets-4 ( ok165 ) worms to wild-type ( Figure 3D , Table 1 and Table S1 ) . In contrast , the extended life span phenotype of ets-4 ( ok165 ) worms was not affected by YFP::ETS-4 expression in neurons ( Figure 3D , Table 1 and Table S1 ) . Additionally , YFP::ETS-4 expression in either cell type did not rescue the altered egg-laying rate phenotype of ets-4 ( ok165 ) worms , indicating that the longevity and egg-laying phenotypes were separable ( Figure S3 ) . Animals expressing YFP::ETS-4 had wild-type brood sizes , suggesting normal fecundity and fitness ( Figure S3 ) . Moreover , because the life span of rescued lines was not shortened beyond wild-type controls , we concluded that YFP::ETS-4 fusion was not toxic in these tissues ( Table 1 ) . In summary , ETS-4 functions in the intestine to modulate life span . Because ETS-4 is implicated to be a transcription factor , a role in life span determination predicts a set of ETS-4 target genes that function in aging . We took advantage of the two distinct strains with a disrupted ets-4 locus to investigate effects on gene expression . We chose late L4 stage larvae because of ease in staging and relevance to adult intestine function . Microarray-based expression profiling of wild-type and ets-4 ( ok165 ) larvae identified 145 genes whose expression was altered with 88 genes down-regulated 2 . 2 fold or more ( Table S2 ) . qRT-PCR analyses in age-matched , one-day old adults confirmed the differential expression observed for nine out of nine genes selected randomly from the top-thirty changed genes ( Figure S4 ) . As predicted by the broader phenotypic consequences observed in ets-4 ( uz1 ) worms , more genes ( 542 ) displayed altered expression in these animals than in ets-4 ( ok165 ) worms ( Table S3 ) . qRT-PCR controls in age-matched , 1-day old ets-4 ( uz1 ) and wild-type adult worms confirmed the expression changes observed for eight out of eight genes ( Figure S4 ) . Because animals carrying either ets-4 deletion displayed similar life span extension , we predicted that the genes overlapping in these data sets would be enriched for aging effectors . A statistically significant overlap of 70 genes with altered expression in ets-4 ( ok165 ) and ets-4 ( uz1 ) worms was identified ( p<0 . 0001 ) ( Table 2 and Table S4 ) . To determine , in an unbiased manner , whether these 70 ETS-4-regulated genes represented a particular biological pathway , we performed gene ontology analysis using GOstat [43] . The top five overrepresented categories yielded by this analysis include lipid transport ( vit-2 , vit-3 , vit-4 , and vit-5 ) , multicellular organismal aging ( vit-5 , vit-2 , thn-1 , and lys-7 ) and fatty acid metabolic process ( acdh-2 , ech-9 , and C48B4 . 1 ) ( Table 3 ) . Five control gene lists of the same size generated randomly from genes represented on the expression arrays did not show these classes to be overrepresented . Restoration of ETS-4 function in the intestine rescued life span of ets-4 null worms to wild-type ( Figure 3D and Table 1 ) . To test whether ETS-4-regulated genes were biased towards expression in a particular cell type , we compared our microarray data set to tissue-specific expression data from previous studies . Strikingly , the ETS-4-regulated gene list was significantly ( p<0 . 0001 ) enriched only with intestinal genes [44] , and not with germ-line [45] , muscle [46] , pharyngeal [47] or neuronal genes [48] ( Table 2 ) . This result did not vary when down- or up-regulated gene sets in the ets-4 mutant animals were analyzed separately for enrichment of cell-type specific gene expression ( data not shown ) . Thus , the molecular signature of ETS-4 , identified by gene expression profiling , substantiates a role for ets-4 in the intestine . Lipid storage and metabolism have been linked to longevity regulation [36] , [39] , [40] , [49] Also , expression of yolk protein genes ( vit genes ) and genes regulating fatty acid β-oxidation , such as an acyl-CoA dehydrogenase ( acdh-2 ) , an acyl-CoA oxidase ( C48B4 . 1 ) and an enoyl-CoA hydratase ( ech-9 ) , were down-regulated in ets-4 mutant animals ( Table S4 ) . Thus , to explore a possible mechanism for life span regulation by ETS-4 , we examined lipid levels in ets-4 null mutant animals . Lipid extracts from synchronized , one-day old adult , wild-type and ets-4 null mutant animals were subjected to thin-layer chromatography ( TLC ) . The phospholipid ( PL ) and triacyglyceride ( TAG ) fractions from ets-4 ( ok165 ) worms visualized on TLC plates were similar to wild-type ( Figure S5 ) . The relative levels of triacylglycerol stores , as well as the fatty acid composition of phospholipid and triacylglycerol fractions , quantified by gas chromatography in age-matched , one-day old ets-4 ( ok165 ) adult animals , were not altered compared to wild-type worms ( Figure S5 ) . Thus , the altered expression of genes involved in lipid metabolism did not affect total lipid levels in ets-4 ( ok165 ) worms . Therefore , lipid homeostasis in ets-4 ( ok165 ) animals could likely be maintained due to the compensatory activities of other enzymes regulating fatty acid metabolism . Alternatively , lipid uptake , transport or storage in different tissues of ets-4 ( ok165 ) worms may be affected without an alteration in the total lipid levels within the whole organism . Discerning the transcriptional targets of ETS-4 enabled us to ask whether ETS-4-regulated genes change expression during the course of normal aging [3] . We found that 24% of the 70 ETS-4-regulated genes were previously identified age-regulated genes ( Table 2 and Table S4 ) . This is a significantly higher overlap than that expected by random chance ( p<0 . 0001 ) , corroborating a function for ETS-4 in normal aging ( Table 2 ) . Interestingly , the GATA transcription factors ELT-3 , ELT-5 , and ELT-6 were shown to direct the age-regulation of a large fraction of genes that change expression with age [3] . Our data suggested that ETS-4 participated in directing the expression of a significant proportion of age-regulated genes . To decipher whether a common set of targets exists for transcription factors that function in life span determination , we compared ETS-4-regulated genes with those that function downstream of the well-characterized insulin/IGF-1 signaling pathway . A comparison to the genes that act downstream of two components of the insulin/IGF-1 signaling pathway ( daf-2 and daf-16 ) [10] indicated a 20% overlap , which represents a significant enrichment ( p<0 . 0001 ) ( Table 2 and Table S4 ) . The overlapping gene set included genes that were down-regulated as well as genes up-regulated in ets-4 ( ok165 ) animals ( Table S4 ) . Also , 50% of the overlapping genes were up-regulated in daf-2 pathway mutant animals and repressed in daf-16; daf-2 double mutant animals , while the other half displayed the opposite expression profile ( Table S4 ) [10] Since ets-4 and daf-2 loss-of-function mutations cause an extended life span phenotype , we predicted that the direction of expression changes for genes involved in aging will be similar in these mutant animals . Indeed , the genes identified by our ontology analyses as functioning in multicellular organismal aging ( Table 3 ) were either down-regulated ( vit-5 and vit-2 ) or up-regulated ( thn-1 and lys-7 ) in both ets-4 ( ok165 ) and daf-2 ( − ) animals ( Table S4 ) [10] . Taken together , our analyses revealed a shared pattern of gene expression changes between in ets-4 mutant animals , insulin/IGF-1 signaling pathway mutant animals , and aging wild-type worms . Thus , a common set of transcriptional targets for ETS-4 , the GATA factors ( ELT-3 , ELT-5 , and ELT-6 ) , and DAF-16 exist . Analyzing global transcriptional changes in ets-4 mutant animals identified a number of genes regulated by ETS-4 . Direct transcriptional targets are predicted to have ETS-4 binding sites in the transcription start site ( TSS ) -proximal regions . As a first step towards testing this hypothesis , we characterized the DNA binding and transcriptional activity of ETS-4 . We tested the binding of ETS-4 to ETS binding sites previously described , which display a 5′-GGAA/T-3′ core recognition sequence [13] , [50] , which is consistent with the reported motif for the vertebrate orthologue , SPDEF [51] . ETS-4 , purified from a bacterial expression system , bound to ETS binding sites displaying either a GGAA or GGAT core motif with similar high affinity ( KD∼10-9 M ) ( Figure 5A and Figure 5B ) . Furthermore , a wild-type , but not mutated , ETS binding site , competed with ETS-4 binding to the labeled DNA ( Figure 5C ) . These results indicated that ETS-4 bound to consensus ETS sites in a sequence-specific manner . ETS proteins demonstrate functional diversity acting both as transcriptional activators and repressors [13] , [52] . To determine the transcriptional activity of ETS-4 , we tested ETS-4 in transcription assays in yeast and cultured mouse fibroblasts . In S . cerevisiae , LexA::ETS-4 ( 1-345 ) activated transcription of a reporter gene via a promoter with LexA binding sites ( Figure S6 ) . ETS-4 ( 1–345 ) also activated transcription in NIH3T3 cells ( Figure S6 ) . In yeast , the N-terminal ETS-4 ( 1–125 ) fragment activated transcription of the reporter gene; however , this activity was lost with the inclusion of the PNT domain in ETS-4 ( 1–200 ) ( Figure S6 ) . Indeed , we have mapped repressive function to the PNT domain including recruitment of repressive co-factors ( data not shown ) . Thus , our data identify two regions of ETS-4 that activate transcription ( amino acids 1–125 and 200–345 ) and suggest the ability of the PNT domain to modulate this activity towards a repressive function . This potential opposing transcriptional function is consistent with the hypothesis that both up- and down-regulated genes are direct ETS-4 targets . To identify candidate direct targets of ETS-4 , we looked for ETS binding sites conserved across six nematode species in the region from 1500 base pairs ( bp ) upstream to 500 bp downstream of the transcription start site ( TSS ) of ETS-4-regulated genes . The position specific weight matrix ( PWM ) used to search for ETS binding motifs encompasses the sites that ETS-4 bound in vitro ( Figure 5 and Table S5 ) . Using the MULTIZ alignment algorithm [53] , at least one conserved ETS binding motif was identified in the transcriptional control regions of 54 of the 70 ETS-4 regulated genes ( Table S5 ) . Although this frequency did not represent an enrichment of conserved ETS sites compared to other randomly selected TSS-proximal regions ( data not shown ) , we suspect this represents the general importance of ETS factors in many control regions and not the lack of biological relevance of these sites on ETS-4 regulated genes [54] , [55] . It was striking that genes with conserved ETS binding motifs included not only 82% of the down-regulated genes , but also 72% of the up-regulated genes in ets-4 mutant animals ( Table S5 ) . This is consistent with the activating and repressive transcriptional functions of ETS-4 noted above . Also , several ETS proteins , in response to signaling pathways , fine-tune their transcriptional activity , functioning as activators or repressors [13] , [52] , [56] . In conclusion , we propose that ETS proteins may bind the transcriptional control regions and , thus , regulate directly the expression of 77% of the genes that were identified to function downstream of ETS-4 . To test the function of ETS-4 as a DNA binding transcription factor in a native context , we focused on vit-5 , which is one of the genes down-regulated in ets-4 mutant worms that bears conserved ETS binding sites ( Table S2 , Table S4 and Table S5 ) . vit-5 encodes a lipoprotein related to mammalian ApoB-100 , a core LDL particle constituent [57] . Also , vit-5 ( RNAi ) was shown previously to extend the mean life span of RNAi-sensitive wild-type worms [10] . Pvit-5::gfp expression was significantly reduced in the intestinal cells of ets-4 ( ok165 ) compared to wild-type worms suggesting that ETS-4 is necessary for inducing vit-5 expression ( Figure S7 ) . We conclude that ETS-4 is a transcription factor , fully competent to help orchestrate a transcriptional network involved in life span regulation . To characterize how ets-4 modulates C . elegans lifespan , we tested whether ets-4 genetically interacted with other known longevity regulators . First , because ETS-4 , ELT-3 , and DAF-16 share common downstream targets , we asked if they function downstream of the same signaling pathway . The insulin/IGF-1 signaling pathway involves the activation of the insulin/IGF-1 receptor , DAF-2 , which triggers a kinase cascade involving the serine/threonine kinases AKT-1 and AKT-2 culminating in the cytoplasmic sequestration and inhibition of the FOXO transcription factor DAF-16 [58] , [59] . In addition to inhibiting DAF-16 , the insulin/IGF-1 pathway also exerts a constant level of regulation on ELT-3 expression [3] , [60] . We tested whether ets-4 modulated life span by acting downstream of the insulin/IGF-1 signaling pathway . To do this , we examined the genetic relationship between ets-4 and two major components of the insulin/IGF-1 signaling cascade - the receptor daf-2 and the kinases akt-1/2 . For the life span comparisons , RNAi was used to inactivate insulin/IGF-1 signaling pathway genes in isogenic ets-4 ( ok165 ) and wild-type strains . If ets-4 regulates life span by acting downstream of the insulin/IGF-1 signaling pathway , then loss of ets-4 should not significantly extend the life span of long-lived worms lacking insulin/IGF-1 signaling . Alternatively , further extension of the life span of long-lived ets-4 null mutant worms upon inhibition of insulin/IGF-1 signaling would indicate that ets-4 functions in parallel to this pathway . As previously reported [1] , [61] , [62] , daf-2 ( RNAi ) animals were significantly long-lived ( Table 1 ) . Notably , daf-2 ( RNAi ) ; ets-4 ( ok165 ) animals lived longer than either ets-4 ( ok165 ) or daf-2 ( RNAi ) worms alone ( Table 1 and Table S1 ) . Similarly , RNAi against the kinases akt-1/akt-2 further extended the life span of the long-lived ets-4 ( ok165 ) worms ( Figure 6A and Table 1 ) . Thus , inhibition of two different components of the insulin/IGF-1 signaling pathway further increased the life span of long-lived ets-4 ( ok165 ) animals . Because RNAi , and not null mutations , was used to inactivate signaling pathway genes , the possibility that the insulin/IGF-1 receptor pathway partially contributes to the life span phenotypes of ets-4 null mutant animals cannot be completely eliminated . However , our data support a model whereby ets-4 functions in parallel to the insulin/IGF-1 signaling pathway to modulate life span ( Figure 7 ) . We further explored the genetic relationship between ets-4 and the insulin/IGF-1 signaling pathway by investigating if ets-4 participated in other physiological processes regulated by the pathway . Signaling through DAF-2 is critical for dauer formation . Hence , the involvement of ets-4 in dauer formation was tested first . Assaying for dauer formation at 25°C and 27°C [63] , [64] , we found no defects in dauer formation in ets-4 ( ok165 ) mutant animals compared to wild-type worms ( data not shown ) , suggesting that ets-4 does not play a significant role in this process . Next , we tested whether ets-4 null mutant animals exhibit altered response to environmental stress stimuli since insulin/IGF-1 signaling also regulates stress resistance . To assay for heat stress response , the survival of adult ets-4 ( ok165 ) worms was monitored after a shift to 35°C . As reported previously[65]-[68] , daf-2 ( e1370 ) animals survived significantly longer than wild-type worms at 35°C , whereas daf-16 ( mgDf50 ) animals died faster ( Figure S8 ) . No significant differences in survival were seen between ets-4 ( ok165 ) and wild-type worms during the heat stress time-course ( Figure S8 ) , indicating that ets-4 is not required for response to heat stress . Extension of life span is often associated with increased resistance to oxidative stress [69] . We determined whether ets-4 null mutant animals show altered response to oxidative stress by monitoring survival when exposed to a powerful oxidant , paraquat . No significant differences in survival were seen between ets-4 ( ok165 ) and wild-type worms during the majority of the oxidative stress time-course ( Figure S8 ) , indicating that ets-4 ( ok165 ) worms display wild-type response to oxidative stress . The participation of ETS-4 in the oxidative stress response pathway was also tested by assessing the genetic interaction of ets-4 with skn-1 . Transcription factor SKN-1 , the ortholog of mammalian Nrf proteins , is critical for oxidative stress resistance and acts in multiple longevity pathways , including the insulin/IGF-1 signaling cascade [70] . To test whether SKN-1 was required to mediate the life span extension observed in ets-4 null mutant animals , we inhibited skn-1 activity by RNAi and monitored the adult life span at 25°C . skn-1 ( RNAi ) was previously reported to alter the expression of genes involved in oxidative stress response [71] . Also , inhibition of skn-1 by RNAi was shown to decrease the life span of daf-2 null mutant animals , but not control RNAi sensitive animals [72] . skn-1 ( RNAi ) on ets-4 ( ok165 ) worms did not alter the extended life span phenotype of these worms ( Figure S8 ) , suggesting that the ets-4 null mutant animals are not susceptible to a partial loss of SKN-1 function . Additionally , to determine whether ETS-4 and SKN-1 shared downstream effectors , we compared a list of SKN-1-dependent target genes involved in oxidative stress response [71] to the 70 ETS-4-regulated genes . There was no significant enrichment for the stress responsive , SKN-1-dependent genes amongst genes that act downstream of ETS-4 ( Table 2 ) . Taken together , these data suggest that ETS-4 does not contribute to all physiological processes regulated by insulin/IGF-1 signaling and supports a model whereby ETS-4 functions , in part , independently of DAF-2 signaling to regulate life span . The FOXO transcription factor , DAF-16 , is a well established regulator of life span that functions in the insulin/IGF-1 signaling pathway [1] , [60] , [73]–[75] and shares downstream longevity effectors with ETS-4 ( Table 2 and Figure 7 ) . We examined the genetic relationship between ets-4 and daf-16 . Consistent with previous studies [1] , [61] , [75] , inhibition of daf-16 activity by RNAi shortened life span ( Figure 6B and Table 1 ) . Further , ets-4 ( ok165 ) worms , when subjected to daf-16 ( RNAi ) , did not display an extended life span ( Figure 6B and Table 1 ) . Thus , daf-16 was required for the longevity phenotype of ets-4 ( ok165 ) worms suggesting that ets-4 functions upstream of , or in parallel to , daf-16 in lifespan regulation ( Figure 7 ) . Because the loss of ets-4 extends life span , whereas , inhibition of daf-16 activity shortens it , our genetic tests indicate that ets-4 antagonizes daf-16 function in longevity regulation by upstream effects , or in parallel . We examined in more detail two possible ways by which ets-4 could function upstream of daf-16 . First , we tested whether ETS-4 reduced DAF-16 levels through a transcriptional effect . Our gene expression profiling experiments indicated that the expression of daf-16 transcripts was not altered in ets-4 mutant worms ( NCBI's Gene Expression Omnibus , accession number GSE17954 ) . Further , upon crossing a daf-16::gfp reporter into ets-4 ( ok165 ) worms , we did not observe an alteration in the expression of DAF-16::GFP due to loss of ets-4 ( data not shown ) . Second , we examined whether ETS-4 antagonized DAF-16 activity by promoting its cytoplasmic retention . The function of the transcription factor DAF-16 is modulated in response to stress conditions , such as heat-shock , by altering its nuclear localization [58] , [76] . The intracellular localization of DAF-16::GFP was not affected by the loss of ets-4 under normal growth conditions and heat-shock ( data not shown ) . These results suggest that ETS-4 does not regulate DAF-16 by altering its expression level or nuclear-cytoplasmic localization . Alternatively , ETS-4 and DAF-16 could function in parallel pathways to modulate distinct targets involved in longevity regulation ( Figure 7 ) . Or , an indirect modulation of DAF-16 activity by ETS-4 is possible , through the regulation of a required transcription co-factor . Taken together , we conclude that ETS-4 is a new life span determinant that functions in parallel to the insulin/IGF-1 signaling pathway , but requires the FOXO transcription factor , daf-16 , to modulate life span .
Loss of ets-4 function led to a substantial extension in mean adult life span . Additionally , ets-4 mutant worms exhibited an altered egg-laying rate . These data are consistent with the precedence of several life span altering mutations that affect multiple aspects of nematode biology , including reproduction and metabolism . This raises the question of whether these phenotypes are causative of increased longevity . Another key question with the identification of ETS-4 as a novel genetic modifier of aging with a broad expression pattern was whether a particular cell type was crucial for its function in longevity regulation . Our data showed that restoring ETS-4 function specifically in the intestine , but not neurons , rescued the extended life span of ets-4 null mutant animals back down to wild-type levels . Interestingly , other transcription factors that function as longevity determinants , including the FOXO protein DAF-16 and the GATA factors ELT-3 , ELT-5 , and ELT-6 , regulate life span primarily through their function in the intestine [3] , [34] . These data thus also support our model of shared transcriptional targets for these factors . Moreover , restoring expression of ETS-4 in the intestine of null mutant worms does not rescue the altered egg-laying phenotype . Thus , the longevity and altered egg-laying rate phenotypes are separable , implying a correlative rather than causative relationship between the two . Because relative to one another , the rates of aging of different tissues appear normal , it is proposed that a network of signaling and feedback regulation is involved in coordinating aging within an animal [33] , [34] . Our data demonstrate that ETS-4 is a transcriptional regulator of this network that functions in the intestine to modulate the rate of aging in C . elegans . Gene expression profiling of long-lived worms carrying two different deletion alleles of ets-4 , identified a robust set of ETS-4-regulated genes . Consistent with the longevity phenotype of ets-4 mutant worms , the ETS-4-regulated gene set was enriched for genes that modulate life span ( Table 3 ) . In addition to the regulation of these known life span determinants by ETS-4 , 29 genes of unknown function were also misexpressed in ets-4 mutant animals and may contribute to the long-lived phenotype of these worms . DNA-binding studies and bioinformatics searches identified conserved ETS binding sites in promoter regions of 54 of the 70 ETS-4-regulated genes , suggesting that these were direct targets ( Table S5 ) . Notably , the ETS-4-regulated gene set was significantly enriched for genes that change expression during normal aging ( Figure 7 , Table 2 and Table S4 ) , the majority of which were proposed targets of the GATA factor ELT-3 [3] . Consistent with the shared pattern of expression changes observed between aged worms and insulin/IGF-1 signaling pathway mutant animals [3] , a significant proportion of ETS-4-regulated genes functioned downstream of the insulin/IGF-1 signaling pathway and the FOXO transcription factor DAF-16 [10] , [11] ( Figure 7 , Table 2 , and Table S4 ) . RNAi against four of these common downstream effectors , vit-5 , vit-2 , thn-1 , lys-7 alters worm life span [10] . These longevity effectors participate in lipid transport ( vitellogenins/yolk proteins ) and innate immune response ( lysozymes and thaumatins ) . In addition to this set of overlapping targets for three transcriptional regulators of aging ELT-3 , DAF-16 and ETS-4 , notable targets were absent . For example , expression of genes mediating response to oxidative stress regulated by the transcription factor SKN-1 ( superoxide dismutases ) [71] , heat-shock or toxicity ( metallothioneins and xenobiotic metabolism genes ) that are altered in daf-16 mutant worms remain unchanged in long-lived ets-4 mutant animals . We propose that ETS-4 participates in some , but not all biological processes regulated by FOXO and GATA factors . In summary , we introduce ETS-4 as a novel transcriptional regulator in the genetic network that modulates life span . Our study is the first demonstration of an ETS factor modulating animal life span , thus illustrating the utility of C . elegans in identifying novel functions for ETS proteins not easily discerned in more complex systems . The mammalian ortholog of ets-4 , SPDEF is expressed in the intestine and in tissues with high epithelial content , like breast and prostate [17] , [77] . A recent study of a mouse mutant strain with a dysfunctional SPDEF allele showed impaired terminal differentiation of specialized intestinal secretory cells derived from the intestinal epithelium [24] . Markers for these secretory intestinal cells , the Paneth and goblet cells , were implicated as SPDEF targets [24] . In an interesting evolutionary convergence , our study illustrated ETS-4 function in the worm intestine , which is a tube comprised of 20 large epithelial cells [78] . The C . elegans intestine executes multiple functions carried out by distinct organs in higher eukaryotes , such as digestion and absorption of nutrients , synthesis and storage of fats , initiation of an innate immune response to pathogens and yolk production [44] , [78]–[80] . Our work demonstrates the regulation of intestinal genes , such as lysozymes and vitellogenins by ETS-4 . Additionally , restoring ETS-4 function in the worm intestine , but not neurons , reduced the life span of ets-4 null worms to wild-type , revealing a link between ETS function in the intestine and longevity regulation . Our finding that ETS-4 has a role in C . elegans aging also provides a new connection between this ETS factor and cancer . Studies in human cell lines have implicated SPDEF function in tumorigenesis . Cell lines from breast and prostate tumors have altered SPDEF expression , although the significance of this misregulation remains controversial . Whereas some studies propose a putative role for SPDEF as a tumor suppressor , others suggest a prometastatic function [17] , [18] , [22] , [23] , [81] , [82] . Given the strong correlation between physiological aging and tumor susceptibility , aging studies in C . elegans have been used to provide genetic insights into tumor biology [83]–[85] . Thus , the role of ETS-4 in aging raises new implications for the physiological role of mammalian ETS factors in development , homeostasis and disease .
C . elegans strains were maintained at 20°C as described previously [86] unless otherwise mentioned . The wild-type reference strain was N2 Bristol . The RB637: ets-4 ( ok165 ) X strain was obtained from the C . elegans Gene Knockout Project at OMRF ( International C . elegans Gene Knockout Consortium ) . ets-4 ( uz1 ) was isolated by PCR-based screening of a library of worms mutated with EMS generated at the University of Utah . The deletion breakpoints of the uz1 on cosmid F22A3 are 14790 and 15861 . Single-worm PCR [87] was used for screening and to determine the genotype of worms during crosses with ets-4 ( uz1 ) and ets-4 ( ok165 ) worms . For ets-4 ( ok165 ) and ets-4 ( uz1 ) identification , nested-PCR was carried out using the following primers: Forward primers: CAATGAACGGTACTGGCTCAG and Primer P5: TGCAATCTTCCAATCCAACCC; Reverse primers: ACTGCCGGAGGACAAATGTC and Primer P6: CATTGCGATTCCCATGTAACC; Primers for sequences deleted in the ok165 and uz1 alleles: GCTAGCCAGCACCAACAATCAA and ACACCAAACGCTGCTTCTTT . The ets-4 ( ok165 ) and ets-4 ( uz1 ) worms were outcrossed to the N2 strain six times before phenotypic analysis , including life span assays . We also used BC11290: dpy-5 ( e907 ) I; sEx11290[rCesC04F6 . 1::GFP + pCeh361] [31] , TJ356: zIS356 IV [Pdaf-16::daf-16-gfp; rol-6 ( su1006 ) ] [76] , lin-15 ( n765ts ) [88] , daf-16 ( mgDf50 ) I , CF1041 daf-2 ( e1370 ) III [66] . RNAi was performed essentially as described previously [89] using HT115 ( DE3 ) E . coli expressing dsRNA from daf-16 , daf-2 , skn-1 ( obtained from Julie Ahringer's RNAi library [90] ) , akt-1 , akt-2 ( obtained from the Marc Vidal's RNAi library [91] ) , ets-4 or carrying the L4440 control plasmid ( EV for empty vector control ) . Each clone was sequenced to confirm its identity . Individual RNAi clones were grown overnight with Ampicillin ( 100 µg/ml ) or Kanamycin ( 25 µg/ml ) seeded onto NGM plates containing 1 mM IPTG ( Sigma ) and 25 µg/ml Carbenicillin ( Sigma ) and allowed to grow for 2 days at room temperature . For akt-1/2 ( RNAi ) , a 1:1 mix of akt-1 and akt-2 equal density overnight bacterial cultures was used to seed NGM plates containing 1 mM IPTG ( Sigma ) and 25 µg/ml Carbenicillin ( Sigma ) . L4 stage larvae were placed on the RNAi plates and observed for phenotypes at 25°C . Total RNA from mixed stage worms was isolated by phenol-chloroform extraction and subjected to DNaseI treatment using the RNeasy kit ( Qiagen ) . 1–2 µg of total RNA was reverse-transcribed by SuperScript III ( Invitrogen ) according to manufacturer's protocol . Ten percent of the resultant cDNA was PCR-amplified by Taq DNA polymerase in a 50 µl reaction . Primer sequences: R1: GGCACAAGTTGTACTGATGTC , P1 ( SL1 primer ) : GTTTAATTACCCAAG TTTGAG , P2: CAGATGACGGAGAATCAGGTC , R2: CTACAAGTTATAAGGAGGCAGG , P3: CTTCAGCCGCCTAGAAACTG and P4: CCAATATCTAGCCAGCAGGAG . The PCR product was sequenced to determine the position of SL1 attachment and splicing pattern . To assess the expression of mRNA in synchronized populations of N2 , ets-4 ( ok165 ) and ets-4 ( uz1 ) worms , levels of cDNA from the reverse transcription reaction were assessed by quantitative PCR according to manufacturer's protocol using the Roche LightCycler 480 . Transcript levels were normalized to the averaged levels of cdc-42 and pmp-3 [92] . See Table S6 for primer sequences . For microarray analysis , total RNA was isolated from L4 stage larvae as described above and reverse transcribed . Cy5 and Cy3 labeled cDNA was applied to C . elegans 22K gene expression arrays ( Agilent ) . Data from three independent repeats of the experiment were analyzed . Lowess-normalized log ( base 10 ) ratios of mutant/wild-type gene expression were obtained from the two-color Agilent C . elegans gene expression microarrays . The log ratios were analyzed using the Rank Products method [93] to identify consistently differentially expressed genes . Genes were selected using a probability of false prediction ( PFP ) cut off of 0 . 1 , i . e . a false discovery rate of 10% . This technique corrects for multiple testing with repeated trials on random permutations of the data set . The microarray data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus ( Edgar et al . , 2002 ) and are accessible through GEO Series accession number GSE21851 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE21851 ) . Life span assays were conducted as described previously [94] . The ets-4 ( ok165 ) and ets-4 ( uz1 ) worms were outcrossed to the wild-type strain six times before the life span assays . Briefly , worms were grown for two or more generations at the assay temperature ( 25°C or 20°C ) prior to the assay . Hermaphrodites were allowed to lay eggs for 6–8 hr on OP50 to obtain synchronous progeny for the experiment . L4 stage larvae were picked to plates spotted with RNAi bacteria containing 1 mM IPTG or OP50 bacteria and allowed to age at 25°C . The animals were moved to fresh plates daily during the reproductive period and every other day for the rest of the assay . The worms were scored for life by assessing movement to touch every 1–2 days [8] . Animals that bagged , exploded or crawled off the plate were excluded from the analysis . The first day of adulthood was counted as day one of the life span experiment . Life span curves and statistical data including p-values from Log-rank ( Mantel-Cox ) test were generated using GraphPad Prism version 5 software ( GraphPad Software , San Diego , California USA ) . Single L4 stage larvae were allowed to lay eggs at 20°C and transferred to a fresh NGM plate every day till the end of the reproductive period . The number of eggs laid and the number of hatched progeny were counted . The average number of eggs laid each day during the egg-laying period and the total number of progeny per worm ( brood size ) were plotted . Development time assays were done as described previously [95] . Briefly , synchronized L1 larvae of each genotype were grown at 20°C . The animals were monitored every 3 hr after they had reached L4 stage till they were pre-fertile adults . Each experiment with at least ten worms per genotype was repeated twice , independently . Unpaired t-test analyses were performed to calculate p-values . | Animal life span is regulated in response to developmental and environmental inputs through coordinate changes in gene expression . Thus , longevity determinants include DNA-binding proteins that regulate gene expression by controlling transcription . Here , we explored the physiological role of the transcriptional regulator , ETS-4 , in the roundworm Caenorhabditis elegans . Our data showed that worms that lack ETS-4 lived significantly longer , revealing ETS-4′s role in the transcription network that regulates life span . We identified 70 genes whose expression was modulated by ETS-4 that function in lipid transport , lipid metabolism and innate immunity . Some of the ETS-4-regulated genes were also controlled by two other regulators of aging , the FOXO and GATA factors . We concluded that a common set of transcriptional targets orchestrate the network of physiological factors that affect aging . ETS-4 is closely related to the human ETS protein SPDEF that exhibits aberrant expression in breast and prostate tumors . Because the genetic pathways that regulate aging are well conserved in other organisms , including humans , our findings could lead to a better understanding of SPDEF function and longevity regulation in mammals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genomics",
"genetics",
"and",
"genomics/animal",
"genetics",
"genetics",
"and",
"genomics/gene",
"discovery",
"molecular",
"biology/transcription",
"initiation",
"and",
"activation",
"genetics",
"and",
"genomics/gene",
"expression",
"developmental",
"biology/aging",
"developmental",
"biology/cell",
"differentiation",
"molecular",
"biology/bioinformatics",
"genetics",
"and",
"genomics/gene",
"function",
"biochemistry/transcription",
"and",
"translation",
"genetics",
"and",
"genomics",
"genetics",
"and",
"genomics/bioinformatics"
] | 2010 | ETS-4 Is a Transcriptional Regulator of Life Span in Caenorhabditis elegans |
Eradication of HIV infection will require the identification of all cellular reservoirs that harbor latent infection . Despite low or lack of CD4 receptor expression on Vδ2 T cells , infection of these cells has previously been reported . We found that upregulation of the CD4 receptor may render primary Vδ2 cells target for HIV infection in vitro and we propose that HIV-induced immune activation may allow infection of γδ T cells in vivo . We assessed the presence of latent HIV infection by measurements of DNA and outgrowth assays within Vδ2 cells in 18 aviremic patients on long-standing antiretroviral therapy . In 14 patients we recovered latent but replication-competent HIV from highly purified Vδ2 cells demonstrating that peripheral Vδ2 T cells are a previously unrecognized reservoir in which latent HIV infection is unexpectedly frequent .
The infecting HIV genome integrates into host chromatin where , transcriptionally silent and unaffected by antiretroviral therapy ( ART ) , it represents a major challenge towards efforts to eradicate infection [1] . Virologic latency is defined as durable , quiescent infection from which replication-competent HIV can emerge after cell activation . To date , resting memory CD4+ T lymphocytes are the major cell type in which latency has been documented in vivo [2–5] . However , efforts to eradicate HIV infection require the identification of all potential cellular reservoirs and therefore , while conventional αβ T cells , which include resting memory CD4+ T cells , constitute the major subpopulation of T lymphocytes , the γδ T cell population merits study as a potentially important site of latent infection . In the absence of pathological conditions such as infection , γδ T cells represent between 2 and 10% of total circulating CD3+ T lymphocytes [6] . Among peripheral CD3+ γδ T cells , those expressing a TCR formed by the Vγ9 and Vδ2 variable regions ( hereafter referred to as Vδ2 cells ) constitute up to 90% of γδ T cells [7] . These Vδ2 cells specifically recognize non-peptidic phosphorylated metabolites of isoprenoid biosynthesis , such as the potent activator ( E ) -4-hydroxy-3-methyl-but-2-enyl pyrophosphate ( HMBPP ) , present in most pathogenic bacteria [8 , 9] , or isopentenyl pyrophosphate ( IPP ) , produced also by the human mevalonate biosynthesis pathway [10] , but are not recognized by conventional αβ T cells . In vitro , both compounds have the same effect on γδ T cells , activating them . In adults , most Vδ2 cells are memory cells that can be further classified according to expression of CD45RO and CD27 surface markers [11–13] . During HIV-1 infection , the peripheral blood Vδ2 cell subset is depleted while Vδ1 cells are expanded , leading to an inversion of the Vδ2/Vδ1 ratio [14 , 15] . An indirect mechanism involving CCR5/α4β7 signalling was hypothesized to explain the specific depletion of Vδ2 cells [16] . However , an additional mechanism may be the direct infection of these cells as productive HIV infection of γδ T cells within peripheral blood mononuclear cells ( PBMC ) [17] as well as infection of γδ T cell clones by the CXCR4-tropic laboratory clone HIVLAI [18] has been reported . Similarly , SIV can infect both Vδ1 and Vδ2 cells , albeit infrequently [19] . Despite the documented capacity of HIV to infect γδ T cells , the potential of γδ T cells to serve as a persistent reservoir of infection has not been studied . Moreover , the memory phenotype of Vδ2 cells suggests that these cells could play a role as durable in vivo reservoirs of HIV infection . Using a viral outgrowth assay to detect latent but replication-competent HIV [20 , 21] , complemented by measures of HIV DNA , we demonstrate for the first time that peripheral Vδ2 cells in ART-treated patients with complete suppression of HIV plasma viremia harbour latent HIV that can replicate following ex vivo induction . We report the discovery of a new reservoir of HIV within peripheral Vδ2 cells , and suggest that infection in this population may be founded by immune activation that transiently upregulates the CD4 receptor on Vδ2 cells .
To study the role of Vδ2 cells as reservoirs of persistent , latent HIV infection , 18 HIV-infected male volunteers , who initiated ART in acute HIV infection ( AHI; n = 9 ) or in chronic HIV infection ( CHI; n = 9 ) and received stable ART for a median of 3 . 4 years [range 1 . 9–9 . 5] were studied . A comparison between AHI and CHI-treated patients’ characteristics at the time of study showed that CHI patients had , as expected , a statistically significant lower nadir CD4 count ( p = 0 . 017 ) and a significantly longer time on ART ( p = 0 . 004 ) . Median CD8+ T cell count was lower and pre-therapy plasma HIV RNA was higher in the AHI patients although these differences did not achieve statistical significance ( Table 1 ) . To ensure that other contaminating cells did not contribute to the recovery of HIV from isolated Vδ2 cells , we incubated freshly isolated patients’ PBMC with raltegravir and abacavir for 24 hours to avoid the possibility that de novo integration events could occur ex vivo after cell donation . γδ T cells were then enriched from PBMC using magnetic immunoaffinity beads , and non-activated ( HLA-DR- ) Vδ2 cells were further purified by FACS-sorting ( Fig 1A and 1B ) . This process excluded αβTCR+ cells ( classical CD4+ T cells ) from pre-sort samples ( Fig 1C ) , as detailed in Materials and Methods . To further confirm that Vδ2 cells were not already activated , aliquots of isolated Vδ2 cells were cultured in 5U/mL IL-2 prior to the addition of target cells in the viral outgrowth assay . HIV p24 measurements from these cultures were uniformly negative . Total HIV DNA levels were then quantified in isolated Vδ2 cells , unfractionated PBMC and total resting CD4+ T ( r-CD4 ) cells , when available ( Fig 2A ) in patients treated in AHI and CHI . As previously published in studies of other cell populations [22] , DNA levels varied widely but interestingly , Vδ2 cells showed the highest level of pol HIV DNA copies per 106 Vδ2 cells ( mean of 873 . 6 HIV copies/106 cells ) . Due to the low number of Vδ2 cells available for analysis the limit of quantitation of Vδ2 cells was 50 . 6 copies/106 cells , and 5 . 1 copies/106 cells for the other cell populations , where more cells could be analyzed . HIV DNA levels within Vδ2 cells were not statistically different between AHI and CHI-treated patients ( p = 0 . 37 ) . Within PBMC and r-CD4 cells HIV DNA levels were higher in CHI patients than in AHI patients , although this difference did not achieve statistical significance ( p = 0 . 06 for PBMC and p = 0 . 65 for resting CD4+ T cells ) . We recovered an average of 638 . 6 HIV DNA copies/million γδ T cells from CHI patients , and an average of 1108 . 7 copies/ million from AHI patient . Conversely , we recovered an average of 30 . 3 HIV DNA copies/ million rCD4 cells from CHI patients and an average of 21 . 4 copies/ 106 rCD4 cells from AHI patients . We calculated the contribution of HIV DNA in Vδ2 cells and r-CD4 cells to the total HIV-DNA+ PBMC ( Fig 2B ) as follows: First , we calculated the total HIV DNA copies in each cell population by multiplying the average HIV copy number per million cells to the percentage of Vδ2 or r-CD4 cells present in total PBMC . Then , this total HIV copy number was divided by the total HIV copy number in PBMC to obtain the proportion of HIV corresponding to Vδ2or r-CD4 populations . Vδ2 cells contributed 1 . 6% and 8 . 1% to the total HIV DNA copy numbers in PBMC of CHI and AHI patients , respectively . Resting CD4 T cells contributed 4 . 9% to the total HIV DNA copy number in PBMC of CHI patients and 1 . 9% in AHI patients . None of the comparisons between cell types or type of patients were statistically different . We performed viral outgrowth assays using highly purified Vδ2 cells to demonstrate the recovery of replication-competent HIV after , but not prior to , activation of Vδ2 cells from HIV-1 infected patients on fully suppressive ART . As expected , percentages of CD3+Vδ2 cells were lower in patients treated in CHI compared to patients treated in AHI ( mean 0 . 25% vs . 0 . 77%; Table 2 ) . Replication-competent HIV was detected in at least one culture replicate in 14 out of 18 patients ( 78% ) , with no virus recovered in two AHI and two CHI patients . Overall , for AHI patients we assayed 94 cultures of which 21 were positive for HIV p24 , compared to 53 total cultures for CHI patients with virus recovered in 20 ( 22 . 3% vs . 37 . 7%; Table 2 ) . Therefore , replication-competent virus was more frequently recovered from patients treated in CHI than from AHI-treated patients . In parallel to these assays and as part of other projects in the lab , viral outgrowth assays with isolated r-CD4 cells were performed as previously described [21 , 23 , 24] . A summary of the outgrowth assays for r-CD4 cells is included in Table 2 . Our results show that Vδ2 cells constitute a novel reservoir for HIV infection that may persist for years as latent Vδ2 cell infection was detected in CHI patients despite long-term suppressive ART and the lack of intermittent low-level viremia ( “blips” ) during the prior two years of clinical follow-up ( Table 1 ) . In addition , we analyzed two patients ( C . 2 and C . 3 ) one year after their first evaluation , and replication-competent HIV was again recovered from their Vδ2 cells ( S1 Fig ) . An expanded longitudinal analysis is underway to assess the stability of this reservoir . In ten of the 14 patients in whom virus was recovered ( A . 2 , A . 3 , A . 5 , A . 6 , A . 7 , C . 2 , C . 3 , C . 6 , C . 7 and C . 9 ) HIV was detected in cultures of only 5000 cells , suggesting a high frequency of infection within Vδ2 cells ( Fig 3 ) . Next , we calculated the frequency of infection expressed as infectious units per million ( IUPM ) isolated Vδ2 cells in 14 patients with available cell dilutions ( seven AHI and seven CHI ) ( Fig 4A ) . In addition , IUPM r-CD4 cells from the same patients were calculated to compare both cell populations . Percentages of r-CD4 cells , number of cells cultured and total number of positive and total cells assayed are shown in Table 2 . As expected , the confidence interval for Vδ2 cells was much greater than for r-CD4 cells due to the lower number of Vδ2 cells assayed and therefore the estimation of the frequency of infection is less accurate ( Fig 4B ) . However , we could not detect statistical differences when IUPM Vδ2 cells were compared to IUPM r-CD4 cells , suggesting that despite the inaccuracy of the co-culture system , Vδ2 cells may be frequently latently infected . Interestingly , while quantifying the frequency of infection within Vδ2 cells by limiting dilution assay , in some patients we found an unusual pattern of recovery of HIV with more positive wells at lower dilutions and no virus recovered when more cells were cultured ( Fig 3 ) . As γδ T cells possess innate , nonspecific antiviral function [25 , 26] , we hypothesized that an antiviral activity of the uninfected γδ T cells might reduce the recovery of HIV in cultures with high cell inputs , yielding the unusual pattern of viral outgrowth seen in some patients . To test this hypothesis , we performed viral inhibition assays , co-culturing Vδ2 cells from a healthy , uninfected donor , with autologous CD4+ T cells that had been HIV-infected ex vivo at ratios of 1 CD4 cell and 0 . 1 or 0 . 01 γδ cells . Vδ2 T cells inhibited HIV production from infected CD4+ T cells , with increased inhibition seen at higher cell inputs in the co-culture system ( Fig 5 ) . In addition , in some wells we blocked the cytotoxic activity of γδ T cells by pre-incubating the isolated Vδ2 cells with a cocktail of antibodies against CD8 , NKG2D and CD16 ( Fig 5 ) . HIV p24 production was 74 . 6% inhibited in the 1:0 . 1 ratio conditions and 41 . 8% in the 1:0 . 01 conditions . When the cocktail of antibodies was used , these percentages decreased to 26 . 9% and 15 . 5% in the 1:0 . 1 and 1:0 . 01 ratios , respectively . Isolated Vδ2 cells can be infected in vitro , [17 , 18] ( S2 Fig ) despite low or absent surface CD4 receptor expression prior to activation and HIV infection of these cells is inhibited by CD4 blockade [17] , ( S2 Fig ) . We further investigated the CD4-dependence of HIV infection in Vδ2 cells . Total PBMC from uninfected donors were activated with IL-2 alone or IPP and IL-2 , and surface marker expression was analyzed by flow cytometry ( Fig 6A and S3 Fig ) . As expected , CD4 receptor expression was detected on <0 . 3% of Vδ2 cells at day 0 , but became detectable on up to 25% of cells after six days of culture in the presence of IL-2 alone , or IPP and IL-2 . All increases were statistically significant ( p <0 . 01; Fig 6A ) . Interestingly , treatment with exogenous IL-2 alone induced CD4 expression to similar levels as did IPP and IL-2 ( mean 15 . 3% and 15 . 9% , respectively ) . In contrast , none of these conditions significantly increased the surface expression of CCR5 after six days in culture ( Fig 6A ) . In addition , the activation status of Vδ2 cells was also assessed in the same cells by analyzing the expression of the MHC Class II HLA-DR receptor , the IL2 receptor alpha chain CD25 , and the activation marker CD38 ( Fig 6B ) . After six days in culture with IL-2 alone or IPP and IL-2 the expression of these markers was significantly increased ( p <0 . 05 in all cases ) although treatment with IL-2 alone induced activation in no more than 20% of Vδ2 cells ( Fig 6B ) . Based on these results we hypothesized that immune activation , driven in this instance by HIV infection , might upregulate CD4 expression on Vδ2 cells in vivo . To test this hypothesis we measured surface expression of CD4 and CCR5 expression in Vδ2 cells donated by three viremic patients diagnosed during the acute phase of HIV infection , prior to the initiation of ART ( Fig 7 ) . Based on history and diagnostic testing , the estimated date of infection in these patients was less than 23 days prior to sampling . Likely related to the pathological immune activation of acute HIV infection , 9 . 5% , 15 . 6% and 15 . 9% of Vδ2 cells expressed CD4 ( Fig 7A ) , as compared to <0 . 3% of Vδ2 cells in healthy donors . In addition , we also analyzed the percentage of Vδ2 cells that coexpressed CD4 and CCR5 ( Fig 7B ) . This observation in these unique patients supports our hypothesis that pathological immune activation in early HIV infection promotes the upregulation of CD4 expression in Vδ2 cells , making them targets for HIV infection in vivo .
We have discovered that peripheral resting Vγ9Vδ2 cells can act as a cellular reservoir of persistent , latent HIV infection . Using a gold-standard coculture assay that defines the presence of latent infection , we find that the frequency of replication-competent virus in these cells is substantial , with virus recovered in as few as 5000 Vδ2 T cells that lack activation markers . This unexpected finding is supported by the detection of HIV DNA within this cell population , implying that at least a fraction of the HIV DNA detected within Vδ2 cells represents replication-competent virus . We propose a mechanism to explain the infection of Vδ2 cells despite the absence of CD4 expression in their surface . As isolated Vδ2 cells can be infected in vitro we hypothesize that the activated immune environment during untreated HIV infection induces transient CD4 upregulation , rendering Vδ2 cells permissive for HIV-1 infection , and founding a substantial population of latently infected γδ cells . This hypothesis is supported in vivo by the detection of expression of CD4 and CCR5 in Vδ2 cells in untreated patients in early acute HIV infection . Latent infection of γδ cells could occur at other times as well , when cellular activation results in upregulation of CD4 expression . Recovery of purified Vδ2 cells was reduced in patients treated in CHI , compared to patients treated in AHI , as there is a dramatic loss of Vδ2 cells early after infection with HIV [14 , 15 , 18] . Such depletion is only partially reversed by ART [27] . Remarkably , replication-competent HIV was quantified in purified Vδ2 cells in 77% of patients who had been treated and suppressed for a median of nearly four years . Although γδ T cells represent only a small fraction of the total CD3+ T lymphocytes , frequency of infection within isolated Vδ2 cells was not statistically different from that in r-CD4 cells . However , it is important to highlight that this estimation is not as accurate as the estimation for r-CD4 cells , as reflected by the 95% confidence intervals , because a significantly lower number of cells were assayed . It is also important to note that maximum activation conditions were differently assayed for both cell populations . Our results may overestimate the frequency of infection in γδ cells compared to r-CD4 cell calculations , as viral outgrowth starts during the first 24 hours of incubation , when γδ cells can spread the virus to the γδ-CD8-depleted-PBMC , required for Vδ2 cell activation . In addition , our data suggest the possibility that the ratio of defective to replication-competent proviruses might be lower in Vδ2 cells although future comprehensive sequence analysis will be required to fully address this possibility . A relative deficiency of restriction factors such as APOBEC3G/3F [28 , 29] in γδ T cells might explain these findings . In addition , a recent study has shown that HIV DNA might be present in cells other than conventional CD4+ T cells and myeloid cells suggesting that HIV may persist in other cell types [30] . As γδ cells can be phenotypically classified using CD45RA and CD27 markers , it is possible that in some prior studies these cells may have been included in the evaluation of HIV infection within the latently infected resting memory cell population . Interestingly , our results and others [22] have shown that in some patients , HIV DNA levels within r-CD4 cells are lower than in total PBMC , suggesting that in those patients , other cell types might have a high contribution to total HIV DNA measurements . In addition , activated T cells have also been reported to have a greater HIV DNA contribution than r-CD4 cells [23 , 24] . Although we found higher levels of HIV DNA within isolated γδ T cells , the rarity of these cells within PBMC makes them to less frequent contributors to the total reservoir than r-CD4 cells . Of note , the custom antibody cocktail used to isolate r-CD4 cells does not completely deplete the γδ T cells and therefore comparisons between r-CD4 cells and γδ T cells are not totally accurate . We have begun studies to evaluate the exact contribution of γδ T cells in assays using total r-CD4 cells . Although we recovered HIV from low numbers of γδ T cells in 14 of 18 patients , in several patients the recovery of replication-competent virus was less than expected in cultures with higher Vδ2 cell input . In these cases , estimates of the frequency of infection cannot be made with these data as such estimates depend on the assumption of a normal distribution of infection . We speculate that in these cultures , sufficient Vδ2 cells were present to exert potent antiviral activity [25 , 26] , leading to inhibition of spread of the virus during the outgrowth phase of the assay . In support of this , we demonstrate a dose-dependent inhibition of the p24 production in culture when isolated in vitro-HIV-infected CD4 cells were cocultured with Vδ2 cells . We also show that blocking CD16 , NKG2D and CD8 receptors , Vδ2 cell cytotoxic capacity is partially inhibited suggesting that other receptors are also involved in exerting this function . HIV can infect isolated γδ T cells in vitro [17 , 18] . In this study , we have shown that this occurs through a mechanism that involves the transient upregulation of the CD4 receptor after activation . CD4 expression is upregulated in Vδ2 cells in vitro following activation with IPP and IL-2 . This transient upregulation of CD4 has also been reported after infection of γδ T cells with human herpesvirus [31] . Treatment with HMBPP and IL-2 also induced the expression of CD4 . Although the vast majority of Vδ2 lymphocytes do not express CD4 , in the setting of lymphopenia , rapid T-cell turnover , or heightened immune activation , increased IL-2 levels could lead to CD4 upregulation in Vδ2 cells , making them susceptible to HIV infection in vivo . We directly observed this phenomenon prior to ART in viremic , newly infected patients , suggesting that this mechanism is plausible in vivo . Therefore , in addition to a previously reported indirect mechanism to explain peripheral Vδ2 cell depletion [16] , we describe a potential direct effect of HIV infection on Vδ2 cell depletion . Direct infection of Vδ2 cells by HIV could lead to depletion of most Vδ2 , while others might survive and establish latent infection . Infection of Vδ2 cells can have significant consequences , as these cells constitute an important bridge between the innate and adaptive immune response [32] . Therefore , due to reduced Vδ2 cell signaling , dendritic cell function [33–36] , follicular B cell help , or CD4+ T cell responses [37] might be impaired . In summary , our results demonstrate that Vδ2 cells are a novel latent reservoir for replication-competent HIV . Although these T lymphocytes are generally rare , the frequency of latent infection in these cells makes it likely that they contribute measurably to the total burden of latent , quiescent HIV infection . Moreover , we demonstrate that isolated Vδ2 cells can be infected in vitro , and illustrate a mechanism that could allow γδ T cell infection despite constitutive low expression of the CD4 receptor . We found that activation of Vδ2 cells upregulates CD4 expression , enabling HIV infection . The durability of latent infection within this novel cell population must still be established in longitudinal studies . However , given the broad efforts to discover reagents that disrupt latency in resting CD4 central memory cells as a first step towards viral eradication therapies , it may be necessary to also address the responsiveness of proviral HIV genomes within resting γδ T cells to such strategies .
All patients provided written informed consent , and studies were approved by the UNC Institutional Review Board . Our criteria to define and select patients treated in acute HIV infection ( AHI ) have previously been reported [38 , 39] . Briefly , patients identified in AHI ( plasma HIV RNA positive and HIV Western blot negative ) were enrolled and initiated ART within 45 days of the estimated date of infection . Serial measurements of plasma viremia and CD4+ T cell count were performed , and when patients were aviremic ( <50 HIV RNA copies/ml ) on ART for >6 months , cells were obtained by continuous-flow leukapheresis . Patients studied after initiation of ART in chronic HIV infection ( CHI ) had a history of stable , successful treatment , and plasma HIV-1 RNA levels <50 copies/mL for >2 years without blips . Buffy coats from uninfected donor volunteers were obtained from the New York Blood Center ( Long Island City , NY , USA ) . As part of the routine preparation for the outgrowth assay , prior to γδ T cell isolation , freshly isolated patients’ PBMC were incubated with raltegravir and abacavir for 24 hours to avoid any potential de novo infection due to HIV reactivation . For infectivity assays , isolated γδ T cells from fresh PBMC from healthy non-HIV infected were used . γδ T cells were first enriched by negative selection using a commercially available cocktail containing monoclonal antibodies ( mAb ) directed against non-γδ cells , including antibodies against granulocytes , red blood cells , dendritic cells , pan-αβ T cells , NK cells , stem cells , monocytes ( StemCell Technologies , Vancouver , Canada ) and afterwards the cells were isolated by fluorescent activated cell sorting ( FACS ) using a Reflection sorter ( iCyt , Champagne , IL , USA ) or a FACSAria II ( BD ) . Each sorting experiment was validated performing instrument quality controls , and running isotype controls and fluorescence minus one control . In addition , gating strategy for each specific experiment was based on the selection of the target cell by CD3 ( clone SK7 , BD ) and Vδ2 ( clone B6 , Biolegend , San Diego , CA ) mAbs , and exclusion of potential contamination of CD4+ T cells in a third channel using the CD4 mAb ( clone RPA-T4 , BD ) , and in some experiments αβTCR mAb ( clone MOPC-21 , Biolegend ) . To identify singlets we performed a 2-step doublet discrimination using HxW of the pulse; first in the SSC direction then in the FSC direction . In all our preparations the singlets were always well separated from the doublets and no contamination with αβ T cells or CD4+ T cells was detected in the presort sample ( Fig 1 ) . In addition , HLA-DR mAb ( clone TU36 , BD ) was used to exclude potential pre-activated γδ T cells , and fixable aqua ( Life Technologies , Grand Island , NY ) was used to exclude dead cells . Cells were collected in RPMI-1640 containing stable Glutamine and Hepes ( Gibco , Life Technologies , Grand Island , NY ) , 10% pooled human AB serum ( Sigma-Aldrich , St Louis , MO ) and 10% PenStrep ( Sigma-Aldrich ) ( hereafter referred to as γδ medium ) . After isolation , an aliquot was used to analyze the purity of the sorted population ( > 99% γδ T cells , < 1% all other cells ) ( Fig 1 ) . After sorting , Vδ2 T cells were centrifuged and resuspended in the suitable volume of γδ medium to perform the standardized co-culture assay [20 , 21] . As part of the standard method , cells were plated in limiting dilution , when possible , and activated with 100nM HMBPP ( kindly provided by Dr . H . Jomaa , Justus-Liebig University , Giessen , Germany ) or 1μM IPP ( Sigma ) and 100U/mL IL-2 ( Peprotech , Rocky Hill , CT ) for 24 hours . In addition , as γδ T cell activation requires the presence of APC [40] but γδ T cells and CD8+ T cells possess anti-HIV activity [26] , allogeneic non-HIV infected PBMC used as APC were first depleted of γδ T cells and CD8+ T cells , added to cultures at a 1:4 ratio ( isolated γδ T cells: PBMC depleted ) , and incubated for 24 hours at 37°C and 5% CO2 . Vδ2 cells were then washed and resuspended in γδ medium containing 10U/mL IL-2 and cocultured with allogeneic PHA-activated PBMC depleted of CD8+ T cells as target cells . CD8+ T cells were depleted by negative magnetic isolation following manufacturer’s instructions ( StemCell Technologies ) . Complete medium containing 10U/mL IL-2 was replaced and refreshed every three or four days . Supernatants were stored at -80°C for analysis of viral p24 production by ELISA ( ABLinc , Rockville , MA , USA ) at days 15 , and 19 . As a control Vδ2 cells were cultured without activation prior to the addition of target cells in 5U/mL IL-2 , following the same coculture protocol , which were uniformly negative . Quantitative viral outgrowth assay for r-CD4 cells was performed in parallel using the standard procedure previously reported in our lab [21 , 23 , 24] and described above also for γδ cells . However , activation procedures were different between both cell populations . Briefly , magnetically isolated r-CD4 cells were activated with 2μg/mL PHA , 60U/mL IL-2 and irradiated PBMC from a non-infected donor , and target cells were added as previously reported and as explained above . Infectious units per million Vδ2 and r-CD4 cells ( IUPM ) were calculated using the R software developed at the University of North Carolina that allows calculating the point estimate of the frequency of infection and the 95% confidence interval . Total HIV pol DNA copies within isolated Vδ2 cells , unfractionated PBMC and resting CD4+ T cells were quantified by droplet digital PCR ( ddPCR ) using primers , probes and conditions previously reported [41] . Briefly , DNA was extracted from frozen cell pellets of 1x105 Vδ2 cells and 1x106 PBMC and r-CD4 cells on average , using the Qiagen DNeasy Blood and Tissue kit ( Qiagen , Maryland , USA ) and concentration of DNA was measured using the Nanodrop ( Thermo Scientific ) . PCR reactions were loaded into the Bio-Rad QX-100 droplet generator . Each reaction consisted of a 20μL mix containing 10 μL ddPCR Probe Supermix , 900 nM primers , 250 nM probe , and template DNA . Following amplification in a standard thermo cycler ( 10 min . at 95°C , 40 cycles of 30 sec . at 94°C , 60 sec . at 58°C and final 10 min . at 98°C ) droplets were immediately analyzed as positive or negative using the Bio-Rad QX-100 droplet reader . The no template controls were used to set the threshold , and copy number was calculated using the manufacturer’s software and normalized to the total number of cells ( Bio-Rad QuantaSoft v . 1 . 2 ) . As a control , dilutions of DNA from J89 cells , which contain a single copy of HIV , was used in all runs . In addition , to control for the different number of cells assayed , we run independent assays to compare HIV DNA levels in PBMC samples of 1x105 cells and 1x106 cells , which showed similar copy number . Samples were run in triplicate and the limit of quantitation ( LOQ ) was calculated based upon the frequency of false positives in the no template control . We calculated the contribution of HIV DNA in Vδ2 cells and r-CD4 cells to the total HIV-DNA+ PBMC as follows: First , we calculated the total HIV DNA copies in each cell population by multiplying the average HIV copy number per million cells to the percentage of Vδ2 or r-CD4 cells present in total PBMC . Then , this total HIV copy number was divided by the total HIV copy number in PBMC to obtain the proportion of HIV corresponding to Vδ2or r-CD4 populations . Vδ2 cells contributed 1 . 6% and 8 . 1% to the total HIV DNA copy numbers in PBMC of CHI and AHI patients , respectively . Resting CD4 T cells contributed 4 . 9% to the total HIV DNA copy number in PBMC of CHI patients and 1 . 9% in AHI patients . None of the comparisons between cell types or type of patients were statistically different . Freshly isolated Vδ2 cells from HIV-negative volunteers were activated using 100nM HMBPP and 100U/mL IL-2 for 24 hours and spinoculated for 2 hours with 1ng/mL of the CCR5 JR-CSF strain . Cells were then washed twice to remove the excess of virus , cultured in triplicate in γδ medium containing 20U/mL of IL-2 for 7 days and supernatants were analyzed for HIV p24 production on days 4 and 7 . As a control for the infection conditions , isolated CD4+ T cells from the same uninfected blood donor volunteer were stimulated in parallel using 2μg/mL PHA and 60U/mL IL-2 . Supernatants from day 0 ( basal levels after virus exposure ) were also stored and measured as a negative control . In some wells , Vδ2 cells and CD4 cells were incubated with 50μg/mL of an anti-CD4 mAb ( clone RPA-T4 , BD ) before exposure to HIV . Experiment was repeated using three different donors . Vδ2 and CD4+ T cells were FACS-sorted from PBMC of healthy donor volunteers . CD4 cells were activated with 2μg/mL PHA and 60U/mL IL-2 for 24h washed and infected by spinoculation with the viral strain JR-CSF following the same protocol described above , and co-cultured in triplicate at different ratios of autologous Vδ2 cells ( 1:0 . 1 and 1:0 . 01 CD4:Vδ2 ) . In some wells cytotoxic activity of γδ cells was blocked using a mixture of purified mAb against CD8 and CD16 from Biolegend , and NKG2D from Miltenyi Biotec . Media was refreshed at days 4 and 7 and supernatants stored until p24 ELISA quantification ( ABLinc , Rockville , MA , USA ) . Experiments were performed in three different donors . PBMC from healthy donor volunteers isolated from fresh buffy coats were incubated with 1μM IPP and 100U/mL IL-2 or 100U/mL IL-2 alone . Expression of CD4 and CCR5 receptors along with expression of activation markers HLA-DR , CD25 and CD38 , was controlled by flow cytometry at days 0 and six of culture . mAb used were; CD4-FITC ( clone RPTA-4 , BD ) , Vδ2-PE ( clone B6 , Biolegend ) , CCR5-V450 ( clone 2D7/CCR5 , BD ) , HLA-DR-FITC ( clone TU36 , BD ) , CD25-V450 ( clone M-A251 , BD ) and CD38-PerCPCy5 . 5 ( clone HIT2 , BD ) . Briefly , cells were blocked with FBS ( Sigma ) for 10 minutes on ice , resuspended in staining buffer ( PBS-2% FBS ) , incubated for 20 minutes on ice in the dark with combinations of the mAb , or suitable isotype controls , and washed twice . Cells were then fixed with 2% paraformaldehyde solution and analyzed in the Attune acoustic cytometer ( Applied Biosystems ) . Differences between patients treated in AHI and patients treated in CHI were compared using the two-tailed Mann-Whitney U-test and comparisons between mean values were performed by the two-tailed Student t-test . Statistical Analyses were performed using the IBM-SPSS version 21 . 0 ( Chicago , Illinois , USA ) and p values <0 . 05 were considered statistically significant . | Antiretroviral therapy ( ART ) has led to a decreased HIV-related morbidity and mortality across the world . While successful ART restores health , it does not cure infection as latent HIV-1 remains integrated within different cell populations , unaffected by ART . To date resting memory CD4+ T cells are the best-characterized cellular reservoir . However , eradication of HIV-1 infection requires the description of all latent cellular reservoirs harboring replication-competent HIV-1 . We describe the discovery of an unexpected cellular reservoir within γδ T lymphocytes . This novel reservoir must be considered as strategies to clear latent HIV are developed and tested . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Peripheral Vγ9Vδ2 T Cells Are a Novel Reservoir of Latent HIV Infection |
Systemic lupus erythematosus ( SLE ) is an autoimmune disease with known genetic , epigenetic , and environmental risk factors . To assess the role of DNA methylation in SLE , we collected CD4+ T-cells , CD19+ B-cells , and CD14+ monocytes from 49 SLE patients and 58 controls , and performed genome-wide DNA methylation analysis with Illumina Methylation450 microarrays . We identified 166 CpGs in B-cells , 97 CpGs in monocytes , and 1 , 033 CpGs in T-cells with highly significant changes in DNA methylation levels ( p<1×10−8 ) among SLE patients . Common to all three cell-types were widespread and severe hypomethylation events near genes involved in interferon signaling ( type I ) . These interferon-related changes were apparent in patients collected during active and quiescent stages of the disease , suggesting that epigenetically-mediated hypersensitivity to interferon persists beyond acute stages of the disease and is independent of circulating interferon levels . This interferon hypersensitivity was apparent in memory , naïve and regulatory T-cells , suggesting that this epigenetic state in lupus patients is established in progenitor cell populations . We also identified a widespread , but lower amplitude shift in methylation in CD4+ T-cells ( >16 , 000 CpGs at FDR<1% ) near genes involved in cell division and MAPK signaling . These cell type-specific effects are consistent with disease-specific changes in the composition of the CD4+ population and suggest that shifts in the proportion of CD4+ subtypes can be monitored at CpGs with subtype-specific DNA methylation patterns .
Systemic lupus erythematosus ( SLE ) is a complex autoimmune disease characterized by an impaired clearance of apoptotic cells , the production of auto-antibodies against nuclear antigens , and the deposition of immune complexes that lead to tissue damage in multiple organs . SLE patients suffer from chronic dermatological , musculoskeletal , renal , and cardiovascular problems , and like many autoimmune diseases , these symptoms typically worsen during periods of active disease , called flares , and improve during quiescent phases of the disease . SLE predominantly affects females ( ∼90% of cases ) , and is more prevalent in individuals of African descent [1] . SLE is known to have a strong genetic basis , with high sibling risk ratios ( λs>8 ) and higher concordance among monozygotic twins compared to dizygotic twins or full siblings [2]–[4] . Recent genetic studies , including genome-wide association studies , have identified multiple common genetic risk factors , the strongest of which are in the MHC region of chromosome 6 , but also include ITGAM , IRF5 , STAT4 , and at least twenty other genes [5]–[10] . While a few rare variants of strong effect have been identified , the currently favored hypothesis is one of complex etiology involving multiple genetic and environmental risk factors . Given the complex nature of SLE etiology , epigenetic analyses are likely to provide new insights into the disease , as chromatin structure and DNA methylation patterns are influenced both by the inherited DNA sequence and by environmental exposures . In fact , the importance of DNA methylation in lupus has been appreciated for over 20 years . T-cells from patients with SLE have reduced expression of DNA methyltransferases [11] , and DNA methylation inhibitors like 5-azacytidine can induce T-cell autoreactivity and lupus symptoms in mice [12] . Furthermore , drug-induced lupus is associated with reduced DNA methylation and aberrant expression of DNA methyltransferases [13] . A few recent studies have been published on genome-wide analyses of DNA methylation patterns in SLE . These include studies of a few thousand CpGs in CD4+ T-cells from discordant monozygotic twins [14] , and either buffy coat DNA or sorted CD4+ T-cells from unrelated individuals [15] , [16] . Here , we report the most comprehensive study to date of SLE epigenetics , where we have analyzed >460 , 000 CpGs , covering >95% of known genes , in CD4+ T-cells , CD19+ B-cells and CD14+ monocytes . Our results uncover a profound hypomethylation of genes regulated by interferon ( type I ) that is present in patients during and after flares , suggesting that this epigenetic state persists beyond stages when circulating interferon levels are at their highest . Our results also suggest a compositional remodeling of the CD4+ T-cell population in SLE patients that can be observed in DNA methylation patterns .
To identify common functional characteristics of genes with aberrant DNA methylation in SLE patients , we performed DAVID Panther GO term analysis and Ingenuity Pathway Analysis ( IPA ) on the genes that were proximal to each of the most significant ( top 100 ) CpGs in each cell type . Both analyses clearly identified interferon signaling as a common feature of the genes showing the most significant changes in methylation among SLE patients . Table 1 and Table S7 list the results of these analyses , including the top Panther GO terms and IPA canonical pathways . IPA also indicated the type-I interferon IFNA2 ( interferon alpha 2 ) as a common upstream regulator , so we suspect that type-I interferon pathways are the targets of the epigenetic changes in lupus . However , IL-29 ( IFNL1 ) was also significant as a potential upstream regulator of these genes , so it is possible that both type-I and type-III interferons are contributing to the epigenetic patterns we observed . As Panther and IPA use different gene annotations , their lists of interferon-regulated genes are not identical . Furthermore , many of the putative interferon-inducible genes ( IFI44 , IFITM1 , etc . ) are not always properly annotated with interferon GO terms . When we combined the gene lists from each software package with type-I interferon annotations and included these “IFI” genes , we found that at least half of the top 50 most significant CpGs in each cell type were proximal to genes involved in interferon signaling ( 50% in T-cells , 60% in B-cells , and 54% in monocytes ) . This represents more than 125 fold enrichment over the ∼0 . 4% of autosomal CpGs represented on the Methylation450 array that are adjacent to interferon type-I genes ( Fisher's exact test p<5×10−46 ) . Remarkably , of the 63 CpGs in T-cells , 58 CpGs in B-cells , and 23 CpGs in monocytes that had highly significant changes in methylation ( p < 1×10−8 ) near an interferon type-I regulated gene , only 1 CpG in B-cells , located at the 3′ end of STAT3 , was hypermethylated in SLE patients . Every other highly significant methylation change near an interferon gene was a hypomethylation effect . This widespread hypomethylation suggests that the primary methylation defect in SLE is a hyper-sensitization of interferon signaling pathways , and this is consistent with gene expression studies that have shown an overexpression of interferon-regulated genes in SLE patients , particularly during flares of the disease [17]–[19] . In addition to the most significant CpGs , we also performed a separate functional analysis of genes in the second phase of the T-cell p-value distribution ( limited to p-values between 1×10−8 and 1×10−11 ) , where we suspect a secondary phenomenon . Both Panther and IPA analyses indicated that these genes were enriched for functions associated with cell division and cancer . IPA specifically identified the p38 mitogen-activated protein kinase pathway as a common feature of these genes , a pathway that has been linked to autoimmune diseases , including SLE [20] . This functional difference between the two phases of the p-value distribution , in addition to the enrichment for hypermethylation effects and intermediate mean methylation levels , is further evidence that two independent phenomena were present in T-cells . Previous reports of increased expression of interferon-regulated genes in SLE patients have indicated that this effect is primarily observed during active phases of the disease , while those patients in quiescent phases have normal levels of expression . This observation coincides well with reports that circulating interferon levels correlate with disease activity [21] . We compared the DNA methylation levels between our active and quiescent SLE patients to identify activity-dependent methylation in these patients that might coincide with this gene expression effect . We performed regression analysis in a case-case comparison of flare versus quiescent SLE patients . As seen in the QQ plot from these association tests ( Figure 3A ) , we found no significant differences between these groups . Regression analyses of methylation versus SLEDAI scores as continuous values were also negative ( data not shown ) . Even the strong hypomethylation at interferon-regulated genes was similar in active and quiescent patients ( Figure 3B ) , with no statistically significant difference between the disease groups in any cell type . These results indicate that the methylation changes in SLE persist beyond flares and may be maintained for many months after interferon levels normalize . It also indicates that SLE patients in quiescent stages remain poised for interferon response at an epigenetic level , with a significant number of immune cells carrying this phenotype . One possible explanation for the persistent hypomethylation of interferon-regulated genes could be the endurance of memory cells that carry this epigenetic state since the last flare . Furthermore , some of the methylation changes we observed might be specific to T-cell subtypes , rather than a general feature of the CD4+ pool . To examine these possibilities , we collected CD4+ T-cells from an independent cohort of 26 SLE patients and 18 controls , and further sorted a fraction of these into CD45RA+RO− naïve , CD45RA-RO+ memory , and CD25+CD127− regulatory T-cells . To ensure that this independent validation set recapitulated the results from our initial cohort , we re-tested for SLE-related methylation changes in CD4+ T-cells using our regression model at 1 , 031 CpGs that were highly significant in the initial cohort ( 2 of the original 1 , 033 failed QC in the validation set ) . Despite the smaller size of the validation set , 76 . 8% of the CpGs were significant at p<0 . 01 in these validation tests ( see Figure 4A , black line , and 4B , gray bar ) . Furthermore , a comparison of the direction and amplitude of the changes in methylation observed in the SLE patients' T-cells indicated a very high correlation with the initial cohort ( R2 = 0 . 92 , see Figure 5A ) . These tests strongly validate our initial findings in an independent cohort . We next tested for SLE-related methylation changes in the sorted T-cell subsets from the same individuals from our validation cohort . If any of the observed methylation changes were specific to memory , naïve , or regulatory T-cells , the enrichment of these cell types should reveal a stronger effect than is seen in the CD4+ pool as a whole . However , when we ran our regression tests on the same 1 , 031 CpGs in the sorted subsets , the distribution of p-values indicated much weaker effects than those seen in the CD4+ pool ( Figure 4A ) . The number of CpGs that validated at p<0 . 01 was less than 38% in all three sorted subtypes , or approximately half of that observed in the CD4+ pool from the same individuals ( Figure 4B ) . Furthermore , the correlations in direction and amplitude of the SLE-related methylation changes were weaker in the sorted CD4+ subtypes , where the R2 dropped below 0 . 70 for each sorted cell type ( Figure 5 B–D ) . When we limited our analysis to only those significant CpGs near interferon-regulated genes , the trend was dramatically different . The number of these CpGs that validated at p<0 . 01 was similar in the sorted subtypes ( 71% in naïve and memory , 81% in regulatory ) compared to the CD4+ pool as a whole ( 87% ) ( Figure 4B ) . Furthermore , the direction and amplitude correlations with the initial CD4+ results were stronger at the interferon CpGs than at the non-interferon CpGs ( Figure 5 B–D , red dots ) , but no stronger than the correlations observed in the CD4+ population as a whole . This suggests that the methylation changes we observed at interferon-regulated genes in CD4+ T-cells are intrinsic to memory , naïve , and regulatory T-cells , but not specific to any one population . So , it is unlikely that the persistence of these changes during quiescent stages of SLE can be explained simply by the endurance of memory cells . Furthermore , since the milder changes in methylation at non-interferon loci that were observed in the CD4+ T-cells , appear to be absent or greatly diminished in the sorted subtypes , the observed differences in methylation are not likely to be intrinsic to memory , naïve or regulatory T-cells . Thus , the most likely explanation for the widespread , moderate changes at thousands of CpGs in the CD4+ T-cells is a change in the composition of the CD4+ pool . Changes in the proportions of CD4+ subtypes in SLE patients would generate disease associations at any loci that had cell type-specific methylation patterns , and as we have observed , these loci would likely have intermediate mean methylation levels due to the mixture of these cell types in the CD4+ population . For example , a 10% methylation difference between SLE patients and controls could be due to a 50% difference in methylation within a CD4+ subtype that makes up 20% of CD4+ cells . Conversely , the same 10% methylation difference could be generated if that same subtype dropped in number among SLE patients to alter the composition of the CD4+ population . Any CpG with a subtype-specific methylation pattern would show this trend . Our data on sorted CD4+ subsets is consistent with the latter , as we observe a reduction , rather than an enrichment of the SLE-control methylation differences , as we purify CD4+ subtypes . The analysis of X-chromosome methylation is hampered by the inherent differences in methylation between males and females , so our disease association tests on this chromosome were limited to females , in which we have the largest sample size . For this reason , it is difficult to compare test statistics to those at the autosomal CpGs . Nonetheless , we ran regression tests at 11 , 122 X-chromosome CpGs to compare female SLE patients to females controls . Only in T-cells did we observe moderately significant associations ( FDR<1% ) , although none were genome-wide significant ( p<1×10−8 ) . Table S6 lists the 43 significant X-chromosome CpGs in T-cells . These include TLR7 and FOXP3 , both of which have been previously linked to SLE .
We have performed a comprehensive analysis of DNA methylation changes in SLE in two lymphoid cell-types ( T- and B-cells ) , and one myeloid cell-type ( monocytes ) . Our analysis has identified a strong hypomethylation of loci involved in type-I interferon signaling , which indicates that SLE patients are hypersensitive to interferon . While this is not entirely surprising , given that interferon-related gene expression changes have been documented in active SLE patients , we have also discovered that the hypomethylation is observed in both active and quiescent patients . This is remarkable because circulating interferon and the expression of the genes it induces , are known to increase during flares of the disease , but return to normal during quiescent periods . So , the epigenetic hypersensitivity at the DNA methylation level appears to be independent of interferon levels and is maintained in the immune system beyond active stages of the disease . Exactly when these epigenetic changes occur is not clear . Studies have demonstrated mildly elevated IFN-α in unaffected relatives of SLE patients , suggesting that there is a genetic basis of higher IFN levels [22] . So it is feasible that SLE patients had higher baseline IFN prior to disease onset , and that chronic exposure could have induced long-lasting epigenetic hypersensitivity . In any case , the persistence of the hypomethylation in patients during quiescence is important , as it may help explain the chronic nature of the disease and the potential for recurrent flares in SLE patients . Our data suggest that these patients are poised for elevated interferon responses , but until some event triggers IFN-α production , the responsive genes remain near normal expression levels . We have also observed the hypomethylation of interferon genes in sorted subpopulations of CD4+ T-cells , including memory , naïve and regulatory T-cells . Given that this appears to be a universal effect , and is apparent in lymphoid and myeloid lineages , the most likely explanation is that a multi-potent progenitor population carries this epigenetic state and produces lineages that are programmed to respond to interferon . Future studies of DNA methylation in early progenitor populations from SLE patients will be needed to establish the responsible cells , and to define the events that might induce this epigenetic state in progenitor cells . In addition to the primary interferon effect , we have identified widespread moderate changes in methylation in T-cells that are best explained by SLE-related compositional changes to the CD4+ population , rather than intrinsic methylation changes in any CD4+ subtype . We did not observe an enrichment of these effects in sorted memory , naïve or regulatory T-cells , although we cannot rule out a role for subtypes such as Th1 , Th2 , or Th17 , as we did not sort along these lines . This is not to suggest that methylation effects are absent from CD4+ subtypes , but rather that the widespread , moderate changes we can observe in the CD4+ population cannot be explained solely by intrinsic methylation changes in memory , naïve or regulatory T-cells . Further sub-fractionation of the CD4+ cells will be required to establish which subtypes are responsible for these subtle changes in SLE patients , either because they carry subtype-specific methylation patterns and are changing in number , or because they carry intrinsic methylation differences in SLE patients . Some studies have indicated that regulatory T-cells are reduced in number in SLE patients . While this may be one contributor to the compositional effect , our quantification of memory , naïve and regulatory T-cells is insufficient to explain the entirety of the methylation changes we observe in CD4+ cells . Our functional analysis of the genes affected by these methylation changes , indicate that they are involved in immune cell signaling and cell division . All of these might be interpreted as part of the T-cell activation process , and perhaps the compositional changes occurring in the CD4+ population are due to increases in the number of activated T-cells that cut across traditional definitions of the CD4+ subsets . A complete characterization of the genome-wide DNA methylation profiles in the CD4+ milieu will be required to understand how different epigenetic states correlate with classic cell type definitions . Finally , while our study was not designed to detect methylation patterns that were induced by medications or might be predictive of a patient's response to medications , this is clearly an area of great interest . The fact that we observe similar methylation patterns in quiescent and active SLE patients , who typically increase their medication levels during a flare , suggests that these medications do not induce a large epigenetic effect . Nonetheless , studies that examine the epigenetic impact of anti-inflammatories , as well as the epigenetic states that modulate their efficacy , may have an impact on the clinical management of SLE .
All patient samples were collected with consent at UAB under compliance with the Institutional Review Board . Patients were recruited through the UAB outpatient Rheumatology clinic . Diagnosis was performed according to revised ACR criteria [23]–[25] and disease activity and SLEDAI scores were collected from each patient , along with gender , age and ethnicity information . Disease activity ( flare versus quiescent ) was defined by a recent increase in SLEDAI without using a specific SLEDAI threshold . However , all patients considered to be active had a SLEDAI > = 4 ( mean = 8 . 5 ) , and all of our quiescent patients had a SLEDAI < = 6 ( mean = 1 . 5 ) . CD4+ T-cells , CD19+ B-cells and CD14+ monocytes were isolated from ∼5 ml each of freshly collected peripheral blood . All three cell types were isolated in parallel using positive selection by antigen-specific Dynabeads ( Invitrogen ) , according to the manufacturer's standard protocol . The cells captured on the beads were lysed and DNA was extracted with QIAGEN DNAeasy kits . Purity of separated populations was verified to be above 95% . For experiments with CD4+ subsets , CD4+ cells were isolated using positive selection ( Invitrogen ) followed by sorting of subsets by flow cytometry ( FACSAriaII , BD Biosciences ) . Very pure populations ( 95–100% ) of memory T cells ( CD45RO+RA− ) , naïve T cells ( CD45RA+RO− ) , and T regulatory cells ( CD25+CD127− ) were collected using anti-CD4-Alexa488 , anti-CD45RO-APC , anti-CD45RA-PE , anti-CD25-PerCP-Cy5 . 5 and anti-CD127-Pacific Blue antibodies ( Biolegend , Inc ) . Cells were then lysed and DNA extracted with QIAGEN DNAeasy kits . 500 ng of each DNA sample was treated with sodium bisulfite ( Zymo EZ DNA ) prior to standard Illumina amplification , hybridization , and imaging steps . To limit confounding from batch effects , we distributed SLE cases and controls equally among the 12 slots on each array . The samples were also grouped on the arrays by cell type . The resulting intensity files were analyzed with Illumina's GenomeStudio , which generated beta scores ( proportion of total signal from the methylation-specific probe or color channel ) and “detection p-values” ( probability that the total intensity for a given probe falls within the background signal intensity ) . Beta scores were generated without background subtraction or Illlumina normalization options . Those beta scores with an associated detection p-value greater than 0 . 01 were removed and samples with more than 1 . 5% missing data points across ∼470 , 000 autosomal CpGs were eliminated from further analysis . Furthermore , any CpG probes where more than 10% of samples failed to yield adequate intensity were removed . The filtered beta scores were then subjected to non-parametric batch normalization with the ComBat package for R software ( http://http://www . bu . edu/jlab/wp-assets/ComBat/Abstract . html ) . To parallelize this process on our computational cluster , normalization was performed on non-overlapping subsets of no more than 20 , 000 CpGs per job ( randomly selected ) , and each array of 12 samples was used as a “batch” . We also separately normalized probes from the Infinium I and II chemistries , as their beta score distributions are slightly different . For example , the 131 , 715 autosomal Infinium I CpGs were split into 6 randomly chosen sets of 20 , 000 CpGs each , plus one set of 11 , 715 CpGs , and each set was batch normalized in parallel . Figure S3A shows QQ-plots of explicit tests for batch effects at each CpG , before and after ComBat normalization . These tests were linear regression tests for batch ID , with disease , age , gender , and ethnicity as covariates . In addition , we compared our subsetting approach of 20 , 000 CpGs to similar ComBat runs with larger numbers of CpGs , but the efficacy of batch correction was virtually identical , while greatly reducing the computational time for normalization . Furthermore , as indicated in Figure S3B , our batch normalization process did not introduce any systematic bias into our data , as our disease-specific regression results applied before and after ComBat were highly similar . Data from the X chromosome was normalized separately for males and females due to the gender-specific effect of X-inactivation on the beta score distribution . After batch normalization , we further adjusted the beta scores for probes that utilized the Infinium II chemistry to better match the Infinium I chemistry using the equation β′ = 0 . 001514 + 0 . 3323* β + 0 . 7411* β 2 . This equation was derived from fitting a second order polynomial to the observed pairs of beta scores across all pairs of probes located <50 bp apart , where one probe was Infinium I and one was Infinium II . At this proximity , within-chemistry correlations are extremely high ( R>0 . 99 ) due to locally correlated methylation patterns , and the non-linear relationship between the two chemistries is easily estimated . Figure S4 illustrates the improved scaling of the two chemistries after our corrections have been applied . Our dataset was further reduced by eliminating any CpGs where the probe sequence either mapped to a location in the genome that was different that the location found in Illumina's annotation file , or where the probe could potentially map to more than one locus . The list of these problematic CpGs was generated by re-aligning all probes ( with unconverted Cs ) to the human reference genome with BLAT . We also maintained a list of probes where known SNPs would fall within the probe sequence or at the CpG itself , but did not explicitly filter out these probes . There was no apparent enrichment for CpG probes that overlapped a SNP in dbSNP 135 among our most significant results . To perform genome-wide association testing , we ran linear regression models at each CpG ( lm package in R ) to test for associations between DNA methylation levels and SLE disease state ( case/control comparison ) or flare status ( case/case comparison ) . Since DNA methylation is influenced by age , gender , and ethnicity , we included these as covariates in our models . ( 1 ) For analysis of the X-chromosome CpGs , females were analyzed separately so gender correction was unnecessary . The p-values and beta coefficients for the disease term in our regression models were used to establish the significance of the association at each CpG , and to estimate the post-correction differences in methylation between cases and controls , respectively . FDR correction was performed on the p-values using R ( p . adjust function ) . We also selected 20 , 000 CpGs at random to perform permutation tests that randomized the disease state variable to estimate empirical p-values ( lmp package ) . After 108 permutations , the permutation-based p-value was compared to the regression estimate , and both p-values were highly correlated . Figure S2 displays the genome-wide QQ plot for CD4+ cells , with the permuted p-values for 20 , 000 random CpGs overlayed in green . The biphasic trend in the QQ plot was recapitulated with permutation-based p-values . We performed two types of analyses ( Ingenuity and DAVID ) to identify gene annotation terms that were enriched among our most significant associations . Our annotation of interferon-regulated genes was expanded to include the “IFI” gene symbols , which have been termed “interferon-inducible transcripts” , but have not all been given GO terms that reflect this functionality . | We have analyzed DNA methylation , an epigenetic modification that influences gene expression , in lupus patients and control subjects . Our analysis was run in three different immune cell types , T-cells , B-cells , and monocytes , to discern common epigenetic effects in lupus from cell type-specific effects . We have identified a lupus-related reduction in methylation around genes that respond to interferon , a cytokine that induces inflammation in response to pathogens . This hypomethylation suggests that lupus patients are hypersensitive to interferon , as DNA methylation is typically an inhibitor of gene expression . We also find that this hypersensitivity is preserved in lupus patients beyond active stages of the disease , and this may help explain the chronic , recurrent nature of the disease . In addition , we have identified DNA methylation changes in T-cells that suggest an alteration in the proportions of these cells in lupus patients , which may help explain the disease process . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systemic",
"lupus",
"erythematosus",
"medicine",
"rheumatology",
"genome",
"analysis",
"tools",
"genomics",
"x",
"chromosome",
"inactivation",
"immune",
"cells",
"b",
"cells",
"chromosome",
"biology",
"genome",
"scans",
"monocytes",
"genetics",
"t",
"cells",
"epigenetics",
"immunology",
"biology",
"dna",
"modification",
"autoimmunity"
] | 2013 | Genome-Wide DNA Methylation Analysis of Systemic Lupus Erythematosus Reveals Persistent Hypomethylation of Interferon Genes and Compositional Changes to CD4+ T-cell Populations |
Short interspersed nuclear elements ( SINEs ) are highly abundant , RNA polymerase III-transcribed noncoding retrotransposons that are silenced in somatic cells but activated during certain stresses including viral infection . How these induced SINE RNAs impact the host-pathogen interaction is unknown . Here we reveal that during murine gammaherpesvirus 68 ( MHV68 ) infection , rapidly induced SINE RNAs activate the antiviral NF-κB signaling pathway through both mitochondrial antiviral-signaling protein ( MAVS ) -dependent and independent mechanisms . However , SINE RNA-based signaling is hijacked by the virus to enhance viral gene expression and replication . B2 RNA expression stimulates IKKβ-dependent phosphorylation of the major viral lytic cycle transactivator protein RTA , thereby enhancing its activity and increasing progeny virion production . Collectively , these findings suggest that SINE RNAs participate in the innate pathogen response mechanism , but that herpesviruses have evolved to co-opt retrotransposon activation for viral benefit .
While only ~1 . 5% of mammalian genomes consist of protein coding sequence , upwards of 75% of the genome is transcribed [1 , 2] . A considerable amount of this transcription generates stable non-protein-coding RNAs ( ncRNAs ) of potential biological relevance . Similarly , transcription from the genomes of many large double stranded ( ds ) DNA viruses is pervasive and can generate an abundance of long and short ncRNAs , a number of which have key roles in viral replication and pathogenesis [3–11] . While there is an increasing appreciation that viruses have adopted ncRNAs as part of their gene regulatory repertoire , with the exception of some small ncRNAs such as microRNAs , how most other cellular ncRNAs may impact the gene expression landscape during infection remains unknown . Given that viruses have provided significant insight into mammalian gene regulation , they have the potential to reveal new features of ncRNA biology . One of the largest potential sources of host-derived ncRNAs is a class of retrotransposons called short interspersed nuclear elements ( SINEs ) , as these comprise greater than 10% of the human and mouse genomes [12–15] . SINEs are non-autonomous and require co-expression of protein products encoded within long interspersed nuclear elements ( LINEs ) for retrotransposition [16] . Alu elements are the predominant SINE family in humans , while the B1 and B2 SINEs are the major families in the murine genome . All SINE families are evolutionarily derived from endogenous RNA Polymerase III ( Pol III ) transcripts: Alu and B1 SINEs are derived from 7SL RNA , the RNA component of signal recognition particle , and B2 SINEs are derived from transfer RNA ( tRNA ) [17–20] . SINE elements contain internal Box A and Box B RNA Pol III promoter elements that drive transcription of a SINE ncRNA . In general , SINE elements are transcriptionally silenced in healthy somatic cells , although they can be activated by a variety of chemical and biological stresses [21] . In this regard , several viruses including herpes simplex virus 1 ( HSV-1 ) [22 , 23] , adenovirus type 5 ( Ad5 ) [24] , and Minute virus of mice ( MVM ) have been shown to induce SINE RNA expression upon infection [25] . SINE elements are thus a robust source of inducible ncRNAs , whose expression could impact the gene expression environment during infection . Indeed , there is precedence for SINE RNA functioning in the regulation of gene expression during heat shock , where transcribed Alu and B2 SINE RNAs participate in transcriptional repression through direct interactions with RNA pol II [26–30] . Additional observations suggest that SINE ncRNA expression can also interface with components of the innate immune system , perhaps in a manner linked to their secondary structure . SINE RNAs are highly structured with multiple regions of long double stranded RNA ( dsRNA ) [31 , 32] , and the majority possess 5’-triphosphate moieties [33] . These ncRNAs thus have the potential to be recognized by cellular dsRNA sensors and could therefore serve as inducible immune signaling molecules . Early studies revealed that Alu RNA is efficiently bound by the double stranded RNA activated protein kinase ( PKR ) [34] , and can function in either an inhibitory or activating capacity depending on the ncRNA concentration [35] . Furthermore , aberrant expression of Alu RNAs within retinal pigmented epithelium induces TLR-independent activation of the NLRP3 inflammasome , leading to geographic atrophy , a form of age-related macular degeneration ( AMD ) [36–38] . Thus , it is possible that mammalian cells have incorporated the regulated induction of SINE ncRNAs as a means to help control immune activation and gene expression , although aberrant or sustained SINE transcription is likely detrimental . Here , we explored potential roles for SINE ncRNAs induced during viral infection using the murine gammaherpesvirus MHV68 , which we found induces SINE ncRNA transcription in a rapid and sustained manner . MHV68 is a widely used model system for probing the in vivo biology and replication of gammaherpesviruses , a subfamily of large , nuclear replicating dsDNA viruses that includes the oncogenic human viruses Kaposi’s sarcoma-associated herpesvirus ( KSHV ) and Epstein-Barr virus ( EBV ) . Unexpectedly , the induction of SINE ncRNAs boosts viral replication and gene expression , suggesting that herpesviruses have co-opted SINEs for proviral functions . The stimulatory effect on viral replication is linked to SINE RNA-based activation of the innate immune system , in particular the IKKβ component of the NF-κB signaling pathway . IKKβ is a known activator of the primary viral transcriptional transactivator protein RTA [39] , and SINE expression enhances the IKKβ-dependent phosphorylation of RTA , thereby boosting its activity . We find that SINE RNAs activate NF-κB through both mitochondrial antiviral-signaling ( MAVS ) protein dependent and independent mechanisms . Collectively , our findings reveal that virus-induced retrotransposon expression contributes to activation of innate immune signaling during infection , but that herpesviruses exploit this pathway to bolster viral gene expression .
In unstressed somatic cells , SINE loci are transcriptionally repressed and thus RNA Pol III-transcribed SINE RNAs are either undetectable or only weakly expressed . To determine whether gammaherpesvirus infection activated SINE RNA expression , NIH3T3 cells were infected with MHV68 and B1 and B2 SINE RNA levels were quantified by primer extension . Indeed , both B1 and B2 SINEs were induced specifically upon infection , with B2 RNA levels exceeding those of the highly abundant RNA Pol III-transcribed 7SK small nuclear RNA ( snRNA ) ( Fig 1A ) . This increase in RNA Pol III transcriptional activity during MHV68 infection was specific for SINE loci , as infected cells displayed no alteration in the levels of multiple other RNA Pol III-derived transcripts , including tRNAVal , 5S ribosomal RNA ( rRNA ) , and 7SL RNA as measured by Northern blotting ( Fig 1B ) . Additionally , B1 and B2 SINE RNAs localized to both the cytoplasmic and nuclear compartments ( Fig 1C ) . SINE RNA induction was both rapid and sustained , occurring by 3 h post infection ( hpi ) at levels similar to heat shock-driven induction and continually increasing throughout the 24 hpi time course ( Fig 1D ) . During the same time course 7SK levels remained constant in both infected and uninfected cells ( Fig 1D ) . To determine whether the increase in SINE RNAs occurred at the level of transcription , nuclear run-on assays were performed using nuclei from uninfected or MHV68-infected cells . A robust transcriptional increase was observed for both B1 and B2 in infected cells , whereas no transcriptional difference was observed for the RNA Pol III transcribed 7SL RNA ( Fig 1E ) . No signal was detected for the negative control ncRNA EBER1 from Epstein-Barr virus , which is not expressed in these cells , confirming the specificity of the run-on signals ( Fig 1E ) . Finally , the increase in SINE RNA was also observed upon infection of primary mouse embryonic fibroblasts ( MEFs ) and in vivo in the lung tissue of MHV68-infected C57BL/6 mice at 5 days post infection , indicating the induction was not restricted to established cell lines ( Fig 1F and 1G ) . Together , these data demonstrate that MHV68 infection results in a rapid , sustained , and specific transcriptional activation of B1 and B2 SINE loci . We next sought to identify aspect ( s ) of the MHV68 life cycle responsible for SINE RNA induction . After entry , the viral genome is delivered to the nucleus and transcription of immediate early and early genes occurs , which in turn enable viral DNA replication and subsequent expression of late viral genes necessary for progeny virion assembly . UV crosslinking of viral particles does not block viral attachment to cells , but inactivates the viral genome thereby inhibiting viral gene expression and replication ( Fig 2A and 2B ) . NIH3T3 cells incubated with UV-inactivated virus still displayed the early SINE RNA induction at 3 hpi , but unlike incubation with infectious virus , B1 and B2 RNA levels rapidly subsided over the 24 hpi time course ( Fig 2C ) . ( It should be noted that the B1 gels in Fig 2 were exposed longer than the B2 gels to better visualize the faint B1 signal ) . This suggests that viral activation of SINEs is biphasic , with the initial induction occurring in response to viral attachment and/or entry , and sustained SINE transcription reliant on downstream aspects of the viral life cycle . To determine whether sustained SINE activation required viral genome replication or late gene expression , cells were treated with the viral DNA polymerase inhibitor phosphonoacetic acid ( PAA ) prior to MHV68 infection ( Fig 2D ) . SINE RNA induction was similar at 24 hpi in control- and PAA-treated cells ( Fig 2E ) . These data suggest that viral early gene expression , but not DNA replication or late gene expression drives sustained transcriptional activation of SINE loci . Although it is possible that a specific viral gene product might be responsible for SINE activation , we have been unable to identify any MHV68 genes whose individual expression in NIH3T3 cells induced SINE transcription . Thus , SINE RNA induction is likely to occur as a consequence of the combined activity of multiple viral genes and/or as a more general cellular response to infection . We hypothesized that the virus-induced noncoding SINE RNAs may play regulatory roles related to infection . To test whether SINE transcription impacts the MHV68 lifecycle , we measured viral replication upon specific knockdown of B1 or B2 SINE RNAs using 2′-O-methylated and phosphorothioate-substituted antisense oligonucleotides ( ASO ) , which direct RNase H-based cleavage of target RNAs . Transfection of B1 and/or B2 ASOs 3 h prior to MHV68 infection of NIH3T3 cells significantly reduced the levels of SINE RNA at 24 hpi ( Fig 3A ) . Control ASOs did not impact B1 or B2 levels , and ASO treatment of uninfected cells did not induce SINE expression ( Fig 3A ) . Remarkably , B2 depletion resulted in delayed viral replication kinetics in a multistep growth curve and a ~15-fold decrease in progeny virion production ( Fig 3B ) . We observed a similar ( ~10-fold ) decrease in viral replication in a single-step growth curve upon treatment of NIH3T3 cells with the specific RNA Pol III inhibitor ML-60218 6 h prior to infection to block SINE transcription ( S1 Fig ) . Although ML-60218 blocks transcription of all RNA Pol III genes , constitutive Pol III transcripts in general have long half-lives and thus , unlike B1 and B2 RNAs , their steady state levels are not appreciably affected by the treatment . Specific ASO-mediated depletion of B1 RNA , which accumulates to far lower levels than B2 RNA during MHV68 infection , had no effect on viral replication , and co-depletion of both B1 and B2 did not enhance the replication defect observed upon B2 knockdown ( Fig 3B ) . The above results indicated that induction of B2 SINE RNA by MHV68 is beneficial to the viral lifecycle . We therefore examined whether the SINE RNAs impacted viral gene expression by measuring the accumulation of viral mRNAs from different kinetic classes in cells treated with control or SINE-specific ASOs . Indeed , RT-qPCR analyses revealed that depletion of B2 RNA resulted in a significant reduction in the levels of all viral RNAs tested ( Fig 3C ) . In agreement with the viral replication data , depletion of B1 RNA had no effect on viral gene expression . Interestingly , the effect of B2 RNA was primarily directed at viral genes , as we observed no decrease in the abundance of a set of 5 cellular mRNAs upon treatment with B1 or B2 ASOs ( Fig 3C ) . However , we did note an increase in the levels of the GAPDH mRNA upon B2 depletion , in agreement with data showing that heat shock-induced B2 RNA can transcriptionally repress select promoters [26 , 27] . We obtained similar results when SINE RNA transcription was inhibited by pre-treatment with ML-60218 ( S1 Fig ) . Thus , virus-induced expression of B2 SINE RNA broadly enhances MHV68 mRNA abundance , and this stimulatory effect primarily impacts viral gene expression even though RNA Pol II transcribes both viral and cellular mRNAs . Several long noncoding RNAs manipulate gene expression through their recruitment to specific cellular gene promoters [40] . In this regard , B2 SINE RNA has been shown to interact with RNA Pol II at select promoters to inhibit transcription [26 , 27] . Although this explains the apparent suppression of GAPDH mRNA by B2 , the stimulatory effect of B2 on MHV68 mRNA suggested that SINE RNAs might not function through their recruitment to promoters . Using chromatin isolation by RNA purification ( ChIRP ) we examined whether B2 RNA was present at viral or cellular promoters . ChIRP-qPCR analyses detected B2 RNA at the promoter of GAPDH and RPL13a ( Fig 3D and 3E ) . In contrast to our findings with cellular promoters , we were unable to ChIRP B2 RNA to any viral promoters , suggesting that B2 SINE RNA modulates viral gene expression indirectly . Our above data suggested that SINE RNAs have promoter-specific effects on cellular genes . We tested this hypothesis by analyzing the effect of B1 and B2 SINE RNA expression on a panel of promoter elements cloned upstream of a luciferase reporter ( Fig 4A ) . To monitor effects specifically linked to SINE RNAs as opposed to secondary effects stemming from infection , we co-expressed each reporter with consensus B1 or B2 SINEs derived from RNA Pol III-driven SINE expression constructs in NIH3T3 cells . SINE RNAs modestly suppressed the AP1 promoter , had no impact on the p53 and ISRE promoters , but significantly activated both the NF-κB-driven and Sp1-driven luciferase reporters ( Fig 4B ) . In each case , B2 RNA had a more potent effect than B1 , and all changes in luciferase reporter levels required SINE RNA expression , as they were blocked upon treatment with the RNA Pol III inhibitor ML-60218 ( Fig 4B ) . MHV68 infection has been shown to activate the NF-κB pathway , components of which are used by the virus for robust induction of lytic gene expression [39 , 41–43] . Though MHV68 blunts the NF-κB transcriptional response by subsequently inducing degradation of the RelA/p65 subunit of NF-κB during the first 4 hpi [41] , the IKKβ kinase component of the pathway is co-opted to promote phosphorylation of the major viral lytic transactivator RTA . Our findings suggested that induction of SINE RNAs might contribute to NF-κB activation during infection , thereby potentiating viral replication . To test this hypothesis , we first examined whether SINE RNA expression was sufficient to activate endogenous components of the NF-κB signaling pathway . Indeed , transfection of plasmid-based SINEs in uninfected cells resulted in increased phosphorylation of the NF-κB p65 subunit , a marker of NF-κB activation ( Fig 4C ) . Furthermore , inhibition of IKKβ by treatment with its specific inhibitor BAY 11–7082 significantly reduced the ability of SINE RNA to activate the NF-κB luciferase reporter , but did not impact SINE-driven activation of the Sp1 reporter ( Fig 4D ) . Thus , SINE RNAs likely activate this pathway in the cytoplasm , upstream of IKKβ . IKKβ activation during MHV68 infection is at least partially mediated through activation of the MAVS adaptor protein [39] . However , the mechanism underlying MAVS activation during MHV68 infection remains unclear . MHV68 is a dsDNA virus , yet MAVS activation generally occurs upon recognition by upstream RIG-I-like receptors of nucleic acid features associated with RNA viruses , including double-stranded RNA with either a 5′ diphosphate or 5′ triphosphate [44] . We therefore explored whether the structured SINE RNAs might serve as the MAVS activation signal to stimulate IKKβ during MHV68 infection . Indeed , the ability of B1 or B2 SINE RNA expression to activate the NF-κB promoter was significantly , though not completely , impaired in MAVS-/- fibroblasts relative to WT fibroblasts ( Fig 5A ) . This effect was specific for the NF-κB promoter , as no decrease in B1- or B2-induced Sp1 promoter activation occurred in the cells lacking MAVS ( Fig 5A ) . We next examined the role of MAVS in endogenous , MHV68-induced SINE RNA-mediated phosphorylation of RelA/p65 . Within 1 hpi , RelA/p65 is phosphorylated at Serine 468 in a MAVS-IKKβ-dependent manner , a mark that primes its degradation via the proteasome , thereby blunting the NF-κB response [41] . The ability of MHV68 to induce phosphorylation of RelA/p65 was significantly reduced in both MAVS-/- fibroblasts and in WT fibroblasts that had been depleted of B2 RNA using a B2-specific ASO ( Fig 5B ) . As anticipated , no effect was observed upon depletion of the B1 RNA , in agreement with the fact that it is only weakly induced relative to B2 RNA by MHV68 ( Fig 5B ) . Additionally , MHV68-induced RelA/p65 phosphorylation was further reduced to background levels when B2 RNA was depleted in the MAVS-/- cells , suggesting that at least a portion of SINE RNA-based IKKβ activation occurs in a MAVS-independent manner ( Fig 5C ) . These results indicate that B2 RNA contributes to the blunting of the NF-κB response during MHV68 infection . We next compared the relative importance of MAVS and B2 SINE RNA in productive MHV68 infection . MHV68 replication was impaired to a similar extent in fibroblasts lacking MAVS as in cells depleted of B2 SINE RNA , as measured in multi-step growth curves ( Fig 5D ) . However , depletion of B2 RNA in a MAVS-/- background caused a further reduction in viral replication , to levels ~30-fold lower than in untreated WT cells . Collectively , these data indicate that the B2 RNAs are major contributors to the IKKβ activation observed during MHV68 infection , and that they stimulate this pathway via both MAVS-dependent and -independent mechanisms to boost viral replication . The broad enhancement of MHV68 gene expression by B2 RNA , coupled with the absence of B2 at viral promoters , suggested that SINE RNAs act at an early stage of the viral lifecycle to boost the subsequent gene expression cascade . Notably , the major viral lytic transactivator protein RTA was recently shown to be phosphorylated by IKKβ in a manner that increased RTA activity [39] . RTA is a viral protein expressed with immediate early kinetics , and its activity broadly impacts the viral gene expression cascade [45–47] . We therefore hypothesized that the pro-viral effects of the SINE RNAs might be mediated through an increase in RTA activity caused by B2-induced NF-κB pathway activation . We tested this idea first by examining whether the activation of IKKβ by B1 and B2 RNA could potentiate RTA phosphorylation , as measured by in vivo P32 orthophosphate labeling . FLAG-RTA was expressed in NIH3T3 cells in the presence or absence of B1 or B2 expression plasmids in 32P orthophosphate containing growth media , and immunoprecipitated with anti-FLAG coupled beads to quantify its phosphorylation status . Expression of B1 and B2 promoted a significant increase in RTA phosphorylation relative to the control plasmid ( Fig 6A ) . Furthermore , 32P orthophosphate labeling in MHV68-infected cells revealed that depletion of B2 RNA during infection reduced FLAG-RTA phosphorylation ( Fig 6B ) . Next , to determine whether this SINE-mediated phosphorylation change resulted in increased RTA activity , we examined the ability of RTA to transactivate several established RTA-responsive promoters in the presence or absence of B1 and B2 SINEs . Cells were co-transfected with RTA and luciferase reporter plasmids containing the RTA-responsive viral promoters from either RTA , ORF57 , or M3 in the presence or absence of B1 and B2 SINE constructs ( Fig 6C ) . Indeed , RTA mediated transcriptional activation of all three promoters was significantly enhanced upon co-expression of either B1 or B2 relative to the control plasmid . This ability of the SINE RNAs to enhance RTA activity was dependent on IKKβ phosphorylation of RTA , as an RTA mutant ( RTA-TTS/A ) [39] in which the IKKβ phosphorylation sites were mutated retained WT basal transcriptional activity but was unresponsive to the SINE RNAs ( Fig 6C ) . Collectively , these data indicate that SINE RNA-mediated activation of IKKβ drives RTA phosphorylation , thereby increasing its activity on viral promoters and enhancing MHV68 gene expression .
Here , we demonstrate that MHV68-induced activation of SINE RNA serves to regulate viral and cellular gene expression through distinct mechanisms ( Fig 7 ) . SINE RNAs reside in the nucleus and in the cytoplasm , and participate in gene regulation in both compartments . Nuclear SINE RNAs appear absent from viral promoters but associate with specific cellular promoters that are repressed upon SINE activation , while SINE RNAs in the cytoplasm drive activation of the antiviral NF-κB signaling pathway . Though this pathway is normally detrimental to viral replication , MHV68 co-opts the IKKβ component of the NF-κB cascade to boost the activity of the viral lytic transactivator RTA , thereby enhancing viral gene expression and replication [39] . Thus , beyond direct regulation of cellular gene expression , induction of this class of ncRNAs may be linked to early immune-based sensing of infection , a process that has been hijacked by gammaherpesviruses to enhance viral gene expression . Activation of SINE expression following MHV68 infection is a biphasic response with an initial phase arising as a result of either viral attachment or entry in to cells , and a second response that requires progression of the infection past entry , and includes immediate early and early viral gene expression . While it is possible that the initial burst of SINE expression observed following infection with U . V . inactivated virus is mediated by protein components of the tegument , we disfavor this scenario as we have been unable to identify an individual MHV68 gene whose expression is sufficient to induce SINE expression . This suggests that the mechanism by which MHV68 induces SINE expression is distinct from that of HSV-1 and Ad5 , in which specific viral proteins have been implicated in promoting SINE induction . For instance , in the context of Ad5 infection , loss of E1a , E1b , or E4 ORFs 3 and 6 results in a significant decrease in SINE expression in infected cells [24] . Additionally , HSV-1 ICP27 has been demonstrated to enhance the activity of the RNA Pol III general transcription factor TFIIIC leading to increased SINE RNA expression [48] . In the case of MHV68 , we believe a more likely model is one that focuses on the cellular stresses imposed on the cell during infection , including disruption of plasma membrane homeostasis during viral attachment and entry , and triggering of the innate immune response . This suggests that at least the initial burst of SINE transcription should occur regardless of whether infection progresses to latency or lytic replication . However , whether a latent infection results in sustained SINE activation similar to a lytic infection is currently unknown . As expression of SINE RNA potentiates a normally antiviral NF-κB response , the coupling of viral entry receptors and/or plasma membrane homeostasis to the production of immunogenic RNAs would constitute a means to rapidly prime the innate immune response for a potential pathogen exposure . SINE induction may also occur as an early component of innate immune sensing . Indeed , many pathogen recognition receptors ( PRRs ) such as the toll like receptors ( TLRs ) , which are present at both the cell surface and endosome , are engaged by herpesviruses [49] . In this regard , cell surface associated TLR2 has been implicated in detecting MHV68 , likely recognizing viral glycoproteins [50] . Alternatively or perhaps concomitantly , the incoming viral DNA may be sensed by nucleic acid sensing PRRs , for which both endosomal TLR9 and the cytoplasmic cGAS-STING sensing pathways have been implicated [51–54] . As both stimulation of PRRs and exogenous overexpression of SINEs can mediate NF-κB , the induction of SINEs in response to PRR stimulation could establish a feed forward or signal amplification mechanism . In support of a model in which SINE RNAs are generated in response to PRR stimulation are the recent findings that lipopolysaccharide ( LPS ) stimulation of TLR4 induces a general activation of RNA Pol III transcription [55] . Though we do not detect a general increase in all RNA Pol III synthesized RNAs , its possible that MHV68 has evolved ways to manipulate the RNA Pol III transcriptional program to selectively drive synthesis of SINE RNAs , which enhance the MHV68 lifecycle . Interestingly , stimulation of TLRs 3 , 4 , 5 , and 9 promote MHV68 reactivation from latently infected B cells [56] . SINE RNAs are present in both the cytoplasm and nucleus , and our and others’ data argue that they regulate gene expression via multiple mechanisms , some of which are linked to their location within the cell . For MHV68 it is the cytoplasmic fraction that is likely important for enhancing viral gene expression through the activation of the NF-κB pathway . Multiple studies have established that the NF-κB pathway is crucial for gammaherpesvirus latent infection [57] , although reports have been more varied as to its roles in the viral lytic cycle . For instance , suppression of NF-κB signaling via expression of the IκBα super suppressor did not impair viral replication [43] , whereas overexpression of the RelA/p65 subunit of the transcriptionally active NF-κB dimer inhibited the MHV68 lytic cycle [58] . However , these apparently inconsistent findings have recently been clarified by Feng and colleagues , who revealed that MHV68 activates the NF-κB pathway immediately following infection but that the downstream NF-κB transcriptional response is blunted as RelA/p65 is robustly targeted for degradation early in infection [39 , 41] . Furthermore , optimal MHV68 gene expression and replication require the role of the IKKβ component of the NF-κB pathway [39 , 41] , which operates upstream of the IκBα super suppressor . IKKβ promotes phosphorylation of RelA/p65 , thereby priming it for proteasomal degradation , as well as phosphorylation of MHV68 RTA , which enhances its transcriptional activating properties . Interestingly , MHV68 activates IKKβ in both MAVS-dependent and independent ways . While some of the MAVS-dependent activation likely comes from MHV68-induced RIG-I deamidation [59] , we establish that B2 SINEs are necessary for robust IKKβ activation , and that this also occurs through both MAVS-dependent and independent mechanisms . Though the RNA ‘sensor’ for SINE RNA is not known , the antiviral dsRNA binding protein PKR has been shown to bind to SINE RNA [34] . Furthermore , PKR is capable of directly associating with MAVS [60 , 61] . Additionally , many other RNA sensors including RIG-I preferentially recognize many features present in SINE RNA , including 5’-triphosphate moieties and double stranded regions . However , we have not detected B2 RNAs in RNA immunoprecipitations of RIG-I , suggesting RIG-I may not be responsible for ‘sensing’ B2 RNAs . Future studies aimed at characterizing the composition of viral-induced SINE ribonucleoprotein complexes may provide insight here . Many viruses , including the related gammaherpesvirus EBV , up regulate host RNA Pol III transcription [62] . In the case of EBV , the up regulation of RNA Pol III transcription is more general , although whether this includes SINE elements is unknown . Interestingly , it was recently shown that EBV infection stimulates RNA Pol III dependent expression of vault RNAs , which can also activate NF-κB and enhance viral establishment [63] . These results suggest that multiple RNA Pol III ncRNAs engage components of the cytoplasmic innate immune system , and their transcriptional induction in response to viral infection could be one of the early stress responses of the cell . Additionally , the production of abundant dsRNAs following reactivation of latent KSHV in iSLK cells was recently reported [64] . Whether the dsRNAs detected are human SINE RNAs is unknown but worthy of investigation . SINE RNA expression also manipulates host gene expression . For example , during the heat shock response , global RNA Pol II transcription is down regulated . This is mediated in part through the direct binding of B2 RNAs to RNA Pol II , which prevent it from establishing contacts with the promoter during closed complex formation [26–28] . Our knockdown data coupled with our ChIRP-qPCR analyses reveal that B2 RNA-mediated transcriptional repression can similarly operate during viral infection , perhaps also through an interaction with RNA Pol II . At the moment we are unaware of the extent to which this occurs globally during infection . However , data from our laboratory suggest that it is unlikely that B2 RNAs globally repress transcription during MHV68 infection , as both WT virus and a viral mutant unable to restrict host gene expression induce B2 RNAs to a similar extent , yet widespread transcriptional repression is only observed during a WT infection [65] . In this regard , future studies will be important to determine the basis for promoter specificity . SINE RNAs are also abundantly expressed early during embryogenesis within stem cells , a cell type in which dsRNA sensing is inefficient . The function of these ncRNAs within this context is unclear , though a fraction of SINE RNAs are processed into endosiRNAs [66 , 67] . Interestingly , it has recently been reported that that efficient induction of induced pluripotent stem cells ( iPSCs ) requires activation of the dsRNA sensor TLR3 [68] . Whether continued low-level activation of the innate immune system is required for iPSC maintenance is unclear , but it is possible that SINE RNA expression provides chronic low-level innate immune stimulation to promote stem cell maintenance . The use of viruses to perturb host systems , such as described here , presents a valuable platform to probe SINE ncRNA biology . SINE ncRNAs are a critical component of the gammaherpesvirus lifecycle , but they are also activated upon infection with multiple other human and murine viruses . Whether in other systems SINE RNAs serve as anti-viral signaling components , as well as if and how they are co-opted by the diversity of viral and non-viral pathogens remain exciting avenues for future research .
NIH3T3 ( ATCC ) , NIH3T12 ( ATCC ) , Vero , MEF , and WT and MAVS-/- fibroblasts ( kindly provided by Russell Vance , University of California Berkeley ) were maintained in Dulbecco's modified Eagle medium ( DMEM; Invitrogen ) supplemented with 10% fetal bovine serum ( FBS; Invitrogen ) . The green fluorescent protein ( GFP ) -expressing MHV68 bacterial artificial chromosome ( BAC ) has been described elsewhere [69] . BAC-derived MHV68 virus was produced by transfecting BAC DNA into NIH3T3 cells using SuperFect ( Qiagen ) . Virus was then amplified in NIH 3T12 cells and titered by plaque assay on NIH3T3 cells . Before infecting mice , the loxP-flanked BAC vector sequence was removed by passaging the virus through Vero cells expressing Cre recombinase ( kindly provided by Dr . Samuel Speck , Emory University ) . For analysis of gene expression by RT-qPCR , total or subcellular fractions of RNA were isolated with TRIzol ( Invitrogen ) in accordance with the manufacturer's instructions . cDNA was synthesized from 1 μg of RNA with random hexamers ( Integrated DNA Technologies ) and SuperScript II reverse transcriptase ( Invitrogen ) . qPCR was performed using the DyNAmo ColorFlash SYBR green qPCR kit ( Thermo Scientific ) with appropriate primers . For small RNA northern blot analysis total RNA was separated on 8% polyacrylamide–7M urea gels and electrotransferred at 4°C to Amersham Hybond-N+ membranes in 0 . 5X TBE buffer for 16h at 15V . Membranes were probed overnight using 32P-end labeled probes overnight at 55°C . Blots were washed three times in 0 . 1X SSC for 10 min each before exposed to phosphoimager screens overnight . For mRNA northern blot analysis total RNA was resolved on 1 . 2% agarose-formaldehyde gels and transferred to Hybond-N+ membranes by capillary action . For primer extension , 15 μg of total RNA was incubated with 6 pmol 32P-labeled primer in 10 μl of Buffer A ( 250 mM KCl , 10 mM Tris , pH 7 . 5 , and 1 mM EDTA ) for 1 h at 55°C . Buffer B ( 40 μl ) and 0 . 5 μl AMV reverse transcriptase ( Promega ) were added and incubated at 42°C for 1 h . Products were phenol-chloroform extracted , ethanol-precipitated , and resuspended in RNA loading dye before being resolved on 8% polyacrylamide–7M urea gels . Nuclear run-on was performed as described previously with minor modifications [70] . Specifically , nuclei were isolated from two 10-cm plates of confluent mock- or MHV68 infected cells . For Luciferase assays NIH3T3 cells were transfected in 6 well dishes with the indicated plasmids using TransIT-3T3 ( Mirus Bio ) . 48 h post-transfection , lysates were prepared from approximately equal number of cells and luciferase activity was determined with the Promega luciferase assay system . ChIRP was performed as previously described with minor modifications [71] . Briefly , ~ 100 million NIH3T3 cells were infected with MHV68 at an MOI 5 . 24 hpi cells were cross-linked with 1 . 1% formaldehyde for 15 min at room temperature . Crosslinking was then quenched with 0 . 125 M glycine for 5 min . Cells were rinsed again with PBS , scraped into Falcon tubes , and pelleted at 1000 g for 5 min . The cell pellet was resuspended in 3 mL nuclei lysis buffer ( 50 mM Tris-HCl [pH 7 . 0] , 10 mM EDTA , 1% SDS , protease cocktail inhibitor [Roche] , and RNAse inhibitor [Fermentas] ) and rotated for 10 min . at 4°C . Cells were dounced 10 times with a B type pestle and separated in to three 1 mL aliquots for sonication . Sonication was performed using a Covaris focused sonicator . After sonication chromatin aliquots were combined and 9 mL of hybridization buffer ( 750 mM NaCl , 1% SDS , 50 mM Tris 7 . 0 , 1 mM EDTA , 15% Formamide , protease inhibitor cocktail , and RNAse inhibitor ) was added . 50 pmol of five separate 3’-TEG biotinylated probes was added to the dilute chromatin and rotated over night for 16 h . Streptavidin-magnetic C1 beads ( Life Technologies ) were washed three times in nuclei lysis buffer , blocked with 500 ng/μl yeast total RNA , and 1mg/ml BSA for 1 hr at room temperature , and washed three times again in nuclear lysis buffer before being resuspended in its original volume . One hundred microliters washed/blocked C1 beads were added to the chromatin mixture and rotated for an additional 4 h at 37°C . Beads:biotin-probes:RNA:chromatin adducts were captured by magnets ( Invitrogen ) and washed five times with 10 mL wash buffer ( 2× SSC , 0 . 5% SDS ) . After the last wash complexes were eluted by resuspending the beads in 500 μL G50 buffer ( 20 mM Tris-HCl , 300 mM NaCl , 2 mM EDTA , 0 . 2% SDS ) plus 50 μg/mL Proteinase K ( Fermentas ) and incubating at 60°C for 1 h . The G50 buffer was separated from the beads , phenol chloroform extracted , and ethanol precipitated . RNA was analyzed by small RNA northern blotting and DNA was analyzed by qPCR , as described above . For in vivo kinase assays in the absence of infection , NIH3T3 cells were cotransfected with FLAG-RTA and either a B1 SINE , B2 SINE , or control expression plasmid using TransIT-3T3 ( Mirus Bio ) . 24 h post-transfection media was replaced with phosphate‐free DMEM ( Life Technologies ) for 1 h and then incubated in the same medium containing [32P]-orthophosphate ( 0 . 5 mCi/ml final concentration , carrier free; PerkinElmer ) for 6 h . After labeling cells were lysed in high salt RIPA buffer ( 20 mM Tris-HCl [pH 7 . 5] , 500 mM NaCl , 1 mM EDTA , 1 mM EGTA , 1% NP-40 , 1% sodium deoxycholate ) containing a phosphatase inhibitor cocktail ( Sigma ) and immunoprecipitated using anti-FLAG ( M2 ) magnetic beads ( Sigma ) overnight at 4°C . Beads were washed extensively with high salt RIPA buffer and then eluted with FLAG peptide ( Sigma ) . The eluate was resolved by SDS–PAGE , transferred to PVDF membrane ( Immobilon; Millipore ) and visualized by autoradiography . Additionally western blot was performed as described below using anti-FLAG ( M2 , Sigma ) . For in vivo labeling during MHV68 infection , NIH3T3 cells were transfected with FLAG-RTA as above . 24 h post-transfection cells were transfected with ASO’s using RNAiMAX as previously described [72] . 3 h post-ASO treatment the media was replaced with phosphate‐free DMEM ( Life Technologies ) for 1 h and then incubated in the same medium containing [32P]-orthophosphate ( 0 . 5 mCi/ml final concentration , carrier free; PerkinElmer ) and infected with MHV68 at an MOI 5 for 6 h . Immunoprecipitations and downstream analysis was performed as described above . Subcellular fractionation was performed using the REAP method with the minor modification of using one 10-cm plate for each fractionation condition [73] . For western blot analysis , cell lysates were prepared in NET-2 buffer ( 50 mM Tris-HCl [pH 7 . 6] , 150 mM NaCl , 3 mM MgCl2 , 10% glycerol , 0 . 5% Nonidet P-40 ) , and protein concentrations were determined by Bradford assay . Equivalent quantities of each sample were fractionated by SDS-PAGE , transferred to a polyvinylidene difluoride membrane , and incubated with the appropriate antibodies . Western blot assays were developed with HRP-conjugated secondary antibodies and ECL reagents ( Pierce ) . For multi-step growth curves , 1 . 5×105 NIH3T3 , WT or MAVS-/- fibroblasts were infected with MHV68 at MOI of 0 . 05 and both supernatant and cells were harvested at 0 , 1 , 2 , and 3 dpi and frozen . Samples were freeze-thawed once before titering by plaque assay on NIH3T3 cells . Female C57BL/6J mice were obtained from The Jackson Laboratory ( Bar Harbor , ME ) and infected when 4–6 weeks old . Mice were anesthetized with isoflourane and inoculated intranasally with 5×104 plaque forming units ( pfu ) in 20 μl DMEM ( Invitrogen ) . Lungs were harvested 5 dpi and homogenized with a tissue homogenizer in 5 mL Trizol and RNA was isolated as described above . This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Committee on the Ethics of Animal Experiments of the University of California Berkeley ( Permit Number: R292-0507 ) . All animals were anesthetized prior to infection with isoflurane , and all efforts were made to minimize suffering . | Short interspersed nuclear elements ( SINEs ) are noncoding mobile genetic elements that are present at ~106 copies per mammalian genome , roughly comprising 10% of mammalian genomic real estate . SINEs are typically transcriptionally silenced , though in some cases viral infection can promote their expression , yet to an unknown functional outcome . Thus , SINE elements represent the largest class of infection-inducible noncoding RNAs that are functionally uncharacterized . Here , we reveal that SINE RNAs play a critical role in the host-pathogen interaction in that they are required for efficient murine gammaherpesvirus 68 ( MHV68 ) replication and gene expression . We demonstrate that SINE RNAs , both exogenously expressed and infection-induced , are robust activators of the IKKβ kinase , a key signaling molecule in the innate immune response . Activation of the IKKβ kinase by SINE RNA is mediated through both MAVS-dependent and independent mechanisms . Moreover , we demonstrate the activation of the IKKβ via SINE RNA is required to drive the phosphorylation of MHV68 RTA , the main viral transcriptional activator , which enhances its transcriptional activating property . Collectively , we reveal the first example of a role for SINE RNAs in the host-pathogen interaction and identify them as a key immune signaling molecule early during infection . Though SINE RNAs activate the innate immune response , MHV68 has co-opted SINE-mediate innate immune activation to enhance the viral lifecycle . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Infection-Induced Retrotransposon-Derived Noncoding RNAs Enhance Herpesviral Gene Expression via the NF-κB Pathway |
The var genes of the human malaria parasite Plasmodium falciparum present a challenge to population geneticists due to their extreme diversity , which is generated by high rates of recombination . These genes encode a primary antigen protein called PfEMP1 , which is expressed on the surface of infected red blood cells and elicits protective immune responses . Var gene sequences are characterized by pronounced mosaicism , precluding the use of traditional phylogenetic tools that require bifurcating tree-like evolutionary relationships . We present a new method that identifies highly variable regions ( HVRs ) , and then maps each HVR to a complex network in which each sequence is a node and two nodes are linked if they share an exact match of significant length . Here , networks of var genes that recombine freely are expected to have a uniformly random structure , but constraints on recombination will produce network communities that we identify using a stochastic block model . We validate this method on synthetic data , showing that it correctly recovers populations of constrained recombination , before applying it to the Duffy Binding Like-α ( DBLα ) domain of var genes . We find nine HVRs whose network communities map in distinctive ways to known DBLα classifications and clinical phenotypes . We show that the recombinational constraints of some HVRs are correlated , while others are independent . These findings suggest that this micromodular structuring facilitates independent evolutionary trajectories of neighboring mosaic regions , allowing the parasite to retain protein function while generating enormous sequence diversity . Our approach therefore offers a rigorous method for analyzing evolutionary constraints in var genes , and is also flexible enough to be easily applied more generally to any highly recombinant sequences .
The human malaria parasite Plasmodium falciparum causes approximately 1 million deaths each year , primarily in young children in sub-Saharan Africa [1] . In endemic regions , individuals develop clinical immunity to severe disease in childhood , but continue to suffer malaria infections and mild illness throughout their lifetimes . This epidemiological pattern is poorly understood , but appears to be caused by the gradual acquisition of a large repertoire of antibodies following sequential exposure to different parasite proteins [2]–[5] . The main candidate for eliciting protective antibodies is the parasite-derived antigen PfEMP1 ( P . falciparum erythrocyte membrane protein 1 ) , encoded in each parasite genome by a large var gene family and expressed during infection on the surface of infected red blood cells in a process of antigenic variation [2] , [6]–[9] . Extremely rapid recombination among var genes generates enormous diversity and complex mosaic structures among these sequences [10]–[14] , and recent field studies have uncovered seemingly limitless var gene diversity in Africa [15] . Superinfection with multiple clones is extremely common , and recombination can occur during meiosis in the mosquito , as well as between var genes on different chromosomes of a single parasite during asexual reproduction [14] . However , these observations are at odds with the rapid acquisition of antibodies to common PfEMP1 variants that are associated with disease [16] , as well as the finding that parasites from different continents share identical sequence blocks despite millions of years of evolutionary separation [17] . The highly recombinant structure of var genes precludes the use of standard phylogenetic tools , and the processes generating this paradoxical relationship between parasite genetic structure and the epidemiology of infection and disease remain unclear [13] , [18] , [19] . Statistically rigorous and scalable techniques to analyze evolutionary relationships between sequences generated through frequent recombination are lacking . Classical phylogenetic analyses are designed to accommodate branching tree-like relationships between genes generated by mutation , and therefore require that highly recombinant regions , where evolutionarily distant sequences may share mosaics , are removed , ignored , or assumed to be absent [20]–[22] . Bockhorst et al . introduced an approach to understanding the most conserved group of var genes based on a segmentation analysis which divides a set of sequences into segments such that polymorphic sites in the same segment are strongly correlated , while nearby polymorphic sites are either weakly or not correlated [18] , [23] . While segmentation analysis is useful to detect mutation-driven diversification following ancient recombination or geographic separation , particularly for subsets of more conserved var genes , it ultimately generates a tree-like relationship between genes and does not accommodate recent and ongoing recombination . Networks provide a mathematical approach to representing and studying complex relationships between genes [24] , and network-based techniques have produced valuable evolutionary insights for many organisms ranging from viruses to eukaryotes [25] . Attempts to introduce recombination within phylogenetic frameworks have led to specialized techniques that produce phylogenetic ( or recombination ) networks for small numbers of sequences when recombination rates are relatively low , but these also focus on conserved regions rather than providing insights into the recombinant regions [26] ( for a review see [27] ) . On the other hand , ancestral recombination graphs have a strong theoretical foundation but lack efficient approximations that are required for rapid inference [28] . Networks have also been used to identify large-scale clusters of global gene sharing and exchange [29] and horizontal gene transfer of the plasmid resistome [30] , as well as differentiating horizontal and vertical flow of information in [31] , [32] . In these approaches , all-to-all BLAST scores are calculated and thresholded for a set of sequences , and the resulting network is generally analyzed visually to assess large-scale structure [22] , [25] , [29]–[32] . These analyses , however , rely on ad hoc parameter choices , uncontrolled assumptions , or prior knowledge of target clustering . And , while potentially useful for hypothesis generation , a reliance on alignment scores contains an implicit model for sequence mutation and substitution that is not justified for highly recombinant var gene sequences . We have previously taken a network approach to analyze clinical var gene domains using short position-specific sequences [13] . Although this approach uncovered distinctive structuring , with clustering that reflected previous var gene classification schemes , it lacked a solid theoretical basis and more importantly it was not generalizable to other domains and genes . Here we take a more sophisticated approach , applying rigorous community detection methods that have primarily been developed in the physics , statistics , and network science literature , to construct and analyze recombinant gene networks in general , and var gene networks in particular . We apply our technique to previously published and annotated sequences of the var Duffy Binding Like-α ( DBLα ) domain [18] , [33] , which unlike other domains is found in almost all var genes sequenced . We show that networks constructed from different mosaic regions across the domain vary widely in their community structure , uncovering a new layer of micromodularity among var genes . Our results imply a lack of coupled evolution within even a single domain . At the same time , clear structures within networks correspond well , and differentially , to previously published classifications that have been linked to disease phenotypes . This structuring therefore provides a mechanism to generate vast diversity while maintaining protein structure and function , reconciling the paradoxical observations of both common serological responses and almost limitless var sequence diversity .
We analyze 307 amino acid sequences from the DBLα domain of the var genes of seven P . falciparum isolates published in [18] . PfEMP1 antigens exhibit modular structures , characterized by between two and nine DBL and CIDR ( Cysteine-Rich Interdomain Region ) domains [18] , [34]–[36] . While there are many different classes of these domains , indexed by α , β , etc . , the N-terminal region of the protein almost always begins with a DBLα and CIDRα pair , each of which has been implicated in the binding of infected red blood cells to various host receptors as well as different disease pathologies . To highlight the diversity and mosaicism of the DBLα domain we first apply standard phylogenetic approaches to the 307 sequences . Sequence length prior to alignment was widely distributed between 357 and 473 amino acids with median 420 and mode 398 . Pair-wise alignments using the standard tool MUSCLE [16] , [37] averaged 5 . 6% ( or 23 ) gaps . Due to the presence of highly variable regions , a multiple alignment required an implausibly large number of gap insertions , yielding an aligned length of 743 . The remarkable diversity in DBLα sequences is further illustrated in the unresolved nature of a phylogenetic tree built from such an alignment ( Figure S3 ) . Instead of using alignments to identify evolutionary signals contained in highly conserved regions , we use them to identify and remove conserved regions in order to focus on recombinant mosaic sequences . The method uses three steps , which we motivate here , and define in detail in the next sections: ( i ) Identify highly variable regions ( HVRs ) across all sequences . ( ii ) Compare sequences pair-wise within each HVR , generating a distinct block-sharing network for each region . ( iii ) Statistically identify communities in each network , which will represent groups of var genes that recombine more frequently with each other than with genes from other communities . These general steps are illustrated in Figure 1 . To identify HVRs , we use the basic premise of an alignment as a starting point: highly variable regions will require gap insertions in order to find an alignment . In contrast with other methods used to identify regions of conservation by discarding poorly-aligned stretches [38] , our explicit goal is to find contiguous poorly-aligned regions , since these are likely to be mosaics resulting from recombination . After identifying HVRs , instead of constructing trees , we generate a complex network for each HVR using an alignment-free process , where each vertex is a sequences and two sequences are connected if they exhibit a pattern of recombination . The structure of this network reflects the constraints and extents of the recombination process . Since any two genes may recombine in the absence of constraints on recombination , deviations from a random network represent structured recombination , function , or evolution between var gene communities . To analyze network structures , we use a community-detection approach that can identify the patterns produced by constrained recombination , by fitting a generative model called a degree-corrected stochastic block model [39] to the network data . The degree-corrected stochastic block model identifies communities by picking out non-random patterns in the network connections making it an appropriate choice among myriad community detection methods . Each step is described in detail below . In the first of three steps , we take a set of amino acid sequences and identify highly variable regions ( HVRs ) . Starting from a multiple alignment , each aligned position t is first assigned an alignment score representing the fraction of input sequences that are aligned at that position ( i . e . not gap insertions ) . This score is used to calculate an alignment indicator a ( t ) such that when all sequences align with no gaps at position t , a ( t ) = 1; if there are any gaps , a ( t ) = 0 . We then identify regions where the sequences align for G or more consecutive positions , that is , a ( t ) = 1 for G or more consecutive t . These well aligned regions will serve as separators between HVRs . We choose G sufficiently short , based on a simple null model of sequences ( SI1 ) , that any highly conserved block of significant length is removed from network construction since it would obscure patterns of recombination . Thus , the HVRs will be the regions in between the conserved regions that we have just identified . However , very short HVRs will have so few amino acids once gaps are removed in subsequent steps that they are unlikely to reflect the mosaicism in which we are interested . So , we define a minimum HVR size H , discarding those that are shorter . These steps are illustrated in Figure 1A . In the second step , we take each HVR and produce an unweighted and undirected recombination network . Each node represents a sequence , and each link represents a shared sequence block , indicating a recombinant relationship between the two sequences . Before comparing sequences , all gap insertions from the alignment process are removed . First , we create a weighted network in which the weight of each link is the length of the longest substring shared between the sequences it connects ( Figure 1B ) . The result is an all-to-all undirected and weighted network for each HVR . Thus , for sequences with multiple HVRs there will be multiple networks . Next , we convert each all-to-all weighted network into a sparse and unweighted network by discarding links with weight below a threshold , and removing weights from the remaining links ( Figure 1C ) . Here we choose the threshold in a way that controls the number of false positive links that may have arisen by chance , as shown in Figures 2A and 2B . The method for computing a noise threshold is based on a null model for randomly assembled sequences using the properties of each HVR , and not derived from network properties . Thus , depending on the confidence one wishes to have in the validity of the network's links , a threshold may be computed from a selected tolerable error rate . Derivation of the function used for this computation is included in Text S1 . In the third and final step , we detect recombination communities , by taking an unweighted and undirected HVR network and applying a degree-corrected stochastic block model [39] to identify community structures , illustrated by Figure 1D . This model takes as its input an unweighted , undirected network and the number of communities k for which it should find a maximum likelihood fit , and provides as an output a list of which nodes belong to which of the k communities , also referred to as a partition . ( Derivation and maximization of the likelihood function are discussed at length in Ref . [39] and efficient code has been made publicly available by Karrer and Newman . ) Because previous classifications included between three and six types , we inferred community structures for k = 3 to k = 6 . Each HVR network may have different community structure , similar to how in a standard approach , different loci may generate different phylogenetic trees . However , clades in trees represent distinct branches in an evolutionary history , while HVR network communities represent distinct clusters of ongoing recombination . Multiple trees may be combined to produce a consensus tree , but HVR networks show no clear consensus . We compare our resulting network communities to previous analyses of var gene sequence groups . Weights are removed from the network prior to community detection for two reasons . First , it is unclear by what principle differences in weights should be interpreted when defining communities . Second , the problem of correctly inferring degree-corrected stochastic block model community structure in sparse and weighted networks is currently unsolved . For these reasons , we interpret each network link as evidence of some recombinant or hereditary history and treat them equally by unweighting networks prior to community detection . We validate this approach carefully as follows . In order to confirm that our method is able to correctly recover recombinant communities , we validate it on synthetic data by creating sequences with varying constraints on recombination between predefined groups . We begin by creating amino acid sequences at random from an empirical amino acid frequency distribution and separating them arbitrarily into three groups . Then , we simulate recombination events in which two parent sequences recombine to produce a child sequence , inheriting the group label of one of its parents . We first choose a parent uniformly at random from the population . Then , with probability p , the other parent is chosen from the same group and with probability ( 1-p ) the other parent is chosen uniformly at random . As p is increased from zero to one , the rate of inter-group recombination goes to zero; the communities within the networks produced by applying our sequence analysis method to the synthetically recombined sequences become more well defined , and the method becomes increasingly accurate in correctly classifying nodes . As shown in Figure 3 , our method is able to recover recombinant communities perfectly in the presence of strong constraints , and performs only slightly better than random guessing when there are no constraints , as expected . Details of the validation process are found in SI3 .
The characteristics and community structures of HVR networks were diverse . Nine HVRs were found in the 307 DBLα sequences [18] , using HVR detection parameters of G = 8 and H = 6 . These parameters were chosen based on the previously described model for false positive links ( SI2 ) , and HVR boundaries were not dramatically affected by small changes to these parameters . The nine HVRs found here corresponded partially to previously identified variable regions over all DBL domains [33] . Since HVRs are by definition highly variable , they consist of mostly gap insertions—this diversity is highlighted by the fact that when gaps were removed after HVRs were identified , sequences shrank by 57% on average . Noise thresholds were computed as shown in Figure 2A such that , in expectation , 10 links ( 0 . 02% ) or less are false positives , yielding cutoff lengths of 5 , 6 , or 7 amino acids , varying by HVR . Each HVR showed a wide range of sequence lengths , and HVRs differed widely from each other in median sequence length . HVR lengths , noise cutoffs , and the percentage of links retained for community detection are found in Table 1 . The fraction of links above the length cutoffs varied by HVR , shown in Figure 2B and Table 1 . For HVRs 2–4 , removal of links below the cutoff fragmented the network and in such cases community structure cannot be inferred . Figure 2C illustrates graphically that there did not exist a threshold that both preserved a large connected component and met our requirements for a low false positive rate for HVRs 2–4 . Each remaining HVR network had identifiable communities , examples of which are illustrated in Figures 4 and 5 . This is consistent with our previous network analysis [13] , but provides greater resolution and statistical certainty that these communities represent genuine constraints on recombination . However , the membership lists of communities derived from different HVRs matched each other for only 38% of nodes on average . In addition to having widely varying community structures , HVR networks also differed from each other in component size and number of components illustrated in Figure 4 and tabulated in Table 2 . Networks are visually illustrated in Figure 4 and the number and size of components are given in Table 2 . Regardless of the number of communities detected , communities corresponded poorly to each other across HVRs . HVR networks lack consensus . Traditional phylogenetic approaches often produce a consensus tree that reflects the most likely evolutionary trajectory of a particular gene . However , if patterns of recombination for two HVRs are relatively independent of each other , we expect the communities of one HVR network to match the communities of the other only to the extent they match by chance . In contrast , HVR networks with similar recombinational constraints will have common community structures . In order to quantify the distance between community assignments , we use the variation of information statistic [40] , which is a distance metric on partitions . A small value indicates that two partitions are “close” to each other , such that the composition of one is highly correlated with the composition of the other . Figure 6A shows the pairwise distances between the inferred communities for k = 3 and partitions defined by UPS and cys/PoLV classifications , across HVRs . ( Plots for all values of k are in Figures S4 and S5B . ) In general , the communities of different HVRs were surprisingly dissimilar , except HVRs 1 and 6 , and , to a lesser extent , HVRs 1 and 5 . We compared the observed distances to an estimated distribution of pairwise distances for a null model in which we held one partitioning constant and computed distances for 10 , 000 random permutations of the other , for each pair , shown as grey symbols in Figure 6A and converted to z-scores in Figure S5B . Two examples of the randomized distributions are shown in detail in Figure 6B: the comparison of HVR 5 with HVR 9 , and the comparison of HVR 1 with HVR 6 . While the distance between HVRs 5 and 9 is smaller than the expected value of a random permutation ( top subplot ) it is significantly closer to its expected value than HVRs 1 and 6 ( bottom subplot ) . We estimated statistical uncertainty in these measurements and found that in all cases , standard deviations were O ( 10−2 ) , much smaller than the size of colored symbols plotted in Figure 6A . However , we note that this estimate is one of many possible measures of statistical uncertainty for network parameters , each of which is flawed in some way , which we discuss fully in Text S4 . Although no pair of community assignments is farther apart than expected at random , most other community assignments are only very weakly similar to each other . The fact that individual HVRs feature clear community structure implies that there are evolutionary constraints on recombination; yet comparisons of community structure between HVR networks reveal only slightly more similarity than random , suggesting that recombinational constraints at different positions are almost completely independent of each other . Thus , variable selection pressures can be accommodated even within a single DBLα domain ( distances are shown as a heatmap for all pair-wise comparisons in supplemental Figure S4 ) . These patterns suggest that mosaic sequences behave as dynamic modules that can be shared among genes relatively intact , with conserved inter-mosaic regions acting as alignment guides in the recombination process . A key finding of this analysis , therefore , is that the relative independence of different HVR networks precludes the use of “consensus” approaches sometimes used to combine trees; HVR networks were sufficiently different that they must be analyzed independently . In the absence of tools capable of handling extremely high rates of recombination , var genes have been variously classified by their domain structure and gene length , sequence characteristics , upstream promoter regions ( UPS ) , position within the chromosome , and direction of transcription [25] , [29]–[32] , [40] , [41] . However , the correspondence of these groups , which only partially overlap , remains ambiguous . As increasing volumes of var gene sequence data are produced from studies in the field , understanding how best to resolve and refine these approaches will be key to interpreting study outcomes . We compared previous classification systems , as well as each parasite genotype , to communities within each HVR network . Individual parasite var repertoires reflect population-level diversity . Examining the sequences of individual parasites , we found that each of the seven parasites' var sequences are evenly spread through the clusters of the network , rather than forming genome-specific communities ( Figure S1 ) , consistent with previous studies [10] . Regardless of how many communities k we choose , each parasite had at least one sequence in each recovered community , showing that a single parasite is not confined to an identifiable genotypic community but instead has samples of all major communities that we identified . This corroborates previous research showing that var genotypic diversity within a single parasite is as high as the diversity of the parasite population [20] . This pattern is also consistent with theoretical work has shown that selective pressure on the var genes should create parasites with as wide a variety of genotypes as possible for immune evasion , while still preserving enough structure for adhesion and sequestration [42] . Thus , each P . falciparum genome contains an antigenic repertoire that is effectively sampled from the diversity of the global pool of var genes . While HVR networks tend to differ from each other , their communities correspond to known upstream promoter sequence ( UPS ) groupings and var gene length . Upstream promoter sequences were previously categorized as UPS A to UPS E or Not Determined ( ND ) [18] . DBLα with UPS D were not present in the 307 sequences examined . The inferred communities in HVRs 1 and 6–8 place nearly all UPS A sequences together , plotted in Figure 7A . The remaining communities comprise a mix of UPS B , C , and ND sequences . This implies that recombinant mixing of UPS A with B or C is comparatively rare , leading to the hypothesis that three ND sequences could be classified as UPSA: IT4var24 , PF07_0048 , and IT4var51 . Furthermore , we found no evidence for a strong separation of var genes into distinct groups for UPS B and C . Although UPS C genes are all found proximate to the centromeres of P . falciparum genomes , recombination appears to occur frequently between centromeric and subtelomeric UPS B genes , consistent with studies of chromosomal positioning during ectopic recombination [14] , [43] , but not yet observed in vitro [44] . Using the variation of information distance measure [40] , Figure 6A shows that the community structures of HVR 1 and UPS communities are particularly close . This is reinforced by a measurement of assortative mixing by label [45] , shown in Figure 6C , a measure of correlation among node labels that takes into account the connections of the network and may be computed without any community assignments . In particular , this reflects the fact that in HVR 1 , many UPS A nodes tend to link almost exclusively to other UPS A nodes . Plots of all HVR networks colored by UPS group are found in Figure S2 . We also found positive assortative mixing by var gene length ( including all NTS , DBL , CIDR , and ATS domains ) implying that genes of the similar length tend to link to each other , consistent with dynamical models of var gene evolution [42] ( Figure 6C ) . Thus , both UPS group and gene length may play roles in constraining recombination , suggesting that future models of evolution and recombination must reproduce these results . We confirmed and extended the Cys2/Cys4 and PoLV classification schemes which previously identified and analyzed patterns within the DBLα domain [10] , [46] , [47] . In an alignment-free study of short tagged sequences corresponding to HVRs 5 and 6 from Kilifi , Kenya , it was found that sequences may be classified based on the number of cysteine residues present in HVR 6 , and that more severe disease phenotypes were correlated with the presence of only two cysteines [10] , [46] . These sequences are referred to as “cys2” sequences , and others , most of which have four cysteines , are referred to as “cysX . ” The cys2/cysX classifications can be further classified according to a set of mutually exclusive sequences motifs called positions of limited variability ( PoLV ) , splitting cys2/cysX categories into six cys/PoLV groups [10] , [47] . Since the HVR network approach creates links based on shared block structure , it captures cys2/cysX and cys/PoLV categorizations . Unsurprisingly , this is reflected most strongly in HVR6 ( Figures 6A and 6C ) , but is also present in HVR1 whose nodes are strongly assortatively mixed by cys/PoLV group ( Figure 6C ) , corroborating the result that HVRs 1 and 6 are structurally similar . There exists one community in HVRs 1 , 6 , and 5 that contains most of the cys2 sequences ( Figure 7B ) . The correspondence of network structures with previous phenotype-associated sequences also suggests that the HVR network approach may be extended to map genotypic patterns to clinical phenotypes when applied to an appropriate data set that includes expression data and clinical information . The DBLα domains we use here were previously analyzed and categorized extensively using tree-based methods , revealing many new subclassifications of the previously identified DBLα0 and DBLα1 [41] , [43] as well as a new classification , DBLα2 [18] . We examined HVR networks for evidence of strong associations between network community structure and subclassifications , but found that only DBLα1 . 3 sequences had a strong tendency to link to other sequences from the same subclassification , particularly in HVRs 7 , 8 , and 9 , where they formed cliques on the periphery of the networks . While the sequence differences of the DBLα1 . 3 subclass have already been noted [18] , we also find that the DBLα0 . 3 subclass tended to link to each other in HVRs 1 and 8 , though not in a clique . No strong correlations were found between network structures and the broader classes of DBLα0 , DBLα1 , and DBLα2 , and networks were not strongly assortative by DBLα class , as shown in Figure 6C . This is perhaps unsurprising , given the different focus of our method , but it highlights the fact that any classification of recombinant genes that is based on a tree-like phylogeny may not be informative about the process of recombination . Our approach highlights and analyzes the within-domain modularity of DBLα . Var genes are by definition modular , with variable numbers and types of DBL and CIDR domains clustered into larger groups that primarily correspond to UPS A and UPS B or C . We have shown that in addition to modular domains that can be shuffled between loci , there are also relatively regularly spaced modular mosaics within the DBLα domain that are shuffled , under constraints , between var genes . Previous studies have suggested that the conserved blocks in DBL domains may provide structural support for the protein while the variable regions in between are loops in the protein designed specifically for antigenic variation under diversifying selection [17] , [33] . Here we offer a more nuanced hypothesis , shown schematically in Figure 8: HVRs under related recombinational constraints may have important functional roles in the PfEMP1 molecule , while other HVRs may exist primarily for purposes of antigenic variation . We find that HVRs 1 , 6 , and to a lesser extent 5 , have similar non-random network community structure that corresponds strongly with structural amino acid residues and classification systems with known associations with severe disease [47] . These regions of the domain may therefore be functionally constrained and play a specific role in binding . Interestingly , HVRs 5 and 6 correspond to the short tag sequences that have previously been amplified from field isolates . If our hypothesis is correct , it would explain why the clustering of these sequence tags exhibit meaningful associations with disease outcome [46] , [48] . The remaining HVRs 7–9 , have highly heterogeneous community structures that bear almost no relation to each other , suggesting that their primary role is in the generation of diversity for the purpose of immune evasion . Having both correlated and uncorrelated recombinational constraints across multiple HVRs thus provides at least two important evolutionary benefits to the parasite: i ) individual mosaics that are functionally important can retain their function without compromising the generation of diversity across the rest of the domain , and ii ) recombination could produce variants through different combinations of modules more rapidly than mutation or random recombination , and without risking the recombinational equivalent of error catastrophe that occurs in systems with very high mutation rates [49] . In other words , the parasite population may be rapidly shuffling ancient sequence mosaics into new combinations across some HVRs , while also preventing the degeneration of structurally important regions of the protein that are involved in binding . The population of genes may thereby balance a need for new diversity with functional requirements . The varying correspondence of HVR communities with previously defined var gene groupings implies that the different classification schemes complement each other , providing insights into different aspects of var gene evolution , likely representing nested or hierarchical recombinant clusters . We measure the extent to which previously defined groupings are reflected in the links of our networks using assortativity , shown in Figure 6C . Importantly , the DBLα group assortativity is much lower than all the others , demonstrating further that while there are clear structures in HVR networks , they are not the same as the classifications based on trees . Such tree-based classifications explain the development of large-scale structure over the relatively longer timescales of mutation , whereas communities detected within recombination networks here shed light on functional or even mechanistic constraints on recombination occurring more recently . The method presented here can accurately extract block-sharing networks from the most highly recombinant regions of protein sequences . While a multiple-alignment is used to identify HVRs ( Figure 1A ) , we find that HVR boundaries are robust to changes in alignment parameters , and the remaining steps of our network extraction process ( Figures 1B–D ) are entirely alignment-free . Since the structure of such networks reveals patterns of recombination , strong communities within networks are indicative of functional or evolutionary constraints on recombination within the underlying population . By applying this technique to the DBLα domain of var genes we find that different locations of the domain produce different communities . Were the constraints identical in each domain , we would expect network communities to be similar to each other and exhibit small variation of information distance . The fact that network communities differ therefore indicates that while constraints exist at each location in the domain , they also vary by location . We suggest that this lack of correlation between constraints allows the DBLα domain to possess exponentially more complexity while simultaneously remaining functional , avoiding recombinant error catastrophe despite extremely high rates of recombination . The combination of principled and alignment-free network construction methods with state-of-the-art generative models for community detection may open the door to new research areas , linking evolution , function , and clinical phenotypes in a range of genetically diverse pathogens . While we demonstrate our method using DBLα , it could also be applied to other highly recombinant var domains , other P . falciparum genes such as the rif and cirs families , or other pathogens , such as HIV , the pneumococcus , or trypanosomes . Methodologically , extensions include the development and application of stochastic block model community detection in weighted networks , treating the separate HVR networks as a single multiplex network [50] , and generalizing the current process to inexact sequence matches . | The human malaria parasite kills nearly 1 million people each year globally . Frequent genetic exchange between malaria parasites creates enormous genetic diversity that largely explains the lack of an effective vaccine for the disease . Traditional phylogenetic tools cannot accommodate this type of diversity , however , and rigorous analytical tools capable of making sense of gene sequences that recombine frequently are still lacking . Here , we use network techniques that have been developed by the physics and network science communities to analyze malaria parasite gene sequences , allowing us to automatically identify highly variable mosaic regions in sequence data and to derive the network of recombination events . We apply our method to seven fully-sequenced parasite genomes , and show that our method provides new insights into the complex evolutionary patterns of the parasite . Our results suggest that the structure of these sequences allows the parasite to rapidly diversify to evade immune responses while maintaining antigen structure and function . | [
"Abstract",
"Introduction",
"Methods",
"Results/Discussion"
] | [] | 2013 | A Network Approach to Analyzing Highly Recombinant Malaria Parasite Genes |
Circadian clocks control many self-sustained rhythms in physiology and behavior with approximately 24-hour periodicity . In many organisms , oxidative stress and aging negatively impact the circadian system and sleep . Conversely , loss of the clock decreases resistance to oxidative stress , and may reduce lifespan and speed up brain aging and neurodegeneration . Here we examined the effects of clock disruptions on locomotor aging and longevity in Drosophila . We found that lifespan was similarly reduced in three arrhythmic mutants ( ClkAR , cyc0 and tim0 ) and in wild-type flies under constant light , which stops the clock . In contrast , ClkAR mutants showed significantly faster age-related locomotor deficits ( as monitored by startle-induced climbing ) than cyc0 and tim0 , or than control flies under constant light . Reactive oxygen species accumulated more with age in ClkAR mutant brains , but this did not appear to contribute to the accelerated locomotor decline of the mutant . Clk , but not Cyc , inactivation by RNA interference in the pigment-dispersing factor ( PDF ) -expressing central pacemaker neurons led to similar loss of climbing performance as ClkAR . Conversely , restoring Clk function in these cells was sufficient to rescue the ClkAR locomotor phenotype , independently of behavioral rhythmicity . Accelerated locomotor decline of the ClkAR mutant required expression of the PDF receptor and correlated to an apparent loss of dopaminergic neurons in the posterior protocerebral lateral 1 ( PPL1 ) clusters . This neuronal loss was rescued when the ClkAR mutation was placed in an apoptosis-deficient background . Impairing dopamine synthesis in a single pair of PPL1 neurons that innervate the mushroom bodies accelerated locomotor decline in otherwise wild-type flies . Our results therefore reveal a novel circadian-independent requirement for Clk in brain circadian neurons to maintain a subset of dopaminergic cells and avoid premature locomotor aging in Drosophila .
Circadian clocks are ubiquitous in the living world , driving rhythms at many levels , from the molecular to the behavioral [1] . The main defining characteristics of these clocks are that: ( i ) they are self-sustained , ticking for many days in constant external conditions , ( ii ) their period in such conditions , called free-run , is close to 24 h . Circadian clocks are normally synchronized to the solar day , ensuring appropriate timing of the processes they control . Light and temperature are the main "Zeitgebers" , or synchronizing signals . The cell-autonomous machineries of circadian clocks have been described in great detail in many taxa , including bacteria , plants , mammals and insects [2 , 3] . They usually include negative transcriptional feedback loops , which are exceedingly well conserved evolutionarily . In Drosophila , one of these feedback loops involves a pair of transcriptional activation factors , Clock ( Clk ) and Cycle ( Cyc ) . These basic-helix-loop-helix ( bHLH ) proteins form heterodimers via their PAS interaction domains ( reviewed in [3] ) . Clk-Cyc dimers activate the period ( per ) and timeless ( tim ) genes , which are subsequently turned off by their own gene products , as these also form dimers that bind to and inhibit Clk-Cyc . Post-transcriptional mechanisms introduce appropriate delays to allow inhibition to start only after the per and tim mRNAs have accumulated to high levels . Both the latter mechanisms and the feedback loop itself are homologous in mammals , except for the replacement of tim by cryptochromes ( and the naming of cyc as Bmal1 ) . Clocks may allow organisms both to better adjust to predictable nycthemeral changes in their environment , and to achieve some temporal order in their functioning , independently of solar cycles [4] . In any case , the functional importance of clocks is attested by the numerous adverse effects of perturbing them on healthspan , both in insects [4–10] and rodents [11–16] . This ever-increasing list is consistent with the presence in all tissues of functional clocks , controlling up to 10% of the genes expressed in any given tissue [17] . Experimental clock disruptions can be genetic , but also environmental , e . g . with constant light ( LL ) that stops the clocks by activating Tim degradation [1 , 2] , or with light-dark ( LD ) cycles of non-24h periodicity , or which are shifted by several hours every few days to generate chronic jet lag-like conditions . In humans , epidemiological data , although only correlative , also suggest that chronic circadian disruption , such as in long-term shift work , increases the incidence of obesity ( and metabolic syndrome more generally ) , cardiovascular diseases , and some cancers , to name only a few pathologies [18] . In large cities , artificial light at night , combined with weak indoor light during the day , may also sufficiently disrupt our circadian clocks to produce similar , if weaker , ill effects on the general population [15] . Age-related locomotor declines or impairments ( ARLI ) , which have a strong impact on life quality , are found in most species . They are thus often used to assay functional aging , particularly in insects ( reviewed in [19] ) . Dopaminergic ( DA ) circuits are important for proper motor control and abilities , both in mammals [20–22] and Drosophila [23–27] . This importance is illustrated by the devastating effect of specific DA neuronal loss on motor control in Parkinson's disease ( PD ) , for which age is the major risk factor [28] . There are perturbations of the circadian system in neurodegenerative diseases , including PD ( sometimes before any motor symptoms ) , and their animal models , as reviewed [29] . The inverse relationship–circadian disruption causing neuronal loss–remains hypothetical , but is consistent with a recent study in mice [30] . Here , we studied mutations in the two Drosophila circadian transcriptional activators Clk and Cyc , and in one of their target genes , tim . Male flies were kept either in standard LD cycles or in LL , and assayed at different ages for survival , locomotor performance , brain reactive oxygen species ( ROS ) levels , and size of brain dopaminergic neuronal subpopulations . Our results confirmed the negative impact of genetically- or environmentally-imposed arrhythmia on healthspan . They also revealed unexpected circadian rhythm- and Cyc-independent effects of Clk gene disruption , namely increased brain ROS levels and a markedly accelerated decline of locomotor responses in aging flies . The latter phenotype was attributed specifically to Clk function in the small lateral ventral neurons ( s-LNv ) , which express the neuropeptide PDF and constitute an important circadian pacemaker in the Drosophila brain , and may be accounted for by the observed effect of Clk inactivation on the PPL1 clusters of brain dopaminergic neurons .
We first examined the effects of genetic and environmental clock disruptions on Drosophila lifespan and ARLI , assessed by monitoring the performance , at successive ages , of groups of flies in a startle-induced negative geotaxis ( SING ) assay . We found that the ClkAR , cyc0 and tim0 mutations each had a similar significant but moderate impact on survival , reducing lifespan by at most 15% ( S1A and S1B Fig and S1 Table ) . In Drosophila , ARLI was found to be faster in arrhythmic per0 mutants [8 , 31] . Here we observed that ARLI was also moderately but significantly accelerated in the arrhythmic tim0 and cyc0 mutants , after 4 weeks of adult life ( Fig 1A ) . Rearing Canton-S flies ( controls ) in LL , which stops the clock , produced a similar ARLI acceleration ( Fig 1B and S1C Fig ) . This could be consistent with accelerated ARLI and reduced survival being general detrimental consequences of long-term circadian arrhythmicity . ARLI in the arrhythmic ClkAR mutant , however , was strikingly more precocious , starting as early as 10 days after adult eclosion ( Fig 1C ) . Interestingly , locomotor decline was less rapid but still significantly faster , compared to control flies , in ClkAR/+ heterozygous flies , which are fully rhythmic [[32] and S2 Table] , while it was not affected in cyc0/+ heterozygotes ( Fig 1D ) . The impact of the ClkAR mutation on ARLI thus appears separate from its effect on rhythmicity . In contrast , average spontaneous locomotor activity in LD during the day displayed little change between young ( 10- to 15-day-old ) and old ( 31- to 36-day-old ) flies in any of the four tested genotypes ( S1D Fig ) . In young ClkAR and cyc0 flies , night-time activity was higher than in controls , as previously reported for ClkAR [32] , or than in tim0 mutants . Although night-time activity strongly decreased with age for both ClkAR and cyc0 mutants , it did not become lower than in the controls ( S1D Fig ) . Circadian disruption also affects sleep , and sleep disruptions by themselves negatively impact the brain [33] , and may thus contribute to accelerated ARLI . Sleep in LD was similarly affected in the ClkAR and cyc0 mutants , with strongly increased latency after lights-off , and reduced total and night-time sleep ( S2A–S2C Fig ) . These results are similar to those previously reported for ClkJrk and cyc0 [34] . However , the sleep profiles of the heterozygous mutants , ClkAR/+ and cyc0/+ , were indistinguishable from controls ( S2D Fig ) . In addition , knocking down wake ( a gene involved in the transition from wake to sleep at the end of the day ) in the LNvs , which was shown to disrupt sleep [34] , did not affect the age-related impairment of the SING behavior ( S2E Fig ) . The ClkAR mutation therefore produces a specific ARLI phenotype , which appears independent from its disrupting effect on circadian rhythmicity , and which does not correlate with accelerated global aging , disrupted sleep , or a major loss of motor ability . In LL , ARLI was not significantly affected by the cyc0 or tim0 mutations ( Fig 1E ) , suggesting that ARLI acceleration in these mutants is due to arrhythmia . In contrast , in LL ARLI was still much faster for the ClkAR mutant than for control flies ( Fig 1F ) . The difference between the ClkAR mutant and the control in LL could be considered as representing the rhythm-independent effect of the ClkAR mutation on ARLI . Analyses of sleep showed no differences between controls and ClkAR mutants in LL ( S3A–S3C Fig ) . Indeed , in LL , total sleep time was similar for control , ClkAR and cyc0 flies ( S3B Fig ) . Sleep fragmentation , as assessed by average night ( or presumptive night ) bout duration , was stronger in LL than in LD ( S3C Fig ) . However , in LL it was now very similar between all three strains . Thus the accelerated ARLI observed for the ClkAR mutants in LL , relative to control or cyc0 flies , again does not correlate with a poorer sleep quality . In order to determine where Clk function is required to maintain a wild-type SING behavior in aging flies , we used the Gal4-UAS system to express a Clk RNAi under the control of various Gal4 drivers . Heterozygous UAS-ClkRNAi/+ and Gal4 driver/+ flies were used as controls . When RNAi expression was driven in all clock cells with tim-Gal4 , lifespan was significantly shortened relative to both controls ( S4A Fig , S3 Table ) . ARLI was greatly accelerated in the ClkRNAi-expressing flies ( S4B Fig ) . The acceleration was weaker than in the mutant , as it was not observed on days 10 and 17 . This could only reflect a weaker loss of function , as ClkRNAi also led to a weaker impact on free-running rhythms compared to ClkAR ( S2 Table ) . We also observed a smaller but significant effect on ARLI for the tim-Gal4/+ control ( S4B Fig ) . When RNAi expression was driven selectively in PDF-expressing neurons with pdf-Gal4 , there was no lifespan reduction ( S4C Fig , S3 Table ) . However , ARLI was greatly accelerated , this time with the two controls behaving similarly to wild-type flies ( Fig 2A ) . This indicates a specific effect of Clk on ARLI , independently of any acceleration in global aging . Accelerated ARLI was also observed with pdf-Gal4 driving another independent ClkRNAi ( UAS-ClkRNAi-R3 ) ( S4D Fig ) . Its weaker effect probably reflected a weaker inhibition of Clk function , as suggested by its also weaker effect on free-running rhythms ( S2 Table ) . cyc knock-down in the LNvs had no effect on ARLI ( Fig 2B ) , although cycRNAi disrupted behavioral rhythms more than ClkRNAi ( S2 Table ) . This is again in line with a specific effect of Clk deficiency on ARLI , unrelated to circadian rhythmicity . The ClkAR mutation may lead to abnormal Clk protein products [32] , but the similarity between the ClkAR ARLI phenotype and that of the two Clk RNAi indicates that accelerated ARLI is likely due to Clk deficiency rather than a gain-of-function effect . We used additional driver lines to further pin down the neurons where Clk function is required for wild-type ARLI in aging flies . Two neuronal groups express both PDF and Clk: the small LNvs ( s-LNvs ) and the large LNvs ( l-LNvs ) . The R6-Gal4 driver allowed us to knock-down Clk in the former , while the C929-Gal4 driver allowed us to do the same in the latter . Knocking down Clk in the s-LNvs did accelerate ARLI ( Fig 2C ) , while knocking down Clk in the l-LNvs had no effect ( S4E Fig ) . Interestingly , we observed that ablating all the PDF neurons through expression of the pro-apoptotic gene hid [see [35]] also had no effect on ARLI ( Fig 2D ) . This indicates that the s-LNvs are not required to preserve normal locomotor aging in Drosophila and that Clk deficiency may induce an alteration in the activity pattern of these cells that would lead to accelerated ARLI . We then asked conversely whether restoring Clk expression selectively in the PDF-expressing LNvs , using the pdf-Gal4 driver , would rescue ARLI in an otherwise ClkAR mutant background . This was indeed the case ( Fig 3A ) , even though these rescued flies were behaviorally arrhythmic ( S2 Table ) , as previously reported [32] . This further demonstrates a specific effect of Clk on ARLI that is unrelated to disruption of behavioral rhythms . We wondered whether overexpressing Clk in the LNvs would modulate ARLI in a wild-type background . It did not ( Fig 3B ) , nor did it affect behavioral rhythms ( S2 Table ) . One factor that has been tentatively linked to ARLI is oxidative stress [19] . We therefore estimated brain ROS levels by dye staining in control , ClkAR and cyc0 flies , at various ages post-eclosion . As previously reported for oxidative damage [8] , ROS levels increase with age , especially between 24 and 31 days post-eclosion , and this was true for both control and ClkAR flies ( S5A Fig ) . This increase may play a part in the onset of ARLI in control flies , as it is concomitant with or slightly precedes it ( Fig 1A , 1C and 1D ) . ROS levels were higher in ClkAR than in controls already at 10 days , and remained higher at 31 days ( Fig 4A and 4B and S5B Fig ) . By 45 days control and ClkAR brains displayed similarly elevated ROS levels ( S5B Fig ) . In contrast , no increase in brain ROS levels was found for the cyc0 mutant relative to control ( Fig 4A and 4B ) . Increased ROS levels are then another ClkAR mutant phenotype which , like strongly accelerated ARLI , is not caused by clock disruption alone . The precociously elevated ROS levels in ClkAR brains could be hypothesized to cause their precocious ARLI phenotype . However , we found no increase in ROS levels in the brains of pdf>ClkRNAi flies relative to their two controls ( Fig 4C ) , even at 31 days of age , although the former flies displayed accelerated ARLI well before that age ( Fig 2A ) . Conversely , whereas restoring Clk expression selectively in the LNvs in a ClkAR background rescued ARLI ( Fig 3A ) , it did not reduce brain ROS levels relative to the two controls ( Fig 4D ) . There is thus no correlation between elevated brain ROS levels and accelerated ARLI in Clk-deficient flies ( ClkAR and pdf>ClkRNAi ) , indicating that a global increase in brain oxidative stress is neither necessary nor sufficient to cause their accelerated ARLI phenotypes . Because dopaminergic circuits are involved in locomotor control in flies ( see above in the Introduction ) , we wondered whether the ClkAR mutation could affect dopaminergic neurons . The adult Drosophila brain contains eight clusters of dopaminergic neurons [36] . We counted the number of these neurons in most classes ( S6 Fig ) except the protocerebral anterior median ( PAM ) cluster , which was difficult to quantify precisely because of its large size . The number of neurons immunoreactive for tyrosine hydroxylase ( TH-IR neurons ) , the rate-limiting enzyme for dopamine synthesis , was not significantly different in the brains of 10-day-old ClkAR , cyc0 and control flies ( Fig 5A ) , at an age when the effect of the ClkAR mutation on SING is still very small ( see Fig 1C ) . In contrast , in 31-day-old flies , we observed a significant and selective loss of TH-IR neurons in the PPL1 cluster of ClkAR mutants relative to control , while cyc0 brains were unaffected ( Fig 5B and S7 and S9A Figs ) . A similar loss was also observed in 31-day-old ClkAR/+ heterozygotes ( Fig 5C and S7 Fig ) , as well as in pdf>ClkRNAI flies where Clk expression was knocked down in the LNvs , relative to both controls ( Fig 5D and S7 Fig ) . That loss , like the accelerated ARLI phenotype ( Fig 3A ) , was rescued by Clk expression restricted to the PDF neurons ( S10A Fig ) . In contrast , the size of several serotonergic clusters in a broad dorso-lateral protocerebral region [see [37]] , where the PPL1 is located , did not differ between aging ClkAR and control flies ( S11 Fig ) . Accelerated ARLI due to impaired Clk function appears thus selectively associated with a loss of TH-IR neurons in the PPL1 cluster , although we cannot rule out that a few PAM neurons are also affected ( see also below ) . We observed that the PPL1 neuronal cell bodies are located close to the s-LNv dorsal projections , as they turn medially in the protocerebrum ( S9A Fig ) , suggesting that these dopaminergic neurons could be influenced by the s-LNvs , through a direct or paracrine connection . Note that projections from the s-LNv appeared wild-type in the ClkAR , ClkRNAi- and cycRNAi-expressing flies , while they were disrupted in cyc0 flies ( S7–S9 Figs ) , as previously reported for both cyc0 and ClkJrk [38 , 39] . Indeed one of the reasons that prompted us to analyze the ClkAR rather than the ClkJrk mutant was to avoid such a developmental effect , another one being the known dominant-negative character of ClkJrk [40] . To test the possibility that PPL1 neuronal loss may involve the activation of apoptotic pathways , we used the H99 deficiency , which removes a pro-apoptotic gene cluster including hid , rpr and grim . H99/+ heterozygous flies are viable , but lack almost all embryonic programmed cell death [41] . No decrease the number of TH-IR PPL1 neurons was observed in H99/+ , ClkAR flies relative to H99/+ controls , which contrasted with ClkAR alone ( S10B Fig ) . Although this does not demonstrate that PPL1 neurons actually undergo apoptosis in aging ClkAR brains , it indicates that the activation of apoptotic pathways is indeed involved in the mutant phenotype . The main signaling molecule secreted by the s-LNv neurons is the neuropeptide PDF . It was recently reported that expression of a dominant-negative form of the circadian kinase doubletime ( dbtK/R ) in the LNvs led to transient daily caspase activation , prominently in the optic lobes , by a circadian-independent mechanism that requires PDF receptor signaling [42] . Transient caspase activation was also observed in head extracts of the ClkJrk mutant [42] . Here we found that the ClkAR mutation was unable to accelerate ARLI in the absence of the PDF receptor ( Fig 6A ) . Remarkably , in a ClkAR mutant background , the lack of PDF receptor also fully rescued the loss of dopaminergic neurons in the PPL1 cluster ( Fig 6B ) . To test the possibility that loss of TH-IR neurons in the PPL1 cluster may impair SING in older flies , we first eliminated at least part of these neurons by expressing the pro-apoptotic gene hid under control of the TH-D1-Gal4 driver [43] . This accelerated SING decline ( Fig 7A ) , comparably to the effect of knocking down Clk in the s-LNvs ( Fig 2A ) . Similar results were obtained with the TH-D'-Gal4 driver ( Fig 7B ) , which is more specific for dopaminergic neurons [43] . However , TH-D'-Gal4 targets other dopaminergic neurons , outside the PPL1 , in the PPL2 and PPM3 clusters , which may contribute to the observed effect on ARLI . Therefore we used an alternative strategy [27]: inactivating Drosophila TH by targeted RNAi . By driving TH RNAi expression with MZ840-Gal4 , which is expressed in only one PPL1 neuron ( MB-V1 ) and in other non-dopaminergic brain cells [44] , the two MB-V1 neurons should be the only affected ones . We observed that TH knock-down with MZ840-Gal4 also accelerated ARLI ( Fig 7C ) , again comparably to the effect of Clk knock-down in the s-LNvs ( Fig 2A ) . The impact of reducing Clk function in the s-LNvs is thus phenocopied by reducing the number or function of PPL1 neurons , and particularly of the MB-V1 pair , consistent with other neurons ( including dopaminergic ones within the PAM ) not playing a major effect in the accelerated ARLI of ClkAR flies .
We report here that both the ClkAR mutation and reducing Clk function in the PDF-expressing LNvs led to faster decline of a startle-induced locomotor response in aging flies , and to dopaminergic neuron disruption selectively in the PPL1 cluster . A significant acceleration of ARLI was also observed in heterozygous ClkAR/+ flies , which are fully rhythmic . Such Clk gene-specific phenotypes were indeed completely clock-independent , as they were not observed when the circadian clock was stopped or perturbed in other ways , such as in LL conditions , or by cyc0 and tim0 mutations . In contrast , we found that spontaneous locomotor activity , sleep , and lifespan were similarly affected in the ClkAR and cyc0 mutants . Other clock-independent functions of clock genes or the LNvs have been reported in Drosophila , e . g . in reproduction [45 , 46] , responses to cocaine [47 , 48] and sensitivity to sleep deprivation [49] . In the latter case , Cyc seems to play a specific role , as Clk , per , and tim mutants do not display hypersensitivity to sleep deprivation like the cyc0 mutant . In what cells Cyc is required to protect the flies from the effects of sleep deprivation is not known . However , in most studies , Clk and cyc mutants had the same or similar phenotypes [34 , 40 , 50–53] . Clk and Cyc are co-expressed in all brain clock neurons , and neither one appears to be expressed in any other brain neurons [54] . Ectopic expression of Clk but not Cyc is sufficient to induce a functional clock in other brain neurons; it does require Cyc function , however [55 , 56] . It remains to be investigated whether Clk requires another partner in circadian neurons to prevent accelerated ARLI and dopaminergic neuron loss . We observed that removing the PDF-expressing neurons had no effect on ARLI , in contrast to the striking phenotype induced by Clk inactivation in the same neurons . This suggests that Clk inactivation does not inhibit but rather overactivates PDF-neuron signaling . Both the locomotor and neuronal loss phenotypes of the ClkAR mutant were indeed rescued in the absence of PDF receptor . Recently , the ClkJrk mutation , but not per0 , was shown to transiently activate the Dronc caspase in fly heads , in response to light during the day . Furthermore , the expression of a dominant negative form of the circadian kinase doubletime ( dbtK/R ) in the LNvs was shown to activate the Dronc caspase in large parts of the brain , independently of the clock itself but dependent on PDF receptor signaling [42] . It is therefore quite possible that Clk inactivation in the s-LNvs can induce PDF receptor-dependent caspase activation in target or neighbouring cells , that will eventually lead , directly or indirectly , to dopaminergic neuron loss ( see below ) . The precocious increase in ROS levels in ClkAR mutant brains could also contribute to caspase activation . Again , the clock and Cyc do not seem to be involved as ROS levels were not affected in the cyc0 mutant . However , our targeted Clk rescue and knock-down experiments suggest that a global increase in brain ROS levels accelerates ARLI only marginally , if at all , in this context . The global ROS increase may contribute to some differential effects of the ClkAR versus cyc0 mutations ( e . g . on sensitivity to sleep deprivation , since elevated ROS levels may induce protective mechanisms , see e . g . [57] ) , or of the ClkAR mutation versus LNv-restricted expression of ClkRNAi ( such as the one we found on longevity ) . Our results suggest the existence of a link between the s-LNvs and the PPL1 cluster of dopaminergic neurons that involves PDF-receptor signaling ( Fig 8 ) . Inverse links between clock and DA neurons have already been demonstrated: dopamine strongly modulates cAMP levels in l-LNvs , via different receptor subtypes , while s-LNvs are much less responsive [58] . The identity of the DA neurons afferent to the l-LNvs is still unknown . We observed that the PPL1 cluster lays close to the s-LNv projections to the dorsal protocerebrum . s-LNv projections were impaired in the cyc0 mutant , as previously shown [38 , 39] , but not in the ClkAR mutant . This difference does not seem to play a part in the differential impact of the two mutants on ARLI , since s-LNv projections appeared completely wild-type when expressing RNAi for either cyc or Clk in the PDF neurons . Although cycRNAi was more effective than ClkRNAi in disrupting circadian activity rhythms , it did not accelerate ARLI at all . As no PDF processes were observed in apposition to TH-IR neurons in the posterior dorsal brain [59] , signaling to the PPL1 may well be paracrine rather than synaptic , consistent with both anatomical [60] and functional [61] evidence for such signaling mechanism by PDF . On the other hand , PDF receptor in that protocerebral region appears exclusively expressed in 3 LNd clock neurons , as judged from anti-MYC labeling of a PDF receptor-MYC fusion [62] . The LNds might thus be intermediary neurons mediating the influence of the s-LNvs on the PPL1 . What happens to the PPL1 neurons when Clk function is compromised in the LNvs ? Similar questions were raised in fly models of PD , where the size of various TH-IR neuronal clusters often appeared reduced ( see e . g . [63–68] ) , but actual cell loss was sometimes contested [69 , 70] . Here we report that reducing TH levels in specific neurons was sufficient to accelerate ARLI , as was also shown in our previous study [27] . In the latter , accelerated ARLI was attributed not to cell loss , but to age-related alterations in contacts onto the mushroom bodies from a subset of ~15 PAM neurons , when they express α-synuclein ( a protein involved in PD ) . In agreement with that , we found that locomotion was already affected in 10-day-old ClkAR mutants , before any visible reduction in dopaminergic cell numbers . The MB-V1 neuron , which is currently our best candidate among PPL1 neurons as mediator of Clk deficiency-induced ARLI acceleration , also projects to the mushroom bodies [71] . It remains to be seen whether this specific neuron is among those that disappear or lose TH-IR in aging flies when Clk is down-regulated in the LNvs , and whether the loss of other PPL1 neurons would be sufficient to accelerate ARLI . The PPL1 cluster was reported to be selectively reduced in flies expressing α-synuclein under control of TH-Gal4 [66] , a driver that does not express strongly in the PAM [23] . That reduction was rescued in an apoptosis-deficient genetic background [66] , similarly to what we observed for the effect of the ClkAR mutation , or by overexpression in dopaminergic neurons of either glutathione S-transferase Gst1 [66] , the Nrf2 transcription factor [67] or the endosomal recycling factor Rab11 [72] . Whether these latter effectors can also protect the PPL1 neurons in Clk-deficient flies , as well as how apoptotic pathways may affect PPL1 neurons in these contexts , remain open questions . Our results show that Clk inactivation in the main Drosophila pacemaker neurons , the s-LNvs , has dramatic consequences for a locomotor response and select dopaminergic neurons maintenance in aging flies , independently of the less prominent but significant defects induced by circadian rhythm disruptions . Such progressive , circadian clock-independent effects both add to , and contrast with , previously described links between Clk and dopaminergic signaling . The acute nocturnal hyperactivity and reduced sleep of ClkJrk mutants , for instance , were attributed to increased dopaminergic transmission [73] . However , they involved Cyc as well , and the l-LNvs rather than the s-LNvs . In mice , a Clk mutant also displays decreased sleep and hyperactivity , as well as a mania-like behavior , presumably owing to increased activity of dopaminergic neurons in the ventral tegmental area [74 , 75] . However , contrary to flies , Clk is expressed in such neurons , where its loss increases expression of genes involved in dopaminergic signaling , including TH [74 , 75] , consistent with the observed phenotypes . Interestingly , the combined deletion of Clk and its paralog Npas2 induced severe age-dependent astrogliosis in the mouse brain , leading to degeneration of synaptic terminals , neuronal oxidative damage and impaired expression of several redox defense genes [76] . Neuron and glia-specific inactivation of BMAL1 , the mouse ortholog of Cyc , produced similar phenotypes . In Drosophila , our results now indicate that Clk regulates s-LNv activity , independently of its role in the circadian machinery , and that the loss of this regulation leads to progressive dysfunction of specific brain dopaminergic neurons , by mechanisms that involve PDF receptor signaling . Further deciphering the wide-ranging effects of Drosophila Clk dysfunction , whether circadian clock-dependent or not , or cell-autonomous or not , could shed new light on the regulation of dopaminergic neuron survival in both flies and mice .
Flies were maintained on standard medium at 25°C , under a 12:12 Light/Dark cycle ( LD 12:12 ) , or under constant light . Light intensity at the level of the flies was in the range 500–3000 lux . The following clock mutants were backcrossed for five generations to a wild-type ( Canton-S ) genetic background before use: ClkAR [32] , cyc0 [77] , pdf0 [35] , tim0 [78] . cyc0 , pdf0 and tim0 are null mutations , whereas ClkAR is a strong hypomorph that is behaviorally arrhythmic but with detectable Per oscillations in peripheral tissues [32] . Other strains included the PDF receptor mutant han5304 [79] , the H99 deficiency ( Bloomington Drosophila Stock Center ( BDSC ) strain #1576 ) and the following Gal4 drivers: C929-Gal4 [80] , MZ840-Gal4 [81] , pdf-Gal4 [35] , R6-Gal4 [82] , TH-D’-Gal4 and TH-D1-Gal4 [43] , tim-Gal4 [83] and UAS strains: UAS-Clk-B19 ( C . Michard-Vanhée , B . Richier and F . Rouyer , personal communication ) , UAS-ClkRNAi ( TriP HMJ02224 , BDSC #42566 ) [84] , UAS-ClkRNAi-R3 ( 7391R-3 strain , National Institute of Genetics , Japan ) , UAS-cycRNAi ( TriP HMJ02219 , BDSC #42563 ) , UAS-hid [85] , UAS-THRNAi ( TriP JF01813 , BDSC #25796 ) , UAS-WakemiR1 [34] . H99 , ClkAR recombinant chromosomes were obtained via standard Drosophila genetics . As they are homozygous lethal , like the H99 parental chromosome , they were identified by assaying the rhythmicity of progeny from crosses between H99 ClkAR / TM3 ( Sb ) and ClkAR flies . Locomotor activity experiments were performed using commercial activity monitors ( TriKinetics ) placed in incubators equipped with standard white , fluorescent low-energy tubes . Young ( 10 day-old male flies ) or older ( 30 day-old males ) males were maintained 5–6 days under LD 12:12 ( Light-Dark 12h:12h ) , and then switched to at least 5–6 days of constant darkness , all on 5% sucrose-agar medium at 25°C . Data analysis was performed with the FaasX software , as described previously [86] . For constant darkness experiments , analysis started the second day of constant darkness . Histograms represent the distribution of the activity through 24 h in 30 min bins , averaged for n flies over 4–5 cycles . All behavioral experiments were repeated 2–3 times to verify reproductibility . 3–4 days old male flies were placed individually in Trikinetics glass tubes , video monitored under infra-red illumination during 3 days in LD , then placed for at least 5 days in LL . The images were processed using pySolo video software [87] to determine the distance travelled ( in pixels ) for each minute of the day . Sleep was defined as 5 min of more of immobility [88] . At least 40 flies in 2–3 replicates were analyzed for each genotype . Male flies were maintained on standard medium at 25°C and under either LD 12:12 , or constant light from adult day 1 until death . 50 animals/bottle in triplicate were tested for each genotype in a given experiment , and each experiment was performed at least twice . Male flies were aged under the same conditions as for the lifespan assays . SING assays were performed as described previously [27 , 64] . Groups of 10 flies were placed in a vertical column ( 25 cm long , 1 . 5 cm diameter ) with a conic bottom end and left for about 20–25 min for habituation . Then , for each genotype , 5 columns were tested individually by gently tapping down the flies ( startle ) , which normally respond by climbing up . Each fly group was assayed three times at 15 min intervals . Results are the mean and SEM of the scores obtained with the 5 independent groups of flies per genotype . The performance index ( PI ) for each column was calculated as follows: ½[1 + ( ntop-nbot ) /ntot] , where ntot is the total number of flies in the column , ntop is the number of flies that have reached at least once the top of the column ( above 22 cm ) during a 1 min interval , and nbot is the number of flies that never left the bottom ( below 4 cm ) . ARLI was monitored as described previously by testing SING performance weekly over 6 weeks , starting on day 10 after eclosion [27] . Dead flies were replaced by substitutes of the same age . Experiments were repeated 2 to 3 times at different periods of the year . Immunostaining was performed essentially as previously described [26] . Adult flies of the desired ages were briefly washed in 70% ethanol before brain dissection in ice-cold Drosophila Ringer Ca2+ free solution . Brains were then fixed for 1h with shaking at room temperature in fixative containing 4% paraformaldehyde in PBS . Brains were then washed 3 x 20 min with PBS and incubated overnight in PBS/0 . 5% Triton X-100 + 2% BSA , at 4°C . Staining with primary antibody was carried out overnight at 4°C in PBS/0 . 5% Triton X-100 + 2% BSA . Brain were then washed 3 x 20 min with PBS/0 . 5% Triton X-100 , and incubated for 2h at room temperature with secondary antibody diluted in PBS/0 . 5% Triton X-100 + 2% BSA . Finally , 3 x 20-min washes were performed using PBS . Primary antibodies used included: mouse monoclonal anti-TH ( Immunostar , 1:1000 ) , rabbit polyclonal anti-PDF ( gift from F . Rouyer's lab , 1:1000 ) , rabbit polyclonal anti-5-HT ( Sigma , 1:1000 ) . Secondary antibodies included: anti-mouse or anti-rabbit conjugated to Alexa Fluor 488 or 555 ( Invitrogen Molecular Probes , 1:1000 ) . In situ ROS detection was performed using a dihydroethidium ( DHE ) dye ( Life technologies ) following a previously described protocol [89] we adapted to whole-mount Drosophila brains . Briefly , male flies were aged under the same conditions as for lifespan assays and dissected in Schneider’s insect medium . The brains were then incubated with DHE for 5 minutes in the dark , fixed for 5 minutes in 7% formaldehyde in 1 X PBS , and immediately imaged on a confocal microscope , as indicated below . Relative ROS levels were measured by quantification of the dye fluorescence using the Fiji software [90] . Fly brains were mounted on slides using as antifade reagent either Prolong Gold ( Life Technologies ) , for brain immunostaining , or Vectashield ( Vector Laboratories ) , for ROS measurements . Brains were visualized and images acquired with a Nikon A1R confocal microscope . A minimum of 10–15 brains was scored over at least 2 trials . Laser , filter and gain settings remained constant within each experiment , and channels were scanned sequentially . Confocal Z-stacks were analyzed using ImageJ software: dopaminergic cells were counted for each cluster , and whole brain average intensity levels were measured for ROS detection . For lifespan assays , survival curves were generated and compared using the log-rank ( Mantel-Cox ) test , and 150 animals were tested per genotype , with each experiment performed 2–3 times . For SING assays and fluorescence quantification , the mean and SEM were calculated for each trial and two-way Anova with post-hoc Tukey comparisons was used . Staining data were analyzed by one-way Anova and Tukey’s pairwise comparisons . Statistical analysis of sleep was performed using the Kruskal-Wallis test . GraphPad Prism 6 was used for all statistical analyses . Significant values in all figures: *: p<0 . 05 , **: p<0 . 01 , ***: p<0 . 001 . | Circadian clocks are highly conserved from flies to humans . They control rhythms in most physiological functions , with free-running periods close to 24 h . Clock disruption , as occurs in shift-work or jet lag , is increasingly suspected to reduce healthspan . Aging and neurodegenerative disorders , like Parkinson’s disease , often disrupt biological clocks early on . It is thus important to understand , and eventually block , a vicious circle that could contribute to accelerated aging . Here we studied the effects of mutations in three Drosophila core circadian genes ( timeless , cycle and Clock ) on lifespan and age-related locomotor impairment , an almost universal hallmark of animal and human senescence . We found that expression of Clock in the main circadian pacemaker neurons is necessary to avoid premature locomotor aging , independently of its circadian function . Clock deficiency in circadian neurons disrupted specific dopaminergic neurons in aging mutant flies , with the involvement of pro-apoptotic pathways , while inhibiting dopamine synthesis in a single pair of these neurons also led to early locomotor decline . This reveals an unexpected link between circadian and dopaminergic circuits in the fly brain , which may be of broad significance to unravel the reciprocal influence between two prominent pathological features of Parkinson's disease: circadian system dysfunction and dopaminergic neurodegeneration . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"sleep",
"dopaminergics",
"neuroscience",
"animals",
"biomechanics",
"biological",
"locomotion",
"animal",
"models",
"physiological",
"processes",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"chronobiology",
"drosophila",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"neurochemicals",
"insects",
"arthropoda",
"biochemistry",
"circadian",
"rhythms",
"cellular",
"neuroscience",
"cell",
"biology",
"phenotypes",
"physiology",
"neurons",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] | 2017 | Drosophila Clock Is Required in Brain Pacemaker Neurons to Prevent Premature Locomotor Aging Independently of Its Circadian Function |
The benefits of ever-growing numbers of sequenced eukaryotic genomes will not be fully realized until we learn to decipher vast stretches of noncoding DNA , largely composed of transposable elements . Transposable elements persist through self-replication , but some genes once encoded by transposable elements have , through a process called molecular domestication , evolved new functions that increase fitness . Although they have conferred numerous adaptations , the number of such domesticated transposable element genes remains unknown , so their evolutionary and functional impact cannot be fully assessed . Systematic searches that exploit genomic signatures of natural selection have been employed to identify potential domesticated genes , but their predictions have yet to be experimentally verified . To this end , we investigated a family of domesticated genes called MUSTANG ( MUG ) , identified in a previous bioinformatic search of plant genomes . We show that MUG genes are functional . Mutants of Arabidopsis thaliana MUG genes yield phenotypes with severely reduced plant fitness through decreased plant size , delayed flowering , abnormal development of floral organs , and markedly reduced fertility . MUG genes are present in all flowering plants , but not in any non-flowering plant lineages , such as gymnosperms , suggesting that the molecular domestication of MUG may have been an integral part of early angiosperm evolution . This study shows that systematic searches can be successful at identifying functional genetic elements in noncoding regions and demonstrates how to combine systematic searches with reverse genetics in a fruitful way to decipher eukaryotic genomes .
Recent technological advances have enabled the sequencing of hundreds of eukaryotic genomes , emphasizing that protein-coding genes constitute only a small fraction of the DNA [1] , [2] . Since then , attention has increasingly shifted to deciphering other parts of the genome , the so-called non-coding regions that are largely composed of transposable elements ( TEs ) . Unlike canonical genes , TEs can persist without benefiting fitness by replicating through transposition within the genome [3] , [4] . There are two major classes of TEs , which transpose by fundamentally different mechanisms: retrotransposons , by reverse transcription of an RNA intermediate; and DNA transposons , by cut and paste transposition . TEs in each active family encode the proteins they need for transposition , which differ by replication strategy and superfamily [5] . Although traditionally viewed as selfish [3] , [4] , evidence continues to mount that TEs affect how genomes evolve in a variety of beneficial ways . One example is molecular domestication , where TEs are repurposed into new genes or other sequences with novel functions [6]–[10] . TE genes are adapted for transposition-related functions , yet they have molecular properties , such as DNA binding and protein-protein interaction , that can be used as raw genetic material and co-opted to perform new functions . Domesticated TEs ( DTEs ) are different from TEs and similar to canonical genes both in terms of their structure and activity , and in that they are subject to phenotypic selection . Most known DTEs have co-linear orthologs and have lost transposition-related features such as flanking terminal repeats and transposase catalytic activity [8] , [10] . However , other features of the original TE that contribute to the beneficial function are maintained , such as conserved domains . DTEs perform various beneficial functions; for example , many DTEs derived from DNA transposons are transcription factors [7] , while others are involved in centromere binding , chromosome segregation , meiotic recombination , heterochromatin formation , TE silencing , programmed genome rearrangement , V ( D ) J recombination , genome stability , and translational regulation [8] , [11] . Molecular domestication has helped to spur remarkable evolutionary innovations , including the mammalian placenta and the vertebrate adaptive immune system [8] . To date , most DTEs have been discovered fortuitously by forward genetics . In plants for example , the FHY3 ( FAR-RED ELONGATED HYPOCOTYL 3 ) family and DAYSLEEPER are the only well-characterized DTEs . FHY3 and FAR1 ( FAR-RED IMPAIRED RESPONSE 1 ) , two members of the FHY3 family , were identified in screens for far-red light mutants [12]–[14] . DAYSLEEPER , which is essential to plant development , was identified in a yeast one-hybrid screen [15] . Even if only a small fraction of the tens of thousands of TE-like genes in plant genome are DTEs rather than TEs , the total number of DTEs may be much higher than currently reported , suggesting that many more may await discovery and that traditional genetic methods may be insufficient to find them , for example due to functional redundancy . These limitations can be overcome by direct bioinformatic searches of genomic data in which DTE genes are discriminated from TEs using genomic signatures that result from differences between how TEs and DTEs function and evolve , such as differences in expression , microsynteny , evolutionary rate , phylogeny , repetitiveness , and TE termini [2] , [16]–[21] . But while theoretically sound , bioinformatics-based searches for DTE genes have not yet been confirmed experimentally . To assess the validity of this approach , we investigated a family of DTEs , MUSTANG ( MUG ) , identified in a previous bioinformatic search of plant genomes [17] , using a reverse genetic approach . MUG sequences are similar to ancestral TEs called Mutator-like elements ( MULEs ) , but unlike MULEs , MUG genes lack signature terminal sequences , are collinear in monocots and eudicots , are functionally constrained , and are differentially expressed [17] , [22] , [23] . Here , we show that MUG genes have been conserved throughout angiosperm evolution and experimentally validate that they are functional in Arabidopsis thaliana by showing that they are essential to flower development and plant fitness .
To investigate the size and distribution of the MUG gene family across a wide range of taxa , we searched for complete sets of MUG paralogs in nine angiosperm species with high quality full-genome sequencing projects at an advanced stage or completed ( Figure S1 ) . Phylogenetic analysis revealed that MUG consists of two major subfamilies , MUGA and MUGB , both present in all nine species , but with different patterns of lineage-specific diversification ( Figure 1 , Figure S2 ) . We found a minimum of four and a maximum of eight MUG genes per species , including the eight previously identified in A . thaliana [17] , which we call At-MUG1 to At-MUG8 . There may be additional MUG genes in unassembled parts of the genomes that we could not identify . These results are consistent with previous analyses of sugarcane ( Saccarum officinarum and Saccarum spontaneum ) [19] , [24] , which found 5 MUGA and 10 MUGB genes , referred to as “Class III” and “Class IV” , respectively , and of grapevine ( Vitis vinifera ) [21] , which found 5 MUGA ( we found only 4 ) and 3 MUGB genes , referred to as MUGvine1–5 and MUGvine6–8 , respectively . Consistent with previous reports showing that MUG genes were present prior to the monocot-eudicot split [17] , [19] , [24] , we found that the MUGA subfamily diversified prior to the divergence of monocots and eudicots into three extant clades , A1 to A3 . Clade A1 has at least one member in each examined species and includes At-MUG1 , At-MUG2 , and At-MUG3 , the latter two belonging to a Brassicales-specific subclade . Clade A2 , which includes At-MUG4 , has exactly one copy in every species , except Zea mays , in which it appears to have a recent duplication . Clade A3 differs from A1 and A2 . Although it was the first of these three clades to diverge , apparently prior to the monocot-eudicot split , it has no member in the examined monocot species , nor is there a clade A3 member in A . thaliana . Conversely , the diversification of the MUGB subfamily did not occur until after the monocot-eudicot split , into two monocot-specific clades ( B-m1 and B-m2 ) and three eudicot-specific clades ( B-e1 to B-e3 ) . Each examined monocot species has one or two members of each monocot-specific clade . The branches leading to these clades are long , suggesting that B-m1 and B-m2 may have diverged early in monocot evolution . Similarly , all examined eudicots have one or two members in the eudicot-specific clade B-e1 , which includes At-MUG5 and At-MUG6 . Clades B-e2 , which includes At-MUG7 , and B-e3 , which includes At-MUG8 , each consist of exactly one member from each examined eudicot species , except Medicago truncatula , which has no member in either B-e2 or B-e3 . Clade B-e3 is the most divergent MUG clade . Together , these results suggest that most MUG clades are conserved within lineages , with MUGA clades encompassing both monocots and eudicots and MUGB clades specific to monocots or eudicots . To further investigate the origin of the MUG family and its distribution among plant taxa , we searched expressed sequence tag ( EST ) libraries deposited at the Ancestral Angiosperm Genome Project ( AAGP; http://ancangio . uga . edu ) and the National Center for Biotechnology Information ( NCBI ) dbEST [25] . We identified multiple ESTs similar to A . thaliana MUG genes in all five ancestral angiosperm species represented in the AAGP ( Table S1 ) , as well as in every order of angiosperm in dbEST with a sufficiently large library ( 5 , 000 ESTs or more ) , except Acorales , Alismatales , and Liliales ( Table S2 ) . No putative MUG ESTs were found in any gymnosperm in the AAGP or dbEST , nor in any other taxa outside angiosperms , although putative MULEs were found in all taxa . Consistent with these EST results , the sequences most similar to MUG in the genomes of Selaginella moellendorffii ( a primitive vascular plant ) and Physcomitrella patens ( a moss ) are repetitive and contain premature stop codons , characteristics indicative of TEs . Phylogenetic analyses of aligned selections of ESTs and genomic sequences confirmed these results ( data not shown ) . All MUG genes contain three conserved domains , an N-terminal MuDR DNA-binding domain ( Pfam PF03108 ) [26] , a core MULE transposase domain ( PF10551 ) , and a C-terminal SWIM zinc-finger domain ( PF04434 ) ( Figure 1B , Figure S2 ) . The same domain architecture is found in diverse transposases of the MULE superfamily [27] . MUGA and MUGB members encode all three domains in a single exon . In addition , MUGB genes have an additional short 5′ exon encoding the Phox and Bem1p ( PB1 ) domain ( PF00564 ) . MUGA and MUGB subfamilies have synonymous nucleotide substitution rates ( dN/dS ) of 0 . 12 and 0 . 10 , respectively , which are significantly less than one ( p = 0 . 0045 and p = 0 . 0085 , respectively ) , showing that they are under purifying selection . To test for MUG functionality in A . thaliana , we used a reverse genetics approach . Two independently derived mutant alleles with T-DNA insertions into the coding regions were obtained from the Arabidopsis Biological Research Center , Ohio State University . When grown under standard conditions , homozygous single mutants show no obvious differences from wild-type ( data not shown ) . This suggests that MUG genes may be functionally redundant , as is often observed in plant gene families . To test this , we crossed single mutants within each subfamily to obtain homozygous double mutants . The mug1 mug2 ( MUGA subfamily ) and mug7 mug8 ( MUGB subfamily ) double mutants exhibit strong phenotypes , consistent with the hypothesis that MUG1 has some degree of functional redundancy with MUG2 , and MUG7 with MUG8 . To characterize the double mutant phenotypes , we measured a number of traits that reflect plant fitness at various life stages ( Table S3 ) . We compared two different allelic combinations each of mug1 mug2 and mug7 mug8 mutants to wild-type A . thaliana plants , ecotype Col-0 ( Figure S3 ) . When grown under standard laboratory conditions , the double mutants differ dramatically from wild-type at all developmental stages from germination to senescence . They exhibit phenotypes that include reduced plant size , an increased incidence of aborted seeds , reduced seed amount , and delays in developmental timing and flowering ( Figure 2A ) . Although the two double mutants have defects in several similar traits , the degree of severity for some traits is stronger in mug7 mug8 than in mug1 mug2 . mug1 mug2 yields 36% of the wild-type seed set , whereas mug7 mug8 yields only 7% . mug1 mug2 stems are more than 2 times shorter than wild-type , whereas those of mug7 mug8 are more than 27 times shorter . Interestingly , mug1 mug2 has a ratio of stem height to rosette diameter ( r = 3 . 44 ) similar to wild-type ( r = 3 . 89 ) , whereas mug7 mug8 has a much-reduced ratio ( r = 0 . 37 ) ( Figure 2B ) . Application of exogenous gibberellic acid ( GA ) failed to rescue reduced stem height or delayed flowering phenotypes ( data not shown ) . The growth defect seems to be restricted to light conditions , since both double mutants show hypocotyls that etiolate normally ( data not shown ) . Some traits are unique to each double mutant . mug7 mug8 leaves are slightly curly , whereas mug1 mug2 leaves have a pale yellow-green coloration , which is most evident in seedlings but remains visible in mature plants ( Figure 2B , 2C ) . mug1 mug2 seedlings contain only about half the chlorophyll per unit mass of wild-type seedlings ( Figure 2D ) . They also fail to develop beyond the cotyledon stage in the absence of exogenous sucrose , and even with exogenous sucrose their growth is restricted ( Figure 2E ) . Each double mutant has defects in floral development and organ morphology that result in reduced fertility ( Figure 3 ) . The flowers are smaller than wild-type . Floral abnormalities are most dramatic at anthesis , especially in mug7 mug8 ( Figure 3 ) . The gynoecium becomes highly elongated relative to the anther filaments ( Figure 3C ) . Scanning electron microscopy ( SEM ) showed that the surface cell layers of certain flower organs , including the stigmatic hair-bearing region and the anthers , are deflated and have abnormal shapes , giving the flowers a shriveled appearance ( Figure 3F ) . These observations suggest that mug7 mug8 flowers may undergo premature senescence . Anthers are bilocular structures that normally produce and hold pollen grains , and upon maturity dehisce longitudinally to release the pollen . Most mug7 mug8 anthers are flat and contain no pollen , and even when they do contain pollen , the locules usually fail to furrow and dehisce , preventing the pollen from being released ( Figure 3L ) . The female organs of mug7 mug8 are also defective , since applying wild-type pollen onto the mutant stigma did not rescue the fertility . mug1 mug2 has less severe defects ( Figure 3 ) . The flowers are not shriveled and both the female and male organs are functional and capable of participating in self-fertilization . However , the anthers do exhibit restricted dehiscence , with only one of the two furrows opening successfully ( Figure 3K ) . The siliques produced by both double mutants have a smaller maximum size than wild-type , restricting the number of seeds they bear , and they have a high incidence of undeveloped ovules , attenuating the seed yield ( Figure 2A , 2F ) .
In silico screens like that in which the MUG gene family was identified [17] have the potential to detect many novel DTEs , but to be convincing their predictions must be validated experimentally . To do so , we exploited a fundamental difference between TE genes and canonical genes . Although TEs can produce phenotypes , for instance by inserting into and disabling a canonical gene , they are not known to themselves encode beneficial functions [11] . Therefore , knocking out a TE gene should have either no effect on fitness or should increase it , whereas knocking out a canonical gene , such as a DTE , may reduce the fitness of the organism . We utilized this difference by examining traits closely tied to fitness in mug T-DNA insertion mutants . Fitness is a measure of how much a genotype contributes to the next generation in a given environment and includes components of survival and reproduction [28] . Both of these aspects are compromised in mug mutants , and certain phenotypes are especially striking . First , the survival of mug mutants is compromised by defects in their physiology ( Figure 2 , Figure 3 ) . mug7 mug8 plants are severe dwarfs , which may disadvantage them in various environments , such as where space or light is limited . The size of mug1 mug2 plants is less reduced , but they have other serious defects . They contain only approximately half of the wild-type amount of chlorophyll per unit mass and their roots barely elongate in the absence of exogenous sucrose ( Figure 2C–2E ) . Sucrose , the major product of photosynthesis , plays a key role in sugar signaling pathways and is required to supply metabolic energy to the roots [29] . Reduced chlorophyll concentrations could impair chloroplast activity and sucrose production , so may be one explanation of why mug1 mug2 plants require exogenous sucrose , without which they would be unlikely to survive to maturity in wild environments . Second , the reproduction of mug mutants is compromised by defects in the floral organ development , fecundity , and reproductive timing ( Figure 2A , Figure 3 ) . These defects are particularly severe in mug7 mug8 mutants . They produce little or no pollen , their stamens and pistil do not elongate normally , preventing contact and limiting self-pollination , and their gynoecium is defective , limiting fertilization even if pollen does adhere to the stigmatic papillae ( Figure 3 ) . This combination of defects renders mug7 mug8 mutants nearly sterile , with an average seed yield of only 7% of wild-type , even under optimal laboratory conditions . The observed phenotypes may result from defects in a variety of physiological processes , so the function of MUG is not yet clear; however , patterns of conservation and expression do provide a few clues . Multiple lines of evidence , in addition to the mutant phenotypes , show that it is highly unlikely that MUG genes function in transposition . They lack the TE termini required for mobilization and are collinear in multiple genomes [17]; Most lack intact DDE motifs , which are required to catalyze transposition ( Figure S2 , Table S4 ) [27] , [30]; Searches of publicly available data show that , unlike TEs [31] , they are not targeted by small RNAs or silenced by DNA methylation [32]–[34] , but are instead expressed in diverse tissues in A . thaliana , sugarcane [19] , [24] , [35] , rice [36] , and other angiosperms ( Table S1 , Table S2 ) . The only other known MULE-like DTEs , the FHY3 family , are transcription factors [30] , [37] , [38] , a common function among domesticated DNA transposons [8] , [11] . Like FHY3 , the WRKY-GCM1 , MULE , and SWIM domains of MUG genes are highly conserved , including key active site residues ( Figure S2 ) , suggesting that they may also have a function involving DNA binding , such as transcription regulation . The severe defects and reduced fitness associated with mutations to MUG genes may explain why they are well conserved . They appear to be ubiquitous among all angiosperms , including basal angiosperms , but are absent from non-angiosperms ( Table S1 , S2 ) , suggesting that MUG was domesticated during early angiosperm evolution to perform what evolved to become a key function . Consistent with previous studies [17] , [19] , [24] , it is clear from their phylogeny that the two MUG subfamilies diverged early in angiosperm evolution , prior to the monocot-eudicot split ( Figure 1 ) . However , phylogenetic and phenotypic differences between MUGA and MUGB suggest that they may have originated from more than one domestication event , similar to the FHY3 family [30] . If all MUG genes did descend from a single domestication event , then while MUGA must have diversified prior to the monocot-eudicot split , MUGB must have acquired a PB1 exon and evolved rapidly prior to the monocot-eudicot split , yet not diversified until after the split . Alternately , if the MUGB subfamily descended from a different domestication event than MUGA , it may have acquired the PB1 domain either through transduplication prior to domestication [39] , gene fusion during domestication [40] , or exon shuffling subsequent to domestication . There may even have been more than two domestication events; in particular , Clade A3 has an unusual phylogenetic pattern , possibly reflecting a separate origin . Although multiple domestication events may help explain differences between the evolution and phenotypes of the two subfamilies , additional evidence , such as from phylogenetic studies of closely related MULEs or the functional characterization of additional MUG genes , will be required to resolve this question . In summary , our results suggest the MUG family originated from TE genes adopting an adaptive function early in flowering plant evolution and are now conserved among angiosperms . Serious defects in mug mutants show that these genes make important contributions to fitness through roles in plant growth , flower development , and reproduction . The approach we used , of evaluating the fitness consequences of mutations to predicted DTEs , effectively couples in silico searches of genomic data with experimental validation . In the future , we expect that similar studies will enable a more complete characterization of TE-derived sequences that have been co-opted to provide fitness benefits .
To confirm the protein sequences of the eight previously identified A . thaliana MUG genes [17] , the sequences of At-MUG1 and At-MUG3 through At-MUG7 were determined from gene models supported by publicly available full-length cDNAs at The Arabidopsis Information Resource ( TAIR ) database , release 10 [41]: At-MUG1 , TAIR AT3G04605 , GenBank AY074390; At-MUG3 , AT1G06740 , AK221278; At-MUG4 , AT5G16505 , AY059842; At-MUG5 , AT3G06940 , AF462806; At-MUG6 , AT5G48965 , AY136382; At-MUG7 , AT3G05850 , BT008628 . At-MUG2 ( AT2G30640 ) and At-MUG8 ( AT5G34853 ) have no publicly available full-length cDNA sequence , so we predicted their structure from genomic DNA using FGENESH ( dicot setting ) [42] . The predicted sequence of At-MUG2 was confirmed by RT-PCR ( data not shown ) and both At-MUG2 and At-MUG8 are consistent with available EST and mRNA-seq data [32] , [41] . To identify additional MUG homologs , we performed comprehensive genomic searches in nine angiosperm genomes: A . thaliana , Carica papaya , V . vinifera , M . truncatula , Mimulus guttatus , Sorghum bicolor , Z . mays , Oryza sativa var . japonica , and Brachypodium distachyon . Because the identity between MUGA and MUGB sequences is low , a representative amino acid query was chosen for each subfamily: At-MUG1 for MUGA , At-MUG7 for MUGB . In A . thaliana , O . sativa , and S . bicolor , we used BLASTP to search the protein databases of TAIR10 [41] , the Rice Annotation Project ( RAP-DB ) [43] , and Sbi 1 . 4 [44] , respectively . To ensure that we found all MUG genes present in the datasets , we calibrated E-value thresholds to well below those needed to find all known MUG genes [17] as well as a few non-MUG sequences , and we validated the results using phylogenetic analyses . At each resulting locus , the highest-ranked annotated gene model was chosen . Conserved domains were identified using the NCBI Conserved Domain Database [45] . Gene models terminating in truncated domains were extended using FGENESH [42] where possible . Because TE-like genes ( including DTEs ) are commonly filtered from protein databases , we confirmed the results in A . thaliana using a TBLASTN [46] search of the TAIR10 Genes database , which includes loci annotated as “transposable element genes” . In S . bicolor , we confirmed the results using a TBLASTN [46] search of the unmasked genome assembly [44] . The remaining six genomes had limited gene models , so instead of searching protein databases , we searched whole genome assemblies: C . papaya , Hawaii Papaya Genome Project , 2007 release [47]; V . vinifera , Genoscope , March 2010 , 12X assembly [48]; M . truncatula , Medicago Genome Sequencing Consortium , release Mt3 . 5 [49]; M . guttatus , Joint Genome Institute ( JGI ) assembly v1 . 0 and gene annotation v1 . 1 [50]; Z . mays , Maizesequence . org , release 4a . 53 [51]; B . distachyon , JGI 8x assembly release v1 . 0 of strain Bd21 [50] , [52] . We used TBLASTN [46] to identify potential MUG homologs and FGENESH [42] to predict corresponding gene models . Conserved domains were identified as above . We conducted phylogenetic analyses on the amino acid sequences identified in the genome searches , including all putative MUG homologs as well as a sample of putative MURA sequences . Putative MURAs typically formed larger , low-identity clusters and the sequences often contained premature stop codons or frameshifts , even within conserved domains , which is characteristic of TEs . Clusters of putative MURA sequences with greater than 95% identity were represented by a single sequence . Alignments were generated using MUSCLE [53] , curated with Gblocks [54] , and phylogenetic analysis performed using PhyML ( 500 bootstraps ) [55] , [56] . All putative MURA sequences were non-monophyletic to the MUG tree . Z . mays MURA ( GenBank AAA81535 . 1 ) was used as the outgroup . Synonymous substitution rates were calculated using PAML CODEML [57] . To determine the taxonomic distribution of MUG homologs , we conducted TBLASTN searches of NCBI dbEST libraries of each major seed plant subgroup not represented in the genome searches , using At-MUG2 and At-MUG7 queries as above ( Table S2 ) . We identified putative MUG ESTs by looking in each species for small clusters of top-ranked ESTs with low E-values and high identity , consistent with results of previously identified MUG homologs . We validated MUG and MURA assignments by phylogenetic analyses , using subsets of ESTs that could be aligned . The mutants mug1-1 ( GK_514B01 ) , mug1-2 ( GK_293B02 ) , mug2-3 ( SALK_090878 ) , mug2-4 ( SALK_055071 ) , mug7-1 ( SALK_012814 ) , mug7-5 ( GK_378C04 ) , mug8-1 ( GK_244B09 ) , and mug8-2 ( GK_155E09 ) were obtained from GABI-Kat ( http://www . gabi-kat . de ) [58] and SALK ( http://www . arabidopsis . org/abrc ) [59] T-DNA insertion populations . The positions of the insertion sites in double mutants used in the phenotypic analyses were confirmed by sequencing of the allele-specific PCR products ( data not shown ) . Wild-type ecotype Col-0 seeds were originally obtained from Lehle Seeds ( www . arabidopsis . com ) . Seeds were sterilized using a 50% bleach solution for 5 min , then washed once in 95% ethanol and 3 times in sterile water . Sterilized seeds were sown on half-strength Murashige and Skoog ( MS ) [60] medium plates containing 0 . 8% agar ( w/v ) and 1% sucrose ( w/v ) . Plated seeds were stratified in the dark for 3 days at 4°C and kept on plates for 2 weeks in a growth chamber ( Conviron model E15 ) at 22°C under a 16 h light/8 h dark photoperiod , ∼100 µmol quanta/m2/s light intensity , 60% relative humidity . Seedlings were transplanted to soil with a composition of PRO-MIX ( Premier Tech Horticulture , Quebec , Canada ) : vermiculite: perlite of 2∶1∶1 in 2 ½ inch square pots and returned to the growth chamber . We selected eight traits for phenotypic analyses previously shown to reflect plant fitness [61]–[65] . A detailed description of each trait is presented in Table S3 . Statistical analysis was performed by two-sample t-tests using the wild-type control ( α = 0 . 05 ) . Scanning electron microscopy was performed using a Hitachi S4700 Field Emission-STEM microscope . Wild-type and homozygous mutant inflorescences were fixed overnight in 2% glutaraldehyde , washed , and dehydrated using a series of graded ethanol solutions ( 30 to 100% ) . Dried samples were sputter coated with gold-platinum . Inflorescences were also photographed using an Olympus DP71 camera attached to an Olympus MVX10 stereomicroscope . For the sucrose assay , Col-0 and mutant seeds were sown on MS medium plates with or without 100 mM sucrose . Seeds were stratified for 3 days in the dark at 4°C and transferred to a growth chamber with settings as above for 12 days with plates vertically-oriented , after which each plate was photographed using a Nikon D3100 camera and scored manually . To measure total chlorophyll content , 200 mg of 17-day-old wild-type and mutant seedlings were extracted by shaking overnight in the dark in 1 ml of 80% acetone . Chlorophyll levels were measured using a 4050 Ultrospec II UV/Vis spectrophotometer ( LKB Biochrom ) and the total amount of chlorophyll was determined using MacKinney's coefficients [66] and the equation: chlorophyll a+b = 7 . 15×OD660 nm+18 . 71×OD647 nm . | The genomes of complex organisms are mostly made up not of ordinary genes but of transposable elements . Transposable elements have been called “selfish DNA” because they normally persist by copying themselves , not by helping the organism to survive or reproduce . Yet transposable elements can help organisms to evolve; for instance , transposable element genes sometimes acquire new functions that do benefit the organism . Because they are difficult to distinguish from transposable elements , little is known about these “domesticated genes . ” Although studies have attempted to identify them computationally , the predictions have not been verified experimentally . Here , we examine some of the first domesticated genes to be predicted computationally , the MUSTANG family of plant genes . We show that the predictions were correct: MUSTANGs are , like ordinary genes , functional . MUSTANG mutations result in serious defects in how plants grow , flower , and reproduce . Since they are present only in flowering plants , MUSTANG probably originated when flowers first evolved , perhaps taking on a key role . This study is important both because it shows that MUSTANG is critical to plant fitness and because , in the future , a similar approach can be used to find additional domesticated genes and to better understand how transposable elements contribute to evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2012 | A Gene Family Derived from Transposable Elements during Early Angiosperm Evolution Has Reproductive Fitness Benefits in Arabidopsis thaliana |
All positive-strand RNA viruses induce the biogenesis of cytoplasmic membrane-bound virus factories for viral genome multiplication . We have previously demonstrated that upon plant potyvirus infection , the potyviral 6K2 integral membrane protein induces the formation of ER-derived replication vesicles that subsequently target chloroplasts for robust genome replication . Here , we report that following the trafficking of the Turnip mosaic potyvirus ( TuMV ) 6K2 vesicles to chloroplasts , 6K2 vesicles accumulate at the chloroplasts to form chloroplast-bound elongated tubular structures followed by chloroplast aggregation . A functional actomyosin motility system is required for this process . As vesicle trafficking and fusion in planta are facilitated by a superfamily of proteins known as SNAREs ( soluble N-ethylmaleimide-sensitive-factor attachment protein receptors ) , we screened ER-localized SNARES or SNARE-like proteins for their possible involvement in TuMV infection . We identified Syp71 and Vap27-1 that colocalize with the chloroplast-bound 6K2 complex . Knockdown of their expression using a Tobacco rattle virus ( TRV ) -based virus-induced gene silencing vector showed that Syp71 but not Vap27-1 is essential for TuMV infection . In Syp71-downregulated plant cells , the formation of 6K2-induced chloroplast-bound elongated tubular structures and chloroplast aggregates is inhibited and virus accumulation is significantly reduced , but the trafficking of the 6K2 vesicles from the ER to chloroplast is not affected . Taken together , these data suggest that Syp71 is a host factor essential for successful virus infection by mediating the fusion of the virus-induced vesicles with chloroplasts during TuMV infection .
The host endomembrane system directly contributes to the formation of virus-induced membrane-bound virus factory for positive-strand RNA virus replication [1] , [2] . Depending on the virus , membranes from distinct cellular organelles such as the endoplasmic reticulum ( ER ) , chloroplast , mitochondrion , endosome , and peroxisome are recruited to house the virus factory [3] . In the past decade , efforts have been made to elucidate the molecular mechanisms underlying the assembly of the virus factory with cellular membranes . For instance , genome-wide screens for host factors affecting Tomato bushy stunt virus ( TBSV ) replication in yeast , has led to the identification of seven ESCRT ( endosomal sorting complex required for transport ) proteins involved in the assembly of membrane-bound replicase complexes through interaction with the viral integral membrane protein p33 [4]–[6] . The characterization of Arabidopsis mutants defective in Tobacco mosaic virus ( TMV ) infection revealed several host integral membrane proteins , including the Tom1/3 proteins , the putative membrane anchor for the virus factory , and the Tom2 protein , which most likely plays an accessory role in replication [7] , [8] . The host vesicle associated protein ( VAP ) was suggested to be involved in the biogenesis of membrane vesicles induced by the 60K protein of Cowpea mosaic virus ( CPMV ) [9] . The VAP protein interacts with the CPMV 60K protein and colocalizes with the ER-derived vesicles that contain the 60K protein in CPMV-infected plant cells [9] . The second 6 , 000-molecular-weight protein ( designated 6K2 or 6K ) of the family Potyviridae , the largest and most agriculturally important family of plant positive-strand RNA viruses , is an integral membrane protein that induces the formation of 6K2-containing membranous vesicles at ER exit sites [10] , [11] . The 6K2-induced vesicles subsequently target chloroplasts to form chloroplast-6K2 complexes that harbour the virus factory for potyvirus replication [12] . In eukaryotic cells , specific membrane fusion between transport vesicles and target membranes is mediated by the SNARE ( soluble N-ethyl-maleimide-sensitive-factor attachment protein receptors ) complex that assembles into a tight cluster of four coiled-coil helices [13]–[16] . In Arabidopsis thaliana , at least 64 SNARE proteins have been identified and classified into the Qa- , Qb- , Qc- , and R-groups [13] , [17] . Of the ternary SNARE complexes identified in plant cells thus far , the Qa-PEN1/Qb+Qc-SNAP33/R-VAMP721 ( 722 ) complex functions in disease resistance and the Qa-SYP22/Qb-SYP51/Qc-VTI11/R-VAMP727 complex operates in seed development [18] , [19] . Previous reports showed that the ER-localized SNARE molecules contain SYP81/AtUfe1 ( Qa ) , SYP71 , SYP72 , and SYP73 ( Qc ) , AtSec22 and AtVAMP723 ( R ) in Arabidopsis [13] . Furthermore , Arabidopsis has 10 VAPs grouped in the VAP33 subfamily of SNAREs which also localize on the ER membrane [20] . These ER-localized SNARE or SNARE-like molecules might selectively function in mediating ER-associated vesicular trafficking , docking and fusion . Due to the fact that the chloroplast-bound 6K2 vesicles housing the virus factory of potyviruses are of ER origin , it is possible that ER-localized SNARE or SNARE-like proteins play a role in the formation of the 6K2–chloroplast complex . In the present study , we show that following the trafficking of 6K2 to chloroplasts , 6K2 induces the formation of chloroplast aggregates and elongated tubular structures at the junction between adjacent chloroplasts . We demonstrate that two ER-localized SNAREs , i . e . , Qc-Syp71 and Vap27-1 , colocalize with the chloroplast-bound 6K2 structure . We further identify Syp71 as a SNARE essential for virus infection via mediating the formation of 6K2-induced elongated tubular structures and chloroplast clumps in infected plants .
Previous studies showed that the potyviral 6K2 membrane protein induces the formation of ER-derived vesicles [10] and potyviral infection can induce the formation of chloroplast aggregates [21] . Recently , we reported that the ER-derived 6K2-containing vesicles in TuMV-infected leaves translocate from the ER to chloroplasts for potyviral genome replication [11] , [12] . To investigate if 6K2 vesicles are involved in chloroplast aggregations , a recombinant TuMV ( TuMV::6K2-GFP ) that contains a green fluorescent protein ( GFP ) -tagged 6K2 was introduced into Nicotiana benthamiana leaf cells via agroinfiltration . At 48 hrs post infiltration ( hpi ) , 6K2-GFP vesicles were found at the periphery of chloroplasts ( Fig . 1A , frame I ) , consistent with our previous observation [12] . At 72 hpi , chloroplast aggregation occurred in about 50% infected cells and 6K2-GFP formed an elongated tubular structure at the junctions between two adjacent chloroplasts ( Fig . 1A , frame II ) . At 96 hpi , more chloroplast aggregates ( >85% of infected cells ) were observed ( Fig . 1A , frames III–VI ) . Typically , about 20 chloroplasts were tightly apposed to each other in a chain or irregularly grouped , and elongated tubular structures highlighted by 6K2-GFP were found between two adjacent chloroplasts ( Fig . 1A , frame III ) . Occasionally , large chloroplast clumps , consisting of up to 50 chloroplasts , were found ( Fig . 1A , frame IV ) . In such clumps , chloroplasts were irregularly grouped but closely apposed to each other to form a spherical amorphous mass . At higher magnifications , an elongated tubular structure enveloped by 6K2-GFP was evident between adjacent chloroplasts in the aggregates ( Fig . 1A , frame V ) . To determine the distribution of 6K2 vesicles in the absence of viral infection , the yellow fluorescent protein tagged 6K2 ( 6K2-YFP ) was transiently expressed in N . benthamiana leaf cells via agroinfiltration . When expressed alone , at 48 hpi , small ring-like vesicles of 6K2-YFP initially accumulated between two closely located , but physically separate chloroplasts ( Fig . 1B , frame I ) . Intriguingly , these vesicles subsequently coalesced , leading to the formation of a bridge across the outer envelopes of the two adjacent chloroplasts ( Fig . 1B , frame II ) . The vesicle bridge apparently promoted the adhesion of the adjacent chloroplasts , with a strong labelling of 6K2-YFP at the junctions ( Fig . 1B , frames III and IV ) , similar to elongated tubular structures induced by 6K2-GFP during viral infection ( Fig . 1A , frames II–IV ) . After 72 hpi , 6K2-YFP induced chloroplast aggregates ( Fig . 1B , frames V–VIII ) . These data therefore demonstrate that the 6K2 protein is able to induce the formation of elongated tubular structures between adjacent chloroplasts independently of any other TuMV-encoded proteins or viral RNA replication . Actin filaments and myosin XI may play a role in the mobility of chloroplasts through direct interactions with the chloroplast outer envelope membrane [22] , [23] . We observed colocalization between actin filaments stained by mTalin-CFP and the elongated tubular structure induced by 6K2-YFP ( in the absence of viral infection; Fig . 2A , frames I–IV ) or 6K2-GFP ( in the presence of viral infection; Fig . 2A , frames V–VIII ) at the junctions between adjacent chloroplasts in N . benthamiana leaf cells ( Fig . 2A ) . The potential role of myosin XI-K in the formation of elongated tubular structures induced by the 6K2 protein was assayed using the myosin XI-K tail , a dominant negative mutant of myosin XI-K [24] . Overexpression of the myosin XI-K tail dramatically inhibited the formation of the 6K2-induced elongated tubular structures at the junction of adjacent chloroplasts as well as chloroplast aggregates ( Fig . 2B , frames I and IV ) , either in the absence of viral infection ( indicated as 6K2-YFP ) or in the presence of viral infection ( indicated as 6K2-GFP ) . Such inhibition was not seen in the N . benthamiana leaf cells overexpressing the myosin XI-2 tail ( Fig . 2B , frames II and V ) or in the negative control leaf cells ( Fig . 2B , frames III and VI ) . Taken together , these data indicate that the formation of chloroplast-bound 6K2 elongated tubular structures and chloroplast aggregates requires the functional actomyosin motility system . As the initial biogenesis of membranous vesicles of 6K2 occurs at ER exit sites ( ERES ) on the ER membrane [11] , [12] , their translocation may require proteins from the SNARE family that mediates specific membrane fusion between transport vesicles and target membranes . To determine if SNARE proteins are involved in the translocation of the 6K2 vesicles , we first cloned seven putative ER-localized SNAREs or SNARE-like proteins from Arabidopsis including Syp71 , Syp72 , Syp73 , Syp81 , VAM723 , Vap27-1 and Vap27-2 and constructed plant expression vectors for the transient expression of these SNARE proteins tagged with YFP . As expected , expression of each of these proteins alone revealed a typical ER reticulated pattern ( Fig . 3A and data not shown ) . When each of these SNARE-YFP fusion proteins was transiently coexpressed with 6K2-CFP ( cyan fluorescent protein ) , only Syp71-YFP and Vap27-1-YFP preferentially colocalized to 6K2-induced elongated tubular structures at the junctions of adjacent chloroplasts ( Fig . 3C , D , F , G ) . Other SNARES tested such as Syp72 did not colocalize with the 6K2-induced structure ( Fig . 3B ) . To investigate if Syp71 and Vap27-1 are recruited to the 6K2-induced elongated tubular structures during viral infection , Syp71-mRFP ( monomeric red fluorescent protein ) or Vap27-1-mRFP was coinfiltrated into N . benthamiana with the recombinant TuMV ( TuMV::6K2-GFP ) . At 72 hpi , the fluorescence of Syp71-mRFP or Vap27-1-mRFP colocalized with the chloroplast-bound 6K2-GFP elongated tubular structures ( Fig . 3E , H ) . To confirm the association of Syp71 or Vap27-1 with the chloroplast , chloroplasts were purified from N . benthamiana leaves coinfiltrated with the following different combinations: ( 1 ) 6K2-CFP and Syp71-YFP ( or Vap27-1-YFP ) ; ( 2 ) the TuMV::6K2-GFP infectious clone and Syp71-YFP ( or Vap27-1-YFP ) ; and ( 3 ) Syp71-YFP or Vap27-1-YFP alone . Western blotting experiments were carried out to detect Syp71 and Vap27-1 in the purified chloroplasts with antisera against Syp71 or Vap27-1 . A protein corresponding to the predicted size for Syp71-YFP was detected by Syp71 antibodies in the chloroplasts purified from the leaves coexpressing Syp71-YFP and 6K2-CFP or expressing Syp71-YFP in the presence of TuMV infection ( TuMV::6K2-GFP ) ( Fig . 3I , lanes 1 and 2 ) . Similarly , a protein with the predicted size for Vap27-1-YFP was also found in the chloroplasts purified from the leaves coexpressing Vap27-1-YFP and 6K2-CFP or expressing Vap27-1-YFP in the presence of TuMV infection ( TuMV::6K2-GFP ) ( Fig . 3J , lanes 1 and 2 ) . In contrast , no detectable Syp71-YFP or Vap27-1-YFP was evident in chloroplasts purified from the leaves expressing Syp71-YFP or Vap27-1-YFP alone ( Fig . 3I and J , lane 3 ) . The purity of chloroplasts was confirmed by immunoblotting to detect chloroplast protein PsbA and ER marker Bip2 ( Fig . 3I and J ) . These data suggest that Syp71 or Vap27-1 are tightly associated with chloroplasts when coexpressed with 6K2-CFP or in the presence of viral infection with TuMV::6K2-GFP and that the association of Syp71 or Vap27-1 with chloroplasts is mediated by the 6K2 protein . To explore how Syp71 and Vap27-1 are recruited by the 6K2 protein , the DUAL membrane system , a variant of the yeast two-hybrid assay , was used to determine if there are any protein-protein interactions among the three membrane proteins 6K2 , Syp71 and Vap27-1 . Our results showed that 6K2 interacted with Vap27-1 , but not with Syp71 ( Fig . 4A ) . Syp71 was also found to bind to Vap27-1 ( Fig . 4A ) . To further confirm these interactions , 6K2 , Syp71 and Vap27-1 were fused with a His or HA tag and the resulting fusion proteins were transiently expressed in N . benthamiana leaves in different combinations , followed by immunoprecipitation . As shown in Fig . 4B and C , positive interactions were confirmed between 6K2-HA and Vap27-1-His and between Syp71-HA and Vap27-1-His . Only a weak interaction signal was detected between 6K2-HA and Syp71-His , when these proteins were co-expressed with the untagged Vap27-1 ( Fig . 4D ) . These data suggest that Syp71 is not likely directly recruited by 6K2 , and the association of Syp71 with the chloroplast-bound 6K2-induced tubular structures may be via a third protein . To investigate whether 6K2-associated proteins Syp71 and Vap27-1 are required for TuMV infection , virus induced gene silencing ( VIGS ) was used to knock down the expression of Syp71 and Vap27-1 . A cDNA fragment of Syp71 or Vap27-1 was cloned into the RNA2-derived vector of Tobacco rattle virus ( TRV ) . The obtained TRV-Syp71 or TRV-Vap27-1 vectors were co-introduced into Arabidopsis with the TRV1 vector via agro-infiltration [25] . Real-time PCR analysis confirmed that Syp71 or Vap27-1 mRNA levels in the treated plants decreased by 85% compared to the negative control ( plants infected with the TRV1 and TRV2 empty vectors ) at 12 days post agro-infiltration ( Fig . 5C ) . Vap27-1-downregulated Arabidopsis plants did not show detectable developmental defects or other phenotypes in comparison with the control plants . Syp71-downregulated Arabidopsis plants were slightly smaller in stature and the upper new rosette leaves were also slightly smaller . The most drastic effect was that the Syp71-silenced plants produced only a few seeds and these seeds were not viable , suggesting that Syp71 may be important for embryogenesis . New leaves from the non-silenced wild type Arabidopsis plants ( either treated with buffer or pTRV2 empty vector ) , and Syp71- and Vap27-1-downregulated plants were inoculated with TuMV::6K2-GFP . After 10 days post inoculation , typical TuMV-induced symptoms such as mosaic and dwarfism were observed in wild type or pTRV2 empty vector-treated plants ( Fig . 5A ) . Vap27-1-downregulated plants also showed similar symptoms ( Fig . 5B ) . In contrast , Syp71-downregulated plants inoculated with TuMV were slightly shorter in stature than the control plants but did not show other noticeable typical TuMV symptoms ( Fig . 5A ) . Real-time RT-PCR was carried out to quantify TuMV viral RNA . The TuMV viral RNA decreased about 10 times in Syp71-silenced plants when compared to the non-silenced control plants treated with the empty VIGS vector or with the inoculation buffer ( mock ) ( Fig . 5D ) . No significant difference in the concentration of TuMV viral RNA was found between the Vap27-1-downregulated plants and the control plants ( Fig . 5D ) . These data suggest that suppression of Syp71 expression effectively inhibits TuMV infection in Arabidopsis , while down-regulation of Vap27-1 does not . To determine if downregulation of Syp71 or Vap27-1 expression affects the formation of chloroplast-bound 6K2 tubular structures , N . benthamiana plants were treated with the TRV-based VIGS vectors to knock down Syp71 and Vap27-1 expression . The N . benthamiana plants were then inoculated with TuMV::6K2-GFP . Syp71 or Vap27-1 expression was reduced in the VIGS vector treated plants to the level similar to the treated Arabidopsis , and the Syp71-downregulated plants showed resistance to TuMV ( data not shown ) . By confocal laser microscopy , strong GFP signals were evident in the newly developed systemic leaves of the control plants ( mock treated or infiltrated with the TRV empty vector ) and of Vap27-1-downregulated plants but not detectable in those of Syp71-downregulated plants . In the inoculated leaves of the TRV empty vector-infected or Vap27-1-downregulated plants , 6K2-induced tubular structures and chloroplast aggregates occurred in a manner similar to those found in the wild type plants infected by TuMV::6K2-GFP ( Fig . 6A ) . However , in the Syp71-downregulated plants , the formation of 6K2-induced plates and chloroplast aggregates was blocked , though the trafficking of 6K2-GFP to the periphery of chloroplasts was apparently not affected ( Fig . 6A ) . These results suggest that Syp71 but not Vap27-1 is required for the formation of the elongated tubular structures and chloroplast clumps . To further determine if inhibition of TuMV infection in Syp71-donwregulated Arabidopsis plants is due to reduced TuMV replication , we transformed the TuMV::6K2:GFP infectious clone into protoplasts isolated from WT or Syp71-downregulated Arabidopsis leaves and then assayed TuMV accumulation . Virus accumulation was significantly reduced in the Syp71-downregulated protoplasts at three different time points ( Fig . 6B ) , suggesting Syp71 is essential for TuMV accumulation .
In a recent report , we established that plant potyviruses initiate viral genome translation on the ER and induces the formation of the 6K2 vesicles at the ER subdomain that subsequently traffic to chloroplasts to generate the chloroplast-6K2 vesicle complex for potyvirus replication [12] . More recently , Grangeon et al . localized the virus-induced complex containing the 6K2 , chloroplasts as well as ER and Golgi markers into the perinuclear region [26] . In the present study , we observed that TuMV infection caused the formation of chloroplast clumps and elongated tubular structures at the junctions between neighbouring chloroplasts in more than 85% of infected N . benthamiana cells at 96 hpi ( Fig . 1A , frames III–VI ) . This is in agreement with an earlier observation that chloroplast aggregates and elongated tubular structures between adjacent chloroplasts of the clump were present in TuMV-infected Chenopodium quinoa leaves [21] . Intriguingly , we also observed that about 30% of the infected cells showed the virus-induced perinuclear structure at 96 hpi ( data not shown ) as described [26] , all containing the chloroplast aggregates . Since the chloroplast and 6K2 interaction was observed first , followed by chloroplast aggregation and then the perinuclear structure , it seems that the chloroplast-6K2 complex leads to chloroplast aggregation and further to the formation of the perinuclear structure . In this study , we demonstrated that expression of the 6K2 protein alone is sufficient to induce the formation of chloroplast aggregates as well as elongated plates at the junctions between adjacent chloroplasts ( Fig . 1B , frames II–VIII ) . As this phenomenon was not observed when other viral proteins including P1 , HC-Pro , P3 , P3N-PIPO , 6K1 , CI , NIa-VPg , NIa-Pro , NIb and CP were expressed alone , it is reasonable to propose that it is the 6K2 protein that accounts for the generation of chloroplast aggregation and elongated tubular structures during potyviral infection . Infections by other plant positive single-stranded viruses can also induce similar cytopathological effects . For instance , the clumping of the chloroplasts is one of the typical cytological effects of Turnip yellow mosaic virus ( TYMV ) infection [27] . TYMV is a rod-shaped small RNA virus in the genius Tymovirus of the family Tymoviridae . TYMV replication complexes colocalize with virus-induced membrane vesicles that are thought to result from the invagination of the chloroplast envelope [28] . The TYMV 140K protein was shown to target the chloroplast envelope , develop vesicles along the chloroplast peripheries and finally induces the clumping of the chloroplasts [28] . As these vesicles contain TYMV viral replication complexes and viral particles , they have been suggested to be the TYMV replication site [27] . In contrast , TuMV sequentially recruits the ER and chloroplasts for its genome replication [12] . In TuMV:6K2-GFP infected N . benthamiana cells , double strand RNAs , a hallmark of viral genome replication , and the viral replicase components such as viral RNA-dependent RNA polymerase ( also NIb ) colocalize with chloroplast-bound 6K2 vesicles [12] . But no TYMV-type invaginations were found in chloroplasts [12] , suggesting a different role of chloroplasts in TuMV infection . In TuMV-infected Chenopodium quinoa leaves , viral particles were found in the elongated tubular structure between clumped chloroplasts [21] . In the case of other potyviruses , chloroplast-bound elongated tubular structures and vesicles containing viral particles were also evident in Wheat streak mosaic virus-infected wheat leaf tissues [29] . In comparison with the highly dynamic ER , relatively stationary chloroplasts and chloroplast clumps may provide replicating potyviruses with an energy- and resource-rich environment effectively protecting them against host defense responses such as virus induced RNA silencing . Taken together , these data support the notion that the 6K2-induced chloroplast-bound tubular structures are the site for TuMV replication and viral particle assembly . We identified a SNARE protein Syp71 and a SNARE-like protein Vap27-1 that were recruited to the 6K2-induced chloroplast-bound elongated tubular structure ( Fig . 3 C–H ) . As TuMV infection was inhibited only in Syp71-downregulated plants ( Fig . 5A ) but not in Vap27-1-downregulated plants ( Fig . 5B ) , Syp71 rather than Vap27-1 plays an essential role in the infection process . We further determined that inhibition of virus infection was due to reduced virus replication ( Fig . 6B ) , highlighting the essential role of Syp71 in TuMV accumulation . Knockdown of Syp1 expression blocked the formation of elongated tubular structures and chloroplast aggregates as well as the virus-induced perinuclear structure , but did not impair the trafficking of 6K2 vesicles to the outer chloroplast envelope ( Fig . 6A ) , suggesting that Syp71 is involved in mediating the association of 6K2 vesicles with the outer chloroplast envelope and further homotypic fusion of 6K2 vesicles to assemble the elongated tubular structures and to join 6K2-bound chloroplasts . Indeed , SNARE family proteins are known for their essential role in specific membrane fusions between the transport vesicles and target membranes [14] , [15] . Syp71 , together with Syp72 and Syp73 , forms a plant-specific SNARE subfamily named Syp7 [30] . Syp71 localizes to the plasma membrane and the ER in Arabidopsis [30] . As no homozygous T-DNA insertion Arabidopsis Syp71 mutant could be isolated from the Syp71/syp71 heterozygote , Syp71 seems essential for the development of Arabidopsis . To date no specific functions have been identified for the Syp7 subfamily except for a recent report showing that Syp71 is essential for symbiotic nitrogen fixation in Lotus japonicas nodules in a yet-unknown manner [31] . To our best knowledge , the current study represents the first to functionally characterize a SNARE family protein essential for virus infection in plants . To explore how Syp71 is recruited to the 6K2 elongated tubular structure , we carried out a yeast two-hybrid assay followed by coimmunoprecipitation experiments to confirm protein-protein interactions in planta . Unexpectedly , Syp71 did not interact with the 6K2 protein ( Fig . 4A ) , suggesting that the recruitment of Syp71 to the 6K2 elongated tubular structure is via an indirect approach . Since Vap27-1 interacted with both 6K2 and Syp71 ( Fig . 4A ) , it is possible that Vap27-1 may function as a linker between the 6K2 vesicle and Syp71 . This assumption was supported by our coimmunoprecipitation assay showing that 6K2 , Vap27-1 and Syp71 indeed can be coimmunoprecipitated ( Fig . 4D ) . Consistently , it is well known that the SNARE-like VAP protein such as Vap27-1 sharing high homology with the mammalian VAP33 can regulate the trafficking , docking and fusion of vesicles through their interactions with SNARE proteins [18] , [32] , [33] . Previously Vap27-1 was shown to colocalize with membranous vesicles induced by the 60K membrane protein of CPMV and to interact with the CPMV 60K protein , and was proposed to play a role in the induction of membrane vesicles and/or the assembly of the replication complex [9] . In this study , knockdown of Vap27-1 expression did not affect the development of elongated tubular structures and chloroplast aggregates in TuMV-infected cells and did not significantly affect virus accumulation ( Fig . 5B ) . These results suggest an alternative recruitment approach for Syp71 to the TuMV chloroplast-associated vesicles . It is possible that other VAPs or proteins that have overlapping functions with Vap27-1 interact with 6K2 and Syp71 and mediate the association of Syp71 with the 6K2 vesicles . Search for the other components of the Syp71 or 6K2 interactome via diverse proteomic tools including yeast two-hybrid screens with Syp71 or 6K2 as bait will certainly help to elucidate the mechanism underlying the formation of chloroplast-bound 6K2 elongated tubular structures and shed new insights into specific host-virus interactions in the infection process .
Gateway technology ( Invitrogen ) except otherwise stated was used to generate clones reported in this study . RNA extraction from Arabidopsis thaliana or Nicotiana benthamiana leaves was performed using an RNeasy Plant Mini Kit ( Qiagen ) . Reverse transcription was catalyzed by Superscript III reverse transcriptase ( Invitrogen ) with oligonucleotide dT20 . Gene sequences were amplified by PCR using Phusion DNA polymerase ( NEB ) . The resulting DNA fragments were purified and transferred by recombination into the entry vector pDONR201 ( Invitrogen ) using BP clonase II ( Invitrogen ) following the manufacturer's protocol . The insert of the resulting pDONR clone was verified by sequencing . The insert was then transferred by recombination to the indicated binary destination vector using LR clonase II ( Invitrogen ) following the standard conditions and procedure recommended by the supplier . SNAREs proteins Syp71 ( At3g09740 ) , Syp72 ( At3g45280 ) , Syp73 ( At3g61450 ) , Syp81 ( At1g51740 ) , VAM723 ( At2g33110 ) , Vap27-1 ( AY364005 ) and Vap27-2 ( AY364004 . 1 ) coding sequences were amplified from cDNA derived from Arabidopsis RNA and recombined into pDONR201 . The inserts of these resulting intermediate clones were further transferred by recombination into the binary destination vectors pEarleygate101 [34] to generate plasmids Syp71-YFP , Syp72-YFP , Syp73-YFP , Syp81-YFP , VAMP723-YFP , Vap27-1-YFP and Vap27-2-YFP , as well as into the binary destination vectors pEarleygate101 and 103 to generate Syp71-HA and Vap27-1-His , respectively . The recombinant TuMV infectious clones TuMV-GFP ( EF028235 . 1 ) consisting of a full-length TuMV cDNA tagged by GFP and TuMV::6K2-GFP containing an additional copy of 6K2 fused to GFP were described previously [35] . The 6K2 cistron of TuMV was amplified by PCR . Gateway technology with the entry vector pDONR201 and destination vectors pEarleygate101 and pEarleygate102 was again used to produce plasmids 6K2-HA-YFP and 6K2-CFP , respectively . The plasmids including mTalin-CFP containing the mouse talin ( mTalin ) sequence [36] , [37] was described previously . The plasmids containing the tails of myosin XI-K and XI-2 from N . benthamiana were described previously [24] . The full-length of TuMV 6K2 sequence was cloned from plasmid TuMV-GFP and engineered into the bait vector pBT3-STE ( Dualsystems Biotech , Schlieren , Switzerland ) using the SfiI restriction site , designated as pBT3-STE-6K2 . The complete open reading frame coding for the Syp71 ( conserved Syp71 domain in sequence accession number AT3G09740 ) and Vap27-1 ( accession number AY364005 ) proteins were amplified by PCR from wild type Arabidopsis cDNA , inserted into pBT3-STE and pPR3-N vectors ( Dualsystems Biotech ) using the SfiI restriction site and designated as pPR3-N-Syp71 , pPR3-N-Vap27-1 , and pBT3-STE-Vap27-1 . Four-week-old N . benthamiana plants grown in a greenhouse at 22°C to 24°C were used for Agrobacterium tumefaciens ( strain GV3101 ) -mediated transient expression as described previously [12] , [38] , [39] . Chloroplasts were isolated from infected N . benthamiana leaves essentially as described [12] . Immunoblotting was carried out with rabbit sera raised against Arabidopsis Syp71 or against Arabidopsis Vap27-1 as described [40] , [41] . Chloroplast purity was monitored by immunoblotting with antisera against the chloroplast marker PsbA and the ER marker Bip2 . Confocal microscopy and subsequent image processing were conducted essentially as described [42] . Plant tissues were imaged at room temperature using a Leica TCS SP2 inverted confocal microscope with a 63 oil immersion objective . For confocal microscopy , a UV laser and a krypton/argon laser were used to examine fluorescence . GFP was excited at 488 nm , and the emitted light was captured at 497 to 510 nm; CFP was excited at 405 nm , and the emitted light was captured at 440 to 470 nm; YFP was excited at 514 nm , and the emitted light was captured at 525 to 650 nm . Images were captured digitally and handled using the Leica LCS software . Post-acquisition image processing was done with Adobe Photoshop 5 . 0 software . Yeast transformation were performed using the lithium acetate-based protocol for preparing and transforming yeast competent cells following the instructions of the DUAL membrane pairwise interaction kit user manual ( Dualsystems Biotech ) . This system provides bait vectors , prey vectors and controls to perform pairwise interaction assays between membrane proteins . In brief , bait and prey vectors were described above . Different pairs of resulting plasmids were cotransformed into yeast strain NMY51 ( MATa his3 trp1 leu2 ade2 LYS2::HIS3 ura3::lacZ ade2::ADE2 GAL4 ) using the DS Yeast transformation kit ( Dualsystems Biotech ) . Transformed colonies were selected in SD-LW medium and incubated for growth of positive transformants . For growth assays , independent positive transformants were selected and grown in SD-LW liquid medium at 30°C overnight . Culture concentrations were adjusted at OD546 = 1 and diluted 10 , 100 and 1000 times . Three microlitres of each dilution was spotted on to SD-LW and SD-AHLW solid media , respectively , and incubated at 30°C for 2 days . After Agrobacterium-mediated transient expression for 48 h , N . benthamiana leaves ( approximately 0 . 3 g ) were harvested , and ground to powder in liquid nitrogen . Ground tissues were resuspended in 3 . 0 mL of IP buffer containing 50 mM Tris , pH 7 . 5 , 150 mM NaCl , 10% glycerol , 0 . 1% Nonidet P-40 , 5 mM dithiothreitol , and 1 . 5× Complete Protease Inhibitor ( Roche ) . The crude lysates were then spun at 20 , 000 g for 15 min at 4°C to precipitate tissues , cells and cell debris . After centrifugation , 1 ml of supernatant was incubated with 0 . 5 mg of the indicated monoclonal antibody for each immunoprecipitation . After incubation for one hr at 4°C , immunocomplexes were collected by the addition of 50 mL of protein G Sepharose-4 fast flow beads ( Amersham ) and incubated for 4 h at 4°C . The immunocomplexes were washed four times with 1 ml of IP buffer and the pellet was resuspended in 3×SDS-PAGE loading buffer ( 12 ) . The obtained protein samples were separated by SDS-PAGE on 12% polyacrylamide gels and transferred by electroblotting to nitrocellulose membranes . Membranes were probed with anti-HA horseradish peroxidase ( Sigma ) or anti-HIS peroxidase ( Abcam ) to detect HA- and HIS-epitope-tagged proteins , respectively . All immunoprecipitation experiments were repeated three times . For TRV-based gene silencing , four pairs of primers AtSyp71F/AtSyp71R , AtVap271F/AtVap271R , NbSyp71F/NbSyp71R , and NbVap27F/NbVap27R were used to amplify Arabidopsis and N . benthamiana leaf cDNAs to generate partial cDNA sequences of Syp71 and Vap27-1 from Arabidopsis and N . benthamiana . The resulting cDNA fragments were digested with EcoRI and BamHI and then ligated into the corresponding sites of TRV2 [43] . To silence Syp71 , TRV1 and TRV2-AtSyp71 ( or TRV2-NbSyp71 ) were coagroinfiltrated into Arabidopsis ( or N . benthamiana ) essentially as described [43] . Similarly TRV1 and TRV2-Vap27-1 ( or TRV2-Nb271 ) were coagroinfiltrated to silence Vap27-1 in Arabidopsis ( or N . benthamiana ) . Ten days post-infiltration , treated seedlings were inoculated with TuMV::6K2-GFP . Total RNA isolation and DNAse I treatment were as described above . RT reactions were performed with SuperScript III First-Strand Synthesis System for RT-PCR kit ( Invitrogen ) according to the manufacturer's instructions . qPCR was performed using the CFX96 real-time PCR system ( BioRad ) following the manufacturer's instructions . Relative amounts of all mRNAs were calculated from threshold cycle values . The ActinII reference gene ( ACT2F: 5′- GCCATCCAAGC TGTTCTCTC- 3′ and ACT2R: 5′- GAACCACCGATCCAGACACT-3′ ) was used for normalization . Primers CP-F ( 5′-TGGCTGATTACGAACTGACG-3′ ) and CP-R ( 5′-CTGCCTAAATGTGGGTTTGG-3′ ) were used for TuMV detection . All results were shown as means of at least three biological replicates with corresponding standard errors . Wild type and Syp71-downregulated Arabidopsis leaves treated with TRV-based gene silencing ( 10 dpi ) were used for isolation of mesophyll protoplasts following an established protocol [44] . TuMV replication assay was carried out essentially as described [45] . Briefly , protoplasts ( about 1 million ) were transfected with 0 . 5 µg of TuMV::6K2-GFP plasmid in the presence of 18% PEG 400 for 10 min . Transfected protoplasts were washed and resuspended in W5 buffer . The transformed protoplasts were then incubated at RT and harvested at planned time points . Viral RNA accumulation was determined by qRT-PCR as described above . | Potyviruses constitute the largest group of known plant viruses which includes many agriculturally important viruses . Like all other positive-strand RNA viruses , potyviruses induce the cytoplasmic membranous-bound virus factory for viral genome multiplication . But the mechanism by which such a factory is formed and associated with cellular membranes remains elusive . Recently we have demonstrated that upon potyvirus infection , the potyviral 6K2 integral membrane protein is responsible for inducing the formation of the ER-derived replication vesicles that subsequently target chloroplasts for robust genome replication . Here , we report that following the trafficking of the TuMV 6K2 vesicles to chloroplasts , 6K2 vesicles accumulate at the chloroplasts to form chloroplast-bound elongated tubular structures followed by chloroplast aggregation . We show that a host protein , Syp71 , is associated with the 6K2 vesicles . When the Syp71 gene is silenced , TuMV infection is inhibited and the formation of the chloroplast-bound 6K2 complex is arrested but the trafficking of the 6K2 vesicles from the ER to chloroplast is not affected . Therefore , Syp71 is a host factor that mediates the fusion of 6K2 vesicles with chloroplasts during TuMV infection . Our study sheds new light onto the mechanism underlying the association of potyviral 6K2 vesicles with chloroplasts . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"science",
"virology",
"plant",
"pathogens",
"plant",
"pathology",
"viral",
"replication",
"complex",
"biology",
"microbiology",
"viral",
"replication"
] | 2013 | The SNARE Protein Syp71 Is Essential for Turnip Mosaic Virus Infection by Mediating Fusion of Virus-Induced Vesicles with Chloroplasts |
In human cytomegalovirus ( HCMV ) , tropism to epithelial and endothelial cells is dependent upon a pentameric complex ( PC ) . Given the structure of the placenta , the PC is potentially an important neutralizing antibody target antigen against congenital infection . The guinea pig is the only small animal model for congenital CMV . Guinea pig cytomegalovirus ( GPCMV ) potentially encodes a UL128-131 HCMV PC homolog locus ( GP128-GP133 ) . In transient expression studies , GPCMV gH and gL glycoproteins interacted with UL128 , UL130 and UL131 homolog proteins ( designated GP129 and GP131 and GP133 respectively ) to form PC or subcomplexes which were determined by immunoprecipitation reactions directed to gH or gL . A natural GP129 C-terminal deletion mutant ( aa 107–179 ) and a chimeric HCMV UL128 C-terminal domain swap GP129 mutant failed to form PC with other components . GPCMV infection of a newly established guinea pig epithelial cell line required a complete PC and a GP129 mutant virus lacked epithelial tropism and was attenuated in the guinea pig for pathogenicity and had a low congenital transmission rate . Individual knockout of GP131 or 133 genes resulted in loss of viral epithelial tropism . A GP128 mutant virus retained epithelial tropism and GP128 was determined not to be a PC component . A series of GPCMV mutants demonstrated that gO was not strictly essential for epithelial infection whereas gB and the PC were essential . Ectopic expression of a GP129 cDNA in a GP129 mutant virus restored epithelial tropism , pathogenicity and congenital infection . Overall , GPCMV forms a PC similar to HCMV which enables evaluation of PC based vaccine strategies in the guinea pig model .
Human cytomegalovirus ( HCMV or Human herpesvirus 5 ) is a member of the Betaherpesvirinae genus and encodes over 165 genes [1] . Viral infection is largely asymptomatic in healthy individual but establishes a lifelong mainly latent state in the host . However , infection of an immune compromised host ( AIDS and transplant patients ) or virus reactivation because of an impaired immune system can have severe consequences of morbidity or mortality [2 , 3] . An additional important aspect of HCMV is congenital infection , where the virus crosses the placenta and infects the fetus in utero . This occurs in approximately less than 1% of live births [4] in the US and causes serious symptomatic disease including mental retardation and sensorineural hearing loss ( SNHL ) in newborns [4–8] . The greatest risk of congenital infection is to mothers who acquire a primary infection during pregnancy and prior immunity can reduce the risk by up to 69% [9] . Hence , the impact of a vaccine is potentially substantial , especially in countries where there is a greater risk of primary infection during pregnancy . These regions include the US , EU and Japan , where up to 50% of women of child bearing age are negative for HCMV [8 , 10] . Licensed HCMV antivirals are available for transplant and AIDS patients but not congenital CMV [11] . Consequently , development of a vaccine against congenital CMV is a high priority . Any proposed intervention for the prevention or treatment of HCMV infection should ideally be evaluated in a pre-clinical model . Unfortunately , HCMV is extremely species-specific . Consequently , animal model pathogenicity , vaccine and antiviral studies are carried out using animal-specific CMVs , including mouse , rat , guinea pig and rhesus macaques [12–16] . The guinea pig is unique insofar as it is the only small animal model to allow the study of congenital CMV infection , where the virus crosses the placenta and infects the fetus in utero unlike the mouse model [17] . Both human and guinea pig placentas are hemomonochorial containing a homogenous layer of trophoblast cells separating maternal and fetal circulation [18–20] . Additionally , as with human pregnancy , the gestation period ( approximately 65 days ) can be divided into trimesters . Importantly , GPCMV congenital infection causes disease in the fetus and in newborn pups similar to those found in humans including SNHL [21–23] . Consequently , the guinea pig model is best suited for testing of intervention strategies aimed at preventing congenital CMV infection [11 , 24 , 25] . A major drawback in GPCMV research has largely been overcome by the recent sequencing of the viral genome and the development of infectious BAC clones of GPCMV [15 , 26–29] . Indeed , manipulation of GPCMV BACs has allowed the preliminary study of some viral genes [11 , 30–36] . Additionally , the guinea pig animal genome has been sequenced ( http://www . ensembl . org/Cavia_porcellus/Info/Index ) which enables the development of new reagents for this model . Analysis of the GPCMV genome [15 , 29] indicated that the virus encoded homologs to the HCMV glycoproteins ( gB , gH , gL , gM , gN , gO ) in genes co-linear with the HCMV genome ( designated GP55 , GP75 , GP115 , GP100 , GP73 and GP74 respectively ) . In HCMV , these six glycoproteins ( gB , gH , gL , gM , gN , gO ) are required for fibroblast cell entry and they form the glycoprotein complexes , gCI ( gB ) , gCII ( gM/gN ) , gcIII ( gH/gL/gO ) on the viral membrane [37–39] . Additionally , in HCMV these complexes are important neutralizing antibody targets and as such potential vaccine candidates [40–44] . We recently demonstrated that GPCMV forms functionally similar glycoprotein complexes and these complexes are essential for infection of fibroblast cells as well as important target antigens [36] . In both HCMV and GPCMV , the viral glycoprotein gB is the immunodominant neutralizing viral antigen [45–49] . A recombinant HCMV gB has been investigated as a candidate subunit vaccine in phase II clinical trials but this provides at best approximately 50% efficacy despite high antibody titers [41] . The process of HCMV entry into the cell was assumed to occur via the mechanism of cell fusion mediated by gB but also requiring other glycoproteins [50–52] . Current studies indicate that it is the triplex ( gH/gL/gO ) and specifically gH/gL that promotes gB cell fusion in fibroblasts . This is in keeping with the general model for herpesviruses with core fusion being related to gB and gH/gL [53–58] . However , gM/gN is considered essential for virus infection of all cell types and is the most abundant complex in the virion [59] . Most of the earlier HCMV cell entry studies were performed on fibroblast cells with lab adapted strains of HCMV ( eg . AD169 ) . These viral strains lack the capacity to efficiently infect other cell types such as endothelial , epithelial cells . Clinical strains of HCMV also encode a pentameric glycoprotein complex ( gH/gL/UL128/130/131 ) that enables viral entry into epithelial , endothelial and myeloid cells via an alternative pathway of cell entry that requires the pentameric complex ( PC ) in association with gB [55 , 57 , 60–66] . Alphaherpesviruses , unlike CMV , only encode one type of gH/gL complex . However , some gammaherpesviruses ( eg . Epstein-Barr virus ) encode two different gH/gL complexes to enable gB fusion into different cell types entry which potentially provides a model for CMV [54] . It is likely that gH/gL based complexes work upstream of gB for cell entry but this is poorly defined [58] . The alternative route of CMV cell entry will be referred to as the PC dependent pathway in this report . However , studies suggest that gO , which is unique to all CMV , also enhances infection of epi/endothelial cells by an undefined mechanism [67] . PC dependent virus infection of epithelial and endothelial occurs via a clathrin-independent endocytosis pathway with the endosomes undergoing an acid flux [61 , 68] . PC dependent virus infection of dendritic cells is independent of pH but dependent upon cholesterol via macropinocytosis pathway [69] . Interestingly , a recent study demonstrated that virus entry into fibroblast cells can also occur via a pH/clathrin independent macropinocytosis pathway in virus devoid of the PC [70] . Undoubtedly , the PC is necessary for efficient entry into epithelial and endothelial cells . The viral locus encoding the PC unique genes ( UL128-131 ) is unstable upon passage of clinical HCMV strains on fibroblast cells and encoded genes rapidly acquire point mutations or deletions with the subsequent loss of epi/endothelial viral tropism associated with the inability to form a functional PC [71] . Lab adapted strains of HCMV ( eg . AD169 ) can infect epi/endothelial cells when the mutated locus is repaired or functional genes are expressed in an ectopic location which enable PC formation [64 , 72 , 73] . The basis for HCMV forming gH/gL/gO triplex or gH/gL PC is poorly defined but both complexes are present on the virion of clinical strains but ratios differ between strains [74] . Potentially , competitive binding of gO or UL128 with gH/gL might be a key stage but UL148 protein has also been suggested to play a role in the balance between these complexes [55 , 75] . The PC is considered an important neutralizing target for HCMV on epithelial/endothelial cells and presumably for congenital infection , given the epi/endothelial structure of the placenta [76–78] . The importance of the PC as a target antigen was confirmed by the isolation of neutralizing human monoclonal antibodies to the PC which had higher potency than antibodies to other target antigens [76 , 77] . In the context of congenital infection , high titer neutralizing antibodies are thought to be effective against transplacental viral transmission [76 , 79 , 80] . Potentially , a delay in the immune response to the PC results in fetal infection [81] . The importance of the PC in virus infection of cells is underscored by a recent finding for the gB subunit HCMV vaccine . In clinical trials , the gB vaccine induces a high titer neutralizing immune response which is effective in neutralizing virus on fibroblasts [41 , 82] . However , in separate studies sera from gB vaccinated individuals is less effective at neutralizing virus infection on endothelial and epithelial cells in comparison to convalescent sera from HCMV infected individuals [78 , 83 , 84] . This demonstrated the importance of other viral neutralizing target antigens for infection on these cell types . Consequently , other target antigens should be considered important in the development of a vaccine against congenital CMV . Importantly , a gB vaccine fails to fully protect against congenital CMV in the guinea pig model [49 , 85 , 86] . Potentially , GPCMV encodes a homolog PC as a UL128-131 homolog locus ( GP128-133 ) was identified in low pass ATCC viral stock of GPCMV , strain 22122 ( ATCC VR682 ) [28 , 87] . The sequence of this virus also matched that of salivary gland ( SG ) GPCMV ( extensively serially passaged in vivo in guinea pigs ) [25] . However , a plaque purified isolate ( designated PP ATCC ) , used to establish the first GPCMV sequence , carried a deletion in this locus [15] . PP ATCC had normal growth kinetics in tissue culture but was attenuated in vivo compared to SG GPCMV [25 , 26] . We hypothesized that the viral attenuation was linked to the inability of the virus to form a homolog pentameric complex which impacted on virus cell tropism to specific cell types ( eg . epithelial cells ) and consequently pathogenicity in the animal . Virus cell tropism was evaluated with a newly established guinea pig renal epithelial cell line . The potential interactions of GPCMV homolog glycoproteins gH and gL with proteins encoded in the GP128-133 locus were studied to provide evidence of a GPCMV PC . [63] . Additionally , a GPCMV mutant virus which encoded an intact locus but a mutated GP129 gene ( UL128 homolog ) was restored for epithelial tropism by expression of a full length GP129 cDNA in an ectopic location . This recombinant virus ( GP129FRT ) incorporated a myc tagged GP129 into the viral particle as part of the PC . Knockout of GP129 , GP131 or GP133 ( UL128 , UL130 and UL131 homologs respectively ) in recombinant GPCMV impaired virus replication on epithelial cells but not fibroblasts . Importantly , virus restored for PC had improved pathogenicity and congenital transmission rates compared to mutant virus lacking the complex . Overall , the similarity of function between HCMV and GPCMV pentameric complexes strengthens the guinea pig model in the development of an effective preclinical vaccine strategy against epithelial and congenital infection based on the CMV PC .
In earlier reported studies of GPCMV , strain 22122 ( ATCC VR682 ) , the pathogenic salivary gland ( SG ) virus was maintained by serial passage in animals but could be attenuated by extensive serial passage of the virus on fibroblast cells , >11–25 passes [21 , 88 , 89] . However , the molecular basis for this attenuation was undefined and the viral stocks of the SG and fibroblast passaged virus are no longer available ( Griffith ( Yale University , CT ) personal communication to AM ) . A potential basis for the viral attenuation in GPCMV might be by modification of a homolog UL128-133 locus [28 , 87 , 90] . In clinical HCMV strains , adaptation of the virus to growth on fibroblast cells rapidly resulted in mutations in this locus and impaired virus tropism to various cell types [60 , 71 , 91] . Inoue and colleagues [87] identified two variants of GPCMV in low pass ATCC stock of GPCMV ( strain 22122 ) . One GPCMV variant was intact for the homolog UL128-131 locus ( GP128-GP133 ) , whereas the other carried a deletion in the locus and was similar in sequence to the tissue culture adapted GPCMV isolate [15 , 29] . GPCMV , strain 22122 ( ATCC ) , was serially propagated in guinea pigs at Children’s Hospital Research Foundation , Cincinnati ( Ohio , USA ) from the late 1980s-2005 . Salivary gland ( SG ) viral stocks were extensively used in congenital GPCMV challenge studies by a number of investigators . This SG virus stock was recently sequenced and shown to encode a full length GP128-133 locus [92] , whereas the tissue culture adapted virus derived from the SG GPCMV stock used to establish the first viral genome sequence carried a deletion in this locus [15] . In this present study , GPCMV salivary gland stock SG11 ( 11 direct serial passages in guinea pigs ) was used to evaluate the sequence of the GP128-133 locus in the virulent virus . Additionally , plaque purified ( PP ATCC ) virus stock extensively passaged on fibroblast cells was used to evaluate the sequence of lab adapted virus . PCR primers ( [87] , S1 Table ) were used to amplify the GP128-GP133 locus and the PCR product cloned prior to sequencing . Fig 1 ( i ) shows the structure of the GP128-133 locus and the analysis of the cloned PCR products for the GP128-133 locus from respective viruses . Sequence analysis of the cloned PCR products ( see Fig 1 ( ii ) A and S1 Fig ) confirmed that the two virus stocks ( SG GPCMV and PP ATCC plaque isolate ) differed in an identical fashion to the two isolates reported in low pass ATCC stock [28 , 87] . The SG GPCMV had a complete GP128-133 locus ( 2 kb PCR product ) , whereas the PP ATCC virus [26] carried a 1 . 6 kb deletion ( 0 . 4 kb PCR product ) which removed the majority of the GP129-133 coding sequence ( 196 , 925–198 , 573 ) as shown by PCR analysis ( see Fig 1 ( ii ) A ) . This potentially supports the hypothesis that a full length GP128-133 locus is necessary for full tropism/pathogenicity in vivo since GPCMV mutated in the GP128-133 locus is attenuated in the animal model [29] . RT-PCR at late stage infection confirmed transcription of genes encoded in the GP128-133 locus in SG GPCMV infected cells ( S2 Fig ) . In HCMV , a consequence of adaptation of clinical strain virus to fibroblast cells is an inability for the virus to form a functional PC ( gH/gL/UL128-131 ) necessary for infection of epithelial or endothelial cells and other cell types [60 , 62–64 , 71] . Potentially , GPCMV encodes a homolog PC but cellular tropism associated with this locus has not been successfully demonstrated [90] except for infection of macrophage [93] . A major limitation in GPCMV studies is the availability of different types of tissue culture cell lines to evaluate virus tropism . Consequently , we generated a novel guinea pig renal epithelial cell line from the kidney . Colonies were clonally isolated , characterized by cytokeratin marker and cell lines immortalized as described in materials and methods . Epithelial cells were characterized by assay for cytokeratin , either by western blot or immunofluorescence assay ( see S3 Fig ) . The guinea pig epithelial cells were positive for cytokeratin unlike GPL cells ( S3 Fig ) . Importantly , SG GPCMV was capable of infecting and replicating on epithelial cells . Fig 1 ( ii ) shows virus infected epithelial cells co-stained for cytokeratin ( cytoplasm ) and GPCMV IE2 protein ( nucleus ) , see panels C and B respectively . In contrast , viral antigens failed to be detected in PP ATCC GPCMV infected epithelial cells ( S4 Fig ) but control studies demonstrated virus infection of fibroblast cells ( S4 Fig ) . Additionally , a growth curve of SG GPCMV vs PP ATCC demonstrated that virus with an intact GP128-133 locus can infect and replicate on epithelial cells , whereas the PP ATCC mutant virus was highly impaired for growth on epithelial cells , see Fig 1 ( iii ) A . In contrast , both viruses were capable of normal growth on fibroblast cells ( Fig 1 ( iii ) B ) . It was concluded that the SG GPCMV has the ability to replicate on epithelial cells , unlike lab adapted GPCMV . Additionally , tropism was a result of a SG GPCMV virus encoding a complete GP128-133 locus unlike lab adapted virus . This could be verified by comparative sequence analysis between isolates which indicated that the only difference between viruses related to the GP128-133 locus [29 , 87 , 94] . In order to study GPCMV pentameric glycoprotein complex formation , full length cDNA clones of GP128 , GP129 , GP131 and GP133 were cloned into mammalian expression vectors or recombinant adenovirus vectors under HCMV MIE promoter control as described in materials and methods [36] . Additionally , ORFs were C-terminal epitope tagged to enable detection: GP129 ( myc ) ; GP131 ( HA ) ; GP128 ( FLAG ) ; and GP133 ( FLAG ) . Also , previously described GP75 ( gH ORF ) and GP115 ( gL ORF ) , C-terminal tagged with GFP or mCherry respectively , in mammalian expression plasmids or recombinant defective adenovirus [36] were utilized in the study . All constructs were verified by sequencing and are summarized in S5 Fig . An initial series of experiments examined the cellular localization of the various GPCMV proteins . Cells transduced with defective adenoviruses expressing gH , gL , GP129 , GP131 and GP133 resulted in a cytoplasmic co-localization of gH and gL with GP129 , GP131 and GP133 ( S5 Fig ) . Next , an immunoprecipitation assay was performed on cells that expressed all the components of the potential pentameric complex to demonstrate protein:protein interactions . Fig 2 ( i ) demonstrated that transduction of epithelial cells with recombinant adenovirus encoding a single glycoprotein enabled their successful detection by western blot analysis using their respective epitope tag marker . Transient expression of gHGFP and gLmCherry have previously been described [36] . Western blot analysis of transiently expressed GP129myc , GP131HA and GP133FLAG produced proteins with larger than expected size: GP129myc ( expected 24 compared to 40kDa ) ; GP131HA ( 25 compared to 31 kDa ) ; GP133FLAG ( 16 . 7 compared to 19 kDa ) ( S2 Table ) . It was presumed that this was a result of post-translation modification such as glycosylation as previously demonstrated for gH and gL [36] . Potentially , both glycosylated and non-glycosylated versions of GP131HA are seen in Fig 2 ( i ) as proteins at two different molecular weights ( approximately 31 and 25 kDa ) are detected . Treatment of cells with the glycosylation inhibitor tunicamycin ( S6 Fig ) demonstrated that GP129 and GP131 were subject to glycosylation as predicted in S2 Table . In the presence of tunicamycin , only lower molecular weight proteins were detected . Tunicamycin treatment also resulted in proteins appearing more aggregated in cellular immunofluorescence studies ( S6 Fig ) . Next , GFP-trap ( Chromotek ) immunoprecipitation ( IP ) assays were used to detect interaction with gH-GFP/ PC formation in cells transduced by all five recombinant adenoviruses ( AdgHGFP , AdgLmCherry , AdGP129myc , AdGP131HA and AdGP133FLAG ) , Fig 2 ( ii ) . The GFP trap approach had previously been used to successfully study GPCMV glycoprotein complexes gM/gN ( via gMGFP IP ) and gH/gL/gO ( via gHGFP IP ) [36] . The GFP-trap IP of lysed cells transduced for all potential PC protein components resulted in the successful IP of gLmCherry , GP129myc , GP131HA and GP133FLAG by gHGFP ( Fig 2 ( ii ) ) . Interestingly , both species of GP131HA were immunoprecipitated . In contrast , a control GFP trap IP of cells transduced with AdGFP , AdgLmCherry , AdGP129myc , AdGP130HA and AdGP133FLAG but not AdgHGFP resulted in successful IP of GFP but none of the components of the PC ( Fig 2 ( iii ) ) , which demonstrated the specificity of the IP assay as well as the importance of gH in the interactions with other PC proteins . A similar series of PC immunoprecipitation reactions were also performed using RFP-trap ( Chromotek ) with gLmCherry and control mCherry . All PC components were able to be precipitated by gLmCherry IP which demonstrated the importance of gL for PC formation ( S7 Fig ) . A control mCherry in place of gLmCherry failed to immunoprecipitate components of the PC ( S7 Fig ) Next , we investigated the impact of C-terminal GP129 ( UL128 homolog ) mutants on the ability to form a pentameric complex with other GPCMV proteins . An initial study was carried out with a natural GP129 mutant . The second generation GPCMV BAC derived virus encodes the full spectrum of viral genes [29 , 95] but contains a 4 bp deletion in the GP129 gene , which places the ORF out of frame and truncated the encoded protein at codon 102 ( NRD13 , see S8 Fig ) . Virus derived from this BAC has normal growth kinetics on fibroblast cells but lacked the ability to grow on epithelial cells ( see section on restoration of epithelial tropism ) . A cDNA clone of the truncated GP129 mutant ( designated GP129NRD13 ) was myc-epitope tagged and cloned into a transient expression vector and assayed for an ability to form a PC in a GFP trap immunoprecipitation assay . Despite detectable GP129NRD13 expression levels , the protein failed to be immunoprecipitated as part of a PC ( Fig 3 ) . This indicated the importance of the C-terminal domain of GP129 in complex formation . A chimeric GP129 C-terminal mutant was also generated synthetically that encoded the C-terminal domain of HCMV UL128 ( Merlin strain ) in place of GP129 after the NRD13 truncation site ( designated GP129UL128 , see S8 Fig ) . In transient expression studies , the GP129UL128 chimeric protein failed to form a PC with other components ( Fig 3 ) . This indicated that there was insufficient conservation between the C-terminal domains of GPCMV GP129 and HCMV UL128 to enable PC formation . Although the GP129 mutants could not be precipitated as part of a PC , the gHGFP immunoprecipitation reactions did pull down other components of the complex ( gL , GP131 and GP133 ) . Potentially , components of the PC were capable of forming subcomplexes with gH/gL . Consequently , the ability of GP129 , GP131 and GP133 to independently form triplexes with gH/gL was investigated . In transient expression immunoprecipitation assays based on gHGFP , gH/gL formed triplex complexes with GP129 , GP131 or GP133 ( Fig 3 ) . Protein cellular co-localization could also been demonstrated in the cytoplasm of epithelial cells ( Fig 4 and S9 Fig ) . The relevance of the various subcomplex triplexes to the PC or to the gH/gL/gO triplex formation and viral assembly remains to be more fully investigated in future studies but would indicate that gH/gL interaction is not solely dependent upon GP129 . Subcomplex formation was also investigated for the GP129 mutant . Triplex formation could not be demonstrated to occur with gH , gL and GP129 NRD13 with a similar outcome to that obtained for interaction with all PC components ( Fig 3 ) . This implied that the C-terminal portion of GP129 was also important for subcomplex formation . Overall , it was concluded that GPCMV forms a homolog pentameric complex and that the UL128 homolog ( GP129 ) , especially the C-terminal domain is important for complex formation . GPCMV serially passaged in animals as salivary gland stock retained an ability to infect epithelial cells unlike lab adapted virus . The second generation GPCMV BAC encoded a full length GP128-GP133 locus but carried a 4 bp deletion that truncated the GP129 ORF ( NRD13 mutant , S8 Fig ) compared to wild type virus [29 , 95] . Transient expression of GP129NRD13 protein together with other components of the pentameric complex ( gH , gL , GP131 and GP133 ) failed to generate a detectable PC ( Fig 4 ) . Additionally , BAC derived GPCMV lacked epithelial tropism but grew normally on fibroblast cells ( see below ) [96] . The capability of BAC derived GPCMV to form a functional pentameric complex and potentially restore epithelial tropism was evaluated by the introduction of a full length GP129 cDNA expression cassette into the GPCMV genome in a non-essential intergenic locus . The intergenic site between UL25 and UL26 homologs ( GP25 and GP26 ) was selected on the basis of co-terminal transcripts ending in this locus with sufficient intergenic sequence to enable insertion of an ectopic cassette without interfering with GP25 or GP26 expression based on previous studies [25] ( see S10 Fig ) . A cDNA myc tagged GP129 ORF used in transient expression studies was cloned initially into a shuttle vector ( pGP2526GP129LinkKm ) which placed the GP129 cDNA under SV40 promoter and SV40 polyA control as described in materials and methods ( S11 Fig ) . Mutant GPCMV BAC clones were selected by kanamycin ( Km ) marker . Full length GPCMV BAC mutants encoding GP129myc in the GP25/GP26 locus were verified by restriction enzyme profile analysis ( S12 Fig ) and by PCR ( S11 Fig ) and sequencing . DNA from correctly identified mutant GPCMV BACs were transfected onto GPL cells to generate virus ( GP129FRT ) . Virus expression of myc tagged GP129 and protein incorporation into the virus particle was demonstrated by western blot analysis of sucrose gradient purified viral particles . Fig 5 demonstrated that GP129myc was expressed in virus infected cells . Additionally , that GP129 was present in the purified viral particles as was GP131 , another unique component of the PC . Glycoprotein gH could also be detected in virus particles , presumably as part of the triplex homolog ( gH/gL/gO ) and also the pentameric complex ( gH/gL/GP129/GP131/GP133 ) . Additionally , gB glycoprotein could be detected as a viral particle component but not GFP , which was expressed in infected cells but not incorporated into the viral particle . A growth curve confirmed that GP129FRT GPCMV was highly trophic for epithelial cells with efficient virus growth ( Fig 6 ) . In comparison , the parental derived GPCMV BAC virus NRD13 which encoded a truncated GP129 but viable GP131 and GP133 failed to efficiently replicate on epithelial cells ( Fig 6 ) . Next , the requirement for other components of the homolog PC for virus growth on epithelial cells was evaluated . Targeted mutagenesis was performed on the GP128-GP133 locus to generate GPCMV BAC mutants with knockout GP128 , GP129-GP131 or GP133 in separate BAC mutagenesis reactions . Final GPCMV BAC mutants encoded an ectopic GP129 cDNA in the GP25/GP26 intergenic locus in addition to specific mutations in the GP128-GP133 locus as described in materials and methods . BAC mutants were characterized by restriction profile analysis , PCR and sequencing of the PCR product ( S12 Fig and S11 Fig ) . Recombinant viruses were designated GP128FRT/GP129Link ( GP128 mutant ) ; GP129-GP131FRT/GP129Link ( GP131 mutant ) ; GP133FRT/GP129Link ( GP133 mutant ) . Although these viral mutants had similar growth kinetics on GPL cells , they lacked an ability to efficiently infect epithelial cells ( S13 Fig ) . The exception was the GP128 mutant which retained an ability to infect epithelial cells ( Fig 7 ) . In the case of GP131 and GP133 GFP tagged mutant viruses ( non-cre BAC excised ) a contrasting infection could be demonstrated between fibroblast and epithelial cells . In separate experiments , a moi of 1 pfu/cell resulted in a 100% infection of GPL cells but only approximately 1 in 106 epithelial cells . Additionally , mutant virus infected epithelial cells failed to result in virus spread to surrounding cells , unlike GP129FRT or SG GPCMV . Examples of impaired GP131 and GP133 mutant virus growth on epithelial cells compared to GPL cells is shown in S13 ( ii ) Fig . Overall , it was concluded that epithelial tropism could be restored to lab adapted GPCMV by ectopic expression of a missing full length GP129 protein . This virus ( GP129FRT ) expressed GP129 and GP131 proteins as part of the viral particle . Knockout of individual genes in the GP128-GP133 locus in the backdrop of virus expressing the ectopic full length GP129 also confirmed the essential role of GP131 and GP133 in PC formation and epithelial tropism . Knockout of the GP128 gene did not prevent epithelial tropism and transient expression of GP128 demonstrated that it was a nuclear targeting protein ( Fig 7 ) . The functional significance of GP128 in the GPCMV life cycle remains unknown . BLAST analysis of the predicted GP128 protein sequence indicated that it was a potential homolog of MCMV IE2 [15] and therefore is unlikely to be relevant to PC formation . In a previous study , it was demonstrated that GPCMV forms a homolog gH/gL/gO glycoprotein triplex and that gO ( GP74 ) was only essential in lab adapted virus which lacked a PC [36] . In an effort to demonstrate that the PC is more important than the triplex complex for infection of epithelial cells , a series of gO ( GP74 ) / GP129 mutant GPCMV BACs were transfected onto GPL or epithelial cells to evaluate virus spread . Three GPCMV BAC mutants ( Fig 8 ) were used: ( 1 ) GP74Km , which contains a GP74 knockout on a GPCMV BAC that lacked a full length GP129 [36]; ( 2 ) NRD13 , GPCMV BAC that lacked full length GP129; ( 3 ) GP129FRT/GP74Km , GPCMV BAC that encoded a GP74 knockout and a full length GP129 cDNA in the GP25/GP26 intergenic locus . Transfection of GP74Km GPCMV BAC DNA ( 1 ) onto GPL or epithelial cells failed to enable the development of viral plaques but instead remained as single transfected cells which could be identified by GFP reporter gene expression ( Fig 8A–8C ) . Transfection of NRD13 ( GP129 mutant ) GPCMV BAC ( 2 ) onto GPL or epithelial cells resulted in the development of viral plaques and spread on fibroblast cells but failed to produce virus on epithelial cells ( Fig 8D–8F ) . Transfection of GPCMV BAC GP129FRT/GP74Km onto GPL and epithelial cells resulted in the development of viral plaques on both cell types ( Fig 8G–8I ) . Importantly , virus derived from BAC GP129FRT/GP74Km demonstrated that gO was not completely essential for epithelial cell infection . as virus spread across the entire monolayer of epithelial cells . Although GPCMV infection of epithelial cells could occur in the absence of gO via the PC , the gB glycoprotein was presumed to be required for epithelial cell infection as demonstrated for fibroblast cells [36] . In an additional experiment , a PC+ /gO+/gB ( GP55 ) negative mutant GPCMV BAC ( GP129FRT/GP55Km ) when transfected onto epithelial cells failed to produce infectious virus . In contrast , a gB rescue virus had restored epi-tropism which emphasized the essential nature of the gB protein for epithelial infection despite the requirement for a PC ( S14 Fig ) [36] Since infection of epithelial cells occurred for a PC+/gO negative virus but not for a PC negative/gO+ virus , we concluded that epithelial cell infection is not absolutely dependent upon gO or the gH/gL/gO triplex but the PC is essential for epithelial cell infection . Admittedly , this particular assay was relatively crude since it fails to fully discriminate between cell to cell spread and infection from cell release virus . However , the strategy demonstrated an importance of the PC for epithelial cell infection . A similar approach defined the essential nature of the GPCMV glycoproteins ( gB , gH , gL , gO , gM and gN ) for fibroblast cells [36] . In HCMV , a PC+/gO+ virus can more easily infect epithelial cells than a PC+/gO negative virus [67] . Therefore in HCMV , gO has an undefined role in infection of cell types other than fibroblasts via an undefined mechanism [36] . The role of gO in potentially enhancing GPCMV infection of epithelial cells and other cells types remains to be evaluated . In HCMV , PC dependent infection of epithelial and endothelial cells occurs via an endocytic pathway that requires an acid flux in the endosome [61] . The antibiotic bafilomycin prevents HCMV infection of epi/endothelial cells by inhibiting the ATPase and subsequent acidification of the endosome [61 , 97 , 98] . Pretreatment of guinea pig epithelial cells with 50nM bafilomycin dramatically inhibited virus infection of epithelial cells but did not greatly impact on fibroblast cell infection ( S15 Fig ) . This potentially indicated that GPCMV entry into epithelial cells is via a similar pathway to HCMV . More focused future studies on the process of GPCMV entry into epithelial cells might provide better insight into the entry pathway for GPCMV and the similarity to HCMV . Analysis of the level of PC compared to gH/gL/gO on GPCMV particles might be worthy of future investigation as this might help to define virus cell tropism . In an effort to determine if expression of the PC increased virus pathogenicity , comparative animal studies were performed with different GPCMV: GP129FRT ( group 1 ) ; SG GPCMV ( group 2 ) ; and NRD13 , BAC derived GPCMV ( group 3 ) . Seronegative animals were randomly divided into three groups ( n = 12 per group ) and animals were each inoculated with 106 pfu of virus ( either GP129FRT , SG GPCMV or NRD13 dependent on their specific group ) . At various time points ( 4 , 8 , 12 and 27 days post infection ) , three animals per group were euthanized and the viral load in target tissue and blood were determined by real time PCR as described in materials and methods . The results of the viral pathogenicity study are shown in Fig 9 . Statistical analysis ( Student t test ) was carried out for GP129FRT vs SG GPCMV and GP129FRT vs NRD13 on tissue from similar target organs at similar time points . Overall , the GP129FRT virus had a dissemination pattern that resembled SG GPCMV during the first 12 days of infection in target organs lung , liver and spleen ( see Fig 9A–9C ) . The viral load in the salivary gland was not evaluated until day 27 and was detected for both GP129FRT and SG GPCMV . The viral load in the salivary gland was approximately 1 log higher for the SG GPCMV compared to GP129FRT which was significant ( p <0 . 005 ) . Additionally , at day 27 the SG GPCMV could be detected in the target organs lung , liver and spleen , whereas GP129FRT could only be detected in the spleen . Overall , in the target organs there was a statistically significant difference ( p <0 . 05 to <0 . 005 ) in viral load between GP129FRT and SG GPCMV except in the liver ( D4 and D8 ) and spleen at D4 . Both SG GPCMV and GP129FRT exhibited similar viremia levels at days 4 , 8 and 12 post infection , with peak levels detected at 8 days post infection . No viremia was detected at day 27 . In contrast to GP129FRT and SG GPCMV , the parental BAC derived NRD13 virus ( encoding a truncated GP129 ) was highly attenuated in the animal model , see Fig 9 . NRD13 was detected in lung , liver spleen at days 4 , 8 and 12 post infection but at substantially reduced levels compared to GP129FRT and SG GPCMV . At all comparable time points , viral titer in all NRD13 tissues were significantly lower with p < 0 . 005 when compared to GP129FRT ( except D4 lung and liver , p < 0 . 05 ) . NRD13 failed to be detected in any target organs at day 27 and was not present in the salivary gland . In contrast to the other viruses , NRD13 viremia at all time points was below the level of detection ( Fig 9E ) . Presumably , the inability of NRD13 to form a pentameric complex and inability to infect a wider range of cell types precluded the ability of NRD13 to effectively disseminate or replicate in target organs . Overall , it was concluded that restoration of an ability to express a missing full length GP129 protein resulted in a virus that could not only infect epithelial cells in tissue culture but also had a greater pathogenicity in the animal model compared to the parental GP129 mutant virus ( NRD13 ) . Importantly , the GP129FRT virus was capable of disseminating to the salivary glands unlike the GP129 mutant . The placenta consists of both epithelial and endothelial cells , the improved tropism of the GP129FRT virus for epithelial cells and other cell types in vivo could also potentially increase the congenital transmission rate of the virus compared to the parental NRD13 virus . Seronegative pregnant dams were challenged with 106 pfu of NRD13 ( group 1 , n = 8 ) or GP129FRT ( group 2 , n = 11 ) at late second trimester via subcutaneous inoculation and animals were allowed to go to term . The viral load in pup target organs ( liver , lung , spleen , brain ) of live or still born animals was evaluated . Table 1 has the mortality outcome for live vs still born pups for the groups . Table 2 has the viral load in the target organs of the pups ( live and dead ) . There was a higher number of still born pups in the GP129FRT group ( 9 pups , 17 . 3% ) compared to the NRD13 group ( 2 pups , 6 . 5% ) . Overall , there was a greater incidence of organs positive for virus in the GP129FRT group compared to the NRD13 group ( p < 0 . 05 ) . In the latter group , only the brain was detected positive for virus in one pup and all organs were negative in all other animals . In the GP129FRT group , 20/52 pups had virus detected in various organs ( brain , liver , spleen and lung ) . The transmission rate for the GP129FRT virus was 38 . 46% compared to 3 . 2% for the NRD13 virus , which was statistically significant ( p = 0 . 0002 ) . It should be noted that none of the still born pups in the NRD13 group were positive for virus and their death in utero was likely a complication of pregnancy and not associated with congenital CMV infection . Only a limited number of term placentas were available for evaluation of viral load ( 2 for GP129FRT group and 7 for NRD13 group ) but only the GP129FRT infected animals had CMV positive placentas ( see Table 2 ) . The presence of GP129FRT virus in the placenta ( 3rd trimester ) of an additional pregnant guinea pig was evaluated by immunohistochemistry at day 22 post-infection . Fig 10 shows the results for immunohistochemistry staining of cryostat sections of placenta for GPCMV gB antigens . The presence of virus in placenta tissue was also verified by DNA extraction and PCR . Fig 10 Panel A shows a cartoon of the placenta structure and indicated regions where viral antigen was detected in cryostat sections ( C and D ) . Panel B shows an agarose gel PCR analysis for DNA extracted from virus infected epithelial cells or placenta section . The PCR was for the GP128-GP133 locus and only a full length locus ( 2 kb ) was detected . Panels C and D are sections stained for GPCMV gB protein . Representative control sections ( no primary antibody ) are shown in panels E and F . Virus was mainly detected in the interface of the labyrinth region of the placenta . These results confirm the presence of viral antigens in the placenta . We concluded that the GPCMV congenital transmission rate was highly dependent upon the virus encoding a functional GP129/ PC . A wider range of virus cell tropism would presumably be a requirement for effective congenital transmission to the fetus since the maternal fetal barrier in the placenta consists of a layer of epithelial syncytiotrophoblast cells [99 , 100] .
Electron microscopy histopathology studies of guinea pig salivary gland ( duct cells ) and placental tissue ( trophoblast cells ) from GPCMV infected animals suggested that GPCMV can infect epithelial cells [21 , 101] . Retrospectively , these were important observations in the context of viral tropism and recent identification of a novel mechanism of HCMV infection of epithelial and endothelial cells and other cell types [60 , 61 , 63 , 64] . Our recent knockout mutagenesis studies of GPCMV glycoprotein genes ( encoding homologs of gB , gH , gL , gO , gM , or gN ) demonstrated a conservation of essential function and homolog glycoprotein complex formation ( gB , gH/gL/gO and gM/gN ) between HCMV and GPCMV [36 , 102] . This present study demonstrated that the UL128-131 homolog locus ( GP129-GP133 ) [28] was necessary for GPCMV infection of guinea epithelial cells in tissue culture and that tropism is dependent upon the ability to form a pentameric homolog complex which is structurally present in the viral particle . GPCMV knockout mutagenesis demonstrated that pentameric complex formation and epithelial tropism was dependent upon an ability of the virus to express wild type GP129 , GP131 and GP133 but that GP128 was non-essential for complex formation , nor epithelial tropism . Furthermore , improved epithelial tropism , pathogenicity and congenital infection in the animal model could be established for a mutant virus ( GP129 mutant in the GP128-133 locus ) via ectopic expression of a wild type GP129 cDNA and subsequent ability to form a homolog pentameric complex . Importantly , GPCMV pentameric complex formation is dispensable for infection of fibroblast cells , as is the case for HCMV . Curiously , Auerbach et al . [90] showed an enhancement of GPCMV fibroblast infection associated with a full length GP129-133 locus . Our studies with both GPL fibroblast cells and in house primary fetal embryo fibroblasts did not determine a dependence upon an intact GP128-133 locus for fibroblast virus infection . Similarly , Inoue and colleagues [26] did not see PC dependence for infection of primary or immortalized guinea pig fibroblast cells [28] . Surprisingly , Auerbach and colleagues failed to demonstrate a specific tropism for endothelial cells in virus with an intact GP128-133 locus compared to virus with a deletion in that locus [90] . Based on our epithelial tropism data and in vivo pathogenicity studies , the prediction would be that PC positive GPCMV would exhibit increased tropism to a wider range of cell types . This increased range would include endothelial cells . HCMV exhibits the same requirement for the PC for cell entry in both epithelial and endothelial cells [63 , 64] . In addition to the current reported epithelial cell line , we have recently isolated two guinea pig placenta derived epithelial cell lines which have the same stringency of requirement for the GPCMV pentameric complex [103] . Importantly , studies on guinea pig leukocyte cells suggest that the PC is necessary for infection on these cell types , which is a similar requirement as seen for HCMV [93 , 104] . Potentially , in the Auerbach et al . study [90] , the endothelial cell population , which were isolated by flow cytometry using cross reacting antibodies , there was a low level of contaminating fibroblast cells , which would prevent a contrasting requirement for the pentameric complex for cellular infection . In our study , epithelial cell isolation was carried out by conventional cloning strategy and antibody staining of cells for cytokeratin marker . Additionally , the immortalized cell lines were re-cloned and characterized ( S3 Fig ) . In contrast , Auerbach et al . [90] , elected to employ RT-PCR analysis of total cell monolayer lysate to characterize their primary endothelial cells despite von Willebrand factor antibody being successfully used to identify guinea pig endothelial cells in previous studies [105] . Consequently , the uniformity of their endothelial cell line remains to be fully confirmed . Importantly , newly isolated epithelial cell lines from the guinea pig placenta demonstrate that GPCMV infection of specialized placental cells requires the PC [103] which further supports the hypothesis that increased congenital transmission is PC dependent . This also explains the increased congenital infection rate observed in virus encoding the full length GP128-133 locus [29 , 106] . Transient expression of the homolog components of the penatmeric complex ( gH , gL , UL129 , UL131 , UL133 ) in epithelial cells resulted in complex formation in the absence of other viral components and this confirmed the results by Auerbach et al . [90] which used purified recombinant baculovirus expressed protein to form pentameric complex in vitro . Importantly , a natural C-terminal deletion mutant of the GP129 protein ( codons 102–179 ) from a virus lacking epithelial tropism ( NRD13 ) was incapable to forming a pentameric complex with other complex components . The importance of C-terminal domain of GP129 for stable PC formation in recombinant GPCMV remains to be further evaluated . In this current report , PC mutant studies were limited to GP129 but the generation of mutants of other PC components ( eg . GP131 or GP133 ) would be worth evaluation in future studies to aid in the definition of critical domains for complex formation . In HCMV , the UL128 protein interaction with gH/gL is a potentially important key stage because of disulphide bond formation with gL [55] . In our GPCMV studies , we demonstrated that interactions with gL as well as gH is important for PC formation ( Fig 2 and S7 Fig ) . Additionally , gH/gL/GP129 triplex complexes occur ( Fig 4 ) , but disulphide bond formation was not investigated . Most certainly , conservation of cysteine amino acids means that the GP129 has the possibility of interacting with gL in the same specific manner . Importantly , the GP129 mutant was unable to form a triplex complex with gH/gL . Triplex complexes with gH/gL and GP131 or GP133 could also be demonstrated . Subcomplex formation has been demonstrated to occur for HCMV between various components of the PC [62] . An evaluation of the stoichiometry of these subcomplexes in various cell types might be important in the determination of their influence on the immune response of the host . The GPCMV PC is highly immunogenic but not individual unique components ( GP129 , GP131 or GP133 ) based on newly developed ELISA assays [103] . Consequently , the homolog PC would appear an important target antigen in GPCMV as is the case for the PC in HCMV . A fundamental question is what dictates formation of a gH/gL/gO triplex , or gH/gL/UL128-131 pentamer in HCMV , and the homolog complexes in GPCMV infected cells . In GPCMV , GP129 , GP131 and GP133 would have a theoretical 3:1 advantage over gO for interaction with gH/gL and the same would be the case for HCMV ( UL128 , UL130 and UL131 ) . In a recent model for HCMV pentamer complex formation , Ciferri et al [55] proposed that UL128 and gO each compete to form a disulphide bond with gL at codon 144 in an exclusive way in the gH/gL complex and that this is the most important interaction since UL130 and UL131 attach to gH/gL via non-covalent bonds . Alternatively , in HCMV an additional protein encoded in the ULb’ region ( UL148 ) has been proposed to regulate gH/gL/gO or gH/gL pentamer formation in an undefined manner [107] . Identification of a homolog UL148 might be worthwhile exploring in future GPCMV studies as would evaluation of the ability of GP129 to bind gH/gL competitively in the presence of gO . Curiously , gO in human and animal CMV is heavily N-glycosylated . In a previous publication [36] , we investigated the importance of the N-glycosylated or non-glycosylated gO protein in GPCMV gH/gL/gO triplex formation . Both wild type ( N-glycosylated ) gO and a gO mutant ( lacking N-glycosylation sites ) were equally capable of forming a triplex . Additionally , in GPL cells infected with GPCMV , the wild type gO could be detected as both glycosylated and non-glycosylated in equal amounts but in transient plasmid expression studies only the N-glycosylated gO could be detected [36] . Potentially , the gO glycosylation state could have ramifications for the steady state of gH/gL available to interact with pentameric complex components . A heavily N-glycosylated gO protein may have a more effective interaction with the endoplasmic reticulum ( ER ) calnexin chaperone protein system [108] . This in turn may enhance the movement of gO or gH/gL/gO complex through the ER resulting in more efficient virion maturation and egress . Consequently , if the glycosylation status of gO in fibroblast vs epithelial cells was different then potentially this might influence the ability of one complex to form over another and subsequently augment viral tropism by the variation of triplex vs pentameric complex on the outside of the viral particle . However , to more fully investigate these possibilities in GPCMV would require the development of a gO specific antibody for full evaluation of these effects . The animal pathogenicity studies demonstrated that a restoration of an ability of the virus to form a pentameric complex not only resulted in epithelial tropism but also increased viral pathogenicity as well as congenital transmission rate ( Fig 9 , Tables 1 and 2 ) . The ectopic expression of GP129 cDNA resulted in a virus with similar pathogenicity pattern to SG GPCMV and importantly the virus reached the salivary glands . In contrast , the parental GPCMV BAC derived virus ( NRD13 ) lacked viremia and poorly disseminated . Additionally , the GP129FRT virus had a congenital transmission rate of 38 . 4% compared to 3 . 2% for NRD13 . Congenital GPCMV challenge studies are normally carried out with virus stock generated by serial passage in the animal and virus generated as salivary gland stock . The SG GPCMV stock of wild type virus has a congenital transmission rate between 55–75% based on previous publications . Potentially , passage of GP129FRT in animals to generate a salivary gland stock might be enhance the congenital rate of the virus and these studies are currently underway . The basis for the viral attenuation occurring in fibroblast cells ( GP128-133 locus mutation ) is undetermined . Most certainly there appears to be a bias in fibroblast cells for the production of cell free virus encoding the triplex compared to virus that also encodes the pentameric complex . The spliced nature of the encoded GP129 and GP131 genes might be a contributing factor to a rate limiting step on expression kinetics . Alternatively , the gO glycosylation status in fibroblast cells compared to epithelial or other cell types might influence triplex ( gH/gL/gO ) formation . Overall , these factors might also be a basis for the rapid generation of GP128-133 locus mutants when SG GPCMV is serially passaged on fibroblast cells . However , this remains to be further investigated and may potentially require the development of additional epithelial and fibroblast cell lines to demonstrate that this phenomenon is not limited to the cell lines used in this study . In summary , this study demonstrated that the UL128-131 homolog locus ( GP129-GP133 ) is necessary for GPCMV to form a homolog PC . Additionally , loss by GPCMV of the capability to form the PC impaired epithelial cell tropism and attenuated the virus in the animal model . Importantly , restoration of GPCMV epithelial tropism resulted in increased pathogenicity and congenital transmission rates . Overall , these findings strengthen the guinea pig as a highly relevant model for HCMV congenital infection and the development of CMV vaccine or intervention strategies against congenital infection . Additionally , these studies emphasize the importance of the PC for congenital transmission and strongly suggest that a vaccine aimed at preventing congenital infection should incorporate the pentameric complex as part of a vaccine design .
GPCMV ( strain 22122 , ATCC VR682 ) , first and second generation GPCMV BAC [26 , 27] derived viruses were propagated on guinea pig fibroblast lung cells ( GPL; ATCC CCL 158 ) in F-12 medium supplemented with 10% fetal calf serum ( FCS , Life Technologies ) , 10 , 000 IU of penicillin/liter , 10 mg of streptomycin/liter ( Life Technologies ) , and 7 . 5% NaHCO3 ( Life Technologies ) at 37°C/5% CO2 . Virus titrations were carried out on six-well plates . Plaques were stained with 10% Giemsa stain or visualized by fluorescence microscopy . High titer stock viruses were generated as previously described [36] . Sucrose gradient purified stocks were initially generated as high titer stock virus and subsequently purified on a sucrose gradient following a previously described protocol [109] . Guinea pig epithelial cells were initially isolated as primary renal epithelial tubule cells from the kidney of a Hartley guinea pig by conventional clonal ring procedure . Primary epithelial cells were initially maintained on collagen ( Life Technologies ) coated plates in epithelial cell growth media ( Applied Technology ) supplemented with 10% FCS ( Life Technologies ) and antibiotics as described for F-12 media . Transformed epithelial cells were maintained in high glucose DMEM with sodium pyruvate ( Life Technologies ) supplemented with 5% FCS ( Life Technologies ) and antibiotics as described for F-12 media . Epithelial cells were maintained at 37°C/5% CO2 and characterized as epithelial cells by characteristic cobbled stone appearance of the cell monolayer as well as staining positive for cytokeratin marker ( anti-pancytokeratin antibody and anti-cytokeratin 18 antibody , Cell Signaling ) . Tissue culture stocks of GP129FRT and SG GPCMV were generated on epithelial cells . All oligonucleotides were synthesized by Sigma-Genosys ( The Woodlands , TX ) and are listed in S1 Table . Primary Hartley guinea pig renal epithelial cells were isolated following protocol by [110] . Subsequently , cells were transformed by transduction with defective lentiviruses encoding SV40 TAg or HPV E6/E7 ( Applied Biological Materials Inc . ) following manufacturer’s protocol . Cells were re-seeded onto collagen coated 100mm dishes and propagated in high glucose DMEM with sodium pyruvate ( Invitrogen ) supplemented with 10% fetal calf serum and 1X antibiotic-antimycotic ( Life Technologies ) and 37°C/5% CO2 . Rapidly growing colonies were isolated by clonal ring procedure and subcloned onto fresh monolayers of collagen coated plates and maintained as separate cell lines and aliquots of cells in cryopreserve media ( Gemini Bioproducts ) and stored in liquid nitrogen . Transformed cells were verified as epithelial by positive staining for cytokeratin marker by western blot or immunofluorescent staining using anti-pancytokeratin ( Cell Signaling Technology ) and anti-cytokeratin 19 ( Applied Technology ) using a previously described protocol [36] . SV40 T antigen transformed cells were poorly supportive of GPCMV infection and were not used in the reported studies . HPV E6/E7 transformed cells were supportive of GPCMV infection and used extensively in the studies . Guinea pig ( Hartley ) animal studies were carried out under approval by IACUC at Texas A&M University , permit 2013#013 . All study procedures were carried out in strict accordance with the recommendations in the “Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . ” Animals were observed daily by trained animal care staff , and animals requiring care were referred to the attending veterinarian for immediate care or euthanasia . Terminal euthanasia was carried out by lethal CO2 overdose followed by cervical dislocation in accordance with IACUC protocol and NIH guidelines . Animals purchased from Charles River Laboratories were verified as seronegative for GPCMV by toe nail clip bleed and anti-GPCMV ELISA of sera as previously described [36] . Blood and tissues ( lung , liver , spleen ) were collected from euthanized guinea pigs to determine the viral load . For pups from congenital infection studies , blood and tissues ( lung , liver , spleen , brain ) were collected within 3 days post birth . Pup specific placenta was collected and preserved for DNA extraction when applicable . For tissue DNA extraction , FastPrep 24 ( MP Biomedical ) was used to homogenize tissues as a 20% weight/volume homogenate in Lysing Matrix D ( MP Biomedicals ) . To obtain DNA from whole blood , 500ul of blood was collected ( by toe nail clip bleed ) into tubes containing ACD anticoagulant and 200μl of blood was subsequently used per extraction . DNA was extracted using the QIAxtractor ( Qiagen ) according to manufacturer’s liquid ( blood ) or tissue protocol instructions . Viral load was determined by real time PCR on Lightcycler 480 ( Roche Applied Science ) . Primers and hydrolysis probe were designed using the Lightcycler Probe Design2 program to amplify a product from the GPCMV GP44 gene: Forward primer 5’TCTCCACGGTGAAAGAGTTGT; Reverse primer 5’GTGCTGTCGGACCACGATA; hydrolysis probe 5’FAM-TCTTGCTCTGCAGGTGGACGA-BHQ1 . PCR master mix contained Lightcycler Probes Master ( Roche Life Science ) , 0 . 4 μM primers and 0 . 1 μM probe , 0 . 4U uracil N-glycosylase ( UNG ) in 25μl total reaction volume including 10 μl of DNA per reaction . Standard controls and no template controls ( NTC ) were run with each assay for quantification . Lightcycler480 amplification parameters were: UNG step for 10 minutes at 40°C followed by activation at 95°C for 10 minutes , then 45 cycles of denaturation at 95°C for 15s , annealing at 56°C for 15s , elongation at 72°C for 10s . Data was collected by ‘single’ acquisition during the extension step . Standard curve was generated using GPCMV GP44 plasmid [33] for quantification and assay sensitivity . The sensitivity of the assay was determined to be 5 copies /reaction . Viral load was expressed as copy number/ml of blood or copy number/mg tissue . Results calculated were a mean value of triplicate PCR runs per sample . The predicted GPCMV coding sequences were based on the complete 22122 viral genome sequence ( Genbank accession #AB592928 . 1 ) . The specific gene coding sequence co-ordinates are: GP74 , gO , ( 117 , 992–119 , 104 ) ; GP75 , gH , ( 119 , 553–121 , 724 ) ; GP115 , gL , ( 180 , 216–180 , 992 ) ; GP128 ( 195 , 713–196 , 768 ) ; GP129 complement ( 196 , 745–197 , 003; 197 , 081–197 , 206; 197 , 285–197 , 439 ) ; GP131 complement ( 197 , 444–197 , 780; 197 , 861–198 , 102 ) ; GP133 complement ( 198 , 102–198485 ) and GP130 ( 196 , 968–197360 ) . Generation of individual shuttle vectors for specific gene knockout ( or intergenic insertion of GP129 ) and construction of transient expression vectors are described in more detail below . An inducible ET recombination system ( GeneBridges ) was introduced into DH10B bacterial cells containing a second generation GPCMV BAC plasmid [26 , 27] using a protocol previously described [32] . Individual GPCMV gene knockout targeting shuttle vectors were linearized with a unique restriction enzyme cutting outside the target gene flanking sequence . Linearized DNA plasmids or PCR products were band isolated and concentrations of DNA were modified to introduce 1μg of linear DNA into each transformation reaction via electroporation [32] . Recombinant bacterial colonies of GPCMV BAC mutants were isolated by chloramphenicol ( 12 . 5 μg/ml ) and kanamycin ( 20 μg/ml ) antibiotic selection in LB agar bacterial Petri dishes . Bacterial plates were initially incubated at 39°C to remove the ts ET recombination plasmid ( Genebridges ) . Mutant GPCMV BAC DNA was purified by maxiprep kit ( Qiagen ) and analyzed by separate EcoR I and Hind III restriction digestions to verify the accuracy of the predicted genome configuration after mutation [26 , 27] . Insertion of the kanamycin ( Km ) drug resistance cassette into the viral genome introduced a novel Hind III restriction enzyme site at the site of mutation to enable verification of locus modification . Specific gene modifications were confirmed by comparative PCR analysis between wild type and mutant GPCMV BACs using common flanking primers . The gene knockout for mutants was further verified by sequencing of the PCR product . In order to enable a second round of GPCMV BAC mutagenesis , the original Km cassette inserted into the genome was removed by FLP recombinase strategy if the Km cassette was flanked by FRT sites . The FLP recombination was accomplished by transforming the BAC positive bacteria with a FLP expression suicide plasmid ( p707 , GeneBridges ) ( permissive conditions with tetracycline ( 3ug/ml ) at 31°C ) and recombinase induction and excision of the FRTKm cassette accomplished following manufacturer’s protocol . A second round of recombination could then be carried out on the GPCMV BAC as described above . GPCMV genes GP128 , GP129 , GP131 and GP133 were individually knocked out by targeted mutagenesis of the GPCMV BAC in bacteria using shuttle vectors carrying a Km drug resistance marker to disrupt each ORF . Targeted recombination knockout of GPCMV genes was performed in the second generation GPCMV BAC [27] . Fig 1 shows the layout of the GP128-133 locus . S1 Fig shows the annotated nucleotide sequence of the GP128-133 locus with the location of the genes shaded and the specific deletions introduced for GP129-GP131 or GP133 mutants indicated . The GP129/GP131 mutant was generated with a BamHI FRT Km insertion between flanking sequence on the synthetic deletion shuttle pSYDGP129/131 , which deleted GPCMV nucleotides 197 , 292–198 , 090 ( 798 bp deletion within the GP129-GP131 coding sequence , see S1 Fig ) . The GP133 mutant was generated by a BamH I FRT Km insertion between flanking sequence on the synthetic deletion shuttle vector pSYDGP133 , which deleted GPCMV nucleotides 198 , 361–198 , 489 ( 138 base deletion which included the GP133 start codon ( S1 Fig ) . The GP128 gene was modified by insertion of a EcoR V FRT Km cassette at a unique EcoR V site ( GPCMV nucleotide 196 , 234 ) in the GP128 gene of the GP128 shuttle vector ( pGP128KmFRTEcV ) , which disrupted the GP128 ORF at codon 176 . The mutant GPCMV BACs were analyzed by restriction enzyme profile analysis as previously described [36] . Insertion of the Km drug resistance cassette into the viral genome introduced a novel Hind III restriction enzyme site at the site of mutation to enable verification of locus modification . Modified GPCMV genomes were analyzed separately by EcoR I and Hind III restriction enzyme profile analysis . In an effort to limit redundancy the profiles shown for each mutant are either Hind III or EcoR I analysis . Additionally , two clonal mutants were generated for each knockout but only one is described . Comparative restriction fragment profiles of wild type and mutant GPCMV BAC genomes correctly demonstrated specific sub-genomic fragment modification for all mutants . Designated GPCMV restriction fragment band nomenclature described by Gao and Isom [111] was used to identify specific band shifts , except that the 5’ and 3’ genome terminal ends were considered linked in a covalent closed circle [26] . S12 Fig shows the GP128 mutant EcoR I profile and Hind III profiles for GP129-131 and GP133 mutants compared to the wt GPCMV BAC profiles . Specific gene locus modifications were further verified by PCR analysis and sequencing as previously described [36] . The GP128 gene encoded in the 4 . 9 kb EcoR I GPCMV genomic fragment ( 192 , 215–197 , 167 ) was modified in the GP128 mutant by insertion of the 1 . 1 kb Km cassette which shifted the fragment to 6 kb ( S12 Fig ) . The GP129 , GP131 and GP133 genes are encoded in the 19 . 7 kb Hind III GPCMV genomic fragment and targeted knockout of these genes introduced a new Hind III site encoded in the inserted Km marker . The GP129-GP131 deletion mutant Hind III fragment was modified from 19 . 7 kb to 5 kb and approximately 15 kb ( S12 Fig ) . In the GP133 deletion mutant the 19 . 7 kb Hind III GPCMV genomic fragment was modified to 6 kb and 14 . 9 kb fragments ( S12 Fig ) . Mutant GPCMV BACs were independently subject to an additional round of mutagenesis to introduce a wild type GP129 ( myc tagged ) cDNA into the GP25/26 intergenic locus under SV40 promoter control as described in an earlier section for ectopic expression of GP129 . This required excision of the Km FRT cassette from the originally mutated locus by Flp recombinase as described ( see above section ) . Excision of the Km cassette was confirmed by patching of colonies for the loss antibiotic resistance and the integrity of the GPCMV BAC confirmed by restriction profile analysis ( S12 Fig ) . GPCMV BACs retained a FRT site at the original site of excision . E . coli ( DH10B ) cells carrying respective GPCMV BAC mutants underwent ET recombination induction and targeted modification with the shuttle vector pGP129limkKmFRT as previously described . GP25/GP26 locus mutants ( carrying GP129myc cDNA ) were isolated by Km marker insertion as previously described . Full length GPCMV BAC clones were identified by EcoR I restriction profile analysis ( S12 Fig ) and subsequently confirmed by PCR of the GP25/GP26 locus and sequencing ( S11 Fig ) . Mutants carrying GP129 cDNA in the GP25/GP26 locus were designated: ( 1 ) GP129FRT ( wt GPCMV BAC ) ; ( 2 ) GP128FRT/GP129Link ( GP128 mutant ) ; ( 3 ) GP129-GP131FRT/ GP129Link ( GP129-GP131 deletion mutant ) ; ( 4 ) GP133FRT/GP129Link ( GP133 deletion mutant ) . In regard to the ectopic insertion of a GP129 cDNA into the GP25/GP26 intergenic locus , the modification generated a characteristic altered GPCMV EcoR I profile . The 5 . 2 kb GPCMV EcoR I genomic fragment ( nucleotides 35 , 537–40 , 739 ) containing the GP25/GP26 locus was modified by the insertion of the SV40 promoter /GP129 expression cassette/SV40 polyA sequence and Km marker into the Bam H I site ( nucleotide 38 , 538 ) . The modified sequence also introduced two novel EcoR I sites . As predicted the original 5 . 2 kb genomic fragment was modified to two novel fragments ( 3 . 5 and 3 . 9 kb ) . S12 Fig shows the modified profiles for wt GPCMV , GP128 and GP129-131 mutants ( GP128FRT/GP129Link and GP129-GP131FRT/ GP129Link respectively ) but not the GP133 mutant to limit redundancy . An additional GPCMV BAC mutant was engineered into GP129FRT GPCMV BAC which introduced a Km cassette into the GP74 gene ( glycoprotein gO ) as previously described [36] . This generated a GPCMV BAC mutant that encoded a GP129 cDNA in the GP25/26 locus ( Km replaced by FRT sites only ) and a GP74 knockout by insertion of a Km cassette which disrupted the ORF at codon 110 [36] ( double mutant was designated GP129FRT/GP74Km ) . The GP74 mutation was confirmed by EcoR I profile analysis of the GPCMV BAC genome . The original 18 kb EcoR I genomic fragment ( nucleotides 102 , 796–120 , 821 ) that encoded the GP74 locus was increased in size by 1 . 1 kb by the insertion of a Km cassette which modified the size of the fragment to 19 . 1 kb , see S12 Fig . The gB ( GP55 ) knockout mutant was generated on the backdrop of GP129FRT GPCMV BAC . The GP55 gene ( 94 , 164–96 , 869 ) was knocked out by insertion of a Km cassette which disrupted the ORF at codon 528 as previously described [36] . The insertion of the Km cassette modified the size of the 4470 EcoR I subgenomic fragment ( 91 , 190–95 , 659 ) by approximately 1 kb generating a novel EcoR I restriction fragment of approximately 5 . 5 kb on the GP129FRT/GP55km GPCMV BAC EcoR I profile . The positions of the wild type and mutant EcoR I restriction fragments are indicated in S12 ( vii ) Fig . For generation of recombinant viruses , large-scale GPCMV BAC DNA was purified from E . coli DH10B strain using a maxi plasmid kit ( Qiagen ) . BAC DNA was transfected onto GPL cells in six well dishes using Lipofectamine 2000 ( Invitrogen ) as previously described [33] . GPCMV BAC transfections were carried out with two independent clones for each gene knockout . Transfections were followed for at least 3–4 weeks for the production of viral plaques . GFP positive viral plaques were detected via microscopy [33] . Non-infectious mutants produced only single GFP positive cells that did not progress to viral plaques . GPCMV mutant BAC transfections were carried out multiple times ( minimum of 6 times ) for each clone . Excision of the BAC plasmid from recombinant viral genome was carried out by co-transfection of BAC DNA with plasmid encoding CRE recombinase ( pCRE ) , a generous gift from Dr . Mike McVoy ( Virginia Commonwealth University ) . Cre BAC excised virus also lost the GFP reporter cassette encoded on the BAC plasmid and therefore GFP negative plaques confirmed successful BAC plasmid excision . Large scale virus stocks were generated as previously described [33] and additionally epitropic viruses were generated as virus stocks on epithelial cells following the same procedure . A gB knockout ( GP129FRT/GP55km ) rescue virus was generated by co-transfection of GP129FRT/GP55km BAC with GP55 rescue plasmid as previously described [36] . Recombinant defective adenoviruses ( serotype 5 ) encoding individual epitope tagged components of the pentameric complex were generated as high titer stocks by Welgen Inc . on HEK293 cells . The C-terminal epitopes tagged ORFs from plasmids pAcGFPNgH , pmCherryNgL , pGP129myc , pGP131HA , pGP133FLAG were each placed under HCMV MIE enhancer promoter control in the E1 locus of the defective Ad vectors using a E1 shuttle vector ( Welgen Inc . ) to generate recombinant defective adenoviruses designated AdgHGFP , AdgLmCherry , AdGP129myc , AdGP131HA , AdGP133FLAG respectively . A defective Ad vector encoding GFP ( AdGFP ) was also used in control expression studies [36] . Time point samples were taken from wild type GPCMV infected GPL cells in a six well dish ( moi = 1 pfu/cell ) at 48 hr post infection . RT-PCR was performed essentially as described in Coleman et al . [36] . Based on the original analysis of the full length GP128-133 locus [28 , 87] , there are five genes designated GP128 , GP129 , GP130 , GP131 and GP133 . Based on co-linearity to HCMV as well as encoded proteins , two of these genes are direct homologs of HCMV UL128 ( GP129 ) and UL130 ( GP131 ) [28] . GP133 has a weak homology to UL131[90] . Gene expression in the GP128-133 locus at late stage infection of SG GPCMV was investigated via RT-PCR assay as previously described [36] using the following primer pairs: RTGP128F/RTGP128R; RTGP129F/RTGP129R; RTGP130F/RTGP130R; RTGP131F/RTGP131R; RTGP133F/RTGP133R and control GAPDH ( GAPDHRTF/GAPDHRTR ) as described in S1 Table . Results demonstrated that the previously identified genes were all expressed in SG GPCMV infected cells ( S2 Fig ) . Immunoprecipitation ( IP ) assays were carried out on plasmid transfected or recombinant Ad transduced fibroblast cells using commercial GFP-trap reagent ( ChromoTek ) or RFP-trap ( ChromoTek ) following manufacturer’s protocols and inclusion of protease inhibitor cocktail ( Pierce ) in cell lysates . Samples were subsequently analyzed by SDS-PAGE ( 4–20% gradient gel ) and western blot using specific anti-epitope tag antibodies: HA ( Novus Biologicals ) ; FLAG ( Novus Biological ) ; GFP ( Santa Cruz Biotechnology ) ; Myc-c ( Novus Biologicals ) ; and mCherry ( Clontech Laboratories ) . Appropriate secondary anti-mouse or anti-rabbit HRP conjugate ( Cell Signaling Technology ) were also used following standard western blot protocol as previously described [36] . Custom antibodies to GPCMV gH , IE2 and GP131 were generated by Genescript to specific peptide sequences ( rabbit polyclonal ) or to purified recombinant ( E . coli ) protein ( mouse monoclonal ) . Rabbit polyclonal antibody sera were separately generated against gH and IE2 using the peptide immunogen of RTDLSSPTEELTSP ( for gH ) or CRKTRPAKRPRSNDE ( for IE2 ) . Mouse monoclonal antibody to GP131 was generated against recombinant purified protein and hybridoma antibody screened for activity to GP131 . Rabbit or mouse IgG was column purified and resuspended at a concentration of 100 μg/ml . Antibody specificity was verified by both western blot and immunofluorescence assay of transiently expressed proteins ( gH , IE2 or GP131 ) in fibroblast/epithelial cells using appropriate plasmid expression vectors for gH , IE2 or GP131 following standard protocols [36] . Epithelial cells or GPL fibroblast cells on coverslips six well dishes were pretreated with complete media containing either 0 or 50nM bafilomycin A1 ( Sigma ) for 1 hr at 37˚C followed by virus infection using FRTGP129 virus ( MOI = 1pfu/cell ) for 1hr at 37 ˚C . All further incubations were performed at the same concentration as pretreatment in complete media . Cells were fixed at 24 hr post infection in 100% methanol at -20˚C and immunostained for IE2 protein and cytokeratin as described in materials and methods . Counts were made of IE2 positive cells in random fields . Statistical analysis was performed using student T-test on the percent of cells infected in thirty random fields of view , each contained ~100 cell nuclei , for each condition . The number of treated cells infected was represented as a percentage of the number of infected untreated cells . Results shown in S15 Fig . In pathogenicity studies , viral load in similar organ tissue from GPCMV infected animals at specific days post infection were compared by Student t-test ( GP129FRT vs SG GPCMV and GP129FRT vs NRD13 ) . In congenital studies , pup outcome and transmission rates were compared by Fisher’s exact test . GPCMV viral load in specific target organs of pups were compared by Student t-test . All comparisons were two-tailed . Student t-test analysis was also carried out for virus infection of cells in bafilomycin virus cell entry studies ( S15 Fig ) . | Congenital CMV is a leading cause of mental retardation and deafness in newborns . An effective vaccine against congenital CMV remains an elusive goal . HCMV encodes a pentameric glycoprotein complex ( PC ) necessary for tropism to epithelial , endothelial and myeloid cells . Given the structure of the placenta , the viral PC is considered important for congenital infection and potentially an important neutralizing antibody vaccine target antigen . The guinea pig , with a placenta structure similar to humans , is the only small animal model for congenital CMV . In this paper , GPCMV is shown to encode a homolog PC which enables epithelial tropism on a newly established cell line . It is likely that the GPCMV PC improves virus tropism to various cell types as PC positive virus has improved virus pathogenicity and congenital infection in vivo . This study lays important foundations for development of a PC based intervention strategy against congenital CMV in this model . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"microbiology",
"vertebrates",
"animals",
"mammals",
"fibroblasts",
"epithelial",
"cells",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"immunoprecipitation",
"connective",
"tissue",
"cells",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"embryology",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"cells",
"placenta",
"connective",
"tissue",
"biological",
"tissue",
"molecular",
"biology",
"genetic",
"loci",
"precipitation",
"techniques",
"guinea",
"pigs",
"rodents",
"viral",
"tropism",
"cell",
"biology",
"anatomy",
"polymerase",
"chain",
"reaction",
"virology",
"genetics",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"amniotes",
"organisms"
] | 2016 | A Homolog Pentameric Complex Dictates Viral Epithelial Tropism, Pathogenicity and Congenital Infection Rate in Guinea Pig Cytomegalovirus |
Correlated changes of nucleic or amino acids have provided strong information about the structures and interactions of molecules . Despite the rich literature in coevolutionary sequence analysis , previous methods often have to trade off between generality , simplicity , phylogenetic information , and specific knowledge about interactions . Furthermore , despite the evidence of coevolution in selected protein families , a comprehensive screening of coevolution among all protein domains is still lacking . We propose an augmented continuous-time Markov process model for sequence coevolution . The model can handle different types of interactions , incorporate phylogenetic information and sequence substitution , has only one extra free parameter , and requires no knowledge about interaction rules . We employ this model to large-scale screenings on the entire protein domain database ( Pfam ) . Strikingly , with 0 . 1 trillion tests executed , the majority of the inferred coevolving protein domains are functionally related , and the coevolving amino acid residues are spatially coupled . Moreover , many of the coevolving positions are located at functionally important sites of proteins/protein complexes , such as the subunit linkers of superoxide dismutase , the tRNA binding sites of ribosomes , the DNA binding region of RNA polymerase , and the active and ligand binding sites of various enzymes . The results suggest sequence coevolution manifests structural and functional constraints of proteins . The intricate relations between sequence coevolution and various selective constraints are worth pursuing at a deeper level .
Coevolution is prevalent at species , organismic , and molecular levels . At the molecular level , selective constraints operate on the entire system , which often require coordinated changes of its components . The most well-known example is the compensatory substitution of nucleic acid pairs in RNA secondary structures [1–6] . Interacting nucleotides vary between AU , CG , and GU pairs in different species in order to maintain the hydrogen bonds . Coordinated changes of amino acid residues have also been investigated . Typically these studies acquired one ( or two ) family ( ies ) of aligned sequences and examined covariation between aligned positions or of the entire sequences . Some of these have applied different covariation metrics including correlation coefficients [7–8] , mutual information [9–13] , and the deviance between marginal and conditional distributions [14] . These studies demonstrate that sequence covariation is powerful in detecting protein–protein interactions [7 , 12] , ligand-receptor bindings [7 , 12] , and the folding structure of single proteins [10 , 13] . In addition to direct physical interactions , distant coevolving amino acid residues are reported to be energetically coupled [14] or subject to the functional constraints of the proteins [8] . A major drawback of many covariation metrics is the lack of phylogenetic information . The sequences manifesting the same level of covariation may arise from either a few independent substitutions in early ancestors or correlated changes along multiple lineages [15 , 16] . In RNA structure prediction , many authors have thereby extended the continuous-time Markov process ( CTMP ) of sequence substitution [17] to coevolving nucleic acid pairs [3 , 4 , 6 , 18] . However , direct application of these models to protein coevolution is intractable due to the large number of parameters ( a 400 × 400 matrix ) in the CTMP of amino acid pairs . This problem was addressed by replacing amino acids in a CTMP with simplified , surrogate alphabet sets such as the presence/absence of a protein in each species [16] or the charge and size of amino acid groups [19] . Yet this simplification deviates from the standard CTMP of sequence substitution , in which a rich set of empirical models are available . All the previous studies of detecting protein coevolution target a few proteins or protein domains , such as myoglobin [19] , PGK [7] , Ntr family [12] , PDZ domain family [14] , Gag , Hsp90 , and GroEL proteins [8] . The availability of large-scale protein sequences and their phylogenetic information allows us to perform a systematic screening on all the known protein families . Such large-scale screening will give comprehensive information of coevolution among all the protein domains and provide insight about their physical/functional couplings . We propose a general coevolutionary CTMP model which requires neither simplification of states nor prior knowledge about interactions , and has only one extra free parameter . Sequence substitution of the two sites is modeled by a continuous-time Markov process . The null ( independent ) model hypothesizes that two sites evolve independently . The alternative ( coevolutionary ) model is obtained from the null model by reweighting the independent substitution rate matrix to favor double over single changes . We apply this model to all the inter- and intra-domain position pairs in all the known protein domain families in Pfam database [20] . Strikingly , from a large number of pairwise comparisons the coevolving domain pairs are highly enriched with domains in the same proteins , protein complexes , or possessing the same functions . Moreover , the coevolving positions demonstrate a tendency of spatial coupling and are mapped to functionally important sites of their proteins .
We extend the CTMP sequence substitution to model coevolution of amino acid position pairs . The state transitions of a CTMP at an infinitesimal time interval follow a matrix differential equation ( Equation 1 ) . The instantaneous transition rates are specified by a 20 × 20 substitution rate matrix Q . A CTMP of an amino acid pair is obtained by concatenating the sequence states of two amino acid positions . The substitution rate matrix of two independent amino acid positions can be directly derived from the CTMP of single sites . However , the rate matrix of a general two-component CTMP has much fewer constraints and a larger dimension ( 400 × 400 ) . We simplify the substitution rate matrix by penalizing all the entries of single changes and rewarding all the entries of double changes with the same weight factors . This coevolutionary model introduces very few extra free parameters , thus it is easy to learn and less vulnerable to overfitting . By applying this general coevolutionary model to RNA sequences , we successfully predicted RNA secondary and tertiary interactions [21] . Figure 1 illustrates the procedures of evaluating the coevolutionary likelihood scores . Given the aligned sequences of two positions in different ( or identical ) protein domains , their joint phylogenetic tree , and the joint substitution rate matrix , we can calculate the marginal likelihood of the observed sequences on the leaves by summing over the sequence states of internal nodes . The level of fitness of the coevolutionary model to the data is measured by the log-likelihood ratio between the coevolutionary and independent models . For each pair of positions in the two families of aligned sequences ( or one family of sequences against itself ) , we can calculate their log-likelihood ratio and mark putative coevolving position pairs . Very often there are multiple coevolving positions between two domains ( or within one single domain ) . To assess the likelihood score of the entire domain pair , we employ a probabilistic graphical model with variables corresponding to specific positions of the protein domains in an ancestral or contemporary species . Using a spanning tree approximation , we evaluate the joint likelihood score in terms of the pairwise and singlet likelihoods ( Equation 5 ) . The method of assessing the likelihood score of multiple coevolving pairs is novel and does not appear in our previous work [21] . Details about the coevolutionary models of position pairs and the entire domain pairs are described in Materials and Methods . The entire Pfam database of aligned protein domain sequences was downloaded [20] ( April 2006 version ) . Overall the dataset contained 8 , 183 domain families . The automatically generated “full alignment'” of each domain family was chosen in order to maximize the coverage and number of sequences in the data . The topology and branch length of the phylogenetic tree for each domain family were also downloaded from Pfam . We considered the 3 , 722 , 468 domain family pairs ( 12% of all family pairs ) which co-appeared in no less than 20 species . Out of the 3 , 722 , 468 domain family pairs , 179 , 117 ( 4 . 81% ) co-appear in the same proteins or share the same GO annotations ( bottom level in the GO hierarchy ) in more than half of the member proteins that have GO annotations . Among each domain family pair , we considered all position pairs . In total there were 1 . 171 × 1011 all-versus-all inter-domain position pairs . We calculated the log-odds score for each position pair in each of the 3 , 722 , 468 domain pairs . We set the threshold of the log-odds scores to be 9 . 0 according to p-values of random CTMP simulation , false discovery rates of multiple hypotheses testing , and functionally coupled domain pairs inferred by the model . First , by randomly simulating 1 million sequences using CTMP ( see Materials and Methods ) we found the p-value for log likelihood ratio 9 . 0 is less than 6 . 0 × 10−5 . Second , by randomly sampling sequences from the 3 , 543 , 351 functionally unrelated family pairs ( see Materials and Methods ) , we plotted the dependence of false discovery rates and log-odds thresholds ( Figure S1 ) . Threshold 9 . 0 yielded the false discovery rate 33 . 00% . Third , when determining the threshold , there was a tradeoff between the number of functionally related domain pairs and the fraction of these “true positives” among all the positive calls ( Figure S2 ) . With threshold 9 . 0 the true positive rate was about 45% . In addition , the results of functional and spatial coupling in the subsequent sections are robust against the choice of threshold ≥9 . 0 . For instance , the top 100 coevolving domain pairs ( Text S1 ) and the distance distribution of inter-domain coevolving position pairs ( Figure 2 ) remain unchanged when the threshold increases to 17 . 0 . With a threshold 9 . 0 , we obtained 3 , 953 position pairs distributed over 582 domain family pairs . We then ranked the 582 inferred domain pairs according to the log-odds scores of the joint model for multiple coevolving positions . The sorted coevolving domain family pairs , their coevolving positions , and the log-odds scores are reported in Text S1 . The coevolving protein domains are highly enriched with functionally coupled domain pairs . Of the 582 domain family pairs ( 44 . 16% ) , 257 share proteins or bottom-level GO annotations in more than half of their members . The enrichment of functionally coupled domain pairs is more than a 9-fold increase compared to the entire dataset ( 4 . 81% ) . The hypergeometric p-value for acquiring ≥257 functionally coupled domain family pairs by randomly choosing 582 domain family pairs is less than 2 . 22 × 10−174 ( see Materials and Methods ) . The functional coupling of the domains , however , may be a trivial consequence of many co-occurring species . To exclude this possibility , we considered the family pairs that overlapped in more than 200 species . The hypergeometric p-value for enrichment is less than 7 . 04 × 10−45 , allowing the null hypothesis of co-occurring species to be rejected . Furthermore , 85 out of the top 100 coevolving domain family pairs are functionally coupled . The enrichment of functionally coupled domains suggests that covariation at multiple positions is a strong indicator for functional coupling . Table 1 lists the functional categorization of coevolving domain families that are functionally coupled among the 582 inferred pairs . The majority of the domain pairs appear in the same proteins or protein complexes , whereas only a small fraction of them ( 26 out of 257 ) are in the same functional pathways . Coevolving domains primarily appear in a few classes of proteins: ribosomal proteins , RNA polymerase , metabolic enzymes , translational apparatus , bacteria conjugal transfer proteins , and virus proteins . Most of these proteins are universally essential from bacteria to human . Proteins which exhibit substantial variability , such as transcription factors , signaling proteins , and receptors are under-represented . Sequence covariation without phylogenetic information can be captured by mutual information . To demonstrate the importance of phylogenetic information , we applied the same inter-domain large-scale screening using pairwise mutual information ( see Materials and Methods ) . We counted the number of inferred domain family pairs that were functionally coupled ( true positives ) or not ( false positives ) . Figure 3 shows the Receiver Operating Characteristic curves of coevolutionary and mutual information scores . The results indicate the coevolutionary model consistently outperforms the mutual information score in identifying functionally related domains . With 582 inferred domain pairs , coevolutionary scores identified 257 functionally coupled domain pairs , whereas mutual information only identified 161 functionally coupled domain pairs . In addition , the top 100 domain pairs inferred by mutual information contained only 40 functionally coupled pairs compared to 85 for coevolutionary scores . Besides functionally coupling coevolving domains , a natural question is whether the coevolving amino acids are also spatially coupled . Of the 582 coevolving domain family pairs , 156 contain the domain pairs co-crystalized in the same proteins or protein complexes . We extracted the 196 protein/protein complex structures of the 156 coevolving domain family pairs from the Protein Data Bank [22] and mapped the coevolving positions to the amino acid residues in their PDB structures ( see Materials and Methods ) . Figure 2 shows the distance distribution between the 4 , 849 coevolving position pairs and the background distance distribution of all 6 , 072 , 873 position pairs between the two domains in the same PDB structures . Clearly , coevolving position distance ( solid blue ) tends to be shorter and more narrowly distributed compared to the background distribution ( dotted red ) . The p-value of the Kolmogorov-Smirnov test is <2 × 10−16 . The significant difference of distance distributions suggests coevolving positions are spatially coupled . The distances of all coevolving positions in the PDB structures are reported in Text S1 . A remarkable example of the spatially coupled coevolving pair is between position 157 of the alpha-hairpin domain ( accession number PF00081 ) and position 61 of the C-terminal domain ( accession number PF02777 ) in iron/manganese superoxide dismutase . This domain pair ranks 82nd on the list ( see Text S1 ) . The amino acids at positions PF00081–157/PF02777–61 exhibit strong covariation between NF and FQ ( N: asparagine , F: phenylaninine , Q: glutamine , see Figure S3 ) . Strikingly , the distances between the two positions in 13 out of the 14 homologous proteins are less than 4Å , suggesting their physical interactions . Figure 4 shows the structures of superoxide dismutase proteins in cyanobacteria ( Anabaena sp . , PDB id 1gv3 , [23] ) , and human ( PDB id 1ap5 , [24] ) and marks the coevolving amino acid residues . Figure 4 was generated by PyMOL . The two coevolving position pairs ( identical in sequence ) link the two subunits of the homo-tetramer . Between cyanobacteria ( NF ) and human ( FQ ) , phenylaninine ( F ) is swapped from the C-terminal domain to the alpha-hairpin domain , and asparagine ( N ) is replaced by glutamine ( Q ) in the same amino acid group . Hence , compensatory substitution between NF and FQ is likely to occur . Unlike PF00081–157/PF02777–61 , the majority of the coevolving positions are not in direct contact: only 4 . 2% ( 203 out of 4 , 849 ) coevolving position pairs are less than 8Å apart . Sequence covariation tends to occur between multiple distant sites of two domains . In large proteins or protein complexes constituting multiple domains ( e . g . , RNA polymerase or ribosome ) , sequence covariation between positions from multiple domains also occurs . This multi-way covariation reflects the structural or functional constraints beyond direct pairwise interactions such as hydrogen bonds . Table S1 gives examples of the multiple coevolving positions and domains . The spatially distant coevolving positions may reflect certain structural or functional constraints of the entire proteins/protein complexes ( e . g . , [8 , 14 , 25] ) . To verify the functional importance of coevolving positions , we examined the coevolving positions from 25 proteins or protein complexes that were derived from the top 100 family pairs and had known structures . Intriguingly , the functionally important sites in about half of these proteins/protein complexes examined ( 13 out of 25 ) either overlapped or were near ( ≤10 Å ) coevolving positions . Table 2 shows the functional sites near or located at the coevolving positions in the 13 proteins . We use four examples to illustrate the spatial relations between inter-domain coevolving positions and functional sites of proteins . There are 43 coevolving positions from ten protein domains in the 30S ribosomal subunit . Ribosomes synthesize proteins by binding tRNAs at three sites: the P ( donor ) site , the A ( acceptor ) site , and the E ( exit ) site . Figure 5 marks the coevolving amino acid residues ( colored spheres ) and the 16S rRNA nucleotides of the tRNA binding sites ( colored ribbons ) in Thermus thermophilus 30S ribosomal subunit ( [26] , PDB id 1fjg ) . Each tRNA binding site is close to some coevolving amino acid residues . Specifically , the S9 portion of the P site , the S12 portion of the A site , and the S7 , S11 portion of the E site partially coincide with the coevolving positions . There are 151 coevolving positions from ten protein domains in RNA polymerase . Figure S4 marks the coevolving positions in yeast RNA pol II ( [27] , PDB id 1i3q ) . These positions are located at the inner core of the macromolecule surrounding the cleft . This region directly binds to DNA ( Figure 10 in [27] ) and is structurally homologous between eukaryotes RNA Pol II and bacterial RNA polymerase ( Figure 12 in [27] ) . There are eight coevolving positions from two protein domains in phosphoglucomutase , an enzyme that transfers the phosphoryl group of glucose or mannose from position 6 to position 1 . Figure 6 marks the coevolving positions , active sites , and ligand binding sites in Pseudomonas aeruginosa phosphoglucomutase ( [28] , PDB id 1k2y ) . All except one of the coevolving position pairs are close in protein structure . Moreover , both coevolving positions and functionally important sites are located at the crevice of the heart-shaped enzyme . Functional sites including the active site , the sugar binding loop , the metal binding loop , and the phosphate binding site are all close to the coevolving positions . There are 16 coevolving positions from two proteins in aspartate/ornithine carbamoyltransferase , an enzyme of the amino acid synthesis pathway [29] . Six out of 16 coevolving positions are close in at least three out of seven homologous protein structures . Specifically , positions 508 in the Asp/Orn binding domain and 346 in the carbamoyl-P binding domain are in contact ( distance ≤4 Å ) in all seven proteins . Figure S5 marks the coevolving positions and the active site in human enzyme ( [29] , PDB id 1c9y ) . Coevolving positions partially overlap with the active binding sites . Other functional sites overlapped with , or close to coevolving positions , include ADP binding sites in carbamoyl-phosphate synthase [30]; Mg2+/pyruvate and nucleotide binding sites of PEP utilizing enzyme [31]; NAD , GLU binding sites , and active site of glutamate/leucine/phenylalanine/valine dehydrogenase [32]; nucleotide and sodarin binding sites of elongation factor [33]; active site of aspartate/ornithine carbamoyltransferase [34]; active site of malic enzyme [35]; NADH binding site of S-adenosyl-L-homocysteine hydrolase [36]; GDP-mannuronic acid binding site of UDP-glucose/GDP-mannose dehydrogenase [37]; and mannitol and NADH binding sites of mannitol dehydrogenase [38] . The annotations of the coevolving sites on the PDB structures of all 25 protein families are given in Text S2 . Each protein domain family has a different phylogenetic tree due to its distinct history of duplication and deletion . The coevolutionary model , however , requires a joint phylogenetic tree of the two families . To calculate the likelihood score , we have to extract a common subtree of the two phylogenetic trees that correspond to the coevolving part along the lineages of the two families . This problem is difficult due to the huge number of possible choices . A common approach to compare two distinct domain ( gene ) trees is to reconcile them with a common species tree: mapping each node in a gene tree to a node in the species tree . There are likely multiple paralogous domains mapped to the same species . Since domains belonging to different species are unlikely to coevolve , we only need to consider the domains in the same species as candidates of the coevolving partners . For simplicity , we also hypothesize that there is at most one pair of coevolving partners from each ( ancient and contemporary ) species . The problem of building a joint phylogenetic tree then becomes the problem of choosing the coevolving partners in each node of the species tree . This problem is still difficult since there are many possible combinations of coevolving partners . We employed a heuristic to construct a joint tree of two domain families and to identify the coevolving partners in each species . The goal of this heuristic is to make the joint tree respect the phylogenetic trees of individual domain families and the species where they reside , to maximize the coverage of the species in the joint tree , and to reduce the spurious covariation from paralogous members . The heuristic is described in Materials and Methods and Text S3 . Despite the advantages of the heuristic , certain covariation from early divergence is amplified when the topology of the domain tree does not conform with the species tree . A typical example is the position pairs between many RNA polymerase and ribosomal proteins ( Figure S6 ) . The pair comprises two amino acid pair sequences denoted by 1 and 2 . The apparent recurrence of sequence 1 in bacteria , plants , and algae actually arises from the early divergence between bacteria/chloroplast and eukaryotes/archaea . This covariation can be structurally and functionally important , since it reflects the difference of transcription and translation apparatus between prokaryotes and eukaryotes . However , it deviates from the original purpose of identifying recurrent covariation across lineages . To further reduce this type of covariation , we trimmed the part of the domain tree which mismatched the topology of the species tree at kingdom level . The enrichment of functionally coupled domain pairs is similar to the untreated version: 219 out of 642 inferred position pairs and 82 out of top 100 inferred pairs were functionally coupled . Most pairs between RNA polymerase domains and between RNA polymerase and ribosomal proteins were absent in the inferred pairs . Although covariation between these domain pairs does not re-occur , it is still important . It is attributed to early divergence of life , and as described previously , maintains the structurally conserved region in RNA polymerase . The inferred domain pairs by removing covariation from early divergence are reported in Text S4 . Our model can also detect the coevolving positions within the same protein domains . Unlike inter-domain screening , the two amino acid residues share a common phylogenetic tree . Hence spurious covariation arising from selection of paralogous proteins does not happen . We calculated the log-odds score for each intra-domain position pair of all 8 , 183 domain families in Pfam . With a threshold value 5 . 0 ( CTMP simulation p-value <3 . 5 × 10−4 ) , we obtained 1 , 444 position pairs from 110 domain families . We also calculated the log-odds scores of the entire domains with multiple coevolving positions and ranked the 110 domains accordingly . The sorted domains , their coevolving positions , and the log-odds scores are reported in Text S5 . Two questions arising from inter-domain screening also need to be answered in intra-domain analysis . First , whether or not coevolving positions within the same domains are spatially coupled . Second , whether or not these coevolving positions overlap with or are close to functionally important sites of proteins . We extracted 401 protein structures of the 110 protein domains from the Protein Data Bank and calculated the distances between intra-domain coevolving positions . As a comparison we also calculated the distances between all position pairs in the same domain families . Figure 7 shows the intra-domain distance distributions of coevolving positions and the background . Similar to Figure 2 , the distances of coevolving positions ( solid blue ) tend to be shorter and narrowly distributed . Both coevolving and background distributions for intra-domain positions are substantially shorter than those for inter-domain positions , as amino acids in the same domains are typically close . Yet the difference between the two distributions is still significant ( p-value for Kolmogorov-Smirnov test <2 × 10−16 ) . About 50% of coevolving positions are less than 10Å apart , whereas only about 20% of background position pairs are within 10Å . The proximity of intra-domain coevolving positions is consistent with previous studies such as [11] . To check the functional importance of coevolution , we examined the intra-domain coevolving positions from the 38 domain families that contain the position pairs with log-odds scores ≥8 . 0 . The coevolving positions from 13 of these 38 domain families overlap with or are close to the functional sites of proteins . The reported functional sites are primarily active or ligand binding sites of enzymes since they are easy to identify in the literature . The coevolving positions on other proteins ( such as virus coat proteins ) might also carry functional information but are not evident by screening the literature . Table 3 shows the functions of intra-domain coevolving sites . Two remarkable instances are domains delta-aminolevulinic acid dehydratase ( PF00490 ) and photosynthetic reaction centre protein ( PF00124 ) . In the PF00490 , there are five coevolving positions . All of them are physically close ( <10Å ) in all three protein structures of the domain family . These positions partially coincide with the active sites and Mn2+-binding sites of Pseudomonas aeruginosa dehydratase protein ( [39] , PDB id 1b4k ) . In PF00124 , there are 41 coevolving positions . Some of these positions are close to the oxygen evolving center of Thermosynechococuus elongatus PSII protein ( [40] , PDB id 1s5l ) , which oxydizes water in photosynthesis . Other functional sites overlapped with , or close to , coevolving positions include proteolytic active site and RNA recognition site of 3C cysteine protease [41]; zinc fingers of Siah ubiquitin ligase [42]; active site and oxamate binding site of pyruvate formate lyase [43]; active site of peptidase family S41 [44]; Zn binding site of S-Ribosylhomocysteinase [45]; active site of UTP-glucose-1-phosphate uridylyltransferase [46]; active site of family 4 glycosyl hydrolase [47]; active site of phosphorylase family [48]; NAD binding site of lactate/malate dehydrogenase [49]; and Dha binding site of Dak1 domain [50] . The complete annotations of intra-domain coevolving sites on the PDB structures are in Text S6 . Analysis in the preceding sections suggests that coevolving domains are likely to be functionally coupled , and coevolving position pairs tend to be spatially coupled and located at functionally important sites . Yet the question in the reverse direction—whether physically interacting amino acid residues are coevolved—are still not answered . Since the majority of the coevolving positions are not in direct contact , we expect the overlap set between physical interactions and coevolving positions to be small . We extracted 223 , 392 physical interactions from Pfam . Interactions corresponding to the same aligned positions in the domain families were collapsed together . To reduce computational time we only considered the interactions where covarying amino acid pairs ( sequences that are distinct at both positions , for example , NF and FQ ) comprise more than half of the members in the domain families . Only about 20% of the interactions ( 45 , 007 out of 223 , 392 ) passed this filtering criterion . We evaluated the log-odds scores of these 45 , 007 interactions . The distribution of the log-odds scores is centered around 0 ( mean 0 . 209 ) with standard deviation 12 . 96 . Only a small fraction of interactions ( 2 . 6% ) have log-odds scores higher than 9 . 0 . The results indicate covariation is not necessary for physical interactions . The majority of physical interactions are dominated by conserved sequences or sequences with unilateral changes .
In this study we propose a probabilistic graphical model to detect coevolution of amino acid residues and invoke large-scale screenings on all the inter-domain , intra-domain position pairs , and known domain residue interactions . Despite the large number of pairwise comparisons executed , the inferred results strongly suggest that coevolving domains and positions are functionally and spatially coupled . The majority of coevolving protein domains appears in the same proteins or shares the same functional categorization . Coevolving positions between and within protein domains are substantially closer than the background distribution . Moreover , the coevolving positions in many proteins coincide with functionally important sites such as the subunit linkers of hydrogen peroxide dismutase , tRNA-binding sites of ribosomes , and active sites of phosphoglucomutase . Most top-ranking coevolving domain pairs are involved in fundamental functions of life: ribosomal proteins , RNA polymerase , carbon metabolism , vitamin B12 dependent enzymes , and so on . This is probably because these ancient proteins have strict structural constraints . Our model implicitly favors the case where covarying sequences maintain the structural constraints . In addition , the stringent filtering criteria of sequence covariation and a wide coverage of species required for significant scores may also exclude the lineage-specific coevolution . To detect coevolution in these variable families ( such as transcription factors , receptors , and signal transduction proteins ) , a targeted search on more extensive sampling of a specific clade and relaxed criteria for covariation are probably required . Since simultaneous changes of multiple nucleic or amino acids are unlikely , there must exist “transition states” between optimal configurations during evolution . These transition states may disappear in contemporary species due to their deleterious effects . In RNAs , however , we do observe non-pairing or wobbling bases in a stem . Transition states also appear in the coevolving protein domains . For example , although position pair PF00081–157/PF02777–61 in superoxide dismutase is dominated by NF and FQ pairs , there are also a few other states including FF , FE , FP , and FR . FF can serve as a transition state between NF and FQ . Intriguingly , the distance between an FR pair is 9 . 46 Å ( PDB id 1coj ) , indicating the two residues are not in contact . This suggests the transition states of amino acids may be accommodated by structural variation . Our inferred results , in agreement with previous studies of protein coevolution , reveal a fundamental difference between protein and RNA coevolution . Typically RNA coevolution occurs in disjoint nucleic acid pairs that form hydrogen bonds and are in direct contact in the 3-D structure . In contrast , there are often multiple coevolving amino acid residues in a protein , and some of them are distant in the 3-D structure . Coevolution of multiple and distant amino acid residues probably results from multiple selective constraints . Some possible explanations include the coupling of binding energy via pathways in the protein , interactions with intermediate molecules such as water , and the global constraints pertaining to the conformation of a region in a protein . The diverse causes of protein coevolution also make validation of computational methods problematic . Unlike RNAs , there is no gold standard for a coevolutionary protein dataset . We validated the findings with indirect evidence such as the enrichment of functionally coupled domains defined by GO categories , distance distribution in protein structures , and annotations of the functions of the coevolving sites . More appropriate validation procedures and datasets may become available as we have better understanding of protein coevolution . The existence of paralogous genes adds difficulty in analyzing coevolution . When there are multiple paralogous domains in a family , we have to assign coevolving partners from all possible combinations . Our heuristic method reduces , yet cannot eliminate , spurious covariation from paralogous families . A better algorithm of dealing with paralogous genes is needed . To facilitate large-scale screening we applied several simplifying assumptions and procedures . First , we applied the same sequence substitution rate matrix ( the Dayhoff matrix ) to all the domain families . Rate variation across domains or different sites within the same domains may create spurious covariation [15] . Second , like other phylogenetic methods of detecting coevolution , the accuracy of the results generated by our model depends on the accuracy of the phylogenetic trees , which is under debate . Third , due to the difficulty of acquiring the parameters of sequence evolution and positive and negative sets of coevolution , the simulation p-values and false discovery rates are subject to error . Refined analysis of specific protein families are needed in order to correct the false predictions from large-scale screening . The distribution of log-odds scores of known physical interactions shows most interacting amino acid residues do not possess covarying sequences , consistent with a recent finding in a yeast protein–protein interaction study [51] . The discrepancy between physical interactions and sequence covariation is attributed to many possible causes . Some interactions may be lineage-specific or have highly conserved sequences . Others may undergo unilateral changes within the same amino acid groups . Coevolution probably only occurs in a small fraction of physical interactions . Nevertheless , we also demonstrate that coevolution manifests spatial and functional constraints other than direct interactions . Hence , the complex relations between coevolution and selective constraints are worth pursuing at a deeper level .
The sequence substitution of a single amino acid is modeled by a CTMP [17] . Denote by x ( t ) the sequence composition at time t . P ( x ( t ) ) is a 1 × 20 probability vector of x ( t ) and follows a Markov process at an infinitesimal time interval: where Q is a 20 × 20 substitution rate matrix . Each row of Q must sum to 0 in order to make components of P ( x ( t ) ) sum to 1 . In this work we used the Dayhoff matrix of amino acid substitution [52] . The transition probability P ( x ( t ) |x ( 0 ) ) at a finite time interval t is given by the matrix exponential eQt , which is the solution of Equation 1: Define x ( t ) = ( x1 ( t ) , x2 ( t ) ) as the joint state of two amino acids . The sequence substitution follows the same equation for the single-site evolution ( Equation 1 ) , but the dimensions of the probability vector ( 1 × 400 ) and the rate matrix ( 400 × 400 ) are much bigger . If two sites are independently evolved , then the joint rate matrix can be derived from the rate matrix of single sites [18]: [ ( a1 , a2 ) , ( b1 , b2 ) ] specifies the sequence substitution rate of the independent model from state ( a1 , a2 ) to state ( b1 , b2 ) . In , the rate of a single amino acid change is equal to the corresponding rate in the single site rate matrix Q , and the rates of double amino acid changes are all zero . For example , [HR , HA] = Q[R , A] and [HR , GX]=0 . This is intuitive since off-diagonal entries of specify the transition probabilities at an infinitesimal time interval . At an infinitesimal time interval , at most one transition occurs for two independent CTMPs . Each diagonal entry of is again −1 and multiplies the sum of other entries in the same row . A true coevolutionary model should reward transitions into the sequence states of selective advantages and penalize the transitions of opposite directions . Due to the difficulty of finding this true model , we constructed a simplified model by reweighting the entries of the independent rate matrix to penalize single transitions and to reward double transitions: Transitions of single amino acids are penalized by multiplying a fixed number ɛ < 1 . Transitions of double amino acids from the same state ( a1 , a2 ) are rewarded by replacing zeros with an identical quantity Its value forces the diagonal entries in to be identical to . favors the sequences that have strong covariation between distinct states . To rank the coevolving domain pairs ( or single domains ) , we need to assess the likelihood scores which take all the coevolving positions between the two domains into account . We treated the model of all coevolving positions as a probabilistic graphical model in both space and time ( Figure 8 , top row ) . Each vertical edge on the phylogenetic tree specifies the temporal dependency between parent and child nodes , whereas each horizontal edge designates the spatial dependency between coevolving positions . These two types of dependencies create a grid-like network with many loops . It is in general difficult to evaluate the marginal likelihood of this network . We simplified the problem by adopting two approximations . First we approximated the spatial dependency network by its maximum spanning tree ( Figure 8 , middle row ) , with the weight of each edge corresponding to its pairwise log-odds score . This approximation removes the loops created by horizontal edges . The likelihood of an undirected tree model can be obtained from the singlet and pairwise marginal probabilities [53 , 54]: where φij and ψi are marginal pairwise and singlet probabilities corresponding to edges and nodes and di is the number of edges incident to node i . This formula can be obtained by assigning consistent directions to the edges and expressing the joint probability as the product of the prior probability of the root and the conditional probabilities of other nodes . The expression in Equation 5 is independent of edge direction assignments . We assumed the conditional probability from the coevolving positions in a parent species to the same set of positions in a child species followed a form similar to Equation 5: whereas P ( xi ( t ) , xj ( t ) |xi ( 0 ) , xj ( 0 ) ) and P ( xi ( t ) |xi ( 0 ) ) were given by the coevolutionary CTMP . Pairwise terms φij ( xi , xj ) and singlet terms ψi ( xi ) in Equation 5 were replaced by conditional probabilities P ( xi ( t ) , xj ( t ) | xi ( 0 ) , xj ( 0 ) ) and P ( xi ( t ) | xi ( 0 ) ) . The first approximation is still intractable since it has to sum over all possible states of all coevolving positions . To further simplify the problem , we performed marginalization for each singlet and pairwise term separately and combined these terms using Equation 5 . The marginal likelihoods of singlet and pairwise terms were calculated using Felsenstein's dynamic programming algorithm [55] . For instance , the marginal likelihood in the middle row of Figure 8 is approximated as where and are the pairwise and singlet marginal likelihood evaluated by dynamic programming . The marginal likelihood of the independent model is the product of the marginal likelihood for each position and can be exactly evaluated . The likelihood ratio in the bottom row of Figure 8 is It is costly to evaluate the coevolutionary likelihood scores . Hence we applied three filtering criteria on all 1 . 171 × 1011 inter-domain position pairs and all 8 . 29 × 107 intra-domain position pairs . First , we discarded the sequences that contained gaps in more than half of their members . Second , we discarded the conserved sequences where one amino acid pair occurred in more than 75% of the members . Third , for each of the remaining position pairs , we identified a maximal set of covarying amino acid pairs ( amino acid pairs which are distinct at both positions , e . g . , NF and FQ ) , and counted the number of occurrences for each amino acid pair . We only considered the sequences where the maximal set of covarying amino acid pairs constituted more than 80% of the members . The first two criteria filtered out the position pairs dictated by gaps and conserved amino acid pairs . The third criterion filtered out the sequences which were expected to have low log likelihood ratios since the coevolutionary model ( Equation 4 ) penalized the sequences with many unilateral changes ( e . g . , NF and FF ) . In all , 3 , 379 , 517 inter-domain position pairs and 196 , 198 intra-domain position pairs passed these criteria . To further reduce computation time and error , we applied the Padé polynomial approximation for matrix exponentiation [56] and pre-computed eQt on each branch length quantized by the following intervals: [0 , 0 . 01 , 0 . 02 , 0 . 05 , 0 . 08 , 0 . 1 , 0 . 2 , 0 . 5 , 0 . 8 , 1 . 0] . To learn the penalty weight ɛ , we chose the joint sequences of the coevolving superoxide dismutase position pairs ( PF00081–157/PF02777–61 ) as the training set and carried out a one-dimensional line search that maximizes its log-odds score . The optimal ɛ = 0 . 7 . To evaluate the coevolutionary likelihood of an inter-domain position pair , a joint phylogenetic tree and representatives from each species in each domain are needed . We selected the species that contained both domains and built a binary species tree on selected species by extracting the hierarchy from the National Center for Biotechnology Information taxonomy [57] . The topology of the species tree was used as the joint tree . For each domain family , we then applied a heuristic to select one representative domain for each species that reduces spurious covariation across paralogous lineages . The idea is to label each internal node of the domain family tree as a speciation or duplication event ( using a reconciliation algorithm , [58] ) and to pick up an orthologous subtree that maximizes species coverage . We then incrementally updated the branch length in the mapped species tree . The procedures of building a joint tree and selecting representatives are described in Text S3 . As a comparison we calculated mutual information between the 3 , 379 , 517 inter-domain position pairs that passed the filtering criteria . Denote x1 and x2 the sequence composition of sites 1 and 2 , P12 ( x1 , x2 ) the frequency of ( x1 , x2 ) among the aligned sequences , and P1 ( x1 ) and P2 ( x2 ) the marginal frequencies of x1 and x2 . The mutual information between x1 and x2 is where 0log0 ≡ 0 . The large-scale screenings , including filtering position pairs by sequence covariation , building the joint phylogenetic tree for domain family pairs , calculating pairwise coevolutionary scores , and evaluating the joint likelihood scores of the entire domains/domain pairs were implemented in C programs and executed on Rackable Linux Cluster ( 2048 AMD Opteron Processors , 2 . 2 GHz ) . The total CPU time was 24 , 000 h for inter-domain screening , 1 , 600 h for intra-domain screening , and 300 h for evaluating the log-odds scores of known interactions . The C codes and sequence data are available per request to the corresponding author . The p-value of the coevolutionary likelihood scores: the significance of log-odds scores was evaluated by random CTMP simulation . In each trial , we first randomly selected a domain family and acquired its phylogenetic tree . A subtree of 50–200 nodes was randomly extracted . We then generated the sequence pairs at leaves by simulating two independent CTMPs using the Dayhoff matrix and the selected tree . The log-odds score of the sampled sequence pairs was calculated . The p-value was the fraction of the 106 random trials which yielded the log-odds scores ≥ threshold θ . The p-value of θ = 9 . 0 is 6 . 0 ×10−5 . The false discovery rate of coevolving position pairs: We evaluated the false discovery rate of multi-hypotheses testing using the approximation procedures in [59] . Given a log-odds score threshold θ , we calculated the false positive rate ( p-value ) by the following procedure . We uniformly selected a random domain family pair which intersected in more than 20 species and did not share the same proteins or bottom-level GO annotations in more than half of their members , and then uniformly drew two random positions . The false positive rate P ( θ ) is the probability of finding a position pair with log-odds score ≥ θ . Notice P ( θ ) is considerably smaller than the p-value of CTMP simulation since many position pairs were filtered out by the pre-processing procedure . Denote m the total number of position pairs and m ( θ ) the number of position pairs with log-odds scores exceeding θ . The false discovery rate q ( θ ) on threshold θ is approximated by The total number of position pairs m = 1 . 17 × 1011 . With threshold θ = 9 . 0 , p ( θ ) = 1 . 114 × 10−8 , and m ( θ ) = 3 , 953 . Thus , q ( θ ) = 1 . 114 × 10−8 1 . 17 × 1011/3953 = 0 . 33 . Figure S1 shows the dependency of q ( θ ) and θ . q ( θ ) varies from 0 . 33 to 0 . 03 as θ varies from 9 . 0 to 30 . 0 . The p-value of enrichment of functionally coupled family pair: we used the standard hypergeometric p-value to assess the significance of enrichment of functionally coupled domain family pairs among inferred domain family pairs . Define N the total number of family pairs considered , n the number of inferred family pairs , K the total number of family pairs that were functionally coupled , and k the number of inferred family pairs that were functionally coupled . The hypergeometric p-value was the probability of randomly drawing n from N pairs ( without replacement ) where more than k pairs were functionally coupled: We downloaded 196 protein structure data from the 582 inter-domain family pairs and 401 protein structures from 110 intra-domain families from the Protein Data Bank [22] . We mapped a position in a domain to an amino acid residue in its PDB structure by aligning the domain sequence and PDB sequence of each chain in the PDB using ClustalW [60] . The closest distance between the atoms from the two amino acid residues was reported . If a protein domain could be mapped to multiple chains of a PDB , then from all possible amino acid residue pairs we reported the one with the smallest distance .
The accession numbers listed in this paper from the Protein Data Bank ( http://www . rcsb . org/pdb ) are alpha-hairpin iron/manganese superoxide dismutase domain , position 157 ( PF00081 ) , C-terminal iron/manganese superoxide dismutase domain , position 61 ( PF02777 ) , delta-aminolevulinic acid dehydratase ( PF00490 ) , and photosynthetic reaction centre protein ( PF00124 ) . | The sequences of different components within and across genes often undergo coordinated changes in order to maintain the structures or functions of the genes . Identifying the coordinated changes—the “coevolution”—of those components in the context of evolution is important in predicting the structures , interactions , and functions of genes . The authors incur a large-scale screening on all the known protein sequences and build a compendium about the coevolving relations of all protein domains—subunits of proteins . The majority of the coevolving protein domains either belongs to the same proteins , appears in the same protein complexes , or shares the same functional annotations . Furthermore , coevolving positions in the same proteins or protein complexes are spatially coupled , as they tend to be closer than random positions in the 3-D structures of the proteins/protein complexes . More strikingly , many coevolving positions are located at functionally important sites of the molecules . The results provide useful insights about the relations between sequence evolution and protein structures and functions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Accession",
"Numbers"
] | [
"eubacteria",
"eukaryotes",
"computational",
"biology"
] | 2007 | Detecting Coevolution in and among Protein Domains |
There is considerable variability in the susceptibility of smokers to develop chronic obstructive pulmonary disease ( COPD ) . The only known genetic risk factor is severe deficiency of α1-antitrypsin , which is present in 1–2% of individuals with COPD . We conducted a genome-wide association study ( GWAS ) in a homogenous case-control cohort from Bergen , Norway ( 823 COPD cases and 810 smoking controls ) and evaluated the top 100 single nucleotide polymorphisms ( SNPs ) in the family-based International COPD Genetics Network ( ICGN; 1891 Caucasian individuals from 606 pedigrees ) study . The polymorphisms that showed replication were further evaluated in 389 subjects from the US National Emphysema Treatment Trial ( NETT ) and 472 controls from the Normative Aging Study ( NAS ) and then in a fourth cohort of 949 individuals from 127 extended pedigrees from the Boston Early-Onset COPD population . Logistic regression models with adjustments of covariates were used to analyze the case-control populations . Family-based association analyses were conducted for a diagnosis of COPD and lung function in the family populations . Two SNPs at the α-nicotinic acetylcholine receptor ( CHRNA 3/5 ) locus were identified in the genome-wide association study . They showed unambiguous replication in the ICGN family-based analysis and in the NETT case-control analysis with combined p-values of 1 . 48×10−10 , ( rs8034191 ) and 5 . 74×10−10 ( rs1051730 ) . Furthermore , these SNPs were significantly associated with lung function in both the ICGN and Boston Early-Onset COPD populations . The C allele of the rs8034191 SNP was estimated to have a population attributable risk for COPD of 12 . 2% . The association of hedgehog interacting protein ( HHIP ) locus on chromosome 4 was also consistently replicated , but did not reach genome-wide significance levels . Genome-wide significant association of the HHIP locus with lung function was identified in the Framingham Heart study ( Wilk et al . , companion article in this issue of PLoS Genetics; doi:10 . 1371/journal . pgen . 1000429 ) . The CHRNA 3/5 and the HHIP loci make a significant contribution to the risk of COPD . CHRNA3/5 is the same locus that has been implicated in the risk of lung cancer .
COPD is expected to be the third leading cause of worldwide mortality and the fifth leading cause of morbidity by the year 2020 [1] . Cigarette smoking is the major risk factor for COPD but smokers show considerable variation in their risk of developing airflow obstruction . Familial aggregation studies suggest a strong genetic component to this risk [2]–[8] . However the only proven genetic risk factor for COPD is severe deficiency of α1-antitrypsin [9] , which is present in only 1–2% of individuals with COPD . This suggests that other genes have yet to be identified that predispose smokers to airflow obstruction . We report the first genome wide association study ( GWAS ) for COPD . Our primary discovery sample was a case-control population collected from Bergen , Norway , and we used three independent study cohorts to further evaluate the top associations emerging from the GWAS analysis .
We used a multi-stage replication design ( Figure 1 ) for this study . The genome-wide association analyses of the COPD case-control status in the Bergen cohort identified several significant associations , including three SNPs on chromosome 5 that reached the level of genome-wide significance ( Table S1 ) . The Q-Q plot showing the distribution of observed P values from the discovery cohort is shown in online Figure S1 . The top 100 SNPs were then evaluated in the ICGN population and 8 were replicated at a nominal p value of 0 . 05 ( SNP rs11219732 showed inconsistent risk alleles in the Bergen and ICGN population and hence was not considered further , Table 2 ) . Two of the three SNPs ( rs7727670 and rs7341022 on chromosome 5 ) that showed genome-wide significance in the Bergen cohort did not replicate in the ICGN population . The SNPs showing the most definitive evidence for replication were rs8034191 and rs1051730 in the CHRNA3/5 locus . Several additional SNPs were later analyzed in the CHRNA3/5 region in the Bergen and ICGN populations ( Table S2 ) . One non-synonymous polymorphism in CHRNA5 ( rs16969968 ) which coded for the substitution of an asparagine for an aspartic acid at amino acid 398 ) was associated with COPD in the Bergen ( p = 8 . 8×10−4 ) and ICGN ( p = 2 . 78×10−6 ) cohorts ( combined p value 5 . 08×10−8 ) . Since this SNP showed a weaker association than both rs8034191 and rs1051730 it was not considered as a causal variant . We then tested the 7 SNPs that showed definite or nominal significance in the NETT-NAS case-control population , and the results are provided in Table 2 . These results further confirmed the association of two SNPs at the CHRNA3/5 locus with COPD ( p = 2 . 5×10−3 , OR = 1 . 43 , combined p value: 1 . 48×10−10 for rs8034191 and p = 0 . 017 , OR = 1 . 32 , combined p value 5 . 74×10−10 for rs1051730 ) . Two SNPs ( rs1828591 and rs13118928 ) at the HHIP locus on chromosome 4 also showed consistent replication across the three cohorts , but the combined p values did not reach genome-wide significance ( 1 . 47×10−7 and 1 . 67×10−7 respectively ) . The only significant associations in the Boston Early-Onset COPD families were with the rs8034191 and rs1051730 SNPs at the CHRNA 3/5 locus ( p = 0 . 03 and 0 . 03 respectively ) and the rs1828591 and rs13118928 SNPs at the HHIP locus ( p = 0 . 0025 and 0 . 0014 respectively ) with post bronchodilator FEV1 . None of the SNPs was significantly associated with a diagnosis of COPD . Since the ICGN cohort had recruited subjects with a wide range of lung function , we also analyzed the association of the CHRNA 3/5 markers with post bronchodilator FEV1 after adjusting for age , height , gender , pack years and smoking status . The results show that CHRNA 3/5 SNPs were associated with FEV1 in the ICGN population ( p values 1 . 04×10−4 and 1 . 75×10−5 for rs8034191 and rs1051730 respectively ) . The COPD associated region on chromosome 15 spans seven genes ( Figure 2 ) . Cholinergic nicotinic receptor subtypes α3 , α5 and β4; IREB2 , PSMA4 , NP_001013641 . 2 ( a gene with unknown function ) and Q9UD29 ( Surfactant protein B ( SP-B ) -binding protein ) . A partial map of the region is shown in online Figure S2 . SP-B binding protein is a DNA binding protein which binds to the promoter of SP-B and affects its expression [10] . Since SP-B is a critical surfactant in the lungs [11] , we sequenced the SP-B binding protein in 30 COPD subjects who are homozygous for the risk allele of rs8034191 but did not identify any polymorphisms in this gene . The p values reported above were based on the adjusted analyses correcting for smoking exposure . The results from the unadjusted association analyses for COPD status were highly significant ( Bergen 2×10−4 and 4×10−4; ICGN 7 . 46×10−7 and 1 . 40×10−6; NETT/NAS , 2 . 0×10−5 and 2 . 5×10−4 and combined p values of 1 . 86×10−12 and 6 . 6×10−11 for rs8034191 and rs1051730 respectively; Table S3 ) . Although the adjustments for smoking exposure attenuated the p values , they still remained highly significant ( Table 2 ) . In the Norwegian discovery cohort , a significant genotype-by-environment interaction ( P = 0 . 002 , Table 3 ) was observed with a substantially higher risk of COPD in current smokers carrying the rs8034191 C allele ( OR = 2 . 00 ) than in former smokers ( OR = 1 . 10 ) . In the overall population , the C allele of rs8034191 was estimated to have a population attributable risk of 12 . 2% for COPD . This risk was 14 . 3% in current smokers and 3 . 1% in former smokers . The p values were attenuated in the ICGN family-based population following adjustment for age , sex , pack-years of smoking and center but remained highly significant ( Table 2 ) . We identified a SNP by pack-years interaction ( p = 0 . 0037 for rs8034191 ) , however no significant SNP by current smoking status interaction ( p = 0 . 85 ) was detected in the ICGN population . Testing directly for an association between the amount of smoking , measured as pack-years , within cases and controls respectively , with the SNP rs8034191 , demonstrated no such association in the Norway discovery cohort ( P = 0 . 63 and 0 . 47 , respectively ) . We also carried out tests comparing allele frequencies for current and former smokers and heavy and light smokers , ( two different definitions , using pack years of smoking and length of smoking history ) within cases and controls separately ( Table 3 ) . The only significant association observed was in comparing current and former smokers among the controls ( p = 0 . 028 ) . Similarly , the rs8034191 SNP was not associated with pack-years smoked in the NETT cases or in the NAS controls .
We have demonstrated and replicated genetic associations between SNPs at the CHRNA3/5 locus and COPD in four study populations . The estimated population attributable risk from this locus was 12 . 2% and represents the discovery of a common major locus contributing to COPD in the general population . However , a potential complication with the interpretation of these findings is the possibility that differences in smoking behavior , likely related to nicotine addiction , between COPD cases and controls may drive the observed association . This is similar to the recently reported association of CHRNA3/5 SNPs with lung cancer [12]–[14] . In the current study populations , only limited assessment of nicotine addiction is available: ( i ) whether subjects were still smoking at the time of study participation , and ( ii ) their lifetime smoking intensity . Thus , we have limited ability to disentangle a genetic determinant of smoking behavior from a genetic determinant of COPD through an alternative pathway . There are several pieces of evidences to suggest that there could be a direct effect of CHRNA3/5 locus on COPD , independent of smoking behavior . First , to the extent that smoking behavior is captured in pack-years , this effect should be factored out by the statistical design in which the discovery analyses used a logistic regression model incorporating pack-years , age and gender as covariates . The adjustments for pack-years smoked , age and gender were also performed in all the replication analyses . However , pack-years smoked only partially capture smoking behavior . Many other factors , such as depth of inhalation , number of puffs per cigarette , and age of starting smoking are also likely to affect the toxicant exposure and effect . Second , we also tested directly for an association between the amount of smoking , measured as pack-years , within cases and controls with the SNP rs8034191 . There was no significant association between the SNPs and pack-years of smoking in the Bergen and NETT/NAS populations . This is consistent with the observed allele frequency among the Norwegian pediatric general population sample ( minor allele frequency = 0 . 326 , n = 551 ) which is between that observed for cases and controls and not significantly different from either . We observed a genotype-by-environment interaction between the risk of the rs8034191 genotype and current smoking status on COPD in the Norwegian sample ( P = 0 . 002 , Table 3 ) , showing a substantially higher risk of COPD in current smokers carrying the rs8034191 C allele ( OR = 2 . 0 ) than in former smokers ( OR = 1 . 1 ) . There are several possible explanations for this gene-by-environment interaction . First , it could relate to nicotine addiction; smokers that have greater difficulty quitting smoking may be more likely to develop COPD . Alternatively , it could indicate that a subset of individuals is at greater risk for developing COPD if they continue to smoke . A similar interaction with current smoking was not identified in the ICGN families . Since all the COPD patients in the NETT population were former smokers , we could not address this question in the NETT/NAS study . The association of smoking dependence was explored in the lung cancer report by Hung and colleagues [13] who did not detect any association with individual Fagerstrom indices of nicotine addiction or when comparing controls with a heaviness of smoking index ( HSI ) . Another lung cancer report by Amos and colleagues [12] did show weak evidence for association with smoking behavior , while a further report by Thorgeirsson and colleagues[14] showed very strong association with smoking behavior . Association of this locus with nicotine dependence has been reported in two other studies [15] , [16] . Thus , it is reasonable to conclude that the variants may affect smoking behavior , at the same time as having a significant effect on COPD and other smoking related diseases such as lung cancer and peripheral arterial disease [12]–[14] . The CHRNA3/5 SNPs were also associated with lung function ( FEV1 ) in the ICGN and BEOCOPD populations . These SNPs were shown to be associated with FEV1 in the British Birth Cohort ( rs8034191 and rs1051730 , p = 0 . 029 and 0 . 023 , respectively ( http://www . b58cgene . sgul . ac . uk/ , accessed [3/7/2008] ) . Historically , nicotinic receptors are classified as neuronal or muscle-type , based on their initial site of identification and composite subunits [17] . Cholinergic activity in the airways primarily induces tracheo-bronchial smooth muscle contraction and mucous secretion . However , there is an increasing body of literature showing the importance of extra-neuronal cholinergic signaling [18] in the lung . The association of the SNPs at the chromosome 4 HHIP ( Hedgehog-Interacting Protein ) locus is also interesting , though it did not reach the stringent genome-wide significance levels in the populations studied in this manuscript . These SNPs were also associated with FEV1 in the BEOCOPD study ( rs1828591 and rs13118928 , p = 0 . 0025 and 0 . 0014 ) . The same SNPs are also associated with FEV1 in the British Birth Cohort ( rs1828591 and rs13118928 , p = 0 . 039 and 0 . 038 , respectively ) but were not associated with FEV1 in the ICGN population . In another manuscript in this issue of the journal , genome-wide association analysis results for FEV1/FVC in the Framingham Heart Study ( FHS ) are reported ( Wilk et al ) . Due to differences in genotyping platforms , the most significantly associated SNPs on chromosome 15 in our study were not genotyped in FHS . Analysis of the genotyped SNPs in the chromosome 15 region in the FHS indicated no significant association with COPD , but association with FEV1 percent predicted was observed with one SNP in LD with rs8034191 ( rs11636431 p value 0 . 007 ) . Evaluation of the imputed data for the most significantly associated SNPs in our populations did not show association with COPD in FHS . Several factors could contribute to the absence of association to the COPD phenotype in FHS: ( 1 ) The FHS cohort is a population-based collection , while our studies evaluated populations ascertained specifically for COPD; ( 2 ) The FHS cohort was recruited over three decades , while our cohorts represent more recent recruitments ( in the last 5–10 years ) -smoking habits have changed over time , and it is also possible that COPD clinical characteristics have changed over this period; ( 3 ) Our cohorts include a greater proportion of severe COPD subjects than in FHS; and ( 4 ) There could be differences in linkage disequilibrium patterns between study populations . Further studies will be required to define the specific genetic determinants influencing COPD on chromosome 15 , the relationship of these genetic factors to smoking behavior , and the characteristics of COPD subjects influenced by these genetic determinants . The association of the Chromosome 4 region in the FHS cohort was genome-wide significant for the FEV1/FVC ratio and was also associated with COPD . This association was subsequently replicated in the Family Heart Study population . Though the HHIP locus association in our study did not reach genome-wide significance , the additional evidence from the FHS and Family Heart Study underscore the importance of this locus on COPD susceptibility . We used independent populations with varying COPD severity , independent genotyping platforms and stringent statistical significance criteria to define genome-wide significant associations . We used consensus criteria for replication using a multi-stage replication design with similar phenotypes , the same genetic model and direction of association [19] . The levels of statistical significance of the association for our most significant results in the CHRNA3/5 region were consistent in all of the populations studied and are unlikely to be false positive results . The p values after adjusting for multiple testing using the most conservative Bonferroni correction were 7 . 3×10−5 and 2 . 83×10−4 for the SNPs rs8034191 and rs1051730 respectively . Though this can be considered as strength , the conservative approach for SNP confirmation that we have used may lead to larger false negative rates . However , with the inconsistent results of previous complex disease genetic association studies , we contend that a conservative approach is appropriate . We selected only the top 100 SNPs from the GWAS for subsequent replication study and a larger number of significant associations may have been uncovered if more of the most promising SNPs had been followed up . A negative association in the replication studies may not rule out a true association , since the power to detect association in the replication populations may be limited . The primary replication cohort ( ICGN ) is moderately powered to detect the replicated associations . Though the sample sizes of the NETT/NAS and BEOCOPD studies are relatively low , these studies include a large percentage of severely affected individuals , who may be enriched for COPD susceptibility genes . This likely account for the high rate of replication in these populations . COPD is a heterogeneous disease and we used a spirometry-based definition for COPD in all of the populations . Differences in smoking exposure , current smoking status , entry criteria and geographic origin of the cohorts might contribute to phenotypic heterogeneity and may lead to lack of replication . The fact that the replicated associations holds-up strongly and consistently in all the populations shows that phenotypic heterogeneity likely has little effect on the most significant results . In summary , we have identified robust evidence of association of COPD with the α-nicotinic receptor ( CHRNA 3/5 ) and HHIP loci . The hedgehog ( Hh ) gene family encodes signaling molecules that play an important role in regulating morphogenesis and the HHIP locus may play a role lung development . Although there is evidence of association of CHRNA 3/5 locus with nicotine addiction , both this study and recent reports of a lung cancer association [12]–[14] with the same alleles suggest that this region may be involved in more than nicotine addiction , and may potentially have direct functional relevance in the development of COPD , lung cancer , peripheral arterial disease , and other smoking related conditions . The first-degree relatives of both lung-cancer patients and COPD patients have higher rates of impaired forced expiratory flow rates than relatives of patients with non-pulmonary disease [20] , suggesting that susceptibility to lung cancer and COPD share common familial components . The association of CHRNA 3/5 locus with COPD , lung cancer , and peripheral arterial disease is powerful enough to make genetic screening of smokers an attractive interventional strategy .
Subjects from a case-control study [21] , [22] recruited from Bergen , Norway were used as the discovery cohort in the GWAS . Baseline characteristics of the subjects are shown in Table 1 . The entry criteria for COPD cases were post-bronchodilator forced expiratory volume in 1 second ( FEV1 ) <80% predicted and FEV1/FVC ( forced vital capacity ) <0 . 7 . The controls were selected based on post-bronchodilator FEV1 >80% predicted and FEV1/FVC >0 . 7 . Individuals with Pi ZZ , ZNull , Null-Null or SZ α1-antitrypsin deficiency were excluded . Subjects with chronic pulmonary disorders other than COPD ( e . g . , lung cancer , sarcoidosis , active tuberculosis , and lung fibrosis ) were also excluded . Because of the potential overlap in susceptibility genes for COPD and asthma , and the difficulty of diagnosing COPD vs . asthma in smokers with chronic airflow obstruction , previous asthma diagnosis was not used as an exclusion criterion . Both cases and controls were required to have a minimum of 2 . 5 pack-years of smoking . A total of 823 COPD cases and 810 controls were included in the present analysis . All of the subjects used in the primary and replication populations were current or former smokers ( Table 1 ) . Although the mean number of pack-years smoked was higher in cases ( mean 32 SD 18 ) compared with controls ( mean 19 SD 13 ) , subjects with a range of smoking intensities were included in the analysis . The distribution of pack-years of smoking in cases and controls in the Bergen cohort is shown in Figure S3 . Subjects from the International COPD Genetics Network ( ICGN ) were used as the primary replication population . In the multi-center ICGN study [22] , [23] subjects with known COPD were recruited as probands , and siblings and available parents were ascertained through the probands . Inclusion criteria for probands were post-bronchodilator FEV1<60% predicted and FEV1/VC <90% predicted at a relatively early age ( 45 to 65 years ) , a≥5 pack-year smoking history , and at least one eligible sibling with a≥5 pack-year smoking history . COPD in siblings was defined by a post-bronchodilator FEV1<80% predicted and FEV1/VC<90% predicted . The same exclusion criteria used in the Bergen study were also applied for the ICGN population . In total , 1891 Caucasian individuals from 606 pedigrees were included in the ICGN family-based association analysis . The second replication cohort included 389 non-Hispanic white COPD cases from the U . S . National Emphysema Treatment Trial ( NETT ) [24] and 472 non-Hispanic white control subjects from the Normative Aging Study ( NAS ) [25] . Subjects in NETT had severe COPD ( FEV1 ≤45% predicted ) and bilateral emphysema on chest CT; all NETT subjects were former smokers . Control subjects from the NAS had normal spirometry and at least 10 pack-years of cigarette smoking history . Subjects from extended pedigrees in the Boston Early-Onset COPD ( BEOCOPD ) study were used as an additional family-based replication cohort . BEOCOPD subjects were recruited through COPD probands with age <53 years , FEV1 <40% predicted , and without severe α1-antitrypsin deficiency [26] . The BEOCOPD analysis included 949 individuals from 127 pedigrees . Finally , to estimate allele frequencies in the general population in Norway we used 551 children ( all non-smoking ) from the Environment and Childhood Asthma ( ECA ) birth cohort study in Oslo [27] . All participants provided written informed consent and local institutional review boards approved the study protocols . All samples in the Bergen discovery cohort were genotyped using Illumina's HumanHap550 genotyping BeadChip ( version 3 ) which contains 561 , 466 single nucleotide polymorphisms ( SNPs ) . All samples that had a call rate <98% , and all SNPs that had a call frequency <99% were deleted . This resulted in the elimination of 23 , 436 SNPs from further analysis ( See Text S1 for more details ) . The ICGN subjects were genotyped using Sequenom's iPLEX SNP genotyping protocol developed for measurement with the MassARRAY mass spectrometer [28] . Genotyping in the NETT/NAS and BEOCOPD studies was performed using Sequenom iPLEX or Applied Biosystems TaqMan assays . Genotyping in the Norwegian ECA Birth cohort was done by TaqMan . For the association analyses COPD affection status in the Norway discovery cohort , we used a logistic regression model to perform single-marker genotype trend tests for the QC-passed SNPs . To control for the possibility of spurious associations resulting from population stratification , we used a modified EIGENSTRAT method [29] ( and Text S1 ) . This showed that there were 12 significant principal component axes , all of which were included in the model . We included age and sex , and since smoking effects are known to influence COPD risk , we also included current smoking status and pack-years of smoking as co-variates . The top 100 SNPs showing the lowest P values in this stage were selected for assessment in replication cohorts . For replication , we used a two stage strategy using three independent cohorts ( Figure 1 ) . In the first stage , family-based association analysis for COPD affection status was conducted in the ICGN data using PBAT version 3 . 6 [30] . Adjustments for age , gender , pack-years of smoking , current smoking status and center were performed in order to take into account the effect of smoking on the association results . Association with FEV1 was also tested using PBAT with age , gender , pack-years of smoking , current smoking status and height as co-variates . Gene-by-environment interaction analyses were also conducted using the PBAT program . Biallelic tests were conducted for SNPs using an additive genetic model . In the second stage the NETT case-control study was analyzed for the presence/absence of COPD using an additive genetic model . An unadjusted analysis and a logistic regression model adjusted for age and pack-years of smoking were conducted; sex was not included as a covariate because all NAS subjects were male , and current smoking was not included because all NETT subjects were ex-smokers . The BEOCOPD family-based study in the validation stage was analyzed using PBAT version 3 . 6 [30] . COPD was defined by post-bronchodilator FEV1/FVC<0 . 7 and FEV1<80% predicted ( GOLD stage 2 or greater ) . Because a broad range of FEV1 values were included in the BEOCOPD study , we also analyzed FEV1 as a quantitative outcome in that population . Analysis of post-bronchodilator values of FEV1 was adjusted for ever-smoking status , pack-years of smoking , age , sex , and height . We assessed genome-wide significance with a Bonferroni correction ( p cutoff = 1 . 013×10−7 considering 493 , 609 independent tests in the Bergen cohort ( see Text S1 ) , 100 tests in the ICGN cohort , 7 tests in the NETT/NAS study and 6 tests in the BEOCOPD study ( Total 493 , 772 tests ) . | There is considerable variability in the susceptibility of smokers to develop chronic obstructive pulmonary disease ( COPD ) , which is a heritable multi-factorial trait . Identifying the genetic determinants of COPD risk will have tremendous public health importance . This study describes the first genome-wide association study ( GWAS ) in COPD . We conducted a GWAS in a homogenous case-control cohort from Norway and evaluated the top 100 single nucleotide polymorphisms in the family-based International COPD Genetics Network . The polymorphisms that showed replication were further evaluated in subjects from the US National Emphysema Treatment Trial and controls from the Normative Aging Study and then in a fourth cohort of extended pedigrees from the Boston Early-Onset COPD population . Two polymorphisms in the α-nicotinic acetylcholine receptor 3/5 locus on chromosome 15 showed unambiguous evidence of association with COPD . This locus has previously been implicated in both smoking behavior and risk of lung cancer , suggesting the possibility of multiple functional polymorphisms in the region or a single polymorphism with wide phenotypic consequences . The hedgehog interacting protein ( HHIP ) locus on chromosome 4 , which is associated with COPD , is also a significant risk locus for COPD . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"respiratory",
"medicine/copd",
"and",
"allied",
"disorders",
"genetics",
"and",
"genomics/gene",
"discovery",
"respiratory",
"medicine/lung",
"cancer"
] | 2009 | A Genome-Wide Association Study in Chronic Obstructive Pulmonary Disease (COPD): Identification of Two Major Susceptibility Loci |
Interactions between antigen-presenting dendritic cells ( DCs ) and T cells are essential for the induction of an immune response . However , during malaria infection , DC function is compromised and immune responses against parasite and heterologous antigens are reduced . Here , we demonstrate that malaria infection or the parasite pigment hemozoin inhibits T cell and DC interactions both in vitro and in vivo , while signal 1 intensity remains unaltered . This altered cellular behaviour is associated with the suppression of DC costimulatory activity and functional T cell responses , potentially explaining why immunity is reduced during malaria infection .
The activation of a protective and highly specific immune response requires a system that can survey , decipher , and quickly respond to infection in an appropriate manner . Dendritic cells ( DCs ) participate in all of these important activities , and upon detection of a “danger signal” , rapidly mature and express molecules required to generate the antigen-specific ( signal 1 ) and costimulatory signals ( signal 2 ) required to induce the activation of antigen-specific CD4+ “helper” T cells [1] . These cellular interactions are an essential step in initiation of adaptive immune responses , and factors that influence signal 1 and 2 upon DCs , such as the dose and duration of antigen presentation [2 , 3] and the level of activation of the DC [4] , can all affect the type of adaptive immune response induced . Importantly , differences in the behaviour of T cells can be observed in vivo by comparing situations in which T cells are primed ( i . e . , activated to induce a protective immune response ) or tolerised/regulated ( i . e . , where they become functionally hyporesponsive ) . Thus , during tolerance induction , clusters of T cells are smaller compared with priming responses [5] , and regulatory T cells ( Tregs ) are known to prevent stable T cell–DC interactions [6 , 7] . Malaria represents a global health challenge , with approximately 500 million clinical cases reported annually [8] . Plasmodium can induce immunosuppression in infected individuals , resulting in an increased susceptibility to secondary infections and reduced vaccine efficacy in patients and in animal models [9–14] . Suppression of immune responses is in part associated with the uptake of the malaria pigment hemozoin ( HZ ) [15 , 16] . Although certain studies have suggested that HZ activates DCs [17 , 18] , others have demonstrated that DCs are functionally less responsive during malaria infection [19] , and that HZ is able to regulate DC activation [20 , 21] . We have recently shown that the modulation of DCs by malaria infection and/or HZ significantly reduces T cell expansion , cytokine production , and migration into B cell follicles [21] , explaining the immunosuppression observed in infected individuals [9–14] . Importantly , the modulation of DC function ( and consequently immunity ) is a highly dynamic phenomenon , which changes throughout the course of infection , with immune suppression most evident soon after the peak of infection . This affects immune responsiveness , with T cell and B cell responses to heterologous antigens altered kinetically during the course of malaria infection [21] . We were therefore interested in examining whether the modulation of interactions between DCs and T cells by malaria infection could account for this reduced immunity . Using in vitro and in vivo systems to examine the behaviour of these cells , we have demonstrated that the immune suppression during malaria infection is associated with impaired DC–T cell interactions .
To dissect the effects of the malaria pigment on DCs , we first examined the uptake of HZ over time as well as DC activation , as assessed by increased expression of the costimulatory molecule , CD40 ( Figure 1 ) . Bone marrow ( BM ) –derived DCs readily phagocytosed HZ in vitro , with pigment taken up within 5 min of addition and accumulating steadily in the majority of DCs over the following 2 h ( Figure 1A and Video S1 ) . Addition of synthetic HZ [22] to DCs not only failed to induce upregulation of CD40 ( Figure 1B ) , but also reduced the subsequent responsiveness of DCs to lipopolysaccharide ( LPS ) stimulation ( Figure 1C ) , confirming our previous observations with HZ isolated from P . falciparum ( Figure S1 and [21] ) . Furthermore , the morphological changes typically associated with LPS-induced activation of DCs were not observed in HZ-treated DCs by time-lapse microscopy ( Videos S2 and S3 ) . The central function of DCs is presentation of antigen to CD4+ T cells in order to initiate the adaptive immune response . We therefore assessed the ability of HZ-loaded DCs to stimulate ovalbumin ( OVA ) -specific CD4+ T cell proliferation . As expected , OVA-specific DO11 . 10 T cells proliferated following incubation with OVA-pulsed DCs or LPS-stimulated OVA-pulsed DCs . This proliferation was significantly reduced if the DCs were treated with HZ prior to addition of antigen ( Figure 1D ) , as was subsequent T cell cytokine production ( unpublished data ) . Thus , DCs exposed to HZ rapidly accumulate pigment , rendering them functionally impaired with reduced LPS responsiveness and a failure to induce efficient T cell responses . The initial interactions between DC and T cell have important implications for the functional outcome of the T cell response [2–4] . Since we had observed reduced T cell responses following stimulation with HZ-treated DCs , we next examined in vitro the interactions between T cells and DCs that are involved in the induction of an antigen-specific response . DCs were labelled with the fluorescent dye CMRA ( orange ) and pulsed with OVA before mixing with CFSE-labelled ( green ) OVA-specific CD4+ T cells . Twenty hours later , T cells were observed clustered around DCs , with several CFSE+ T cells remaining closely associated with CMRA+ DCs for up to 2 h ( Figure 2A and Video S4 ) . However , in cultures in which the DCs had been treated with HZ prior to antigen pulse , these clusters were not evident , with CFSE-labelled T cells making transient contacts with CMRA+ DCs lasting only a few minutes ( Figure 2B and Video S5 ) . To quantify this clustering of T cells around DCs , we employed an assay in which fluorescently labelled DCs and T cells were co-cultured as described above and then fixed using paraformaldehyde prior to analysis by flow cytometry . In this way , CMRA+ DCs , CFSE+ OVA-specific T cells , and clusters of DCs and T cells could be individually identified ( Figure S2 ) . Addition of antigen to this system increased the proportion of clustered cells ( detected as CMRA+ CFSE+; Figure 2C ) . However , this colocalisation of T cells with DCs was reduced in cultures in which the DCs were pre-treated with HZ ( Figure 2C ) , confirming the above observations . The above data suggest that HZ-treated DCs are altered such that their ability to interact with naïve CD4+ T cells is suppressed , resulting in reduced T cell proliferation . One possible explanation for this observation is that HZ-loaded DCs are simply unable to take up , process , or present antigen . Use of GFP-labelled Eα antigen [23] allowed us to measure antigen uptake by DCs , as well as antigen presentation using the YAe antibody that specifically recognises the complex of Eα-peptide in the context of class II major histocompatibility complex ( MHC ) [24] . In this system , DCs pre-treated with HZ were able to process and present antigen as effectively as untreated DCs ( Figure 2D ) . Importantly , DCs co-cultured with P . chabaudi–infected erythrocytes ( that produce HZ deposition within DCs and recapitulate the described effects on T cells [21] ) also presented Eα-peptide to the same extent as resting DCs or DCs cultured with uninfected erythrocytes ( Figure S3 ) . Thus , whilst HZ-treated DCs are able to process antigen efficiently and provide a “signal 1” in the form of antigen/MHC for the T cell , their interaction with CD4+ T cells is altered such that stable clusters do not form to initiate a functional T cell response . In order to confirm the above in vitro observations in vivo , we used multi-photon laser scanning microscopy to examine the interactions between DCs and T cells in intact lymph nodes ( LNs ) . BALB/c mice received CMTPX-labelled DO11 . 10 T cells and were subsequently immunised subcutaneously with CFSE-labelled DCs . Twenty-four hours after immunisation with untreated DCs , a small proportion of these cells had migrated to the draining LN and could be seen making transient contacts with OVA-specific CD4+ T cells ( Figure 3A and Video S6 ) , as previously described [25–27] . When DCs were pulsed with OVA prior to transfer , more interactions between DCs and antigen-specific T cells were observed in the draining LN , and these contacts appeared to last longer ( Figure 3B and Video S7 ) . While interactions between antigen-specific T cells and DCs were also observed following injection of OVA-pulsed , HZ-treated DCs , these appeared more motile , and less stable , than the clustering seen with normal antigen-pulsed DCs ( Figure 3C and Video S8 ) . The 4-dimensional movement of the OVA-specific T cells was characterised by software-based tracking . As previously described [28] , the 3-dimensional velocity of antigen-specific CD4+ T cells was reduced by pulsing DCs with OVA prior to transfer as T cells clustered around DCs ( Figure 3D–3F ) . However , T cells stimulated by HZ-treated , OVA-pulsed DCs had higher mean velocities than those of T cells undergoing an effective priming response , although this was also significantly slower than naïve CD4+ T cells ( Figure 3D ) . Despite this difference in velocity , the meandering index ( a measure of directionality of cell movement ) of T cells in recipients of normal OVA-pulsed or HZ-treated OVA-pulsed DCs was reduced to the same extent compared with naïve T cells ( Figure 3E ) . Furthermore , responding T cells upregulated the early activation marker CD69 in recipients of OVA-pulsed DCs , irrespective of whether they were loaded with HZ or not ( unpublished data ) , demonstrating that T cells were recognising antigen presented by the DCs . In order to quantify the interactions between DCs and T cells , we analysed the degree of colocalisation between green DCs and red T cells to calculate a colocalisation coefficient representing the proportion of green voxels that were also red . Whilst T cell interaction with normal DCs was increased following antigen recognition , the colocalisation of DCs and T cells was inhibited by HZ treatment of DCs ( Figure 3F ) . Thus , it appears that the in vivo interactions between CD4+ T cells and HZ-loaded DCs are altered relative to a priming response , despite recognition of antigen by T cells , confirming our in vitro observations . We next wanted to confirm the importance of the above observations during malaria infection . We have previously demonstrated that around day 12 after infection with P . chabaudi , OVA-specific CD4+ T cells fail to proliferate effectively to challenge with antigen and that as a consequence , T cell migration into B cell follicles is suppressed and antibody responses are not induced [21] . We therefore characterised the migratory behaviour of OVA-specific CD4+ T cells following immunisation of malaria-infected animals ( Figure 4A–4C and Videos S9–S11 ) . When the movement of multiple cells in several samples was analysed , it was evident that these T cells moved less rapidly and migrated shorter distances 8 and 20 h following immunisation , compared with naïve T cells ( Figure 4D–4F ) . Although T cells transferred into malaria-infected animals immunised with OVA moved more slowly than naïve T cells , their velocities were significantly higher than those of primed T cells in uninfected recipients at both time points following immunisation ( Figure 4D ) . T cells activated in P . chabaudi–infected animals also moved greater distances away from their point of origin than T cells activated in uninfected animals ( Figure 4E and 4F ) . These differences were observed at a time when lymphoid architecture was essentially normal [21] and no differences were apparent between the behaviour of naïve T cells in uninfected and P . chabaudi–infected animals ( Figure S4 ) . Despite this difference in behaviour , OVA-specific T cells upregulated the early activation marker CD69 in response to immunisation in both uninfected and P . chabaudi–infected animals ( Figure 4G ) , suggesting that antigen is presented effectively to T cells during malaria infection . In order to confirm that antigen presentation was unaltered by malaria infection , P . chabaudi–infected C57BL/6 mice were immunised with 500 μg of Eα-GFP plus 50 ng of LPS on day 12 of infection . Eight or 20 h later , splenic DCs isolated from malaria-infected animals showed similar surface levels of peptide/MHC ( signal 1 ) as DCs from uninfected animals ( Figure 4H ) . Therefore , the lack of formation of long-lasting interactions between DCs and T cells that is associated with reduced effector functions of T cells in malaria-infected animals is signal 1 independent . Several reports have demonstrated immunosuppression during malaria infection in patients and animal models [9–14] . We [21] and others [20] have shown that DC activation and function is suppressed by infection or malaria pigment and , as a consequence , T cell and B cell responses fail to develop effectively . Here we have shown that uptake of HZ by DCs modulates their ability to interact effectively with T cells , despite presenting normal levels of peptide/MHC on their surface . The failure of T cells to fully interact with DCs appears to result in a lack of efficient T cell priming , leading to the subsequent failure of a protective immune response . Thus , these observations may explain why T cell responses fail to develop during malaria infection and clearly demonstrates the significance of these early interactions in the generation of an immune response . It is important to highlight that the impact of malaria on the ability of the immune system to respond changes markedly during the different phases of the infection . Thus , DC function is impaired immediately following the initial burst of parasitaemia . As a consequence , T cell proliferation , effector function , and migration are suppressed , and thus B cells do not receive help for expansion or antibody production [21] . We were interested in examining why T cell responses fail at this important time point and reasoned that the interactions between DCs and T cells might play an important role . By analysing the behaviour of T cells in malaria-infected animals following the peak of parasitaemia ( a period of immune suppression ) , we demonstrated that the failure of immunity is associated with reduced interaction between T cells and endogenous antigen-presenting cells ( Figure 4 ) . Such immune suppression is not as evident at earlier and later time points of infection , but it will be interesting to examine how DC–T cell interactions are affected at these stages . Importantly , transfer of antigen-pulsed , HZ-treated DC into naïve , uninfected recipients recapitulated these observations , suggesting that while pro-inflammatory infections such as malaria may be followed by a wave of anti-inflammatory cytokine production [29] , failure of T cells to interact with DC effectively is sufficient to suppress immune priming . In this investigation , we confirmed several previous reports that DCs pulsed with HZ displayed an impaired ability to undergo subsequent activation . It is important to note that others have suggested a pro-inflammatory role for HZ , possibly via the pathogen recognition receptor TLR9 [17] . Although , in our hands , DCs treated with HZ show a minor upregulation of CD40 , this is significantly lower than the activation seen by stimulating DCs with LPS ( Figure 1B ) , and these HZ-treated DCs fail to respond to subsequent activation ( Figure 1C ) . It is possible that these differences reflect the different subsets of DCs used , since TLR9 is primarily expressed by Flt3L-generated , plasmacytoid-like DCs [30] , as opposed to the granulocyte/monocyte colony stimulating factor ( GM-CSF ) -induced , predominantly myeloid DCs used here . It is also possible that the observed differences reflect the methods used to produce synthetic HZ from β-hematin ( unpublished data ) . Recent work has demonstrated the TLR9/MyD88-dependent activation of DC by parasite-derived HZ to be due to the binding of malarial nucleic acids to HZ [31] . In the current study we therefore used synthetic HZ . Importantly , synthetic HZ produced by our method recapitulates the effect upon DC function of both HZ isolated from parasites as well as the effect of Plasmodium infection in vivo ( Figure S1 and [21] ) . One explanation for our findings that HZ-loaded DCs are unable to form stable interactions with T cells is that the DCs express fewer costimulatory or adhesion molecules on their surface . Indeed , despite normal signal 1 , HZ-treated DCs fail to upregulate costimulatory molecules in response to both TLR ligands and CD40 ligation ( Figure 1 and [21] ) . Thus , although T cells are initially activated through antigen/MHC ( achieving sufficient signal to express CD69 ) , other molecules involved in synapse formation or T cell activation are not expressed by HZ-loaded DCs and stable clustering does not occur . This shows that signal 1 alone is not sufficient to drive T cell clustering and other factors , most likely costimulatory or adhesion molecules , are required for stable interactions between T cell and DC . Alternatively , active suppression of the T cell response may occur by the involvement of Tregs , which have been implicated in malaria infection [32] . Recent evidence has demonstrated that Tregs are able to suppress T cell activation by reducing their interactions with DCs in vivo [6 , 7] , and this may be a possible mechanism for immune suppression during malaria infection . Our results demonstrate the significance of the early interactions between T cells and DCs in the priming of effector T cell responses . Whilst the T cell–DC clustering dynamics are altered by HZ , DCs still present antigen to T cells , as demonstrated directly by YAe antibody staining and by the upregulation of CD69 by T cells . Thus , it is an inability to form stable , long-lasting clusters with HZ-laden DCs , independently of signal 1 , that results in the reduced effector function of CD4+ T cells seen during malaria infection .
Female BALB/c and C57BL/6 mice were purchased from Harlan Olac . The DO11 . 10 transgenic mice , with CD4+ T cells specific for OVA323–339 peptide in the context of I-Ad recognised by the KJ1 . 26 clonotypic antibody [33] , were obtained originally from N . Lycke , University of Göteborg , Sweden , and backcrossed onto the scid background such that all lymphocytes were OVA-specific CD4+ T cells . All mice were maintained at the Biological Procedures Unit , University of Strathclyde , under specific pathogen-free conditions and first used between 6 and 8 wk of age in accordance with local and UK Home Office regulations . To initiate a malaria infection , mice were inoculated with 1 × 106 P . chabaudi AS-infected erythrocytes intraperitoneally . Parasitaemia was monitored by thin blood smears stained with Giemsa's stain . Peak parasitaemia occurred at 5–6 d post-infection , after which time parasite levels declined and remained at low but usually detectable levels for the remainder of experiments ( Figure S5 ) , as previously described [34] . Malaria-infected ( day 12 of infection ) and control mice were immunised intravenously with 500 μg of OVA ( Sigma-Aldrich ) or Eα-GFP [23] , along with 50 ng LPS ( from Salmonella equi abortus; Sigma-Aldrich ) . DCs were prepared from BM as previously described [35] . Cell suspensions were obtained from femurs and tibias of female BALB/c mice . The BM cell concentration was adjusted to 5 × 105 cells/ml and cultured in 6-well plates ( Corning Costar ) in complete RPMI ( cRPMI: RPMI 1640 supplemented with L-glutamine [2 mM] , penicillin [100 μg/ml] , streptomycin [100 μg/ml] [all from Invitrogen] , and 10% FCS [Labtech International] ) containing 10% of culture supernatant from X63 myeloma cells transfected with mouse GM-CSF cDNA . Fresh medium was added to the cell cultures every 3 d . On day 6 , DCs were harvested and cultured at the required concentration for each individual experimental procedure as described below . This technique generated a large number of CD11c+ DC largely free from granulocyte and monocyte contamination , as previously described [35] . DCs were antigen loaded for 6 h with 5 mg/ml OVA ( Worthington Biochemical ) , 5 μg/ml OVA323–339 peptide ( Sigma-Genosys ) , or 100 μg/ml Eα-GFP [23] , and/or stimulated with 1 μg/ml LPS prior to use , as indicated in individual experiments . In imaging experiments , DCs were fluorescently labelled with Cell Tracker Orange ( CMRA; Invitrogen ) or 5 , 6-carboxy-succinimidyl-fluorescein-ester ( CFSE; Invitrogen ) immediately before use [36] . For in vivo imaging , 1 million DCs were immunised subcutaneously in the footpad of T cell recipient animals and the draining popliteal LN imaged 20–24 h later . DO11 . 10/scid LNs and spleens were homogenised and resulting cell suspensions washed twice and resuspended in RPMI . Cells were labelled with the fluorescent dye CFSE or Cell Tracker Red ( CMTPX; Invitrogen ) immediately before use [36] . Syngeneic BALB/c recipients received 3 × 106 to 6×106 antigen-specific cells . For functional in vitro studies , OVA-specific T cells were mixed with HZ-treated or control BM DCs at a 1:1 ratio in 96-well tissue culture plates ( Corning Costar ) . T cell proliferation was assessed after 72 h of culture and assessed by incorporation of [3H] thymidine ( 0 . 5 μCi/well ) for the last 24 h of culture . To measure in vitro clustering of T cells with DC , cells were co-cultured in 12-well plates for 20 h and then fixed with 4% paraformaldehyde ( Sigma-Aldrich; 20 min at 4 °C ) . Gly-Gly ( 0 . 06%; Sigma-Aldrich ) was added and briefly incubated for 1 min to neutralise residual paraformaldehyde . Cells were then harvested and analysed by flow cytometry . Aliquots of 1 × 106 cells in 12 × 75 mm polystyrene tubes ( BD Biosciences ) were resuspended in 100 μl of FACS buffer ( PBS , 2% FCS and 0 . 05% NaN3 ) containing Fc Block ( 2 . 4G2 hybridoma supernatant ) as well as the appropriate combinations of the following antibodies: anti-CD4-PerCP ( clone RM4–5 ) , anti-CD11c-PE ( clone HL3 ) , anti-CD40-FITC ( clone 3/23 ) , anti-CD69-PE ( clone H1 . 2F3 ) , PE-hamster IgG isotype control and FITC-rat IgG2a , k isotype control ( all BD Biosciences ) , biotinylated KJ1 . 26 antibody , or biotinylated-Y-Ae . Biotinylated antibodies were detected by incubation with fluorochrome-conjugated streptavidin ( BD Biosciences ) . After washing , samples were analysed using a FACSCanto flow cytometer equipped with a 488-nm Argon laser and a 635-nm red diode laser ( BD BioSciences ) and analysed using FlowJo software ( Tree Star ) . Synthetic HZ was produced using the method of Egan et al [22] . Briefly , hemin chloride ( Sigma-Aldrich ) was polymerised using 4 . 5 M sodium acetate at 60 °C and the product extensively washed with deionised water and filtered using 0 . 22-μm cellulose nitrate filtration units . Endotoxin-free buffers and solutions were used throughout . Plasmodium HZ was isolated from supernatants obtained from cultures of P . falciparum gametocytes , kindly provided by Lisa Ranford-Cartwright , Division of Infection and Immunity , University of Glasgow , UK . Supernatants were centrifuged for 20 min at 450g . The pellet was washed three times in 2% SLS and resuspended in 6 M guanadine HCl . Following five to seven washes in PBS , the pellet was resuspended in PBS and sonicated for 90 min using Soniprep 150 ( Sanyo Scientific ) at an amplitude of 5–8 μm . Again , endotoxin-free buffers were used throughout . Total haem content was determined as previously described [37] by depolymerising haem in 1 ml of 20 mM NaOH/2% SDS , incubating the suspension at room temperature for 2 h , and then reading the OD at 400 nm using UV visible spectrophotometer ( Thermo Spectronic , Heλios ) . Prior to use , the HZ was sonicated to minimise aggregation and maintain the HZ in suspension . DCs were pulsed with 1–40 μM HZ , a range similar to that seen when DCs were cultured at 1:100 ratio with P . chabaudi–infected erythrocytes . For in vitro imaging , 5 × 104 CMRA-labelled DC and 5 × 104 CFSE-labelled DO11 . 10 T cells were co-cultured on an Ibidi μSlideVI ( Thistle Scientific ) . Time-lapse images were acquired using an Axiovert S-100 Zeiss microscope using a ×63 oil immersion lens 20 h after mixing of cells . For the images in Figure 1 , 40 μM HZ was added in RPMI at the initiation of imaging . To image cellular interactions in LNs , the excised LNs were transferred into CO2-independent medium ( Invitrogen ) at room temperature . The LN was bound with veterinary glue ( Vetbond , 3 M ) onto a coverslip that was then adhered with grease to the bottom of the imaging chamber and continuously supplied with warmed ( 36 . 5 °C ) and gassed ( 95% O2 and 5% CO2 ) RPMI 1640 before and throughout the period of microscopy . Excised LNs were imaged on the following system , as previously described [5 , 38] . The two-photon excitation source was a solid-state , tunable Titanium: sapphire laser system ( 5W Chameleon; Coherent Laser Group ) . The laser beam was routed into a multi-photon excitation laser scanning system ( Radiance; Bio-Rad Laboratories ) . The objective lens used for all imaging investigations was the CFi-60 Fluo-W 40X/0 . 8 water-dipping objective lens ( Nikon ) . The sample was illuminated with 780–830 nm light , and the emission spectrum was separated with a 550-nm dichroic mirror ( Chroma Technologies ) . Each imaged volume consisted of between 11 to 18 planes 2 . 55 μm apart . Volumes were acquired every 18 to 38 s . Images were analysed using Volocity software ( Improvision ) . Objects were tracked for at least eight time points and the mean velocity , displacement , and meandering index calculated for each . The interaction between DCs and T cells was measured by quantifying the colocalisation of green voxels with red to generate the colocalisation coefficient—a measure of the proportion of DC volume in contact with T cells . Results are expressed as mean ± standard deviation . Significance was determined by Student's t-test using Minitab . A p-value of p ≤ 0 . 05 was considered significant . | Malaria is a major infectious disease , affecting 500 million people and causing 2 . 7 million deaths each year . The severity of malaria is , in part , due to the failure of the host immune system to effectively clear an infection and generate protective immunity . Dendritic cells ( DCs ) are central to the immune system; by presenting components of pathogens to circulating T cells , they are able to initiate a highly specific immune response to clear an infection . Importantly , the quality of the interaction between T cell and DCs can affect the functional outcome of the immune response . However , previous work has demonstrated that DCs are modified by malaria parasites , resulting in inefficient priming of the adaptive immune system . Here , we have visualised the interactions between DCs and T cells in the context of malaria and demonstrate that infection is able to prevent priming of immune responses by antagonising these cell–cell contacts . Importantly , the failure to form long-lasting interactions is not due to reduced presentation of antigens by the DC , suggesting that other mechanisms may be involved . These studies provide a visual insight into the mechanism by which parasites may suppress immunity and highlight the importance of early cellular interactions in the immune response . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"mus",
"(mouse)",
"immunology",
"plasmodium"
] | 2007 | Malaria Impairs T Cell Clustering and Immune Priming despite Normal Signal 1 from Dendritic Cells |
During meiosis , chromosomes align with their homologous pairing partners and stabilize this alignment through assembly of the synaptonemal complex ( SC ) . Since the SC assembles cooperatively yet is indifferent to homology , pairing and SC assembly must be tightly coordinated . We identify HAL-2 as a key mediator in this coordination , showing that HAL-2 promotes pairing largely by preventing detrimental effects of SC precursors ( SYP proteins ) . hal-2 mutants fail to establish pairing and lack multiple markers of chromosome movement mediated by pairing centers ( PCs ) , chromosome sites that link chromosomes to cytoplasmic microtubules through nuclear envelope-spanning complexes . Moreover , SYP proteins load inappropriately along individual unpaired chromosomes in hal-2 mutants , and markers of PC-dependent movement and function are restored in hal-2; syp double mutants . These and other data indicate that SYP proteins can impede pairing and that HAL-2 promotes pairing predominantly but not exclusively by counteracting this inhibition , thereby enabling activation and regulation of PC function . HAL-2 concentrates in the germ cell nucleoplasm and colocalizes with SYP proteins in nuclear aggregates when SC assembly is prevented . We propose that HAL-2 functions to shepherd SYP proteins prior to licensing of SC assembly , preventing untimely interactions between SC precursors and chromosomes and allowing sufficient accumulation of precursors for rapid cooperative assembly upon homology verification .
Generation of haploid gametes during sexual reproduction depends critically on the ability of homologous chromosomes to identify and form pairwise associations with their appropriate partners . Pairing is essential to enable homologs to orient and segregate away from each other at the meiosis I division , thereby achieving the reduction in ploidy that is necessary to ensure restoration of the diploid state upon fertilization . Pairing between homologous chromosomes is reinforced by a highly ordered tripartite structure known as the synaptonemal complex ( SC ) that assembles at the interface between aligned homologs [1] . The lateral elements ( LEs ) of the SC are comprised of cohesin complexes and meiosis-specific axial components that coalesce along the length of each homolog during early meiotic prophase . SC central region ( CR ) proteins , which contain extended coiled-coil domains , then load to create transverse connections between the axes of the paired homologs , conferring the characteristic zipper-like appearance of mature SC . SC proteins collaborate with meiosis-specific recombination proteins to promote crossover recombination events between the homologs and to promote maturation of chromosome structure surrounding crossover sites into temporary connections known as chiasmata that enable homolog segregation at the meiosis I division [2] . Although the SC normally assembles only between paired homologous chromosomes , the SC structure itself is indifferent to homology . Moreover , SC assembly is highly cooperative and processive , such that once synapsis is initiated , it can proceed along nonhomologous chromosome segments [1] . Further , CR proteins have a propensity to self-assemble into aggregates known as polycomplexes even when they are not associated with chromosomes [3]–[6] . Thus , homolog pairing and synapsis must be tightly coordinated to ensure that SC assembly only occurs in a productive fashion , linking the axes of correctly aligned homologs . During Caenorhabditis elegans meiosis , much of the coordination between pairing and synapsis is mediated by specialized chromosomal domains termed pairing centers ( PCs ) that are located near one end of each chromosome [7]–[9] . PCs function both in conferring SC-independent local stabilization of pairing and in promoting SC assembly [8] , [10] . A family of zinc finger proteins ( HIM-8 , ZIM-1 , ZIM-2 and ZIM-3 ) that bind to specific PCs mediates functional connections between chromosomes and cytoplasmic dynein and microtubules through a conserved nuclear envelope-spanning complex ( which includes KASH-domain protein ZYG-12 and SUN-domain protein SUN-1 ) that promotes dramatic intranuclear chromosome movements [11]–[17] . In addition to facilitating timely pairing between homologs , these movements also appear to play a role in licensing SC assembly [14] . Dynein-dependent forces acting at PCs have been proposed to function in a checkpoint-like mechanism that makes SC assembly contingent upon homology verification [14] , [18] . Both the ZYG-12/SUN-1 complex and the HORMA domain LE component HTP-1 have been implicated in the operation of this proposed coupling mechanism , serving to impose a barrier to SC assembly that can be overcome by interactions between homologous PCs , thereby preventing SC assembly between nonhomologous chromosomes [14] , [16] , [18] . Further , HTP-1 has also been implicated in a second coordination mechanism that couples termination of chromosome movement with SC installation [18] . Here we identify a nucleoplasmic component of the meiotic machinery , HAL-2 ( homolog alignment-2 ) , that provides a distinct contribution to the coordination of homolog pairing and synapsis . We find that HAL-2 promotes pairing between homologous chromosomes predominantly by counteracting detrimental inhibitory effects of CR precursors ( SYP proteins ) . By preventing uncontrolled behavior of SYP proteins that interferes with PC function and antagonizes HTP-1/2 localization , HAL-2 enables chromosome movement and homology assessment . Additionally , we find that HAL-2 also has SYP-independent roles in regulating progression of meiotic prophase events , by contributing to the coupling mechanism that maintains chromosome mobilization in response to unsynapsed chromosomes and by promoting initiation of recombination .
An essential role for HAL-2 in promoting nuclear reorganization and homolog pairing was revealed through analysis of the hal-2 ( me79 ) mutant , which was isolated in a screen for mutants with defective meiotic chromosome segregation [19] , [20] . Cytological analysis of the hal-2 ( me79 ) mutant revealed defects in chromosome organization during meiotic prophase . In wild-type germ cell nuclei entering meiotic prophase ( in a region of the gonad called the transition zone , TZ ) , chromosomes are dramatically reorganized into a clustered configuration that reflects active chromosome movement ( Figure 1A ) [14] , [15] , [21] , [22] . In the hal-2 mutant , germ cell nuclei entering meiotic prophase lack this characteristic TZ organization and chromosomes are instead dispersed around the nuclear periphery ( Figure 1A ) . Further , whereas wild-type nuclei at the mid-pachytene stage exhibit parallel tracks of DAPI-stained DNA corresponding to aligned homolog pairs , hal-2 mutant nuclei at this stage exhibit disorganized single DAPI tracks ( Figure 1B ) . This organization reflects a severe impairment of homolog pairing in the hal-2 mutant , as assessed at the 5S locus on chromosome V ( by fluorescence in situ hybridization [FISH] ) and at the X chromosome PC ( X-PC ) ( by immunofluorescence [IF] for X-PC binding protein HIM-8 ) ( Figure 1C ) . In both assays , the hal-2 mutant exhibited two widely separated foci in most nuclei ( see quantitation below ) , indicating a failure to achieve homolog pairing . As expected given the pairing defect , diakinesis oocytes in the hal-2 mutant contain 12 univalents , indicating a lack of chiasmata connecting homologous chromosomes ( Figure 1B ) . Absence of chiasmata results in impaired meiotic chromosome segregation , with hal-2 mutant hermaphrodites producing both a high frequency of inviable embryos ( 87 . 2% , n = 1147 ) , indicative of autosomal aneuploidy , and a high incidence of males ( Him ) among the surviving progeny , indicative of X chromosome missegregation ( 32 . 7% , n = 147 ) ( Table S1 ) [23] . Nuclear reorganization and initiation of homologous pairing in the TZs of wild-type worms are accompanied by dramatic chromosomal movements that are mediated by association of PCs and their binding proteins with patches of nuclear envelope ( NE ) proteins that are in turn connected to cytoplasmic microtubules [14]–[16] . SUN-1 , an inner NE protein , is phosphorylated on multiple residues upon meiotic entry in a CHK-2-dependent fashion , and together with the outer NE protein ZYG-12 , is reorganized into NE patches in wild-type TZ nuclei [11] , [14]–[17] . Further , polo-like kinase PLK-2 , which promotes phosphorylation of Ser12 ( S12-Pi ) of SUN-1 , colocalizes during early meiotic prophase with PCs , ZYG-12 and phosphorylated SUN-1 in these NE patches and promotes their formation [24] , [25] . IF analysis revealed that PLK-2 is not localized to NE patches in early prophase nuclei in hal-2 mutants , and SUN-1 S12-Pi is not detected ( Figure 2A ) . Further , even though ZYG-12::GFP was localized to the nuclear envelope in the hal-2 mutant , it was not reorganized into patches , and CHK-2-dependent phosphorylation of Ser8 ( S8-Pi ) of SUN-1 was severely reduced and/or delayed ( Figures 2B and S1 ) . Finally , while live imaging of TZ nuclei in wild-type control germ lines showed chromosome-associated ZYG-12::GFP patches moving rapidly along the nuclear envelope , no evidence of mobile ZYG-12::GFP NE patches was visible in hal-2 mutant nuclei ( Videos S1 and S2 ) . Thus , the modification and reorganization of the NE proteins required for PC-mediated chromosomal movement is severely abrogated in hal-2 mutants . We also assessed recruitment of autosomal PC-binding proteins ZIM-2 ( chromosome V ) and ZIM-3 ( chromosomes I and IV ) in the hal-2 mutant . Whereas the X-PC binding protein HIM-8 is associated with the X-PCs and the nuclear envelope throughout most of the germ line , association of the ZIM proteins with their respective PCs and the nuclear envelope occurs most prominently in the TZ during wild-type meiosis , beginning just prior to the onset of homolog pairing [12] , [13] . We found that while bright NE-associated HIM-8 foci were detected in hal-2 mutant nuclei ( Figures 1C and S2 ) , no bright ZIM-2 or ZIM-3 foci were detected ( Figures 2C and S3A ) , indicating a failure to concentrate ZIM-2 and ZIM-3 to their respective PCs and the nuclear envelope . In summary , our data indicate that hal-2 mutants are defective in the assembly of the linkage between PCs and cytoplasmic microtubules ( via the NE bridge ) required for PC-mediated chromosome movement . During wild-type meiosis , assembly of the SC between successfully paired homologs is initiated in the TZ . LE components such as HIM-3 and HTP-3 [26] , [27] coalesce to form discrete chromosome axes , and these axes are then linked by loading of CR components such as SYP-1 and SYP-3 between them [10] , [28] , [29] . In standard high-resolution IF images , LE components and SYP proteins colocalize at the interface between aligned homologous chromosomes , beginning in the TZ and during the pachytene stage ( Figures 3A and 3B ) . In meiotic prophase nuclei from similar regions of hal-2 mutant germ lines , we detected HTP-3 and HIM-3 localized in tracks along the individual chromosomes , indicating that LEs assemble along the unpaired chromosomes in the hal-2 mutant . Whereas SYP proteins normally load only between paired LEs , however , we found that SYP proteins colocalized with LE components along the lengths of individual unpaired chromosomes in the hal-2 mutant ( Figures 3A and 3B ) . Colocalization of SYP-1 with HTP-3 in the hal-2 mutant was detected at about the same time or soon after HTP-3 began to coalesce into discrete axial structures ( Figure 3A ) . These data indicate that in the absence of HAL-2 , SYP proteins are able to load onto unpaired chromosome axes . Images obtained using 3D-structured illumination microscopy ( 3D-SIM ) [30] support this interpretation . In 3D-SIM images of wild-type pachytene nuclei , the two LEs of the aligned homologs ( visualized using an α-HTP-3 antibody ) can be spatially resolved as two separate entities along much of their lengths , with SYP-1 localized between the two LEs and the associated chromatin of the paired homologs ( Figure 3C and Video S3 ) . In 3D-SIM images of hal-2 mutant pachytene nuclei , SYP-1 was detected mainly in association with single LEs ( rather than between resolvable pairs of LEs ) along the lengths of the conjoined sister chromatids of unaligned individual chromosomes ( Figure 3C and Video S4 ) . Further , in nuclei where short stretches of parallel LE tracks were observed , these corresponded to regions of fold-back near chromosome ends or regions where LEs of two different chromosomes were in close proximity ( Figure S4 ) . Thus , the 3D-SIM images are most consistent with SYP proteins loading along individual LEs in the hal-2 mutant , rather than with formation of SCs between sister chromatid pairs that each had assembled their own LE . Intersister SC formation has been reported to occur in Rec8 null mutant mice [31] and in the budding yeast pds5 meiotic-null mutant [32]; in both cases , the widths of the intersister SCs in the mutants were similar to those of interhomolog SCs in wild-type controls [31] , [32] . While the lack of extensive resolvable pairs of LEs in our 3D-SIM images of hal-2 mutant nuclei rules out the formation of intersister SCs of normal width , however , we cannot exclude the possibility that intersister SC-like structures of smaller width might form in hal-2 mutants . The loading of SYP proteins onto the axes of unaligned individual chromosomes in hal-2 mutants contrasts with a distinct class of C . elegans pairing mutants ( including htp-1 ) in which chromosomes engage in extensive nonhomologous synapsis , with SYP proteins linking the axes of nonhomologous chromosomes [16] , [18] , [26] , [33] . This aspect of the hal-2 phenotype also contrasts with chk-2 mutants , in which pairing is abolished and SYP loading is very delayed [18] , [20] . We found that hal-2; htp-1 double mutants load HTP-3 and SYP-1 extensively onto unpaired chromosome axes ( as in the hal-2 single mutant; Figure S5 ) , indicating that nonhomologous synapsis in the htp-1 mutant is dependent on HAL-2 and suggesting that HAL-2 may be required for SYP proteins to connect chromosome axes together . Likewise , the hal-2; chk-2 double mutant also resembles the hal-2 single mutant with respect to loading of HTP-3 and SYP-1 ( Figure S5 ) , implying that the ability to inhibit SYP loading on unpaired axes requires HAL-2 . The inappropriate loading of SYP proteins onto unpaired homologs in hal-2 mutants raised the possibility that the defects related to homologous pairing might be a consequence of the improper association of SYP proteins with chromosomes . To test this hypothesis , we evaluated several features associated with homologous pairing in hal-2 mutants lacking the SYP proteins . First , we found that removal of SYP proteins ( in hal-2; syp double mutants ) substantially restored multiple features associated with PC-mediated chromosome movement that were eliminated in hal-2 single mutants . Nuclei with clustered chromosomes and NE patches of PLK-2 , ZYG-12::GFP and phosphorylated SUN-1 ( S12-Pi and S8-Pi ) were detected in hal-2; syp-2 and/or syp-3; hal-2 double mutants ( Figures 4A , S3B and S6 ) . Further , live imaging of ZYG-12::GFP in hal-2; syp-2 germ lines indicated that these restored NE patches are competent for movement ( Video S5 ) . After the TZ region , hal-2; syp-2 nuclei also resembled wild-type nuclei in that most nuclei retained a single bright NE focus that was closely associated with HIM-8 , indicating that the ZYG-12/SUN-1 NE patch attached to the X chromosomes persists longer than those associated with the autosomes ( Figure S7 ) [14] , [15] . In addition , localization of the autosomal PC-binding proteins ZIM-2 and ZIM-3 to bright foci at the nuclear periphery in nuclei with clustered chromosomes was also observed in hal-2; syp-2 double mutants ( Figures 4B and S3A ) . Finally , restoration of chromosome clustering and SUN-1 S8-Pi in the syp-3; hal-2 mutants is dependent on CHK-2 , as these features were absent in syp-3; hal-2; chk-2 triple mutants ( Figure S6 ) . Taken together , these findings indicate that CHK-2-/PLK-2-dependent PC-mediated chromosome mobilization can be inhibited by SYP proteins , and that HAL-2 antagonizes this inhibitory effect of SYP proteins on PC function . Second , we found evidence that significant homolog pairing occurred in hal-2 mutants that lacked the SYP proteins . The syp mutants are proficient for establishment of pairing at PCs and can achieve a significant degree of lengthwise homolog alignment , but are defective in stabilization and maintenance of homologous associations [10] , [28] , [34]–[36]; this made it possible to test whether significant homolog pairing occurred in hal-2; syp double mutants . We quantified levels of pairing at the X chromosome PCs ( using IF for HIM-8 ) , at chromosome V PCs ( V-PCs ) ( using IF for ZIM-2 ) and at an internal position of chromosome V ( using FISH for the 5S rDNA locus ) ( Figures 4D–F ) . For all of these analyses , gonads were divided into 5 zones of equal lengths , as indicated in Figure 4C . Pairing at the X-PC was quantified from Zones 2–5 ( Figure 4D ) , as HIM-8 localization at the X-PCs is most robust from the TZ through late pachytene [13] . Whereas pairing at the X-PCs was effectively abolished in the hal-2 single mutant , high levels of pairing were restored in the hal-2; syp-2 double mutant , with X-PC pairing levels peaking in Zones 3 and 4 at levels approaching those of syp-2 and wild-type controls ( Figure 4D ) . Although X-PC pairing was substantially restored in the hal-2; syp-2 double mutant , however , pairing in all zones was significantly lower than in the syp-2 single mutant , suggesting that restoration of synapsis–independent maintenance of PC pairing may be incomplete . ZIM-2 pairing levels were analyzed specifically for Zone 2 , the zone where ZIM-2 foci are most prominently detected during normal meiosis ( Figure 4E ) . In wild-type gonads , Zone 2 includes some premeiotic nuclei , the TZ , and some early pachytene nuclei; hence , the wild-type data set included nuclei with two bright ZIM-2 foci , nuclei with a single focus ( indicative of paired PCs ) , and some nuclei without ZIM-2 foci . As described above , hal-2 mutants lack nuclei with clear bright ZIM-2 foci , and these foci are restored in the hal-2; syp-2 double mutants ( Figure 4B ) . Among the nuclei with ZIM-2 foci in Zone 2 , the proportion of nuclei in which foci were paired in the hal-2; syp-2 double mutant was similar to the proportions observed for both wild-type and syp-2 mutant gonads ( Figure 4E ) , indicating highly successful pairing at the V-PCs in early meiotic prophase nuclei with ZIM-2 foci in the hal-2; syp-2 double mutants . Whereas ZIM-2 foci ( and other markers of PC-mediated movement ) persisted into Zones 3 and 4 in syp-2 mutant gonads , ZIM-2 foci were not detected beyond Zone 2 in the hal-2; syp-2 double mutants ( data not shown ) , thus precluding assessment of pairing at V-PCs throughout a larger portion of the hal-2; syp-2 mutant germ lines . Pairing at the 5S locus on chromosome V was assessed for all five zones of the gonad ( Figure 4F ) . While pairing at this locus was abolished throughout the gonad in the hal-2 single mutant , the syp-3; hal-2 double mutant exhibited a highly significant increase in pairing over the hal-2 single mutant specifically in Zone 2 ( p<0 . 0001 ) , the same zone in which chromosome V PCs were shown to be active in the hal-2; syp-2 double mutant . However , increases in pairing were not detected in other zones and pairing levels for syp-3; hal-2 for Zones 2–5 were consistently significantly lower than in syp-3 controls . As markers of active PCs persist beyond Zone 2 at the X-PCs but not at the autosomal PCs ( in both hal-2; syp-2 and wild-type ) , it is likely that the lower level of pairing restoration observed for chromosome V upon removal of SYP proteins reflects a difference between the X chromosome and autosomes in the duration of PC activity . Taken together , our analyses show that significant homologous pairing at the PCs can occur in hal-2 mutants when SYP proteins are absent , indicating that the SYP proteins can interfere with pairing , likely through inappropriate association with unaligned chromosomes . These findings imply that HAL-2 promotes PC function by counteracting inhibitory effects of the SYP proteins . However , incomplete restoration of pairing also indicates that HAL-2 plays additional role ( s ) in homolog alignment beyond antagonizing inhibitory effects of SYP proteins on PC function ( see below ) . hal-2 mutants show several phenotypic similarities with worms lacking both HTP-1 and HTP-2 ( LE components that are paralogs of HIM-3 and HTP-3 ) , including association of SYP proteins with the axes of unpaired chromosomes in pachytene nuclei [33] , [37] . Thus , we assessed localization of HTP-1/2 in hal-2 mutant gonads by IF . ( Since HTP-1 and HTP-2 are recognized by the same antibody , they are referred to as HTP-1/2 [37] . ) In wild-type germ lines , HTP-1/2 colocalize with other SC components ( including the SYP proteins ) along the full lengths of paired homologs until late pachytene , when HTP-1/2 and SYP proteins begin to become redistributed in a crossover-dependent fashion to reciprocal chromosomal domains such that their localization is almost mutually exclusive by diakinesis , with HTP-1/2 localizing on the long arms of the bivalent and SYP proteins localizing on the short arms ( Figure 5 ) [37] , [38] . The localization of HTP-1/2 on diakinesis chromosomes is markedly different from the other axis components , such as HTP-3 , which localize to all four arms of the bivalent ( Figure 5B ) [27] , [37] . We found that HTP-1/2 localization was impaired in the hal-2 mutant . During the pachytene stage , faint HTP-1/2 staining was detected along the axes of the unpaired chromosomes , but the level of HTP-1/2 relative to HTP-3 was much lower than in wild-type controls ( Figure 5A ) . Further , while HTP-3 was present on univalent chromosomes of diakinesis-stage oocytes in the hal-2 mutant , HTP-1/2 staining was undetectable ( Figure 5B ) . The fact that SYP proteins load improperly onto unaligned chromosome axes in hal-2 mutants raised the possibility that the SYP proteins might be preventing the normal localization of HTP-1/2 . Consistent with this hypothesis , HTP-1/2 localization was substantially restored in hal-2; syp-2 double mutant gonads , both at the pachytene and diakinesis stages ( Figure 5 ) . These results provide additional support for a previously proposed idea that SYP proteins and HTP-1/2 are incompatible , each antagonizing the chromosomal localization of the other [37] . Further , the data suggest that HAL-2 plays a role in promoting a temporary compatibility between SYP proteins and HTP-1/2 that enables them to coexist along the lengths of chromosomes from zygotene through mid-pachytene stages during wild-type meiosis . Mapping of the hal-2 ( me79 ) mutation , phenocopy by RNAi , and identification of a nonsense mutation in the me79 allele by sequence analysis identified T16H12 . 11 as the hal-2 gene ( see Materials and Methods ) ( Figure 6A ) . An independently generated deletion allele , tm4960 , is missing 596 bp of the 927-bp coding region ( Figure 6A ) and fails to complement me79 , confirming the molecular identity of hal-2 as T16H12 . 11 . The two alleles cause similar phenotypic defects and both are likely null alleles; all of the analyses reported here were performed using the me79 mutant unless otherwise specified . Although the T16H12 . 11 gene model in WormBase ( release WS227 ) predicts a 295 aa protein , we propose a modified gene model ( that uses an upstream ATG ) based on upstream conservation of amino acid sequence in other Caenorhabditis orthologs ( Figure S8 ) . The revised gene model encodes a predicted 308 aa protein with MW of 36 . 1 kDa . An antibody raised against the C-terminal 100 amino acids of HAL-2 detects a band of slightly less than 37 kDa in Western blot analysis of wild-type whole worm lysates , consistent with the protein size predicted by our revised gene model . This band was absent in lysates from both hal-2 ( me79 ) and hal-2 ( tm4960 ) mutant worms ( Figure 6B ) , indicating that no full-length protein is produced in these mutants . Bioinformatics analyses using PSI-BLAST and several protein structure prediction servers yielded little information regarding HAL-2 function . No homologs were detected outside of the Caenorhabditis genus , and no conserved motifs were detected by searches of the Conserved Domain Database . Further , while HAL-2 orthologs were found in other Caenorhabditis species , they exhibited relatively low sequence conservation ( 52% amino acid identity between ortholog pairs ) . This differs from conservation levels observed within Caenorhabditis for proteins implicated directly in meiotic recombination ( which range from 74–98% amino acid identity ) and is more in line with conservation levels for SYP proteins ortholog pairs ( which range from 41–61% amino acid identity ) . Analysis using the COILS and Paircoil2 programs [39] , [40] suggests that a short domain ( 33–46 amino acids ) in HAL-2 orthologs may have the capacity to adopt a coiled-coil conformation , although the probability of this conformation differed widely among HAL-2 orthologs and reached significance threshold only for a subset of them . IF analysis using our α-HAL-2 antibody demonstrated that the HAL-2 protein concentrates in the nuclei of germ cells . HAL-2 staining was first detected in germ cell nuclei before they entered the TZ and gradually weakened during diplotene until no signal was observed by the end of diakinesis ( Figures 6C–E ) . As the majority of the IF signal did not colocalize with DAPI-stained chromatin , nucleoli ( data not shown ) or SC ( see below ) , we infer that HAL-2 is predominantly nucleoplasmic . This nucleoplasmic staining was detected in wild-type gonads but was absent in hal-2 mutant gonads ( Figure 6E ) , further verifying the specificity of the antibody . We also generated a functional GFP::HAL-2 transgene that rescues the Him and progeny lethality phenotypes of hal-2 mutants ( Table S1 ) . GFP::HAL-2 expressed from this transgene exhibited a similar nucleoplasmic localization in germ cells , although the protein persisted beyond diplotene ( Figure 6F and data not shown ) , likely because of heterologous promoter and 3′UTR sequences . Since our phenotypic analyses indicated that HAL-2 plays an important role in restraining the behavior of the SYP proteins , we used IF to investigate their spatial relationship in situ . In wild-type germ cell nuclei , little or no overlap was detected between the localization of HAL-2 , which concentrates in the nucleoplasm , and SYP-1 , which concentrates at the interface between aligned chromosomes ( Figure 7 ) . However , when we analyzed the localization of HAL-2 in the scc-3 cohesin mutant , in which chromosome axes and SC assembly are severely disrupted and LE precursors ( including the other cohesin subunits ) and SYP proteins form aggregates ( presumably polycomplexes ) in the nuclei instead of loading extensively onto chromosomes , we detected HAL-2 both in the nucleoplasm and concentrated together with SYP-1 in the nuclear aggregates ( Figure 7 ) [27] , [41] . Localization of HAL-2 in nuclear aggregates is dependent on SYP-1 , as HAL-2 nuclear aggregates are not observed in scc-3; syp-1 germ cell nuclei ( Figure 7 ) . Further , we found that HAL-2 colocalized with SYP-1 in nuclear aggregates in an htp-3 mutant , in which LEs do not assemble along the chromosomes but cohesin components are also not present in the nuclear aggregates ( Figure 7 ) [42] . Finally , we examined HAL-2 localization in a him-3 null mutant , which does assemble LEs that contain cohesin complexes , HTP-1/2 and HTP-3 , but is severely defective for SC assembly [26] , [27] , [37] , [42] . In the him-3 mutant , SYP-1 appears as small nuclear aggregates at early pachytene that elongate into short stretches by late pachytene , and HAL-2 was found both in the nucleoplasm and colocalized together with SYP-1 in these structures ( Figure 7 ) . HAL-2 is not required for formation of these SYP-1 nuclear aggregates , however , as SYP-1 aggregates were still formed in hal-2; him-3 double mutants ( Figure S9 ) . Together , these analyses of SYP-dependent localization of HAL-2 to nuclear aggregates in various mutants with defective LE and SC assembly raise the possibility that HAL-2 might interact with SYP proteins or other SC-associated proteins; however these IF analyses do not address whether such potential interactions are direct or indirect . Consistent with our finding that homolog pairing in the hal-2; syp double mutants is not restored to the levels observed in syp single mutants , closer examination of DAPI-stained chromosome morphology , SUN-1 phosphorylation , ZYG-12::GFP patches and PLK-2 localization in hal-2; syp-2 gonads provided further evidence that HAL-2 has additional roles in meiosis beyond regulating SYP loading ( Figures 8A , S10 and data not shown ) . In contrast to the dramatically extended TZ observed in syp-2 mutant gonads , in which nuclei with clustered chromosomes and multiple bright NE patches reflecting ongoing chromosome mobilization persist until near the very end of the pachytene region ( Figure 8A ) [14] , [15] , [21] , [34] , hal-2; syp-2 mutant gonads lack persistent chromosome clustering , and only the single NE focus associated with HIM-8 persists beyond the TZ , as in wild-type gonads ( Figures 8A and S7 ) . The fact that hal-2; syp-2 nuclei can exit from the clustered chromosome organization despite lack of synapsis suggests that HAL-2 may be required for normal functioning of a checkpoint-like mechanism that operates to make redispersal of clustered chromosomes contingent upon SC assembly [18] . We also found that hal-2 mutants exhibit severely reduced levels of immunostaining for DNA strand exchange protein RAD-51 [34] , [43] . Whereas multiple RAD-51 foci were detected in most early/mid-pachytene nuclei in wild-type germ lines , most nuclei in hal-2 mutant germ lines lacked RAD-51 foci ( Figure 8B ) , suggesting that initiation of recombination may be impaired . Further , this deficit of RAD-51 foci was not suppressed in the syp-3; hal-2 double mutant , which also lacked RAD-51 foci in most nuclei , indicating that this phenotype is not a secondary consequence of inappropriate SYP loading . The reduced levels of RAD-51 foci in the syp-3; hal-2 double mutant contrasts with the increased levels of foci observed in the syp-3 single mutant ( Figure 8B ) [44] , indicating that absence of HAL-2 prevents accumulation of RAD-51 foci . Reduction in RAD-51 foci in hal-2 mutants suggests either that hal-2 mutants have a reduced number of double-strand DNA breaks ( DSBs ) or that HAL-2 is needed for efficient loading of RAD-51 onto DSBs . To test whether HAL-2 is required for accumulation of RAD-51 on DSBs , we induced DSBs with ionizing irradiation in hal-2 animals and dissected them 1 hour post irradiation for RAD-51 immunostaining . Abundant RAD-51 foci were detected in nuclei throughout the irradiated hal-2 mutant germ lines ( Figure 8C ) , indicating that HAL-2 is not required for accumulation of RAD-51 on DSBs . Taken together , these data suggest that HAL-2 has a role in promoting initiation of recombination that is distinct from its role in regulating SYP loading . Variable defects in PC-mediated chromosome clustering in combination with impaired DSB formation had been reported previously for the C . elegans him-19 [45] mutant . Moreover , some of the defects in PC function in him-19 mutants were rescued by the introduction of DSBs by irradiation [45] , raising the possibility that defects in PC function in hal-2 mutants might be similarly rescued by artificial induction of DSBs . We tested this by inducing DSBs in hal-2 mutants by ionizing irradiation and dissecting the animals 2 hours post irradiation for IF analysis . No chromosome clustering , recruitment of bright ZIM-3 foci to the PCs and nuclear envelope , or NE patches of phosphorylated SUN-1 S12-Pi were observed in the irradiated hal-2 gonads ( Figure S11 ) , indicating that these defects in the hal-2 mutant are not secondary consequences of its defect in recombination initiation .
The work reported here highlights the potential of SC central region ( CR ) precursors to exert detrimental effects that interfere with the process of homolog pairing . This potential emphasizes the need for SC precursors to be carefully shepherded during early meiotic prophase to avoid antagonizing the very process that they normally serve to stabilize . One way that CR precursors can cause trouble is by inappropriately linking the axes of nonhomologous chromosomes . If transient nonhomologous interactions that occur during early prophase are stabilized by synapsis , chromosomes will then have fewer chances to associate and pair with their correct partners . This is illustrated by the improved pairing observed when SYP proteins ( and nonhomologous synapsis ) were eliminated in a mutant with impaired SUN-1 function [16] . Nonhomologous synapsis may also prevent and/or restrain the active prophase chromosomal movements that promote pairing , as illustrated by the amelioration of restrained SUN-1::GFP NE patch movement in an htp-1 mutant by the removal of SYP proteins [21] . Our comparative analyses of hal-2; syp double mutants with hal-2 single mutants have revealed several additional ways in which CR precursors can be detrimental . Firstly , SYP proteins can interfere with pairing by inhibiting CHK-2-dependent PC activities at the onset of meiotic prophase . In hal-2 mutants , NE protein SUN-1 is not phosphorylated on Ser12 ( and S8-Pi is severely reduced and/or delayed ) , PCs fail to recruit PLK-2 , autosomal PC-binding proteins fail to concentrate at their respective PCs or the NE , and no ZYG-12::GFP NE patches or PC-mediated chromosome movements are observed . Coordinate loss of these features suggests that lack of pairing in hal-2 mutants likely results primarily from the abrogation of vital CHK-2-dependent PC functions . Moreover , the fact that all of these problems are alleviated by removal of SYP proteins implies that the unconstrained behavior of SYP proteins in the absence of HAL-2 is responsible for preventing PC function in this context . Secondly , SYP proteins can interfere with proper chromosomal association of HORMA domain proteins HTP-1/2 . An incompatibility between CR proteins and ( some ) meiosis-specific HORMA domain proteins is evident from observations of apparently opposing spatial relationships of these classes of proteins along chromosomes in several different organisms . In S . cerevisiae , HORMA protein Hop1 and CR protein Zip1 exhibit largely complementary patterns of high and low abundance along pachytene chromosomes [46] . Further , mouse HORMAD1 and HORMAD2 lose their chromosomal localization upon installation of CR protein SYCP1 [47] . In C . elegans , all four HORMA proteins ( HTP-3 , HIM-3 and HTP-1/2 ) coexist with the SYP proteins along the lengths of chromosomes for most of the pachytene stage , but at late pachytene HTP-1/2 and SYP proteins are triggered by ( nascent ) COs to relocalize to reciprocal chromosomal domains [27] , [37] , [38] . These patterns all suggest a mutually antagonistic relationship between CR proteins and ( some ) HORMA proteins . In C . elegans , the coexistence of HTP-1/2 and SYP proteins from zygotene until late pachytene further implies a mechanism that temporarily counteracts this antagonism . The temporary compatibility between these proteins during early prophase requires HAL-2 , as our data show that the unconstrained behavior of SYP proteins in hal-2 mutants inhibits axis association of HTP-1/2 during early prophase . Since HTP-1/2 is essential for homolog pairing [18] , [33] , reduced chromosomal HTP-1/2 likely contributes to the pairing defect in hal-2 mutants . Taken together , our data indicate that HAL-2 plays a major role in homolog pairing by restricting the detrimental behavior of SYP proteins and constraining their loading to the appropriate context . Moreover , the contributions of HAL-2 to the coordination of pairing and synapsis are distinct from and parallel to the contribution of a previously identified checkpoint-like mechanism that ensures that SC central region assembly is contingent upon successful homology verification [14] , [18] . LE component HTP-1 , NE protein SUN-1 , and dynein-mediated chromosome movements were shown to be involved in this mechanism [14] , [16] , [18] . Inhibition of dynein causes a significant delay between pairing and SYP loading ( implying that motion and/or exertion of tension is required to license SYP loading ) , while HTP-1 loss or impaired SUN-1 function bypasses the dynein requirement and leads to nonhomologous synapsis ( suggesting that these proteins are required to generate or respond to a “wait synapsis” signal ) . In contrast , neither pairing nor nonhomologous synapsis occurs in hal-2 mutants , and instead , SYP proteins load along the axes of individual unpaired chromosomes . This suggests the possibility that HAL-2 might function in promoting maturation of CR precursors , such that upon licensing of SC assembly , SYP installation invariably occurs in a manner that links LEs together . Thus , HAL-2 may have two potentially inter-related roles in regulating the behavior of SYP proteins: i ) shepherding/restraining CR subunits to prevent them from wreaking havoc in the nucleus and ii ) enabling their timely maturation into a form that can be assembled cooperatively into mature SCs . These proposed roles suggest that HAL-2 might serve in the capacity of a chaperone to help ensure timely and appropriate SC assembly . Although the HAL-2 sequence does not identify it as a member of previously known chaperone families , there are numerous striking parallels between HAL-2 and Fpr3 , a prolyl isomerase that regulates SC assembly during budding yeast meiosis [48] . Like HAL-2 , Fpr3 localizes in the nucleoplasm of meiotic cells and colocalizes with CR precursors in polycomplexes that form when SC assembly is prevented [48] . Further , like HAL-2 , Fpr3 functions in parallel with other mechanisms to prevent assembly of CR precursors onto chromosomes until licensing conditions have been met [48] . Thus , both HAL-2 and Fpr3 can be seen as guardians of CR precursors , inhibiting their inappropriate behavior . The cooperative and processive nature of SC assembly and the inherent tendency for self-assembly of its precursors [1] present a challenge for meiotic cells: they must accumulate large pools of precursors needed to accomplish rapid SC assembly , while at the same time preventing these precursors from aggregating into nonfunctional structures and/or interfering with prerequisite events . In C . elegans , the SC central region consists of four interdependent coiled-coil SYP proteins that interact to span the distance between the LEs [10] , [28] , [29] , [34] , [35] . Thus , involvement of regulatory proteins not included in the mature SC ( such as HAL-2 ) represents a practical solution for controlling maturation of these CR precursors and constraining their loading to occur only in the appropriate context . In addition to HAL-2 , assembly of the SC central region also requires CRA-1 [49] , the C . elegans ortholog of the non-catalytic subunit of the NatB N-terminal acetyltransferase complex , which acetylates the N-terminal methionine of proteins containing Met-Asp- , Met-Glu- or Met-Asn- [50] , [51] . As in hal-2 mutants , SYP proteins associate with unpaired LEs in cra-1 mutants [49] , raising the possibility that maturation of SC precursors may require N-terminal acetylation of one or more subunits . While we have also suggested a role for HAL-2 in promoting CR subunits maturation , it is clear that HAL-2 and CRA-1 make different contributions to regulating SYP behavior . Loading of SYP proteins onto chromosome axes in cra-1 mutants occurs later than in hal-2 mutants and does not interfere with homolog pairing [49] . Further , whereas association of SYP proteins with unpaired chromosomes in cra-1 mutants is dependent on DSB formation and progression of meiotic DSB repair processes [49] , SYP proteins load onto unpaired chromosomes in hal-2 mutants in the absence of DSBs ( as in hal-2; spo-11 double mutants; Figure S5 ) , indicating that the chromosomal association of SYP proteins in hal-2 mutants is not dependent on DSBs or recombination intermediates . In addition to the checkpoint-like mechanism that couples SC assembly to homology verification , a related but distinct mechanism operates during C . elegans meiosis to render redispersal of clustered chromosomes and termination of active chromosome motion contingent upon SC assembly [18] , [28] . HAL-2 may also play a role in this coupling mechanism that is distinct from its role in preventing SYP proteins from interfering with pairing . This is suggested by the fact that chromosome clustering and PC-mediated chromosome mobilization do not persist in hal-2; syp double mutants , in which synapsis cannot occur . It has been postulated that chromosome redispersal is regulated by a signaling mechanism that monitors synapsis progression , and HTP-1 has been suggested to be involved in generating and/or responding to an inhibitory “wait dispersal” signal that maintains chromosome clustering [18] . Since HTP-1/2 loading onto chromosomes is largely restored by removal of SYP proteins ( in hal-2; syp double mutants ) , the inability to maintain chromosome clustering is not a consequence of lack of HTP-1 in the hal-2 mutant , but it may reflect a cooperation of HAL-2 with HTP-1 in regulating or responding to the “wait dispersal” inhibitory signal . Implicit in such a coupling mechanism is that progression of synapsis must generate “start dispersal” activation signals to counteract the existing inhibitory signals . The SYP proteins themselves likely contribute , as suggested by analysis of the syp-3 ( me42 ) mutant , which expresses C-terminally truncated SYP-3 [28] . In contrast to the persistent chromosome clustering observed in syp null mutants , a premature exit from chromosome clustering is observed in the syp-3 ( me42 ) mutant; it was proposed that association of SYP proteins with unpaired chromosomes in this mutant may act as the trigger for chromosome redispersal , as removal of other SYP proteins in the syp-3 ( me42 ) mutant restored persistent chromosome clustering [28] . This contrasts with the hal-2 mutant , in which elimination of SYP proteins did not lead to persistent clustering , further supporting the notion that HAL-2 may play a SYP-independent role in the coupling mechanism and/or the maintenance of chromosome movement .
C . elegans strains were cultivated at 20°C under standard conditions [52] , unless otherwise specified . For experiments involving meiotic mutants , homozygous mutants were picked from the progeny of heterozygous balanced parents by the absence of dominant markers associated with the balancers and/or by the presence of a recessive marker cis-linked with the mutant allele . For Figure S5 , the spo-11 single mutant controls imaged were heterozygous for the hal-2 mutation as they were siblings of hal-2; spo-11 double mutants , both of which were derived from doubly balanced heterozygous parents . All experiments involving transgenic strains were performed in the presence of the wild-type gene unless otherwise noted . A list of strains used in this study is provided in Text S1 . hal-2 ( me79 ) was generated by EMS mutagenesis and isolated by screening for “Green eggs and Him” [19] and defects in prophase chromosome morphology and organization [20] . hal-2 ( me79 ) was mapped to a ∼126 kb region between positions 10057573 and 10183653 on chromosome III using standard crosses and SNP mapping [53] . The interval contained 28 known or predicted genes; complementation tests and RNAi led to the identification of T16H12 . 11 as the likely hal-2 gene . RNAi targeted to the conserved region of T16H12 . 11 produced a partial hal-2 phenocopy: no apparent TZ , unaligned chromosomes at pachytene and 8–10 DAPI-stained bodies at diakinesis . Sequencing of T16H12 . 11 in the hal-2 ( me79 ) mutant revealed a C-to-T transition resulting in a premature stop ( instead of glutamine ) encoded at codon 96 of the 308 amino acid coding sequence . hal-2 ( tm4960 ) was provided by Dr . S . Mitani at the C . elegans National BioResource Project , NIG , Japan; the 761 bp deletion was confirmed by PCR and sequencing . L4 animals were selected and placed at 25°C for 20 hours before dissection . Worms were dissected in 10 µl of M9 containing tricaine ( 0 . 1% ) , tetramisole ( 0 . 01% ) and Hoechst 33342 ( Invitrogen; 1∶1000 ) on a coverslip . The coverslip with the dissected worms was flipped onto a 2% agarose pad and imaged immediately using the DeltaVision deconvolution microscopy system with a 60× oil objective with 1 . 5× optivar . Images were acquired as stacks of 3 optical sections at 1 . 95 µm intervals every 15 s over a 5 min time period . Exposure times were kept constant for all images . Videos show maximum intensity projections of the 3D stacks displayed at 2 frames/s and were made with the Volocity ( PerkinElmer ) software . For RAD-51 immunostaining , 20 hours post-L4 adults were exposed to 1 krad of γ irradiation from a Cs-137 source and dissected and fixed one hour post irradiation ( 21 hours post-L4 ) . For ZIM-3 and SUN-1 S12-Pi immunostaining , 20 hours post-L4 adults were exposed to 5 krad of γ irradiation and dissected two hours post irradiation ( 22 hours post-L4 ) . Unirradiated age-matched adults were dissected for immunostaining as controls for all irradiation experiments . | For successful segregation of homologous chromosomes during sexual reproduction , homologs must first identify and pair with their correct partners . Further , many organisms stabilize and maintain alignment between paired homologs through assembly of a highly ordered structure known as the synaptonemal complex ( SC ) . Pairing and synapsis must be tightly coordinated to ensure that SC assembly only occurs in a productive manner , linking the axes of correctly aligned homologous chromosomes . In this work , we identify HAL-2 , a protein that concentrates in the nucleoplasm of germ cells , as a key player in mediating this coordination . We find that precursors of the SC have the potential to inhibit homolog pairing , interfering with the very process that the SC normally serves to stabilize . Moreover , we show that HAL-2 promotes homolog pairing and associated chromosome movement primarily by counteracting these detrimental inhibitory effects of SC precursors . Our data suggest that HAL-2 serves to prevent inappropriate association of SC precursors with chromosomes prior to licensing of SC assembly , and we propose that HAL-2 may enable precursors to accumulate in a manner that allows rapid , cooperative SC assembly upon homology verification . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"chromosomal",
"inheritance",
"phenotypes",
"heredity",
"gene",
"identification",
"and",
"analysis",
"genetics",
"molecular",
"genetics",
"biology",
"genetics",
"and",
"genomics",
"gene",
"function"
] | 2012 | HAL-2 Promotes Homologous Pairing during Caenorhabditis elegans Meiosis by Antagonizing Inhibitory Effects of Synaptonemal Complex Precursors |
Fatal outcomes of Ebola virus ( EBOV ) infections are typically preceded by a ‘sepsis-like’ syndrome and lymphopenia despite T cells being resistant to Ebola infection . The mechanisms that lead to T lymphocytes death remain largely unknown; however , the degree of lymphopenia is highly correlative with fatalities . Here we investigated whether the addition of EBOV or its envelope glycoprotein ( GP ) to isolated primary human CD4+ T cells induced cell death . We observed a significant decrease in cell viability in a GP-dependent manner , which is suggestive of a direct role of GP in T cell death . Using immunoprecipitation assays and flow cytometry , we demonstrate that EBOV directly binds to CD4+ T cells through interaction of GP with TLR4 . Transcriptome analysis revealed that the addition of EBOV to CD4+ T cells results in the significant upregulation of pathways associated with interferon signaling , pattern recognition receptors and intracellular activation of NFκB signaling pathway . Both transcriptome analysis and specific inhibitors allowed identification of apoptosis and necrosis as mechanisms associated with the observed T cell death following exposure to EBOV . The addition of the TLR4 inhibitor CLI-095 significantly reduced CD4+ T cell death induced by GP . EBOV stimulation of primary CD4+ T cells resulted in a significant increase in secreted TNFα; inhibition of TNFα-mediated signaling events significantly reduced T cell death while inhibitors of both necrosis and apoptosis similarly reduced EBOV-induced T cell death . Lastly , we show that stimulation with EBOV or GP augments monocyte maturation as determined by an overall increase in expression levels of markers of differentiation . Subsequently , the increased rates of cellular differentiation resulted in higher rates of infection further contributing to T cell death . These results demonstrate that GP directly subverts the host’s immune response by increasing the susceptibility of monocytes to EBOV infection and triggering lymphopenia through direct and indirect mechanisms .
Ebola virus ( EBOV ) is one of the deadliest pathogens known to exist as evidenced by the latest outbreak in West Africa that resulted in more than 28 , 000 confirmed and suspected infections including more than 11 , 000 fatalities [1] . Currently , experimental EBOV candidate vaccines and monoclonal antibody-based therapies are being tested in clinical trials [2 , 3]; however , none have yet to be approved for treatment of infected patients . Gaining an in depth understanding of the mechanisms of EBOV’s unparalleled ability to counteract and disrupt the immune response is critical to developing targeted approaches aimed at reducing the pathogenesis directly or indirectly caused by the virus . A characteristic feature of EBOV infection is the rapid onset of lymphopenia , which is observed in both humans and experimentally infected non-human primates ( NHP ) [4–11] . Development of lymphopenia is typically observed in EBOV patients that succumb to disease , whereas survivors have been shown to maintain CD3+ T lymphocyte populations throughout the course of disease [12 , 13] . Strikingly , lymphopenia occurs despite the inability of EBOV to infect lymphocytes [4] . On the other hand , dendritic cells ( DCs ) and cells derived from monocyte-macrophage lineages are among the primary targets of EBOV infection in vivo [11 , 14] . EBOV infection of these cells results in their aberrant activation [15–18] and induction of Fas and tumor necrosis factor related cell death inducing factor ( TRAIL ) pathways [19 , 20] . We and others recently demonstrated that the lack of proper maturation of these critical antigen presenting cells ( APCs ) results in a limited activation of antigen-specific T lymphocytes [21 , 22] further contributing to the deficient adaptive immune response . In addition , infection of monocyte-macrophage lineage may also contribute to EBOV-associated pathogenesis by several mechanisms including the release of inflammatory mediators , which may contribute to their apoptosis and necrosis [23] . The only EBOV envelope glycoprotein ( GP ) was shown to bind and activate the TLR4 signaling pathway [24] . TLR4 is known to trigger both apoptotic and necrotic pathways via direct and/or indirect activation of infected cells or bystander cells , which may contribute to these inflammatory mechanisms [25] . Lastly , TLR4 stimulation of cells of the monocyte/macrophage lineage and DCs results in their activation and/or differentiation [26 , 27] . Cellular differentiation may further contribute to the release of inflammatory mediators associated with the onset of a cytokine storm , which is a characteristic feature of EBOV infection [28–30] . Since lymphocytes are resistant to EBOV infection , the mechanisms causing lymphopenia during EBOV infection remain largely unknown . Hence , the primary goal of this study was to examine whether EBOV directly stimulates T lymphocytes and determine the direct and indirect role of TLR4 in mediating T cell death in the pathogenesis of EBOV infection .
As lymphopenia is a common feature observed in fatal cases following EBOV infection [4–11] , we first sought to determine if a direct interaction of EBOV with DCs can cause T cell death . To visualize the infection , we used a recombinant EBOV expressing enhanced green fluorescent protein ( GFP ) from an added transcriptional cassette ( EBOV-GFP ) ; this virus replicates in cultured cells at the same level as wt EBOV [31] . Human monocyte-derived DCs and autologous T lymphocytes were co-cultured with EBOV-GFP for 7 days , and the percentages of apoptotic CD4+ or CD8+ T cells were determined by annexin-V staining . We also included primary lymphocyte cultures in which highly purified naïve or CD3/CD28 activated CD4+ T cells were exposed directly to EBOV-GFP in the absence of DCs or monocytes . Culturing of CD4+ T cells in the presence of EBOV-infected mature ( by adding TNFα ) or immature DCs resulted in a significant increase of apoptotic cells ( Fig 1A ) . Unexpectedly , the highest level of cell death was observed when isolated CD4+ T cells were cultured alone in EBOV-containing medium and this effect was observed to be dose-dependent ( Fig 1B , S1A and S1B Fig ) . Similar results were observed with isolated CD8+ T cells ( S1C Fig ) . We also observed an elevated level of proliferation of CD4+ T cells cultured with EBOV alone or with EBOV-infected immature DCs , but not CD3/CD28 bead-stimulated or mature DCs ( Fig 1C ) . Similarly , EBOV induced proliferation of CD8+ T cells ( Fig 1D , S1D Fig ) . These data suggest that EBOV is capable of inducing non-specific proliferation of lymphocytes . Five days after addition of EBOV-GFP , we detected a dose-dependent increase in the percentages of dead CD4+ T cells or cells positive for activated caspase-8 and caspase-9 , as well as proliferated cells ( Fig 1E ) . The addition of the inhibitor of apoptosis z-VAD-FMK significantly reduced cell death associated with EBOV; however , proliferation remained unabated ( Fig 1F and 1G ) . Similarly , the addition of the pro-survival cytokines IL-7 and IL-15 significantly reduced the percentages of apoptotic CD4+ and CD8+ T cells cultured in the presence of EBOV ( S1E and S1F Fig ) . The infectivity of EBOV incubated under the experimental conditions used for these studies was determined by daily collection of aliquots from cell-free medium , their flash freezing , infection of Vero E6 cells and flow cytometry analysis of the percentages of GFP+ cells . A moderate reduction of infectivity not exceeding 32% was detected on days 1–3 , followed by ~66% reduction of infectivity on days 4–6 ( data now shown ) . These data demonstrate that EBOV directly triggers apoptotic death of T cells , despite the lack of their infection . We next tested if death of T lymphocytes exposed to EBOV is caused by binding of GP . We used the chimeric parainfluenza virus type 3 in which its envelope proteins HN and F are replaced with EBOV GP ( HPIV3/ΔF-HN/EboGP ) ; the structure of GP at the surface of this chimeric virus has been shown to be identical to that found on the surface of EBOV particles [32] . Total PBMCs or purified CD3+ T lymphocytes were cultured for 1 , 4 or 7 days in the presence of HPIV3 or HPIV3/ΔF-HN/EboGP and stained with annexin-V . Cultivation with HPIV3/ΔF-HN/EboGP resulted in a significant increase compared to HPIV3 in the percentages of dead ( annexin V+ ) cells in gated CD3+ T cells in total PBMCs ( Fig 2A ) and in purified CD3+ T cells directly exposed to EBOV ( Fig 2B ) . We also sought to evaluate cell death and the role of EBOV GP by using the chimeric vesicular stomatitis virus VSVΔG/ZEBOVGP in which the sole envelope protein G was replaced with EBOV GP [33] . A 4 day-long incubation of SupT1 , a lymphoblastoid CD4+ T cell line , with this virus resulted in a dose-dependent cell death ( S2A and S2B Fig ) . The findings suggest that EBOV-induced apoptosis of T cells is directly associated with EBOV GP . We next determined if exposure of lymphocytes to GP activates apoptotic pathways . Jurkat cells were cultured in presence of HPIV3/ΔF-HN/EboGP , HPIV3 or staurosporine , which is a strong inducer of apoptosis , for 7 days . We found that culturing with HPIV3/ΔF-HN/EboGP or staurosporine caused a strong increase in the levels of activated caspases 3 , 8 and 9 , which were higher than that in cells treated with HPIV3 or mock treated cells ( Fig 2C–2E ) . To characterize caspase activation at a single-cell level , the cultured Jurkat cells were also analyzed for activated caspase 8 by flow cytometry; we found that culturing with HPIV3/ΔF-HN/EboGP and HPIV3 resulted in 60 . 3±1 . 9% and 28 . 7±0 . 9% of cells , respectively , positive for active caspase 8 ( Fig 2F and 2G ) . Taken together , this data indicates that EBOV triggers cell death pathways through GP-dependent mechanisms . Since T lymphocytes express TLR4 [34] , we hypothesized that the virus can attach to T lymphocytes . Previous reports have indicated that EBOV GP binds to and activates TLR4 signaling in DCs [24 , 35] . We confirmed binding following transfection of 293T human embryo kidney cells with plasmids expressing EBOV GP , VP40 and/or TLR4/TLR4-FLAG . EBOV GP , but not VP40 , efficiently co-precipitated with anti-TLR4 antibodies , suggesting that GP specifically interacts with TLR4 ( Fig 3A and 3B ) . To determine if GP mediates binding to T lymphocytes , we used confocal microscopy and flow cytometry; to distinguish the role of GP as opposed to whole EBOV , we utilized HPIV3/ΔF-HN/EboGP . As expected , both confocal microscopy and flow cytometry demonstrated direct binding of EBOV and HPIV3/ΔF-HN/EboGP to control 293T cells expressing TLR4 from a transfected plasmid [36] , as wild-type 293T cells do not express TLR4 [37] . Importantly , binding of EBOV and HPIV3/ΔF-HN/EboGP to purified human CD4+ T cells , Jurkat cells and SupT1 cells previously shown to possess a strong TLR4 signaling pathway following activation [38] was observed ( Fig 3C and 3D ) . EBOV stimulation of naïve and CD3/CD28-bead activated CD4+ T cells reduced the relative density of TLR4 on cell surface ( Fig 3E ) , suggesting that TLR4 may be internalized . Moreover , blocking of TLR4 with polyclonal anti-TLR4 antibodies significantly reduced binding of EBOV to SupT1 T cells ( Fig 3F ) . These data demonstrate that EBOV GP binds to T cells via TLR4 resulting in stimulation of cells despite the lack of infection . A functional TLR4 response in primary T lymphocytes has been demonstrated [39]; we therefore sought to determine if TLR4 signaling was the initial trigger of primary CD4+ T cells death exposed to EBOV GP . We previously confirmed that EBOV binds to SupT1 cells ( Fig 3D ) . The relative binding efficiency of EBOV appeared to be similar to that observed on primary CD4+ T cells and Jurkat cells . SupT1 cells or monocytes , which were used as control cells susceptible to EBOV infection , were cultured in the presence of EBOV , the natural TLR4 ligand lipopolysaccharide ( LPS ) [40] , HPIV3 , its derivative expressing EBOV GP from an added transcriptional cassette HPIV3/EboGP [41] , HPIV3/ΔF-HN/EboGP described above , or the TLR3 agonist polyI:C , which was used as a control for TLR specificity . Cells were cultured in the presence or absence of the TLR-4 inhibitor CLI-095 to determine the role of TLR4 signaling . Activation of TLR4 signaling cascade was examined by analysis of the phosphorylation state of TLR4 adapter proteins p-TRAM1 , phosphorylated following endosomal translocation of TLR4 and activating MyD88-independent pathway [42] , dephosphorylated IRAK4 , p-Pyk2 or p-p38 . The data demonstrated activation of TLR4 signaling following stimulation with EBOV or its GP protein in both SupT1 cells and monocytes with a reduction in TLR4 signaling being observed when cells were cultured with CLI-095 ( Fig 4A ) . To more specifically demonstrate that EBOV stimulated TLR4 through its glycoprotein GP , we evaluated the capacity of recombinant GP-bound beads or empty beads to activate both the TRAM1 and MyD88-dependent pathways using THP-1 , THP-1 MyD88-/- and SupT1 cells ( Fig 4B , S3A Fig ) . While in THP-1 and SupT1 cells GP beads were able to activate TLR4 signaling , in THP-1 MyD88-/- cells they were unable to trigger dephosphorylation of IRAK4 , phosphorylation of Pyk2 or p38 , while phosphorylation of TRAM1 still occurred . The addition of CLI-095 , which blocks both the TRAM1 and MyD88 pathways , significantly reduced TLR4-associated signal transduction . These data demonstrate that EBOV GP induces TLR4 signaling in CD4+ T cells by triggering both MyD88-dependent and MyD88-independent pathways . We next examined activation of NFκB , a downstream transcription factor known to be activated following TLR4 stimulation [36] , by stimulating SupT1 cells , monocytes , THP-1 and THP-1 MyD88-/- cells with GP delivered by HPIV3/ΔF-HN/EboGP , virus-like particles ( VLPs ) or GP beads ( Fig 4C–4E ) . Cell lysates were examined for the phosphorylation status of the p65 subunit of NFκB , which plays a major role in immune and inflammatory responses and whose phosphorylation is indicative of NFκB signaling activation [43] . Peak phospho-p65 ( p-p65 ) was detected 2 h after stimulation ( S3B Fig ) . Stimulation of SupT1 cells , monocytes , THP-1 and THP-1 MyD88 -/- cells with EBOV , HPIV3/ΔF-HN/EboGP , VLPs , GP beads or LPS , as well as control CD3/CD28 activation beads , resulted in a marked increase in phosphorylated p65 ( Fig 4C–4E ) . Treatment of cells with CLI-095 for 1 h prior to stimulation reduced the levels of phosphorylated p65 associated with EBOV , HPIV3/ΔF-HN/EboGP , VLPs , GP beads or LPS stimulation . As expected , CLI-095 did not affect phosphorylation of p65 in control CD3/CD28 activated cells ( Fig 4C and 4D ) . These findings further confirm that EBOV stimulation results in TLR4-mediated signal transduction in primary CD4+ T cells by both MyD88-dependent and MyD88-independent signaling pathways . Next , we determined the role of TLR4 signaling in the previously observed EBOV GP-mediated T lymphocytes death . Purified primary CD4+ T cells were cultured in the presence of EBOV or LPS with or without the TLR4 inhibitor CLI-095 . EBOV and LPS induced similar rates of cell death; however , cell death associated with EBOV induced extensive activation of caspase 9 , whereas LPS primarily triggered activated caspase 8 ( Fig 4F , S3C Fig ) . Interestingly , the rate of proliferation was increased by EBOV but not by LPS . Importantly , inhibition of TLR4 significantly reduced the percentages of dead cells with a correlative decrease in activated caspase 9 being observed in the presence of CLI-095 ( Fig 4F , S3C Fig ) . Taken together , these findings indicate that death of CD4+ T cells exposed to EBOV is associated with both intrinsic and extrinsic pathways . Based on these findings , we conducted a series of experiments aimed at determining the mechanisms by which EBOV induces T lymphocyte cell death . First , transcriptome analysis was utilized to determine the global response of CD4+ T cells to EBOV stimulation . Deep sequencing was performed on RNA samples extracted from highly purified CD4+ T cells cultured in medium alone , with EBOV or LPS at 24 and 96 h . Differential expression ( DE ) analysis comparing EBOV- and mock-infected samples at 24 h resulted in 2 , 591 DE genes using a 1 . 5 fold change cutoff and an adjusted p-value of 0 . 05 as criteria . A significant portion of these DE genes were related to cell death and innate immunity sensing . Specifically , exposure of T lymphocytes to EBOV resulted in 265 DE genes related to pathways specific for apoptosis , necrosis and TLR4 signaling ( S4A–S4C Fig ) . These results are consistent with the observations in Fig 1E demonstrating increased caspase-dependent cell death and TLR4 activation following exposure of T cells to EBOV . Overall , a global transcriptome profile of CD4+ T cells cultured in the presence of EBOV or LPS resulted in a much greater number of upregulated genes than downregulated genes and a greater number of differentially regulated genes involved in necrosis than apoptosis ( S4A–S4C Fig ) . In addition , the number of genes whose expression was differentially regulated was remarkably similar between EBOV and LPS ( Table 1 ) . However , the pattern of gene response to EBOV was clearly distinct of that induced by LPS ( S4A Fig ) . As indicated in the heat map and gene network in S4A and S4B Fig , TLR4 activation is associated with both apoptotic and necrotic pathways . Induction of multiple cell death pathways by EBOV explains why this virus but not HPIV3 strongly induces cell death ( Fig 2A and 2B ) even though both viruses induce TLR4 ( Fig 4A ) . We next investigated the contributing roles of each cell death pathway on EBOV-mediated T cell death . Due to the essential role of TNFα as an immune modulator following TLR4 activation and its role as an inducer of both apoptotic and necrotic pathways [44] , we hypothesized that TNFα inhibition may reverse cell death induced by EBOV . We first determined the capacity of isolated CD4+ T cells to release TNFα when cultured with EBOV . Indeed , a 96 h-long incubation of cells with EBOV or a positive control staphylococcal enterotoxin B ( SEB ) resulted in increased levels of TNFα , 47 . 5±10 . 6 and 4 , 913±738 pg/ml , respectively , which were significantly greater than in mock-treated samples , 4 . 0±0 . 8 pg/ml ( Fig 5A ) . To confirm that TNFα expression is TLR4-dependent , we treated or mock-treated SupT1 T cells with CLI-095 or anti-TLR4 neutralizing antibodies , incubated with EBOV or 12-O-tetradecanoylphorbol-13-acetate ( TPA ) /ionomycin for 24 h and analyzed for TNFα or IFNγ , as an additional marker of T cell activation . Our results demonstrated an increase in intracellular TNFα+ SupT1 cells following cultivation with TPA/ionomycin ( 19 . 2% of TNFα+ cells ) and EBOV ( 6 . 3% ) when compared to mock ( 0 . 3% ) ( Fig 5B and 5C ) . Treatment of cells with CLI-095 reduced the percentages of TNFα+ cells only when they were infected with EBOV , to 4 . 3% . Treatment of SupT1 cells with TPA/ionomycin or EBOV resulted in an increase in percentages of TNFα+ and IFNγ+ cells ( 17 . 2% and 24 . 5% for TPA/ionomycin and 8 . 9% and 18 . 1% for EBOV , respectively ) when compared to mock ( 3 . 6% and 6 . 0% , respectively ) ( S5A and S5B Fig ) . Pre-incubation of cells with anti-TLR4 reduced the percentages of TNFα+ and IFNγ+ cells in EBOV treated cells by 22 . 4% and 27 . 7% , respectively . Daily treatment of cells with TNFα at 80 pg/ml significantly increased the percentages of dead cells ( Fig 5D ) , while the addition of an inhibitor of TNFα , TNFα antagonist III , which blocks TNFα receptor-adapter interactions and prevents downstream signaling [45] , completely reversed TNFα-associated cell death ( Fig 5E , left panel ) . Interestingly , treatment with TNFα antagonist III also reduced cellular proliferation suggesting that TNFα may promote cell activation ( Fig 5E , right panel ) . Transcriptome analysis further supported these findings , as 24 h-long exposure of CD4+ T cells to EBOV resulted in a 5 . 1±1 . 6 fold increase in the levels of TNFα transcripts ( p<0 . 05 , Student’s T-test , based on log2 values ) ( S4A Fig ) . As TNFα is also involved in necrosis , we further confirmed the role of necrotic pathways in the observed EBOV-mediated induction of cell death using necrosis inhibitors . The addition of Necro X5 , geldanamycin or 1400W drastically reduced CD4+ T cells death following exposure to EBOV without affecting cell proliferation ( Fig 5F ) suggesting induction of necrosis . These findings demonstrate that T lymphocyte death induced by EBOV is associated with the engagement of TLR4 and subsequent production of TNFα which lead to induction of both apoptotic and necrotic pathways . Therefore , targeting of TLR4 or TNFα signaling cascades may provide therapeutic intervention strategies against EBOV infections . As noted , EBOV disease is characterized by infection of multiple types of cells including DCs and cells of the monocyte/macrophage lineage , a high and uncontrolled inflammatory response and depletion of T cells . Previous findings demonstrated that monocyte activation and differentiation increases their susceptibility to EBOV [46] . Furthermore , stimulation of TLR4 by EBOV GP was demonstrated to activate 293T cells [24] and DCs [35] . We hypothesized that the increased susceptibility of monocytes to EBOV infection is related to their differentiation caused by EBOV GP triggering TLR4 signaling . To test the hypothesis , we used THP-1 cells; differentiation was characterized by analysis of activation/differentiation markers CD14 , CD11b [47] and CD68 [48] by flow cytometry . First we tested if GP-mediated activation of TLR4 can result in differentiation of THP-1 cells . As TLR4-mediated stimulation of monocytes typically leads to differentiation into macrophages [26] , THP-1 cells were cultured with LPS , EBOV , HPIV3/ΔF-HN/EboGP and GP beads in the presence or absence of CLI-095 for 24 h or 96 h ( S6A Fig ) . Our results demonstrated a significant increase in the levels of markers of differentiation at both time points compared to mock-stimulated or empty beads-stimulated cells ( Fig 6A–6C , S6B and S6C Fig , S7A–S7H Fig ) . Higher levels of differentiation were observed at 96 h compared to 24 h post-infection ( S6D–S6F Fig , S7I–S7K Fig ) , which was consistent with the increase of infected cells as determined by the percentage of GFP+ cells ( S6G Fig , S8A–S8D Fig ) . CLI-095 treatment significantly reduced the expression of the differentiation markers , which was more pronounced at 96 h , clearly implicating the role of TLR4 in EBOV-induced differentiation of THP-1 ( Fig 6A–6C and S6B and S6C Fig ) . Specifically , in cells treated with LPS , EBOV , HPIV3/ΔF-HN/EboGP and GP beads , we detected a reduction in the percentages of cells positive for CD14 and CD11b . We next determined the relative rates of infection of THP-1 cells following TLR4-mediated differentiation . Following culture in the presence of LPS , EBOV , HPIV3/ΔF-HN/EboGP or GP beads for 24 or 96 h with or without CLI-095 , cells were infected for 48 h with EBOV-GFP ( Fig 6D ) . Consistent with the data on the effects of LPS , EBOV , HPIV3/ΔF-HN/EboGP and GP beads on cell differentiation , these stimulations also increased the percentages of infected cells , which similarly , was more pronounced at 96 h ( Fig 6E and 6F , S8A–S8D Fig ) . Again , adding of CLI-095 reduced the rates of infection at both 24 h and 96 h ( Fig 6E and 6F , S8A–S8C Fig ) further confirming the role of TLR4-dependent differentiation in the susceptibility of THP-1 cells to EBOV-infection . Furthermore , the analysis of THP-1 cells stimulated by HPIV3/ΔF-HN/EboGP and GP beads suggests that the observed induction of cell differentiation leading to increased susceptibility to EBOV infection is specifically related to interaction of GP with TLR4 . Overall , these results demonstrate that engagement of TLR4 by EBOV GP increases their differentiation , which subsequently leads to their increased susceptibility to the virus .
This study demonstrates , for the first time , that despite the lack of infection of T lymphocytes , EBOV directly binds and induces T cell death . In addition , this study demonstrates that interaction of EBOV GP with TLR4 stimulates differentiation of monocytes , which results in an increased susceptibility to EBOV infection . We show that following EBOV infection , monocytes undergo activation , which is known to lead to secretion of TNFα [49] . TNFα can contribute to activation and bystander death of T lymphocytes , which is consistent with the previously demonstrated effects of TNFα on T cells [50] . Furthermore , we show a direct binding of EBOV to T cells partially involving TLR4 and presumably involving additional ligands , as previous studies have identified the lectins DC-SIGN and L-SIGN [51–53] , folate receptor-α [54] , Tyro3 receptor tyrosine kinases [55] as attachment factors for EBOV . We demonstrate that the binding triggers the activation of numerous inflammatory signaling pathways including interferon , TLR and cell death signaling pathways as evidenced by the transcriptional profile of EBOV-stimulated T cells . The observed effects of TLR4 inhibitors conclusively demonstrated the role of TLR4 in lymphocyte cell death . Finally , using vectored and bead delivery of GP we demonstrate the direct role of the protein in the induction of both MyD88-independent ( TRAM1 ) and MyD88-dependent activation of the TLR4 signaling pathway and subsequent initiation of cell death pathways . Previous reports demonstrated the involvement of TNFα in apoptosis [56] as well as in necrosis following accumulation of reactive oxygen species [57] . Necrotic and apoptotic pathways are closely related , and following TLR4 activation , MyD88-dependent signaling pathway has also been reported to trigger necroptosis [58] , which also can occur in EBOV-cultured T cell culture as we demonstrated MyD88-dependent activity ( Fig 4A ) . The fact that multiple cell death mechanisms occur in the same EBOV-cultured T cells environment conjugated to the fact that different cell death mechanisms are interconnected between each other [59] could explain the drastic diminution of cell death observed in association with known specific inhibitors of apoptosis , necrosis or TNFα following EBOV-cultured T cells . Of note , HPIV3 also induced some TLR4 signaling ( Fig 4A ) but only low-level apoptosis ( Fig 2 ) that can be explained by some differences in signaling pathways induced by direct engagement of TLR4 by HPIV3 versus EBOV or by different strength of TLR4 signaling ( S4A Fig ) . A recent study demonstrated a wide-spread T lymphocyte activation in EBOV-infected patients receiving experimental antibody-based therapies; however , the percentage of CD4+ and CD8+ T cell responders was limited in comparison to the total percentage of activated T lymphocytes [60] . It was suggested that the activation may result from stimulation with immune complexes formed by the administered EBOV therapeutic monoclonal antibodies or convalescent plasma . However , based on the present data , activation may also be attributed to the activator role of GP . In parallel with these data , the activation of the TLR4 signaling pathway , sensitivity to TLR4 inhibitors and production of TNFα are routinely associated with LPS-induced bacterial sepsis [61] . Activation of innate immune pathways via pattern recognition receptors ( PRR ) including TLR4 has been shown to lead to systemic inflammation [62] . Previous studies indicated that successfully blocking TLR4 signaling significantly reduces the pathogenesis associated with bacterial sepsis and lethal influenza infection indicating the critical role of this innate immune-signaling pathway in exasperating immunological responses [63 , 64] . Our findings suggest that TLR4 inhibitors may be of therapeutic value for the treatment of EBOV patients . For example , our recent study demonstrates a high level of protection against EBOV and the closely related Marburg virus in vivo by the TLR4 receptor antagonist Eritoran [65] . Lymphopenia is consistently observed in human and nonhuman primate models following infection with viruses that cause viral hemorrhagic fever ( VHF ) [4 , 66 , 67] and therefore , this phenomenon is not restricted to EBOV infection . Furthermore , lymphopenia has been observed during fatalities from several non-VHF-related pathogens or pathologies , including that caused by highly pathogenic influenza virus , West Nile virus and bacterial sepsis [68–70] . Overall , remarkable similarities exist between EBOV-associated symptoms and the pathological features associated with the diseases caused by these non-VHF-related etiological agents including immune suppression , toxic effects due to inflammatory mediators and high viremia in the case of viral infections [66] . As the onset of lymphopenia is highly correlative with fatal outcomes following infection with aforementioned pathogens , identification of factors contributing to T lymphocyte cell death may enable the development of broad acting therapeutics . Lastly , we note that all current EBOV candidates rely on GP as a sole antigen [71]; with all requiring extremely high doses to achieve protection . For example a recent clinical study utilizing the VSV-vectored EBOV vaccine VSVΔG/ZEBOVGP , which is known to replicate at high levels , required a vaccination dose of 2x107 PFU [72] . The current study demonstrated induction of cell death by VSVΔG/ZEBOVGP ( S2A and S2B Fig ) . It is possible that the high vaccine doses required for the induction of protective immune responses are necessary to compensate for the reduced immunogenicity associated with the immune-modulating effects of GP presented in this study . Furthermore , another clinical trial of this vaccine demonstrated induction of a transient arthritis and dense CD4+ T lymphocytic vasculitis suggesting a pathophysiological role of vaccine-induced T lymphocytes . Both effects were related to the EBOV GP component of the vaccine , but not the VSV vector [73] . In addition , HPIV3/ΔF-HN/EboGP , which demonstrated toxic effects associated with GP in this work ( Fig 2A and 2B ) was also developed as a vaccine candidate [32] . Thus , GP exerts multiple immune modulating effects on immune cells , which may adversely affect the quality of the T cell response although it is also possible that the magnitude of lymphocytes death following vaccinations is modest and does not significantly affect the vaccine efficacy . The combination of our transcriptome analysis performed on CD4+ T cells and the perturbation of signaling cascades associated with cell death following CLI-095 treatment suggest an important role of the GP-TLR4 interaction in the pathogenesis of EBOV infection . Our study highlights diverse strategies used by EBOV to perpetrate lymphopenia through direct and indirect mechanisms , which results in both apoptotic and necrotic T cell death despite the lack of infection . We expanded these studies to include monocytes , which are permissive to EBOV . We found that the increase in activation/differentiation following EBOV-mediated stimulation of TLR4 resulted in a significantly increased rate of infection of monocytes . We note , however , that the effects of TLR-4 on differentiation are likely to be only partial , and other mechanisms involving GP may also contribute it . As mentioned above , several attachment factors have been identified for EBOV; their engagement as well as release of cytokines ( e . g . TNFα ) are likely to contribute cellular differentiation in an autocrine and/or paracrine manner . These data indicated that infection of monocytes and other cells amplify lymphocytes death by producing more viral particles and also by secreting TNFα and other proteins contributing death of lymphocytes . As TLR4 is expressed by multiple cell types , we suggest that the interactions of GP with TLR4 may have profound effects in vivo , and therefore , examination of TLR4 inhibitors as therapeutics for EBOV-infection is warranted . Overall , these data contribute to understanding of the ‘immune paralysis’ during EBOV infections .
Human embryonic kidney 293T ( 293T ) , SupT1 , THP-1 and Jurkat cell lines were obtained from the American Type Culture Collection . THP-1 MyD88-/- cells were obtained from InVivoGen . 293T were cultured in Dulbecco’s modified Eagle medium ( DMEM ) supplemented with 10% heat-inactivated fetal bovine serum ( HI-FBS ) ( ThermoFisher Scientific ) , 1% HEPES ( Corning ) , 1% nonessential amino acids ( Sigma-Aldrich ) , 1% sodium pyruvate ( Sigma-Aldrich ) and 2% PenStrep mix ( ThermoFisher Scientific ) . THP-1 , THP-1 MyD88-/- , SupT1 and Jurkat cell lines were cultured in RPMI 1640 ( ThermoFisher Scientific ) supplemented with 10% HI-FBS and 1% HEPES . Viruses and VLPs used in the study are briefly described in S1 Table . The recombinant EBOV , strain Mayinga , expressing green fluorescent protein ( EBOV-GFP ) or wild-type ( EBOV ) were generated as described in our previous study [17] . The recombinant wild type human parainfluenza type 3 ( HPIV3 ) [74] , strain JC , was provided by Drs . P . Collins and M . Skiadopoulos ( National Institutes of Health ) . Chimeric HPIV3 , in which EBOV GP has been added to the envelope ( HPIV3/EboGP ) or where the HN and F genes were replaced with that of GP ( HPIV3/ΔF-HN/EboGP ) were generated in our previous study [32] . All viruses were quantitated by plaque titration in Vero-E6 monolayers as previously described [17] . EBOV VLPs were generated as previously described [75] using mammalian cell codon optimized plasmids expressing EBOV GP pWRG7077:64755-2010-233-1_GP_optVP40 ( codop-EBO-GP ) and VP40 pWRG7077:64759-2010-233-1-4_VP40_optVP40 provided by Dr . Sina Bavari ( U . S . Army Medical Research Institute of Infectious Diseases ) . His-tagged GP were purchased from Integrated BioTherapeutics ) and bound to Dynabeads ( ThermoFisher Scientific ) following manufacturer’s instructions . Beads were washed with RPMI 1640 supplemented with 10% HI-FBS and 1% HEPES and used at a concentration of 3 beads/cell . THP-1 cells were plated at 1x106 cells/ml in 24-well plates in medium alone or medium containing CLI-095 ( 100 ng/ml ) for 1 h . Thereafter LPS ( 500 ng/ml ) , EBOV ( no GFP ) , EBOV GP beads or HPIV3/ΔF-HN/EboGP were added at MOI 3 PFU/cell and cells were incubated for 24 h or 96 h . To analyze markers of differentiation , cells were harvested , stained with antibodies specific for CD14-BUV395 ( BD Biosciences #563561 ) and CD11b-FITC ( BD Biosciences #562793 ) , permeabilized , fixed and stained with antibodies specific for CD68-PE/Cy7 ( BD Biosciences #565595 ) . To analyze susceptibility to infection , following stimulation cells were centrifuged for 5 min at 250 g and supernatants were removed , cells were infected with EBOV-GFP at MOI of 3 PFU/cell , and incubated for an additional 48 h . CLI-095 ( 100 ng/ml ) was added 1h prior cells were cultured with conditions . Cells were harvested , fixed and analyzed and analyzed using for markers of differentiation or GFP by FACS Fortessa flow cytometer ( BD Biosciences ) . SupT1 cells were cultured at 1x106 cell/ml in 24-well plates in medium with or without CLI-095 at 1 μg/ml or anti-TLR4 antibodies at 50 μg/ml for 1 h . Thereafter , cells were mock-treated or treated with TPA ( Sigma-Aldrich ) ( 25 ng/ml ) / ionomycin ( Sigma-Aldrich ) ( 0 . 5 μM ) or EBOV ( MOI 3 PFU/cell ) . Then , cells were treated with Brefeldin-A ( Sigma-Aldrich ) ( 10 μg/ml ) 1 h post treatment for 24 h . Cells were harvested , stained for intracellular TNFα or IFNγ using anti-TNFα-Pacific Blue ( Biolegend #502920 ) and anti-IFNγ-PE ( eBiosciences #12-7319-42 ) and analyzed by flow cytometry . Supernatants were analyzed for TNFα by Multiplexing LASER Bead Technology by Eve Technologies ( Calgary , Canada ) . Buffy coats were obtained from anonymous healthy adult donors according to a clinical protocol approved by the University of Texas Medical Branch at Galveston ( UTMB ) Institutional Review Board . Peripheral blood mononuclear cells ( PBMCs ) were isolated by Histopaque ( Sigma-Aldrich ) gradient as recommended by the manufacturer . CD14+ monocytes were isolated from fresh PBMCs , which were subsequently used for isolation of CD3+ , CD4+ and CD8+ T lymphocytes by positive selection using magnetic microbeads separation kits ( all from Miltenyi ) . In experiments where only CD4+ T cells were isolated , a negative selection CD4+ T cells isolation kit ( Miltenyi ) was used . In some experiments , negative selection of CD3+ , CD4+ or CD8+ T cells ( Stem Cell Technologies ) was used . Purity of the isolated lymphocytes typically ranged from 93 to 95% as determined by flow cytometry . CD14+ monocytes were cultured in Monocyte-DC Differentiation Medium ( Miltenyi ) for 7 days to obtain immature DCs . To differentiate immature DCs to mature DCs , Mo-DC Maturation Medium ( Miltenyi ) was added and cells were incubated for an additional 3 days . Immature and matured DCs were assessed using Mo-DC Differentiation Inspector ( Miltenyi ) and analyzed by flow cytometry . For co-culture experiments , DCs were combined with autologous cryopreserved CD4+ or CD8+ T lymphocytes at a 1:1 ratio in 96-well U-bottom plates in RPMI 1640 medium supplemented with 10% HI-FBS . LPS ( Invivogen ) was added as indicated . T lymphocyte activation was induced with Dynabeads Human Transactivator CD3/CD28 Beads ( ThermoFisher Scientific ) according to manufacturer’s recommendations . SupT1 cells were plated at the concentration of 1x106 cells per well in U-Bottom 96-well plates ( ThermoFisher Scientific ) and placed on ice ( to prevent internalization of viruses without infection ) , and EBOV at MOI 1 PFU/cell was added . Cells were incubated for 2 h at 4°C and washed with PBS containing 2% HI-FBS . Thereafter , cells were immunostained with rabbit antibodies raised against EBOV VLP ( Integrated BioTherapeutics ) . After staining , cells were washed three times with PBS containing 2% HI-FBS , fixed in 10% formalin ( ThermoFisher Scientific ) and stained with goat anti-rabbit antibodies labeled with Alexa-Fluor 647 ( ThermoFisher Scientific ) and washed again 3 times in PBS with 2% HI-FBS . Flow cytometry was performed using a LSRII Fortessa flow cytometer ( BD Biosciences ) available at the UTMB Flow Cytometry Core Unit . The inhibitors of necrosis NecroX5 ( Enzo Life Sciences ) , geldanamycin ( Invivogen ) or N- ( 3-aminomethyl ) benzylacetamindine ( Santa Cruz Biotechnology ) were used at concentrations 20 μM , 10 μM and 10 μM , respectively . The inhibitor of apoptosis z-VAD-FMK ( Affymetrix eBioscience ) was used at 20 μM . The inhibitor of TNFα , TNFα antagonist III ( Santa Cruz ) , was used at 1 μM . The TLR4 inhibitor CLI-095 ( Invivogen ) was used at 100 ng/ml . The inhibitors were added to cell cultures 1 h prior to the addition of EBOV ( MOI 1 PFU/cell ) . The T lymphocyte prosurvival cytokines , IL-4 , IL-7 and IL-15 ( all R&D Systems ) were used at 100 ng/ml , 10 ng/ml , and 100 ng/ml , respectively . In all experiments , inhibitors of cell death and prosurvival cytokines were added 30 minutes prior to the addition of EBOV ( MOI 1 PFU/cell ) . Monocytes , THP-1 , THP-1 MyD88-/- or SupT1 T lymphocytes were plated at a concentration of 1x106 cells per well in 96- or 24-well plates and mock-treated or treated with CLI-095 at 1 μg/ml for 1h . Then cells were stimulated with HPIV3 WT , HPIV3/EboGP , HPIV3/ΔF-HN/EboGP , EBOV GP beads , EBOV at MOI 0 . 1 , 1 or 3 PFU/cell or mock-stimulated , transfected Poly I:C ( 10 μg/ml ) , VLPs at 10 μg/ml or 25 μg/ml , CD3/CD28 beads at a ratio of 1 bead per 3 cells , or LPS at 100 or 500 ng/ml and harvested at the indicated time points . Cell were collected at the indicated time points and lysed in RIPA buffer ( ThermoFisher Scientific ) . Proteins were separated by SDS-PAGE using gradient 4–12% gels ( ThermoFisher Scientific ) and transferred to nitrocellulose membranes ( ThermoFisher Scientific ) using the I-blot system ( ThermoFisher Scientific ) . Membranes were blocked with 5% milk and 0 . 1% Tween-20 in PBS for 1 h at 37°C and stained with antibodies specific for the following molecules: TRAM1 ( Abcam #ab96106 ) , phosphorylated TRAM1 ( FabGennix #PTRAM-140AP ) , MyD88 ( #4283S ) , IRAK4 ( #4363S ) , phosphorylated IRAK4 ( #11927S ) , Pyk2 ( #3090S ) , phosphorylated Pyk2 ( #3291S ) , p38 ( #8690S ) , phosphorylated p38 ( #4511S ) , NFκB ( #6956S ) , phosphorylated NFκB ( #3033S ) and GAPDH ( #8884S ) ( all Cell Signaling Technology ) diluted according manufacturer’s recommendations in PBS with 0 . 1% of Tween-20 . Primary T lymphocytes , lymphoid cell lines or PBMC were stained with CFSE ( ThermoFisher Scientific ) to monitor cell proliferation as recommended by manufacturer . The activity of caspase-8 and caspase-9 was determined at the indicated time points by flow cytometry using Vybrant FAM Caspase-8 kit ( ThermoFisher Scientific ) or CaspGLOW Active Caspase-8 Staining kit ( eBioscience ) and Red FLICA Caspase-9 Assay kit ( Immunochemistry Technologies ) . Thereafter , cells were stained with a combination of the following antibodies: anti-CD3 clone UCHT1 , labeled with BUV395 , anti-CD3 clone UCHT1 , labeled with Pacific Blue , anti-CD4 clone OKT4 , labeled with PerCP-Cy5 . 5 , anti-CD8 clone RPA-T8 , labeled with PerCP-Cy5 . 5 , anti-CD8 clone RPA-T8 , labeled with APC , and also with Annexin-V labeled with PE ( all BD Biosciences ) . As the study was done under BSL-4 biocontainment , and due to biosafety regulations , cells analyzed by flow cytometry must be fixed with paraformaldehyde . Since apoptotic cells are positive for Annexin V , co-staining of cells with Annexin V and Live/Dead stain instead of PI provides the opportunity to discern apoptotic from necrotic cells [76] . As both PI and 7-AAD are poorly compatible with fixation protocols , Live/Dead staining which is compatible with formalin fixation and similarly to PI and 7AAD stains dead cells , was used . Following surface receptor staining , Live/Dead ( ThermoFisher Scientific ) staining was performed according manufacturer’s recommendation and cells were fixed with 10% formalin . Flow cytometery was performed using a FACS Fortessa instrument ( BD Biosciences ) . Jurkat T cells were plated at 1x106 cells per well of a 24-well plate and HPIV3 or HPIV3/ΔF-HN/EboGP was added at a MOI 1 PFU/cell . Cells were incubated for 7 days at 37°C . As a positive control for activation of caspase-3 , -8 and -9 , cells were treated with staurosporine ( Sigma-Aldrich ) at 1 μM for 6 h . After 7 days , cells were harvested , washed three times with PBS and lysed in RIPA lysis buffer supplemented with 4x Laemmli buffer ( ThermoFisher Scientific ) . Western blot analysis was performed using anti-caspase-3 ( Santa Cruz Biotechnology sc-271028 ) , anti-caspase-8 ( Santa Cruz Biotechnology sc-81657 ) , and anti-caspase-9 ( Cell signaling #9502 ) antibodies . Densitometric analyses of active caspase-3 , active caspase-8 , active caspase-9 were performed using ImageJ software ( NIH ) and normalized using GAPDH . SupT1 cells were infected with VSVΔG/ZEBOVGP at 1 or 3 PFU/cell , incubated for 4 days at 37°C , stained with Live/Dead and analyzed by flow cytometry as described above . TLR4 expression in purified T lymphocyte populations and cell lines was determined by western blot analysis using anti-TLR4 antibody ( Santa Cruz Biotechnology , #sc-293072 ) . GP-TLR4 binding was determined by co-transfecting 293T cells with mammalian codon-optimized plasmids encoding EBOV GP or VP40 , as well as the plasmids expressing TLR4 ( Addgene , #20863 ) or TLR4 FLAG-tagged protein ( TLR4-FLAG ) ( Addgene , #42646 ) using TransIT-LT1 reagent ( Mirus ) for 48 h at 37°C . Cells were lysed with 500 μl of RIPA lysis buffer supplemented with Protease and Phosphatase Inhibitor Cocktail ( ThermoScientific ) for 30 min at 4°C . Lysates were centrifuged at 400 x g at 4°C for 10 minutes . Then , 50 μl of cleared supernatants were kept for protein expression analysis while the remaining 450 μl were utilized for immunoprecipitation assays . Supernatants were incubated with monoclonal antibodies specific for TLR4 ( Santa Cruz Biotechnology , #sc-293072 ) or FLAG , clone M2 ( Sigma-Aldrich ) and incubated for 2 h at 4°C with rotation , followed by addition of protein G agarose beads ( ThermoFisher Scientific ) and overnight incubation at 4°C on a rotating platform . Beads were centrifuged at 2 , 500 x g at 4°C , washed with RIPA buffer ( ThermoFisher Scientific ) three times , and resuspended in 50 μl of Laemmli lysis buffer ( ThermoFisher Scientific ) for western blot analysis . The following antibodies ( all Integrated BioTherapeutics ) were used for western blot analysis: rabbit anti-GP ( #0301–015 ) , rabbit anti-VP40 ( #0301–010 ) and rabbit anti-EBOV VLP ( #01–0004 ) . HRP-conjugated Secondary ( Santa Cruz Biotechnology ) antibodies were used to visualize bands following the addition of ECL reagent ( ThermoFisher Scientific ) . 293T cells transfected with a TLR4-expressing plasmid ( as described above ) , THP-1 , SupT1 , Jurkat and primary CD4+ T cells were plated at 1x105 cells/ml and HPIV3/ΔF-HN/EboGP and EBOV or EBOV GP beads were added at MOI 5 PFU/cell or 3 beads/cell respectively . Cells were incubated for 2 h on ice or 2 h at 37°C , washed with PBS with 2% HI-FBS , fixed with 10% formalin and loaded on positively charged slides ( ThermoFisher Scientific ) and dried overnight . Following rehydratation , cells were permeabilized with PBS 0 . 5% Triton X100 ( Alfa Aesar ) for 15 minutes . Cells were then washed with PBS and incubated with 0 . 5 M glycine in PBS for 30 minutes at room temperature before performing antigen blocking using 5% donkey serum diluted PBS with 1% BSA and 0 . 1% Triton X100 ( PBS-BSA-TX100 ) for 1 h . Anti-EBOV VLP serum ( Integrated BioTherapeutics ) was diluted at 1:100 , anti-LAMP1 and anti-Rab7 were diluted at 1:12 . 5 and anti-TLR4 was diluted at 1:50 using PBS-BSA-TX100 and put on the slides for 1 h . Then , slides were washed with PBS with 0 . 1% Triton X100 and incubated with secondary donkey anti-rabbit antibodies conjugated with AlexaFluor 647 ( ThermoFisher Scientific ) diluted at 1:200 in PBS-BSA-TX100 for an additional 1 h before being washed as above . Next , cells were incubated with 6-diamin-2-phenylindole-dihydrochloride ( DAPI ) ( ThermoFisher Scientific ) at 1 μg/ml for 2 minutes and washed with PBS . The coverslips were mounted onto microscope slides using PermaFluor mounting medium ( ThermoFisher Scientific ) . Laser scanning confocal microscopy was performed on Olympus FV1000 confocal microscope housed in the Galveston National Laboratory . Laser beams with 405 nm wavelengths were used for DAPI excitation , and 635 nm for AlexaFluor 647 excitation . Emission filters were 425/25 nm for DAPI and 610/50 nm for AlexaFluor 647 detection , respectively . All images were acquired using a 60x oil objective . Isolated CD4+ T cells from four donors were cultured in RPMI1640 medium in the presence or absence of EBOV at MOI 3 or LPS at 500 ng/ml . 24 and 96 h post stimulation , cells were washed with PBS , lysed in 1 ml of TRIzol ( ThermoFisher Scientific ) and stored at -80°C . RNA isolation was performed using Direct-zol RNA MiniPrep kit ( Zymo Research ) . mRNA libraries were constructed following analysis with an Agilent 2100 Bioanalyzer ( nanochip format ) . Libraries were constructed using the Kapa Stranded mRNA-Seq Kit ( Kapa Biosystems ) according to the manufacturer’s instructions before being quality controlled and quantitated using the BioAnalzyer 2100 system and QuBit ( Invitrogen ) . The libraries were clonally amplified and sequenced on an Illumina NextSeq 500 to achieve a target density of approximately 200K-220K clusters/mm 2 on the flow cell with dual indexed paired end sequencing at a 75 bp length using NextSeq 500 NCS v1 . 3 software . Raw reads ( 75 bp ) had their adapter sequences removed . FastQC ( http://www . bioinformatics . babraham . ac . uk/projects/fastqc/ ) was used for general quality control of the raw reads . Bowtie ( v2 . 1 . 0 ) was utilized to remove ribosomal RNA using an index of human , mouse , and rat rRNA sequences [77] . Reads were then mapped against a human reference genome ( hg19 , build GRCh37 , from the UCSC genome browser ( http://genome . ucsc . edu ) using STAR ( v2 . 4 . 0h1 ) [78] . From this alignment , quantitative gene counts were produced using HTSeq ( http://www-huber . embl . de/users/anders/HTSeq/doc/overview . html ) utilizing the human annotation associated with the genome [79] . Gene counts for 25 , 237 genes for each sample were loaded into R ( http://www . r-project . org/ ) . Counts across samples were normalized with edgeR ( version 3 . 10 . 2 ) using the weighted trimmed mean of M-values while genes with no counts or at least three sample with counts were removed [80] leaving 14 , 932 genes with an average of ~106 reads per sample . Differentially expressed genes were identified using edgeR between treatments and time points and defined by using an absolute fold change cutoff of 1 . 5 and a p-value of ≤ 0 . 05 after adjustment using the Benjamini-Hochberg multiple testing correction . Additional clustering , creation of heatmaps , and other statistical analyses were performed using R . Functional analysis of the differential gene expression data was performed with QIAGEN’s Ingenuity Pathway Analysis ( IPA , QIAGEN Redwood City , www . qiagen . com/ingenuity ) . Functional annotation of genes for specific biological functions was assigned through querying AmiGO ( version 2 . 20 ) [81] . Gene names were collected by searching for human genes with search queries “cell death” , “necrosis” , “apoptosis” , and “TLR” . Functional enrichment analysis using GO terms was performed using the PANTHER Overrepresentation Test ( release 20150430 ) from the GO Ontology database released 2015-08-06 against a reference of all Homo sapiens genes using the complete GO biological processes annotation set [82] . Molecules from both the experimental expression data and the IPA Knowledge base were added to the networks that were then assigned a score derived from its p-value . IPA network scores of 2 or higher have at least a 99% confidence interval of not being generated by chance alone . Genes contributing to the construction of the TLR4 and necrosis networks were created from stricter absolute log2 fold change cutoffs of > 2 fold to focus on the interactions between the most highly activated genes . Molecular activity prediction ( MAP ) analysis was also performed on the cell death network . MAP analysis uses the IPA Knowledge Base to predict the upstream and downstream effects of activating or inhibiting a molecule in the network . Each independent experiment was performed in triplicate to rule out experimental bias or random error . Statistical methods used were described in Figure Legends using GraphPad Prism 6 ( GraphPad Software ) . P values of <0 . 05 were considered statistically significant . Mean and standard error of the mean ( SE ) were calculated for all graphs . All work with EBOV was performed within the Galveston National Laboratory biosafety level 4 laboratories . All staff had the appropriate training and U . S . government permissions and registrations for work with EBOV . | The latest outbreak of Ebola virus ( EBOV ) in West Africa resulted in more than 28 , 000 human infections including more than 11 , 000 deaths thus highlighting the necessity for the development of countermeasures . Monocytes and dendritic cells are among the primary targets of EBOV infection; infection of these critical antigen presenting cells contributes to the immune deficiency observed in Ebola virus disease ( EVD ) . In contrast , lymphocytes are resistant to EBOV infection; however , in fatal EVD , pronounced lymphopenia is uniformly observed . Here we report that T lymphocyte cell death in the absence of detectable infection was observed in an EBOV glycoprotein ( GP ) -dependent manner . Using transcriptome analysis of EBOV-stimulated CD4+ T cells we show upregulation of both toll-like receptor 4 ( TLR4 ) and cell death associated pathways . Furthermore , we demonstrate that EBOV increases susceptibility of monocytes to infection by promoting cellular differentiation . Both EBOV-induced monocyte differentiation and cell death of T lymphocytes result from a direct interaction between GP and TLR4 . Blocking of TLR4 signaling significantly reduced both EBOV-induced T cell death and infection of monocytes . These data contribute to understanding of the ‘immune paralysis’ during EBOV infections and provide evidence for the development of targeted therapies for the treatment of EVD . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"flow",
"cytometry",
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"cell",
"processes",
"cell",
"differentiation",
"immune",
"receptor",
"signaling",
"developmental",
"biology",
"membrane",
"receptor",
"signaling",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"necrotic",
"cell",
"death",
"animal",
"cells",
"t",
"cells",
"spectrophotometry",
"cytophotometry",
"signal",
"transduction",
"cell",
"biology",
"monocytes",
"apoptosis",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling",
"spectrum",
"analysis",
"techniques"
] | 2017 | Ebola virus glycoprotein directly triggers T lymphocyte death despite of the lack of infection |
Visceral leishmaniasis ( VL ) is a vector-borne disease whose factors involved in transmission are poorly understood , especially in more urban and densely populated counties . In Brazil , the VL urbanization is a challenge for the control program . The goals were to identify the greater risk areas for human VL and the risk factors involved in transmission . This is an ecological study on the relative risk of human VL . Spatial units of analysis were the coverage areas of the Basic Health Units ( 146 small-areas ) of Belo Horizonte , Minas Gerais State , Brazil . Human VL cases , from 2007 to 2009 ( n = 412 ) , were obtained in the Brazilian Reportable Disease Information System . Bayesian approach was used to model the relative risk of VL including potential risk factors involved in transmission ( canine infection , socioeconomic and environmental features ) and to identify the small-areas of greater risk to human VL . The relative risk of VL was shown to be correlated with income , education , and the number of infected dogs per inhabitants . The estimates of relative risk of VL were higher than 1 . 0 in 54% of the areas ( 79/146 ) . The spatial modeling highlighted 14 areas with the highest relative risk of VL and 12 of them are concentrated in the northern region of the city . The spatial analysis used in this study is useful for the identification of small-areas according to risk of human VL and presents operational applicability in control and surveillance program in an urban environment with an unequal spatial distribution of the disease . Thus the frequent monitoring of relative risk of human VL in small-areas is important to direct and prioritize the actions of the control program in urban environment , especially in big cities .
Visceral leishmaniasis ( VL ) is a vector-borne disease highly influenced by social and environmental factors . The majority ( >90% ) of cases is concentrated in six countries: Bangladesh , Brazil , Ethiopia , India , Nepal and Sudan [1] . In Brazil , the VL is caused by Leishmania infantum , belonging to the Leishmania donovani complex that is mainly transmitted by the sand fly Lutzomyia longipalpis . Dogs are considered to be the principal parasite reservoir , playing an important role in the transmission cycle in urban areas . The VL urbanization has been documented since the 1980s [2]–[8] . This trend represents a challenge for the control of the disease in urban areas [6]–[7] , [9] . The average incidence rate of VL in Brazil was 1 . 9/100 , 000 inhabitants between 1994 and 2009 [10] . During the last decades , an increasing number of clinical VL cases have been reported for large Brazilian cities , including Belo Horizonte . In 1994 , the first human VL cases occurred in BH ( n = 29 ) , with incidence rate of 1 . 4 cases/100 , 000 inhabitants and case fatality rate of 20 . 7% . From 1994 to 2009 , the highest incidence rate was 7 . 2/100 , 000 ( in 2008 ) , the case fatality rate 23 . 6% ( in 2009 ) and the proportion of infected dogs was 9 . 9% ( in 2006 ) [11] . The presence of the vector and of the canine reservoir has been described throughout Belo Horizonte with an unequal geographic distribution [11]–[15] , which is likely due to the intra-urban differences present in large cities . In Belo Horizonte , the Health Vulnerability Index incorporates intra-urban differences represented by indicators of sanitation , housing , income , education , and health [16] . The Health Vulnerability Index is useful in the identification of areas with unfavorable socioeconomic conditions , which are priorities for interventions and the allocation of resources for public policies . In Brazil , the Visceral Leishmaniasis Control and Surveillance Program ( VLSCP ) is based on reducing the morbidity and case-fatality rates through the early diagnosis and treatment of human cases and on decreasing the transmission risk by controlling the population of both domestic reservoirs and the vector [9] . However , in urban areas , the VLSCP has encountered difficulties , including: logistics and a high cost for a chemical control of the vector; insufficient environmental management for vector control; large interval between diagnosis and the elimination of infected dogs; dissatisfaction among the human population with the elimination of infected dogs; insufficient accuracy of the tests to detect infection , thereby allowing asymptomatic dogs to persist as a source of infection for the vector; substitution of culled dogs by a new susceptible canine population; and high canine infection and infectiveness rates [8] , [17]–[19] . The spread of the disease in Belo Horizonte has occurred despite systematic interventions since 1994 , which reflects the difficulty of control in urban areas . Thus , the identification of areas with a greater disease risk may help to direct and prioritize the actions of the VLSCP . This study aimed to identify the greater risk areas for human VL and the risk factors involved in transmission , using spatial analysis , in Belo Horizonte , Minas Gerais State , Brazil , during 2007–2009 .
This study was approved by the research ethics committees of the Universidade Federal de Minas Gerais-UFMG ( No 211/09 ) and of the Belo Horizonte City Hall ( No 075 . 2008 ) . Given the assumptions of research ethics , we maintained the confidentiality of data during processing . Analyses were performed anonymously; hence the Informed Consent Form was not necessary . This is an ecological study on the relative risk of human VL , whose spatial units of analysis were the coverage areas of the Basic Health Units of Belo Horizonte . Belo Horizonte , capital of Minas Gerais State , has 2 , 375 , 151 inhabitants , and a population density of 7 . 2 inhabitants/km2 . It is the sixth most populous city in Brazil according to the census of the Brazilian Institute of Geography and Statistics ( Instituto Brasileiro de Geografia e Estatística-IBGE ) [20] . The city is located at 852 meters above sea level , between latitude 19°49′01″S and longitude 43°57′21″W . It has a dry winter and a hot and rainy summer , with an average annual temperature of 21°C , average relative air humidity of 65% , and average annual rainfall of 1 , 500 mm [21] . In Belo Horizonte , the health services are organized territorially into nine health districts that are subdivided into 146 areas covered by Basic Health Units ( Figure 1 ) . These coverage areas are the result of aggregating 2 , 563 census tracts defined by the Brazilian Institute of Geography and Statistics ( IBGE ) [20] . The division between these coverage areas considers aggregation and geographical barriers , continuity occupation , transportation facilities and characteristics of homogeneity . It is noteworthy that the divisions are mainly administrative . The population of these areas varies from 2 , 197 to 45 , 171 inhabitants with an average of 15 , 331 inhabitants . Human VL cases were selected with an onset of symptoms between 2007 and 2009 ( n = 412 ) , which were available in the Brazilian Reportable Disease Information System . The cases were georeferenced at the household level , which allowed identifying the respective coverage areas in which they were contained . During 2007–2009 , the incidence rates were 4 . 9 , 7 . 2 and 6 . 6/100 , 000 inhabitants and the case fatality rates were 8 . 2 , 12 . 4 and 23 . 6% , respectively . In Belo Horizonte , the domestic dog population was 301 , 593 dogs in 2009 ( 1 dog per 8 inhabitants ) . The actions of the Visceral Leishmaniasis Control and Surveillance Program related to the dogs are recorded in the Information System on Zoonosis Control . The canine blood samples are screened for antibodies against Leishmania by an enzyme-linked immunosorbent assay ( ELISA ) and confirmed by an indirect immunofluorescent antibody test ( IFAT ) . In the studied period , 470 , 479 blood samples were tested , of which 7 . 8% were reactives . Some negative dogs were tested more than once . The data on canine infection were evaluated in two ways: number of infected dogs and number of infected dogs per inhabitants . Among the components of the Health Vulnerability Index , we selected those that have been described in the literature as being associated with the occurrence of VL: indicators of urban services ( water supply , sanitary sewage , and the destination of waste ) and of the socioeconomic level ( income and education ) . These data , available for the census tracts , were grouped according to the coverage areas by the average weighted by the population or by the number of households , depending on the main relation of the indicator . We used a contour map of Belo Horizonte to calculate the average altitude of each area . The vegetation coverage was characterized by means of the Normalized Difference Vegetation Index ( NDVI ) [22] . The NDVI varied from −1 to +1 . In general , the positive and negative values indicate the presence and absence of green vegetation , respectively . To calculate the NDVI , images were obtained by the Thematic Mapper sensor aboard the LANDSAT-5 satellite in July 2008 , since this was the intermediate year in the analyzed period , and there were higher-quality images due to the relative absence of cloud cover that month . The images were adjusted to the spatial conformation of the coverage areas , and the minimum , maximum , mean , standard deviation , and median values of the NDVI were calculated . The spatial statistical modeling used does not consider the time variable [23] . Therefore , we studied a short time series to minimize the potential effect of time in the estimation of the incidence rates . The period 2007–2009 was selected due to the availability of georeferenced data on dogs . The analyses were performed using the MapInfo® 8 . 5 and WinBUGS 1 . 4 softwares and the geographical information system of Belo Horizonte . The descriptive analyses were carried out in two steps . First , thematic maps were used to visualize the spatial distribution of cumulative incidence rates of human VL per 100 , 000 inhabitants ( crude rates ) and the ratio of infected dogs per 1 , 000 inhabitants in 146 coverage areas during 2007 to 2009 . Second , scatter plots graphics were generated with the log-relative risk of VL ( log-RR-VL ) and the studied covariates . This process allowed for the selection of the covariates that were most correlated with the log-RR-VL for further statistical modeling . Data from neighboring areas were used to estimate the VL incidence rate of a certain area ( smoothed rates ) [23] . This approach was intended to minimize the instability in the incidence rates calculated for small-areas . Moreover , it produces more reliable estimates and smoother maps , which are easier to view and interpret from an epidemiological point of view [24]–[27] . As a definition for neighboring areas , we used the adjacency: areas which share borders were considered neighbors . The dependent variable was the number of VL cases observed in area i , Oi , i = 1 , 2 , … , N , where N is the number of areas in the map . We assume that Oi follows a Poisson distribution with mean λi . Our aim was to model the relative risk , which is defined as , where Ei is the expected number of VL cases in area i if all VL cases were homogeneously distributed over the entire map . To obtain a generalized linear model , the logarithmic link function was used: . The covariates are incorporated into the model by , where are coefficients to be estimated and bi is a random effect . Then , ( 1 ) where is an offset term . To model the random effect bi , we used the convolution conditional autoregressive ( CAR ) model [23] , which was defined by a set of conditional distributions . The random effect bi is decomposed into two components , , such that μi are unstructured i . i . d . N ( 0 , σ2μ ) random effects . The component si is spatially structured and follows an intrinsic conditional autoregressive model as follows:where N is the total number of areas in the map and is the ( N-1 ) -dimensional vector of observations in all N areas but the observation of area i . Let be a N×N neighborhood matrix so that if the areas i and j are neighbors and otherwise . By definition , . We define the matrix so that , where is the number of neighbors of area i . Each area must have at least one neighbor because islands are not allowed here . Therefore , . The parameter is the area i prior variance , and σs2 is the common variance . Hence , a larger number of neighbors of an area imply a smaller prior variance . It is worth noting that the expected value of si takes into account not only information for area i but also information for its neighbors . In this sense , the approach [23] uses the spatial information in the neighborhood to estimate parameters related to each area of the map . Using a Bayesian approach and the Markov Chain Monte Carlo ( MCMC ) sampling method , a posterior distribution was generated for each coefficient of the model in ( 1 ) . The average of the sampled values was used as an estimate for the coefficient , and 95% credibility intervals were used as a criterion to determine whether the covariates should remain into the model . As a prior distribution to the intercept and the covariate coefficients , it was adopted a flat distribution and a normal distribution with a zero mean and a variance equal to 1 . 0×10−5 , respectively . For the precision parameters σ2μ and σ2s , we adopted a gamma distribution ( 0 . 5; 0 . 0005 ) . The simulations were made using the software WinBugs 1 . 4 . Details pertaining to spatial statistical modeling for the relative risk of VL are available in supporting information .
Geocoding was possible for 93% of the infected dogs ( 34 , 127/36 , 627 ) and 93% of the human VL cases ( 384/412 ) . The remaining cases were excluded due to inconsistencies in the addresses . Among the 28 cases of human VL whose residential address was not geocoded ( 7% ) , 11 were homeless and 17 could be allocated to a sanitary district , but not in the geographic units of analysis ( 146 small-areas ) . The loss on geocoding can be considered homogeneous in all health districts of Belo Horizonte . This allows us to consider that the lack of geocoding of cases is not associated with the occurrence of VL or other variables . The Figure 2 shows the spatial distribution of the number of infected dogs per 1 , 000 inhabitants ( A ) and the cumulative incidence rates of the human visceral leishmaniasis cases per 100 , 000 inhabitants ( B ) . Because of the missing cases , this figure underestimates the number of infected dogs and human VL cases . In Figure 2-A , the higher levels of canine infection are concentrated in five of the nine health districts of the city . In Figure 2-B , the highest cumulative incidence rates of human VL ( crude rates ) are located in six health districts . These maps show the overlap of human VL cases and infected dogs . The scatter plots in Figure 3 present the log-RR-VL on the vertical axis and the covariates on the horizontal axis . The points represent each of the 146 coverage areas . We observed a positive linear trend between the log-RR-VL and the canine infection represented by the covariate “number of infected dogs to inhabitants” ( Figure 3-A ) . The graphs for Health Vulnerability Index ( Figure 3-B ) and its components that displayed a positive linear trend with log-RR-VL are also presented ( Figure 3-C to 3-F ) , as follows: percentage of illiterate people; percentage of householder with fewer than four years of education; percentage of householder with an income fewer than twice the Brazilian minimum wage ( US$ 200 . 00 ) and average income ( inverted ) of the householder . Since a smaller income usually results in a higher vulnerability of the population of an area , the inverted average income was used in the analysis to preserve a possible positive relation with the relative risk . All these covariates were selected for further statistical modeling of the log-RR-VL . The following Health Vulnerability Index components did not display evidence of an association with the occurrence of VL: proportion of the householders aged 10–19 years and indicators of inadequate or absent services of water supply , sanitary sewage and waste disposal . It was observed lack of association between the log-RR-VL and the NDVI ( Figure 3-G ) . However , the NDVI was included in the modeling spatial because of the vegetation coverage is favorable to the vector presence . The locations of the residence of the human VL cases exhibited between 716 and 1 , 143 meters of the altitude . A non-linear relation between the log-RR-VL and altitude was found ( Figure 3-H ) . In general , a lower altitude indicated a higher log-RR-VL in BH , with a higher frequency of cases below 949 meters . Table 1 shows the results of the estimates parameters of the univariate spatial models for the log-relative risk of VL . Except for the NDVI and altitude , the other covariates displayed coefficients whose means may be considered to be significantly different from zero , since their 95% credibility intervals do not include the value of zero . The full models with the best fit to the data to explain the relative risk of VL in Belo Horizonte comprised two variables: number of infected dogs per inhabitants and education or income ( models 1 to 4 ) . Model 1 presented the lowest deviance information criterion ( DIC ) value and , therefore , fit slightly better to the data ( Table 2 ) . According to this model , an increase of 0 . 01 infected dogs per inhabitant leads to an average increase of 13% in the relative risk of VL in areas with the same mean householder income . Considering areas that have the same number of infected dogs per inhabitants , a decrease of 10% in the mean householder income , which means an increase of 11 . 1% in the inverted mean income , makes the relative risk of VL to be multiplied by 6 . 6 , on average . The map of this model highlights the 14 areas with the highest RR-VL in Belo Horizonte ( Figure 4 ) . The estimates of RR-VL ranged from 0 . 42 to 2 . 55 and in 54% of the areas ( 79/146 ) they were greater than 1 . 0 . The estimates of RR-VL are significantly higher than 1 . 0 ( inferior limit of the 95% credibility interval ) in 14 areas and 12 of them are concentrated in the northern region of the city .
In this study , the relative risk of human visceral leishmaniasis was shown to be correlated with income , education and the number of infected dogs per inhabitants in Belo Horizonte . The indicators of income and education of the Health Vulnerability Index were significantly associated with the relative risk of VL , which suggests the usefulness of this index in the planning and prioritization of areas for control actions in Belo Horizonte . It should be stressed that income and education are related to each other and to several health problems . Therefore , income can serve as a proxy for the socio-economic-cultural context . Although the ecological correlations do not allow for causal inferences [28] , this study has permitted a joint analysis of the intra-urban differences that are potentially associated with the occurrence of human VL cases . In Belo Horizonte , other studies suggest intra-urban differences related to risk of VL . These studies suggest high indices of vector diseases in the poorest regions of the city [29] , and a low socioeconomic level of dog owners and canine infection by L . infantum [19] . Ecological studies performed in Teresina ( the capital of Piauí State , northeastern Brazil ) suggest a spatial correlation between a high incidence of VL and areas with less urban infrastructure and precarious life conditions [6] , [30]–[31] . In Belo Horizonte , the wide distribution of water supply ( 98 . 7% of the population ) , sewage ( 91 . 1% ) , and waste collection ( 97 . 9% ) services [20] likely reflect the higher degree of urbanization of the city . In Belo Horizonte , the ratio between the number of infected dogs per inhabitants in geographic areas and the occurrence of the human VL cases in these areas has not been described in the literature . The use of this indicator was only possible due to the large number of tested dogs ( >150 , 000 dogs/year ) , which was equivalent to more than half the dog population , and to the existence of an information system to record control activities of the canine reservoir . In Iran , an increase in seropositivity in children was associated with an increase in the canine population and the dog/human ratio [32] . However , canine infections were not evaluated , which is the main methodological difference with the present study . In Belo Horizonte , a spatial correlation between human cases and canine infections was suggested [13] in addition to a greater likelihood of human cases due to the presence of animals in the neighboring area [4] . Study conducted in Araçatuba ( São Paulo State , southeastern region ) , showed that a higher concentrations of human cases were detected in areas with a higher prevalence of infected dogs [33] . Ecological association between canine infections and human VL cases were influenced by the worse socioeconomic conditions in a study carried out in Teresina [30] . All these results highlight the need for new control strategies aimed at infected dogs in urban areas . As pointed out in our results , in the multivariate spatial analysis , the altitude did not remain in the final-models . Nevertheless , another study conducted in Belo Horizonte suggested a concentration of infected dogs and human VL cases between 780 and 880 m [14] . In Sudan , the average rainfall and the altitude were the best predictors of VL incidence [34] . In addition , in India , altitude was associated to the risk of VL , and poverty was cited as a determinant of transmission of the disease [35] . Vegetation , represented by the NDVI , was not shown to be associated with the relative risk of VL in Belo Horizonte . Elsewhere , it was identified as an environmental factor predictive of disease risk [30]–[31] , [34]–[36] . In Teresina , higher VL incidence rates in areas with worse socioeconomic conditions , high population growth , and abundant vegetation suggested a relation between the occupation of the area and its vegetation coverage , which was represented by the NDVI [30] . It is probable that the differences in levels of urbanization and infrastructure of Teresina and Belo Horizonte have leaded to these different results . In Brazil , VL is a disease requiring mandatory reporting and whose drugs for treatment are provided exclusively by the government and released only after reporting the case to the Brazilian Reportable Disease Information System . In addition , a three-decades of the occurrence of the disease in Belo Horizonte , the capacity to care for cases through the healthcare network , and the existence of an organized health surveillance system allow for the assumption that there is very little underreporting of cases . However , because of the missing geocoding ( 7% ) , the estimates in Figure 2 include an underestimated number of cases . The use of spatial statistical modeling allowed for estimations of incidence rates smoothed by the spatial dependence between neighboring areas . This Bayesian approach minimizes the instability of rates resulting from a low frequency of cases in small-areas , eliminates a large part of the randomness not associated with the risk factors , and overcomes the political-administrative delimitations of the areas [6] , [23]–[24] , [26]–[27] . Therefore , this approach proved to be useful for the identification of small-areas with a higher risk of VL and exhibited its operational applicability in surveillance and control in an urban environment with an unequal spatial distribution of the disease . Some limitations of this study should also be highlighted . One limitation is related to the temporality of the Health Vulnerability Index , which uses socioeconomic indicators taken from the demographic census of the year 2000 [20] . The updating of this index depends on the results of the census from 2010 , which was no available during this study . The broad distribution of the vector L . longipalpis has been described in Belo Horizonte and , more precisely , in the peridomiciles of the households [12] , [14]–[15] . Based on this knowledge , households were considered to be likely infection sites . Therefore , the geocoding of human VL cases in the level of the household can be a limitation of this study because there would be an ideal indicator of the risk of transmission for all situations . In spite of the fact that Belo Horizonte has high coverage of geocoding system , some problems related to the absences or inconsistencies in addresses could be happen at the time of human VL case reporting and during the canine survey ( eg . homeless individuals; house number inexistent ) . Loss in geocoding of human cases and canine infection was around 7% and it can be considered homogeneously distributed over the city . Hence , it is likely that the loss was not associated with geographical locations and suggested absence of selection bias in spatial analysis [37] . Areas of the municipalities neighboring to Belo Horizonte were not included in the Bayesian analysis to estimate incidence rates . This point would be other limitation of our study . It is noteworthy that georeferenced data of VL patients from neighboring counties may not be available due to the lack of the geocoding system . In Belo Horizonte , it is likely that the effectiveness and sustainability of the visceral leishmaniasis control program are influenced by the complexity of the disease in a large territorial area with high human and canine population densities and intra-urban differences . In this context , identifying higher risk areas may help in the surveillance of VL , direct the prioritization of small-areas for specific interventions [30] , [38] , and contribute to the effectiveness and reduction of operational costs of the Visceral Leishmaniasis Control and Surveillance Program [6] , [39]–[40] . | Visceral leishmaniasis ( VL ) is a vector-borne disease whose factors involved in transmission are poorly understood , especially in more urban and densely populated counties . In Brazil , the increasing occurrence of human VL cases in urban centers is a challenge for the control program . We aimed to identify the risk areas for VL and the risk factors involved in transmission in Belo Horizonte , a large urban area of the Brazil . At the same geographical space , we analyzed human VL cases ( n = 412 ) , canine infection and socioeconomic and environmental features . We identified a concentration of high-risk small-areas of human VL cases in the northern part of the city , marked by worse levels of education and income , and higher number of infected dogs per inhabitants . The spatial analysis used is useful for the identification of small-areas with a greater risk of VL and displays operational applicability in the control program in an urban environment with an unequal spatial distribution of the disease . Thus , the frequent monitoring of risk of human VL according to small-areas is important to direct and prioritize the actions of the control program in urban environment , especially in big cities . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Relative Risk of Visceral Leishmaniasis in Brazil: A Spatial Analysis in Urban Area |
Developing embryos exhibit a robust capability to reduce phenotypic variations that occur naturally or as a result of experimental manipulation . This reduction in variation occurs by an epigenetic mechanism called canalization , a phenomenon which has resisted understanding because of a lack of necessary molecular data and of appropriate gene regulation models . In recent years , quantitative gene expression data have become available for the segment determination process in the Drosophila blastoderm , revealing a specific instance of canalization . These data show that the variation of the zygotic segmentation gene expression patterns is markedly reduced compared to earlier levels by the time gastrulation begins , and this variation is significantly lower than the variation of the maternal protein gradient Bicoid . We used a predictive dynamical model of gene regulation to study the effect of Bicoid variation on the downstream gap genes . The model correctly predicts the reduced variation of the gap gene expression patterns and allows the characterization of the canalizing mechanism . We show that the canalization is the result of specific regulatory interactions among the zygotic gap genes . We demonstrate the validity of this explanation by showing that variation is increased in embryos mutant for two gap genes , Krüppel and knirps , disproving competing proposals that canalization is due to an undiscovered morphogen , or that it does not take place at all . In an accompanying article in PLoS Computational Biology ( doi:10 . 1371/journal . pcbi . 1000303 ) , we show that cross regulation between the gap genes causes their expression to approach dynamical attractors , reducing initial variation and providing a robust output . These results demonstrate that the Bicoid gradient is not sufficient to produce gap gene borders having the low variance observed , and instead this low variance is generated by gap gene cross regulation . More generally , we show that the complex multigenic phenomenon of canalization can be understood at a quantitative and predictive level by the application of a precise dynamical model .
C . H . Waddington inferred developmental canalization of gene expression by noting that differentiation leads to the formation of discrete types of tissue , rather than a continuous blend , and that genotypic and environmental variation is suppressed at the phenotypic level in wild-type but not mutant organisms . These points led him to state that “developmental reactions , as they occur in organisms submitted to natural selection , are in general canalized . That is to say , they are adjusted so as to bring about one definite end-result regardless of minor variations in conditions during the course of the reaction” ( p . 563 in [1] ) . These adjustments manifest themselves as a reduction in the variation of phenotypes . In this article , we make a detailed experimental and theoretical study of the canalization process in the gap gene system of Drosophila . Interestingly , this investigation puts us in contact with a number of recent investigations of variation in the gap gene system as we now explain . Among contemporary geneticists and evolutionary biologists , the buffering of phenotypic variation to underlying genotypic variation in wild type is well known , and recent experimental studies have identified individual genes that are responsible for this aspect of canalization [2 , 3] . Theoretical studies have demonstrated that this phenotypic buffering is an intrinsic consequence of the adjustments of developmental trajectories postulated by Waddington [4] , but direct evidence of developmental canalization was still lacking . Direct evidence of developmental canalization must meet two requirements . First , the system must show variation between individuals that decreases significantly over the course of development . Second , this decrease in variation must be demonstrated to be an inherent property of the system under study , rather than being imposed from a different part or from outside the organism . The first requirement implies the need for data on the dynamics of developmental determinants , some or all of which will be molecular . The demonstration that reduction in variance of a developmental system is inherent to that system requires precise understanding of the consequences of the interaction of many developmental determinants . Understanding such consequences requires a quantitative model . The segmentation system of Drosophila melanogaster [5 , 6] is highly suitable for studies of canalization because it is already known that the first of the above requirements , decrease in phenotypic variability over time , is satisfied . In work reported elsewhere [6] , we performed a quantitative analysis of the expression of segmentation genes expressed from the maternal genome only ( bicoid ) , the maternal and zygotic genomes ( caudal and hunchback ) , and the zygotic genome ( Krüppel , giant , knirps , tailless , fushi tarazu , even skipped , runt , hairy , odd skipped , paired , and sloppy paired ) . These expression data have cellular resolution in space and about 6 . 5-min resolution in time , and comprise either the expression levels of even skipped ( eve ) and two other genes in individual embryos or the integrated averaged expression of all 14 protein gene products . These data show that segmentation gene expression is highly variable among individual embryos in cleavage cycle 13 and the early part of cycle 14 . There is extensive variation in expression levels , locations of domain borders , and the time and order of the appearance of individual domains . However , the variation in the expression patterns reduces over time and is significantly lower at the onset of gastrulation than at earlier times . In this article , we restrict our attention to a particularly important class of phenomena concerning the variation in the location of the boundaries of zygotic gene expression domains . These boundaries shift when maternal gradients are perturbed , and hence are at least in part under the control of these gradients [7 , 8] . Nevertheless , the variation in the boundary positions of gap and pair-rule expression domains is much lower than the variation in the maternal gradient of Bicoid ( Bcd ) protein [6 , 9 , 10] . Under the simplest model of specification , expression borders would form at a fixed threshold of its concentration [11] . The variation of the Bcd gradient can be measured in terms of the range ( ρBcd ) or standard deviation ( σBcd ) of the position ( xBcd ) where it crosses the threshold concentration . In our data σBcd = 4 . 6% egg length ( EL ) , whereas gap gene domain border positions have σ ∼ 1% EL [6] . Pair-rule stripe border positions also have similar low variation [6] compared to the Bcd threshold position . Of the large number of gene expression borders that have lesser variation than Bcd , the posterior border of the anterior hunchback ( hb ) domain has received intense scrutiny [9 , 12–18] . Gregor and colleagues [16] measured the absolute concentration of Bcd in the nuclei at the hb border using a Bcd-GFP fusion rescue construct . This measurement led the authors to two important but mutually contradictory conclusions . First , using a result from bacterial chemotaxis theory [19] , it was shown that noise due to small number of Bcd molecules is high enough so that the hb border cannot be accurately specified by Bcd alone . Second , by measuring Bcd profiles from several live embryos in parallel , it was found that σBcd ∼ 2% EL . The amount of variance reduction implied by this value of σBcd is significantly smaller than the estimates from fixed-tissue experiments [6 , 9] , presumably because the GFP measurements do not have scaling error introduced in the setting of microscope gain in separate fixed-tissue experiments . Under the assumption that Bcd is the only regulator of hb , it was argued that these results can be reconciled if spatial averaging from the diffusion of Hb molecules is taken into account , and the authors draw the conclusion that Bcd alone is sufficient to specify the hb border accurately . The assumption that hb is under the sole control of Bcd is incorrect , because its border position changes in embryos mutant for giant ( gt ) [9] , Krüppel ( Kr ) [20] , Krüppel;knirps ( Kr;kni ) [20] , and in embryos lacking Nanos [9] . Apart from the average position , the variation of the position is also increased in embryos lacking chromosome arm 3L or Nanos to ∼2% EL and ∼1 . 6% EL respectively [9] , close to the Bcd variation measured in live embryos . The Bcd-dependent response of a fragment of the hb promoter has variation close to Bcd variation ( see Figure 4 in [21] ) ( ρBcd ∼ 9% EL; compare with Figure 1A ) and higher than the variation of endogenous hb ( Figure 1B ) . All these experiments suggest that endogenous hb expression has lower variation than Bcd , which is increased to the level of Bcd variation when other inputs to hb are removed . To make the relationship between Bcd variation and hb border variation clear , we have plotted the data of Gregor et al . and compare it with fixed-tissue Hb data ( Figure 1 ) . The hb border is steep , and its position can be measured unambiguously in fixed tissue since it is not sensitive to scaling variation , unlike a Bcd threshold concentration . This comparison shows that the Bcd threshold position at which the hb border forms is about twice as variable as the border position . It is important to note that the hb border is just one among many that have low positional variance . As an example , consider the anterior border of the posterior kni domain , which is activated by Bcd [22] . This border is located ∼10% EL posterior to the hb border and has a positional variance of ∼1% EL [6] . At this position , the Bcd-GFP data have a variance of ∼4% EL ( see Figure 5C in [16] ) , which is much larger than that of the kni border . This example clearly contradicts a picture of positional specification in which Bcd specifies all of its targets accurately . Like hb , kni is not regulated by Bcd alone , but also by Caudal ( Cad ) [22 , 23] , Hb [20 , 24] , and Gt [25] . The examples of the hb and kni borders show that it is not possible to explain the variance of a single expression border position in isolation . It is necessary to take into account all the genetic interactions required for generating the expression patterns over space and time . Gene circuits [26 , 27] are dynamical models that can accurately reproduce observed gene expression patterns by reconstituting the required set of genetic interactions in silico . In the gene circuit method , ( 1 ) a gene circuit model , in which one real number characterizes the regulatory effect of one gene on another , is ( 2 ) fit to quantitative gene expression data by ( 3 ) parallel Lam simulated annealing [28 , 29] or some other nonlinear optimization method [30–32] , and finally ( 4 ) biological conclusions are obtained . This method has been successfully used to analyze both the pair-rule and gap systems [27 , 33–36] and performed better than other models in a comparative study [31] . In what follows , we briefly describe the gene circuit used in the Results section “Gap Gene Circuits”; a full description is provided in Protocol S1 ( Section S1 ) . We then demonstrate in the Results section “Simulation of Bcd Variation and Size Variation” that a gene circuit model for the gap genes hb , Kr , gt , and kni under maternal control and including mutual gap–gap interactions is sufficient to explain the reduction in the variance of gap gene domain border positions , showing that it is not necessary to postulate an undiscovered gradient [12 , 14 , 15] or active transport [13] to explain this property . In the Results section “Variance Reduction by Gap Gene Cross Regulation and Experimental Verification , ” we characterize the regulatory interactions underlying variance reduction . This analysis shows that the reduction in variance of the hb border is a consequence of repression by Kr and Kni , and hence predicts an increase in variance when these factors are removed . We present new experimental data from Kr;kni double mutant embryos that validate this prediction and show the correctness of the gene circuit approach . The validation of the gene circuit approach shows that the gap gene system also satisfies the second requirement for demonstrating canalization , which is that variance reduction arises from intrinsic properties of the system .
We constructed a gene circuit that models the expression patterns of the gap genes hb , Kr , gt , and kni in a region of the blastoderm extending from 35% EL to 92% EL along the anteroposterior ( A–P ) axis . The protein products of these genes are transcription factors [37–40] and localize to nuclei . The model computes the time course of nuclear protein concentrations in a time interval beginning at the start of cleavage cycle 13 and ending at the onset of gastrulation [41] ( Figure S2 in Protocol S1 ) . The initial conditions were specified using cleavage cycle 12 gene expression data [6] ( Section S1 in Protocol S1 ) . The 13th nuclear division occurs in the duration of the model ( Figure S2 in Protocol S1 ) after which the nuclei in the modeled region are divided , and daughter nuclei inherit the state of the mother nucleus . The gene circuit equations model the processes of protein synthesis , degradation , and diffusion ( Section S1 . 1 in Protocol S1 ) . During the mitosis preceding the 13th nuclear division , the protein concentrations are governed only by diffusion and degradation as synthesis shuts down [42] . During interphase , the synthesis rate of a protein depends on the regulation of the gene by the other gap proteins , the maternal gradient Bcd , and the upstream regulators Cad and Tailless ( Tll ) . The regulatory effect of a protein is characterized by the product of its concentration and its regulatory strength . The concentration varies from nucleus to nucleus , but the regulatory strength , a parameter of the model , is invariant throughout the embryo , reflecting the fact that the zygotic genome is the same in each blastoderm nucleus . The sum of all the regulatory terms for a gene , called the total regulatory input ( u ) , determines its synthesis rate through the sigmoidal regulation-expression function ( Figure S1 in Protocol S1 ) . To fully specify the model the concentrations of Bcd , Cad and Tll must be provided . The concentrations of Cad and Tll were provided by the interpolation of average data [6] in time ( Section S1 . 2 in Protocol S1 ) . A representative profile of Bcd was provided as follows . The Bcd gradient is essentially stationary during cleavage cycles 13 and 14 [6 , 43 , 44] , and hence , its concentration vBcd is assumed to be constant in time . It is known through antibody studies [6 , 9 , 12 , 43] and a recent GFP-Bcd [44] study that the Bcd profile is an exponential function of A–P position x , so that The arithmetic mean of exponential curves is not exponential . Thus , it is not possible to generate a representative Bcd profile by taking an arithmetic average over embryos . Instead of averaging , a representative individual Bcd profile vBcd ( x ) was chosen in the following manner . We obtained data from 88 cycle 13 embryos immunostained for Bcd and removed the background signal from the Bcd profiles as described [45] . Taking the logarithm of Equation 1 , we get The background-removed profiles were then fit by linear least-squares to Equation 2 . This procedure yielded two parameters for each profile , γ ( length scale ) , and A ( concentration scale ) . Figure 2A is a scatter plot of log A with γ . A profile ( “median” profile ) was chosen such that its parameters lie in the middle of the scatter plot ( Figure 2A ) . Since this study concerns only embryo-to-embryo variation and not nucleus-to-nucleus noise , the exponential fit of the profile was used in the model . Using background-removed profiles directly in the model yields circuits with the same properties as those of circuits with exponential fits ( See section “Simulation of Bcd Variation and Size Variation” below and Table S4 in Protocol S1 ) . To compare with quantitative gene expression data [6] , we calculate the solution of the model at multiple time points . The data are classified into time classes to give nine time points for comparison ( see Table 1 and Figure S2 in Protocol S1 ) . We obtained specific gene circuits by performing a least-squares fit of the model to gene expression data from nine time points ( see Methods ) . We used a median Bcd profile from an individual embryo as described in the previous section , together with averaged expression profiles from the other genes considered . Twenty-three circuits were obtained with almost identical network topology ( Table S3 in Protocol S1 ) and expression patterns . One of these circuits was chosen for further study . Its parameters are given in Protocol S1 ( Tables S1 and S2 ) . The circuit's gap gene patterns ( Figure S4A and S4B in Protocol S1 ) are consistent with data except for two minor defects ( Section S1 . 3 in Protocol S1 ) and the circuit has the same network topology that was obtained previously [31 , 35 , 36] . In the previous section , it was shown that the model correctly predicts the positional variance of six borders across a 50% EL region of the embryo . In this section we use the model to characterize the gene interactions responsible for this canalizing behavior of the embryo . This was done in two steps . First , we determined the regulators that set each of the correctly predicted domain borders . The basic idea of this method is that at a border , the relative activation of a gene g ( ua ) changes from a value close to zero to a value close to one as a result of changes in regulatory input . Those regulatory inputs responsible for changing the state of a gene from off to on can be read from an appropriate graph [33–36] ( see Section S3 in Protocol S1 for further details ) . In the second step , we ascertained how the regulation of a gene at its domain border is affected when subjected to Bcd variation by comparing the strengths of the regulatory inputs at the border position under differing individual Bcd profiles . The results of the regulatory analysis are shown in Figure 4A–4C for the posterior borders of the hb anterior domain , the central Kr domain , and the posterior kni domain; and in Figure 5A–5C for the anterior border of the posterior kni domain and both borders of the posterior gt domain . This analysis is consistent with earlier results [31 , 36] . We briefly summarize the results here; see the captions of these figures and previous studies [31 , 36] for details . In general , these borders are set up by an activator ( either bcd or cad ) and two repressors . There are two tiers of repression . Two pairs of genes with complementary domains in the modeled region , Kr and gt as well as kni and hb ( Figure S4B in Protocol S1 ) , have strong mutual repression . Genes with overlapping domains in the modeled region , gt and kni for example , have weak mutual repression . One border , the anterior border of the posterior gt domain , is set by a single repressor , Kr . This is well supported by the very large anterior shift of this border in Kr– embryos [58] . We studied the effects of varying the Bcd profile in the gap gene circuit as follows . The position of each border was calculated in the 88 simulations using Bcd profiles from individual embryos . With the exception of the borders of the posterior gt domain , the simulations were pooled into 1% EL bins according to the position of each border . We then averaged the Bcd activation and gap repression levels at the border position in each group . To understand how the regulation of a border is affected under Bcd variation , we plotted the pooled Bcd activation and gap repression levels together for each border under consideration ( Figures 4D–4I and 5D–5G ) . The low positional variation of the posterior border of the anterior hb domain has been studied extensively [9 , 12 , 14–16 , 59] . Our analysis shows that this border forms by Bcd activation and repression from Kr and Kni ( Figure 4A ) . The results of varying the Bcd profile on the regulation of the hb border are presented in Figure 4D and 4G . It shows that both activation and repression levels remain correlated for different positions of the hb border , with embryos having hb borders in more posterior positions having greater activation by Bcd and greater repression by Kr and Kni . This correlation suggests an epistatic balance between activation and repression at the hb border . These results are in apparent contradiction to a previous report indicating that the standard deviation of the location of the hb border was unchanged in single mutants for either Kr or kni [9] . We found the same result for single mutants ( unpublished data ) but in double mutants , uncharacterized previously , the standard deviation of the location of the anterior hb border doubles and that of the posterior border of the third anterior gt domain increases significantly ( Figure 6 ) . This experimental result strongly supports the model and rules out a picture in which Bcd provides the sole input to hb [16] . The same epistatic balance is seen in the posterior borders of the Kr and kni domains ( Figure 4E , 4H , 4F , and 4I ) . Of the other correctly predicted domain borders , the anterior border of kni is set by repression from Kr and Hb ( Figure 5A ) . Reduced levels of Bcd in the circuit cause this border to form at more anterior positions ( Figure 5D ) ; however , the shift is limited by increased levels of Kr and Hb repression toward the anterior ( Figure 5E ) . Both borders of the posterior gt domain show epistatic balance ( Figure 5F and 5G ) , but it does not correlate with position , because Bcd is not a morphogen in this region and does not provide positional information .
We have shown that the reduction in the variation of gene expression in the gap gene system takes place because of cross regulation between zygotic gap genes . Cross regulation is an intrinsic property of the system , and hence we have demonstrated that the system canalizes . More specifically , the experimentally supported analysis of the Results section “Variance Reduction by Gap Gene Cross Regulation and Experimental Verification” shows that the Bcd gradient is not sufficient to generate gap gene borders with the observed low variation . Our results , together with those of other investigators , rule out three classes of theoretical models that have been invoked to settle certain questions associated with the positional variation of hb . The first two classes of models invoke either active transport [13] or a second gradient at the posterior pole of the embryo [12 , 14 , 15] as a mechanism for the reduction in variation of the hb border . With one exception [15] , these studies only considered the hb border , while the work reported here correctly predicts the variation of six borders , some of which are not under the control of bcd . Moreover , the border positions in the model reported here scale with egg length ( Figure 3E ) without the need to invoke an undiscovered gradient . A third type of model [21] invokes transient behavior of the Bcd gradient and its associated thresholds during cycles 10–13 as a mechanism of variance reduction . This model has been invalidated by the experimental demonstration that intranuclear Bcd concentrations are in fact constant during this period of time [44] . Furthermore , threshold concentrations of Bcd propagate toward the posterior of the embryo during the establishment of the Bicoid gradient , but gap domains in the posterior , in fact , shift in an anterior direction [35] . A third contradiction with experiment is that the model used by the authors of [21] produces incorrect mutually exclusive gap domains in the presence of diffusion . Many seemingly puzzling experimental results concerning the variance of expression borders driven by Bcd can be understood in a simple and unified manner by considering our results together with two recent studies of the Bcd gradient , one in vivo [16] and the other in fixed tissue [18] . Our comparison of Bcd-GFP data with fixed tissue hb data ( Figure 1 ) shows that the range of xBcd in vivo is 10% EL , but that of xHb in fixed tissue data is 5% EL . This comparison finds support in the fixed-tissue study [18] in which a modified staining protocol was used and data were acquired in a single imaging cycle to minimize experimental error . The Bcd data thus obtained had the same level of error as Bcd-GFP data [16] , yet a direct comparison between the Bcd and Hb data showed that the posterior border of the anterior hb domain was half as variable . The observation that the range of xBcd is twice that of xHb is extremely important , because there are many experimental manipulations that can double the range or standard deviation of xHb , making it equivalent to the range or standard deviation of xBcd . First , these results indicate that the increased variance of xHb in Kr;kni double mutants ( Figure 6A caption ) can be interpreted as arising from the unmodified action of Bcd without reduction in variance mediated by other genes . This is a key point , because the larger standard deviation of xBcd observed in data from fixed tissue here and elsewhere [9] appeared to imply that variance reduction still took place in double mutants , albeit at a reduced level . Second , a 9% range of border positions has been observed from reporters driven by a fragment of the hb promoter believed to contain only Bcd binding sites ( see Figure 4 in [21] ) , suggesting that this expression reflects the underlying variation of xBcd . Third , when chromosome arm 3L is completely removed [9] , the standard deviation of xHb doubles . We suggest that all three of these experimental manipulations uncover unmodified action of Bcd on hb by removing additional modulators respectively in trans , in cis , and by a currently uncharacterized mechanism . Other important experimental results can be understood by recalling that the maximum accuracy of the Bcd gradient is inversely proportional to the square root of the Bcd concentration [16 , 19 , 60] , suggesting that borders under unmodified Bcd control will have lower variation at more anterior positions . This fact explains the observation that a synthetic reporter construct driven only by Bcd has the same standard deviation as xHb [59] . The synthetic construct forms its border at 28 . 6% EL , where the in vivo data ( see Figure 2B in [16] ) indicate that the Bcd concentration is about three times greater than at the native hb border . If the standard deviation of 2 . 2% EL seen here in Kr;kni double mutants is the unmodified response to Bcd at 47% EL , then the expected standard deviation at 29% EL is 1 . 3% , which is close to the 1 . 6% observed ( see Figure 5C in [59] ) . Finally , the fact that the standard deviation of the posterior border of the anterior gt domain increases by less than that of hb in Kr;kni mutant embryos can be understood by noting that it is located at about 40% EL . At this position , the in vivo data appear to indicate that the Bcd concentration is 1 . 69 times greater than that at the hb border . This concentration ratio predicts that the standard deviation of this border in Kr;kni double mutants will be 1 . 7% EL , while the actual value is 1 . 9% EL . In recently published work , He et al . [18] report that the Bcd threshold at which the posterior border of the anterior hb domain forms is not constant but varies in a correlated manner with the slope of the Bcd profile . Figure 4D shows that in our simulations , the Bcd level at which the hb border forms changes as the Bcd profile is varied . Figure 4G shows that this effect is a consequence of the dynamic adjustment of the repression levels , arising from gap gene cross regulation . He et al . speculate that the stability of the Bcd protein might affect its ability to activate hb . However , our results suggest that the dynamic adjustment of the activation level is in fact an indirect consequence of gap gene cross regulation . One remaining question concerns the reason for the correct prediction of gap gene domain border variation presented here , as it is clear that the fixed tissue Bcd data used for the theoretical study exaggerate the standard deviation of xBcd [16] . We believe that the high degree of canalization observed ( Figure 3A ) makes the theoretical model quite insensitive to initial variation , as is true of the biological system itself . In a companion work [61] , we show that robust insensitivity to initial conditions is an inherent mathematical property of the gene circuit equations for this system . Moreover , the observed in vivo standard deviation of xBcd was obtained in an approximately isogenic stock grown at fixed temperature , while natural populations have much more variable genotypes and develop in the presence of temperature fluctuations . These factors alter the profile of the Bcd gradient but have little effect on the border position of downstream targets [9 , 62] . We believe that the fixed-tissue Bcd profiles used in the present study capture this extra level of variability characteristic of natural populations . This analysis of canalization has focused on the reduction of variation of border positions of the gap genes with respect to the threshold positions of the maternal determinant Bcd ( Results sections “Simulation of Bcd Variation and Size Variation” and “Variance Reduction by Gap Gene Cross Regulation and Experimental Verification” ) . The gap gene borders are specified at a low enough molecular number of Bcd [16] that intrinsic fluctuations are a significant source of variation [16 , 60] . The buffering of phenotypic variation against intrinsic fluctuations in molecular number , while not considered by Waddington , is likely to be an important aspect of canalization in biological systems . It is noteworthy that we were able to model canalization using a deterministic model with what were , essentially , stochastic initial conditions . Future work will need to treat the role of intrinsic fluctuations in the formation of maternal gradient systems [63 , 64] . While stochastic interactions undoubtedly occur among zygotic gap gene products , understanding the functional role of stochastic processes in the segmentation system will require the formulation of a fully stochastic model . Such a model must also produce the observed reduction in variation in the gap gene system , although the numerical details will differ . Waddington hypothesized that canalization arises from gene interactions [1 , 65–68] . We have identified specific regulatory interactions among the gap genes responsible for canalization . It is noteworthy that variation reduction is an emergent property of the system and does not require special built-in mechanisms such as a posterior gradient . The mutual repression between the gap genes that is required for correct spatiotemporal expression is also responsible for canalization , providing support for the “intrinsic” [67] or “single-mode” [69] canalization hypothesis . Our results make it possible to characterize in a more precise way how canalization works . The reduction of gap gene expression variation over time [6] suggests that canalization occurs because the possible developmental trajectories of the system tend to converge over time , thus reducing variation . In an accompanying manuscript [61] , we show that the trajectories of the gap gene system in individual nuclei do tend to converge and are stable against perturbations . This convergence and stability are a consequence of the fact that the regulatory interactions of the gene circuit cause the state of the system to approach a stable steady state [70 , 71] , also called an attractor . A given nucleus can potentially approach one of many attractors , a property called multistability , and a particular attractor is chosen based on the initial state of the nucleus . Waddington inferred canalization , in part , from the observation that there are discrete tissue types rather than a continuous blend . Multistability provides a rigorous mathematical justification for the connection between canalization and the discreteness of tissue type . If the initial state is perturbed by a small amount , the system is attracted by the same attractor , providing variation reduction . However , if the perturbation is large enough , or if attractors are created or destroyed , the trajectory can switch to a different attractor providing a discrete response . Our analysis [61] suggests that switching between attractors occurs in the gap gene system and is the mechanism by which the posterior border of the anterior hb domain forms . These results show that the gap gene network is a system with the intrinsic property of reducing variation , a form of error correction , and that this canalization property is a consequence of the mutual regulatory actions of the gap gene network .
The methods used to obtain and characterize the quantitative data are as described in earlier work [6] . All gene expression levels are on a scale of 0–255 , chosen to maximize dynamic range without saturation . The Kr;kni double mutant embryos were made by crossing Kr1 and Df ( 3L ) ri-79c flies . 3–4-hours-old embryos were fixed and stained as described [72] using guinea pig anti-Kr and anti-Kni antibodies with Alexa Fluor 488 conjugated secondary antibody , rabbit anti-Gt and rat anti-Hb antibodies with Alexa Fluor 555 and 647 conjugated secondary antibodies ( Molecular Probes ) , respectively . Embryos null for Kr and Kni showed no signal on the first channel . Double mutants in lateral orientation were scanned using a Leica TCS SP2 confocal microscope and the images were then processed as described [73] . Domain border positions for wild type and mutants were determined by calculating the A–P position at which the relative concentration is half its maximum value in the domain . The local maximum in a domain was determined using a quadratic spline approximation [74] . The numerical implementation of the gene circuit equations is as described [35 , 36] with the addition of time varying external inputs ( see Section S1 . 2 in Protocol S1 ) . The gap gene circuit was fit to integrated gap gene data [6] using Parallel Lam Simulated Annealing ( PLSA ) [28 , 29] . PLSA minimizes the root mean squared ( RMS ) difference between model output and data . The minimized RMS difference for a circuit is a measure of the quality of fit , and is called the RMS score . The RMS score indicates the average difference between the model and the data . Search spaces , penalty function , and other annealing parameters were as described [27 , 36] . The circuit analyzed in detail had an RMS score of 10 . 76 , corresponding to a proportional error in expression residuals of about 4–5% . Protocol S1 ( Section S1 . 3 ) contains further details . | Animals have an astonishing ability to develop reliably in spite of variable conditions during embryogenesis . More than 60 years ago , it was proposed that this property of development , called canalization , results from genetic interactions that adjust biochemical reactions so as to bring about reliable outcomes . Since then , a great deal of progress has been made in understanding the buffering of genotypic and environmental variation , and individual mutations that reveal variation have been identified . However , the mechanisms by which genetic interactions produce canalization are not yet well understood , because this requires molecular data on multiple developmental determinants and models that correctly predict complex interactions . We make use of gene expression data at both high spatial and temporal resolution for the gap genes involved in the segmentation of Drosophila . We also apply a mathematical model to show that cross regulation among the gap genes is responsible for canalization in this system . Furthermore , the model predicted specific interactions that cause canalization , and the prediction was validated experimentally . Our results show that groups of genes can act on one another to reduce variation and highlights the importance of genetic networks in generating robust development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"developmental",
"biology",
"mathematics",
"computational",
"biology",
"evolutionary",
"biology",
"biophysics",
"genetics",
"and",
"genomics"
] | 2009 | Canalization of Gene Expression in the Drosophila Blastoderm by Gap Gene Cross Regulation |
Comparative genomics usually involves managing the functional aspects of genomes , by simply comparing gene-by-gene functions . Following this approach , Mushegian and Koonin proposed a hypothetical minimal genome , Minimal Gene Set ( MGS ) , aiming for a possible oldest ancestor genome . They obtained MGS by comparing the genomes of two simple bacteria and eliminating duplicated or functionally identical genes . The authors raised the fundamental question of whether a hypothetical organism possessing MGS is able to live or not . We attacked this viability problem specifying in silico the metabolic pathways of the MGS-based prokaryote . We then performed a dynamic simulation of cellular metabolic activities in order to check whether the MGS-prokaryote reaches some equilibrium state and produces the necessary biomass . We assumed these two conditions to be necessary for a living organism . Our simulations clearly show that the MGS does not express an organism that is able to live . We then iteratively proceeded with functional replacements in order to obtain a genome composition that gives rise to equilibrium . We ruled out 76 of the original 254 genes in the MGS , because they resulted in duplication from a functional point of view . We also added seven genes not present in the MGS . These genes encode for enzymes involved in critical nodes of the metabolic network . These modifications led to a genome composed of 187 elements expressing a virtually living organism , Virtual Cell ( ViCe ) , that exhibits homeostatic capabilities and produces biomass . Moreover , the steady-state distribution of the concentrations of virtual metabolites that resulted was similar to that experimentally measured in bacteria . We conclude then that ViCe is able to “live in silico . ”
The search for LUCA , the Last Unknown Common Ancestor , is an open problem in evolutionary theory , which has been addressed using many different approaches . After the completion of several bacterial genomes , some authors tried to infer a possible minimal genome ruling out of non essential genes from existing small bacterial genomes . Dispensable genes were detected using both wet-lab techniques ( e . g . , see [1 , 2] ) and comparative genomics methods ( e . g . , see [3] ) . The implicit hypothesis underlying these approaches is that the ancestor genome is made of singular elements only , and therefore would have a minimum size . We are aware of the criticisms raised about this hypothesis ( e . g . , see [4 , 5] ) , but a discussion on this subject would be off-topic for the present paper . Instead , we shall examine how such a simplified organism can be inferred by a comparative genomics approach , specifically following Mushegian and Koonin [3] . They considered the two very small genomes of Haemophylus influenzae and Mycoplasma genitalium , and manually scanned the two correspondent gene lists , so as to remove any element that looked redundant for biological function . The final result of this work was the so-called Minimal Gene Set ( MGS ) , made of 254 singular genes ( the original paper declared 256 genes , but two genes , -mg297 and mg336- , have been counted twice ) . This hypothetical minimal genome was claimed to specify for a very essential prokaryote , but no argument was provided to address the fundamental question of whether a cell equipped with MGS ( call it MGS-prokaryote ) is able to live or not . A direct , biological approach to answer this question could consist in synthesizing this genome , in cloning it in a ghost bacterium , and in evaluating the overall cell viability . However , there are many severe technical problems along this way , which make it hard to get an answer quickly . We instead described this hypothetical cell as a computer program and simulated its behavior in silico . We then tested whether it shows some fundamental properties of living organisms . First of all we checked whether it enjoys homeostasis , i . e . , the capability to reach a steady state in which the concentration of all the chemical species inside the cell fluctuates within a narrow range . We also investigated the capability of a MGS-prokaryote to produce biomass . To model the MGS-prokaryote , we used ( a variant of ) the π-calculus , a process calculus designed to specify concurrent processes , which has already been used to describe biological phenomena [6 , 7] . We represented a complete set of metabolites , metabolic pathways , etc . , involving the genes of the MGS-prokaryote . We ran in silico simulations and we observed the concentration of fundamental metabolites ( ATP , NADH , etc . ) , checking the trend of their time courses toward constant values .
To specify the metabolites and their relationships in terms of biochemical reactions , we used an enhanced version of the π-calculus , which has already been shown to be suitable for describing biological entities [8 , 9] . We refer the reader to [10] and [11] for a complete presentation of the ( enhanced ) π-calculus , and here we survey very briefly its fundamentals . The π-calculus was designed to express , run , and reason about concurrent systems . These are abstract systems composed of processes , i . e . , autonomous , independent processing units that run in parallel and eventually communicate , by exchanging messages through channels . A biochemical reaction between two metabolites , catalyzed by an enzyme , can be modeled in π-calculus as a communication . The two metabolites are represented by two processes , and , in our approach , the enzyme is modeled as the channel which permits the communication . In addition to communications , the π-calculus also allows us to specify silent internal actions , used to model those activities of the cell , the details of which we are not interested in ( e . g . , the pure presence of a catalyst in a reaction , where it is not actively involved ) . The calculus has the means to express alternative behavior , when a metabolite can act in different possible manners: the way to follow is chosen according to a given probability distribution . The main difference between the standard π-calculus and the enhanced version we used in this work is the notion of address . An address is a unique identifier of a process , totally transparent to the user , automatically assigned to all of its child subprocesses . This labeling technique helps in tracking the history of virtual metabolites and reasoning about computations , in a purely mechanical way . In particular , stochastic implementation or causality are kept implicit and are recovered as needed . The enhanced π-calculus shares with other language-based approaches a number of advantages with respect to other formal descriptions . The very specification of the cell is actually a program and can be executed , giving rise to a virtual experiment , unlike other static descriptions such as the SBML [12] . Additionally , specifications turn out to be rather compact when compared , for example , with those expressed by P-systems [13] , which , however , also describe membranes and their activities that we neglect here . Also , the specification of a whole organism is given by composing its constituents in a remarkably straightforward way . This is sometimes not the case with other approaches , e . g . , those based on Petri Nets and used since [14] , that have a nice graphical notation , but lack a linguistic framework . For a survey on process calculi for modeling biological entities , see also [15] . We specified in the π-calculus all the elements of the molecular machinery of the cell . Each element is specified in isolation , only defining its potential interactions with the environment . Then these pieces are put together in a compositional , holistic fashion . We wrote an interpreter for the enhanced π-calculus in Java , and we used it to run simulations . Simulations play the role of virtual experiments , performed according to given different initial conditions . The input file contains the definitions of all the metabolites inside the cell , the initial inner concentrations of the metabolites , and the rates of enzymatic activities , derived from the available real experimental data . The interpreter stores and displays some information about the virtual experiment , typically the concentrations of all the virtual metabolites ( i . e . , the number of the corresponding processes ) or the usage of the different enzymes ( i . e . , the number of accesses of each channel ) at given instants . With the first output , we determined the time course of the concentration of any virtual metabolite during the simulation; with the second one , we inspected the usage rate of the enzymes specified in the definitions , and , therefore , we tested the presence of unused metabolic pathways .
The MGS-prokaryote has been exhaustively described in the enhanced π-calculus . We represented the 237 genes , their relative products , and the metabolic pathways expressed and regulated by the genes , as the corresponding processes and channels . In particular: the Glycolytic Pathway , the Pentose Phosphate Pathway , the pathways involved in nucleotide , aminoacids , coenzyme , lipids , and glycerol metabolism . Moreover , MGS genes encode for a set of membrane carriers for metabolite uptake , including the PTS carrier . We placed this virtual cell in an optimal virtual environment , in which all nutrients and water were available , and where no problems were present in eliminating waste substances . A large number of simulations ( about 5 , 000 ) have been run , differing in the values of the initial parameters . We independently varied the amount of glucose in the extracellular environment ( the number of copies was in the range 100–5 , 000 ) and the time interval of observation T ( in the range 10–10 , 000 ) . Recall that in our model time steps correspond to the occurrence of transitions , so we set T establishing the length of the computations performed by the simulator . In all the studied cases , the MGS-prokaryote could not reach a steady state; most of the essential metabolites fell to zero in a short period , as is clearly shown in Figures 1 and 2 , which display the typical time course of ATP and 2-Acyl-Glycerol ( 2AG ) . These results lead us to the conclusion that this MGS-based cell was not able to live , at least in silico . Our approach to a functional screening of genomes was shown to be valid . In particular , our results have been obtained very cheaply with respect to a possible wet-lab approach involving de novo synthesis of the examined genome . Clearly , if a hypothetical genomes does not pass the in silico test , it will be unlikely to give rise to a living organism . It is hard to sustain the opposite: we cannot affirm that a hypothetical genome passing the test is able to sustain a living organism , and only a wet-lab approach can validate the proposal . Indeed , in silico experiments can help us to select which proposals are coherent , and thus more promising . As evidence of this , our work shows that the minimal genome we proposed for ViCe is surely more biologically reliable than an MGS . | The origins of life represent a fascinating problem that has been addressed using different approaches and a wide variety of technologies . A theoretical approach consists of inferring a possible oldest ancestor genome from a well-defined comparison of current ones . A crucial problem concerns the validation of the proposed genome . The direct solution of synthesizing such a genome in a laboratory is often extremely difficult , due to the great complexity of a biological cell . In this paper , we present an approach for evaluating the chances a hypothetical organism has to be really viable , relying on computer simulations . Our method is based on a certain formal language , through which we specify a whole metabolic network , and we study its dynamics , in particular for verifying if a living organism has some fundamental properties , e . g . , homeostasis . This approach is not equivalent to a wet-lab one , but it allows for early pruning of most of the inconsistently designed hypothetical organisms , thus saving biologists time and resources . | [
"Abstract",
"Introduction",
"Methods",
"Results/Discussion"
] | [
"none",
"computational",
"biology"
] | 2007 | A Computational Approach to the Functional Screening of Genomes |
Of all the age-related declines , memory loss is one of the most devastating . While conditions that increase longevity have been identified , the effects of these longevity-promoting factors on learning and memory are unknown . Here we show that the C . elegans Insulin/IGF-1 receptor mutant daf-2 improves memory performance early in adulthood and maintains learning ability better with age but , surprisingly , demonstrates no extension in long-term memory with age . By contrast , eat-2 mutants , a model of Dietary Restriction ( DR ) , exhibit impaired long-term memory in young adulthood but maintain this level of memory longer with age . We find that crh-1 , the C . elegans homolog of the CREB transcription factor , is required for long-term associative memory , but not for learning or short-term memory . The expression of crh-1 declines with age and differs in the longevity mutants , and CREB expression and activity correlate with memory performance . Our results suggest that specific longevity treatments have acute and long-term effects on cognitive functions that decline with age through their regulation of rate-limiting genes required for learning and memory .
A guiding proposition of longevity research is that treatments that extend survival will also be generally beneficial to the health of the organism . However , many specifics of this concept remain to be tested . In humans , aging is often accompanied by declines in cognitive function . An understanding of the molecular mechanisms underlying the initiation and progression of age-related neuronal decline requires an experimental system to quickly test early symptoms , rather than the correlative downstream effects , of neuronal decline and disease . Although C . elegans' neural system is relatively simple compared with higher organisms , it has been an important model system for the study of neuronal development , synapse formation and function , and neuron-mediated behaviors . C . elegans has also been invaluable in the study of aging , revealing several longevity-modifying pathways that have proven to be conserved in higher organisms [1]–[6] . C . elegans is particularly useful as a model of post-mitotic cellular aging; because the cells do not turn over , maintenance of neuronal function must be due to cell and protein maintenance with age , as is the case for the majority of human brain cells . With its short lifespan and simple stereotyped nervous system , a C . elegans model characterizing the age-related neuronal decline of neurodegenerative disease may allow the identification of novel therapeutic targets for the earliest-onset cognitive disorders in humans . Electron microscopy studies reveal that while C . elegans muscle tissue degrades with age , neuronal cells maintain their structural integrity [7] . However , this may not indicate a retention of function with age: humans display short-term memory loss that appears to be independent of neuronal degeneration [8] . Functional studies show that Drosophila also experience declines in olfaction and olfactory learning with age [9] . C . elegans displays age-related declines in chemotaxis [10] and isothermal tracking , a type of associative memory recalling the temperature at which an animal was raised . However , these declines significantly overlap with age-related declines in motility and may be related to degradation of muscle function [10] . Age-related decline in habituation ( desensitization to mechanical stimulus ) occurs late in adulthood as well , also overlapping with declines in muscle function [11] . Thus , it remains to be determined whether C . elegans experiences early age-related declines in higher-order neuronal function despite the structurally intact appearance of neurons . Two of the primary regulators of longevity , Insulin/IGF-1 Signaling ( IIS ) and Dietary Restriction ( DR ) , have been well-studied in C . elegans . The DAF-2 insulin receptor ( WBGene00000898 ) and its downstream target , the transcription factor DAF-16/FOXO ( WBGene00000912 ) , regulate survival , stress resistance , and the maintenance of youthful movement in C . elegans [1] , [12] , [13]; its homologs in other organisms , including humans , also regulate aging , suggesting significant conservation of this pathway's functions [4] , [5] , [14] . The C . elegans mutant eat-2 ( WBGene00001133 ) is a model of Dietary Restriction and lives up to 50% longer than wild type [2]; DR increases survival in every organism tested [15] . Low insulin signaling in daf-2 mutants maintains isothermal tracking and chemotaxis abilities with age better than wild type [10] , [16]; conversely , high insulin levels decrease locomotion and spatial memory in mice [17] , suggesting that insulin signaling's effects on cognition may also be conserved . DR has also been suggested to attenuate age-related cognitive decline [18] , but the molecular bases for such effects are not yet known . Here we have designed positive olfactory associative assays to measure C . elegans learning and memory . We have found that C . elegans long-term associative memory ( LTAM ) requires the same molecular components , such as transcription , translation , and CREB activity , as long-term memory in other organisms . Our aging results suggest that long-term olfactory memory is the first function to be lost with age and that olfactory learning , chemotaxis , and motility decline later and sequentially , prior to any obvious structural defects . We then tested these behaviors in the insulin-signaling and DR longevity mutants , both in young and aged worms , and found that these mutations have surprisingly different effects on age-related declines in learning and memory . We find that CREB levels and activity correlate well with long-term memory , suggesting an underlying molecular mechanism determining memory performance . Our results suggest that the regulation of the degeneration or maintenance of these behaviors may be conserved in higher organisms and may also be manipulable through specific longevity treatments .
To examine cognitive decline in C . elegans , we developed simple Pavlovian appetitive associative learning and memory assays using the AWC neuron-sensed odorant butanone ( Figure 1 ) , and tested these behaviors with age in wild-type animals and in longevity mutants . Briefly , after a short starvation , worms are fed in the presence of butanone at a concentration that normally elicits a low chemotactic response ( similar to Toroyama et al . [19]; Figure 1B ) , and then are tested for their attraction to butanone ( Figure 1A ) . We found that after a single ( “massed” ) training , wild-type animals' chemotaxis to butanone increased ∼0 . 6 chemotaxis index units , which is its “Learning Index” ( LI ) . This massed associative learning was saturated by 30 min of training ( Figure 1C ) and was dependent on the simultaneous presence of food and butanone during training ( Figure 1B ) . Memory can be separated into distinct classes based on duration and molecular requirements; in Aplysia , Drosophila , and mice , short-term memory lasts minutes to hours [20] , [21] , while long-term memory lasts hours to days and requires new protein synthesis and gene transcription [22] . To assess the duration of the learned association , we held worms on a plate with food but no butanone after a single training session . We found that the memory of the food-butanone association was retained less than two hours ( Figure 1D ) , which is similar to the duration of C . elegans salt-starvation association [23] . Starvation after massed training only slightly extended this short-term associative memory ( STAM ) ( Figure S1A ) . In flies , mice , and Aplysia , training paradigms in which conditioning stimuli are presented to animals several times with rest periods between presentations ( “spaced training” ) yield longer-lasting memory than does massed training [22] . We found that spaced training also greatly enhanced the duration of C . elegans' memory of the food-butanone association: while the number of training blocks did not affect initial ( “spaced” ) learning ( 0 h , Figure 1E ) , recall increased with the number of training blocks . After seven training blocks , the learning index 16 h post-training was the same as that immediately after conditioning ( Figure 1E ) . ( Although the 16 h time point is arbitrary , it is similar to the time frame used in mammalian long-term memory studies [24] . ) In our spaced-training paradigm , worms are starved in the “rest” period between conditioning training sessions and put onto food after training ( the post-conditioning period ) . Therefore , any decline in LI after training is not due to adaptation since butanone ( the conditioned stimulus ) is not present between training and testing for memory . In terms of classical conditioning , holding worms on food after spaced training may be considered to be re-exposure to the unconditioned stimulus; however , in our assays it is critical to return animals to food , the neutral state , after conditioning , since significant transcriptional changes in response to starvation can occur as soon as 1 h after the removal of food [25] . Moreover , we find that holding naïve worms on plates without food for 16 h greatly increases their attraction to butanone ( Figure S1B ) . Thus , starving the worms during the post-conditioning period would not allow a fair test of how well the association between butanone and food is retained . Previous studies in Drosophila have demonstrated that varying the duration of the rest period during spaced training ( either mechanically or through genetic manipulation ) can affect recall performance [26] , [27] , [28] . We found that doubling or halving the length of time between training intervals appears to have no effect on long-term memory ( Figure S1C ) . Long-term memory in other organisms requires gene transcription and protein synthesis [22] . We found that cycloheximide treatment and cold shock , which interrupt protein synthesis , and actinomycin D treatment , which interrupts transcription , all abrogated 16 h memory but had no effect on the 0 h LI ( Figure 1F ) , indicating that both protein translation and gene transcription are required for long-term memory but not for spaced learning . Thus , our spaced-training memory paradigm greatly enhances the duration of recall compared with the massed-training paradigm , and meets the transcriptional and translational requirements of classical long-term associative memory ( LTAM ) . Several genes that are required for olfactory learning have been identified , including casy-1 ( WBGene00000403 ) , a calsyntenin [29]; glr-1 ( WBGene00001612 ) , an AMPA-type glutamate receptor [30] , [31]; and hen-1 ( WBGene00001841 ) , a secretory protein required for cue integration and olfactory learning [32] . We found that these mutants performed normally in benzaldehyde chemotaxis assays ( Figure S2A ) ; however , these animals , especially casy-1 and glr-1 , exhibited defects in massed learning ( Figure 2A ) and long-term ( 16 h ) memory ( Figure 2B ) . By contrast , we found that CREB , a bZIP transcription factor required for long-term memory in Aplysia , Drosophila , and mammals [22] , is required specifically for LTAM: deletion allele mutants of CREB ( crh-1 , WBGene00000793 ) had normal benzaldehyde chemotaxis ( Figure S2B ) , massed learning ( Figure 3A ) , short-term memory ( Figure 3A ) , and spaced learning ( Figure 3B ) , but were defective for long-term memory ( Figure 3B ) . In fact , crh-1 recall is lost by 2 h post-spaced training ( ) , while wild type shows no decrease at this point , highlighting the requirement for CREB activity in long-term memory . Expression of CREB under a neuronal-specific promoter ( crh-1 ( tz2 ) ;cmk-1::crh-1β [33] ) completely rescued the long-term memory defect of the crh-1 ( tz2 ) deletion mutant ( Figure 3C ) . Moreover , neuronal overexpression of CREB in a wild-type background both increases long-term memory duration ( Figure 3D ) and reduces the number of training sessions to achieve 16 h memory ( Figure 3E ) . Together , our results suggest that learning is molecularly distinct from but required for subsequent memory , and that CREB is specifically required for long-term associative memory . The observation that C . elegans neurons do not display obvious age-dependent structural degeneration [7] leads to the question of whether worms experience functional neuronal decline . Thus , we tested the effect of aging on various neuronally-controlled behaviors . While motility and chemotaxis were unaffected through the first week of adulthood ( as shown previously [10] , [34] ) , we found that massed learning , spaced learning , and long-term memory abilities declined quickly ( Figure 4 ) . Strikingly , 16 h long-term memory was impaired significantly by Day 2–3 and was completely lost by Day 5 . Thus , not only do learning and long-term memory require different gene activities ( Figures 2 , 3 ) , but these behaviors also decline at different rates , suggesting that the molecularly distinct mechanisms of learning and memory are also differently susceptible to aging . daf-2 insulin/IGF-1 receptor mutants are long-lived and morphologically youthful [1] and thus might be predicted to maintain cognitive abilities with age . In our positive appetitive assay , daf-2 mutants displayed no defects in chemotaxis to butanone or to another AWC-sensed odorant , benzaldehyde ( Figure S2B ) , consistent with daf-2 performance in chemotaxis adaptation assays [10] , and no learning defects ( Figure S3A ) . Strikingly , daf-2 ( e1370 ) , daf-2 ( e1368 ) , and daf-2 ( RNAi ) animals ( Figure 5A , B; Figure S3B–E ) displayed greatly increased duration of memory on the first day of adulthood: daf-2's short-term memory lasted more than 3 times as long as wild type's ( Figure 5A , Figure S3B , C ) , and daf-2's long-term memory at 40 h is still more than 60% of its initial learning levels ( Figure 5B , Figure S3D , E ) . daf-2's short- and long-term memory extensions both require the activity of the downstream transcription factor daf-16/FOXO ( Figure 5A , B ) . Are daf-2 worms simply less plastic , acquiring and losing information more slowly than wild-type worms do ? To answer this question , we measured the rate of learning in both the massed and spaced-training paradigms . In the massed training paradigm , daf-2 worms learned at a rate similar to wild type , with maximum learning achieved after 30 min of conditioning ( Figure 5C ) , suggesting that daf-2's basic massed learning ability is similar to wild type's . However , daf-2 worms established LTAM faster than wild type , reaching maximum 16 h memory with only five training blocks ( Figure 5D ) , similar to the performance of CREB overexpression animals ( Figure 3E ) . These results suggest that reduced insulin signaling does not change plasticity but can both establish the long-term memory of an association more quickly and prolong the duration of this association . daf-2's cognitive phenotypes could be specific to IIS or could be general for all longevity pathways . To differentiate these possibilities , we examined the acetylcholine receptor mutant eat-2 , a model of the well-established DR longevity mechanism . eat-2 encodes a nicotinic acetylcholine receptor ( nAChR ) that functions postsynaptically in pharyngeal muscle to regulate the rate of pharyngeal pumping [35] , [36] . eat-2 mutants ingest food ( E . coli ) poorly and extend life span through a daf-16-independent DR pathway [2] , [37] . We found that Day 1 adult eat-2 mutants displayed normal benzaldehyde chemotaxis ( Figure S2B ) and normal massed learning ( Figure 6A , Figure S4A ) , suggesting that its decreased food ingestion does not affect its ability to form food-olfactory associations or to chemotax toward odorants . eat-2's short-term memory duration was the same as wild type's ( Figure 6A , Figure S4A ) , in contrast to daf-2's significant STAM extension ( Figure 5A , Figure S4B , C ) . However , in two point mutation allele mutants , eat-2 animals' long-term memory was significantly impaired , with a complete abrogation of memory by 24 h ( Figure 6B , Figure S4B ) . eat-2's neutral effect on STAM and negative effect on LTAM were unexpected , based on our observations that starvation extends STAM ( Figure S1A ) and that daf-2 mutations extend both STAM and LTAM ( Figure 5A , B , –E ) . Increasing the number of training blocks from seven to ten improves eat-2's LTAM to a level similar to wild type's after 7× spaced training ( Figure 6C ) , suggesting that eat-2 mutants can form long-term memories but require more training to do so . To rule out the acetylcholine receptor itself as the source of eat-2's memory impairment , we fed eat-2 mutants smaller , “easier to eat” bacteria , Comamonas sp . [38] ( Leon Avery , personal communication ) . Comamonas had no effect on the growth or longevity of wild-type worms but suppressed eat-2's small size and long life span ( Figure 7A–C , Figure S4C ) ; thus , these worms still had the mutant acetylcholine receptor but were not dietarily restricted . Strikingly , Comamonas also suppressed eat-2's long-term memory defect ( Figure 7D ) . All of eat-2's phenotypes were also suppressed by treatment with RNAi of pha-4 , the FoxA transcription factor that mediates eat-2's effects on longevity ( Figure 7E ) [2] , [37] . Together , these results suggest that the memory impairment we observe in eat-2 mutants is indeed due to DR rather than to acetylcholine receptor dysfunction . Thus , while Dietary Restriction and reduced insulin signaling both increase longevity , the two pathways influence cognitive ability of young adults in an opposite manner . To test the roles of IIS and DR in the maintenance of cognitive ability with age , we measured daf-2 and eat-2 mutants' learning and memory abilities later in adulthood . We found that daf-2 mutants retain the ability to learn longer than do wild-type or daf-16 worms , with no significant loss in massed learning ability at Day 5 , when wild-type massed learning ability has completely ceased ( Figure 8A , Figure S5A ) . Surprisingly , however , daf-2 mutants lose long-term memory with age at the same rate as wild type: on Day 4 , daf-2 mutants had better spaced learning than wild-type worms but exhibited no significant 16 h memory ( Figure 8B ) . Thus , despite extending longevity and learning ability with age , reduced insulin signaling does not appear to maintain memory performance with age . Like daf-2 , eat-2's learning ability is maintained with age: on Day 4 , eat-2 mutants learned better than Day 4 wild-type worms after spaced training ( Figure 8C ) . However , in contrast to daf-2 mutants , eat-2 mutants maintain both short- and long-term memory with age , as Day 4 eat-2 animals exhibited no significant decline from their performance on Day 1 ( Figure 8C , Figure S5B , C ) . This maintenance of long-term memory can be attributed to DR , as Comamonas feeding suppressed both the aged learning and memory phenotypes of eat-2 mutants ( Figure 8D ) . To determine whether DR strictly in adulthood can rescue age-related memory phenotypes , we raised eat-2 animals on Comamonas until early adulthood , then switched them to E . coli to induce DR . When switched post-developmentally , the animals were still large and exhibited normal ( wild-type-like ) Day 1 memory ( Figure S5D–F ) but retained Day 4 memory better than wild-type ( Figure 8E , Figure S5G ) , suggesting that memory loss was alleviated by DR . Thus , while DR and reduced IIS both increase longevity , the two pathways have very different effects on cognitive behaviors , both early in adulthood and with age . To identify the underlying molecular mechanisms that might distinguish IIS and DR effects on long-term memory maintenance with age , we examined the transcriptional levels of key learning and memory genes . While the expression of the learning genes glr-1 and casy-1 did not change significantly with age ( Figure S6A ) , we found that CREB/crh-1 expression levels correlate with memory performance: crh-1 levels are higher in young daf-2 than in wild-type ( Figure 9A ) or daf-16;daf-2 animals , lower in young eat-2 than in wild-type ( Figure 9A ) , fall with age in both wild-type and daf-2 worms ( Figure 9B ) , and are maintained with age in eat-2 mutants ( Figure 9B , Figure S6B ) . To determine whether changes in CREB transcriptional levels reflect changes in activity , we used an anti-phosphorylated CREB antibody to assay levels of activated CREB . First , we found that naïve crh-1 overexpression worms have a higher level of phosphorylated CREB ( P-CREB ) than did wild-type animals ( Figure 9C , Figure S6C ) . Secondly , both wild-type and CREB overexpression worms showed increased P-CREB levels post-training ( Figure 9D , Figure S6C ) ; these levels increased with training sessions ( Figure 9E , Figure S6D ) , parallel to LTAM activity ( Figure 1E ) . Increases in CREB activity with training sessions also parallels the LTAM performances of daf-2 and eat-2 mutant worms ( Figures 5 , 6 ) : while daf-2 P-CREB levels increased fairly linearly through six training trials ( Figure S6F ) , P-CREB levels increased only with additional training sessions in eat-2 animals ( Figure S6G ) . To determine how well crh-1 expression levels and LTAM activity correlate , we plotted the ratios of LTAM activities against the ratios of crh-1 levels for eight pairs of samples ( Figure 9F ) ; the R2 value of 0 . 8 indicates that crh-1 ratios are a reasonable predictor of relative LTAM activity . An even stronger correlation was found when we plotted the ratios of LTAM activities against ratios of P-CREB levels ( R2 = 0 . 9; Figure 9G ) . Thus , CREB activity appears to be the major molecular mechanism determining LTAM performance , and crh-1 levels may be an excellent predictor of long-term memory performance .
While it is known that many neuronal structures remain intact with age [7] , [8] , previously it was not clear how higher-order neuronal functions are affected by aging . While basic motor skills and chemotaxis abilities continue through later stages of adulthood , we find that higher-level cognitive abilities are lost much earlier in adulthood . Our assays are able to distinguish between the processes of massed learning , spaced learning , short-term memory , and long-term memory , and our results suggest that not only do these processes have distinct molecular requirements , but they also decline differentially with age . We find that long-term memory , which inherently requires learning and chemotaxis abilities , declines particularly early in adulthood , prior to the decline of learning , chemotaxis , and motility . LTAM likely involves complex synaptic machinery [20] and thus may be particularly susceptible to age-related damage . Associative olfactory learning and memory appear to be more sensitive to age-related decline than other behaviors , such as isothermal tracking [10] or habituation [11] . Interestingly , anosmia is recognized as one of the earliest symptoms of neurodegeneration , including Alzheimer's and Parkinson's disease [39] , and declines in taste are linked to olfactory defects [40] . Therefore , food-smell associations may be extremely effective in evaluating changes in learning and memory with age and neurodegeneration in humans as well as in worms . We have also demonstrated for the first time , to our knowledge , the requirement of CREB activity in C . elegans memory . The differential effects of the IIS and DR pathways on learning and memory decline with age appear to be attributable to their differential regulation of CREB/crh-1 expression levels and activity , rather than to changes in learning-associated genes such as glr-1 , casy-1 , and hen-1 . In general , we find that CREB expression level changes can largely account for the decline in wild-type memory with age and its maintenance in longevity mutants ( Figure 9 ) , suggesting that CREB levels may be a good indicator of long-term memory function . CREB levels and activity also decline with age in the mammalian brain [41] , [42] , and over-expression of CREB in the hippocampus increases the performance of aged animals in several long-term memory tasks [43] . Thus , the molecular mechanisms underlying C . elegans long-term memory , particularly CREB's importance , are likely conserved in higher organisms . Our results imply that specific types of longevity treatments could have either positive or negative effects on learning and memory , and therefore , it will be crucial to examine the effects of specific longevity treatments on maintenance of human cognitive behaviors with age .
Animals were cultivated at 20°C on HGM plates on OP50 E . coli or Comamonas sp . ( DA1877 ) using standard methods [44] and developmentally synchronized by hypochlorite treatment . Worms were moved to HGM + 50 mM FUdR at the L4 stage when tested for learning or memory after Day 1 of adulthood . Wild type: ( N2 Bristol ) ; mutant strains: RB888 ( casy-1 ( ok739 ) ) , KP4 ( glr-1 ( n2461 ) ) , JC2154 ( hen-1 ( tm501 ) ) , DA465 ( eat-2 ( ad465 ) ) , DA1116 ( eat-2 ( ad1116 ) ) , CF1041 ( daf-2 ( e1370 ) ) , CF1038 ( daf-16 ( mu86 ) ) ; CF1043 ( daf-16 ( mu86 ) ;daf-2 ( e1370 ) ) ; MT9973 ( crh-1 ( n3315 ) ) ; YT17 ( crh-1 ( tz2 ) ) ; and YT50 ( crh-1 ( tz2 ) ;cmk-1::crh-1β ) . The tz2 mutation lacks 979 nucleotides/38 residues at the C-terminus of CREB's bZIP region , and no functional protein is expressed [33] . Alkema and Horvitz report that n3315 is a loss of function deletion allele that eliminates the expression of all four crh-1 isoforms ( Wormbase ) . The “crh-1 OE” strain ( CQ71 ) was made by crossing N2 with YT50 animals and selecting worms carrying the cmk-1::crh-1β transgene . Chemotaxis assays were performed according to previously described methods [45] . >200 developmentally-synchronized worms were placed at origin , and the number at butanone ( 1 µL 1:10 butanone:ethanol + 1 µL NaN3 ) , ethanol control ( + 1 µL NaN3 ) , and origin were counted after 1 h . Chemotaxis Index ( CI ) = [ ( nattractant ) − ( ncontrol ) ] / [ ( Total − norigin ) ] . Mobility was measured on each day of adulthood by calculating the percentage of worms that remained at the origin of a chemotaxis assay plate after 1 h . Synchronized Day 1 adult hermaphrodites were starved in M9 buffer for 1 h , transferred to a 60 mm NGM plate with 500 µL freshly-seeded OP50 or DA1877 and 2 µL of 10% butanone on lid , trained for 1 h , then tested for chemotaxis to butanone . LI = CITrained − CINaive . After 1× massed training , worms were transferred to 60 mm NGM plates freshly seeded with 500 µL OP50 or DA1877 ( “holding plate” ) for specified intervals . After 1 h of starvation , worms received seven training blocks ( 30 min on training plates with food and butanone , followed by two M9 washes and 30 min on plates without food ) . Worms were then tested immediately for spaced learning ( “0 h” ) or transferred to holding plates for 16 or 24 h . SEM and student's t test was used to assign p values in all assays . Protein synthesis inhibition: animals were cold shocked at −20°C [30] for 15 min , then were returned to the conditioning temperature ( 20°C ) for 15 min , or treated with 800 µg/mL cycloheximide [46] , during the starvation period of each training block . Transcription inhibition: animals were treated with 100 µg/mL Actinomycin D during the starvation period of each training block [47] . Wild-type or eat-2 ( ad465 ) worms were cultivated and life span assays were carried out at 20°C on NGM + 50 µM FUdR with OP50 ( E . coli ) or DA1877 ( Comamonas sp . ) . The first day of adulthood was defined as t = 0 . n>70 for each strain . Standard Kaplan-Meier survival analysis was used to assess significance ( [48] , GraphPad , Prism 5 . 01 ) . RNAi clones were PCR-verified . RNAi-sensitive eri-1 ( mg366 ) ;lin-15B ( n744 ) ( daf-2 RNAi ) or eat-2 ( ad465 ) ( pha-4 RNAi-treated ) animals were synchronized and cultivated on vector control or RNAi bacteria on NGM plates with 0 . 1 M IPTG ( final concentration ) at 20°C until Day 1 of adulthood . Data for gene expression analyses with age in wild-type , daf-2 , and daf-16;daf-2 conditions was provided by Murphy et al . [49] . eat-2 ( ad465 ) and daf-16 ( mu86 ) mutants were collected and analyzed as previously described [49] , [50] . Data were filtered for quality , and replicates were collapsed to an average value ( PUMAdb; http://puma . princeton . edu ) . RT-PCR was carried out to verify expression results ( Figure 9A , Figure S6B ) . cDNA was made from total worm RNA ( checked for 230/260/280 quality before processing ) using TaqMan Reverse Transcription Reagents ( Applied Biosystems ) . Serial dilutions of 0 . 5 µg/mL cDNA were used in 20 µL PCR reactions . For crh-1 RT-PCR experiments , the primers used ( forward: ATGTCAGCGAAAGGTAACGG , reverse: CGTTTTGTTGTGGTCCTCCT ) amplify a 442 bp fragment located at 30–471 bp in the 1 , 197 bp mRNA sequence ( NCBI reference sequence NM_001027690 . 1 ) , a region that lies upstream of the deletion described for crh-1 ( tz2 ) . Worms were washed in M9 , collected , and frozen in liquid nitrogen . Lysates were prepared by freeze/thawing worm pellets in lysate buffer ( 50 mM HEPES , 1 mM EDTA , 150 mM NaCl , 1 mM NaFl , 10% glycerol , 1% Triton X-100 , proteinase inhibitor ) , and sonication to break cells . Protein concentrations were quantified using Coomassie Plus ( Pierce ) . Anti-Phospho-CREB ( Ser133 ) rabbit mAb ( Cell Signaling Technology 87G3 , #9198 ) was used to probe for P-CREB; Anti-α-Tubulin mouse mAb ( Sigma-Aldrich , #T9026 ) was used as a probe for the loading control , as we were unable to find a working antibody for total CREB in C . elegans . Antibodies were diluted 1:1 , 000 in 1× TBS-T , 5% BSA . Quantification of Western blot results was performed using “GeneTools” software from SynGene; P-CREB levels were compared to the α-tubulin loading control . | In humans , aging is often associated with a decline in cognitive function . Progress toward an understanding of the molecular mechanisms underlying the initiation and progression of age-related neuronal decline could be hastened by the development of experimental systems that quickly test early and true symptoms ( rather than the correlative downstream effects ) of neuronal decline and disease . In contrast to muscle degradation , the nervous system of C . elegans is structurally remarkably well-preserved , leaving open the question of how to define age-related changes in neuronal function . To address this problem , we have established a novel system to study associative learning , short-term associative memory , and long-term associative memory in C . elegans . Through chemotaxis assays , we measured worms' ability to learn a positive association of a neutral chemoattractant with food . We found that long-term , but not short-term , associative memory is dependent on crh-1 , the C . elegans homolog of the transcription factor CREB . Furthermore , we find that worm learning and long-term associative memory decreases with age and is influenced differently by insulin/IGF-1 and Dietary Restriction longevity pathways . These effects can be largely attributed to changes in expression of crh-1 , which correlate with memory performance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/disease",
"models",
"developmental",
"biology/aging",
"neurological",
"disorders/cognitive",
"neurology",
"and",
"dementia"
] | 2010 | Insulin Signaling and Dietary Restriction Differentially Influence the Decline of Learning and Memory with Age |
TPL-2 ( COT , MAP3K8 ) kinase activates the MEK1/2-ERK1/2 MAPK signaling pathway in innate immune responses following TLR , TNFR1 and IL-1R stimulation . TPL-2 contributes to type-1/Th17-mediated autoimmunity and control of intracellular pathogens . We recently demonstrated TPL-2 reduces severe airway allergy to house dust mite by negatively regulating type-2 responses . In the present study , we found that TPL-2 deficiency resulted in resistance to Heligmosomoides polygyrus infection , with accelerated worm expulsion , reduced fecal egg burden and reduced worm fitness . Using co-housing experiments , we found resistance to infection in TPL-2 deficient mice ( Map3k8–/– ) was independent of microbiota alterations in H . polygyrus infected WT and Map3k8–/–mice . Additionally , our data demonstrated immunity to H . polygyrus infection in TPL-2 deficient mice was not due to dysregulated type-2 immune responses . Genome-wide analysis of intestinal tissue from infected TPL-2-deficient mice identified elevated expression of genes involved in chemotaxis and homing of leukocytes and cells , including Ccl24 and alternatively activated genes . Indeed , Map3k8–/–mice had a significant influx of eosinophils , neutrophils , monocytes and Il4GFP+ T cells . Conditional knockout experiments demonstrated that specific deletion of TPL-2 in CD11c+ cells , but not Villin+ epithelial cells , LysM+ myeloid cells or CD4+ T cells , led to accelerated resistance to H . polygyrus . In line with a central role of CD11c+ cells , CD11c+ CD11b+ cells isolated from TPL-2-deficient mice had elevated Ccl24 . Finally , Ccl24 neutralization in TPL-2 deficient mice significantly decreased the expression of Arg1 , Retnla , Chil3 and Ear11 correlating with a loss of resistance to H . polygyrus . These observations suggest that TPL-2-regulated Ccl24 in CD11c+CD11b+ cells prevents accelerated type-2 mediated immunity to H . polygyrus . Collectively , this study identifies a previously unappreciated role for TPL-2 controlling immune responses to H . polygyrus infection by restricting Ccl24 production .
Heligmosomoides polygyrus is a natural murine intestinal helminth , used to model chronic human helminth infections . Resistance to H . polygyrus is mediated by genetic strain specific responses [1] , as well as protective immune mechanisms attributed to the strength of the type-2 immune response [2] . These include the activation of alternatively activated macrophages leading to the killing of tissue dwelling larvae [3] , production of IgG1 antibodies that limit parasite fecundity and protect against reinfection [4 , 5] , and production of the anti-parasitic protein RELM-β by intestinal epithelial cells [6] . Despite these mechanistic observations of resistance to infection by H . polygyrus , there remains a considerable gap in our understanding of the molecular pathways regulating natural or vaccine mediated immunity to H . polygyrus . Tumor Progression Locus 2 ( TPL-2; also known as MAP3K8 and COT ) , activates the ERK1/2 MAP kinase pathway in response to stimulation of Toll-like receptors ( TLRs ) , TNF receptor ( TNFR ) and IL-1 ( IL-1R ) [7] . TPL-2 is an important kinase for pro-inflammatory type-1 and Th17 responses [8 , 9] . TPL-2/ERK1/2 signaling is required for cytokine and chemokine production , with TPL-2-deficient ( Map3k8–/– ) mice being protected in models of endotoxic shock , experimental autoimmune encephalomyelitis ( EAE ) [9] , pancreatitis , liver fibrosis and thrombocytopenia [10–15] . Additionally , TPL-2 is required for protective immunity to intracellular bacterial and protozoan infection , with substantially greater susceptibility in Map3k8–/–mice compared to wild type [8 , 16] . In contrast to these pro-inflammatory roles for TPL-2 in type-1/Th17 responses , we recently identified a regulatory role for TPL-2 in type-2 immune settings , including house dust mite ( HDM ) mediated airway allergy . Map3k8–/–mice demonstrated significantly increased production of IL-4 , IL-5 and IL-13 , increased airway eosinophilia , severe lung inflammation and increased serum IgE compared to WT mice [17] . TPL-2 also regulated alternative activation of macrophages , with Map3k8–/–mice developing significantly greater fibrotic granulomas with enhanced type-2 immune responses following Schistosoma mansoni infection [18] . While , increased type-2 responses contributed to increased immunopathology following HDM allergen challenge or S . mansoni infection , in this study we tested the hypothesis that TPL-2 regulated type-2 immune responses contributed to susceptibility to intestinal helminth infection . In the present study , we demonstrate that Map3k8–/–mice were resistant to a primary H . polygyrus infection , with significantly fewer worm and fecal egg burdens compared to wild type ( WT ) infected mice . Resistance to H . polygyrus in Map3k8–/–mice was not due to differences in type-2 immune responses with comparable adaptive and cytokine responses . Given the importance of intestinal microbiota in regulating local and systemic immune responses [19] and the fact that TPL-2 is downstream of TLR-4 [10] , we investigated the impact on intestinal microbial communities following H . polygyrus infection of WT and Map3k8–/–mice . Although intestinal microbiota composition was altered following infection and that distinct differences were observed in Map3k8–/–mice , co-housing experiments that corrected for these between-genotype variations ruled out any role for microbiota alterations in mediating protective immunity to H . polygyrus in Map3k8–/–mice . Transcriptional analysis of intestinal tissue from infected mice revealed increased expression of Ccl24 alongside a significant increase in the expression of type-2 memory signature genes associated with alternatively activated cells , including Arg1 , Chil3 and Ear11 in Map3k8–/–mice , compared to WT mice . Using lineage specific conditional knockout ( KO ) mice , we identified a role for TPL-2 in CD11c+ cells and not Villin+ epithelial cells , LysM+ myeloid cells or CD4+ T cells in controlling susceptibility to H . polygyrus . CD11c+ CD11b+ cells from H . polygyrus infected Map3k8–/–mice expressed increased Ccl24 compared to WT mice . Increased Ccl24 expression correlated with an increase in the frequency of eosinophils , neutrophils , monocytes and Il4GFP+ Th2 cells in Map3k8–/–mice , compared to WT mice . Furthermore , blocking Ccl24 in vivo resulted in a significant decrease in the expression of type-2 memory markers , Arg1 , Retnla , Chil3 and Ear11 and led to loss of resistance to H . polygyrus in Map3k8–/–mice . These data suggest that TPL-2-regulated Ccl24 is an important pathway contributing to susceptibility to H . polygyrus . These results demonstrate a previously unidentified role for TPL-2 in restricting protective Ccl24-dependent pathways during intestinal H . polygyrus infection .
To test whether TPL-2 contributed to immunity to H . polygyrus , we infected WT and Map3k8–/–mice with 200 H . polygyrus L3 stage larvae . Adult luminal worms and fecal eggs were evaluated on day 14 ( D14 ) and D28 post infection . Map3k8–/–mice harbored significantly fewer worms at D14 compared to WT mice , with a further significant reduction in worm burden by D28 ( Fig 1A ) . Parasite egg burden in Map3k8–/–mice was also significantly reduced compared to WT mice at D14 and D28 ( Fig 1B ) . To test whether the deficiency of TPL-2 in the host directly affected the fitness of the remaining worms , we measured the level of ATP [6] in the remaining adult worms harvested from the lumen of D14 H . polygyrus infected WT and Map3k8–/–mice . The worms harvested from the lumen of Map3k8–/–mice had significantly lower ATP levels compared to WT mice ( Fig 1C ) , suggesting that the absence of TPL-2 led to lower worm burden likely by compromising the fitness of H . polygyrus . It has previously been suggested that T cell intrinsic TPL-2 regulates Th2 differentiation [20] and regulatory T cell development [21] . IL-4+ Th2 cells are important for generating a protective type-2 response in H . polygyrus infection [22 , 23] and Foxp3-expressing regulatory T cells ( Tregs ) are important for regulating Th2 immune responses during H . polygyrus infection [24] . Therefore , to determine if Th2 and Treg frequencies and numbers were affected in WT and Map3k8–/–mice following H . polygyrus infection , we crossed Map3k8–/–mice onto an Il4GFPFoxp3RFP dual-reporter background to simultaneously monitor Th2 and Treg cells during H . polygyrus infection . Analysis of Th2 cells in the spleen , mesenteric lymph node ( mLN ) or Peyer’s patches ( PP ) revealed there was no significant difference in the frequency of Il4GFP+ CD4+ Th2 cells between infected WT and Map3k8–/–mice at D14 ( Fig 1D ) . Similarly , we found no significant difference in the frequency of Foxp3RFP+ Treg cells between the mice at D14 post infection ( Fig 1D ) . Of note , there was no significant difference in the total cell numbers among the different tissues between WT and Map3k8–/–mice ( Fig 1E ) . Anti-parasitic immunity has been correlated with serum IgG1 [5] . We therefore measured H . polygyus antigen extract ( HEX ) -specific IgG1 in the serum of infected mice . Both WT and Map3k8–/–mice mounted significant , but indistinguishable amounts of HEX-specific IgG1 ( Fig 1F ) . Similarly , there were no significant differences in the production of HEX specific type-2 cytokines , IL-5 and IL-13 in mLN cell culture supernatants from infected WT and Map3k8–/–mice ( Fig 1G ) . These data indicated that TPL-2 deficiency promoted immunity to H . polygyrus , but this is not associated with any change in T cell-associated responses . A close relationship between helminth infections and intestinal microbiota has previously been reported [25] , with a recent study demonstrating a direct correlation between abundance of Lactobacillus species and susceptibility to infection by H . polygyrus [26] . To evaluate whether changes in intestinal microbiota contribute to resistance in Map3k8–/–mice , we setup a co-housing experiment between WT and Map3k8–/–mice as depicted in the schematic ( Fig 2A ) . WT and Map3k8–/–mice were co-housed in two groups . Group 1 allowed for genotype-specific influences on microbiota to emerge , while group 2 aimed to normalize any impact on microbiota . Briefly , group 1 mice were co-housed together for 14 days , normalizing the microbiota between all mice and then separated into their individual genotypes for 14 days for genotype associated differences in microbiota to emerge . For the remainder of the experiment , group 1 mice were left in their genotype-associated separate housing . Group 2 mice were kept separated for 14 days before being co-housed together for 14 days . For the remainder of the experiment , mice in group 2 were co-housed . Fecal pellets were collected on day 14 and day 28 ( as depicted by “S” in Fig 2B ) . Both groups of mice were subsequently infected with H . polygyrus on D28 and fecal samples were harvested at 1 ( day 35 ) , 2 ( day 42 ) and 4 ( day 56 ) weeks-post infection , as indicated . Analysis of the fecal microbiota composition by 16S ribosomal RNA sequencing revealed no significant differences in the fecal microbiota of the WT and Map3k8–/–mice at day 14 of the experimental setup ( Fig 2B ) . However , a genotype-associated influence on Bacilli spp . emerged in group 1 at day 28 , with a significant proportional increase in Map3k8–/–mice compared to WT mice , which was further increased following 1 week ( day 35 ) of H . polygyrus infection . Conversely , Alphaproteobacteria spp . , Clostridia spp . and Erysipelotrichia spp . were proportionally increased in WT mice compared to Map3k8–/–mice following 1-week of H . polygyrus infection ( Fig 2B ) . This time point corresponds with larval migration through the mucosal surface of the small intestine . These differences correlated with an increase in overall microbial diversity in WT , but not Map3k8–/– , mice ( S2 Fig ) . At 2-weeks ( day 42 ) post-infection , there was a secondary shift in the microbial composition , with a significant proportional increase in Betaproteabacteria spp . and a relative reduction in Clostridia spp . in Map3k8–/–mice ( Fig 2B ) . Correlating with the changes in microbial composition described in group 1 , Map3k8–/–mice expelled significantly more worms ( Fig 2C ) with significantly lower eggs detected in the feces ( Fig 2D ) , similar to previous observations ( Fig 1 ) . In group 2 , co-housing mice normalized all of the above-mentioned differences between the genotypes following infection , indicating that the co-housing setup had successfully corrected for genotype-associated influence on microbial communities . Normalization of microbiota between WT and Map3k8–/–mice had no impact on the observed resistance in Map3k8–/–mice , with fewer worms ( Fig 2C ) and fewer eggs detected in fecal pellets of Map3k8–/–mice ( Fig 2D ) . These data indicated that there were changes in microbial composition following H . polygyrus infection and that these changes were regulated , in part , by TPL-2 . However , changes in microbial composition in Map3k8–/–mice did not contribute to the resistance phenotype observed . Following the observation that fewer adult worms were present at D14 ( Fig 1A ) , we assessed the early type-2 immune response at D5 and D7-post infection to assess early Th2 priming events . No differences in cellularity , or the frequency of CD4+ Il4GFP+ Th2 cells or Foxp3RFP+ CD4+ cells were observed in the spleen , mLN or PP between infected WT and Map3k8–/–mice at D5 or D7 ( S1A–S1F Fig ) . Similarly , HEX-specific IL-5 and IL-13 production from mLN cell culture supernatants of infected WT and Map3k8–/–mice were similar at D7 ( S1G Fig ) . These observations suggested that resistance to infection in Map3k8–/–mice was not due to dysregulated Th2 cell priming . Following infection , infective larvae penetrate the mucosal layer and embed in the external muscularis layer of the duodenum [1] . To determine whether a similar number of larvae were infecting and penetrating through the mucosal layer , we evaluated number of H . polygyrus larvae in the duodenal wall at D5 post infection . Significantly fewer worms were present in the intestinal wall of Map3k8–/–mice , compared to WT mice ( S1H Fig ) . These results indicated an early event in the tissue might impact the emergence and fitness of H . polygyrus leading to increased resistance in Map3k8–/–mice . To identify gene expression differences that might contribute to the reduced larval establishment in Map3k8–/–mice , we isolated RNA from duodenal tissue at D5-post H . polygyrus infection of WT or Map3k8–/–mice . Transcripts in infected WT and Map3k8–/–mice were compared to their respective naïve controls , generating a fold-change from naïve value ( Fig 3A , y-axis ) . These fold change values were then compared to each other ( Fig 3A , x-axis ) creating a ratio of ratios plot to identify genes that were significantly different in Map3k8–/–mice at D5-post infection ( x-axis ) ( Fig 3A ) . This analysis identified several genes that were significantly upregulated in infected Map3k8–/–mice relative to infected WT mice , including Ccl24 , Ear11 , Gsdmc2/Gsdmc4 , Riok3 , Chil3 , Gsdmc2 , Pla2g4c , Arg1 , Kif5b and Cfi ( Fig 3A and Table 1 ) . Pathway analysis identified increased chemotaxis of leukocytes , chemotaxis , homing of leukocytes and homing of cells in Map3k8–/–mice , which were completely absent in infected WT mice ( Fig 3B ) . These observations suggested that an early upregulation of genes/pathways involved in chemotaxis and type-2 memory responses might be responsible for the enhanced resistance observed in Map3k8–/–mice . Next , to determine which cellular compartment was responsible for mediating resistance to H . polygyrus in Map3k8–/–mice , we restricted TPL-2 deficiency to cells primarily responsible for expressing the genes upregulated in D5 H . polygyrus infected Map3k8–/–mice ( Table 1 ) . These included CD11c+ cells , which have been reported to express Ccl24 [27 , 28] , LysM+ myeloid cells/macrophages , which express Ccl24 , Ear11 , Arg1 and Chil3 [27–29] and Villin+ epithelial cells , which have been shown to express Gsdmc2/ Gsdmc4 [6 , 30] . To determine the role of TPL-2 in these cells in vivo , we assessed the adult worm burden in Cd11cCreMap3k8fl/fl , VillinCreMap3k8fl/fl and LysMCreMap3k8fl/ko conditional knockout mice . We also evaluated the adult worm burdens in Cd4CreMap3k8fl/ko mice to formally test whether TPL-2 had a T cell-intrinsic role in mediating resistance to H . polygyrus . Significantly fewer worms were observed in Cd11cCreMap3k8fl/fl mice , comparable to Map3k8–/–mice , but not in VillinCreMap3k8fl/fl ( Fig 4A ) , LysMCreMap3k8fl/ko or Cd4CreMap3k8fl/ko mice ( Fig 4B and 4C ) . This suggested that a CD11c+ cell-intrinsic function of TPL-2 regulated resistance to H . polygyrus infection . Ccl24 was expressed in both H . polygyrus infected WT and Map3k8–/–mice , but was upregulated ~5 . 0 fold more in Map3k8–/–mice compared to WT mice ( Table 1 ) . Ccl24 is a chemotactic factor for eosinophils , granulocytes and lymphocytes [31 , 32] . Following the observation that CD11c+ cells deficient in Map3k8 mediate resistance to H . polygyrus infection ( Fig 4 ) and the recent observation that elevated Ccl24 from CD11c+ MHC-II+ cells in Map3k8–/–mice contributed to enhanced HDM-driven airway allergy [17] , we hypothesized that elevated Ccl24 in Map3k8–/–mice could be responsible for accelerated resistance to H . polygyrus . We therefore determined the expression of Ccl24 in CD11c+ cells from the lamina propria of the duodenum and jejenum of H . polygyrus infected WT and Map3k8–/–mice ( Fig 5A ) . CD11c+ CD11b+ cells isolated from D5 H . polygyrus infected Map3k8–/–mice expressed significantly increased Ccl24 compared to similar cells from infected WT mice ( Fig 5B ) . In addition , CD11c+ CD11b+ cells from H . polygyrus infected Map3k8–/–mice also expressed significantly higher levels of Ear11 ( Fig 5B ) , Retnla ( S3A Fig ) , and had a trend for an increase in the expression of Arg1 and Chil3 ( S3B and S3C Fig ) compared to WT cells . To determine whether the increased expression of Ccl24 in CD11c+ CD11b+ cells was associated with enhanced cellular influx of CCR3 expressing cells , including eosinophils , granulocytes and lymphocytes [31–33] , we evaluated the myeloid and lymphoid cell infiltration in the small intestinal lamina propria ( LP ) of D5 H . polygyrus infected Map3k8–/–and WT mice . Concurrent with increased Ccl24 , we observed a significant increase in the frequency of eosinophils ( Fig 5C ) , neutrophils ( Fig 5D ) and Ly6C+ monocytes ( Fig 5E ) in D5 H . polygyrus infected Map3k8–/–mice compared to WT mice . This increase in frequency was accompanied with a trend for an increase in the number of these cells in Map3k8–/–mice compared to WT mice ( Fig 5C–5E ) . We also observed a significant increase in the frequency of Il4GFP+ Th2 cells in the LP of D5 H . polygyrus infected Map3k8–/–mice compared to WT mice ( Fig 5F ) . There was no change in the total number of Il4GFP+ Th2 cells in H . polygyrus infected Map3k8–/–mice compared to WT mice ( Fig 5F ) most likely due to a decrease in lymphocytes ( CD4+ T cells , CD8+ T cells and trend for a decrease in CD19+ B cells ) in Map3k8–/–mice compared to WT mice ( S3D–S3F Fig ) . There was a trend for a decreased macrophages ( CD11b+ F4-80+ ) in Map3k8–/–mice compared to WT mice ( S3G Fig ) , however this failed to reach statistical significance . Similarly , here was a trend for a reduced frequency of KLRG1+ Sca1+ group 2 innate lymphoid cells ( ILC2 ) ( S3H Fig ) , which also failed to reach statistical significance . The total numbers of ILC2s were unchanged . Finally , to determine whether elevated Ccl24 in Map3k8–/–mice contributed to enhanced type-2 memory responses and resistance to H . polygyrus , we neutralized Ccl24 using anti-Ccl24 antibodies in H . polygyrus infected WT and Map3k8–/–mice . In accordance with previous observations ( Fig 3 , Table 1 ) Map3k8–/–mice had significantly increased expression of signature genes associated with alternatively activated macrophages [29] and helminth associated type-2 memory responses [3 , 34] , including Arg1 ( Fig 6A ) , Retnla ( Fig 6B ) , Chil3 ( Fig 6C ) and Ear11 ( Fig 6D ) . Neutralization of Ccl24 led to a significant decrease in the expression of most of these type-2 memory markers in Map3k8–/–mice , compared to Map3k8–/–mice treated with control antibodies ( Fig 6A–6D ) . Ccl24 blockade had no observable effect in WT mice . Similarly , neutralizing Ccl24 had no significant effect on the luminal worm burden in infected WT mice compared to isotype treated WT control mice ( Fig 6E ) . However , Ccl24 blockade significantly reversed the resistance phenotype in infected Map3k8–/–mice , compared to isotype treated Map3k8–/–mice ( Fig 6E ) , correlating with the loss of type-2 memory markers ( Fig 6A–6D ) . These observations indicated that immunity to H . polygyrus in Map3k8–/–mice were determined , in part , by increased Ccl24 , correlating with accelerated expression of type-2 memory responses .
Soil transmitted helminth infections remain a significant global burden , especially in tropical and subtropical countries . Anthelmintic drugs remain the only treatment option to control helminthiasis . At present the immune mechanisms regulating natural resistance to infection , and thus a putative vaccine-driven immune pathway capable of eliciting protective immunity , are still unclear . In the present study we demonstrated that TPL-2 deficiency accelerates resistance to H . polygyrus infection . Using transcriptional analysis , we identified that Ccl24 was upregulated almost 5-fold more in infected Map3k8–/–mice , compared to WT mice . We also observed a significant increase in genes associated with alternatively activated macrophages and helminth associated type-2 memory responses , including Arg1 ( 3-fold ) , Chil3 ( 3 . 2-fold ) and Ear11 ( 4 . 5-fold ) , in infected Map3k8–/–mice , compared to WT mice . Similar to our previous observations [17] , we observed that TPL-2 functions in a cell intrinsic manner in CD11c+ cells and that CD11c+ CD11b+ cells from D5 H . polygyrus infected mice expressed elevated Ccl24 . Consistent with the hypothesis that Ccl24 is an important chemotactic mediator for eosinophils , granulocytes and lymphocytes , we observed a significant increase in eosinophils , neutrophils , monocytes and Il4GFP expressing CD4+ T cells in the LP of D5 H . polygyrus infected Map3k8–/–mice compared to WT mice . Finally neutralizing Ccl24 led to a significant reduction in the expression of Arg1 , Retnla , Chil3 and Ear11 , correlating with a loss of resistance in H . polygyrus infected Map3k8–/–mice . In summary , these observations identify that expression of TPL-2 in CD11c+ cells and Ccl24-dependent pathways can accelerate expulsion of H . polygyrus . Upon infection , H . polygyrus larvae breach the epithelium and mucosa of the small intestine , exposing the host to pathogen-associated molecular patterns ( PAMPs ) , and eliciting anti-microbial responses . As a consequence , changes in microbiota composition have been observed following H . polygyrus infection [26 , 35 , 36] . Furthermore , helminth-associated intestinal microbiota alterations have been linked to helminth-mediated immunoregulation in distal sites [37] . Given that TPL-2 is downstream of TLR4 and IL-1R-signaling and that TPL-2-deficent mice are more susceptible to bacterial infection due to reduced TNFα and IL-1β secretions [12] , we analyzed the changes in microbiota following infection of WT and Map3k8–/–mice . In accordance with previous studies , we observed changes in the fecal microbiota composition of H . polygyrus infected WT mice . However , several of these changes were not observed in Map3k8–/–mice ( grey portion of Fig 2B ) , suggesting that host TPL-2-dependent pathways contributed to H . polygyrus-associated alterations in microbiota composition . Specifically , we observed increases in the proportional abundances of Alphaproteobacteria spp . and Clostridia spp . following infection , which were unchanged in Map3k8–/–mice . Normalizing these differences by co-housing , however , did not impact the accelerated resistance observed in Map3k8–/–mice , indicating that differences in the microbial composition between WT and Map3k8–/–mice were not responsible for resistance to H . polygyrus infection in Map3k8–/–mice . Previous studies have demonstrated the role of TPL-2 in preventing severe airway allergy [17 , 20] and limiting Th2 mediated immunopathology [18] . To date , there have been no studies investigating the role of TPL-2 in regulating helminthiasis , wherein type-2 responses are protective . Furthermore , mechanisms of resistance to H . polygyrus infection [3 , 4 , 6] have been primarily observed in secondary challenge infections post drug treatment . In the present study , we demonstrate a previously unidentified immunoregulatory mechanism with TPL-2 deficiency promoting spontaneous worm expulsion during primary H . polygyrus chronic infection . Resistance to H . polygyrus infection in TPL-2 deficient mice was not due to a detectable increase in adaptive type-2 immunity in the spleen , mLN and PP , determined by Il4GFP expression from CD4+ T cells , IL-5 and IL-13 secretion from HEX stimulated mLN cultures or levels of antigen-specific serum IgG1 , all of which have been implicated in mediating protection from H . polygyrus [5] . This contrasts with the role of TPL-2 in airway allergy and S . mansoni induced immunopathology where we have previously observed enhanced Th2 responses and increased serum antibody responses in Map3k8–/–mice compared to WT mice [17 , 18] . It is likely these differences arise from site and cell specific roles of TPL-2 in regulating inflammatory responses . Our new observations suggest that TPL-2- regulated innate pathways upstream of Th2 cell differentiation , class switch recombination and antibody secretion , orchestrating mechanisms of resistance to intestinal H . polygyrus . Early transcriptional differences in Map3k8–/–mice pointed towards genes associated with chemotaxis and homing of leukocytes to infected tissue , in addition to genes associated with alternatively activated macrophages / dendritic cells . These genes and gene associated pathways have previously been linked with challenge infections with an influx of neutrophils , alternatively activated macrophages , CD4+ T cells and CD11c+ dendritic cells in a type-2 granuloma leading to the killing of tissue larvae [34] . Thus it appeared that TPL-2-deficient mice mounted an accelerated ‘memory-like’ type-2 inflammatory response , rather than a primary response . Accelerated innate pathways may result in more effective killing of invading larvae during a primary infection compared to adaptive immune pathways , that primarily target established adult worms . One such accelerated TPL-2-regulated pathway involved Ccl24 . It was recently reported that IL-33 secretion , which is released soon after H . polygyrus infection [38] , can activate group 2 innate lymphoid cells ( ILC2 ) and drive Ccl24 expression at mucosal surfaces [39] . Although we did not detect differences in Il33 transcripts or differences in the number or the activation states of ILC2 cells , we observed significantly more Ccl24 in TPL-2-deficient mice . Although the precise mechanisms regulating these upstream events remain unclear , it is possible that early IL-33-ILC2 derived type-2 cytokines could activate TPL-2-deficient cells leading to enhanced production of Ccl24 . In support of this activation of myeloid cells , GM-CSF matured dendritic cells and lung resident CD11b+ myeloid cells respond to IL-4R signaling to produce Ccl24 [27 , 40] and IL-33 and ILC2 expansion is actively inhibited by H . polygyrus-derived excretory secretory products [41] , suggesting that H . polygyrus may have evolved to directly inhibit this otherwise anthelmintic response . Using a cell-specific deletion system , we identified that accelerated expulsion of H . polygyrus tracked with deletion of Map3k8 in CD11c-expressing , but not Villin- , Cd4- or LysM-expressing cells . We have previously observed that TPL-2-regulated Ccl24 expression in CD11c+ dendritic cells following HDM challenge [17] . Here we demonstrate that LP-derived CD11c+ CD11b+ cells express increased Ccl24 . Taken together with previous observations , we hypothesized that an early inflammatory model , where IL-33 [38] , and/or IL-4 [42] induced Ccl24 secretion by CD11c+ cells [27 , 43 , 44] orchestrates an early inflammatory response . In accordance with this hypothesis elevated Ccl24 was concurrent with an increase in the frequency of eosinophils , neutrophils , monocytes and activated Il4GFP+ Th2 cells in the LP of H . polygyrus infected Map3k8–/–mice . CCR3 expressing myeloid cells and T lymphocytes can contribute to anthelminthic immunity [1 , 31–33] . In support of this , we demonstrate that increased Ccl24 correlated with increased parasite killing in Map3k8–/–mice . In addition to an increased frequency of myeloid cells and Il4GFP+ Th2 cells in Map3k8–/–mice , we observed a significant decrease in the frequency and a trend for a decrease in the number of other lymphocyte subsets including , CD4+ T cells , CD8+ T cells and CD19+ B cells in Map3k8–/–mice . These observations suggest a change in the cellular landscape in the LP of D5 H . polygyrus infected TPL-2 deficient mice towards a myeloid and Th2 cell-biased environment , which may contribute to an early type-2 ‘memory-like’ inflammatory response . Ex-vivo TPL-2 deficient CD11c+ CD11b+ cells expressed significantly increased Ear11 , Retnla and had a trend for increased Arg1 and Chil3 , markers associated with type-2 alternatively activated macrophages / dendritic cells . Neutralization of Ccl24 led to a significant reduction in Arg1 , Retnla and Ear11 and a trend for a decrease in Chil3 , in H . polygyrus infected TPL-2 deficient mice . Correlating with the loss of these markers , we observed a loss of resistance to infection in TPL-2 deficient mice . Taken together , an accelerated ‘memory-like’ type-2 responses , mediated by CD11c+ cells in H . polygyrus infected TPL-2 deficient mice , led to an influx of inflammatory myeloid and Th2 cells , via Ccl24 , accelerating worm expulsion . These observations point to a previously unidentified role of TPL-2 in restricting Ccl24-mediated immunity to H . polygyrus .
All animal experiments were carried out following UK Home Office regulations ( project license 80/2506 and 70/8809 ) and were approved by The Francis Crick Institute Ethical Review Panel . | Helminth infections remain a huge global burden , causing significant morbidity in both animals and humans . Morbidity and recurring infections are associated with limited access to anthelmintic drugs . While vaccination remains the best available solution to treat helminthiasis , mechanisms of natural or vaccine-mediated immunity to helminths are unclear and efforts are being made to understand genetic factors and immune responses that mediate protection from infection . In this study , we tested and identified the role of a kinase , TPL-2 in regulating protective immunity to the intestinal roundworm , Heligmosomoides polygyrus . Using murine models with deletion of TPL-2 in all cells ( Map3k8–/– ) , we identified the absence of TPL-2 protein was important for mediating protection against infection by the worm . We identified this role for TPL-2 in regulating immunity to H . polygyrus infection was not due to changes in the classical immune responses or intestinal microbiota between TPL-2 deficient and TPL-2 sufficient-wild type ( WT ) mice . Using genome-wide analyses and murine models of infection we discovered that TPL-2 restricted the expression of Ccl24 and the influx of innate immune cells and T cells in the small intestines of H . polygyrus infected mice . Finally we demonstrated TPL-2 mediated expression of Ccl24 is important for developing accelerated immune responses to the worm finally leading to resistance to infection by H . polygyrus . These results reveal a previously unappreciated role for TPL-2 in limiting protection to H . polygyrus infection . Thus , targeting TPL-2 could be advantageous to the development of anti-helminth therapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"t",
"helper",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"microbiome",
"immunology",
"cell-mediated",
"immunity",
"microbiology",
"parasitic",
"diseases",
"microbial",
"genomics",
"digestive",
"system",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"gene",
"expression",
"t",
"cells",
"immune",
"response",
"gastrointestinal",
"tract",
"helminth",
"infections",
"anatomy",
"cell",
"biology",
"immunity",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"genomics"
] | 2017 | TPL-2 restricts Ccl24-dependent immunity to Heligmosomoides polygyrus |
Cutaneous Leishmaniasis ( CL ) is a parasitic infection classified by the WHO as one of the most uncontrolled spreading neglected diseases . Syria is endemic for Leishmania tropica and Leishmania major , causing CL in the Eastern Mediterranean . The large-scale displacement of Syrian refugees exacerbated the spread of CL into neighboring countries . Therapeutic interventions against CL include local , systemic and physical treatments . The high risk for drug-resistance to current treatments stresses the need for new therapies . Imiquimod is an immunomodulatory drug with a tested efficacy against L . major species . Yet , Imiquimod efficacy against L . tropica and the molecular mechanisms dictating its potency are still underexplored . In this study , we characterized the effect of Imiquimod against L . tropica and L . major , and characterized the molecular mechanisms dictating its anti-leishmanial efficacy against both strains . We also investigated the potency and molecular mechanisms of an Imiquimod analog , EAPB0503 , against these two strains . We have tested the effect of Imiquimod and EAPB0503 on macrophages infected with either L . major , L . tropica strains , or patient-derived freshly isolated L . tropica parasites . The anti-amastigote activity of either drugs was assessed by quantitative real time PCR ( RT-PCR ) using kinetoplast specific primers , confocal microscopy using the Glycoprotein 63 ( Gp63 ) Leishmania amastigote antibody or by histology staining . The mechanism of action of either drugs on the canonical nuclear factor kappa- B ( NF-κB ) pathway was determined by western blot , and confocal microscopy . The immune production of cytokines upon treatment of infected macrophages with either drugs was assessed by ELISA . Both drugs reduced amastigote replication . EAPB0503 proved more potent , particularly on the wild type L . tropica amastigotes . Toll-Like Receptor-7 was upregulated , mainly by Imiquimod , and to a lesser extent by EAPB0503 . Both drugs activated the NF-κB canonical pathway triggering an immune response and i-NOS upregulation in infected macrophages . Our findings establish Imiquimod as a strong candidate for treating L . tropica and show the higher potency of its analog EAPB0503 against CL .
Cutaneous leishmaniasis ( CL ) is caused by Leishmania parasite and is classified by the World Health Organization ( WHO ) as one of the most common neglected tropical diseases [1] . During the past decade , an alarming increase in the incidence of CL was documented , ranging from 2 . 1 million cases in 2002 , to approximately 4 million cases in 2015 [2] . In the Eastern Mediterranean , Leishmania tropica and Leishmania major cause CL [3] . In Syria , the prevalence lately doubled due to chronic conflicts [4] . The displacement of Syrian refugees to the neighboring countries , including under-endemic ones like Lebanon , promoted the dissemination of this infection [5] . CL treatment varies among patients [6] , and include local , systemic and physical approaches [7] . Meglumine antimoniate ( Glucantime ) is widely used [8] , but yet presents with many disadvantages such as the painful intra-lesional injections to be repeatedly injected in each lesion , on a weekly basis and for up to 8 weeks [9] . An intramuscular injection of Glucantime was proposed to overcome this painful process , however it was associated with high hepatic and cardiac toxicity [10] . Imiquimod is an FDA approved imidazoquinoxaline against skin infections , with great anti-viral/anti-tumor activities [11] . Imiquimod proved potent in CL treatment [12 , 13] . It was used in combination with systemic antimonials [14] , and presented with cure rates exceeding 90% in refractory patients [15] . Accordingly , it was introduced by the WHO to the guidelines of CL treatment [16] . Among several synthesized Imiquimod analogs [11] , EAPB0503 ( 1- ( 3-methoxyphenyl ) -N-methylimidazo[1 , 2-a]quinoxalin-4-amine ) exhibited higher potency than Imiquimod in several cancer models [17 , 18 , 19] . This study addressed the effect of Imiquimod and its analog , EAPB0503 , in the context of CL , against amastigote stages of L . tropica and L . major parasites . The mechanism of action as well as the elicited immune response were also investigated . This work gives a better insight about the effect of immunomodulatory drugs derivatives on CL , and opens horizons for new and promising treatment paradigm .
To compare the effect of Imiquimod and EAPB0503 on L . major amastigotes , macrophages were infected at the ratio of 5 parasites per cell . Treatment was performed with different concentrations of either drugs for 24 hours . Amastigotes replication was evaluated by real time PCR , using kinetoplast specific primers . Starting the concentration of 0 . 1 μM , L . major amastigotes transcription levels decreased in a concentration-dependent manner following treatment with either drugs , and leading to 80% inhibition of parasite replication at the concentration of 1μM ( Fig 1A ) . L . tropica , the most endemic species causing anthroponotic CL ( ACL ) in the Middle East area [7] , showed that both drugs exert an anti-amastigote activity in a concentration dependent manner . Strikingly , a concentration of 0 . 1 μM was obtained upon treatment with EAPB0503 as compared to 1μM of Imiquimod ( 10 folds higher concentration ) ( Fig 1B ) . This decrease in amastigotes was also more prominent at 0 . 5 and 1 μM of EAPB0503 , compared to the same doses of Imiquimod ( Fig 1B ) . No effect was observed using the vehicle alone ( S1 Fig ) . This promising data clearly shows a different response of leishmanial strains to treatment with either drugs , and a better response obtained upon treatment of L . tropica strain with EAPB0503 . Based on our concentration screening results , we chose the optimal concentration of 0 . 1 μM for further analysis . We examined the effect of this concentration at an earlier time point of 10h . Imiquimod induced a decrease in L . major amastigotes replication by 50% at 10h post-treatment , and by 65% at 24h post-treatment ( Fig 1C ) . More interestingly , EAPB0503 showed a more prominent decrease of amastigotes expression at 10h or 24h post-treatment , where only 10% of amastigote transcripts were detected by RT-PCR ( 5 folds less than Imiquimod ) ( Fig 1C ) . Imiquimod reduced L . tropica amastigotes transcription levels to around 60% at 10h post-treatment and to around 20% at 24h post-treatment ( Fig 1D ) . Interestingly , EAPB0503 reduced amastigotes transcript levels to 30% ( almost 2 folds less than Imiquimod ) at 10h post-treatment and to 10% ( around 2 folds less than Imiquimod ) at 24h post-treatment ( Fig 1D ) . We then assessed the effect of both drugs on amastigotes of L . major and L . tropica strains histologically . Consistent with the transcript results , both drugs had a leishmanicidal effect on both strains ( Fig 1E and 1F ) . Whilst Imiquimod displayed similar results against L . major at both time points ( Fig 1E ) , EAPB0503 was more potent against L . tropica strain ( Fig 1F ) . Altogether , these data show that EAPB0503 acts at the low dose of 0 . 1 μM and as early as 10h , when compared to its parental compound Imiquimod . Imiquimod belongs to the class of Toll-like receptor ( TLR ) agonists with high affinity to TLR7 , commonly involved in pathogen recognition ( Fig 2A ) [14 , 20] . We investigated the molecular mechanisms underlying the potency of Imiquimod and its analog EAPB0503 against Leishmania amastigotes . We focused on L . tropica and adopted the concentration of 0 . 1 μM at both time points 10 and 24h post-treatment . Our results showed that TLR7 protein levels increased after treatment with either drugs , in comparison to uninfected or untreated infected macrophages ( Fig 2B ) . Same results were obtained on L . major infected macrophages upon treatment with either drugs ( S2 Fig 2 ) . In case of L . tropica and consistent with published data , the upregulation was maximal upon treatment with Imiquimod for 10h ( Fig 2B ) . EAPB0503 induced a higher protein expression of TLR7 . Nonetheless , the highest induction of TLR7 was obtained upon treatment with Imiquimod . Our results confirm the mechanism of action of Imiquimod via this receptor in the context of CL . The lower expression of TLR7 upon treatment with EAPB0503 , seemingly shows a potential mode of action through a different TLR . Following recognition of pathogens , TLRs trigger the NF-κB pathway activation ( Fig 2A ) [21] inducing immune inflammatory responses [22] . Imiquimod activates the canonical NF-κB pathway ( Fig 2A ) [23] . We explored this pathway in the context of CL . Western blot analysis clearly showed an activation of the multimeric IKK complex ( IKKα/IKKβ ) after 10 or 24h treatment with either drugs ( Fig 2C ) . Furthermore , an induction of the phosphorylated form of the IκBα at both time points was obtained , presumably leading to its degradation ( Fig 2C ) . We then examined whether this NF-κB activation involves the canonical pathway . Our results demonstrate that the p50 subunit was upregulated especially upon 24h post-treatment with either drugs ( Fig 2C ) . This led to the nuclear translocation of p65 ( Fig 2D ) , which represents the active NF-κB subunit , and known to activate immune response genes . Collectively these results showed that both Imiquimod and EAPB0503 inhibit amastigote replication via activation of the canonical NF-κB pathway . We investigated the expression of pro- and anti-inflammatory mediators after treatment . Macrophage Inflammatory Proteins ( MIP-1α and β ) and Monocyte Chemoattractant Protein ( MCP-1 ) levels increased upon treatment with both drugs ( Fig 3A ) . The secreted levels of depicted pro-inflammatory cytokines namely Interleukin-12 ( IL-12 ) , IL-1β , TNF-α and IL-6 , important in CL clearance [24] , were increased upon treatment with Imiquimod or EAPB0503 ( Fig 3A ) . Macrophage-derived nitric oxide ( NO ) is effective against microbes , and synthesized by Nitric Oxide Synthase ( i-NOS ) . i-NOS is induced in response to pro-inflammatory cytokines [25] and , in CL-infected macrophages , iNOS is protective against L . major [26] . Both drugs increased i-NOS transcripts in macrophages infected with either L . tropica or L . major strains , with EAPB0503 inducing a 5-fold higher expression ( Fig 3B , S2B Fig respectively ) . This presumably leads to higher NO production , hence enhanced leishmanicidal activity . In CL , pro-inflammatory cytokines are linked to resistance against leishmaniasis; whereas anti-inflammatory cytokines relate to disease progression [27] . We examined the secretion levels of two depicted anti-inflammatory cytokines , IL-10 and IL-4 after treatment with either drugs . In comparison to non-treated L . tropica infected macrophages , secretion levels of IL-10 and IL-4 decreased by around 4 folds after treatment with Imiquimod ( Fig 3C ) . More interestingly , treatment with EAPB0503 showed a significant decrease by 4 and 15 folds of IL-10 and IL-4 respectively and as compared to non-treated infected macrophages ( Fig 3C ) . Altogether , our results show that NF-κB activation by Imiquimod and EAPB0503 induces secretion of pro-inflammatory cytokines . This leads to i-NOS upregulation , presumably leading to NO production and leishmanicidal activity . In addition , and concomitantly with the upregulation of pro-inflammatory cytokines , a decrease in the anti-inflammatory cytokines is obtained , diminishing macrophage susceptibility to L . tropica infection , and triggering the leishmanicidal effect of the tested drugs . To eliminate the potential doubt due to the susceptibility of cultured L . tropica and L . major strains to our tested treatments ( e . g genetic drift and less virulent strains after long term culture ) , we investigated the effect of Imiquimod and its analog EAPB0503 on freshly isolated parasites from untreated patients’ biopsies . The infection with L . tropica was confirmed in all used patients by PCR and Restriction Fragment Length Polymorphism ( RFLP ) ( S3A and S3B Fig respectively ) . Both drugs inhibit amastigote replication in a time dependent manner . As a control of potency , we used Glucantime , alone or combined with Imiquimod , since these drugs were clinically used in the treatment of CL [15 , 16 , 28] . Upon treatment with Imiquimod , amastigotes transcription levels decreased to reach around 10% at 72h post-treatment as compared to 60% upon treatment with Glucantime alone , or around 25% upon treatment with Glucantime combined to Imiquimod ( Fig 4A ) . Treatment with EAPB0503 induced a more prominent effect where , L . tropica amastigote transcripts decreased by 70 and 90% , at 10 and 24h post-treatment respectively ( Fig 4A ) . This effect was identical at 72h post-treatment ( Fig 4A ) . Interestingly , the effect of EAPB0503 was more profound than Glucantime alone , Imiquimod alone or Glucantime combined to Imiquimod at all tested time points , and starting 10h post-treatment ( Fig 4A ) . We then assessed the effect of both drugs on amastigotes histologically ( Fig 4B ) and by immunofluorescence confocal microscopy ( Fig 4C ) . Our results were very consistent with the transcript data with less amastigotes detected upon treatment . Consistently , and using the Leishmania Glycoprotein Gp63 marker for quantification of amastigotes [29] , Imiquimod treatment led to a decrease in amastigotes percentage reaching 60% at 10h and 35% at 24h post-treatment ( Fig 4C ) . More interestingly , EAPB0503 induced a more prominent decrease in amastigotes number , to 40% after 10h of treatment , and to 25% after 24h of treatment ( Fig 4C ) . Altogether , these data show that Imiquimod and mostly EAPB0503 are highly active at the low dose of 0 . 1 μM and as early as 10h on patients’ derived L . tropica stages , confirming the obtained results on in vitro cultured strains . The mechanism of action of either drugs on L . tropica obtained from patients , was evaluated for TLR7 protein expression and showed an increase after treatment , in comparison to uninfected or untreated infected macrophages . Consistently with the cultured strain , Imiquimod induced the highest TLR7 protein levels ( Fig 5A ) . Moreover , i-NOS transcript levels were increased reaching the highest levels after 10h treatment with EAPB0503 ( Fig 5B ) . These results indicate a similar mode of action of both drugs on freshly isolated parasites from patients’ biopsies and confirm the higher potency of EAPB0503 against CL .
CL is one of the most common neglected tropical diseases worldwide . Globally , the annual incidence of CL is estimated to be 0 . 7 to 1 . 2 million new cases per year . This disease is still endemic in many countries [30] . In the Eastern Mediterranean , Syria presents with the highest number of CL cases [31] . The Syrian conflict exacerbated the spread of the infection to the surrounding areas . In Lebanon , 85% of infected Syrian refugees were diagnosed with L . tropica whilst the remaining 15% were infected with L . major [5] . Pentavalent antimony compounds remain the treatment choice for CL . However , these compounds associate with high cost , poor availability , drug resistance and systemic toxicity [32] . We focused on testing novel drugs’ efficacy on L . tropica and L . major . Imiquimod activates macrophages [15] , the main host cells for Leishmania replication . In CL , Imiquimod was mainly tested against L . major amastigote [33] . In Imiquimod treated mice infected with L . major , an increased protection was obtained [34] . In CL patients , Imiquimod combined to Glucantime induced a high healing rate in refractory patients [13 , 16 , 34] . We showed that both drugs affected amastigotes . Conversely , EAPB0503 was more potent on L . tropica strain [4 , 5] . Imiquimod acts via binding TLR7 , leading to the activation of the NF-κB pathway . Imiquimod protective effect , on L . major infected mice , was coupled with the induction of NO synthesis [26 , 35] . Consistent with the published data , but on L . tropica strain , we showed that Imiquimod and EAPB0503 upregulated TLR7 expression . Nonetheless , the highest induction was obtained upon Imiquimod treatment . This finding suggests that EAPB0503 may partially act via TLR7 , or through other TLRs . The canonical NF-κB pathway was activated by both drugs , leading to increased secretion levels of pro-inflammatory cytokines . MIP-1α and MIP-1β , involved in resistance against infections [36] were both secreted at higher levels upon treatment with either drugs . Consistently with the known role of MIP-1α and MIP-1β in recruiting other cytokines such as TNF-α and IL-6 [37] , levels of secretion for these two cytokines were also increased . TNF-α increased secretion levels were consistent with its protective role against CL [38] . Previous studies have shown that MCP-1 is highly expressed in lesions of patients with self-healing localized cutaneous leishmaniasis whereas it is scarce in those of chronic diffuse cutaneous leishmaniasis [39] . This suggests its role in the parasites elimination via induction of reactive oxygen intermediates ( ROI ) . Our results showed that MCP-1 levels increased upon treatment with either drugs , but more importantly with EAPB0503 , presumably explaining its higher potency . However , the potential involvement of ROIs on the clearance of treated parasites remains to be elucidated . We also succeeded to test the activity of Imiquimod and EAPB0503 on freshly isolated L . tropica from skin lesions of CL patients . We confirmed the results obtained on cultured strains , thus eliminating any potential doubt about a lower virulence or a genetic drift obtained from long term cultures . Moreover , EAPB0503 showed a better anti-leishmanicidal activity than the clinically used Glucantime , whether alone or combined to Imiquimod [15 , 16 , 28] . These results highlight the promising potency of EAPB0503 for CL treatment . Nitric Oxide production by activated macrophages is known to play a major role in fighting against infections [40] , including Leishmania [41] . Inhibition of i-NOS reduced L . infantum burden in human macrophages [42] . In addition , the increase of i-NOS and NO generation in response to IFN-γ and TNF-α is crucial to control CL [43] . We checked for i-NOS transcripts in treated macrophages infected with either L . major or patients’ derived L . tropica and showed an important increase with either drugs . Interestingly , the highest levels were obtained upon EAPB0503 treatment , presumably explaining its higher leishmanicidal efficacy . TLRs are important pattern recognition receptors expressed abundantly on macrophages . Early studies concluded that TLR2 , TLR4 , and TLR9 , are involved in the recognition of L . major and that TLR2 ligands play a protective immune role against Leishmaniasis [44] . However , recent studies on C57BL/6 mice deficient in either TLR2 , 4 , or 9 , showed that only TLR9-/- mice are more susceptible to L . major infection , indicating TLR2 and TLR4 related immunity to murine leishmaniasis requires re-evaluation [45] . In this study , we confirmed that Imiquimod displays its anti-amastigote activity via TLR7 upregulation , leading to NF-κB activation and pro-inflammatory cytokine production . EAPB0503 effect on TLR7 was less prominent . Whether EAPB0503 acts via any of the important TLRs in CL or not , requires further investigation . Collectively , our results did not only show a promising efficacy of a new compound , EAPB0503 against CL , but also highlighted the mechanism of action through which Imiquimod and its analog acted against the aggressive L . tropica strain . We also described the molecular mechanisms of these drugs against amastigotes highlighting the importance of immune-modulatory therapy against CL .
Leishmania major ( MHOM/MA/81/LEM265 and MMER/MA/81/LEM309 ) and Leishmania tropica ( MHOM/LB/76/LEM61 , MRAT/IQ/72/ADHANIS1 ) were purchased from the CRHU “Montpellier” . Parasites were maintained in RPMI1640 ( Lonza ) supplemented with 10% Fetal Bovine Serum ( FBS ) , 100IU/ml streptomycin/penicillin ( Sigma ) . Imiquimod was purchased from Molekula ( Wessex House ) and EAPB0503 was synthesized using microwave-assisted chemistry as described by Khier et al . [46] . Drugs were prepared as a 0 . 1 M stock in dimethylsulfoxide ( DMSO ) and stored at -80°C . Glucantime ( 1 . 5g/5ml ) was used at the final concentration of 100 μg/mL . Working solutions of 0 . 1 μM were freshly prepared in culture media . Human monocytic THP-1 cells ( American Type Culture Collection ( ATCC TIB-202 ) , Manassas , VA ) were grown in RPMI1640 medium with L-Glutamine , 25 mM Hepes ( Lonza ) , supplemented with 10% of fetal bovine serum ( FBS ) , 1% penicillin-streptomycin , 1% kanamycin and 1% glutamine ( Invitrogen ) . 1 million THP-1 cells were differentiated into macrophages , using 50 ng/mL of phorbol 12-myristate 13-acetate ( PMA , Sigma ) overnight . Following their adherence , differentiated macrophages were then activated with 1 μg/mL of LPS for 4h , then infected with L . major or L . tropica at the ratio of 5 parasites/macrophage , and incubated for 24h at 37 ᴼC . Non-internalized promastigotes were removed by two gentle washes with PBS . Punch biopsies ( 4 mm of diameter ) from three CL patients were performed in 2016 , and incubated in sterile physiological serum , supplemented with Penicilline G ( 100 IU/ml ) . Specimens were incubated in a semi-solid culture media ( 10g agar , 3g NaCl , 500 mL water ) . 3 weeks later , promastigotes were transferred to liquid medium . Macrophages infected with L . major , L . tropica , or patients’ derived L . tropica parasites were treated with Imiquimod and EAPB0503 ( 0 . 01 μM , 0 . 05 μM , 0 . 1 μM , 0 . 5 μM and 1 μM ) for 24h . Total RNA was extracted using Trizol ( Qiagen ) . cDNA synthesis was performed using a Revert Aid First cDNA synthesis Kit ( #K1622-Thermo Scientific ) . Syber green qRT PCR was performed using the BIORAD-CFX96 machine . Primers for the housekeeping Glyceraldehyde-3-Phosphate dehydrogenase GAPDH , and i-NOS are listed in Table 1 . Primers for amastigotes detection target the minicircle kinetoplast DNA ( kDNA ) ( Table 1 ) . Percentage of expression was calculated according to Livac method [47] . Supernatants of infected macrophages in presence or absence of either drugs were collected 10h and 24h after treatment , and ELISA was performed using Multi-Analyte ELISArray Kit ( Qiagen ) according to the manufacturer’s instructions . Briefly , supernatants of L . tropica infected macrophages ( untreated or treated with 0 . 1 μM of Imiquimod or EAPB0503 ) were collected . Supernatants were spun for 10 min at 1000g and transferred to new Eppendorf tubes , and diluted using a specific cocktail of antigens ( IL-12 , IL-1β , IL-6 , and TNF-α , MIP-1α , MIP-1β , MCP-1 , IL-10 and IL-4 ) provided by the kit ( Qiagen ) . Samples were then loaded in the coated ELISA plaque , and were incubated for 2 hours . 3 washes were performed , and the detection antibody was added and incubated for 2 hours . Then , Avidin-HRP was added for 30 min , and 4 washes were performed . Development solution was then added in dark and kept for 15 min , before addition of the stop solution . The secreted levels of the following cytokines and chemokines were then assessed . The optic density ( O . D ) was determined at 450 and 570 nm and calculated according to the standard values of a positive control provided by the kit . For Immunofluorescence assay , p6 well plates were seeded with activated macrophages infected with L . tropica ( 5p/c ) for 24h and treated with Imiquimod or EAPB0503 for 10 or 24h . At these time points , coverslips were fixed in 4% paraformaldehyde for 20 minutes . Permeabilization was performed in Triton ( 0 . 2% ) for 10 minutes . Following one PBS wash , blocking for 30 min with PBS-10% FBS was performed . Primary antibody directed against the NF-κB p65 subunit ( Santa Cruz , Sc-8008 ) was used at the dilution of 1:50 . For Leishmania parasite staining inside macrophages , an anti-Gp63 ( LifeSpan BioSciences , LS-C58984 ) was used at the dilution 1:50 . Anti-mouse secondary antibodies ( Abcam , ab150116 ) were used at the concentration of 1:100 . Staining of nuclei was performed using 1 μg/mL of Hoechst 33342 , trihydrochloride trihydrate solution ( Invitrogen , H33342 ) for 5 min and then coverslips were mounted on slides using a Prolong Anti-fade kit ( Invitrogen , P36930 ) . Z-section images were acquired by confocal microscopy using a Zeiss LSM 710 confocal microscope ( Zeiss , Germany ) and all images were analyzed using Zeiss LSM 710 software . H&E staining was performed as described by Grosset et al . , 2017 [50] . Briefly , hematoxylin ( Fisher Scientific , Canada ) was added on cells , and a counterstaining for 30 seconds was performed followed by a water rinse for 5 minutes . Slides were then dipped in 50% ( vol/vol ) alcoholic eosin Y solution ( Leica Microsystems , Canada ) then rinsed in ethanol before slide mounting . Giemsa staining was performed using Ral 555 Kit ( RAL Diagnostics ) . Briefly , cells were fixed with methanol for 1 min , stained with solution 2 for 40 seconds then with Solution 3 for 25 seconds as per the manufacturer . Cells were then mounted using Prolong Anti-fade ( Invitrogen , P36930 ) . Activated macrophages infected with L . tropica ( 5p/c ) for 24h were treated with 0 . 1 μM of Imiquimod or EAPB0503 for 10 or 24h . Cells were scrapped , washed with PBS , and pellets were re-suspended in 1x Laemmli buffer . Following denaturation , samples were run on 10% polyacrylamide gels . Proteins were then transferred to nitrocellulose membranes ( BIO RAD Cat# 162–0112 ) at 30V overnight using a BioRad transfer unit . To verify the protein transfer , nitrocellulose membranes were stained with Ponceau Red . Blocking was performed for 1h in 5% of Bovine Albumin serum ( BSA ) in wash buffer and probed with specific primary antibodies against TLR7 ( sc- 57463 Santa Cruz Biotechnology , 1:100 ) , NF-κB p65 ( sc-8008; Santa Cruz Biotechnology , 1:250 ) , or p52 ( sc-7386 , Santa Cruz Biotechnology , 1:250 ) . Equal loading was tested following probing with the anti- GAPDH antibody ( MAB5476; abnova , 1:20 000 ) . Nitrocellulose membranes were then washed three times with wash buffer for 5 minutes each , before incubation with the appropriate anti-mouse secondary antibody conjugated to Horseradish peroxidase ( HRP ) ( m-IgGk BP-HRP , Santa Cruz , sc-516102 , 1:5000 ) . Bands were visualized by autoradiography , following incubation with luminol chemiluminescent substrate ( Bio-Rad , Cat# 170–5061 ) . DNA extraction from patient’s biopsies and PCR were performed as previously described using ITS-1 primers ( Table 1 ) [49] . Following PCR amplification , 10 μl of the remaining volume of the amplicon was digested with 2 μl MnII enzyme , in 2 μl in 10x Buffer G ( Fermentas Life Sciences , Thermo Fisher Scientific ) and 18μl of nuclease-free water . Digestion was carried out using the TS100 thermal cycler ( Biorad ) by incubating for 6h at 37°C , followed by enzyme inactivation for 20 minutes at 65°C . 25 μl of the digestion products were electrophoresed on 1 . 5% agarose gels at 100 V in 1× TBE buffer ( 0 . 04 M Tris-acetate and 1 mM EDTA , pH 8 . 0 ) for 30 minutes . The restriction fragments were visualized under UV light . Images were then captured using the GelDoc-IT TM Imaging System and RFLP patterns interpreted for sub-speciation . Leishmania tropica is characterized by the presence of two bands of 300 base pairs ( bp ) and 50bp respectively . Sample collection was approved by the Institutional Review Board of the American University of Beirut ( PALK . IK . 01 ) . A parent of each of the three children participants provided written informed consent on the child’s behalf before sample collection . Continuous variables were analyzed by the unpaired Student’s t test . P value was determined and values for p < 0 . 05 were considered as significant . | Cutaneous Leishmaniasis ( CL ) is a parasitic infection caused by Leishmania ( L . ) parasites . In the Old World and the Near East , CL is mainly caused by L . major and L . tropica . The ongoing Syrian war and the resulting massive population displacement led to an alarming increase in the incidence of CL , in Syria and its surrounding countries . Current therapies against CL lead to partial or complete cure in L . major infections but are less effective against L . tropica . These therapies associate with several limitations , including patients’age , immune system , repetitive painful injections , high cost , poor availability , and mainly systemic toxicity . Therefore , it is of high interest to seek for novel drugs against CL . We assessed the activity of an immunomodulatory drug and its analog against L . major and L . tropica parasites and showed their potency . Importantly , the analog proved more efficient against the wild type L . tropica strain . These results highlight the promising efficacy of immuno-modulatory drugs against CL . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"immunology",
"tropical",
"diseases",
"microbiology",
"parasitic",
"diseases",
"protozoan",
"life",
"cycles",
"parasitic",
"protozoans",
"developmental",
"biology",
"protozoans",
"pharmaceutics",
"leishmania",
"molecular",
"development",
"neglected",
"tropical",
"diseases",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"zoonoses",
"animal",
"cells",
"proteins",
"life",
"cycles",
"protozoan",
"infections",
"amastigotes",
"immune",
"system",
"toll-like",
"receptors",
"biochemistry",
"signal",
"transduction",
"macrophages",
"eukaryota",
"cell",
"biology",
"physiology",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"protozoology",
"drug",
"therapy",
"immune",
"receptors",
"organisms"
] | 2018 | EAPB0503: An Imiquimod analog with potent in vitro activity against cutaneous leishmaniasis caused by Leishmania major and Leishmania tropica |
Successful execution of the meiotic program depends on the timely establishment and removal of sister chromatid cohesion . LAB-1 has been proposed to act in the latter by preventing the premature removal of the meiosis-specific cohesin REC-8 at metaphase I in C . elegans , yet the mechanism and scope of LAB-1 function remained unknown . Here we identify an unexpected earlier role for LAB-1 in promoting the establishment of sister chromatid cohesion in prophase I . LAB-1 and REC-8 are both required for the chromosomal association of the cohesin complex subunit SMC-3 . Depletion of lab-1 results in partial loss of sister chromatid cohesion in rec-8 and coh-4 coh-3 mutants and further enhanced chromatid dissociation in worms where all three kleisins are mutated . Moreover , lab-1 depletion results in increased Aurora B kinase ( AIR-2 ) signals in early prophase I nuclei , coupled with a parallel decrease in signals for the PP1 homolog , GSP-2 . Finally , LAB-1 directly interacts with GSP-1 and GSP-2 . We propose that LAB-1 targets the PP1 homologs to the chromatin at the onset of meiosis I , thereby antagonizing AIR-2 and cooperating with the cohesin complex to promote sister chromatid association and normal progression of the meiotic program .
Timely establishment and subsequent removal of Sister Chromatid Cohesion ( SCC ) between sister chromatids is necessary to facilitate faithful segregation of chromosomes in mitosis and meiosis . Failure to correctly segregate chromosomes in either mitosis or meiosis has been associated with tumorigenesis , miscarriages , and congenital defects [1] , [2] . Premature loss of SCC prevents the correct bipolar attachment of sister kinetochores to the mitotic spindle , whereas a delay in removing SCC may prevent segregation and therefore result in mitotic arrest and aneuploidy ( reviewed in [3] , [4] ) . During meiosis , the control of SCC establishment and removal is even more intricate , and many meiotic processes fail when SCC is compromised . In this specialized cell division program , one cycle of chromosome replication is followed by two consecutive rounds of chromosome segregation , thus reducing the chromosome number by half to produce haploid sperm and oocytes . At the onset of meiosis I , homologous chromosomes pair and undergo synapsis mediated by the formation of a proteinaceous scaffold called the synaptonemal complex ( SC ) . When the SC disassembles , homologs remain attached to each other through chiasmata as a result of earlier crossover recombination events underpinned by flanking SCC . A tight regulation of the establishment of SCC is required for the normal progression of these meiotic events . However , while studies from a number of organisms have revealed key insights into the regulation of SCC removal , far less is known about how the establishment of SCC is regulated . Cohesin is a highly conserved multisubunit complex that establishes SCC by binding the newly formed chromatid as it is synthesized . The four core subunits of cohesin are the structural maintenance of chromosomes proteins Smc1 and Smc3 , the Scc1/Rad21 kleisin , and the accessory protein Scc3 ( reviewed in [5] , [6] ) . During meiosis in monocentric organisms such as flies , vertebrates , and yeast , dissolution of chromosome cohesion occurs in a stepwise manner and involves phosphorylation and degradation of Rec8 , a meiosis-specific Scc1 paralog . First , cohesin is removed at the chromosome arms during meiosis I , but actively maintained at the pericentromeric regions until anaphase II segregation when the remaining cohesin subset is degraded [7]–[9] . At the end of meiosis I , members of the Shugoshin protein family prevent cohesin removal from the centromeres by recruiting protein phosphatase 2A ( PP2A ) , which counteracts the phosphorylation of Rec8 by Aurora B Kinase , thereby sparing cohesin from separase-mediated degradation [10]–[19] . In C . elegans , the holocentric homolog pairs undergo a single crossover event located at the terminal thirds of chromosomes [20] , [21] . Upon chromosome condensation , this off-center crossover leads to the characteristic cruciform shape of the bivalents , which display long and short arms that play an important role for correct alignment on the metaphase I plate [22] , [23] . As in monocentric organisms , cohesin is also removed in a two-step process in C . elegans . While REC-8 is lost on the short arm in meiosis I , REC-8 is preserved on the long arm past homolog segregation . This process coincides with the differential loading of the Aurora B Kinase homolog AIR-2 , which during diakinesis is observed exclusively on the short arms where it is proposed to phosphorylate REC-8 , thus licensing its cleavage by Separase [24] , [25] . Recently , two other kleisin homologs , COH-3 and COH-4 , were found to also play a part in SCC during C . elegans meiosis , but their specific roles and localization are still unknown [26] . In both mammals and fission yeast two Shugoshin paralogs function in cohesin protection as well as in the spindle assembly checkpoint [14] , [19] , [27]–[29] , yet it was suggested that the latter is the ancestral role of Shugoshin , and that the protection of sister chromatid cohesion evolved as its consequence [27] . So far only a single sequence-predicted Shugoshin homolog was identified in C . elegans , SGO-1 , but neither AIR-2 localization nor sister chromatid association are compromised during meiosis I in sgo-1 mutants [16] , [30] . Instead , our studies suggested that the worm-specific LAB-1 ( Long Arms of the Bivalent ) protein participates in protecting REC-8 at the long arms during the metaphase I to anaphase I transition [30] . LAB-1 progressively forms continuous tracks throughout the full length of chromosomes starting at the onset of meiosis and co-localizes with the synaptonemal complex at pachytene [30] . During chromosome remodeling , LAB-1 becomes restricted to the long arms of the bivalents and , like Shugoshin in monocentric species , is finally removed from chromosomes in early anaphase I . lab-1 hypomorphic mutants show a spreading of AIR-2 signals to both arms of the bivalents , similar to mutants of the protein phosphatase 1 ( PP1 ) homolog gsp-2 . These results are consistent with a model in which LAB-1 impacts sister chromatid cohesion in late meiosis I by regulating cohesin phosphorylation . Based on these and other observations , as well as the lack of evidence for a direct role of Shugoshin in protecting meiotic cohesin in C . elegans , we have speculated that due to the holocentric nature of C . elegans chromosomes , Shugoshin maintained its roles in the spindle attachment checkpoint , but LAB-1 evolved as part of a process to protect cohesin during meiosis in this organism [30] . However , both how and when PP1 function is directed and regulated remained open questions . Here we describe an earlier and distinct role for LAB-1 in the establishment of SCC via PP1 regulation . Moreover , we demonstrate how failing to properly establish SCC influences various downstream meiotic events . Depletion of lab-1 by RNAi reduces SCC and consequently impairs homolog pairing , alters the progression of meiotic recombination , and results in an increase in recombination intermediates ( MSH-5 and ZHP-3 foci ) . We found that LAB-1 , together with REC-8 , is required for proper loading of SMC-3 and consequently for proper SC polymerization , which requires normal axis morphogenesis . While the different cohesin members and LAB-1 show some degree of interdependence with respect to either their initial localization or the subsequent maintenance of their localization on chromosomes , LAB-1 can promote partial SCC even in the absence of all three SCC-1 meiotic paralogs . Finally , underscoring a role in the regulation of phosphorylation , LAB-1 directly interacts with GSP-1 and its paralog GSP-2 . Moreover , depletion of lab-1 results in reduced GSP-2 and increased AIR-2 signals in early meiotic nuclei . We propose that LAB-1 specifically targets PP1 to chromosomes in early meiotic stages to antagonize AIR-2 phosphorylation and to promote sister chromatid cohesion , thus supporting the normal progression of downstream meiotic events .
Our analysis of lab-1 ( RNAi ) worms revealed the presence of >12 DAPI-stained bodies in 1 . 1% ( n = 88 ) of oocytes at diakinesis , compared to the six DAPI-stained bodies corresponding to the six pairs of attached homologous chromosomes observed in wild type or the 7 to 12 univalents indicative of lack of chiasmata , suggesting instead a defect in sister chromatid cohesion ( this study and [30] ) . To determine whether this defect results from a role for lab-1 in early meiosis , we examined the effects of lab-1 depletion on the various processes that take place earlier during prophase I ( Figure S1 ) . We first examined homologous chromosome pairing , a process that occurs upon entry into meiosis in most organisms ( reviewed in [31]–[34] ) . To follow the progression of pairing in the germline , we used fluorescence in situ hybridization ( FISH ) with probes labeling the pairing center end of chromosome I . FISH signals either ≤0 . 75 µM or >0 . 75 µM apart represent paired and unpaired homologs , respectively ( Figure 1A ) . In control gonads , homologous chromosomes are unpaired prior to meiotic entry and therefore only a few nuclei ( n = 17/81 ) show a single FISH signal ( Figure 1B and 1C , zones 1 and 2 ) . Upon entry into meiotic prophase I ( transition zone , which corresponds to the leptotene and zygotene stages; zone 3 ) , the frequency of nuclei carrying a single focus increases ( n = 52/72 ) , and by pachytene , pairing is completed ( Figure 1C , zones 4–7 ) . In contrast , in lab-1-depleted gonads , pairing levels were reduced . Only 52% ( n = 44/84 ) of nuclei in the transition zone carried paired homologs ( Figure 1C , zones 3 ) , and this was further decreased until only 24% ( n = 9/37 ) of nuclei were observed with paired homologs at late pachytene ( Figure 1C , zone 7 , p<0 . 05 , by the two-sided Fisher's Exact Test , 95% C . I . ) . Using a probe targeting the X chromosome pairing center revealed similar , albeit milder , results ( Figure 1D ) . Although the reason for a stronger impairment of pairing on the autosomes compared to the X chromosome remains unclear , this has also been observed in other meiotic mutants [35]–[37] . Thus , depleting lab-1 reduces homolog pairing , suggesting a previously unknown function for LAB-1 during prophase I . In 20% ( 14/69 ) of premeiotic nuclei from lab-1-depleted gonads , we also noticed the presence of three or four FISH signals ( Figure 1A ) , consistent with a defect in sister chromatid association ( Figure 1C ) . At later stages , the frequency of nuclei with 3–4 signals decreased in lab-1 ( RNAi ) gonads ( 5% , n = 2/37 , at late pachytene ) . This temporal reduction could be explained either by residual LAB-1 or by a LAB-1-independent mechanism . Taken together , these observations suggest that LAB-1 may affect homologous pairing due to its role in the early establishment and/or maintenance of SCC . To examine whether lab-1 depletion affects the progression of meiotic DNA double-strand break repair ( DSBR ) , we utilized an antibody against RAD-51 , a protein involved in strand invasion/exchange during DSBR [38] . In control gonads , the levels of RAD-51 foci peaked in early/mid-pachytene ( zone 5 ) , and progressively decreased in later stages ( Figure 2A and 2C ) . In contrast , levels of RAD-51 foci were elevated throughout mid to late pachytene ( zones 5–7 ) and persisted in 88% ( n = 95 ) of early diplotene ( zone 8 ) nuclei in lab-1 ( RNAi ) gonads compared to only 21% ( n = 97 ) of nuclei in control gonads ( Figure 2B and Figure S2 ) . The elevated levels of RAD-51 foci observed in lab-1 ( RNAi ) gonads depend on the formation of meiotic DSBs by SPO-11 and are therefore indicative of a meiosis-specific DSBR defect ( Figure S3 ) . These results can be explained by either a delay in meiotic DSBR or an increase in the levels of DSB formation upon lab-1 depletion . Consistent with the interpretation that nuclei in lab-1 ( RNAi ) germlines contain unrepaired recombination intermediates , we observed a CEP-1/p53-dependent 2- to 3-fold increase in germ cell apoptosis in lab-1 ( RNAi ) gonads compared to control , suggesting the activation of a late pachytene DNA damage checkpoint ( p<0 . 0001 by the two-tailed Mann-Whitney test , 95% C . I . ; Figure 2D and Figure S4 ) . These results show that depletion of lab-1 perturbs normal meiotic DSBR . Meiotic crossovers are tightly regulated such that at least one crossover always forms between homologs while additional crossovers nearby are discouraged [39] . Due to the decreased levels of homologous pairing observed , we hypothesized that following lab-1 depletion , crossover levels would also be reduced . To highlight crossover precursor sites , we used a ZHP-3::GFP transgene [40] , [41] . In C . elegans , six ZHP-3::GFP foci are observed in >78% of late pachytene nuclei , correlating with the expected one crossover event per bivalent ( Figure 2E ) . Surprisingly , we observed a mean of 9 . 2 foci per nucleus in lab-1 ( RNAi ) pachytene nuclei ( n = 63 , p<0 . 0001 by the two-tailed Mann-Whitney test , 95% C . I ) , and 4/63 had >12 foci ( Figure 2E ) . To verify that these foci represent recombination events , we immunostained both control and lab-1 ( RNAi ) gonads with an antibody that recognizes MSH-5 , a conserved meiosis-specific protein required for crossover formation [42]–[45] . We found that the pattern of distribution of MSH-5 foci was almost identical to that of ZHP-3 ( Figure 2F ) . These results suggest that lab-1 depletion may disrupt crossover control and lead to more crossover events . Utilizing a similar cytological approach to that in Rosu et al . [46] we did not observe any bivalents with two or more chiasmata in diakinesis oocytes from lab-1-depleted gonads ( n = 0/88; Figure S5 ) . This outcome , coupled with the significant reduction in homologous pairing observed in these gonads , suggests that some of these recombination events may represent crossovers between sister chromatids as opposed to between homologs . The SC plays an essential role in promoting the maturation of DSBs into crossovers [31] , [33] , [34] . Therefore , we examined whether impaired chromosome synapsis might account for the altered DSBR progression observed in lab-1 ( RNAi ) gonads . Specifically , we examined whether LAB-1 is required for the localization of SYP-1 , a central region component of the SC . In wild type pachytene nuclei , SYP-1 is localized throughout the full length of thick DAPI-stained tracks representing paired and aligned homologous chromosomes ( Figure 2G ) [47] . In the lab-1-depleted gonads , SYP-1 signal was observed associating with chromosomes with wild type kinetics upon entry into meiosis . However , SYP-1 was not detected along the full length of DAPI-stained chromosomes in many nuclei at pachytene , as exemplified by co-staining with HTP-3 , an axial element component [48] that is observed continuously throughout chromosome axes ( Figure 2G ) . In contrast , LAB-1 is still observed localizing throughout chromosomes in syp-1 mutants ( Figure S6B ) . This suggests that while assembly of lateral element components of the SC is apparently LAB-1-independent at this level of cytological observation , assembly of the central region components of the SC requires LAB-1 function . FISH analysis and the number of DAPI-stained bodies ( >12 ) in diakinesis oocytes suggested that the impaired DSBR progression and chromosome synapsis observed in lab-1 ( RNAi ) gonads may be due to an earlier role of LAB-1 in sister chromatid cohesion . To determine whether LAB-1 executes this role through interactions with the cohesin complex , we first examined if LAB-1 localization depends on cohesin . Depletion of the cohesin member smc-3 resulted in an overall decrease in LAB-1 signal throughout prophase I nuclei ( Figure 3 ) . In late pachytene nuclei , only short tracks of LAB-1 were observed on the chromosomes , and LAB-1 was detected on univalents at diakinesis ( Figure 3 ) . A similar pattern was observed following depletion of the cohesin complex subunit scc-3 ( Figure S7 ) . We also examined whether meiosis-specific cohesin subunits were required for LAB-1 localization . In addition to REC-8 , two other kleisins , COH-3 and COH-4 , mediate meiotic sister chromatid cohesion in C . elegans [26] , [49] . LAB-1 chromosomal localization was delayed in rec-8 ( ok978 ) , but improved as nuclei proceeded through pachytene , and by late pachytene a mixture of both long and discontinuous tracks of LAB-1 were present throughout chromosomes ( Figure 3 ) . Interestingly , LAB-1 signals were no longer associated with chromosomes in 40% ( n = 15/37 ) of −1 oocytes at diakinesis , and instead were distributed diffusely throughout the nuclei of rec-8 mutants ( Figure 3 ) . This is reminiscent of the early loss of chromosome-associated REC-8 signal observed in the lab-1 hypomorphs on metaphase I , suggesting some degree of interdependence between these proteins [30] . In coh-4 ( tm1857 ) coh-3 ( gk112 ) double mutants , LAB-1 associated with the chromosomes at early pachytene , but failed to form tracks ( Figure 3 ) . When all three meiotic kleisins were mutated , LAB-1 localization was further impaired as only very few and faint LAB-1 foci were detected on either pachytene or diakinesis chromosomes ( Figure 3 ) . This effect on LAB-1 localization was specific to members of the cohesin family , as we found no change in LAB-1 localization in mutants for either smc-5 , which is a structural maintenance of chromosomes family member , but not part of the cohesin complex , or hcp-6 , a gene encoding a member of the condensin II complex , which is structurally similar to cohesin ( unpublished data ) [50] . These results suggest that LAB-1 recruitment to the chromosomes depends on the cohesin complex , and that its association with meiotic chromosomes only completely fails when all three kleisins are absent . To gain further insight into the function and regulation of LAB-1 , we set out to identify the proteins interacting with LAB-1 . Utilizing a transgenic line expressing GFP-LAB-1 and GFP antibodies for mass spectrometry analysis , we identified 19 proteins that were specifically co-immunoprecipitated from whole worm lysates with GFP-LAB-1 , but not unrelated controls ( Table S1 ) . Prevalent among these were the axial element proteins HIM-3 ( 56 . 4% ) , HTP-1 ( 43 . 2% coverage ) , HTP-2 ( 38 . 9% coverage ) , and HTP-3 ( 15 . 8% coverage ) . All four proteins carry a HORMA domain that is also present in proteins involved in DSBR , synapsis , and mitotic spindle checkpoints from yeast to mammals [31] , [51]–[53] . Importantly , HTP-1 is the only other known long arm-specific protein in C . elegans and HTP-3 was shown to be critical for meiotic sister chromatid cohesion , as multiple cohesin members fail to load onto chromosomes in htp-3 ( y428 ) mutants [26] , [48] , [54] . To assess the functional relevance of this set of interactions we examined the interdependency between the localization of LAB-1 and the HTP proteins . The localization of HTP-1 , HTP-2 , and HTP-3 in early prophase I was not altered in lab-1 ( RNAi ) gonads compared to control ( Figure 2G , Figure S6A , and Figure S6C ) . A reciprocal analysis revealed that LAB-1 localization is indistinguishable from wild type in htp-1 gonads ( Figure S6B ) . In contrast , LAB-1 signals were not observed associated with chromosomes in htp-3 mutant germlines at any meiotic stage ( Figure 3 and Figure S6D ) . Therefore , our analysis suggests that LAB-1 forms a complex with the HORMA domain proteins , and that HTP-3 is required for LAB-1 localization either directly or through its role in cohesin loading . Our findings that lab-1 depletion results in reduced sister chromatid cohesion , and that the localization of both LAB-1 and REC-8 are partially co-dependent in late meiosis I , raise the question of whether LAB-1 could be involved in cohesin complex localization during early prophase I . Immunolocalization of either SMC-3 or REC-8 showed no differences between wild type and lab-1-depleted gonads ( Figure 4 , Figure S8 , and [55] ) . Moreover , SMC-3 localization was indistinguishable from wild type in rec-8 mutants ( Figure 4 and [26] ) . Therefore , we reasoned that as both REC-8 and LAB-1 are required for normal sister chromatid cohesion , yet seem to be dispensable for SMC-3 localization , they might work in parallel . Indeed , in lab-1 ( RNAi ) ; rec-8 worms , SMC-3 loading was significantly impaired and , as expected due to abrogation of cohesin loading , the SC central region protein SYP-1 was restricted to mostly a single large aggregate per nucleus in mid-pachytene and few long tracks in late pachytene nuclei ( Figure 4 ) . Therefore , REC-8 and LAB-1 work in parallel to enable the loading of SMC-3 and facilitate SC formation . If LAB-1 and REC-8 cooperate in SMC-3 loading during meiosis , then lack of both should increase the premature loss of sister chromatid cohesion detected at diakinesis . Indeed , the number of diakinesis oocytes carrying 13–24 DAPI stained bodies is significantly increased in lab-1 ( RNAi ) ; rec-8 gonads compared with either lab-1 ( RNAi ) or rec-8 ( 48% , 1% , and 7% , n = 46 , 88 , and 28 , respectively ) . Nevertheless , many chromatids were still held together in lab-1 ( RNAi ) ; rec-8 as demonstrated by the 52% of oocytes that had 7–12 DAPI stained bodies ( average 13 . 1±2 . 5; Figure 5 ) . This could be explained by either residual LAB-1 that was not depleted or by other factors that contribute to sister chromatid cohesion independently . To examine whether the other two meiotic kleisins contribute to sister chromatid cohesion in parallel with lab-1 , we looked at the effects of lab-1 depletion on coh-4 coh-3 double mutants . 7–12 DAPI stained bodies are observed in the diakinesis oocytes of coh-4 coh-3 double mutants ( Figure 5 and [26] ) , suggesting that sister chromatids are still held together . However , when we depleted lab-1 in these worms , the average number of DAPI-stained bodies increased from 11±1 to 15±3 ( Figure 5 , n = 30 and 12 , respectively , p<0 . 0001 by the two-tailed Mann-Whitney test , 95% C . I ) . These results suggest that LAB-1 affects both REC-8 and COH-3/COH-4 cohesin complexes . Moreover , the role of LAB-1 in SCC may be greater than suggested by these results , since this analysis relied on RNAi depletion and therefore residual LAB-1 activity cannot be ruled out . Two models for lab-1 contribution to sister chromatid cohesion can be envisioned from these results . In one , lab-1 ensures sister chromatid cohesion solely through rec-8 , coh-4 , and coh-3 . Alternatively , lab-1 can contribute to sister chromatid cohesion even in their absence . To distinguish between these two possibilities , we examined the effect of lab-1 depletion when all three meiotic kleisins are mutated . Similar to previous observations [26] , we found that most of the oocytes at diakinesis in the rec-8;coh-4 coh-3 worms had 13–24 DAPI stained bodies ( Figure 5 ) . However , the average number of bodies was only 15±3 , indicating that sister chromatid cohesion was not completely lost . When lab-1 was depleted in these worms , the number of DAPI-stained bodies increased to 18±3 , and 97% had 13–24 bodies ( Figure 5 , n = 44 and 36 for control and lab-1 ( RNAi ) , respectively , p<0 . 0001 by the two-tailed Mann-Whitney test , 95% C . I . ) . This result suggests that lab-1 acts to promote sister chromatid cohesion in parallel with rec-8 , coh-4 , and coh-3 . We have previously hypothesized that LAB-1 targets the PP1 homologs GSP-1 and GSP-2 to the long arms of the bivalents , thereby antagonizing AIR-2 localization to that region [30] . To test whether LAB-1 can directly interact with either PP1 homolog we utilized the far-western assay [56] . Bacterially expressed and purified GSP-1 and GSP-2 proteins transferred to nitrocellulose membranes bound purified LAB-1 ( Figure 6A ) . Reciprocally , LAB-1 transferred to membranes bound GSP-1 ( Figure 6A ) . These results suggest that LAB-1 binds to the PP1 homologs in vitro . A highly degenerate motif [ ( R/K ) ( V/I ) X ( F/W ) ] has been previously associated with the binding , localization , and function of PP1 phosphatases [57]–[59] . To test if the putative PP1 binding motif present in LAB-1 [30] is required for the in vitro interaction detected between LAB-1 and the PP1 homologs , we used purified LAB-1 protein either lacking the motif ( ΔPP1 ) or carrying two alanine substitutions in this motif ( KAIA ) . GSP-1 and GSP-2 proteins transferred to membranes were still able to bind both mutant LAB-1 proteins , and reciprocally , membrane-bound mutant LAB-1 proteins could bind GSP-1 ( Figure 6A ) . Thus , the putative PP1 binding motif is probably not necessary for LAB-1 binding to GSP-1 and GSP-2 . To verify the interaction between LAB-1 and the PP1 homologs we utilized the yeast two-hybrid system . Full-length GSP-1 fused to the activation domain of GAL4 interacted with LAB-1 as well as with the LAB-1 PP1 mutants fused to the DNA binding domain of GAL4 ( Figure 6B ) . Interestingly , GSP-2 could only weakly interact with the LAB-1 PP1 mutants but not with wild type LAB-1 ( Figure 6B ) . It is possible that LAB-1 binding to GSP-2 is masked in yeast cells by the host's endogenous PP1 homologs , and only when these unrelated interactions are removed , LAB-1 and GSP-2 interaction can be detected . Taken together , these results support a direct interaction between LAB-1 and the PP1 homologs . The mislocalization of AIR-2 to the long arms of the bivalents during diakinesis in lab-1 hypomorph mutants [30] prompted us to test whether depletion of lab-1 by RNAi results in changes in AIR-2 localization during early meiotic stages as well . Indeed , unlike in most control gonads ( n = 45/53 ) , in which AIR-2 signal was not observed in transition zone nuclei , clear AIR-2 patches were observed in most transition zone nuclei upon lab-1 depletion ( n = 16/30 ) ( Figure 6C , p<0 . 0005 , by the two-sided Fisher's Exact Test , 95% C . I . ) . Moreover , consistent with AIR-2 localization in early prophase I , increased histone H3 phosphorylation , a well-characterized chromosomal target for Aurora B was also observed at that stage ( Figure 6D , mean relative fluorescence = 1 . 2 , n = 284 , and 1 . 5 , n = 175 , for control and lab-1 ( RNAi ) , respectively; p<0 . 0001 by the two-tailed Mann-Whitney test , 95% C . I . ) [24] , [60] . Since LAB-1 binds the PP1 homologs and was suggested to restrict AIR-2 through GSP-2 in metaphase I [30] , we tested if depletion of lab-1 also changes GSP-2 localization in early meiotic nuclei . In control gonads , GSP-2 is found in foci throughout transition zone nuclei and the syncytial gonad ( n = 12/12 ) ( Figure 6E and Figure S9 ) . However , in most lab-1 depleted gonads ( n = 7/12 ) , the level of nuclei-associated GSP-2 foci was reduced ( Figure 6E and Figure S10 , p<0 . 05 , by the two-sided Fisher's Exact Test , 95% C . I . ) . These results suggest that LAB-1 directly targets GSP-2 to the chromatin during meiotic onset , thus restricting aberrant AIR-2 accumulation at that stage and promoting normal establishment and maintenance of sister chromatid cohesion .
The dynamic nature of chromosome interactions during the cell cycle requires the ability to establish , mobilize , and remove SCC . Since the discovery of the core components of the cohesin complex , a growing number of proteins have been found to take part in all aspects of cohesin function: proteins involved in loading cohesin , maintaining its binding , and removing it from chromosomes . The importance of SCC is highlighted in meiosis , due to the intricacy and complexity of this process . During prophase I , homologous chromosomes pair , synapse , undergo programmed meiotic DSBs and recombine [32] . Thus , while a small reduction in chromatid cohesion may still provide for normal mitotic segregation , it would be unable to support the structural requirements for the meiotic processes to proceed . For example , in the yeast smc3-42 temperature-sensitive mutant , both SC and crossover formation are perturbed and cells fail to undergo any meiotic division , even in the mitotic permissive temperature [61] . In C . elegans , lack or depletion of either cohesin members or their interactors have also been shown to alter pairing , synapsis , DSBR , and accurate chromosome segregation [26] , [55] , [62]–[64] . Another level of cohesin control must be employed in meiosis due to the requirement for protection of cohesin at specific chromosomal regions , namely at centromeres in the case of monocentric organisms and along the long arms of the bivalents in the case of the holocentric C . elegans chromosomes . Along the long arms , cohesin must be preserved during the first meiotic division , while in all other parts of the chromosomes it must be removed [6] , [29] , [65] . Without this protection , faithful chromatid segregation cannot take place during the second meiotic division . It is therefore not surprising that various different components are involved in regulating SCC . However , how these proteins modulate the way SCC is either enforced or relieved , and how these two processes are coordinated , is not completely understood . Most of the factors implicated in SCC depend on the cohesin complex , yet some reports have also suggested cohesin-independent pathways [66]–[68] . Although LAB-1 can maintain some degree of SCC even in the absence of the meiotic kleisins , most of our data support a cohesin-dependent mechanism . Here we show that LAB-1 and different cohesin members are partially interdependent in their localization throughout prophase I and at metaphase I . The cooperation between LAB-1 and cohesin to ensure SCC can be observed cytologically through the number of DAPI-stained bodies at diakinesis when meiotic kleisins are mutated . Mutations in either rec-8 or coh-4 coh-3 do not result in precocious sister separation in most nuclei , yet depletion of lab-1 in those mutants significantly increases the frequency of unbound sister chromatids at diakinesis . These results are consistent with LAB-1 acting to maintain cohesion in cooperation with the cohesin complex . Yet it is possible that LAB-1 can promote SCC in a pathway that does not require the meiotic kleisins , since we observed a significant increase in sister chromatid separation when we depleted lab-1 in the rec-8 , coh-3 , and coh-4 triple mutant . The factors taking part in this pathway remain to be uncovered . Interestingly , our preliminary analysis did not reveal increased loss of sister chromatid cohesion following lab-1 depletion in either scc-1 or smc-3 depleted backgrounds ( Y . B . T . and M . P . C . unpublished results ) . However , the lack of fully separated chromatids in the rec-8; coh-4 coh-3 triple mutant could be due to mechanisms involving yet other kleisins and/or the formation of tangles between sister chromatids . Yan and colleagues have recently reported the finding of sisters on the loose ( SOLO ) , a protein that together with stromalin on meiosis ( snm ) is required for centromere cohesion and SMC1 localization during Drosophila male meiosis [69] . We suggest that LAB-1 plays a similar role , in that it is required to both appropriately load and maintain the cohesin complexes on specific chromosomal subdomains during pre-meiotic S phase and prophase I , respectively , in C . elegans , thereby enabling the meiotic program . In this context ( Figure S11 ) , LAB-1 acts to properly load and maintain the association of cohesin complexes along the chromosome axes . Therefore , the meiotic defects observed in lab-1 ( RNAi ) gonads are diagnostic of problems in SCC . First , lack of LAB-1 perturbs homolog paring . In this state , interhomolog repair is impaired , DSBR progression is altered and many nuclei undergo apoptosis . Second , nuclei that are not eliminated by apoptosis contain univalents as well as single chromatids at diakinesis . Finally , in the bivalents where SCC is not lost in early prophase I , reduced LAB-1 on the long arm is ultimately insufficient to prevent unchecked AIR-2 loading on all chromosome axes [30] . Upon metaphase I entry , REC-8 removal occurs at both short and long arms and accurate homolog segregation fails ( Figure S11 ) . How does LAB-1 affect SCC ? Our finding that LAB-1 and HTP-3 are present in the same complex raises the possibility that LAB-1 acts to maintain SCC via HTP-3-mediated control of cohesin loading [26] . Alternatively , HTP-3 may target LAB-1 either directly or indirectly to chromosome axes , and LAB-1 in turn recruits other factors that maintain SCC . The presence of other axial element proteins in LAB-1 immunoprecipitates as well as our finding that LAB-1 directly interacts with the PP1 homologs lead us to favor the latter . We found that in the yeast two-hybrid system GSP-2 could only weakly interact with the LAB-1 protein carrying mutations in the putative PP1 binding motif , leading us to hypothesize that GSP-2 is a weaker binding partner of LAB-1 than GSP-1 . Weak or transient binding is probably the reason why we did not detect GSP-1 and GSP-2 in our LAB-1 IPs , which were done using stringent conditions . Nevertheless , lab-1 depletion affected the localization of GSP-2 , suggesting that LAB-1 indeed targets GSP-2 to early meiotic nuclei . LAB-1 localization in the germline is highly dynamic . We propose that at the onset of meiosis , the chromosomal association of LAB-1 opposes cohesin-removing proteins whose levels gradually increase during pachytene . Indeed , AIR-2 signal is only first observed in late pachytene in wild-type C . elegans [23] , whereas it accumulates in transition zone nuclei following lab-1 depletion ( this work ) . Given that AIR-2 promotes cohesin removal during mitotic prophase [24] , [70]–[74] , we suggest that a role for LAB-1 is to restrict AIR-2's ability to remove cohesin during early meiosis . According to this model , the system that protects cohesin during metaphase I , and involves LAB-1 antagonizing AIR-2 probably via the PP1s along the long arms of the bivalents , also operates throughout the early stages of meiosis to maintain SCC ( Figure 7 ) . This may be achieved by targeting the PP1s to chromosome axes and restricting AIR-2 . In late prophase I , LAB-1 is lost from part of the chromosomes , permitting the removal of cohesin at these subdomains . In this article we show that LAB-1 plays an important role in protecting sister chromatid cohesion by localizing a phosphatase ( PP1 by LAB-1 instead of PP2A by Shugoshin ) and antagonizing Aurora B phosphorylation activity . Thus , our studies have revealed key conserved principles that guide proper regulation of meiotic sister chromatid cohesion . The use of PP1 to maintain and protect cohesin , while possibly the result of a necessary co-evolution with the holocentric nature of C . elegans chromosomes , may also be required for monocentric organisms . Similar to LAB-1 , Sgo2 was found to maintain cohesion in mouse bivalents during late prophase I [75] . This raises the possibility that Shugoshin proteins in other metazoans may also play a role earlier in meiosis in establishing and/or maintaining sister chromatid cohesion . In support of this possibility , it was recently shown that PP2A has early meiotic roles , in addition to its role of protecting centromeric cohesin during metaphase I [76] . Therefore , the use of mouse meiotic conditional alleles may aid in assessing the effect of these proteins in early prophase I . In conclusion , we have shown that LAB-1 has an earlier role in regulating the establishment and maintenance of SCC . Thus , LAB-1 emerges as a central protein in the regulation of SCC , exerting this role from the start of prophase I through homolog segregation at the metaphase I to anaphase I transition .
The N2 Bristol strain was used as the wild-type background . C . elegans strains were cultured at 20 °C under standard conditions as described in [77] . The following mutations were used: LGI: htp-3 ( tm3655 ) , cep-1 ( lg12501 ) , lab-1 ( tm1791 ) , LGIV: htp-1 ( gk174 ) , rec-8 ( ok978 ) , spo-11 ( ok79 ) , LGV: coh-4 ( tm1857 ) , coh-3 ( gk112 ) , and syp-1 ( me17 ) ( [26] , [30] , [35] , [47] , [78]–[80] ) . The GFP::LAB-1::HA line has been previously described in [30] . Feeding RNAi experiments were performed at either 20 °C ( for smc-3 and scc-3 ) or 25 °C ( for lab-1 ) as described in [30] , [63] , [81] . Control RNAi was performed by feeding HT115 bacteria carrying the empty pL4440 vector . A feeding vector from the ORFeome RNAi collection [82] was used for smc-3 RNAi experiments . Successful depletions were verified at every single experiment by immunostaining with either LAB-1 or SMC-3 antibodies for lab-1 ( RNAi ) and smc-3 ( RNAi ) , respectively , and by RT-PCR for scc-3 ( RNAi ) . cDNA was produced from single-worm RNA extracts using the Thermoscript RT-PCR system ( Invitrogen ) . The effectiveness of scc-3 RNAi was determined by assaying the expression of the scc-3 transcript in at least four individual animals subjected to RNAi . Expression of gpd-1 ( T09F3 . 3 ) was used as a control . Whole mount preparation of dissected gonads , DAPI staining , immunostaining , and analysis of germline nuclei were carried out as in [41] , [83] . A rabbit polyclonal antibody against a C-terminal peptide of C . elegans GSP-2 ( TPPRNAPAAQPKKGAKK ) was generated by Sigma-Genosys . The antiserum was affinity-purified against the original peptide-antigen as described in [84] . Primary antibodies were used at the following dilutions: α-LAB-1 ( 1∶300; [30] ) , α-REC-8 ( 1∶50; Abcam ) , α-Histone H3 phospho-Ser10 ( 1∶300; Upstate Biotechnologies ) , α-HTP-3 ( 1∶100; [48] ) , α-HTP-1/2 ( 1∶200; [54] ) , α-SYP-1 ( 1∶200 [47] ) , α-RAD-51 ( 1∶100; [83] ) , α-SMC-3 ( 1∶500; Chemicom ) , α-MSH-5 ( 1∶100000; SDI ) , α-AIR-2 ( 1∶100 [30] ) , and α-GSP-2 ( 1∶100 ) . The secondary antibodies used were: Cy3 α-rabbit , Cy3 α-rat , Cy3 α-mouse , FITC α-rabbit , and FITC α-guinea pig ( Jackson Immunochemicals , 1∶200 ) . FISH was performed as in [37] utilizing a probe recognizing the left end of the X chromosome , derived from YAC Y51E2 , and a probe recognizing the right end of chromosome I derived from pooled cosmids F32A7 and F14B11 , prepared as in [36] . Immunofluorescence images were collected at 0 . 2 µm increments with an IX-70 microscope ( Olympus ) and a cooled CCD camera ( model CH350; Roper Scientific ) controlled by the DeltaVision system ( Applied Precision ) . Images were subjected to deconvolution analysis using the SoftWorx 3 . 0 program ( Applied Precision ) as in [23] . For germ cell apoptosis , worms were transferred onto a drop of M9 on 1 . 5% agarose pads on slides and assayed using the Leica DM5000 B microscope ( 100× objective ) . Mass spectrometry analysis was done as described in [84] . Specificity was further supported by analysis of >7 other C . elegans proteins precipitated in the same approach ( I . Cheeseman , personal communication and [85]–[87] ) . Control and lab-1 ( RNAi ) worms were mounted on the same slides , but either the heads or tails were dissected to distinguish between genotypes . Immunostaining and imaging were performed as described above . Images were acquired from gonads still attached to carcasses . Fluorescence intensity was measured using ImageJ . Values were normalized by dividing the fluorescence intensity level detected in a rectangular area encompassing each nucleus with the intensity level detected within the same size area of the adjacent cytoplasm . For far-westerns , 0 . 8 µg ( as determined by Micro BCA protein assay kit , Pierce Biotechnology Rockford , IL ) of each protein were resolved on 12 . 5% SDS-PAGE and transferred to nitrocellulose membranes . Membranes were stained with Ponceau S and then washed with PBS . The membranes were then blocked for 1 h in PBSTBM ( PBS containing 1% Tween 20 , 5% dry milk , and 1% BSA ) , and incubated overnight with 2 µg/ml of the appropriate binding protein diluted in PBSTBM at 4°C . After four washes with PBSTBM , the membranes were incubated for 1 h at RT with primary antibody diluted 1∶100 , 000 in PBSTMB for 1 h , washed four times with PBST , and incubated for 1 h with peroxidase-conjugated secondary antibodies diluted 1∶10 , 000 in PBST . Incubation with an N-terminus HIM-18 antibody was used as a negative control ( Figure S12 ) . Yeast two-hybrid assays were performed as in [41] . Quantitative analysis of RAD-51 foci was performed as in [83] except that eight instead of seven zones composing the germline were scored . The eighth additional zone included in this study consists of early diplotene nuclei . The average number of nuclei scored per zone ( n ) from six gonads each for control and lab-1 ( RNAi ) were: zone 1 ( n = 247 ) , zone 2 ( n = 311 ) , zone 3 ( n = 267 ) , zone 4 ( n = 204 ) , zone 5 ( n = 181 ) , zone 6 ( n = 160 ) , zone 7 ( n = 134 ) , and zone 8 ( n = 96 ) . Germ cell corpses were scored in adult hermaphrodites 18 h post-L4 using acridine orange as described in [43] . A minimum of 35 gonads were scored for each genotype . Statistical analysis was performed using the two-tailed Mann-Whitney test , 95% C . I . | A critical step for achieving successful cell division is the regulation of how the cohesin complexes that bind sister chromatids are initially deposited , then maintained , and finally removed to allow the chromatids to separate into daughter cells . This is particularly challenging during meiosis , when the sister chromatids must remain partially connected to each other through the first division . In organisms that have a single focal centromere on each chromosome , such as mammals and flies , cohesin is protected through the first meiotic division by the protein Shugoshin , which binds the PP2A phosphatase . PP2A counteracts phosphorylation by the Aurora B kinase; if certain cohesins are phosphorylated by Aurora B they become targeted for removal , which allows the chromatids to separate . In the nematode C . elegans , the chromosomes lack a localized centromere and the predicted Shugoshin homolog is not required for protection of cohesins; instead , this function is executed in metaphase of the first meiotic division by the protein LAB-1 . But it is not completely understood what leads to the deposition of cohesin prior to entry into meiosis and to its maintenance throughout early meiosis I . In this study , we show that LAB-1 is also required for the loading and maintenance of the cohesin complex . LAB-1 ensures that the chromatids are not separated prematurely , and thus enables the proper progression of events through prophase I of meiosis . We propose that LAB-1 may act at the onset of meiosis in a manner akin to Shugoshin , by recruiting the PP1 phosphatase to counteract Aurora B kinase , thereby ensuring sister chromatid cohesion . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2012 | LAB-1 Targets PP1 and Restricts Aurora B Kinase upon Entrance into Meiosis to Promote Sister Chromatid Cohesion |
To understand how excitable tissues give rise to arrhythmias , it is crucially necessary to understand the electrical dynamics of cells in the context of their environment . Multicellular monolayer cultures have proven useful for investigating arrhythmias and other conduction anomalies , and because of their relatively simple structure , these constructs lend themselves to paired computational studies that often help elucidate mechanisms of the observed behavior . However , tissue cultures of cardiomyocyte monolayers currently require the use of neonatal cells with ionic properties that change rapidly during development and have thus been poorly characterized and modeled to date . Recently , Kirkton and Bursac demonstrated the ability to create biosynthetic excitable tissues from genetically engineered and immortalized HEK293 cells with well-characterized electrical properties and the ability to propagate action potentials . In this study , we developed and validated a computational model of these excitable HEK293 cells ( called “Ex293” cells ) using existing electrophysiological data and a genetic search algorithm . In order to reproduce not only the mean but also the variability of experimental observations , we examined what sources of variation were required in the computational model . Random cell-to-cell and inter-monolayer variation in both ionic conductances and tissue conductivity was necessary to explain the experimentally observed variability in action potential shape and macroscopic conduction , and the spatial organization of cell-to-cell conductance variation was found to not impact macroscopic behavior; the resulting model accurately reproduces both normal and drug-modified conduction behavior . The development of a computational Ex293 cell and tissue model provides a novel framework to perform paired computational-experimental studies to study normal and abnormal conduction in multidimensional excitable tissue , and the methodology of modeling variation can be applied to models of any excitable cell .
One of the major challenges in the field of cardiac electrophysiology is quantifying the electrical dynamics of myocytes in the context of their environment . The electrical activity of cardiomyocytes in vivo is modulated by the other excitable and unexcitable cells to which they are coupled , as well as the complex interstitial space in which they are embedded . Making multisite measurements of the transmembrane potential is technically difficult to perform in situ and hence limited information is available to characterize the cells’ complex response to stimuli and drugs . One approach to studying excitable cells in context is to develop detailed in silico computational models of isolated excitable cells and tissues , and to use these models to infer the behavior of the real cells on which they are based . In most cases , computational membrane models are derived from experimental data obtained using various patch clamp techniques , often performed in different labs , under different conditions and often in cells from different species [1] . Moreover , because the details of the complex native tissue environment are poorly understood , most computational tissue models make use of significantly simplified representations of the native 3D tissue structure . An alternative approach for studying cells’ electrical dynamics in context with other cells is to use in vitro cell cultures . While typically limited to two dimensions and lacking a defined interstitial space , cultured cell monolayers can reproduce many features of the natural tissue through the manipulation of cell orientation , spacing and shape [2–5]; engineered monolayers have previously been used to study complex phenomenon such as conduction block , re-entry and spiral wave formation in 2D [6] . At present , these methods are limited to the use of neonatal cells as culturing of adult cardiac cells into confluent , electrically coupled monolayers has proven difficult . Unfortunately , the intrinsic currents of neonatal cells have been difficult to model as they change rapidly through development . As a result , there is currently no robust and tractable framework for the careful comparison of computational predictions with biologically analogous experimental measurements in multidimensional tissue . Kirkton and Bursac recently demonstrated the ability to genetically engineer a synthetic excitable cell line through the addition of only two ion channels ( Nav1 . 5 and Kir2 . 1 ) into the immortalized and non-excitable HEK293 cell line . They further were able to electrically enhance the intercellular connectivity of these excitable HEK293 cells ( named Excitable-293 or Ex293 cells ) by overexpressing connexin-43 gap junctions to form an excitable , engineered monolayer capable of propagating action potentials [7] . These excitable monolayers were subsequently used to study a wide range of electrophysiological behaviors such as reentry and conduction failure [8 , 9] . The limited number of ionic currents in these novel biological constructs as well as their relative stability over time ( compared to maturing neonatal cells ) suggest the strong potential to be modeled computationally with high fidelity . Despite their monoclonal origin and relatively simple electrophysiology compared to adult cardiomyocytes , Ex293 cells and tissues exhibit moderate variability in their action potential characteristics ( e . g . action potential duration , maximum upstroke velocity , etc ) and conduction properties [7 , 9] . This observed variability results from a combination of beat-to-beat variability in single cells , cell-to-cell variability within a monolayer , and variability between different tissue-cultured monolayers . While biological variability has a relatively moderate impact on macroscopic conduction under well-coupled normal conditions [10–12] , it will likely play a more significant role in experimental scenarios replicating disease states such as fibrosis , cellular uncoupling , reduced excitability , and premature stimuli [13] . As such , it is important to consider intrinsic variability when developing computational membrane models that are intended for the study of complex conduction behavior . In general , most computational models are constructed using mean experimental-derived properties , and efforts at modeling variability have been focused on regional differences ( epicardial vs endocardial , or apex vs base ) or on variability between isolated cells [14 , 15]; the concept of simulating tissues with variable membrane properties has only recently been explored [12] . In this study , we expanded on the work of Kirkton and Bursac [7] by using published experimental single cell and monolayer data to develop a computational model of the Ex293 cell that can be used for future paired experimental-computational studies . In doing so , we examined what sources of variability were required in the model of this highly-simplified excitable cell in order to reproduce experimental behavior . Our results show that the incorporation of cell-to-cell and inter-monolayer ionic conductance variation as well as inter-monolayer conductivity variation was necessary to reproduce the behavior of propagating action potentials in Ex293 monolayers over a range of experimental conditions . Moreover , we demonstrate that non-random spatial organization of cell-to-cell variation does not significantly affect macroscopic conduction , indicating that random spatial distribution of cell-to-cell ionic variation can adequately capture the impact of experimental ionic variability .
The Ex293 membrane model includes four constitutive currents: the inward rectifying potassium current ( IK1 , carried by the Kir2 . 1 channel ) and the fast voltage-gated sodium current ( INa , carried by the Nav1 . 5 channel ) , both of which are transfected into HEK293 cells to create the Ex293 line; as well as two endogenous HEK293 currents , a voltage gated sodium current [16] and a delayed-rectifier potassium current [17–19] . As described in Methods , mathematical descriptions of each current were formulated directly from previously reported mean experimental data from Kirkton and Bursac [7] and others [16 , 17] . The inward rectifying potassium current was modeled using a single activation gate and the resulting model is able to reproduce the current-voltage curve for the peak K+ current ( Fig 1A , dotted line ) as well as the time courses of potassium current elicited by a step change in membrane potential at room temperature ( Fig 1B ) . The transfected voltage gated sodium current was modeled with three identical activation gates , and one inactivation gate with fast and slow components . At room temperature ( 23°C ) , the model is able to recapitulate the experimentally observed current-voltage curve for the peak Na+ current ( Fig 1C , dotted line ) as well as the dynamics of the sodium current in response to step changes in membrane potential ( Fig 1D ) . Because the published electrophysiological data provides insufficient information to fully specify the Ex293 membrane model at 35°C , a multiobjective genetic search algorithm was used to determine optimal values for several free parameters in order to match experimentally observed conduction properties . Eight free parameters were fit using the genetic search technique , including the maximal current densities for each of the four currents and the bulk tissue conductivity ( see Table 1 for all fitted parameters ) . For each trial parameter set , conduction was simulated in a two dimensional continuous monodomain ( Fig 2A ) using the Cardiowave system [20] , and two error functions: ( 1 ) the root mean square error of the simulated action potential compared to a representative experimental action potential , and ( 2 ) the absolute error in simulated conduction velocity ( CV ) were calculated . The genetic algorithm was executed nine times , yielding different solutions , and the parameter set from each Pareto front that minimized the mean square action potential error was selected . Each run of the genetic algorithm required an average of 112 generations to converge; the mean parameter values identified by the genetic algorithm across multiple runs are shown in Table 1 . While the genetic algorithm searched a large parameter space , multiple runs resulted in parameter estimates with relatively little variability , indicating a strong likelihood that the fit results recreate the biological behavior . The parameter set that provided the lowest mean square error of action potential fit was used in the 35°C model . Fig 2B shows that action potential generated by these parameter values reproduces the representative experimentally recorded action potential with a high degree of accuracy , with a root-mean-square error of 1 . 109 mV and a CV error of 0 . 0001 cm/s . In addition , the simulated action potential replicates several other metrics characterizing experimental action potentials obtained from multiple cells ( S1 Table , Columns 1 and 2 ) . Based on the genetic algorithm fits , the change in temperature from 23°C to 35°C resulted in an approximately 3-fold increase in the INa maximum conductance , and a 50% increase in the IK1 maximum conductance , which is in line with previous studies [21–23] . The total temperature-induced changes in the current-voltage relationships of the INa and IK1 currents , due to both temperature-dependent conductance changes ( as determined by the genetic algorithm fits ) and temperature-dependent shifts in activation and inactivation ( as determined based on previous studies ) , are shown in Fig 1A and 1C ( dashed lines ) . The model qualitatively matches the net ionic current recorded experimentally during single-cell action potential ( AP ) clamp recordings at 23°C ( S1A and S1C Fig ) . At 35°C , the model shows temperature induced changes in ionic currents ( S1D Fig ) including a substantial increase in inward sodium current . Examination of the role of the individual currents shows that the INa current is responsible for rapid depolarization while the IK1 current resists depolarization and is responsible for rapid repolarization . In addition , the model suggests that an endogenous HEK293 potassium current plays an important role in the repolarization of the Ex293 cell as its low outward current gives the plateau phase of the action potential its shape , and lowers the membrane potential from peak voltage ( ~ 20 mV ) to a voltage where the transfected IK1 current activates and initiates the rapid depolarization phase . In contrast , a relatively small endogenous sodium current within HEK293 cells appears to only play a minor role in the Ex293 action potential . To replicate the observed experimental variability in action potential and conduction properties , variation of ionic conductances and of tissue conductivity was incorporated into the model by scaling model parameters by a factor randomly selected from a normal distribution with a mean of one and a specified standard deviation . Action potential shape properties were measured from a single location in the monolayer ( i . e . a single node in the computational grid ) , to compare to experimental data from sharp intracellular electrode recordings; conduction properties were measured by simulated optical mapping ( see Methods ) of 2-D computational monolayers . Experimentally observed variability in [7] was attributed either entirely to cell-to-cell variation ( each cell has slightly different properties ) , entirely to inter-monolayer variation ( each monolayer has slightly different properties ) , or to a combination of both types of variation . The addition of cell-to-cell ionic conductance variation to the model with standard deviations of variation as high as 0 . 50 led to a small degree of variability in single cell maximum upstroke velocity that was insufficient to match that seen experimentally ( variances unequal by Levene’s test , p < 0 . 01 ) and almost no variability in either single cell action potential duration ( APD ) or macroscopic monolayer CV ( Fig 3 , Column 2 ) . Cell-to-cell variation was then eliminated and all experimentally observed variability was instead modeled as due to inter-monolayer current variation . Random normal variation of the ionic conductances for each monolayer with a standard deviation of 0 . 125 approximately replicated variability in single cell APD; however , insufficient variability was seen in mean monolayer CV ( p < 0 . 01 ) , and the lack of cell-to-cell variation resulted in insufficient variability in maximum upstroke velocity compared to experimental observations ( p < 0 . 05 ) ( Fig 3 , Column 3 ) A combination of inter-monolayer conductance variation and cell-to-cell conductance variation ( termed “dual variation” ) was explored . Random normal variation with standard deviation of 0 . 125 was used to select each monolayer’s mean conductances , and within each monolayer , further random normal variation with standard deviation of 0 . 125 was used to select ion channel conductances of individual tissue nodes ( Fig 4B ) . The resulting conductances , when pooled across multiple monolayers , were normally distributed a coefficient of variation of 0 . 177 , comparable to that reported for current density in isolated cells ( 0 . 22 for IK in Ex293 [7] , 0 . 13 for the endogenous potassium current in HEK293 [17] ) , and well within the range of intraclonal protein expression variation seen in monoclonal cell lines [24] . This formulation was found to result in simulated variability that approximately matched experimentally observed single cell upstroke velocity variability , as well as experimental APD variability; however , dual variation failed to capture the degree of variability seen in the mean monolayer CV ( p < 0 . 05 ) ( Fig 3 , Column 4 ) . Inter-monolayer conductivity variation ( i . e . , variation of bulk tissue conductivity for each monolayer ) was then added to the dual variation model to yield “triple variation” . Each monolayer’s tissue conductivity was perturbed with random normal variation with standard deviation of 0 . 25 . Ultimately , this combination of model parameter variation was able to faithfully replicate the degree of variability observed experimentally in macroscopic conduction properties ( mean CV and APD ) as well as the variability in experimental single cell action potential properties ( APD and maximum upstroke velocity ) ( Fig 3 , Column 5; and S1 Table ) . The effects of each type of variation on measured variability are summarized on Fig 4A . While cell-to-cell variability is likely to remain constant between experimental studies , the degree of experimental inter-monolayer variability , which depends on culture conditions , initial cell seeding and monolayer confluence , will vary between experimental preparations . Indeed , a moderate decrease in mean monolayer APD and CV variability during 2Hz pacing was noted between experimental results in [7] and later studies in [9] . A 50% reduction in the inter-monolayer variability of ionic conductances and tissue conductivity was necessary to match the experimental variability noted observed in [9] . A sensitivity analysis was performed to understand how variation in each of the Ex293 conductances ( see Eqs 3 , 5 , 7 and 8 and S2 Table ) affect macroscopic conduction properties , and to ensure that small perturbations in model parameters led to physiologically reasonable behavior . Maximum channel conductances in monolayers with no variability were scaled from 0 . 5 to 1 . 5-fold , individually . Perturbations in the exogenous sodium current ( INa ) had a substantial , non-linear effect on CV and APD ( S3 Fig ) , with a 40% decrease in conductance leading to a 35% decrease in CV and 24% decrease in APD , while a 40% increase in conductance led to 17% conduction speeding and a minimal increase ( 6% ) in APD . The effect of perturbations in the exogenous potassium current ( IK ) on CV was nearly linear ( R2 = 0 . 995 ) with a slope of -0 . 17 ( % change in CV / % change in conductance ) , while the effects on APD were drastically non-linear ( 40% increase and decrease in conductance lead to 25% decrease and 79% increase in APD , respectively ) . The effect on APD is consistent with the dominant role of the INa and IK1 currents in the upstroke and downstroke of the action potential , respectively . In addition , during the action potential upstroke , the inward INa current that causes depolarization is opposed by the outward IK1 current , which drives the membrane potential towards rest; the effects of perturbations in INa and IK1 on CV are consistent with the roles of these currents during the action potential upstroke . In contrast with the effects of perturbation of the exogenous currents , perturbations in endogenous currents ( IK , wt and INa , wt ) substantially affected APD without significantly altering CV . This is consistent with the activity of the IK , wt and INa , wt currents in the plateau and early repolarization of the action potential and the lack of activity during the depolarization and late repolarization . The Ex293 model was validated over a range of experimental conditions that had been studied previously [7–9] . For example , Kirkton and Bursac measured the CV and APD restitution by examining the response of a monolayer to a premature stimulus delivered following pacing of the monolayer at a constant rate . The restitution properties were obtained in the model by stimulating a strip of tissue at 2Hz ( S1 ) and applying a premature stimulus ( S2 ) at incrementally earlier times . Fig 5A shows that baseline mean Ex293 model is able to reasonably reproduce the experimentally observed CV restitution ( Fig 5A ) and APD restitution ( Fig 5B ) curves , with R2 values of 0 . 97 and 0 . 82 , respectively . The significant variability noted in experimental APD restitution curves [9] is approximately replicated by the inclusion of cell-to-cell and inter-monolayer variability in the Ex293 model ( Fig 5 , dashed lines represent one SD above and below the mean ) . Kirkton and Bursac also explored the effects of the ion channel blockers tetrodotoxin ( TTX ) and barium chloride ( BaCl2 ) on the Ex293 cells . Because both the exogenous and endogenous sodium currents are sensitive to TTX [7 , 16] , simultaneous perturbation of both sodium channels is analogous to application of TTX in an experimental preparation . TTX blocks sodium current in a dose dependent but not voltage sensitive manner [25] . While suppression of the sodium currents in the model cannot be correlated with a specific TTX dose in the absence of an Ex293 TTX dose-response curve , the effect of simulated sodium channel blockade ( Fig 6B ) is qualitatively similar to the response of Ex293 monolayers to TTX in Kirkton and Bursac ( Fig 6A ) [7] , with increasing sodium block leading to accelerating decreases in CV and APD until conduction failed when sodium conductance was reduced by more than 45% ( Fig 6B ) . Barium chloride acts as a blocker of some potassium channels: while the transfected Kir2 . 1 potassium channels are sensitive to barium chloride [7] , the endogenous potassium currents in HEK-293 cells are not affected [26]; therefore , reductions in the transfected potassium conductance are analogous to treatment of Ex293 cells with BaCl2 . However , barium chloride blocks potassium current not only in a dose-dependent manner , but also in a voltage-sensitive manner [27–29] . Based on data from published figures from [27–29] , a description of barium chloride block was incorporated that reflects the variation in block as a function of membrane potential ( S4 Fig ) . Using this model , our simulated Ex293 monolayers behave similarly to experimental monolayers when exposed to barium chloride ( Fig 6C and 6D ) . As the potassium current is blocked , there is an increase in CV up to 27 . 1 cm/s due to decreased outward current ( i . e . , that opposes depolarizing inward sodium current ) during the upstroke of the action potential . When the potassium current at low membrane potentials is decreased by more than 75% ( corresponding to a decrease by 26% at 0 mV ) , conduction slowing occurs , due to sodium channel inactivation as the resting potential rises . Simulated barium chloride-induced reduction in potassium current also results in a up to 20-fold monotonic , exponential increase in APD , as shown by Kirkton and Bursac [7] . Finally , because the IK1 conductance is dependent on extracellular potassium concentration ( [Ko] ) , the effect of varying this concentration was also examined . While neonatal rat ventricular myocytes require a 12 mM increase in extracellular potassium before conduction fails , experimental Ex293 monolayers experience conduction block when extracellular potassium is increased by as little as 2 mM from 5 . 4 mM to 7 . 4 mM [30] . In our simulated Ex293 monolayers , increases in extracellular potassium result in conduction slowing by up to 35 . 2% with a 1 . 6 mM increase in [Ko]; any further increase in extracellular potassium results in a failure to fire and propagate action potentials ( Fig 7 ) . This effect is due to a combination of increased IK1 conductance [31 , 32]; changes in the IK1 driving force due to an raised potassium reversal potential; and sodium channel inactivation due to elevation of the resting membrane potential . The threshold for causing conduction failure is lower than that observed experimentally . However , when the model is adjusted such that the change in resting membrane potential due to changes in extracellular potassium are not perfectly Nernstian ( based on data from Bailly et al . [32] , using a change in reversal potential of 51 mV per decade change in [K+]o , rather than 58 mV as predicted by the Nernst equation ) , the extracellular potassium concentration must be increased by at least 2 . 2 mM to trigger conduction failure , comparable to the threshold observed experimentally . Because monolayers and in vivo tissues arise from a smaller number of parent cells that divide and grow to confluence , it is possible that cell-to-cell level variability is spatially organized rather than randomly distributed . The impact of this spatial organization of variation on macroscopic conduction was examined using an idealized scenario with a central region of prolonged APD ( due to reduced mean IK1 conductance ) and reduced cell-to-cell variation . The distribution of ionic properties in the remainder of the monolayer was adjusted such that the overall distribution of IK1 conductance was not altered . The addition of the central region of prolonged APD and reduced variance does not significantly affect APD or CV during 1Hz pacing ( Fig 8B ) . In addition , no statistically significant change in restitution behavior or in minimum viable S1-S2 interval were observed ( Fig 8C–8E ) . Because it is well established that cell-to-cell variation is masked by strong coupling of the tissue , we examined how spatial organization of variation affected macroscopic conduction in tissue with functional decoupling via simulated , idealized fibrosis . Fibrosis was simulated via the addition of a regular field of non-conductive obstacles ( Fig 8A ) , which results in a 18 . 1% reduction of mean CV and 6 . 0% reduction in mean APD during 1 Hz pacing ( Fig 8B ) , as well as exaggerated conduction slowing at shortened diastolic intervals below 300 ms when compared to non-fibrotic tissues ( normalized to 1Hz CV ) ( Fig 8D ) . In addition , the tissue with regular fibrosis-like obstacles exhibits increased variability in conduction failure behavior compared to the control tissue: only 2 . 5% monolayers fail to conduct at an S1-S2 interval longer than 100 ms in the control case , while in the fibrotic monolayer , 25% fail at S1-S2 intervals greater than 100 ms ( p < 0 . 05 ) . The addition of the central region with reduced variation and prolonged APD to the fibrotic tissue results in no significant change in macroscopic conduction behavior at 1Hz or in behavior at shortened diastolic intervals , beyond those observed in the fibrotic tissue alone ( Fig 8B , 8C and 8D ) . Further , the addition of a central region of prolonged APD and reduced variance does not further increase failure variability ( 30% failure rate at S1-S2 interval greater than 100 ms vs 25% in fibrotic tissue without central region ) .
Various groups have reported extensively on the presence of endogenous potassium ( [17–19 , 33–35] ) , sodium ( [16 , 36] ) , calcium ( [37 , 38] ) and chloride ( [34] ) currents in HEK293 cells [39] . In our model , we chose to include two of these currents: the endogenous potassium current , which has been implicated in allowing spiking behavior in HEK293 cells expressing exogenous sodium channels [40]; and the endogenous sodium currents of relatively large magnitude whose presence would affect plateau behavior of the action potential . An in-depth discussion of the description of individual membrane currents is presented in Supplementary Text 1 . We have shown that Hodgkin-Huxley style currents definitions , as built using published literature current descriptions and adapted to physiological temperatures , are sufficient to faithfully reproduce both the single-current ( Fig 1 ) and whole-cell properties ( S1 Fig ) seen in experimental recordings . In our studies , the membrane model’s free parameters ( including current densities of each current and temperature-dependent effects on the endogenous potassium currents ) were fitted to match a representative action potential recorded in a tissue cultured monolayer using a microelectrode , by coupling a multiobjective genetic search algorithm [41–43] with 2-D continuous model simulations . While membrane models have traditionally been constructed in the context of a single isolated cell , several groups have recently attempted to fit a propagating action potential rather than one obtained from an isolated cell in order to account for the electrotonic coupling of neighboring cells [44 , 45] . Kaur et al . demonstrated that two sets of parameters that generate nearly identical action potentials in an isolated cell model can generate drastically different action potentials in a model of 2-D propagation [42] . In this work , we utilized an approach of fitting both the action potential morphology and the tissue CV simultaneously by searching for the optimal membrane free parameters and the tissue bulk conductivity . Recently , Johnstone et al . showed that fitting with a single action potential was sufficient to accurately estimate up to 6 channel conductances [46] , and as such , we believe that our methodology that faithfully reconstructed the Ex293 action potential shape and CV ( Fig 2 , S1 Table ) also allowed us to accurately estimate true channel conductances . Variability in ionic currents can lead to changes in steady state macroscopic conduction properties , as well changes in dynamic behaviors such as restitution [47] , making its incorporation into computational models critical to accurately predict behavior under arrhythmogenic conditions . Variability in cardiac electrophysiological models has typically been considered on a regional basis ( i . e . atria vs ventricles , epicardial vs endocardial [48] ) ; more recent work has recognized the important of modeling other sources of variability including beat-to-beat variation , cell-to-cell variation , and inter-subject variation , as each of these impact measured electrical variability ( Fig 8A ) . Several studies have developed populations of cell models with cell-to-cell variability where model parameters are distributed in a range around the original model parameter values [14 , 49] . However , the majority of these efforts have focused on modeling and studying isolated single cells , and studies that have developed models of tissues with cell-to-cell variability have generally introduced variability to only a single current [10 , 50] . Recently , Walmsley et al . conducted a more comprehensive examination of the effects of beat-to-beat variability and cell-to-cell variability in two dimensional tissue simulations and found that while the effects of both are muted in well-coupled tissues , the effect of cell-to-cell variability predominates as tissue coupling is reduced [12] . Finally , it is well-known that electrophysiological properties vary between subjects , and models have been developed to capture this variability by recreating experimental action potentials from different subjects [13 , 43 , 51] , but no models known to us have used the combination of inter-subject and within-subject variation to explain experimentally recorded variability . In order to recreate previously reported experimental variability in Ex293 behavior , we sought to identify the type and degree of variation that was required in the Ex293 model [7 , 9] . We chose to model only cell-to-cell variation and inter-subject variation because the effect of beat-to-beat variation is largely masked by that of cell-to-cell variation [12 , 52] . Inter-subject ( with monolayers being considered as different subjects ) variation of ionic currents could result from small differences in culture conditions , as well as from variability in the properties of cells used to initially seed each monolayer . In addition to inter-monolayer ionic conductance variation , we also considered inter-monolayer conductivity variation , which would result from variability in monolayer confluence and degree of coupling at the time of experimental recordings . We found that only combined “triple variation” in cell-cell conductance , inter-monolayer conductance , and inter-monolayer conductivity was necessary and sufficient to replicate experimental variability in single cell properties , and mean monolayer APD and CV ( Fig 3 ) , as well as several other properties including resting membrane potential and action potential amplitude ( S2 Table , S5 Fig ) . The need for multifactorial variation is congruent with analysis of experimental intra-monolayer variability ( S2 Fig ) . Experimental variability in APD within a single monolayer is substantially smaller than the variability between monolayers , indicating that the strong coupling within each monolayer masks any inherent cell-to-cell APD variability . However , the degree of cell-to-cell upstroke velocity variability within a single monolayer is comparable in magnitude to inter-monolayer upstroke velocity variability , indicating that both forms of variation are essential in recreating overall experimental variability . While a model that only considered inter-monolayer variability could feasibly reproduce the range of macroscopically observed conduction behavior under normal conditions , small differences in local upstroke velocity could be crucially vital to determining macroscopic behavior ( i . e . whether conduction block occurs ) under critical regimes of conduction such as reduced excitability and poor coupling . While we incorporated variation as a normal distribution around the baseline model parameters , others have previously used uniform variation on the range of [-100% , +100%] and performed a post-hoc screening to identify those models whose action potential properties fall within predefined , experimentally-based inclusion criteria [15] . This approach is useful for studying the relative contributions of each current and the interactions between currents , but the post-hoc screening step makes it challenging to use this approach for the generation of numerous tissue models with randomly generated cell-to-cell variability . Instead , we chose to vary each model parameter around a baseline value determined using a representative action potential recording , as described above , with a degree of variation selected to match the distribution , rather than the range , of output properties . Both this method , and the uniform distribution method with post-hoc selection resulted in a normal-like distribution of output properties ( S5 Fig , and Fig 3 of [15] ) . While our method of normally distributed variation might be less suited for mechanistic or sensitivity analysis , it allows for the simulation of tissues where the properties of each node are randomized and determined at the time of simulation initiation without the need for calibration or screening . In order to use computational simulations and derive meaningful conclusions from their results , the underlying model must be validated under conditions beyond a single action potential measured at a regular pacing rate . It has previously been noted that variation in channel conductance can lead to significant variation in restitution behavior [47] and this effect is clearly seen in experimental restitution behavior . The Ex293 membrane model with triple variation was able to closely match both the mean degree and the variability of experimental restitution behavior ( Fig 5 ) with CV restitution being recapitulated more closely than APD restitution . We note that the model CV restitution curve is steeper than that of APD restitution , a behavior that is seen across other experimental and computational studies; however , the Ex293 experimental APD restitution curve is much steeper than the CV restitution curve , suggesting that additional study of this anomalous relationship is warranted . In addition to modeling restitution properties , we examined how tetrodotoxin ( TTX ) , barium chloride ( BaCl2 ) , and the extracellular potassium concentration ( [K+]o ) affect conduction in simulated Ex293 monolayers compared to experimental observations . The simulated response of the Ex293 model to TTX and BaCl2 was qualitatively similar to that in experimental monolayers ( Fig 6 ) . Under BaCl2 treatment , however , the minimum achieved CV before failure was lower experimentally than in the model , and consequently , the experimental results exhibited longer APD than the model . This is likely because the discrete nature of experimental monolayers leads to local variation in conductivity that allows for propagation of a slow , non-planar wavefront that cannot occur in a continuous representation of tissue . We also note that direct comparison of experimental and simulated results is not possible in the absence of a dose-response relationship; as such differences in the shape of the BaCl2 response curves may be due to scaling of figure axes , or to uncertainty in voltage-dependent model of BaCl2 block . Furthermore , the Ex293 model and experimental monolayers behave similarly when subject to increased [K+]o , although the model failed to conduct at lower [K+]o than experimental monolayers . This may be because the changes in reversal potential due to changing [K+]o are slightly less than predicted by the Nernst formula [32] , which would lead to less sodium inactivation and facilitate conduction in cases where the model predicted failure . Such modification of the model resulted in a [K+]o failure threshold comparable to that observed experimentally , suggesting that modification of biophysical models to include “real-world” behavior is necessary in order to faithfully simulate the complexity of experimental preparations . The importance of cell-to-cell ionic conductance variation in reproducing experimental variability raises the question of whether ionic properties should be varied randomly across a tissue or whether a more complex spatial organization of variation is needed . Because in vitro and in vivo tissues develop through the repeated growth and division of an initial population of cells , it is conceivable that each tissue contains regions of reduced variance due to common cellular lineage , paracrine effects , and local metabolic conditions . We thus analyzed an extreme case of spatial organization , where the central region of the tissue exhibits reduced IK1 conductance and reduced ionic conductance variance ( Fig 8A ) and found that the presence of a spatial organization of variation did not significantly impact conduction behavior in well-coupled monolayers ( Fig 8B–8D ) . Because the strong coupling can mask effects of cell-to-cell variation , the impact of functional decoupling of the monolayer through simulated fibrosis was examined . Simulated conduction in fibrotic tissue with random cell-to-cell variation showed conduction slowing at basal pacing rates , a further exaggerated slowing at shorter diastolic intervals , and an increase in variability in the minimum diastolic interval able to sustain conduction across the tissue ( Fig 8B–8E ) , compared to control well-coupled tissue , in line with previous studies [53 , 54] . The addition of spatial organization of variation to fibrotic tissue had no significant impact on macroscopic conduction at 1 Hz pacing ( Fig 8B ) , or at shortened diastolic intervals ( Fig 8C and 8D ) . In addition , conduction failure behavior in these tissues was similar to fibrotic tissues without a central region ( Fig 8E ) . These results suggest that cell-to-cell variation can be incorporated randomly into a tissue model without the consideration of the spatial distribution of that variation . In addition , while it is clear that fibrosis increases arrhythmogenic potential by slowing conduction and inducing premature conduction failure , it appears unlikely that the presence of fibrosis unmasks any additional pro-arrhythmogenic effect from the variability of cellular properties of the underlying tissue . This study describes a new computational model of the engineered excitable Ex293 cell that reproduces experimentally observed behavior in a range of normal and abnormal conduction conditions . We have identified the key components of experimental variability that are necessary to include in the model–namely , inter-monolayer conductivity variation , and cell-to-cell and intra-monolayer ionic conductance variation—and implemented a simple yet novel method of stochastic normal random variation to allow for the simulation of the full range of experimental outcomes rather than simply the mean . While experimental approaches are limited in their ability to simultaneously gather data at high spatial and temporal resolution , computational simulations can provide tissue-wide high resolution recordings that can help elucidate the subcellular electrophysiological mechanisms behind observed macroscopic behavior . As such , we believe that this paired experimental-computational platform will enable unique future insights into the effects of microstructural variation on microscopic and macroscopic impulse conduction The computational model of Ex293 cell was developed under the assumption that the tissue is a continuum rather than discrete structure with individual cells and sub-cellular regions . Such a formulation may fail to capture the effects of changes in tissue properties or electrical behavior that occur on the spatial scale of individual cells ( for example , [55] ) . However , the validation studies were performed in well-coupled tissue and most tissue structural changes either occur on larger spatial scales ( e . g . collagen deposition ) or can be simulated via alteration of local properties in the continuous model ( e . g . tissue decoupling , cell death etc ) . In addition , the Ex293 model can easily be incorporated into a discrete model of tissue structure , as previously described by our group [56] , if necessary for further replication of experimental observations . Our model incorporates only two of the endogenous currents that have been identified in HEK293 cells—the endogenous potassium and sodium currents . Chloride currents were excluded from the model because they are poorly characterized under physiological conditions , and there remains significant uncertainty as to their rectification behavior , calcium dependence , and peak density . In addition , we chose not to include endogenous calcium currents because of their relatively small magnitude compared to the other currents [35] . The voltage dependence of the described endogenous calcium currents appears qualitatively similar to that of the endogenous and transfected sodium currents [37] , and given the transfected sodium current conductance in our model is approximately 100x the literature-reported conductance of the endogenous calcium current , exclusion of the calcium current from the model likely had minimal effect , even in cases of partial sodium channel blockade . While the degree of channel conductance variation is likely different for each channel type , we considered a single degree of variation for all channels because of limited information into each channel’s variability . In addition , our model does not incorporate cell-to-cell variability in tissue conductivity . While the variations used in this work were able to reproduce the variability seen in experimental results , the degrees of inter-monolayer variation in different experimental set will need to be modified to match specific experimental variability due different culture conditions , monolayer seeding and handling .
The excitable cell membrane of a single cell is modeled by the following differential equation: CmdVmdt = − ( Istim+Iionic ) ( 1 ) where Cm is the membrane capacitance , Vm is the transmembrane voltage , Istim is the externally applied stimulus current , and Iionic is the sum of the individual ionic currents ( whose dynamics are described by a system of ordinary differential equations ) that contribute to the action potential: Iionic = INa+ IK+ INa , wt+ IK , wt ( 2 ) A multi-objective genetic search algorithm was used to identify the optimal ion channel conductances , tissue conductivity and other free parameters necessary to reproduce the experimental action potential waveform and CVs in Ex293 monolayers . An initial population of 600 “parent” parameter sets was generated by randomly choosing values from the physiologically reasonable search range for each parameter ( Table 1 ) . A custom MATLAB script ran the 2-D tissue simulation for each trial parameter set ( see Numerical Methods ) , and computed the values of the two error functions: the root mean squared error ( RMSE ) between an experimental AP recording and the simulated AP , and the absolute difference between the experimental and simulated CV . 2-D simulation was performed in a 140 node x 140 node monolayer ( dx = dy = 50 micron ) with no-flux boundary conditions . The tissue domain was paced at 1Hz at one corner and the action potential tracing of the third action potential from a node 0 . 6 cm diagonally from the stimulus site was recorded . CV was determined from activation times ( 50% AP amplitude ) at nodes 0 . 2 cm and 0 . 8 cm from the stimulus site . RMSE was calculated by aligning the experimental action potential ( recorded via sharp electrode in [7] ) and simulated action potentials at the upstroke crossing of -40 mV . The multi-objective genetic algorithm was configured to run in parallel across 32 cores using MATLAB’s Parallel Computing Toolbox and Global Optimization Toolbox [60] , with heuristic crossover , tournament selection and a small Pareto fraction , and terminated when the average change in the spread of the Pareto front was less than 0 . 001 over 50 generations . The genetic algorithm was run 9 times to examine diversity of results; the parameter set that minimized the RMSE was selected from each Pareto front . In order to compare macroscopic conduction properties of simulation results with those obtained experimentally using a voltage sensitive dye and an optical fiber recording array , averaging of local simulated membrane potentials in small circular regions was performed to simulate optical recording of model results . While the experimental optical fiber array has fibers with diameter of 750 μm arranged in a 20-mm diameter hexagonal bundle , spacing between the optical fiber array and the tissue sample results in a wider effective field of view for each fiber . As a result , while our simulated optical sensors were spaced with 750 μm center-to-center spacing , each sensor averaged potentials over a circular region of diameter 1100 μm . The resulting voltage traces for each of the 504 simulated optical sensors was analyzed using custom MATLAB software developed for analysis of experimental optical mapping recordings , and the CV and APD80 ( action potential duration at 80% repolarization ) were calculated , as previously described [5 , 7] . Inter-monolayer variation of conductances and conductivity was generated by selecting scaling factors for each monolayer from random normal distributions with mean 1 and specified standard deviation . Base model parameters were then scaled by these factors and the new monolayer-specific mean parameters were provided to the Cardiowave simulator . Cell-to-cell variability was incorporated directly into the membrane model . Each node’s channel conductance is randomly selected at the start of the simulation from a random normal distribution with the selected mean monolayer channel conductance and a specified standard deviation ( see Fig 4B ) . Simulated variability was compared to experimentally variability measured in [7] including macroscopic CV recorded via optical mapping in n = 39 independent monolayers , and single cell APD and maximum upstroke velocity recorded via sharp electrode recording in 6 different ( biological replicate ) monolayers ( n = 4–5 cells per monolayer; 27 total recordings ) . The sensitivity of the membrane model to perturbation was assessed by independently altering the conductance of each current in a range from 50% to 150% , and measuring the impact on conduction properties . Sensitivity was measured in a simulated strand of tissue ( 600 x 10 nodes; dx = dy = 20 micron , no-flux boundary conditions ) . In addition to the macroscopic conduction properties measured using simulated optical sensors , several addition properties were measured using the action potential traces from the center-point of the strand , analogous to experimental sharp electrode recording . These properties included the maximum upstroke velocity ( dVm dt-1 ) , resting membrane potential , APD and action potential amplitude . Restitution of APD and CV was measured in a 2D strand continuous monodomain model ( 600 x 10 nodes; dx = dy = 20 micron ) . The standard ( S1-S2 ) protocol was used wherein the strand was stimulated from one end at 2 Hz for 10 pulses ( S1 ) followed by a premature stimulus of the same amplitude ( S2 ) , and the CV and APD resulting from the S2 stimulus were recorded . The S1-S2 interval was decreased until the S2 pulse no longer elicited an action potential . Pharmacological channel blockade due to tetrodotoxin ( TTX ) was simulated by simultaneously altering the conductance of both the INa and INa , wt currents . Blockade of the IK1 channel due to BaCl2 was simulated by scaling the IK1 conductance based on the degree of block at -100 mV and the membrane potential ( S4 Fig ) . The effect of varied extracellular potassium concentration was simulated by scaling the sodium conductance proportional to the square root of extracellular potassium , as included above in eq 3 , and by altering the reversal potential of potassium , as predicted by the Nernst equation . The physiological intracellular potassium concentration was estimated by assuming that the resting potential during sharp electrode recordings of HEK cells transfected with only the Kir2 . 1 channel ( with control extracellular potassium concentration of 5 . 4 mM ) is equivalent to the potassium Nernst potential [7] . The reversal potential as a function of extracellular potassium is then calculated using this estimate of intracellular potassium concentration . Conduction was simulated in a 2D continuous monodomain model ( 600 x 200 nodes; dx = dy = 10 micron ) . Non-conductive fibrosis-like obstacles were added to the tissue domain , with each obstacle 100 μm x 100 μm in size and with 100 μm spacing between obstacles . Obstacles were decoupled from neighboring tissue nodes to establish no-flux boundary conditions . A central region of homogeneity with a diameter of 140 μm ( 70% of strand width ) was established . Within this region , the mean potassium conductance was one standard deviation ( 12 . 5% ) below the monolayer mean , and the standard deviation of cell-to-cell variation was narrowed to 0 . 0625 for all ionic properties to create relative homogeneity with preserved but reduced cell-to-cell variability . Cell-to-cell ionic variation in the remainder of the tissue was resampled in order to preserve the distribution of cell-to-cell variation across the full tissue . The central region remained fully coupled to the surrounding tissue . The previously described standard S1-S2 protocol was used to assess restitution behavior . All simulations were performed using the Cardiowave software package [20] , a cardiac simulation system that incorporates numerous modules for various membrane models , time integration methods and linear solvers ( available online at cardiowave . duke . edu ) . The governing equations were discretized using finite differences and propagation was simulated using a semi-implicit Crank-Nicholson scheme with adaptive time steps between 5 μs and 100 μs . A biconjugate gradient stabilized method solver with tridiagonal preconditioner was used to simulate each time-step . Potentials were recorded at intervals of 10 μs at selected individual points and across the domain using spatial averaging across simulated optical sensors , as described earlier . All data is presented as mean +/- standard deviation unless otherwise specified . Comparison of model and experimental variability was performed using Levene’s test for equality of variances with an alpha value of 0 . 05 . Comparison of macroscopic CV , APD and failure behavior in the presence of fibrosis and spatial organization of variability was performed using a two-way ANOVA with two between-subjects measures ( fibrosis and organization ) . Comparison of restitution behavior was performed using three-way ANOVA with one within-subjects measure ( S1-S2 interval ) and two between-subjects measures ( fibrosis and organization ) , and post-hoc pairwise comparison was performed using Fisher’s LSD . | One of the major challenges in trying to understand how arrhythmias can form in cardiac tissue is studying how the electrical activity of cardiac cells is affected by their surroundings . Current approaches have focused on studying cardiac cells in vitro and using computational models to elucidate the mechanisms behind experimental findings . However , tissue culture techniques are limited to working with neonatal , rather than adult , cells , and computational modeling of these cells has proven challenging . In this work , we have a developed a new approach for conducting paired experimental and computational studies by using a cell line engineered with the minimum machinery for excitability , and a computational model derived and validated directly from this cell line . In order to create a model that reproduces the diversity , rather than simply the average behavior , of experimental studies , we have incorporated a simple yet novel method of inherent variability , and explored what types of experimental variation must be incorporated into the model to recapitulate experimental findings . Using this new platform for paired experimental-computational studies with inherent variability , we will be able to study and better understand how changes in cardiac structure such as fibrosis and heterogeneity lead to conduction slowing , conduction failure , and arrhythmogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"applied",
"mathematics",
"membrane",
"potential",
"fibrosis",
"electrophysiology",
"genetic",
"algorithms",
"neuroscience",
"simulation",
"and",
"modeling",
"algorithms",
"developmental",
"biology",
"probability",
"distribution",
"mathematics",
"research",
"and",
"analysis",
"methods",
"barium",
"chemistry",
"chlorides",
"probability",
"theory",
"normal",
"distribution",
"physiology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"neurophysiology",
"chemical",
"elements"
] | 2017 | Modeling an Excitable Biosynthetic Tissue with Inherent Variability for Paired Computational-Experimental Studies |
Bacterial uptake by phagocytic cells is a vital event in the clearance of invading pathogens such as Streptococcus pneumoniae . A major role of the P-selectin glycoprotein ligand-1 ( PSGL-1 ) on leukocytes against invasive pneumococcal disease is described in this study . Phagocytosis experiments using different serotypes demonstrated that PSGL-1 is involved in the recognition , uptake and killing of S . pneumoniae . Co-localization of several clinical isolates of S . pneumoniae with PSGL-1 was demonstrated , observing a rapid and active phagocytosis in the presence of PSGL-1 . Furthermore , the pneumococcal capsular polysaccharide and the main autolysin of the bacterium ―the amidase LytA― were identified as bacterial ligands for PSGL-1 . Experimental models of pneumococcal disease including invasive pneumonia and systemic infection showed that bacterial levels were markedly increased in the blood of PSGL-1−/− mice . During pneumonia , PSGL-1 controls the severity of pneumococcal dissemination from the lung to the bloodstream . In systemic infection , a major role of PSGL-1 in host defense is to clear the bacteria in the systemic circulation controlling bacterial replication . These results confirmed the importance of this receptor in the recognition and clearance of S . pneumoniae during invasive pneumococcal disease . Histological and cellular analysis demonstrated that PSGL-1−/− mice have increased levels of T cells migrating to the lung than the corresponding wild-type mice . In contrast , during systemic infection , PSGL-1−/− mice had increased numbers of neutrophils and macrophages in blood , but were less effective controlling the infection process due to the lack of this functional receptor . Overall , this study demonstrates that PSGL-1 is a novel receptor for S . pneumoniae that contributes to protection against invasive pneumococcal disease .
Streptococcus pneumoniae ( pneumococcus ) is one of the major causes of invasive disease accounting for more deaths than any other vaccine-preventable bacterial infection . This microorganism colonizes the human nasopharynx , being one of the leading causes of acute otitis media , community-acquired pneumonia and invasive pneumococcal disease ( IPD ) including sepsis and meningitis [1] . The World Health Organization estimates that nearly 14 million episodes of serious pneumococcal disease occur every year with a critical impact in childhood population as pneumonia kills more children than AIDS , malaria and measles combined [1 , 2] . Resolution of pneumococcal disease is regulated by the efficient recognition and clearance of the invading pathogen by professional phagocytes [3] . Leukocytes play an important role in inflammatory and immune responses against bacterial infection and bacterial clearance depends on the efficacy of different receptors on phagocytic cells to recognize , internalize and kill the pathogen [4–7] . Phagocytic receptors on the cell surface trigger phagocytosis following direct recognition of particulate targets . Interaction between selectins and selectin-ligand molecules is essential for the host-pathogen encounter due to its crucial role in leukocyte extravasation [8] . In this sense , expression of P-selectin and E-selectin by the endothelium provides protection against invading pathogens such as S . pneumoniae [9 , 10] . P-selectin glycoprotein ligand-1 ( PSGL-1 ) on leukocytes mediates interactions with P-selectin and E-selectin expressed by endothelial cells [11] . PSGL-1 is a homodimeric mucin-like glycoprotein expressed on the surface of almost all circulating leukocytes with a great importance in leukocyte adhesion and transmigration as it is responsible for the initial steps of the extravasation cascade [8 , 12] . However , certain intracellular pathogens have developed sophisticated strategies exploiting specific receptors for their own benefit to enter eukaryotic cells and replicate intracellularly [13] . This is the case of the obliged intracellular pathogens Anaplasma phagocytophilum , Ehrlichia sp . , and enterovirus 71 that get access inside the cell by binding PSGL-1 , causing granulocytic anaplasmosis/ehrlichiosis and hand-foot-mouth disease respectively [14–16] . However , there is no experimental evidence indicating that PSGL-1 could act as a receptor on leukocytes participating in the recognition and clearance of extracellular invading pathogens such as S . pneumoniae . In this sense , the main goal of this study was to investigate the protective contribution of PSGL-1 in host defense against IPD .
The plasma membrane of phagocytes expresses an array of receptors that interact with specific microbial ligands promoting the internalization and clearance of the potential pathogen . To evaluate the impact of PSGL-1 in pneumococcal phagocytosis , HL-60 cells differentiated to granulocytes were used as target cells because they express the same phagocytic receptors as peripheral blood neutrophils including PSGL-1 [14–17] ( S1 Fig ) . To assess the role of PSGL-1 in the phagocytosis of S . pneumoniae , the receptor function was blocked using the specific monoclonal antibody KPL-1 . This is an accepted method to assess the impact of PSGL-1 in microbial interaction [14–16] . To determine the generic role of PSGL-1 in host defense against this important pathogen , clinical isolates of S . pneumoniae belonging to different serotypes were assessed . Phagocytosis was significantly impaired when PSGL-1 was blocked , indicating that pneumococcal phagocytosis is more efficient when this receptor is fully active ( Fig 1A and 1B ) . The contribution of FCγ-receptors was evaluated indicating that the effect of PSGL-1 in phagocytosis is independent of FCγ-receptors activity ( S1 Fig ) . In addition , bacterial killing mediated by PSGL-1 was examined using three different clinical isolates . Our results showed that pneumococcal survival increased when PSGL-1 in phagocytic cells was blocked , demonstrating that this receptor is involved in the clearance of S . pneumoniae ( Fig 1C ) . Finally , phagocytosis experiments using neutrophils obtained from the spleen of wild-type and PSGL-1–/–mice by cell sorting , confirmed that PSGL-1 is involved in the phagocytosis of S . pneumoniae ( Fig 1D and 1E ) . To further analyze the kinetics of pneumococcal phagocytosis , cells with PSGL-1 ―either active or antibody-blocked― were infected with the D39 ( serotype 2 ) strain expressing the green fluorescent protein ( GFP ) , and the phagocytosis process was monitored using live imaging confocal microscopy ( Fig 2A and 2B , S1 and S2 Movies ) . When PSGL-1 was active , a rapid and active phagocytosis was observed , with the majority of the cells containing pneumococcal cells by the end of the process ( Fig 2A and S1 Movie ) . However , when PSGL-1 receptor was blocked , the recognition and engulfment of S . pneumoniae was impaired , which confirmed the importance of PSGL-1 in pneumococcal phagocytosis ( Fig 2B and S2 Movie ) . To confirm the interaction of S . pneumoniae with PSGL-1 , fluorescently-labeled pneumococcal isolates of serotypes 2 , 3 , 6B , 11A , 23F and 19A were used to observe co-localization with PSGL-1 ( Fig 3 and S2 Fig ) . Hence , our findings show that PSGL-1 is involved in the recognition and phagocytosis of a major human pathogen such as S . pneumoniae contributing therefore , to the variety of receptors on professional phagocytes that are needed to efficiently identify and destroy invading pathogens [4–6 , 18] . Phagocytosis requires receptor-mediated recognition of microbial ligands that are usually expressed in the surface of invading pathogens . These ligands are frequently known as pathogen-associated molecular patterns ( PAMPS ) which are recognized by specific receptors of the innate immune system [19] . As LytA ―the main cell wall hydrolase of S . pneumoniae― is located on the bacterial surface and it is essential to interact with critical components of the host immune response including neutrophils and macrophages [20 , 21] , we explored the possible interaction of PSGL-1 with LytA . Pneumococcal recognition by PSGL-1 was hindered in the absence of LytA , suggesting that LytA might be a bacterial ligand for PSGL-1 ( Fig 4A and 4B ) . Direct interaction between purified LytA and PSGL-1 molecules was observed confirming that this receptor recognizes LytA ( Fig 4C ) . This interaction was dependent on the concentration of PSGL-1 , suggesting that recognition of S . pneumoniae is enhanced when PSGL-1 levels are increased . Preincubation of HL-60 cells with purified LytA before infection reduced the phagocytosis in a similar way than KPL-1 antibody , supporting additional evidence that LytA interacts with PSGL-1 ( S3 Fig ) . To confirm the interaction of LytA and PSGL-1 we included the non-capsulated strain ( M11 ) and an isogenic lytA mutant strain ( Fig 4D ) . Our results demonstrated that in the absence of LytA , the binding of S . pneumoniae to PSGL-1 is impaired . Although the capsular polysaccharide ( CPS ) is one of the major virulence factors of S . pneumoniae , resistance to phagocytosis can vary with the capsular type , which might explain differences of invasiveness among strains [22 , 23] . Capsule recognition by PSGL-1 was investigated using a non-capsulated strain ( M11 ) and several isogenic transformants of M11 expressing different CPSs ( Fig 4E ) . The absence of CPS caused impaired recognition by PSGL-1 , in comparison to the corresponding encapsulated transformants , confirming that PSGL-1 recognizes the pneumococcal CPS ( Fig 4E ) . However , different levels of binding were observed depending on the CPS with the weakest recognition for the pneumococcal strain expressing serotype 19A ( Fig 4E ) . Experiments including purified CPS of type 3 and different concentrations of PSGL-1 were included to investigate the binding of PSGL-1 to the pneumococcal capsule ( Fig 4F ) . This CPS was assessed because is included in the current PCV-13 vaccine and clinical isolates of serotype 3 are a major cause of IPD [24] . Direct recognition of purified CPS by PSGL-1 was observed showing a concentration-dependent pattern , which confirms that PSGL-1 recognizes the capsule of S . pneumoniae ( Fig 4F ) . The development of IPD depends of the complex interplay of many factors including virulence determinants of the pathogen and the efficacy of the host immune response to clear the infection process . A failure to efficiently detect and destroy S . pneumoniae in the lower respiratory tract or the systemic circulation lead to severe pneumonia or disseminated infection which are associated to increased mortality rates [3] . Next , the protective role of PSGL-1 against IPD was investigated using pneumonia and sepsis models of infection ( Fig 5 ) . In pneumonia , bacterial counts were slightly higher in the bronchoalveolar lavage fluid ( BALF ) and lung of PSGL-1−/− mice , and much higher in the blood of KO mice ( Fig 5A ) . These bacterial levels were markedly elevated at 24 h in the blood of PSGL-1−/− mice ( with the progression of the infection ) , confirming that PSGL-1 contributes to control bacterial load by reducing the severity of pneumococcal dissemination ( Fig 5A ) . In the sepsis model , PSGL-1−/− mice had greater levels of bacteria in blood ( Fig 5B ) . In addition , lethal infection developed faster in PSGL-1−/− mice than in wild-type mice indicating that PSGL-1 plays a critical role in host defense against IPD by controlling bacterial infection in the systemic circulation ( Fig 5B and5C ) . To confirm this hypothesis and exclude the contribution of cellular migration mediated by PSGL-1 , mice were directly infected by the intravenous route ( Fig 5D ) . Mice lacking PSGL-1 had increased levels of bacteria in blood than wild-type mice both at 6 and 24 h confirming that a major function of PSGL-1 in host defense is to clear the bacteria in the bloodstream controlling the dissemination ( Fig 5D ) . To extend the importance of PSGL-1 in the clearance of S . pneumoniae from the systemic infection , a sepsis model was repeated using a lethal dose of a different serotype such as TIGR4 strain . Our findings corroborated the results obtained with the D39 strain demonstrating that lack of PSGL-1 was associated with increased bacterial counts in blood and a more severe infection compared to wild-type mice ( Fig 5E and 5F ) . Overall , our findings demonstrate that PSGL-1 plays an important role against IPD . The inflammatory response to infection with S . pneumoniae in PSGL-1−/− and wild-type mice was characterized in BALF and serum . The pattern of the major cytokines associated to infection was similar in BALF ( Fig 6A ) , although in serum of PSGL-1−/− mice there were significant increased levels of IL-5 , IL-6 , and IFN-γ ( P <0 . 05 ) which is compatible with the higher bacterial levels found in the blood of these mice ( Fig 6B ) . Neutrophils predominate within cellular infiltrates in pneumococcal pneumonia and the consequences of the neutrophil influx for the host can be advantageous or detrimental , depending on the degree of cellular influx and the ability of the pathogen to successfully avoid the immune response [3] . Leukocyte infiltration into lungs and circulating leukocytes in blood , were measured by flow cytometry , using wild-type and PSGL-1−/−mice infected with S . pneumoniae D39 strain ( Fig 7A and 7B ) . In the pneumonia model ( intranasal inoculation ) , the number of neutrophils and macrophages were similar in the lungs of both types of mice whereas T cell counts were higher in PSGL-1−/−mice ( Fig 7A ) . In a systemic model of infection ( intravenous inoculation ) , however , the number of T cells were similar , although the proportion of neutrophils and macrophages were significantly higher in PSGL-1−/−mice ( Fig 7B ) , which is compatible with the higher levels of bacteria in the blood of these mice ( Fig 5D ) . Overall , these results indicate that PSGL-1−/−mice , despite having greater numbers of leukocytes in blood , had an impaired ability to clear the bacteria from the bloodstream confirming the importance of PSGL-1 in the recognition and killing of S . pneumoniae in the systemic circulation . Immunohistochemical characterization of thin sections from lung tissues confirmed that mice deficient in PSGL-1 had greater infiltration of T cells and neutrophils compared to wild-type mice , which are consistent with the severity of the infection process developed in PSGL-1−/− mice ( Fig 7C ) . To demonstrate that pneumococcal LytA is involved in the physiological effects mediated by PSGL-1 , phagocytic assays were performed using HL-60 cells exposed or not to the KPL-1 antibody and a pneumococcal strain lacking LytA ( Fig 8A and 8B ) . Hence , our findings confirmed that phagocytosis of a LytA deficient strain is not dependent on PSGL-1 on HL-60 cells ( Fig 8A and 8B ) . To confirm the relevance of LytA in the interaction with PSGL-1 , pneumonia and sepsis models of infection were repeated in wild-type and PSGL-1−/− mice using a D39 lytA deficient strain ( Fig 8C and 8D ) . In contrast to mice infected with wild-type D39 , there were no differences in lung or BALF CFU between PSGL-1−/− and wild-type mice infected intranasally with the D39 lytA mutant strain . Similarly , in the sepsis model there were no differences in recovered D39 lytA deficient strain CFU between PSGL-1−/− and wild-type mice . These results confirm that the interaction of LytA with PSGL-1 is important for innate immunity against S . pneumoniae . ( Fig 8C and 8D ) .
Neutrophils are key players in the innate and adaptive immune responses to microbial cells , since they are critical for rapid clearance of invading bacteria [25 , 26] . For this process , neutrophils must first detect the microorganisms using surface receptors that are essential to engulf and kill the pathogen [18 , 27] . PSGL-1 is a ligand of P- , E- and L-selectins , and is able to mediate the tethering and rolling of circulating leukocytes on the activated endothelium prior to their extravasation [8 , 11 , 12] . The role of P , E and L-selectin as well as certain integrins against pneumococcal infection has been previously characterized [9 , 10 , 28 , 29] . However , the direct role of PSGL-1 on leukocytes in host defense against S . pneumoniae including its contribution to the recognition and clearance of this microorganism is unknown . As this bacterium is highly variable with up to 96 serotypes described so far [30] , we included different serotypes of S . pneumoniae to investigate the role of PSGL-1 in phagocytosis . Hence , our findings showed that PSGL-1 is involved in the recognition and phagocytosis of a major human pathogen such as S . pneumoniae contributing therefore to the variety of receptors on professional phagocytes that are needed to efficiently identify and destroy invading pathogens [4–6 , 18] . For the detection of the pathogen it is necessary the interaction of phagocytic receptors with bacterial ligands that are usually exposed on the surface of the microorganism [19] . Using LytA-deficient mutants , we have recently demonstrated that this surface-exposed protein is a major determinant in the virulence of S . pneumoniae by interacting with essential components of the immune system including phagocytes [20] . Our results show now that the LytA autolysin is recognized by PSGL-1 and this effect is dependent on the level of PSGL-1 suggesting that variations in the expression of PSGL-1 on leukocytes might contribute to the efficiency of this interaction . One of the major concerns in the outcome of the infection is that pneumococcal disease can be produced by clinical isolates of a wide variety of polysaccharide capsules , the main virulence determinant of S . pneumoniae [22 , 23] . In this sense , using pneumococcal transformants expressing the same genetic background but different CPS , we have demonstrated that PSGL-1 recognizes the capsule of S . pneumoniae . This effect was variable depending on the CPS expressed , with the lowest binding related to the strain expressing serotype 19A . In this sense , incidence of IPD cases caused by this serotype has dramatically increased in the last few years and it has been linked to the emergence of vaccine escape variants that arise by switching the capsular locus from serotype 4 to 19A [31 , 32] . These results are important in terms of pathogenesis because differences in the recognition of pneumococcal CPS by receptors such as PSGL-1 , might explain why certain serotypes of S . pneumoniae are more associated to IPD and dissemination worldwide than others [31–33] . The repertoire of host receptors involved in the binding , uptake , signaling and response to invading pathogens is critical for the outcome of the infection . PSGL-1 is the main selectin receptor involved in neutrophil adhesion and migration [8] . Although the importance of leukocyte extravasation is relevant in pneumococcal infection [9 , 10 , 29] , the contribution of PSGL-1 to the resolution of IPD was previously unknown . In this study we have used PSGL-1−/−mice to investigate the in vivo role of this receptor in host defense against pneumococcal infection . Our findings confirm that PSGL-1 may act as a pathogen-recognition receptor of the immune system [34] . In this sense , PSGL-1 acts in host defense controlling bacterial proliferation , dissemination and tissue injury which are critical aspects of IPD . This is of great relevance from the respiratory perspective as invasive pneumonia caused by S . pneumoniae is one of the major causes of mortality in children and adults [1 , 3] . Lack of PSGL-1 has been linked to a greater susceptibility against the intracellular pathogens Salmonella typhimurium or Citrobacter rodentium , which is associated with dramatically increased levels of pro-inflammatory cytokines after intestinal infection [35 , 36] . In the case of S . pneumoniae , bacterial recognition by immune cells generates an array of cytokines which may play a significant role in host defense . Increase of IL-5 , IL-6 and IFN-γ was observed in PSGL-1–/–mice sera . According to our findings , it has been described that increased serum levels of IL-5 and IL-6 were associated with reduced microbial clearance and higher mortality rates in sepsis [37 , 38] . In addition , the increased levels of IL-12 and IFN-γ in the lungs of wild-type mice might be protective as IFN-γ is involved in the IL-12 regulation of neutrophil-mediated host defense against pneumococcal pneumonia [39] . Bacterial infections of the lower respiratory tract are characterized by massive accumulation of neutrophils in the alveolar spaces [3 , 40] . Our results show that similar numbers of neutrophils and macrophages were observed in the lungs of PSGL-1−/− mice . Our results might be unexpected as PSGL-1 is involved in cellular migration of these cells and therefore , increased numbers should be present in the lungs of wild-type mice [8] . Hence , our findings can be explained in the context of a bacterial infection as certain pathogens ―including S . pneumoniae― have the ability to impair neutrophil migration to the site of infection by cleaving PSGL-1 [41 , 42] . Interestingly , this is in line with previous findings confirming that the absence of endothelial selectins E , L and P is not associated with impairment of leukocyte emigration to infection sites after pneumococcal disease [10 , 43] . In addition , PSGL-1 negatively regulates CD4+T cell immune responses in vivo which can explain the increased levels of T cells observed in the lungs of PSGL-1−/− infected mice [44] . Using an intravenous infection model when no collateral migration effects were expected , higher numbers of neutrophils and macrophages but not T cells , were observed in PSGL-1−/− mice . This is consistent with previous observations showing that blood of PSGL-1−/− mice contained similar numbers of lymphocytes although they had a significant increase in the proportion of leukocytes , with enhanced levels of granulocytes and monocytes in comparison to wild-type mice [45] . This is relevant from the phagocytosis perspective as PSGL-1−/− mice , despite having increased numbers of these phagocytic cells in the bloodstream , were unable to control bacterial replication in the blood , leading to the rapid development of fatal infection . Overall , these results confirm that PSGL-1 on leukocytes plays a critical role in host defense against pneumococcal infection . As a consequence of pathogen-recognition by PSGL-1 , S . pneumoniae is efficiently engulfed and destroyed , reducing bacterial replication and dissemination in the host , contributing to control the severity of IPD .
All the experiments involving the use of animals have been performed following the guidelines of the Bioethical and Animal Welfare Committee of Instituto de Salud Carlos III ( ISCIII ) that reviewed and approved protocol CBA PA 52-2011-v2 , to be performed at the National Centre for Microbiology of ISCIII . Animals were bred at Universidad Autónoma de Madrid animal facility following institutional guidelines for animal use and care . Infection experiments conformed to the Spanish government legislation ( RD 1201/2005 ) and European Community regulations ( 86/609/EEC ) . The S . pneumoniae strains used were D39 ( NCTC 07466 , serotype 2 ) , TIGR4 ( ATCC BAA-334 , serotype 4 ) , and clinical isolates of different serotypes obtained from the Spanish Pneumococcal Reference Laboratory; 957/12 ( serotype 3 ) , 1515/97 ( serotype 6B ) , 450/12 ( serotype 11A ) , 3347/12 ( serotype 19A ) , 69 ( serotype 19F ) and 48 ( serotype 23F ) . The non-encapsulated S . pneumoniae strain M11 and their isogenic transformants expressing CPS of serotypes 3 , 6B and 19A were also included in this study [46] . S . pneumoniae D39 strain expressing the GFP was constructed by transformation with plasmid pMV158GFP [47] and used for confocal microscopy experiments . S . pneumoniae strains were cultured on blood agar plates at 37°C in a 5% CO2 atmosphere , or in Todd-Hewitt broth supplemented with 0 . 5% yeast extract , to an optical density at 550 nm ( OD550 ) of 0 . 5 , and stored at −70°C in 10% glycerol as single use aliquots . HL-60 cells ( CCL-240; ATCC ) differentiated to granulocytes were used and the general conditions of the assay were based on those described previously [17 , 48] . Briefly , S . pneumoniae strains were fluorescently labeled by incubation with FAM-SE ( Molecular Probes ) in 0 . 1 M sodium bicarbonate buffer for 1 h at 37°C , washed five times with Hanks balanced salt solution ( HBSS ) and stored at −70°C in 10% glycerol as aliquots for further assays . HL-60 cells were harvested by centrifugation and washed twice with HBSS and once with HBSS in the presence of calcium and magnesium ions . Infection assays were performed in the absence of serum to avoid complement-dependent phagocytosis and 105 HL-60 cells were infected with 2×106 colony forming units ( CFU ) of viable FAM-SE labeled bacteria . To block PSGL-1 , HL-60 cells were incubated for 1 h at room temperature with 25 μg/ml of the KPL-1 antibody ( mouse anti-human PSGL-1; MBL ) or IgG isotype control ( mouse anti-human IgG; Novus Biologicals ) as previously described [15 , 16] . A similar approach was performed using purified LytA . A minimum of 6 , 000 cells were analyzed using a FACS Calibur flow cytometer ( BD Biosciencies ) . Using cytochalasin D , an inhibitor of actin polymerization , we have previously shown that the majority of the effect on the association of fluorescent S . pneumoniae with HL-60 cells is due to phagocytosis rather than to adhesion of bacteria to the cell surface [49] . Results were expressed as a fluorescence index defined as the proportion of positive cells for fluorescent bacteria multiplied by the geometric mean of fluorescence intensity which correlates with the amount of bacteria phagocytosed per cell [48] . Opsonophagocytosis killing assays were performed in the absence of serum using 105 HL-60 cells with the PSGL-1 receptor either active or blocked as mentioned above and 2 . 5 × 104 CFU/ml of S . pneumoniae as previously described [50] . Serial dilutions of culture supernatants were plated on blood agar plates for bacterial counts determination and results were expressed as bacterial survival after 45 min incubation of the pneumococcal strains with HL-60 cells expressing ( or not ) PSGL-1 . Phagocytosis assays were repeated using neutrophils purified from the spleen of wild-type and PSGL-1−/− mice by FACS using a FACSAria I ( BD Biosciences ) dispositive with DIVA version 6 . 1 software as previously described [51] . Briefly , single-cell suspensions were prepared in staining buffer ( 2% fetal calf serum in PBS ) , and non-specific binding was blocked with Fc block ( BD Biosciences ) . Staining was performed using standard protocols with the following antibodies diluted 1/200 in staining buffer including propidium iodide ( rat anti-mouse CD11b-allophycocyanin ( APC ) and rat anti-mouse GR-1-phycoerythrin; Biolegend ) . S . pneumoniae strains labeled with FAM-SE were used for immunofluorescence microscopy . HL-60 cells previously infected as described above were seeded on 12-mm circular coverslips for immunofluorescence staining . As HL-60 cells are in suspension , cells were cytofuged at 70 × g for 2 min using a Cytospin centrifuge ( Thermo Electron , Pittsburgh , PA ) , as described elsewhere [16] . For the detection of PSGL-1 in HL-60 cells differentiated to granulocytes , cells were fixed with 3% paraformaldehyde ( PFA ) for 10 min at room temperature and after two washes with PBS , coverlips were kept in a solution 1 M NH4Cl-PBS solution . Coverslips containing the infected cells were washed twice in PBS containing 0 . 1% saponin and once in PBS and incubated for 30 min with the primary antibody . Staining was performed in PBS containing 10% horse serum , 0 . 1% saponin and the primary antibody using a mouse anti-human PSGL-1 monoclonal antibody ( KPL-1; MBL ) diluted 1/300 . Cellular DNA was stained with Hoechst ( Invitrogen ) diluted 1/2500 . After 30 min incubation with the primary antibody at room temperature , coverlips were washed twice with PBS-saponin 0 . 1% , and once with PBS pH 7 . 0 before incubation during 30 min at room temperature with a dilution 1/200 of the secondary antibody ( goat anti-mouse Texas Red; Serotec ) . Finally , coverslips were washed twice in PBS containing 0 . 1% saponin , once in PBS , and once in H2O , mounted with Aqua Poly/Mount ( Polysciences ) , and analyzed with a Leica spectral SP5 confocal microscope using the Leica software ( LAS-AF ) . Binding of PSGL-1 to S . pneumoniae , purified LytA or CPS was analyzed by ELISA as previously described [20] . Briefly , whole cell ELISA was performed by coating 96-well plates with 200 μl of exponentially growing bacteria and resuspended in PBS to an OD550 of 1 . 0 . Plates were air dried at room temperature and blocked with 200 μl of PBS-0 . 5% BSA-NaN3 for 1 h before 50 μl of different concentrations of PSGL-1 ( R&D systems , USA ) were added to each well . After overnight incubation at 4°C , plates were washed 5 times with PBS-Tween 0 . 1% and incubated overnight at 4°C with 50 μl of mouse anti-human PSGL-1 ( KPL-1; MBL ) diluted 1/4000 . After 5 washes with PBS-Tween 0 . 1% , plates were incubated with goat anti-mouse IgG HRP ( Southern Biotech ) for 30 min at room temperature and developed with o-phenylenediamine ( Sigma-Aldrich ) . Plates were measured at OD492 using a microtiter plate reader ( Anthos 2020 ) . Direct binding of PSGL-1 to purified LytA protein or type 3 CPS ( ATCC 169-X , Merck Sharp & Dohme ) was performed as described above except that the 96-well plates were coated with 50 μg of purified LytA protein or CPS per well . Purified LytA protein was obtained by overexpression in Escherichia coli [52] . Wild-type C57BL/6 mice and PSGL-1−/− mice were bred in a conventional animal facility at the School of Medicine , Universidad Autónoma de Madrid ( UAM ) . PSGL-1−/− mice were kindly provided by Dr . D Vestweber and Dr . MK Wild ( Max Plank Institute for Molecular Biomedicine , Münster , Germany ) . Wild-type C57BL/6 mice obtained from the Jackson Laboratory and PSGL-1−/− mice were backcrossed and the wild-type and PSGL-1−/− littermates obtained from crosses of the resulting heterozygous mice were used to breed our wild-type and PSGL-1−/− colonies used in this study . Animal procedures were approved by the Animal Care and Use Committee of ISCIII . All mice used were 8–16 weeks old , and within each experiment , groups of mice were matched for age and sex . Studies investigating pneumococcal sepsis or pneumonia were performed using groups of at least 5 mice and infected as previously described [48] . Briefly , for sepsis , mice were challenged with 5 × 106 CFU/ml for D39 strain or 3 × 104 CFU/ml for TIGR4 strain ( in a volume of 200 µl ) by the intraperitoneal route , whereas for pneumonia , mice under anesthesia with isofluorane were inoculated intranasally with 50 µl containing 107 CFU/mouse of D39 strain . For intravenous inoculation , mice were infected with 2 × 107 CFU/mouse of D39 strain through the tail vein . At 6 h and 24 h after challenge , a lethal dose of pentobarbital was administered and bacterial counts were determined from samples recovered from BALF , lung and blood . Experiments were repeated twice using 5 mice in each group and results were expressed as Log10 CFU/ml of bacteria recovered from the different sites . Cytokines were measured from BALF and blood of wild-type mice and PSGL-1−/− mice infected with D39 strain by the intranasal route as explained above . Cytokines levels ( IL-2 , IL-4 , IL-5 , IL-6 , IL-10 , IL-12p70 , GM-CSF , TNF-α and IFN-γ ) were determined by using a Luminex magnetic bead array assay ( Bio-Rad ) according to manufacturer protocols . Experiments investigating cellular populations in lungs and blood after pneumococcal infection were performed in wild-type and PSGL-1−/− mice infected as explained above using a FACS assay as previously described [51 , 53] . Briefly , single-cell suspensions were prepared in staining buffer ( 2% fetal calf serum in PBS ) , and non-specific binding was blocked with Fc block ( BD Biosciences ) . Staining was performed using standard protocols with the following antibodies diluted 1/400 in staining buffer ( rat anti-mouse CD11b-allophycocyanin ( APC ) , Biolegend; rat anti-mouse CD4-FITC , Biolegend; rat anti-mouse CD8-APC , Biolegend; rat anti-mouse GR-1-phycoerythrin; Biolegend ) [51 , 53] . Cells were analyzed on a FACSCanto I using the FlowJo version 6 . 3 . 4 software package . Mice were euthanized with pentobarbital and lungs were inflated and fixed with 4% PFA in PBS . Lungs were paraffin-embedded and 5-μm sections were obtained . Infiltrates of granulocytes and T cells were measured by staining with anti-Ly-6G/6C ( antigen retrieval pH 6 . 5 , 1/400 , Abcam ab2557 ) and anti-CD3 ( antigen retrieval pH 6 . 0 , 1/200 , Santa Cruz Biotechnology sc-1127 ) antibodies respectively . Immunohistochemistry was performed with the Dako LSAB+ System-HRP following manufacturer’s instructions . Data are representative of results obtained from repeated independent experiments , and each data point represents the mean and standard deviations ( SD ) for 3 to 5 replicates . Statistical analysis was performed by using two-tailed Student’s t test ( for two groups ) , whereas analysis of variance ( ANOVA ) followed by a Dunnett’s post hoc test was chosen for multiple comparisons . Survival was analyzed by the log-rank test . GraphPad InStat version 5 . 0 ( GraphPad Software , San Diego , CA ) was used for statistical analysis . Differences were considered statistically significant with P <0 . 05 ( * ) and highly significant with P <0 . 01 ( ** ) and P <0 . 001 ( *** ) . | S . pneumoniae is one of the most important and devastating human pathogens worldwide , mainly affecting young children , elderly people and immunocompromised patients . In terms of host immune defense against invasive pneumococcal isolates , professional phagocytes require receptor-mediated recognition of certain ligands on the bacterial surface for the uptake and clearance of the microorganism . In this study , we demonstrate that the P-selectin glycoprotein ligand-1 ( PSGL-1 ) on leukocytes is involved in the phagocytosis process of S . pneumoniae by targeting the capsule and the surface protein LytA as pathogen-associated molecular patterns . To explore this process in more detail , we have used wild-type mice and mice deficient in PSGL-1 demonstrating that lack of PSGL-1 is detrimental for the host by increasing the susceptibility to the infection and the severity of the pneumococcal invasive disease . Overall , these data show the importance of PSGL-1 on leukocytes in host defense against S . pneumoniae and confirm that PSGL-1 plays a critical protective role against invasive bacterial disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"enzyme-linked",
"immunoassays",
"pathogens",
"cell",
"processes",
"immunology",
"microbiology",
"pulmonology",
"pneumonia",
"sepsis",
"signs",
"and",
"symptoms",
"immunologic",
"techniques",
"neutrophils",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"medical",
"microbiology",
"microbial",
"pathogens",
"immunoassays",
"phagocytosis",
"hematology",
"blood",
"cell",
"biology",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
] | 2016 | PSGL-1 on Leukocytes is a Critical Component of the Host Immune Response against Invasive Pneumococcal Disease |
Odors are initially represented in the olfactory bulb ( OB ) by patterns of sensory input across the array of glomeruli . Although activated glomeruli are often widely distributed , glomeruli responding to stimuli sharing molecular features tend to be loosely clustered and thus establish a fractured chemotopic map . Neuronal circuits in the OB transform glomerular patterns of sensory input into spatiotemporal patterns of output activity and thereby extract information about a stimulus . It is , however , unknown whether the chemotopic spatial organization of glomerular inputs is maintained during these computations . To explore this issue , we measured spatiotemporal patterns of odor-evoked activity across thousands of individual neurons in the zebrafish OB by temporally deconvolved two-photon Ca2+ imaging . Mitral cells and interneurons were distinguished by transgenic markers and exhibited different response selectivities . Shortly after response onset , activity patterns exhibited foci of activity associated with certain chemical features throughout all layers . During the subsequent few hundred milliseconds , however , MC activity was locally sparsened within the initial foci in an odor-specific manner . As a consequence , chemotopic maps disappeared and activity patterns became more informative about precise odor identity . Hence , chemotopic maps of glomerular input activity are initially transmitted to OB outputs , but not maintained during pattern processing . Nevertheless , transient chemotopic maps may support neuronal computations by establishing important synaptic interactions within the circuit . These results provide insights into the functional topology of neural activity patterns and its potential role in circuit function .
The brain continuously processes patterns of sensory input to extract relevant information about the environment . In most or even all sensory systems , afferent activity patterns exhibit a topological organization in which defined stimulus variables are mapped along spatial coordinates in the brain . A continuous change in a stimulus variable therefore results in a continuous shift of activity along a distinct trajectory on the brain surface . Topological sensory maps are often elaborated during successive stages of processing [1 , 2] and are assumed to play important roles for information processing although their precise functions are not well defined [3] . Odors differ from most other sensory stimuli because they span a high-dimensional stimulus space and cannot be described by a small number of continuous variables [4 , 5] . It is therefore debated whether a systematic mapping of chemical stimulus features , referred to as “chemotopy , ” plays a role in the processing of odor-encoding activity patterns in the brain . Odor information is conveyed from the nose to the first central processing center , the olfactory bulb ( OB ) , by specific patterns of activation across the array of glomeruli [6–13] . Each glomerulus receives convergent input from sensory neurons expressing the same odorant receptor [14] . As a consequence , the array of glomeruli constitutes a discrete spatial map of odorant receptor expression . Because individual odorant receptors respond to overlapping sets of molecules [5 , 15 , 16] , glomerular activity patterns are combinatorial and spatially distributed . However , odors sharing certain molecular properties preferentially activate glomeruli within defined areas , as shown in various vertebrate species [9–13 , 17–20] . The gross functional organization of glomerular activity patterns is therefore described as a “chemotopic map” because molecular stimulus features are associated with glomerular activity in defined regions . In rodents and zebrafish , primary molecular properties ( e . g . , characteristic functional groups ) are mapped onto relatively large domains , whereas secondary molecular features ( e . g . , chain length ) are mapped onto subregions within these domains [9 , 10 , 12 , 13 , 17] . Chemotopic maps are therefore organized hierarchically such that “fine” maps of secondary features are nested within “coarse” maps of primary features . Within a given region , however , not all glomeruli respond to all stimuli with the associated feature , and a given stimulus usually activates glomeruli in more than one region . Compared to topological maps in other sensory systems , the chemotopy of glomerular activity patterns therefore appears rough and fractured , possibly as a consequence of reducing a high-dimensional molecular feature space onto two spatial dimensions [13 , 21] . Although chemotopic activity patterns have been described at the level of glomeruli , much less is known about the topological organization of odor-evoked activity patterns downstream of glomerular afferents . Patterns of glomerular input activity are processed in the OB by neuronal circuits that are conserved within vertebrates [22 , 23] and similar to the first olfactory processing center in insects , the antennal lobe [4] . The principal neurons of the OB , the mitral cells ( MCs ) , receive direct excitatory input from sensory neurons and convey the output of the OB to higher brain regions . In addition , the OB contains different types of interneurons ( INs ) including periglomerular and granule cells . Most INs are GABAergic and mediate inhibitory lateral interactions between MCs over a wide range of spatial scales [24–26] . These network interactions create temporal patterns of activity on multiple time scales and result in dynamic changes of OB output activity patterns during an odor response [27–34] . It is , however , unclear how this circuitry processes chemotopically organized patterns of glomerular input . A detailed analysis of topological activity patterns downstream of glomeruli requires recordings of odor responses from a substantial fraction of MCs and INs to a representative panel of stimuli with high spatial and temporal resolution . Single-unit extracellular recordings from anesthetized rabbits revealed that MCs within certain regions of the OB tend to respond to molecules sharing chemical features , suggesting that MC activity patterns exhibit at least some chemotopy [35 , 36] . However , the dynamics of MC activity patterns over time have not been investigated , and the spatial resolution and sample size obtained with this approach are necessarily limited . Moreover , electrophysiological recordings in mice indicate that MC activity patterns are influenced substantially by anesthesia [37] . The visualization of odor-evoked activity by 2-deoxyglucose uptake or immediate early gene expression suggests that the activation of small groups of glomeruli is associated with spatially restricted activity in deeper layers of the OB [6 , 11 , 38–41] . The comparison of 2-deoxyglucose uptake patterns from multiple animals , each stimulated with one odor , could provide information about chemotopic maps , but has been performed only for the glomerular layer [11] . Moreover , this technique has relatively low spatial resolution and lacks temporal resolution . Patterns evoked by multiple stimuli within the same OB have been visualized by functional magnetic resonance imaging to study chemotopic maps in the glomerular layer of rodents [20 , 42] . Signals in deeper layers , however , appeared weak and have not been analyzed . Optical imaging techniques appear promising but cannot penetrate deep into the OB and provide only a limited field of view . In the salamander , odor-evoked activity patterns throughout multiple layers have been measured by voltage-sensitive dye imaging [43] , but chemotopy has not been studied systematically . Results obtained with different methods therefore indicate that odor-evoked activity across MCs and INs is not uniformly distributed but have not enabled a systematic analysis of chemotopy . Activity patterns across large numbers of individual neurons in the intact vertebrate brain have recently been measured by two-photon Ca2+ imaging [44 , 45] after bolus loading of a Ca2+ indicator [46–50] . The temporal resolution of raw Ca2+ signals is limited by the kinetics of unitary somatic Ca2+ transients . However , changes in firing rates can be reconstructed from somatic Ca2+ signals by temporal deconvolution with a kernel representing a unitary Ca2+ transient [51] . This technique , referred to as temporally deconvolved Ca2+ imaging ( TDCa imaging ) , substantially increases the effective temporal resolution of Ca2+-based activity measurements and facilitates their interpretation in terms of neural activity . Here , we used TDCa imaging to analyze the topological organization of odor-evoked activity patterns across MCs and INs during pattern processing in the OB of zebrafish . The zebrafish is an attractive model system because the architecture of neuronal circuits in the OB is similar to other vertebrates , but contains fewer neurons [41 , 52–54] . Unlike in other fish species , most MCs in zebrafish innervate a single glomerulus within a distance of less than 40 μm from the soma [53] ( Figure S1 ) . Because the zebrafish OB is only about 500 μm in diameter , most neurons are optically accessible . The hierarchical chemotopic organization of glomerular activity maps has been studied previously [9 , 10] . TDCa imaging now permits the analysis of spatiotemporal odor response patterns across a large fraction of the OB neurons downstream of glomerular inputs . In this study , we addressed two questions . First , we quantified basic odor response properties of INs , because this knowledge is important to understand how inhibitory interactions in the OB shape odor-encoding activity patterns across output neurons . The results show that IN responses exhibit a broader range of odor selectivities than MCs and develop differently over time during an odor response . Second , we directly analyzed the topological organization of MC and IN response patterns during the dynamic reorganization of OB output activity . Shortly after response onset , chemotopic activity maps were present throughout all layers and particularly pronounced across MCs . During the subsequent few hundred milliseconds , however , the chemotopy of MC activity patterns became substantially reduced , indicating that the chemotopic organization of odor representations is not maintained during pattern processing . Nevertheless , the transient chemotopy of MC activity patterns may be involved in computations that enhance pattern discriminability and thus may play an important role in circuit function .
We measured neuronal activity patterns in the absence of anesthetics in an explant preparation of the entire zebrafish brain , including the nose and other sensory organs [32] . Neurons in the OB were loaded with the red-fluorescent Ca2+ indicator , rhod-2-AM , by bolus injection [46 , 47 , 51] , and fluorescence was measured by two-photon microscopy [44 , 45] . MCs were identified by colocalization of rhod-2 fluorescence signals with a yellow-fluorescent MC marker protein ( yellow cameleon [YC] ) in a transgenic line ( HuC:YC ) [51 , 54] ( Figure 1A ) . YC-negative neurons were collectively classified as INs . Because most recorded INs were located in the deep ( granule cell ) layer , we assume that IN datasets contained predominantly granule cells . Glomerular neuropil was identified as regions in the peripheral layer of the OB that were devoid of somata and contained a high density of MC dendrites ( Figure 1A ) . Application of odors to the nose caused robust changes in rhod-2 fluorescence in stimulus-specific subsets of neurons ( Figure 1A and 1B1 ) . Responses from up to approximately 350 individual neurons could be detected simultaneously in a given focal plane . Patterns of Ca2+ signals evoked by repeated stimulation with the same odor were reproducible and stable over time ( Figures 1B and S2; average correlation coefficient [mean ± standard deviation ( SD ) ]: MC patterns , 0 . 84 ± 0 . 03; IN patterns , 0 . 87 ± 0 . 03 ) . Somatic Ca2+ signals were low-pass filtered and temporally deconvolved using single-exponential kernels with appropriate time constants to convert Ca2+ signals into firing-rate changes ( Figure 1C; see also Materials and Methods and [51] ) . The temporal resolution was limited by the frame rate ( 128 or 256 ms/frame ) under our experimental conditions . This procedure yielded time series of frames , each representing the instantaneous firing rate of many neurons relative to the baseline firing rate ( Figure 1B2 ) . Individual MCs and INs responded to the same stimulus with different latencies and time courses ( Figure 1B2 , 1C , and 1D ) . As a consequence , the pattern of activity across the population changed over time . The sequence of patterns was reproducible upon repeated stimulation with the same odor , but distinct in response to different stimuli ( Figure 1B2; see below ) . In order to measure activity patterns across large , three-dimensional populations of neurons , odor stimuli ( duration , ~2 . 4 s ) were applied repeatedly , and Ca2+ signals were measured at different focal planes . We concentrated on the ventrolateral OB , which contains approximately 120 small glomeruli responding preferentially to amino acids [9 , 10 , 19] . Amino acids are important odors for most or all aquatic species [55] and thus represent a well-defined subspace of the natural odor space . The concentration used ( 10 μM ) is in the intermediate physiological range [55] and does not saturate glomerular responses in zebrafish [9] . Three datasets were collected: dataset 1 contained responses from 1 , 313 MCs in n = 9 different OBs ( mean ± SD: 146 ± 45 MCs per OB ) to 16 amino acids ( 10 μM ) , acquired at 128 ms/frame . Dataset 2 contained responses of 5 , 111 INs ( n = 3 OBs; 1 , 704 ± 161 INs per OB ) to the same stimuli . Frame time was 256 ms/frame to allow for a larger field of view . Dataset 3 contained experiments in which responses to nine amino acids were measured from both MCs and INs in the same OBs ( n = 3 OBs; mean ± SD: 265 ± 52 MCs and 1 , 388 ± 254 INs per OB ) at 256 ms/frame ( Figure 1C ) . The total number of neurons in the adult zebrafish OB is estimated to be about 20 , 000 , including approximately 1 , 500 MCs [53 , 54] . Responses were therefore obtained from approximately 10% of all MCs or INs in each dataset . The fraction of amino acid–responsive neurons contained in our datasets was probably substantially higher because experiments were performed specifically in the amino acid–sensitive region . Although odor responses of MCs have been studied extensively in various species , response properties of INs are not well understood in vertebrates . We therefore compared basic response properties of MCs and INs . Peak times and rise times of excitatory responses were determined in all trials in which the TDCa signal exceeded a threshold well above noise level ( TDCa signal ≥ 5; dataset 3; n = 2 , 218 MC responses and n = 21 , 754 IN responses; Materials and Methods ) . On average , MC responses became maximal 415 ± 327 ms ( mean ± SD ) after stimulus onset , whereas IN responses peaked significantly later ( 738 ± 492 ms; p < 0 . 001; Wilcoxon rank sum test ) . The mean rise time of the TDCa signal was also slightly , but significantly , longer for IN responses ( MCs: 332 ± 214 ms; INs: 403 ± 333 ms; mean ± SD; p < 0 . 001 , Wilcoxon rank sum test ) . In response to different odors , peak and rise times varied slightly , but were consistently lower for MC responses than for IN responses ( unpublished data ) . To estimate the mean population firing rates of MCs and INs , TDCa signals were averaged across all odors and neurons ( datasets 1 and 2 ) , scaled using factors derived from simultaneous electrophysiological recordings [51] to yield firing-rate changes , and offset corrected for spontaneous firing rates determined by electrophysiological recordings ( see Materials and Methods ) . The estimated mean firing rate across the MC population increased during the initial phase of the odor response and decreased again thereafter ( Figure 1E ) . The absolute firing-rate change was , however , small because individual responses could be excitatory or inhibitory , and because responses were sparse ( see below ) . This observation is consistent with electrophysiological results [32] . The estimated population activity of INs increased more slowly , reached its peak slightly later , and was always substantially lower than that of MCs ( Figure 1E ) . The response selectivity of MCs and INs was quantified by two approaches . First , we determined how many of the 16 amino acids evoked a response for each neuron in datasets 1 and 2 . A response was counted when the distribution of TDCa signals between 0 and 1 . 5 s after stimulus onset was significantly different from the distribution of TDCa signals during spontaneous activity ( Kolmogorov-Smirnov test; p < 0 . 01 ) . Using this criterion , MCs responded , on average , to 2 . 7 ± 3 of the 16 stimuli ( mean ± SD ) , and very few MCs responded to more than ten stimuli . INs , in contrast , showed a broader range of odor selectivities ( Figure 2A ) , and some INs responded to most or all stimuli . On average , INs responded to significantly more stimuli than MCs ( 4 . 1 ± 4 . 5; p < 0 . 001 ) . Similar differences in response selectivity between MCs and INs were observed with a less stringent criterion for the detection of responses ( Kolmogorov-Smirnov test; p < 0 . 05; unpublished data ) . Second , we quantified response selectivity as a function of time by the sparseness of excitatory response profiles ( Figure 2B ) . This measure is one when a neuron responds with excitation to only one odor and zero when a neuron does not discriminate between stimuli [32 , 56] . The mean sparseness of MC response profiles was relatively high and almost constant throughout the odor response , consistent with electrophysiological results [32] . Sparseness of IN response profiles , in contrast , was lower , particularly during the early phase of a response . Hence , INs responded , on average , less selectively to odors than MCs . To quantify the density of population activity , we calculated the sparseness of excitatory TDCa signals across the population of MCs or INs in successive time windows . This “population” sparseness is one when only one neuron is excited by a given stimulus and zero when all neurons respond equally . The mean population sparseness of MC activity was relatively high and almost constant throughout the odor response , consistent with electrophysiological observations [32] . The population sparseness of IN activity patterns , in contrast , was substantially lower and increased at later time points ( Figure 2C; see also Figure 1B2 ) . The density of the IN population response was therefore higher than that of the MC population response . MC activity patterns evoked by closely related stimuli in zebrafish are initially similar but become continuously more distinct during the first few hundred milliseconds of an odor response [32–34 , 51] . This pattern decorrelation may be involved in odor discrimination behaviors and other tasks [57–60] . The similarity relationships between IN activity patterns are , however , unknown . We therefore measured IN activity patterns by TDCa imaging and analyzed similarity relationships between IN activity patterns by correlation analysis and factor analysis . We first quantified the pairwise similarities between IN activity patterns evoked by the 16 odors ( dataset 2 ) by their correlation coefficients in successive time windows ( length , 256 ms ) . Results were plotted as a series of correlation matrices in which odors are arranged according to their presumed molecular similarity ( Figure 3A ) . During the first few hundred milliseconds of the odor response , all correlation coefficients were positive , indicating that even dissimilar stimuli evoked partially overlapping IN activity patterns ( e . g . , His and Lys , Figure 1B2 ) . However , highest correlation coefficients occurred near the diagonal of the correlation matrix , indicating that IN activity patterns evoked by related odors were most similar to each other . Subsequently , correlation coefficients decreased but remained elevated for a substantial period of time after stimulus onset ( >1 . 7 s ) . Highest correlations between IN activity patterns were evoked by stimulus pairs that also evoked high correlations between MC activity patterns during the initial phase of the odor response [32 , 33 , 51] . The temporal development of correlations was , however , different for MC and IN activity patterns . In order to compare the time course of correlation changes between MC and IN activity patterns , the average correlation coefficients for the ten pairs of stimuli whose activity patterns were most highly correlated during the initial phase of the odor response were plotted as a function of time ( datasets 1 and 2; Figure 3B ) . Correlations between MC activity patterns decreased substantially during the first few hundred milliseconds to an intermediate level and remained stable thereafter . Correlations between IN activity patterns , in contrast , remained high during the decorrelation of MC activity patterns and slowly decreased only thereafter ( Figure 3B ) . Similarity relationships between activity patterns were further characterized by factor analysis [61 , 62] . This method extracts a small set of “elementary” patterns , called factors , from the original set of measured patterns . The association between each original pattern and each factor is quantified by “factor loadings . ” When a dataset contains groups of similar patterns , the corresponding stimuli show high loadings on one common ( “dominant” ) factor , but low loadings on other factors . When the original patterns are dissimilar ( decorrelated ) , factor dominance is low , i . e . , each original pattern shows moderate loadings on multiple factors . Similarly , the fraction of variance explained by the factors ( “communality” ) is usually high when groups of patterns are similar , and lower otherwise . Factor dominance and communality are therefore indicators for “similarity groups” within a set of patterns . We extracted four factors from activity patterns evoked by the 16 amino acids across MCs and INs ( datasets 1 and 2 ) in successive 256-ms time windows ( Figure S3 ) . For MC activity patterns , factor dominance was initially high but decreased during the subsequent few hundred milliseconds of the response ( Figure 3C ) . For IN activity patterns , factor dominance was also maximal during the early phase of the response but decreased more slowly . Similar observations were made for the communality ( Figure 3D ) . Factor analysis results therefore show that related IN activity patterns remain similar as MC activity patterns decorrelate and become only gradually more dissimilar at later times , consistent with results from correlation analysis . Response patterns across MCs and INs were reconstructed in three spatial dimensions to analyze their topological organization . We first concentrated on the initial phase of the odor response and analyzed TDCa signals time-averaged over the first 768 ms ( shorter time windows yielded similar results ) . MC and IN activity patterns recorded in the same OB ( one experiment from dataset 3 ) are shown in Figure 4A . MC activity was widespread but not uniformly distributed . Rather , activity patterns often contained volumes in which MC activity was particularly dense ( Figure 4A , blue arrows ) . IN activity was more widespread but also exhibited volumes of dense activity that were usually associated with dense MC activity in the overlying MC layer ( Figure 4A , green arrows ) . Hence , MC and IN activity patterns contained topologically related foci of activity . The focality of activity patterns was quantified by an index that is zero when active neurons are randomly distributed and approaches one when active neurons are tightly clustered ( Materials and Methods ) . The average focality of IN activity patterns was lower than that of MC activity patterns ( Figure 4B ) . To test whether the focalities were significantly different from randomly distributed activity , the positions of neurons in measured patterns were randomly permuted . The focality of randomized patterns was always near zero and significantly different from that of activity patterns across MCs or INs . The gross distribution of activity across MCs and INs during the initial phase of the odor response is therefore odor-dependent and topologically related across layers . We also estimated the focality of glomerular activity patterns from Ca2+ signals measured in glomerular neuropil regions ( Figure 1A ) in the same OBs as MC activity patterns ( dataset 1 ) . Ca2+ signals were time-averaged over 2 . 4 s and not temporally deconvolved , because transmitter release from olfactory sensory neurons is nearly linearly related to the intracellular Ca2+ concentration [63] , and because responses of sensory neurons in zebrafish do not change much over time [32 , 34] . The focality of glomerular activity patterns was not significantly different from that of MC activity patterns but higher than that of IN activity patterns ( Figure 4B ) . Because Ca2+ signals from glomerular neuropil may not exclusively reflect the activity of sensory inputs , we also quantified the focality of glomerular Ca2+ signals evoked by the same set of stimuli after selectively loading a dextran-coupled Ca2+ indicator into sensory neurons ( data from ref . [9]; measured using wide-field optics ) . Again , the focality of glomerular activity patterns was not significantly different from the focality of MC activity patterns ( p = 0 . 38 ) but significantly higher than the focality of IN activity patterns ( p < 0 . 001 ) . These results suggest that the focality of MC activity patterns during the initial phase of the odor response reflects , at least in part , the focality of glomerular inputs . Biochemical and physiological studies in different fish species indicate that groups of amino acids sharing certain chemical features stimulate overlapping , albeit not identical , sets of odorant receptors and olfactory receptor neurons ( ORNs ) , whereas amino acids with different chemical features stimulate largely non-overlapping sets of receptors and ORNs [9 , 64 , 65] . Based on these studies , amino acids can be assigned to chemical groups termed “long-chain” ( e . g . , Ile and Val ) , “aromatic” ( e . g . , Trp , Tyr , and Phe ) , and “basic” ( e . g . , Lys and Arg ) , after the feature that seems to influence receptor binding . Stimuli of the same chemical group tend to activate common glomeruli within a defined region so that the associated chemical features are represented in a chemotopic fashion across the array of glomeruli [9] . We therefore examined whether stimuli of these chemical groups evoke overlapping and localized response patterns also in neurons downstream of glomeruli . MC activity patterns evoked by stimuli of the same chemical group ( e . g . , Ile and Val , Trp and Tyr , and Lys and Arg ) overlapped substantially and had a focus in the same region ( Figure 4A ) . Moreover , foci of activity evoked by stimuli from different chemical groups were spatially separated from each other . Foci of activity evoked by long-chain , aromatic , and basic amino acids were consistently found in a posterior , central , and anterior location , respectively ( Figure 4A ) , in different animals . These locations correspond to the location of foci in glomerular activity patterns associated with the same chemical features [9] . IN activity evoked by stimuli of different chemical groups was generally more widespread but was also densest in regions corresponding to foci of glomerular or MC activity . Hence , the rough chemotopic map observed at the level of glomerular inputs appears to be reflected in MC activity patterns and , to a lesser extent , in IN activity patterns shortly after response onset . We next analyzed three-dimensional activity patterns as time series in successive 256-ms time windows to examine how the topological organization of activity patterns changes during the reorganization of OB output activity . Shortly after response onset , MC activity patterns contained foci of activity at locations typical for the chemical feature of the stimulus ( Figure 5A , blue arrows ) , consistent with the observations in activity patterns time-averaged over the first 768 ms ( Figure 4A ) . During the subsequent few hundred milliseconds , however , foci became less pronounced and activity became more evenly distributed ( Figure 5A; Videos S1–S3 ) . IN activity patterns also changed over time , but broad foci were still observed a few hundred milliseconds after response onset ( Figure 5B; Videos S1–S3 ) . We then quantified the focality of MC and IN activity patterns in successive time bins . The average focality of MC activity patterns decreased significantly during the first few hundred milliseconds ( r = −0 . 92 , p < 0 . 001 ) . The decrease in focality was particularly pronounced for those patterns that were initially highly focal ( Figure S4A ) . The focality of IN activity patterns was lower initially , remained almost constant during the subsequent phase of the odor response , and decreased only slightly at late time points ( Figure 6A; r = −0 . 93; p > 0 . 001 ) . Focalities did , however , remain significantly different from the focality of randomized patterns . Different focality indices yielded similar results ( Figure S4B and S4C ) . OB output activity is therefore topologically reorganized and becomes more uniformly distributed during the first few hundred milliseconds of an odor response . Odor-evoked MC activity may become more evenly distributed because activity is sparsened within foci , or because it becomes denser outside the foci . To distinguish between these possibilities , we quantified population sparseness separately within and outside the volumes corresponding to foci . For each activity pattern , the centroid ( “center of mass , ” where the mass is representing the TDCa signal ) was determined 256 ms after response onset . For each stimulus , MCs within a 50-μm radius from the centroid were classified as belonging to the focal volume , whereas MCs more than 50 μm away from the centroid were classified as outside the focal volume ( Figure 6B ) . The average sparseness of activity within the focal volume was initially lower than that outside the focal volume ( Figure 6C ) , reflecting the high density of active MCs in the focus . During the first few hundred milliseconds , however , sparseness in the focal volume increased markedly and approached the sparseness outside the foci . Outside the focal volume , the sparseness remained almost constant . Consistent with these observations , the average TDCa signal of neurons within foci was initially higher than outside foci but then decreased substantially ( Figure 6D ) . Hence , OB output activity becomes more evenly distributed because MC activity patterns are locally sparsened within the initial foci . This result is consistent with the visual inspection of three-dimensional MC activity patterns in successive time bins ( Figure 5A ) . The overlap of MC activity within foci is likely to contribute significantly to the initially high correlation between MC activity patterns evoked by chemically similar stimuli . We therefore examined whether the local sparsening of MC activity patterns contributes to pattern decorrelation during the first few hundred milliseconds . Local sparsening in the focus would promote decorrelation if it is odor-specific , i . e . , if different MCs continue to be active in response to chemically similar stimuli . To address this question , we overlaid time series of three-dimensional activity patterns evoked by chemically related stimuli . As shown in Figure 7A , the overlap of MC activity patterns evoked by stimuli from the same chemical group ( Trp and Tyr; aromatic ) was initially highest in the focus ( black spheres in overlay; arrow ) and decreased as activity patterns were locally sparsened . Similar observations were made when activity patterns evoked by stimuli from other chemical groups were compared ( Figure 7B; “long-chain”: Ile and Val; and “basic”: Arg and Lys ) . Activity patterns evoked by stimuli from different chemical groups ( e . g . , Lys and Tyr; Figure 7C ) showed little overlap throughout the odor response . Hence , local sparsening of MC activity patterns within foci is stimulus-specific and contributes to the decorrelation of OB output activity patterns . Overlays of IN activity patterns evoked by chemically similar stimuli ( e . g . , Lys and Arg ) showed little or no reduction in the overlap during the initial phase of the odor response ( Figure 8A ) , consistent with the observation that correlations between IN activity patterns remain high during the decorrelation of MC activity patterns ( Figure 3A and 3B ) . The overlap between IN activity patterns evoked by dissimilar stimuli ( e . g . , Lys and Tyr ) was lower but still substantial ( Figure 8B ) . These observations are consistent with results from correlation analysis ( Figure 3A ) . In a chemotopic map , different molecular features are associated with neural activity in different regions . The visual inspection of activity patterns revealed that stimuli of different chemical groups evoke foci of activity in distinct regions during the early phase of an odor response ( Figures 4A , 5A , and 7 ) , suggesting the presence of a transient chemotopic map . In order to directly address this question , we tested whether neurons representing a molecular feature are indeed spatially clustered , using factor analysis . A factor is a hypothetical activity pattern that represents the “elementary” molecular feature shared by a set of stimuli associated with the factor [61 , 62] . The spatial distribution of activity within factors therefore directly reflects the topology of feature representations . Moreover , the focality of factors provides a quantitative measure of chemotopy . We first concentrated on the initial phase of the odor response and analyzed MC and IN response patterns time-averaged over the first 768 ms ( shorter time windows yielded similar results ) . In addition , we extracted factors from glomerular activity patterns , estimated as described above . Four factors were extracted from response patterns to the 16 odors in datasets 1 and 2 and to the nine odors in dataset 3 . The association between stimuli and factors was very similar for glomerular , IN , and MC activity patterns: in all OBs except for one OB in dataset 1 , three of the four factors were consistently associated with the chemical groups of stimuli defined previously ( “basic , ” “aromatic , ” and “long–chain”; Figures 9A and S3 ) . The fourth factor was associated with variable groups of odors in different animals and therefore was excluded from further analysis . Three-dimensional reconstructions of the extracted factors had clear foci of activity ( Figure 9B , arrows ) . The focalities of factors extracted from glomerular and MC activity patterns were not significantly different , whereas the focality of IN activity patterns was lower ( Figure 9C ) . However , the average focality of factors extracted from all datasets was significantly different from the focality of randomized patterns ( Figure 9C ) . Foci of activity in IN factors were associated with foci in the corresponding MC factors in more superficial layers ( Figure 9B ) . Moreover , foci of activity in factors corresponded to foci in patterns evoked by stimuli associated with each factor ( compare Figures 9B and 4A ) . Hence , activity patterns across MCs and INs contain topologically related chemotopic maps during the early phase of an odor response . The extracted factors accounted only for approximately 50%–70% of the variance in the measured activity patterns ( Figure 3D ) . Each individual activity pattern therefore also had a unique component . These unique components ( “residuals” ) contain not only trial-to-trial variability , but also information about the precise identity of a given odor . We therefore reconstructed residuals in three dimensions and found that they were substantially more widespread than factors ( Figure 9D and 9E ) . Active neurons in residuals were found within the volumes corresponding to foci in factors as well as in the surrounding areas . Thus , information about precise odor identity appears to be widely distributed . Previous low-resolution imaging of sensory input to glomeruli revealed foci of activity representing “basic , ” “aromatic , ” and “long-chain” features also in patterns of glomerular input activity in zebrafish [9] . These foci are segregated along the anterior–posterior axis and preserved across individuals . We therefore examined whether chemotopic foci of MC and IN activity during the initial phase of the odor response are preserved between individuals and related to glomerular input activity . For each OB , the coordinates of glomeruli , MCs , and INs were centered on the centroid of the activity in the corresponding aromatic factor . Activity in the three factors was then binned in 40 × 40 × 40-μm3 voxels and averaged over individuals . When the geometric arrangement of chemotopic foci is consistent in different individuals , the distribution of activity in the averaged factors should be focal; otherwise , activity should be uniformly distributed . Figure 10 shows three-dimensional plots of averaged factors . For easier inspection , two-dimensional projections into a horizontal plane are shown separately in the floor patterns of three-dimensional plots ( Figure 10 ) and in Figure S5 . In all averaged factors , activity was concentrated within a circumscribed volume . Foci in different factors were segregated in an anterior-to-posterior direction and partially along the ventral-to-dorsal axis also , consistent with glomerular activation patterns recorded previously [9] . The geometrical arrangement of foci representing long-chain , aromatic , and basic features was consistent in factors extracted from glomeruli , MCs , and INs . Hence , chemotopic maps during the initial phase of an odor response are at least partially preserved between individuals and topologically related across layers of the OB . We then examined how chemotopy is affected by the local sparsening and topological reorganization of activity patterns during the initial phase of the odor response . Factors were extracted from MC and IN activity patterns in successive 256-ms time windows , and chemotopy was quantified by the focality of factors . The focality of factors extracted from IN activity patterns remained relatively stable . The focality of MC factors , in contrast , was initially high but then decreased during the first few hundred milliseconds after response onset ( Figure 11A ) , similar to the focality of MC responses to individual odors ( Figure 6A ) . The focality of factors extracted from both MCs and INs did , however , remain significantly different from the focality of randomized patterns ( Figure 11A ) , indicating that chemotopy did not vanish completely . Similar results were obtained using different focality indices ( Figure S4D and S4E ) . These data show that the chemotopy of OB output activity is substantially reduced during the topological reorganization of activity patterns in the OB . To confirm these results by an independent analysis , we quantified the similarity of response profiles of individual neurons as a function of their distance . In chemotopically organized patterns , response profiles of nearby neurons are expected to be , on average , more similar than response profiles of distant neurons . Shortly after response onset , the mean correlation of response profiles was highest for nearby neurons and decreased with distance ( Figure 11B , yellow curve ) . During the subsequent reorganization of activity patterns , response profiles of nearby neurons became less similar and the decrease in response profile similarity with distance became less pronounced ( Figure 11B , orange to black curves ) . The distance dependence of the average correlation between IN response profiles was not as pronounced as for MCs , and the decrease in response profile similarity occurred later during the odor response ( Figure 11B ) . These data confirm that the chemotopy of OB output activity patterns decreases during the topological reorganization of activity patterns . As chemotopic maps disappeared , the fraction of the variance contained in unique pattern components ( residuals ) increased ( Figure 3D , dashed lines ) and OB output activity becomes more informative about the precise identity of odors [32–34] . One possible mechanism underlying this effect is that after odor-specific local sparsening , unique MC activity patterns emerge within areas that previously contained foci . In this scenario , residuals would be expected to become more focal over time because unique components would be increasingly dominated by neurons within the initial foci . Alternatively , the increased uniqueness of MC activity patterns may be caused by an increased contribution of odor-specific firing patterns outside foci because MCs within foci are generally inhibited . In this scenario , residuals should also become more focal because they would be dominated by activity outside initial foci . Both mechanisms could , however , also coexist . In this case , the focality of residuals should approach zero because odor-specific information would be conveyed by widely distributed MCs . For both MC and IN activity patterns , the average focality of residuals was low throughout the odor response . The focality of MC residual patterns approached zero and became indistinguishable from that of randomized patterns ( Figure 11C ) . Hence , information about precise odor identity appears to be conveyed by distributed sets of neurons both within and outside areas of initial foci . This result is consistent with the visual observation that MCs responding specifically to one of two similar stimuli are widely distributed after local sparsening ( Figure 7A and 7B ) . In summary , the decorrelation of MC activity patterns ( Figure 3B ) , the reduction in overlap between related MC activity patterns ( Figure 7A and 7B ) , the decrease in factor dominance ( Figure 3C ) , and the decrease in the communality of extracted factors ( Figure 3D ) indicate that MC firing-rate patterns become less informative about the molecular features that were initially represented in spatial maps . Factor analysis and the distance dependence of response profile similarity directly demonstrate that chemotopy of OB output activity patterns becomes substantially reduced during the first few hundred milliseconds of an odor response . The initial chemotopic mapping of chemical features is therefore not maintained during the topological reorganization of OB output activity .
In insects , INs were found to respond less selectively to odorants than projection neurons [66 , 67] . In vertebrates , biophysical properties of INs have been characterized in vitro [e . g . , 26 , 68–70] , and synaptic input and action potential firing of individual INs during odor responses have been examined in vivo [71–75] . Odor-evoked activity in populations of INs has been visualized at low resolution by 2-deoxyglucose uptake or immediate early gene expression [6 , 11 , 38–40] . To our knowledge , however , response profiles of individual INs to defined panels of relevant odors have not been studied systematically before . Although MCs usually responded with excitation only to a few amino acids , a subset of INs was excited by most or all stimuli . This broad tuning may result from the convergence of MCs with different tuning profiles onto individual INs , from nonspecific centrifugal inputs , or both . Other INs , however , responded more selectively to odors . The population of INs therefore exhibited a wider range of odor selectivities than the population of MCs . Conceivably , INs with different tuning properties could subserve different functional roles . For example , broadly tuned INs may mediate general operations , such as synchronization or global gain control , whereas more selective INs may shape OB output patterns in an odor-dependent manner . It will now be interesting to determine whether INs with different response selectivities indeed subserve different functions , and whether they correspond to different types of neurons , such as periglomerular and granule cells . Activity patterns across the population of INs were less sparse than MC activity patterns , particularly during the initial phase of the odor response . The spatial distribution of responding INs was not uniform , but more widespread than MC activity shortly after response onset . This may be the consequence of the divergent , yet spatially restricted , connectivity between MCs and INs . IN response patterns evoked by chemically related stimuli were more similar to each other than response patterns evoked by structurally different stimuli , as observed in MCs [32 , 51] . Nevertheless , the similarity relationships between IN activity patterns differed from those of MC activity patterns in at least two ways . First , IN response patterns were positively correlated even when odors were structurally dissimilar . MC response patterns evoked by dissimilar stimuli are , in contrast , nearly uncorrelated during the initial phase of the odor response . Second , correlations between IN activity patterns remain high during the first few hundred milliseconds , when MC activity patterns are decorrelated , and gradually decrease only at later time points . As MC activity patterns become decorrelated , INs therefore provide distributed , yet somewhat focal , patterns of inhibitory feedback that is related , but not identical , for similar stimuli . This topologically organized feedback may ensure dense inhibitory interactions between MCs receiving similar inputs . Our results also demonstrate that IN activity patterns change during the course of an odor response in a stimulus-specific manner , consistent with the observation in brain slices that individual granule cells respond to brief electrical stimulation in superficial layers with reproducible , glomerulus-dependent latencies of up to approximately 900 ms [76] . Hence , the pattern of IN activity changes in a stimulus-dependent fashion in response to sensory input , although the pattern changes are qualitatively different from those observed in MCs . The resulting dynamics of inhibitory feedback onto the MC population is likely to contribute to the temporal patterning of OB output activity . MC activity patterns were focal shortly after response onset but became more evenly distributed during the subsequent few hundred milliseconds . Hence , activity patterns transmitted from the OB to higher brain regions are topologically reorganized during pattern processing . This topological reorganization of OB output activity was associated with a disappearance of chemotopic maps . The chemotopic organization of glomerular activity patterns is therefore only transiently reflected at the next stage of the olfactory pathway . The relationships between activity patterns across different populations of neurons provide initial insights into the mechanisms underlying pattern reorganization . During the early phase of an odor response , MC and IN activity patterns were topologically related to glomerular activity maps , suggesting that MC and IN activity patterns are initially driven to a large extent by glomerular input . The subsequent reorganization of activity patterns , however , implies that the chemotopic organization is unstable . This may be expected because inhibitory feedback from INs modifies MC activity , which in turn changes IN activity patterns . Because patterns of inhibitory feedback evoked by different stimuli are not identical , the pattern of inhibitory feedback onto MCs would modify OB output activity in an odor-specific manner . The reorganization of activity patterns is therefore likely to result , at least in part , from interactions between MCs and INs . After a few hundred milliseconds , OB output activity patterns stabilize [34] , implying that feedback interactions between the populations of MCs and INs approach a dynamic equilibrium . An important mechanism contributing to the topological reorganization of MC activity patterns is the local sparsening of MC activity within initial foci . Foci in the MC layer were associated with dense IN activity in deeper layers , suggesting that local sparsening of MC activity patterns is caused by topologically related inhibitory feedback from INs . One prediction from this hypothesis is that the focality of IN activity should be lower than the initial focality of MC activity because connections from MCs to INs are divergent . This was indeed observed . A second prediction is that inhibitory feedback should reduce , but not completely abolish , the focality of MC activity patterns to maintain a dynamic equilibrium . Indeed , the focality of MC and IN activity patterns remained significantly different from that of randomized patterns ( Figures 6A and 11A ) . A third prediction is that MC activity should not spread from the focus to adjacent regions because , unlike in the insect antennal lobe [67 , 77] , extensive interglomerular excitatory interactions between principal neurons have not been described in the vertebrate OB ( but see [78 , 79] ) . Consistent with this hypothesis , the density of activity outside foci was stable during the odor response ( Figure 6C ) . Finally , the reduction of chemotopy by topologically related inhibitory feedback and local sparsening could be reproduced in simulations ( unpublished data ) . Hence , the topological reorganization of MC activity can be explained , at least in part , by local sparsening through divergent , yet topologically related , inhibitory feedback from INs . The general architecture of the OB and the hierarchical chemotopy of glomerular activation patterns are conserved in different vertebrate classes , suggesting that a similar topological reorganization of OB output patterns may also occur in other vertebrate species . How may the transient chemotopic organization of activity patterns influence the function of neuronal circuits in the OB ? In parallel with the topological reorganization of activity patterns , initially similar MC activity patterns become decorrelated [32–34 , 51] . This computation increases the discriminability of odor representations and may be involved in odor discrimination behavior [57–59] . In theory , the topological reorganization and the decorrelation of MC activity patterns could occur independently . For example , chemotopy could be maintained during pattern decorrelation if activity was redistributed locally within foci . Likewise , correlations could be maintained despite a topological reorganization if all patterns were reorganized in the same manner . Alternatively , the topological reorganization of activity may be functionally involved in pattern decorrelation . In this case , the two processes would depend on each other and cannot be separated . Detailed comparisons of three-dimensional activity patterns revealed that the overlap between MC responses to related stimuli is initially high within foci . Subsequently pattern overlap was decreased by odor-specific local sparsening , presumably because INs inhibit different subsets of MCs in response to each stimulus . Hence , local sparsening contributes both to the topological reorganization and to the decorrelation of MC activity patterns . This conclusion is further supported by factor analysis results . Using this technique , we extracted the pattern components that accounted for the high initial similarity of related patterns ( factors ) , as well as the components that were unique to each pattern ( residuals ) . The initial focality of factors demonstrated that pattern overlap was most pronounced in those regions where MC activity was dense , indicating that activity within foci dominates the initial correlation of related patterns . Unique components of activity patterns containing stimulus-specific information were , in contrast , widely distributed . Together , these results indicate that local sparsening contributes to pattern decorrelation in two ways: first , local sparsening decreases pattern overlap within regions of initially dense activity . Second , local sparsening reduces the overall activity within regions of greatest overlap so that the contribution of more-specific pattern components becomes enhanced . Local sparsening therefore promotes both the decorrelation and the topological reorganization of OB output patterns , indicating that the two phenomena are linked . One function of the transient chemotopy in the OB may therefore be to enable local sparsening and thereby enhance the decorrelation of related odor representations . Further experiments are required to explore this hypothesis . The topological reorganization of activity patterns occurred less than 1 s after response onset . Individual components of activity patterns may , however , be reorganized more rapidly . In electrophysiological recordings , complete pattern decorrelation is observed approximately 800 ms after response onset when activity patterns are analyzed across all recorded MCs [32] . However , decorrelation specifically occurs within odor-specific components of the overall firing-rate patterns that are defined by the absence of phase locking to the local field potential oscillation . More specific analyses of these components revealed that pattern decorrelation is nearly complete already after 400 ms [33] . The time window during which a topological reorganization of the overall activity patterns was observed should therefore be considered as an upper time limit for the processing of specific pattern components . We concentrated our analysis on the ventrolateral region of the zebrafish OB where secondary features of amino acids are mapped in “fine” chemotopic patterns . In addition , the OB of zebrafish and other vertebrates also exhibits a “coarser” chemotopic organization in which primary chemical features ( “chemical classes” ) are mapped to large , distant areas [11 , 12 , 17 , 21 , 80] . In zebrafish , for example , amino acids and bile acids activate glomeruli predominantly in the ventrolateral and dorsomedial OB , respectively [10] . This “coarse” chemotopic organization may not be affected by the topological reorganization observed here because the underlying sparsening of MC activity patterns is more local . It will now be interesting to examine whether “coarse” chemotopy is maintained and enables the parallel processing of distinct odor classes . Within the amino acid–sensitive region , the chemotopic organization of OB output patterns is transient , raising the question as to what the function of fine chemotopic maps may be for odor processing . One possibility is that downstream neurons use the transient chemotopic map in the OB as a coordinate system to access sensory information about distinct chemical features during the initial phase of an odor response . Alternatively , the transient chemotopy may organize synaptic interactions within the OB . In general , a topographic feature map does not , per se , encode information , because the information conveyed by an activity pattern is independent of the neurons' positions . In a neuronal circuit , however , connections are usually formed between neurons within a limited range . As a consequence , the set of neurons that interacts in response to a given stimulus is defined not only by the given connectivity , but also by the spatial arrangements of inputs . One potential function of topographic feature maps is therefore to specify important functional interactions between neurons in a circuit . In the OB , chemotopy would favor mutual inhibition of MCs receiving similar inputs , which is expected to promote local sparsening and pattern decorrelation . We therefore propose that the chemotopic organization of activity patterns in the OB is an important feature that configures computational properties of the circuit even though it is not maintained during pattern processing .
Zebrafish were kept at 26–27 °C on a 13/11-h light/dark cycle under standard conditions . The transgenic line , HuC:YC [81] , expressed the fluorescent protein , YC 2 . 1 [82] , under the control of a fragment of the HuC promoter . In the adult zebrafish OB , HuC:YC is expressed selectively in MCs [54] . YC fluorescence did not change in response to odor stimulation and was used exclusively as an anatomical marker . No obvious differences were observed between MC odor responses in HuC:YC or wild-type fish recorded by loose-patch recordings , whole-cell recordings , or Ca2+ imaging . Experiments were performed in an explant preparation of the intact brain , nose , and other sensory organs [32] . Briefly , adult ( >3 mo old ) zebrafish were cooled to 4 °C and decapitated in artificial cerebrospinal fluid [83] . After removal of eyes , jaws , and bones covering the ventral telencephalon , the preparation was placed in a flow chamber and slowly warmed up to room temperature . All animal procedures were performed in accordance with the animal care guidelines issued by the Federal Republic of Germany . Odors were delivered through a constant flow directed at the ipsilateral inflow naris using a computer-controlled , pneumatically actuated high performance liquid chromatography ( HPLC ) injection valve ( Rheodyne , http://www . rheodyne . com ) . From amino acids of the highest available purity ( Fluka; Sigma-Aldrich , http://www . sigmaaldrich . com ) , frozen stock solutions ( 1 or 10 mM ) were diluted to a final concentration of 10 μM immediately before the experiment . The minimum interstimulus interval was 90 s to avoid adaptation . The standard stimulus set included 16 amino acids used in previous studies [9 , 32 , 33] and a blank , which did not evoke responses . Neurons in the OB were loaded with the red-fluorescing Ca2+ indicator , rhod-2-AM , by bolus injection and imaged by two-photon microscopy as described [51] . Fluorescence imaging was performed with a custom two-photon microscope equipped with a mode-locked Ti:Sapphire laser ( Mira900 , 100 fs , 76 MHz; 830–850 nm; pumped by a 10-W Verdi laser; Coherent , http://www . coherent . com ) , a 20× , N . A . 0 . 95 objective lens ( Olympus , http://www . olympus . co . jp/en/ ) , and external detection optics . Rhod-2 and YC fluorescence were detected simultaneously in separate emission channels using bandpass filters of 515/30 nm and 610/75 nm , respectively . Laser intensity was adjusted to minimize photobleaching in each focal plane . Time series of rhod-2 images were converted to image series representing the relative change in fluorescence , F , in each pixel after stimulus onset ( ΔF/F ) Image series in dataset 1 were acquired at 128 ms/frame and contained 128 × 256 pixels , covering a field of view of 163 × 81 . 5 μm2 . Images in datasets 2 and 3 were acquired at 256 ms/frame and contained 256 × 256 pixels , covering 163 × 163 μm2 . Time values represent the starting time of frames . The dataset to quantify the stability of responses ( Figure S2 ) was acquired using the same parameters as in datasets 1 ( MCs ) or 2 ( INs ) . t = 0 designates stimulus onset , which was determined as the first frame in which a response was observed . Because it is unclear when the response started within this frame , the first frame was omitted from analyses . In each dataset , responses to the same sets of odors were collected in multiple optical sections at different x , y , and z positions throughout the volume of interest . The z-distance between focal planes that overlapped in x and y was 15 μm or greater . The average soma diameter of MCs is approximately 10 μm [32]; INs are smaller . Three-dimensional activity patterns were reconstructed from responses measured in 5–18 different optical sections in each dataset . In dataset 1 , optical sections were distributed throughout the glomerular/MC layer in the ventrolateral OB . In datasets 2 and 3 , optical sections also covered the central OB , where many granule cells are located . Dataset 1 is identical to the MC dataset described in [51] . The number of MCs recorded in each optical section was lower than that of INs because the density of MCs is much lower . Neuronal somata were outlined manually based on YC fluorescence ( MCs ) or rhod-2 fluorescence and Ca2+ signals ( INs ) . Firing-rate changes were reconstructed from Ca2+ signals by temporal deconvolution as described [51] . Parameters used for temporal deconvolution were τdecay = 3 s and thrnoise = 1% for MCs , and τdecay = 6 s and thrnoise = 1% for INs . These parameters are within the range yielding optimal reconstruction of firing-rate patterns , as determined previously [51] . As shown previously [51] , differences in dye concentration or variability of parameters across neurons only minimally affects the reconstruction of population activity patterns . Sparseness is a measure for the “peakiness” of a distribution that was derived by Vinje and Gallant [56] from a related measure developed by Rolls and Tovee [84] . The sparseness S was calculated as described [32 , 56] after setting negative values to zero: S = {1 − [ ( Σrn/N ) 2/Σ ( rn2/N ) ]}/[1 − ( 1/N ) ] , where rn is the nth response ( response of the nth MC for population sparseness or response to the nth odor for sparseness of response profile ) and N is the total number of responses ( responses of different neurons to the same stimulus for population sparseness or responses to different stimuli for sparseness of response profiles ) . Sparseness is a value between zero and one and is independent of the number of responses N in the dataset . To test whether the slow decrease in correlation between IN activity patterns ( Figure 3A and 3B ) is a consequence of the increasing population sparseness of IN activity ( Figure 2C ) , we artificially increased the sparseness of IN response patterns . Random subsets of IN responses ( up to 50% ) were replaced by noise with an amplitude and distribution identical to the spontaneous activity of each IN . The sequence of correlation matrices remained similar after artificial sparsening , indicating that the slow decrease of correlation coefficients cannot be attributed to the increase in population sparseness . Factor analysis was performed by a principal component analysis followed by varimax rotation and promax transformation , as described previously [9 , 32 , 62] . The number of factors was set to four based on results from correlation analysis and previous studies [9 , 32 , 33] . Factor dominance was quantified as described [32] and normalized onto the interval ( 0 , 1 ) by dividing by the maximum possible value ( Imax = 0 . 1875 for four factors ) . For the quantification of the rise time of excitatory responses , a sigmoid curve was fitted to the onset of each response . Rise time was then quantified as the time between 10% of the maximum in the fit and the peak time determined from the original trace . Inhibitory responses were not considered . To estimate mean population firing rates of MCs and INs , all individual responses were averaged for each cell type and scaled to yield approximate mean population firing-rate changes . Scaling factors were determined previously by simultaneous TDCa imaging and electrophysiological recordings [51] . Because the TDCa signal reflects the firing-rate change relative to baseline , spontaneous firing rates , determined by electrophysiology in separate experiments , were added ( MCs: 8 . 8 Hz [34]; INs: 0 . 7 Hz [B . Judkewitz; unpublished data] ) . The estimate of the mean population firing rate is sensitive to errors in determining the scaling factors . However , the time course of the estimated MC population firing rate was similar to that determined previously by electrophysiological recordings [32] . We therefore conclude that estimates of population firing rates based on TDCa imaging are valid , although precise measurements require extensive electrophysiological recordings . The standard focality index was based on a similar index used previously [54] . Neurons with response intensities equal to or greater than 50% of the maximum response were selected from each activity pattern . This threshold is arbitrary and serves to exclude weakly active neurons that may contribute noise but are unlikely to have a substantial influence on focality . The distances between all pairs of selected neurons were weighted by the product of their response intensities ( normalized so that the sum of all weights was one ) and averaged , yielding <dselected> . The focality index f was calculated as 1 − ( <dselected>/<dall> ) , where <dall> is the average distance between all neurons . This focality index f is related to the index used previously fprevious [54] by f = 1 − fprevious; it is therefore more intuitive because low focalities are represented by values close to zero whereas extremely high focalities are represented by values close to one . The significance of focality was assessed by a statistical comparison of the measured focalities to focalities after randomly permuting the positions of all neurons . In order to test whether results depend critically on the procedure used to quantify focality , we also tested a variety of modifications of the focality index . However , all procedures yielded similar results ( Figure S4B–S4E ) , indicating that the quantification of focality using the standard index is robust . | Many sensory brain areas contain topographic maps where the physical location of neuronal activity contains information about a stimulus feature . In the first central processing center of the olfactory pathway , the olfactory bulb , chemically distinct odors often elicit spatially segregated input activity so that general chemical features are initially represented in a topographic fashion . It is , however , unclear whether this “chemotopic” organization of odor representations is maintained at subsequent stages of odor processing . To address this question , we visualized activity patterns across thousands of individual neurons in the intact olfactory bulb of zebrafish over time using two-photon calcium imaging . Our results demonstrate that odor-evoked activity across the output neurons of the olfactory bulb is chemotopically organized shortly after stimulus onset but becomes more widely distributed during the subsequent few hundred milliseconds of the response . This reorganization of olfactory bulb output activity is most likely mediated by inhibitory feedback and reduces the redundancy in activity patterns evoked by related stimuli . These results indicate that topographically organized activity maps in the olfactory bulb are not maintained during information processing , but contribute to the function of local circuits . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"danio",
"(zebrafish)",
"computational",
"biology",
"neuroscience"
] | 2007 | Topological Reorganization of Odor Representations in the Olfactory Bulb |
In countries with a high prevalence of tuberculosis there is high coincident of helminth infections that might worsen disease outcome . While Mycobacterium tuberculosis ( Mtb ) gives rise to a pro-inflammatory Th1 response , a Th2 response is typical of helminth infections . A strong Th2 response has been associated with decreased protection against tuberculosis . We investigated the direct effect of helminth-derived antigens on human macrophages , hypothesizing that helminths would render macrophages less capable of controlling Mtb . Measuring cytokine output , macrophage surface markers with flow cytometry , and assessing bacterial replication and phagosomal maturation revealed that antigens from different species of helminth directly affect macrophage responses to Mtb . Antigens from the tapeworm Hymenolepis diminuta and the nematode Trichuris muris caused an anti-inflammatory response with M2-type polarization , reduced macrophage phagosome maturation and ability to activate T cells , along with increased Mtb burden , especially in T . muris exposed cells which also induced the highest IL-10 production upon co-infection . However , antigens from the trematode Schistosoma mansoni had the opposite effect causing a decrease in IL-10 production , M1-type polarization and increased control of Mtb . We conclude that , independent of any adaptive immune response , infection with helminth parasites , in a species-specific manner can influence the outcome of tuberculosis by either enhancing or diminishing the bactericidal function of macrophages .
Infection with helminth parasites and microbial pathogens present very different challenges to the mammalian immune system , and distinct immune effector mechanisms have evolved to combat infection with these different classes of organisms . Typically , infection with microbial pathogens requires the mobilization of professional phagocytes and Th1-dominated immunity , whilst some of these effectors may play a role in the response to helminth parasites , effective eradication of metazoans is the remit of Th2 immunity and its effectors , such as antibody , mucus and eosinophils [1–3] . The geographic distribution of tuberculosis ( TB ) and endemic helminth infections are almost superimposable and many individuals with TB will be , or will have been , infected with helminth parasites [3–5] . Given the general paradigm of the reciprocal inhibition of Th1 and Th2 immune responses and increase in TB globally , a comprehensive understanding of the impact of infection with helminth parasites on the response to Mycobacterium tuberculosis ( Mtb ) and the outcome of TB is essential . Co-infection with helminth parasites and Mtb in mice and analysis of co-infected individuals has provided important , and often contrasting data , which may reflect host-parasite specificity in response to the helminths . For example , it was shown that mice infected with helminths and M . tuberculosis had a greater bacterial burden in their lungs [6–8] , and contrarily , early control of M . bovis BCG in the lungs has been reported in helminth co-infected mice [9] . Where co-infection was shown to enhance susceptibility to TB , increased Th2 cytokines were implicated [6–8]; as for instance in the case of IL-4 promoting alternatively activated macrophages ( AAMs ) that accumulated in the lungs , correlating with deficient innate anti-tuberculosis protection [8] . Similarly , AAMs were found to be less effective than IFN-γ-treated macrophages in controlling M . tuberculosis [10] . Although no specific microbicidal mechanism was defined , it is likely that the polarization status of macrophages in helminth-infected mice affects the outcome of concomitant mycobacterial infection . Furthermore , helminth-derived products can directly reduce the LPS activation of macrophages , decreasing the expression or levels of pro-inflammatory cytokines [11–15] . Although the effect of helminth products in the context of macrophage Mtb infection was not tested those findings strengthen the notion that helminths , even without the amplification or signals via the adaptive immune response , could directly stimulate a regulatory M2-like macrophage that has suppressed mycobactericidal properties . Many clinical studies indicate that infection with helminths modulate an individual’s susceptibility to TB , by , for example , increasing the risk of becoming latently infected with Mtb [16] , and co-infected patients often present with more advanced disease [17] . Clinical studies of helminth-M . tuberculosis co-infection have focused mainly on documenting ( often at a single time-point ) the levels of Th1 and Th2 cytokines , and that helminth-induced Th2 polarization ultimately reduces cellular immunity to M . tuberculosis [17 , 18] . While murine models highlight macrophages as an important target cell affected during helminth-mycobacterial co-infection [9 , 19] , the direct microbicidal mechanisms of human macrophages have not been studied . A number of scenarios would allow helminth-derived antigens ( i . e . secretory/excretory ( E/S ) products ) and mycobacteria to access the same macrophage: some intestinal helminths migrate through the lung as part of their life cycle; E/S products liberated from gastrointestinal helminths , filarial worms in lymphatics or Schistosoma species in the blood vessels would facilitate local and systemic delivery of antigens; and , the common mucosal immune system allows for the potential of phagocytes in the gut to traffic to the airways [20 , 21] . Hypothesizing that human monocyte-derived macrophages ( hMDMs ) co-exposed to M . tuberculosis and helminth antigens would have a decreased ability to handle the bacteria , a series of investigations were performed with antigens from a nematode ( Trichuris muris , T . m ) , a cestode ( Hymenolepis diminuta , H . d ) and a trematode ( i . e . Schistosoma mansoni , S . m ) –representatives of the major groups of helminth parasites that infect humans . hMDMs co-treated with H . d or T . m displayed a decrease in the maturation of M . tuberculosis-phagosomes although , remarkably , after 1h of pre-exposure with the antigens , the number of intracellular bacteria were not different between hMDMs treated with M . tuberculosis ± worm antigen; however , after 48h of pre-exposure the co-treated hMDMs had increased bacterial burdens . Thus , extending the work of others showing that helminth antigens can directly affect macrophage function , we demonstrate that antigens from specific helminth parasites diminishes the bactericidal functions of human macrophages against M . tuberculosis , and that this effect occurs independent of any adaptive immune response .
Monocytes were obtained from heparinized peripheral human blood ( Linköping University Hospital Blood Bank ) from healthy donors who had given written consent for research use of the donated blood . Blood donation is classified as negligible risk to the donors and only anonymized samples were delivered to the researchers in accordance with the Declaration of Helsinki , not requiring a specific ethical approval according to paragraph 4 of Swedish law ( 2003:460 ) on Ethical Conduct in Human Research . Whole blood was added onto a density gradient and centrifuged for 40 min at 480xg at room temperature . The layer of peripheral blood mononuclear cells ( PBMCs ) was collected , washed and seeded into flasks to adhere for 1-2h before the lymphocytes were washed away . The adherent monocytes were left to differentiate into macrophages for 7 to 9 days in DMEM containing 10% pooled non-heat inactivated natural human serum , with full medium change twice during the culture period giving a mature macrophage population [22] . To confirm that the monocytes had differentiated into macrophages and not into dendritic cells , CD1c ( dendritic cell marker ) and CD14 staining was routinely performed ( along with CD209 , CD1a , and other markers ) . With this macrophage protocol cells were CD14high and less than 2% expressing CD1c . Prior to experiments the hMDMs were seeded into either 96-well plates or 24-well plates containing cover slips . For creating Mtb-specific CD4+ T cells from naïve CD4+ T cells , in order to analyze changes in antigen presentation from Mtb infected cells , we utilized our recently established system [23] using either the entire culture filtrate from H37Rv ( e . g . PPD , purified protein derivative ) or purified Ag85B protein , the major secretory and highly immunogenic protein of Mtb [24 , 25] . To generate antigen presenting cells ( APCs ) freshly isolated monocytes were cultured in RPMI-1640 containing 5% heat inactivated human AB serum supplemented with rhGM-CSF ( 10 ng/ml ) and rhIL-4 ( 10 ng/ml ) ( both from Peprotech , USA ) for 3 days . The APCs generated in this way were CD1a/c+ . For T cell priming the generated APCs were harvested and γ-irradiated ( 25 Gy ) before being co-incubated with naïve CD4 T-cells purified from PBMCs of the same donors using the human naive CD4+ T-cell isolation kit ( Stem Cell Technologies ) , according to instructions provided by the manufacturer . Naïve CD4+ T cells ( 1x106/well ) were co-cultured with APCs ( 2 . 5x105/well ) in a 24-well plate and were stimulated with 10 μg/ml of purified protein derivative ( PPD; culture filtrates from Mtb strain H37Rv obtained from the Staten Serum Institute , CPH , Denmark ) or purified Ag85B protein ( Rv1886c from Mtb strain H37Rv obtained from BEI Resources , Manassas , USA ) . Fresh media supplemented with IL-2 ( 20 IU/ml ) were replenished once a week . The specificity test was carried out 3–4 weeks after generation of the CD4+ T-cell lines . For that , the CD4+ T cells and thawed autologous APCs were co-cultured at a 5:1 ratio along with PPD ( 10 μg/ml ) or Ag85B ( 10 μg/ml ) or Staphylococcal enterotoxin B ( SEB; 1 μg/ml , used as a positive control ) or ovalbumin ( 10 μg/ml , used as background control ) , and the level of IFN-γ in cell free supernatant was analyzed after 48h . The H . diminuta antigens were prepared from whole worm crude extract [26] , and T . muris antigens ( a kind gift from Dr . W . Khan ( McMaster Uni . Hamilton , ON , Canada ) ) from excretory and secretory products ( E/S ) from the worm into 5% PenStrep containing medium [27] . S . mansoni soluble egg antigen was from Professor Mike Doenhoff , Nottingham University , Nottingham UK . Stock concentrations of antigens in PBS were confirmed by Bradford assay and aliquots stored at -80°C until use . The hMDMs were treated with the helminth antigens at the concentrations indicated in the legends , starting with concentrations previously used for H . diminuta ( 100 μg/ml; [26] ) and T . muris ( 50 μg/ml; [11] ) . Antigens were added 1h ( Figs 1–6 ) or 48h ( Figs 6–8 ) prior to infection with M . tuberculosis , as specifically stated in the figure legends . Since concentrations above 3 μg/ml of T . muris had a direct mycobactericidal effect , 1 . 5 μg/ml across all antigens were chosen when evaluating the intra-macrophage killing capacity of Mtb . All helminth antigens were free from LPS contamination , i . e . found below the detection limit of Pierce LAL Chromogenic Endotoxin Quantification Kit using 1 mg/ml of the individual antigens . M . tuberculosis ( Mtb ) H37Ra or H37Rv ( only for Fig 7 ) were grown at 37°C in Middlebrook 7H9 broth supplemented with 0 . 05% Tween-80 and 10% ADC enrichment ( Becton Dickinson ) with 20 μg/ml kanamycin for green fluorescence protein ( GFP ) -expressing Mtb or 100 μg/ml hygromycin for luciferase-expressing Mtb . Log phase bacterial cultures was centrifuged two times for 5 min at 5000xg and the bacteria were separated by needle shearing first in PBS-Tween-80 ( 0 . 05% ) and then in serum-free medium . The OD value was measured to determine the concentration and for calculation of multiplicity of infection ( MOI ) . After pretreatment with the antigens , Mtb was added to the hMDMs at different MOI ( 1 , 2 , 5 or 10 ) and time-points as indicated in the figure legends . Antigen presentation experiments ( Fig 7 ) were performed with H37Rv that was heat inactivated for 1h at 70°C prior use; for all other experiments , live H37Ra was used . We have recently shown that both M . tuberculosis H37Ra and H37Rv infect and can replicate in hMDMs ( although H37Rv replicate to a greater extent ) , and that both the strains can manipulate/block the autophagy pathway [28] . After 1h pretreatment with antigens and 2h infection with GFP-Mtb , LysoTracker deep red ( LTDR ) ( Life technologies ) was added to a final concentration of 75 nM for 30 min to visualize acidic compartments . The cells were detached with accutase ( Stem Pro accutase , Gibco by life technologies ) and fixed in 4% PFA before being run in a Gallios flow cytometer ( Beckman coulter ) . The quantitative LTDR signals were analyzed with Kaluza software version 1 . 2 ( Beckman Coulter , USA ) . Qualitative assessment of phagosomal acidification using LTDR was performed with hMDMs adhered to coverslips [28] . hMDMs were pretreated with helminth antigens for 1h before infection with GFP-Mtb , or FITC-labelled yeast as a positive control for phagolysosomal maturation . Negative controls were pretreated with 100 nM bafilomycin ( from streptomyces griseus , Sigma Aldrich ) prior to infection with Mtb . LTDR was added for the last 2h of the 4h Mtb infection before fixation and staining with wheat-germ agglutinin ( WGA ) AF350 ( Life technologies ) 1 μg/ml for 20 sec followed by mounting . All cover slips were analyzed in a LSM 700 Zeiss inverted confocal microscope and the images were observed in a blinded manner . The brightness of the images was increased for visualization purposes only , after the completion of the analysis . Helminth treated hMDMs were infected with Mtb at different MOI for 2h or 24h . Time-points for analyzing cytokine secretion was based on our previous analysis for pro- and anti-inflammatory cytokines secreted from stimulated hMDMs ( including Mtb exposure ) , were TNF-α was secreted as early as 2h after exposure whereas other cytokines were more robustly secreted and detected after 18-24h of stimulation [29] . The medium supernatants from triplicate wells of each treatment were pooled , cleared from cellular debris , and stored in -70°C until assayed . TNF-α , IL-12p40 , IL-6 , IL-1β , and IL-10 levels were determined by cytometric bead array analysis , performed according to the manufacturer’s protocol ( BD Biosciences ) . Detection of cytokines was performed by flow cytometry ( Becton Dickinson ) and cytokine concentrations were analyzed using Kaluza software version 1 . 2 ( Beckman Coulter , USA ) . hMDMs exposed to helminth antigens and infected with different MOI of Mtb were stained with different macrophage polarization markers . For positive controls of M2 macrophages , IL-4 ( 20 ng/ml ) or IL-10 ( 2 ng/ml ) was added to induce M2a or M2c macrophages , respectively [30] . A cocktail of 100 ng/ml LPS and 100 U/ml IFN-γ was added as a positive control for induction of pro-inflammatory M1 macrophages . Stimulated and infected hMDMs were detached with accutase ( Stem Pro accutase , Gibco by life technologies ) and centrifuged at 900xg for 5 min prior to adding 10 μl DC-SIGN/CD209 PerCP for 15 min at room temperature ( RT ) . Twenty μl of a mix of the antibodies; CD206-FITC , CD163-PE , CCR7-AF647 , was added with the dilutions 1:26 . 6 for CCR7 and 1:6 . 6 for the rest and incubated for 25 min RT . Fluorescence minus one ( FMO ) -controls received an antibody mix lacking one of the respective antibodies . The samples were acquired on a Gallios flow cytometer ( Beckman Coulter ) , and the hMDMs were analyzed using Kaluza software 1 . 2 where the % marker positive hMDMs were evaluated based on their respective FMO-control . hMDMs exposed to helminth antigens were infected with Mtb ( MOI = 5 ) for 24h and treated with brefeldin A ( 5μg/ml ) the last 22h before being detached with trypsin EDTA and stained intracellularly using a PE mouse anti-human CCL22 antibody ( clone T51-719 ) using Cytofix/Cytoperm according to the manufacturer’s protocol ( BD Biosciences ) . A PE isotype matched control antibody was used at same concentration as the specific CCL22 antibody ( 4μg/ml; determined by titration ) to set the positive gate . IL-4 and a LPS/IFN-γ cocktail was used as positive stimuli for M2 and M1 macrophage , respectively , at concentrations indicated above . The samples were acquired on a Gallios flow cytometer ( Beckman Coulter ) , and the hMDMs were analyzed using the software FlowJo ( version 10 . 1 ) where the % CCL22 positive hMDMs were evaluated based on gates set by the PE isotype control antibody . After antigen treatment for 1h or 48h , Mtb expressing luciferase were added for 1 . 5h . Extracellular bacteria were washed away and new medium with re-added antigens was added and hMDMs incubated for 5 days . hMDM viability was measured by calcein-AM uptake , using 0 . 4% calcein in PBS incubated for 30 min , that was added after removal of supernatants and washing of cells . The calcein fluorescence was measured in a Modulus microplate reader prior to lysing of hMDMs to measure the bacteria in the lysate as described previously [31] . The luminescence from live bacteria in both supernatant and lysate was measured . hMDMs ( 3x104/ well ) were treated with helminth antigens ( 10 μg/ml of H . d , T . m or S . m ) for 48h , thereafter stimulated with heat killed Mtb ( MOI = 5 ) , PPD , SEB , or ovalbumin for 24h . Supernatants were discarded and hMDMs were co-cultured with autologous Mtb antigen-specific ( PPD or Ag85B ) CD4+ T-cells ( 1 . 2x105 cells/ well ) . Cell free supernatants were examined for IFN-γ production after 48h of co-culture . hMDMs were untreated or treated with 10 μg/ml of the antigens for 48h prior to Mtb infection for 4 . 5h . The cells were lysed in boiling 2x Laemmli sample buffer and western blot was run as previously described [28] using the primary antibodies rabbit monoclonal anti-LC3 ( D11 ) ( Cell signaling ) diluted 1:5000 , mouse monoclonal anti-SQSTM1 D-3 ( Santa Cruz Biotechnology ) diluted 1:1000 and mouse monoclonal anti-β-actin ( clone AC-74 , Sigma Aldrich ) diluted 1:10 000 . The secondary antibodies were polyclonal goat anti-rabbit and anti-mouse immunoglobulins/HRP ( Dako Cytomation ) diluted 1:2000 except for the one against anti-β-actin which was diluted 1:10 000 . Inhibition in autophagosome maturation ( i . e . functional autophagy ) was evaluated based on increases in either the LC3BII/actin ratio or the SQSTM1/actin ratio according to our previous experience with M . tuberculosis-infected hMDMs [28] . Statistical analyses were performed with Graph pad prism ( version 5 . 0f ) . Data from multiple treatments were analyzed using Repeated Measures ANOVA with Dunnett post test , and for single treatment a paired Student t-test was used as indicated . p values < 0 . 05 were considered significant .
To study the direct effect of helminth antigens on hMDMs function we first evaluated Mtb phagosomal maturation in macrophages simultaneously exposed to helminth antigens . The first approach was to quantitatively measure the lysosomal marker LysoTracker Deep Red ( LTDR ) -signal within uninfected and infected macrophages ( the GFP+ fraction ) using flow cytometry . The helminth antigens did not significantly affect the basal level of the LTDR-signal in uninfected hMDMs . In hMDMs treated with Mtb only , there was a clear decrease in LTDR-signal with increasing MOI of Mtb ( Fig 1; and previously shown [32] ) , indicating that Mtb-infection alone blocks phagosome acidification . hMDMs infected with Mtb at MOI = 1 and co-exposed with antigens from H . diminuta or T . muris showed a ≥ 50% drop in LTDR MFI , compared to Mtb only infection ( p < 0 . 05 for both H . diminuta and T . muris , respectively ) . Similar results were obtained at MOI = 10 with a 42% reduction in LTDR MFI for H . diminuta co-treated and 36% reduction for T . muris co-treated hMDMs ( p < 0 . 05 for both H . diminuta and T . muris treatment ) . Thus , antigens from these helminths effectively suppress acidification and phagosomal maturation in Mtb infected hMDMs , irrespective of the bacterial burden . Schistosoma soluble egg antigen co-treatment did not affect the LTDR-signal at any MOI tested . Using confocal microscopy the co-localization of the phagosomal maturation marker and the bacteria was analyzed ( Fig 2A ) . Areas with phagosomes contributed to the strongest LTDR-signal , whereas remaining parts of the cells did not , indicating that the LTDR-signal measured by flow cytometry ( Fig 1 ) is maturing phagosomes and not general hMDMs acidification . Bafilomycin , the v-ATPase inhibitor used as a negative control , strongly inhibited LTDR co-localization to Mtb-phagosomes ( Fig 2B ) , further verifying the specificity of this probe . hMDMs ingesting FITC-labeled yeast , used as positive control for phagosome maturation ( Fig 2B ) , showed a 1 . 5-fold increase in LTDR co-localization compared to that of Mtb phagosomes ( from 54% with Mtb to 83% LTDR-positive phagosomes with yeast ) , consistent with Mtb virulence mechanisms being active in preventing phagosomal maturation ( Fig 2B ) [33] . Significantly less LTDR-Mtb co-localization was observed in macrophages co-exposed with H . diminuta ( p < 0 . 05 ) or T . muris ( p < 0 . 05 ) antigens . Thus , while Mtb can obstruct phagosomal maturation , concomitant exposure to helminth antigens can further reduce the capacity of hMDMs to handle and efficiently process Mtb phagosomes . Again , schistosoma soluble egg antigen co-treatment did not affect the Mtb-LTDR co-localization . No differences in number of intracellular Mtb were seen in helminth antigen treated or untreated hMDMs ( Fig 2C ) , indicating that the reduced acidification and phagosome maturation was not due to differences in total bacterial uptake by the macrophages . Cytokine secretion was monitored in uninfected and infected hMDMs at increasing bacterial loads ( MOI = 1 , 2 , 5 , and 10 ) ( Fig 3A–3D ) . We evaluated the early cytokine secretion at 2h , and the delayed cytokine secretion at 24h post-treatment/infection . Untreated uninfected hMDMs showed low secretion of TNF-α at 2h ( <300 pg/ml ) , whereas H . diminuta and T . muris treatment of infected and uninfected hMDMs induced an immense TNF-α secretion ( ~9500 pg/ml and ~8500 pg/ml , p < 0 . 01 and p < 0 . 05 compared to untreated uninfected , respectively ) . After 24h , the levels of TNF-α had decreased in the H . diminuta and T . muris-treated cells although still exhibiting significant increase in uninfected and infected up to MOI = 2 , but not for the higher MOIs were the Mtb-infected only cells had caught up with those of the co-exposed groups . The initial low levels of IL-6 at 2h ( untreated <30 pg/ml , H . diminuta and T . muris-treated ≤150 pg/ml , irrespective of infection ) had increased substantially at 24h showing a significant increase with helminth-treatment in uninfected hMDMs ( p < 0 . 05 for both H . diminuta and T . muris treatment ) , and for H . diminuta or T . muris co-exposed hMDMs at MOI = 1 ( p < 0 . 05 for T . muris co-exposed ) and MOI = 5 ( p < 0 . 05 for both H . diminuta and T . muris co-exposed ) . Except for IL-6 and TNF-α no other cytokines measured showed significant release above background at 2h . Unlike the other cytokines measured , IL-1β was not secreted in any conditions under MOI = 5 , and H . diminuta exhibited a strong augmenting effect on the Mtb-triggered response that was 14x-fold at MOI = 5 and 4 . 3x-fold at MOI = 10 ( p < 0 . 05 ) . Evaluating secretion of the anti-inflammatory cytokine IL-10 , the helminthic antigens H . diminuta and T . muris exhibited a synergistic effect with increasing MOI of Mtb . From these analyses we conclude that H . diminuta and T . muris antigens can trigger an early pro-inflammatory response with increased TNF-α both in the absence and presence of Mtb-infection which is then shifted towards an anti-inflammatory response with a synergistic increase of IL-10 . S . mansoni-antigen treatment of hMDMs did not induce any cytokine secretion by itself and did not augment the Mtb-induced TNF-α cytokine secretion ( Fig 3C and 3D ) , but instead lowered the Mtb-induced IL-10 secretion ( Fig 3D ) . To elucidate whether helminth antigens , in the absence of polarizing modulators from the adaptive immune response , have the capacity to polarize human macrophages toward M1 ( classically activated or pro-inflammatory ) or M2 ( alternatively activated or anti-inflammatory ) , hMDMs surface expression of CD206 ( mannose receptor; M2a , indicative of IL-4 macrophages ) , CD163 ( M2c , also referred to as IL-10 macrophages ) , CCR7 ( M1-marker for macrophages ) , and DC-SIGN ( marker for regulatory alternatively activated M2-macrophages ) [30 , 34] were investigated ( Fig 4 ) . We used the same experimental setup and rational with increasing bacterial MOI as for detecting cytokine secretion . Of the several polarization patterns that the helminth antigens induced without or with infection , we focus on the dominant effect for each individual antigen during co-stimulation with Mtb . T . muris induced a polarization towards M2a-like ( CD206+ ) hMDMs with elevated DC-SIGN expression . H . diminuta induced polarization towards M2c-like ( CD163+ ) hMDMs . S . mansoni soluble egg antigen triggered increased DC-SIGN expression in combination with enhanced expression of the M1-marker CCR7 at the highest MOI . In the absence of antigen , CD206 , DC-SIGN , and CCR7 expression was gradually elevated with increasing bacterial MOI , whereas CD163 decreased with increasing bacterial MOI . Despite Mtb being immunomodulatory on macrophage polarization itself , helminth-antigen exposure further exacerbated the Mtb effect as seen by the increasing expression of CD206 ( T . muris ) , CD163 ( H . diminuta ) , DC-SIGN ( T . muris and S . mansoni soluble egg antigen ) , and CCR7 ( S . mansoni soluble egg antigen ) . Expression of CCL22 , being specifically expressed under M2-stimulation ( e . g . IL-4 ) at 24h post-stimulation [35 , 36] was used to further decipher the shift in macrophage polarization upon helminth antigen challenge ( Fig 5 ) . The data for CCL22 indicate that antigen from H . diminuta and T . muris ( but not from S . mansoni ) significantly increase its production in uninfected ( H . diminuta , p < 0 . 001 ) or in combination with M . tuberculosis infection ( T . muris , p <0 . 05 ) . Taken together with the surface expression of the M2a-marker ( CD206; augmented by T . muris co-exposure ) and the M2c-marker ( CD163; augmented by both T . muris and H . diminuta co-exposure ) this indicated that these antigens are driving polarization towards an AAM phenotype . Our data thus far indicate that some helminth antigens can have both an early pro-inflammatory effect and a long-lasting immunoregulatory effect pertaining to polarization of cells towards an AAM phenotype with reduced phagolysosome fusion . To investigate these rather opposing effects on hMDMs ability to handle Mtb infection , hMDMs were exposed to helminth antigens for various periods before infected by Mtb . hMDMs were preincubated with helminth antigens for 1h to investigate the early effects of worm exposure , and for 48h to investigate the long-term or chronic effects of worm infection on Mtb infection . Neither exposure time affected the uptake of Mtb into hMDMs , nor the viability of hMDMs ( Fig 6 , upper and lower horizontal panel , respectively ) . Exposure with the T . muris antigen for 1h prior to infection with Mtb induced a slight , yet statistically significant ( p < 0 . 05 ) , decrease in the bacterial burden in hMDMs at 5 days post-infection ( compared to untreated ) ( Fig 6A , middle panel ) . hMDMs exposure with helminth antigens for 48h prior to infection , instead caused a 2 . 8-fold increase ( p < 0 . 05 ) in the bacterial burden for T . muris and 1 . 7-fold increase in the bacterial burden for H . diminuta ( p < 0 . 05 ) , whereas S . mansoni antigen induced increased control of Mtb showing a bacterial burden of 0 . 7-fold compared to untreated infected hMDMs ( Fig 6B , middle panel ) . This suggests that during chronic helminth infection the direct immunomodulatory properties of helminthic antigens , can either facilitate growth of Mtb inside human macrophages , or help macrophages maintain control over Mtb depending on the helminth species . To elucidate whether the immunomodulatory effects of the individual helminth antigens affected hMDMs ability to present Mtb-antigens we generated Mtb Ag-specific CD4+ T-cells which were cultured with autologous hMDMs . As helminth antigens were seen to affect the proliferation of Mtb inside macrophages ( Fig 6 ) , and the Mtb Ag-specific presentation would be influenced by the number of bacteria for the net availability of mycobacterial antigens , these experiments were performed with inactivated M . tuberculosis H37Rv thereby keeping the source of antigen ( e . g . the bacteria ) equal throughout the treatments . H . diminuta and T . muris exposed and infected hMDMs co-cultured with autologous PPD- or Ag85B-specific CD4+ T cells significantly reduced the Mtb-induced IFN-γ secretion by the Mtb Ag-specific CD4+ T cells ( Fig 7A and 7B ) . S . mansoni antigen exposure of hMDMs did not affect their capacity to stimulate the Mtb Ag-specific CD4+ T cells . The positive control SEB markedly induced IFN-γ , whereas the negative control ovalbumin did not induce any IFN-γ production above that of uninfected hMDMs . Besides the helminth-driven skewing effect in antigen presentation ( i . e . IFN-γ release from Mtb Ag-specific CD4+ T cells ) when hMDMs were stimulated with intact Mtb bacteria ( Fig 7 ) , H . diminuta pre-exposed hMDMs were further seen to also reduce the release of the Th1-cytokines TNF-α and IL-2 upon mycobacterial protein stimulation ( S1 Fig ) . In agreement with intact bacterial stimulated hMDMs , both H . diminuta and T . muris exposed hMDMs reduced the IFN-γ release from Mtb Ag-specific CD4+ T cells when mycobacterial proteins were used for stimulating the hMDMs . Acidification of the phagosome contributes to degradation of bacteria and generation of bacterial peptides delivered for antigen presentation [37] . Since autophagy is involved in delivering antigens to the MHC class-II loading compartment [37] , we tested if the helminth antigens affected autophagy in Mtb infected hMDMs ( Fig 8 ) . With a 48h helminth antigen pretreatment alone the antigens did not significantly affect the autophagy proteins LC3B and SQSTM1 . However , H . diminuta and T . muris-antigen pretreatment and Mtb co-exposure markedly enhanced accumulation of LC3BII , and significantly accumulated the autophagy substrate SQSTM1 ( p < 0 . 05 for both H . diminuta and T . muris-antigen treatment ) , compared to unexposed Mtb infected hMDMs . Buildup or accumulation of autophagy proteins in hMDMs during infection with M . tuberculosis is caused by M . tuberculosis blocking autophagosome maturation of the bacteria containing vacuoles as previously shown [28] . The presented data indicate that co-exposure with helminth derived antigens further obstruct a functional autophagy , needed for both elimination of the bacteria and generation of Mtb-antigens for MHC class-II loading .
There are conflicting data regarding the interplay between helminth and Mtb infections , with some studies showing an increased bacterial burden in co-infected animals [7 , 8 , 19] , while others show no effect [38–40] or a decreased burden of mycobacteria [9 , 41] . We hypothesized that helminths would make the hMDMs less capable of controlling Mtb infection , and used antigens from three distinct groups of helminths to investigate this . It is important to note that the effects documented here are in the absence of T cells ( or any other cell of the adaptive immune response ) . We found that antigens from different helminths cause different responses against Mtb in human macrophages with respect to lysosome function , macrophage polarization , Mtb burden and antigen processing of Mtb . Antigens from the nematode T . muris and the cestode H . diminuta induced similar responses in hMDMs leading to an increased burden of Mtb , while soluble egg antigens from the trematode S . mansoni induced a response that favored the host by decreasing Mtb burden . This was consistent with the decreased phagosome maturation seen in hMDMs co-exposed to H . diminuta and T . muris antigen and the unaffected Mtb-phagosome acidification with S . mansoni soluble egg antigen . When exposing the macrophages to antigen directly prior to Mtb infection , the macrophage burden of Mtb was largely unchanged independent of the helminth antigen used . However , upon 48h pretreatment ( mimicking a chronic infection ) with the helminth antigens , an increase in the Mtb burden could be seen upon H . diminuta and T . muris antigen treatment while exposure to soluble egg antigens from S . mansoni increased the control of Mtb . The production of cytokines was accordingly more pro-inflammatory at early exposure time-points and shifted to a more anti-inflammatory response , with increased IL-10 , with a longer exposure to T . muris and H . diminuta antigens , while S . mansoni antigens decreased the IL-10 production from Mtb-infected hMDMs . This indicates that some helminths prime the innate immune response towards a more pro-inflammatory response while others push it towards an anti-inflammatory response which could affect the outcome of a bacterial infection that usually is dominated by a Th1 pro-inflammatory response . This is in accordance with another study showing that T . muris evoked an increase in pro-inflammatory cytokines such as TNF-α along with an increase of IL-4 and IL-10 compared to BCG infection and co-infection [38] . This initial pro-inflammatory response may be due to the activation of TLR4 by the glycans from the helminths , as shown by Goodridge et al [42] . However it seems that after a longer infection period with helminths , the response becomes dominated by Th2 cytokines with a concurrent decrease in Th1 cytokines [43] . As seen previously , helminths can induce AAMs with increased expression of arginase-1 in mice , which can be less capable of combating infection with bacteria [10 , 19 , 43 , 44] , although this kind of type 2 response is needed to expel the helminths [21] . IL-10 could promote the development of these alternative activated macrophages [44] . This cytokine is also associated with a higher sensitivity to Mtb in mice , causing increased bacterial load and mortality [44] . Similarly , we observed polarization towards a M2-type macrophage in H . diminuta and T . muris antigen-treated hMDMs , with elevated CCL22 expression ( specifically expressed during M2-polarization; [36] ) , increased IL-10 levels , and reduced capacity to combat Mtb . On the other hand , S . mansoni treated hMDMs expressed higher levels of CCR7 ( M1 macrophages; [35] ) , produced less IL-10 during infection , showed no elevation in CCL22 and were also more fit to control Mtb . A study with T . muris and Mtb co-infected mice showed no effect on the bacterial load , although there was a decreased response to both pathogens which lead to delayed expulsion of the helminth and a decreased Th1 response in the lung [38] . A previous study showed reduced expression of co-stimulatory molecules in dendritic cells and macrophages along with decreased numbers of IL-10 and IFN-γ producing CD4+ T cells upon Brugia malayi microfilariae exposure [45] . During moderate Mtb co-exposure ( MOI 1–5 ) , we observed that the T . muris antigen induced higher IL-10 levels than H . diminuta , and that H . diminuta antigen stimulation of hMDMs enhanced their CCL22 expression in the absence of Mtb but that T . muris antigen required co-exposure with Mtb for CCL22 expression . To further investigate whether the helminth antigens polarized hMDMs into different AAM phenotypes additional markers were employed . One of the markers investigated here was CD206 ( the mannose receptor ) and expression of this marker indicates a M2a phenotype in humans and mice [3 , 35] . Besides the increase in IL-10 , we observed that T . muris treated macrophages had increased CD206 expression indicative of a M2a-like macrophage . The mannose receptor has been found to be important in the binding of T . muris antigen and uptake of S . mansoni antigens by macrophages , causing a reduced production of pro-inflammatory cytokines [20 , 21] . In our study only T . muris treated hMDMs had increased expression of CD206 which was not seen in S . mansoni co-exposed hMDMs . The expression of DC-SIGN was also increased in hMDMs co-exposed to T . muris indicative of a regulatory M2-like macrophage . This receptor is induced on alveolar macrophages during Mtb infection and its expression has been associated with increased susceptibility to Mtb , since the pathogen binds more easily to the macrophages [46] . The Mtb burden was highest in the T . muris co-exposed hMDMs , possibly due to the high IL-10 production and the lack of IL-1β compared to H . diminuta co-exposed cells that exhibited an augmented IL-1β production and a less dramatic effect on Mtb replication . Furthermore , during co-infection with Mtb these antigens promoted different types of regulatory macrophages: T . muris antigen increased CD206 expression ( M2a-like macrophages ) whereas H . diminuta antigen increased CD163 expression ( M2c-like macrophages ) . Infection with helminths ( as modeled here by helminth antigens ) will impact how the host deals with TB , and as we show this is a complex and species-specific interaction such that in the field it will be important to determine not only if the individual is Th2 skewed but also if the helminth-infection is a trematode , a cestode or a nematode as well as determining the species , since it is likely that there will be a species-specific differential effect on the outcome of TB . Several studies have shown an impact on the polarization of the adaptive immune response upon helminth and mycobacterial co-infection , with reduced levels of Th1 cytokine expressing T cells [17 , 18 , 38 , 45] and increased levels of regulatory T cells [47 , 48] . To further elucidate the response towards Mtb during co-exposure to helminth antigens in hMDMs , Mtb-antigen presentation was measured by the activation of Mtb antigen-specific CD4+ T cells . hMDMs co-exposed to Mtb and antigen from H . diminuta or T . muris caused less activation of the CD4+ T cells , indicating reduced efficiency in Mtb-antigen presentation by the hMDMs to the T cells . Together with the reduced LysoTracker co-localization to Mtb and the accumulation of autophagy proteins , this implies deficient processing of Mtb antigens in the co-exposed hMDMs that would lead to a decreased activation of CD4+ T cells . In contrast , hMDMs co-exposed to Mtb and antigen from S . mansoni did not lead to reduced T cell activation or reduced LysoTracker co-localization , which is in accordance with the increased control of Mtb . However , this is in contrast to another study showing that S . mansoni antigen impaired Mtb specific T cell responses with a reduction of IFN-γ and reduced control of Mtb [19] . The reason for the contradictory results might be due to the fact that the first response to a schistosoma infection is dominated by Th1 events , while the production of eggs later during infection causes a shift towards a Th2 response [49] . Additionally , the differences between other studies and data herein are that we assessed the direct effect of helminth antigens on macrophages without the involvement of a Th2 response . A recent example of helminth exposure of Mtb-specific T cells , showed that S . mansoni soluble antigen exposed T cells of TB infected individuals produced increased levels of anti-inflammatory IL-10 that caused a phagosomal arrest in Mtb infected human macrophages [50] . In conclusion , our study shows that different helminth antigens can have direct effects on macrophages and cause different responses to Mtb in co-exposed hMDMs . H . diminuta antigens and to a greater degree T . muris antigens caused an anti-inflammatory response with M2-type polarization and increased IL-10 secretion , along with decreased T cell activation , in Mtb infected cells . These co-exposed hMDMs also exhibited reduced bactericidal functions as shown by reduced phagosome maturation and an increased Mtb burden . Antigen from S . mansoni had the opposite effect on macrophages , causing a decrease in IL-10 output , a M1-type polarization and an increased control of Mtb . As expected the interaction of helminths ( mimicked by use of helminth antigens ) and Mtb is complex and species-specific and while the mechanism ( s ) of this trans-kingdom interaction need to be fully defined , it is clear that in helminth-endemic areas the outcome of TB will be influenced by the helminth burden . Assuming the in vitro data presented herein translate to infected humans the challenge will be to develop effective therapy for TB that considers the patients co-infection status . | The innate immune system is the first response against invading pathogens like the bacterium Mycobacterium tuberculosis ( Mtb ) or parasitic worms ( helminths ) . The adaptive immune response takes over after being primed by the innate immune response . Infection with Mycobacterium tuberculosis typically gives rise to a pro-inflammatory T-helper ( Th ) -1 response while helminths promote a Th2 response which is needed to combat the infection . Co-infection with both of these pathogens could lead to reduced immunity contributing to worsening of tuberculosis due to an increased Th2 response caused by helminths . We found that antigens from different helminth species ( a nematode , a cestode and a trematode ) caused different responses towards Mtb in macrophages . Depending on the helminth species , the macrophages can be more or less capable of combating Mtb infection and priming the adaptive immune response , which in turn would influence the outcome of tuberculosis . Thus , exposure to helminth antigens , in a species-dependent manner , could lead to a better control of Mtb infection or worsening of tuberculosis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"helminths",
"immunology",
"parasitic",
"diseases",
"animals",
"developmental",
"biology",
"molecular",
"development",
"bacteria",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"actinobacteria",
"immune",
"response",
"immune",
"system",
"helminth",
"infections",
"cell",
"biology",
"mycobacterium",
"tuberculosis",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
] | 2017 | Species dependent impact of helminth-derived antigens on human macrophages infected with Mycobacterium tuberculosis: Direct effect on the innate anti-mycobacterial response |
Bioactive peptides ( i . e . , neuropeptides or peptide hormones ) represent the largest class of cell-cell signaling molecules in metazoans and are potent regulators of neural and physiological function . In vertebrates , peptide hormones play an integral role in endocrine signaling between the brain and the gonads that controls reproductive development , yet few of these molecules have been shown to influence reproductive development in invertebrates . Here , we define a role for peptide hormones in controlling reproductive physiology of the model flatworm , the planarian Schmidtea mediterranea . Based on our observation that defective neuropeptide processing results in defects in reproductive system development , we employed peptidomic and functional genomic approaches to characterize the planarian peptide hormone complement , identifying 51 prohormone genes and validating 142 peptides biochemically . Comprehensive in situ hybridization analyses of prohormone gene expression revealed the unanticipated complexity of the flatworm nervous system and identified a prohormone specifically expressed in the nervous system of sexually reproducing planarians . We show that this member of the neuropeptide Y superfamily is required for the maintenance of mature reproductive organs and differentiated germ cells in the testes . Additionally , comparative analyses of our biochemically validated prohormones with the genomes of the parasitic flatworms Schistosoma mansoni and Schistosoma japonicum identified new schistosome prohormones and validated half of all predicted peptide-encoding genes in these parasites . These studies describe the peptide hormone complement of a flatworm on a genome-wide scale and reveal a previously uncharacterized role for peptide hormones in flatworm reproduction . Furthermore , they suggest new opportunities for using planarians as free-living models for understanding the reproductive biology of flatworm parasites .
Platyhelminthes ( flatworms ) inhabit a variety of aquatic and terrestrial environments and members of the phylum are thought to parasitize most vertebrate species [1] . The remarkable ability of flatworms to maintain plasticity in their reproductive cycles is a key to their success . As an example , free-living planarian flatworms are capable of reproducing sexually as cross-fertilizing hermaphrodites or asexually by transverse fission [2] . Some planarian species even maintain the ability to switch between modes of sexual and asexual reproduction , resorbing and regenerating their reproductive organs , depending on the environmental context [3] . This dynamic regulation of reproductive development is not limited to free-living platyhelminths; parasitic flatworms can also undergo dramatic changes in their reproductive development in response to external stimuli . In dioecious parasites of the genus Schistosoma , female reproductive development requires pairing with a male worm [4]–[8] . Thus , female schistosomes derived from single-sex infections have underdeveloped ovaries and accessory reproductive organs when compared to females from mixed sex infections . Interestingly , the reproductive organs of mature females deprived of their male counterpart regress and are capable of regrowing once male-female pairing is reestablished [9] . Because flatworms , including schistosomes , are responsible for causing important neglected tropical diseases , understanding the mechanisms that coordinate the reproduction of both free-living and parasitic members of the phylum is of fundamental importance . Peptide hormones ( i . e . neuropeptides ) are among the most structurally and functionally diverse class of metazoan signaling molecules [10] . In vertebrates , a neuroendocrine axis involving peptide hormone signaling between the brain and the gonads controls the maturation and long-term maintenance of reproductive development and function [10]–[13] . A similar role for neuroendocrine signals in controlling flatworm reproduction is suggested by studies exploiting the well-known regeneration abilities of planarians . Head amputation ( i . e . removal of the brain/cephalic ganglia ) of sexually reproducing planarians results in regression of the testes [14] , [15] to their germ cell primordia [16] , which re-grow only when cephalic ganglia regeneration is complete . These observations suggest that neural signals control the dynamics of planarian reproduction . Thus , flatworms may employ peptide-based mechanisms , similar to vertebrates , to synchronize their reproductive development . To date only limited data exist to support a “vertebrate-like” role for peptide hormones in invertebrate reproductive maturation . Insulin-like peptides influence germline stem cell proliferation in Drosophila [17] , [18] and C . elegans [19] and promote oocyte maturation in the starfish Asterina pectinifera [20] and the mosquito Aedes aegypti [21] . In locusts , treatment with the peptide hormones ovary maturing parsin ( OVP ) [22] or short Neuropeptide F ( sNPF ) [23] , [24] can stimulate ovarian development and vitellogenesis . Because of this paucity of data linking neuroendocrine function to invertebrate reproductive development , additional studies are required to determine how invertebrates modulate their reproductive output in response to external and metabolic cues . Peptide hormones are processed proteolytically from longer secretory prohormone precursors and often require covalent modifications before becoming biologically active [10] , [25] . As a result of this extensive processing , and because the biologically relevant signaling units are encoded by short stretches of amino acid sequence ( usually 3–40 amino acids ) , predicting genes encoding these molecules represents a major challenge for bioinformatics-driven genome annotation . The recent application of bioinformatic approaches coupled to mass spectrometry-based peptide characterization techniques ( an approach called peptidomics [26]–[28] ) has revolutionized discovery efforts , uncovering hundreds of new genes encoding metazoan bioactive peptides . Among invertebrates , however , much of this recent progress has been focused on genome-wide studies of nematodes [29]–[31] , arthropods [32]–[36] , and mollusks [37] , [38] . Thus , little is known of the peptide hormones present in phyla such as Platyhelminthes . Despite recent bioinformatic efforts to characterize flatworm peptide-encoding genes [39] , [40] , only three distinct peptides have been characterized extensively at the biochemical level in all flatworms [41] . Owing to a wealth of functional genomic tools [42] and a sequenced genome [43] , the planarian S . mediterranea represents an ideal model to characterize flatworm neuropeptides . Furthermore , this species exists as two distinct strains: an asexual strain that lacks reproductive organs and propagates exclusively by fission and a sexual strain that reproduces as cross-fertilizing hermaphrodites [44] . This dichotomy presents a unique opportunity to explore the extent to which peptide hormones are associated with distinct reproductive states . To address the possibility that peptide signals influence planarian reproductive development , we began by disrupting a gene encoding a prohormone processing enzyme , Smed-prohormone convertase 2 ( Smed-pc2 , GB: BK007043 ) , in sexual planarians . Consistent with a role for peptide hormones in controlling planarian reproduction , knockdown of Smed-pc2 led to a depletion of differentiated germ cells in the planarian testes . To identify potential peptide mediators of this effect , we used peptidomic approaches to characterize the peptide hormone complement of S . mediterranea . This analysis identified 51 genes predicted to encode more than 200 peptides , 142 of which we characterized biochemically by mass spectrometry . Global analysis of the expression of these genes by whole mount in situ hybridization revealed a distinct distribution of some peptide prohormones between sexual and asexual strains of S . mediterranea . We find one prohormone gene , npy-8 , to be enriched in the nervous system of sexual planarians and show that this gene is required for the proper development and maintenance of reproductive tissues . These results demonstrate the utility of S . mediterranea as a model to characterize metazoan peptides and suggest that flatworm reproductive development is controlled by neuroendocrine signals .
To explore potential roles for peptide signaling in regulating planarian reproductive physiology , we characterized Smed-pc2 ( Figure S1 ) , whose orthologues are required in both vertebrate and invertebrate models for the proteolytic processing of prohormones to mature neuropeptides ( in the interest of brevity , we will drop the prefix “Smed” from the remainder of the genes described below ) [30] , [45] , [46] . A large-scale RNA interference ( RNAi ) screen determined that this gene was essential for coordinated movement and normal regeneration in asexual planarians [47] . Whole-mount in situ hybridization in sexual planarians revealed expression of pc2 in the central nervous system [48] , the pharynx , sub-muscular cells , the photoreceptors , the copulatory apparatus , and the testes ( Figure 1A–C ) . To determine if peptide signals are likely to play a functional role in coordinating reproductive development , we monitored the effects of pc2 RNAi on the dynamics of germ cells within the planarian testes . Individual testis lobes consist of an outer spermatogonial layer in which cells divide to form cysts of eight spermatocytes that , after meiosis , give rise to spermatids and , ultimately , sperm [44] , [49] . After 17 d of RNAi treatment , pc2 ( RNAi ) animals displayed a decrease in both testis size ( Figure 1E ) and the number of animals producing mature sperm ( 28/29 for controls versus 2/36 for pc2 RNAi; p<0 . 0001 , Student's t test ) . To establish which cell types are affected by pc2 RNAi , we performed fluorescence in situ hybridization ( FISH ) to detect germinal histone H4 ( gH4 ) ( GB: DN306099 ) and nanos ( GB: EF035555 ) mRNAs , which are expressed in spermatogonia and germline stem cells ( GSCs ) , respectively [16] , [50] , [51] . In developed testes of control animals , relatively few cells within the outer spermatogonial layer are identifiable as nanos-positive GSCs ( Figure 1F ) . However in pc2 ( RNAi ) animals , regressed testes clusters almost always co-expressed both gH4 and nanos ( Figure 1G ) ( n = 16/17 animals ) . These results suggest that pc2 is required for proper germ cell differentiation and/or for the maintenance of differentiated germ cells in the testes . Since our analysis of pc2 implicated peptide signaling in regulating planarian reproductive development , we characterized the peptide hormone complement of S . mediterranea . We employed bioinformatic and mass spectrometry ( MS ) -based methodologies to identify peptide prohormone genes from the S . mediterranea genome [43] and predict their processing into bioactive peptides ( Figure 2A ) [52] . With these approaches , we identified 51 prohormone genes in S . mediterranea , with peptides from 40 prohormones detected by MS ( Tables S1–S5 , gene names and abbreviations are shown in Table 1 ) . In most cases , MS confirmed multiple distinct peptides from a single prohormone , and in five prohormones we detected every predicted peptide by MS ( Figure 2B ) . In total , we characterized 142 peptides biochemically , corresponding to ∼45% of the distinct peptides predicted from our collection of 51 prohormone genes ( Table S5 ) . This analysis identified genes encoding relatives of all previously characterized flatworm neuropeptides ( YIRFamide [53] , spp-11; FRFamide [54] , npp-4; and neuropeptide Y-like [55] , npy-1 to npy-11 ) and provided biochemical validation for 10 prohormones previously predicted from the S . mediterranea genome [39] . The neuropeptide Y ( NPY ) -superfamily represents a large family of neuropeptides that influence diverse processes in both vertebrate and invertebrate taxa [10] , [41] , [56] . This family is considered to consist of two types of peptides: the NPY-like peptides that possess a C-terminal amidated tyrosine ( Y ) residue and the NPF peptides that possess a C-terminal amidated phenylalanine ( F ) residue [55] . Vertebrate genomes typically encode NPY-like peptides [57] , whereas invertebrate genomes encode NPF peptides [55] , [58] , [59] . Our studies found that the planarian genome possesses an expanded family of npy genes predicted to encode both NPY-like and NPF-like peptides ( Figure 2C ) . Prohormones NPY-5 , -7 , -9 , and -10 possess a C-terminal tyrosine residue , similar to vertebrate NPY peptides , and prohormones SMED-NPY-1 , -2 , -3 , -4 , -6 , and -8 contain a C-terminal phenylalanine residue , similar to invertebrate NPF peptides . Three of these planarian npy genes ( npy-1 , -4 , and -9 ) have been described previously [39] , [60] . Additionally , our studies , and those of others [39] , [61] , find evidence of conservation in the genomic organization of flatworm NPY genes . NPY genes from vertebrates possess an intron that separates the exon encoding the RXR motif from the penultimate amidated amino acid residue ( Figure 2D ) [62] . We found an identical architecture for S . mediterranea genes npy-1 , -2 , -3 , -4 , -5 , -6 , -8 , -9 , -10 , and -11 , indicating a close evolutionary relationship between chordate and platyhelminth npy genes ( Figure 2C , D ) . The planarian genome also encodes peptides with sequence similarities to those from other invertebrate taxa , including mollusks ( ppp-1 , GB:BK007041; ppp-2 , GB:BK007018; mpl-1 , GB: BK007017; mpl-2 , GB: BK007016; and , cpp-1 , GB: BK007012 ) and arthropods ( ppl-1 , GB: BK007007 ) . Furthermore , our analysis found that previously characterized , novel planarian genes encode peptide prohormones . Homologues of prohormones eye53-1 , 2 ( GB: BK007033 and GB: BK007024 , respectively ) and 1020-1 , 2 ( GB: GU295180 and GB:BK007025 , respectively ) from the planarian Dugesia japonica are required for proper visual system function following amputation; knockdown animals show no morphological defects after injury yet are unable to respond properly to light [63] . These previous observations , together with our findings that these genes encode neuropeptides , suggest a role for peptide signaling in the functional recovery of the planarian nervous system following injury . To examine if pc2 is required for prohormone processing in planarians , we disrupted pc2 expression using RNAi and performed MS to analyze the peptide complement of pc2 ( RNAi ) animals . Consistent with pc2 encoding a genuine prohormone convertase , analysis of peptide profiles in planarian tissue extracts by MALDI-TOF MS analysis demonstrated that pc2 RNAi resulted in a significant decrease in the signal intensity of a specific set of peptides in sexual animals ( Figure 2E , F and Table S6 ) . Interestingly , the levels of some peptides were increased following pc2 ( RNAi ) ; whether this alteration reflects a compensatory mechanism for regulating peptide levels or an altered threshold of detection for certain peptides caused by a global reduction in neuropeptide levels remains to be determined . However , these data parallel studies of pc2 knockout mice , in which the abundance of some peptides was either increased or decreased [45] . Given that the S . mediterranea genome is predicted to encode at least three additional proteins with similarity to prohormone convertases ( Figure S2 ) , it is possible that compensatory mechanisms are responsible for the observed elevation in the levels of some peptides . This redundancy among prohormone convertases is also likely to explain why we only observed changes in a subset of peptides following pc2 RNAi . These data suggest that the reproductive defects observed in pc2 ( RNAi ) animals may be due to altered levels of specific peptides . To determine the extent to which peptides may regulate flatworm reproduction , we took advantage of the fact that S . mediterranea exists as both sexually and asexually reproducing strains . By comparing prohormone gene expression between these strains we sought to uncover expression patterns specific to sexually or asexually reproducing animals . Thus , we began by performing comprehensive whole-mount in situ hybridization analyses of prohormone genes in asexual planarians ( Figure 3 ) . Our studies indicate that in asexual planarians ∼85% ( 44/51 ) of prohormone genes are expressed in the central nervous system ( CNS ) ( Table S5 ) , which consists of bi-lobed cephalic ganglia and two ventral nerve cords ( VNCs ) that run the length of the body [64] . Of the prohormones expressed in the CNS , 20% ( 10/51 ) were detected only in the cephalic ganglia . Notably , the expression of individual prohormones was often enriched in specific cell types or regions within the CNS . For example , the expression of some prohormones was enriched in either lateral ( e . g . npp-4 , GB: BK007037; npp-8 , GB: GU295189; spp-4 , GB: GU295179; and 1020HH-2 ) , medial ( e . g . spp-2 , GB: BK007032; and spp-6 , GB: GU295177 ) or posterior ( e . g . npy-1 , GB: GU295175 ) regions of the cephalic ganglia ( Figure 3 ) . Strikingly , a large fraction of prohormone mRNAs were detected in restricted cell populations within the CNS ( e . g . npy-1; npy-2 , GB: BK007019; cpp-1; spp-6; spp-9 , GB: BK007026; spp-10 , GB: BK007028; grh-1 , GB: GU295185; and ilp-1 , GB: BK007034 ) ( Figure 3 ) . Consistent with peptide signaling having a role in processes outside the CNS , we also detected prohormone expression in: the pharynx ( e . g . npp-1 , GB: BK007036; npp-22 , GB: BK007038; npy-11 , BK007021; and ppp-1 ) ; photoreceptors ( e . g . eye53-1 , -2; npp-12 , GB: GU295182; and mpl-2 ) ; sub-epidermal marginal adhesive glands ( e . g . mpl-2 ) ; an anterior domain between the VNCs ( e . g . spp-6; spp-7 , GB: GU295178; spp-8 , GB: GU295181; spp-9; cpp-1; and spp-10 , GB: BK007028 ) ; cells surrounding the ventral midline ( e . g . npp-5 , BK007015 ) ; the intestine ( e . g . npp-8 , GB: GU295189; and npy-10 , GB: BK007011 ) ; and various sub-epidermal cell types ( e . g . npp-18 , GB: BK007027; spp-4; spp-16 , GB: BK007042; and npy-4 , BK007039 ) ( Figure 3 ) . To investigate the extent to which prohormones are expressed in overlapping or distinct cell types in the CNS , we compared the expression of prohormone genes using triple FISH . Prohormone genes spp-1 ( GB: GU295176 ) , npp-2 ( GB: BK007035 ) , and ppp-1 encode unrelated peptides ( Tables S1 and S5 ) that appear to be expressed ubiquitously in the CNS ( Figure 3 ) . Comparison of the expression domains of these prohormone genes revealed that spp-1 , npp-2 , and ppp-1 are expressed in largely non-overlapping populations of cells of the cephalic ganglia and VNCs ( Figure 4A–C ) . We also analyzed the expression of a family of paralogous prohormone genes ( spp-6; spp-7; spp-8; spp-9; and spp-17 , GB: GU295183 ) that encode similar neuropeptides ( Figure S3 ) . Because this gene family has been expanded in the S . mediterranea genome , we refer to these prohormones as the Planarins . Examination of Planarins spp-6 , -7 , and -9 expression by FISH demonstrated that these genes are expressed in a common set of cells distributed between the VNCs and surrounding the pharynx ( Figure 4D ) . Despite being co-expressed in cells outside the CNS , spp-6 and spp-9 transcripts were detected in distinct groups of cells within the cephalic ganglia ( Figure 4E , F ) . These findings , with earlier observations [48] , [64] , suggest a level of complexity not previously appreciated for the patterning of the flatworm nervous system ( see Figure S4 ) . We also examined four prohormone genes ( eye53-1 , -2; npp-12 , and mpl-2 ) expressed within the photoreceptors . The planarian photoreceptors are comprised of two distinct cell types: neuronal photoreceptive cells and pigment cells that envelop the rhabdomeric projections of the photoreceptor neurons [2] , [65] , [66] . Analysis of prohormone gene expression within the photoreceptors revealed that the planarian photoreceptor neurons are patterned along the anterior-posterior axis . Specifically , prohormone genes npp-12 and eye53-1 were expressed exclusively in the anterior photoreceptor neurons , whereas mpl-2 and eye53-2 were expressed exclusively in posterior neurons ( Figure 5A ) . These findings are consistent with dye-tracing studies demonstrating that anterior and posterior photoreceptor neurons project to distinct anatomical regions [67] . In addition , we detected mpl-2 expression in a ventral population of cells that was separate from the expression of eye53-2 ( Figure 5B ) ; this result suggests that the photoreceptors are also patterned along the dorsal-ventral axis . Together , these data indicate that at least three chemically and anatomically distinct sets of neurons are present in the planarian photoreceptors . To determine if peptide expression is correlated with reproductive state , we next analyzed the expression of a subset of prohormones in the sexual strain of S . mediterranea . The reproductive system of this animal is comprised of a pair of ovaries located posterior to the cephalic ganglia , numerous dorsolateral testes lobes , as well as a variety of accessory reproductive organs ( i . e . oviducts , sperm ducts , copulatory apparatus , and accessory glands ) ( Figure 6A ) . We found several prohormones expressed in sexual reproductive organs , including the oviducts ( Figure 6B , C ) , the copulatory apparatus ( Figure 6B , C , and D ) , gland cells surrounding the copulatory apparatus ( Figure 6E , F ) , and the testes ( Figure 6G , H ) . These expression patterns implicate peptide signaling in reproductive processes such as copulation , fertilization , egg-laying , and gonadal function . Our expression analyses also found evidence of differential prohormone expression within the nervous system of sexual S . mediterranea . ppl-1 encodes peptides related to the pyrokinin peptides originally isolated from arthropods [68] , [69] . In contrast to asexual planarians in which ppl-1 expression was detected almost exclusively in the cephalic ganglia and the distal region of the pharynx ( Figure 3 ) , ppl-1 was expressed widely in the VNCs and surrounding the copulatory apparatus of mature sexual animals ( Figure 7A ) . To explore if ppl-1 expression was linked to sexual maturation , we determined the distribution of ppl-1 in immature sexual animals . In sexual animals analyzed within one week of hatching from the egg capsule , ppl-1 was expressed in a pattern similar to that of asexual animals ( Figure 7A ) ; thus , ppl-1 expression undergoes a change in spatial distribution during the process of maturation . The prohormone gene npy-8 ( GB: BK007010 ) is predicted to encode a 29 AA NPF-like peptide ( NPY-8A ) and a novel C-terminal peptide ( NPY-8B ) ( Figure 8A ) . By in situ hybridization we failed to detect npy-8 expression in asexual animals ( Figures 3 and 7B ) . In mature sexual animals , however , npy-8 RNA was detected in a variety of cells within the central and peripheral nervous systems including the cephalic ganglia , the VNCs , the sub-muscular plexus , and the pharyngeal nervous system ( Figure 7B ) . Additionally , in a majority of animals ( 13/18 ) we detected npy-8 RNA in a dorsal population of cells ( Figure 6C ) . Analysis of this dorsal cell population by FISH localized npy-8 expression to cells often , but not exclusively , found in association with testes lobes ( Figure 7D ) . To determine if npy-8 levels changed with sexual maturation we examined npy-8 expression in sexual hatchlings . In recently hatched animals npy-8 was detected in tissues similar to those of mature sexual animals including the cephalic ganglia , the VNCs , the sub-muscular plexus , and the pharyngeal nervous system ( Figure 7B ) . Furthermore , we observed dorsal cells expressing npy-8 in a majority of animals ( 8/13 ) ( Figure 7C ) . The lack of observable expression of npy-8 in asexual animals by in situ hybridization suggested a relationship between npy-8 expression and the ability to reproduce sexually . Because we initially cloned the npy-8 gene by 3′ RACE with cDNA derived from asexual animals ( Table S5 ) , we wished to confirm our in situ hybridization results using an alternative approach . Therefore , we performed northern blot analyses to detect npy-8 transcript in asexual , recently hatched sexual , juvenile sexual , and mature sexual animals ( Figure 7E ) . Consistent with our in situ hybridization results , we detected high levels of npy-8 in sexual animals of all developmental stages but not in asexual animals , suggesting that npy-8 is expressed at negligible levels in asexual planarians . Because npy-8 was expressed at high levels only in sexually reproducing planarians , we reasoned that peptides encoded from this gene may be important for reproduction . Therefore , we determined the knockdown phenotype of npy-8 using RNAi . For this analysis we employed two distinct RNAi feeding regimens . First , we measured the effect of npy-8 depletion on the maintenance of the reproductive system by feeding mature sexual animals bacterially expressed npy-8 dsRNA and observing the structure of the reproductive system at 4- and 7-wk time points . As a complementary approach , we fed juvenile sexual planarians in vitro synthesized dsRNA and observed the development of the reproductive system after 1 mo of feeding . Mature sexual animals fed npy-8 dsRNA over the course of 4–7 wk displayed a range of phenotypes consistent with loss of sexual maturity ( data are summarized in Table 2 ) . Specifically , in comparison to controls , a majority of npy-8 ( RNAi ) animals had regressed testes and failed to produce mature sperm ( 1/18 for controls versus 14/21 for npy-8 RNAi ) ( Figure 8B ) . In addition to testes defects , npy-8 ( RNAi ) treatment resulted in regression of the copulatory organs ( 0/18 for controls versus 13/20 for npy-8 RNAi ) ( Figure 8B , C ) and a decrease in the size ( or complete disappearance ) of the gonopore ( unpublished data ) . Similar to mature sexual animals , juvenile planarians fed npy-8 dsRNA for 1 mo displayed stunted testes growth , failed to produce mature sperm ( 0/8 for controls and 6/8 for npy-8 ( RNAi ) ) , and had shrunken or absent gonopores ( 0/20 for controls and 16/20 for npy-8 ( RNAi ) , Figure 8D ) . Importantly , these effects on reproductive maturation were not due to an overall defect in growth since npy-8 ( RNAi ) and control animals grew to similar sizes over this time period ( Figure S5A ) . Since npy-8 is a member of an expanded family of NPY-like genes in S . mediterranea ( Figure 2C ) , we examined both the effectiveness and the specificity of our npy-8 knockdowns . We fed juvenile planarians dsRNA specific to npy-8 and monitored the transcript levels of npy-8 and its closest relative , npy-1 , by quantitative RT-PCR . This analysis found that npy-8 RNAi treatment resulted in a statistically significant decrease in npy-8 transcript levels while having no effect on npy-1 mRNA levels ( Figure S5B ) . To further explore the specificity of the npy-8 ( RNAi ) phenotype , we performed a long-term feeding experiment in which we fed juvenile animals dsRNA specific to npy-8 or either of its two closest relatives , npy-1 or npy-2 . In contrast to npy-8 RNAi , neither npy-1 nor npy-2 RNAi treatments produced observable defects in the maturation of the planarian reproductive organs ( Figure S5C ) . Collectively , these studies suggest that the effects of npy-8 ( RNAi ) on reproductive development are due to specific disruption of npy-8 function and suggest that off-target effects are unlikely . To examine the regressed testes of npy-8 ( RNAi ) animals , we performed FISH to detect nanos and gH4 expression . This analysis uncovered a range of phenotypes associated with npy-8 RNAi ( Figure 8E ) . Some npy-8 ( RNAi ) animals had clusters of gH4-positive cells that were also nanos-positive; these testes clusters are similar to those observed in pc2 ( RNAi ) animals ( Figure 1G ) . In other animals we found gH4-positive clusters in which a subset of cells expressed nanos . We interpret the former to represent a “severe” npy-8 knockdown phenotype , whereas we suggest that the latter represents an “intermediate” phenotype resulting from incomplete npy-8 knockdown and/or perdurance of the peptide . In the most severe cases , the testes regression phenotypes seen in pc2 ( RNAi ) or npy-8 ( RNAi ) animals were similar . One model to explain this observation is that PC2 is required for proteolytic processing of the NPY-8 prohormone , and loss of a mature peptide ( or peptides ) encoded by npy-8 results in loss of the ability to achieve or maintain sexual maturity . Since our MS analysis did not identify any peptides encoded by npy-8 in extracts from either asexual or sexual animals ( Tables S1–S3 ) , we used FISH to determine if npy-8 and pc2 transcripts are localized to similar cell types in the planarian nervous system . We found that npy-8-expressing cells within the cephalic ganglia , the VNCs , the pharynx , and the sub-muscular plexus also express high levels of pc2 ( Figure 8F; and unpublished data ) . This observation is consistent with PC2 being required for the processing of peptides encoded by the npy-8 gene . Related flatworms of the genus Schistosoma currently infect over 200 million people worldwide [70] . Because of their complicated life cycles , schistosomes are not readily amenable to the types of large-scale biochemical analyses that we have employed to characterize the planarian peptidome . As an indirect means of biochemically validating peptide sequences from these animals , we compared our MS-validated prohormones with predicted proteins from the genomes of the trematodes Schistosoma mansoni [40] and Schistosoma japonicum [71] . With this approach we validated the sequences of peptides from eight previously characterized schistosome prohormone genes ( Tables 3 and S7 ) [39] , [40] . Furthermore , we identified eight additional Schistosoma genes not previously annotated as peptide prohormones ( Tables 3 and S7 ) . Among these newly annotated prohormones are schistosome genes that encode the peptide YIRFamide , a well-conserved flatworm peptide that has potent stimulatory effects on schistosome muscle fibers [41] that was not identified in previous bioinformatic efforts [39] , [40] . Together , these data provide biochemical validation for roughly half of the predicted prohormones in Schistosoma and demonstrate the utility of using planarians to understand flatworm parasites .
Although previous studies have characterized the expression of subsets of prohormones or their corresponding peptides [73]–[76] , a comprehensive accounting of the expression of these genes at the level of the whole animal has not yet been performed . Here we describe the distribution of all known neuropeptide-encoding genes in the planarian S . mediterranea by whole mount in situ hybridization . One surprising finding from these studies was the complexity of prohormone expression within the planarian CNS , which is considered to be among the most primitive centralized nervous systems in the animal kingdom [77] . We find that prohormone gene expression is localized to distinct regions of the cephalic ganglia and that many individual prohormones are expressed in unique CNS cell types . These results parallel observations in the planarian D . japonica in which small molecule neurotransmitters ( e . g . serotonin and dopamine ) are found in separate CNS cell populations [64] . The expression of prohormone genes in distinct regions/cell-types in the CNS suggests that processing centers for different neural functions ( e . g . sensory , motor , and neuroendocrine ) may be localized to chemically and spatially distinct domains of the flatworm CNS . In support of this idea , a “visual center” has been proposed to exist at the medial regions of the cephalic ganglia to which visual axons send their projections [78] . Elucidation of the functions of peptides expressed in these discrete CNS foci may help relate specific anatomical positions to distinct neural functionalities and allow for the dissection of planarian neural circuits . Our analysis of prohormone expression also revealed that many prohormone genes are expressed in tissues of the reproductive tract . Expression of peptide prohormones has also been observed in the somatic reproductive organs of C . elegans [73] . Interestingly , the expression pattern of some planarian prohormones parallels the immunohistochemical localization of similar gene products in other invertebrates . The NPY family member Smed-npy-9 was expressed in the cement glands ( or shell glands ) surrounding the copulatory apparatus that are thought to be involved in egg capsule synthesis and deposition [2] , [79] . Studies of S . mansoni observed NPY-like immunoreactivity in the region of Mehlis' gland [80] , which is morphologically , and likely functionally [79] , similar to the glands labeled by npy-9 . cpp-1 encodes VPGWamide and TPGWamide , peptides that are related to the APGWamide peptides first described in molluscs [81] . We found cpp-1 to be expressed around the penis papillia and the oviducts of sexual planarians , which mirrors APGWamide localization in the molluscan oviducts and male copulatory organs [82] , [83] . While specific functions for any of these peptides in planarian reproductive function remain to be elucidated , these results suggest evolutionarily conserved roles for peptides in several reproductive organs . Two prohormone genes ( ppl-1 and npy-8 ) were expressed differentially in the nervous systems of mature sexual versus asexual planarians . The expression of ppl-1 was similar in asexual and immature sexual animals but underwent a dramatic change in distribution during sexual maturation . Conversely , npy-8 expression was detected at similar levels and distribution in sexual animals yet was not detected in asexual animals . Interestingly , our biochemical analyses detected a number of peptides uniquely in either mature sexual or asexual planarians ( Tables S1–S3 ) . Taken together , these results indicate that sexually mature planarians possess unique signatures in both the composition and spatial distribution of peptide hormones relative to asexual and immature sexual animals . To address the role of peptide signaling in planarian reproductive physiology we first examined the planarian prohormone convertase 2 orthologue , pc2 . This analysis suggested that prohormone processing is required for regulating the dynamics of germ cell differentiation . A similar requirement for prohormone processing in germ cell development has not been described in other animal models . Loss-of-function mutations in the C . elegans pc2 orthologue egl-3 result in a range of neuromuscular defects [84] , [85] , but mutant animals are capable of germ cell development since they produce viable progeny . The role of the Drosophila pc2 orthologue Amontillado has not been assessed in adult reproductive development due to a requirement for this gene at multiple points during embryonic and larval development [86] , [87] . Despite the fact that peptide hormones are known to regulate vertebrate germ cells [11] , [12] , extensive studies of prohormone convertase knockout mice have also not revealed roles for prohormone processing in germ cell development [46] . Therefore , it is likely that functional redundancies exist among the enzymes responsible for processing hormones involved in vertebrate reproduction . Given this possibility of genetic redundancy in vertebrates , we suggest systematic characterization of prohormone processing in other invertebrate models ( e . g . C . elegans and Drosophila ) may help address the extent to which peptide signaling regulates reproductive development in other animals . Our studies suggest that NPY-8 may be among the prohormones processed by PC2 that are required for normal sexual development . At present it is not known which of the two predicted peptides encoded by NPY-8 influence planarian reproductive physiology . Prohormones that encode NPY-like peptides , including NPY-8 , often also encode a C-terminal peptide or CPON ( C-flanking peptide of NPY ) [39] , [58] , [88] , [89] . Because the functions of both vertebrate and invertebrate CPON peptides remain elusive , we speculate that the NPY-related peptide NPY-8A is the functional unit of this prohormone . In vertebrates , NPY signaling is thought to elicit diverse effects on the neuroendocrine axis regulating reproduction . Depending on the hormonal milieu , NPY administration can either promote or inhibit surges of luteinizing hormone [90] , a gonadotropin that regulates multiple functions in the male and female reproductive systems [10] , [11] , [13] . The hypothalamic gonadotropin-releasing hormone , which promotes luteinizing hormone release from the pituitary , can also be influenced by NPY [91] , [92] . Additionally , NPY may influence the timing of sexual maturation in mammals since it has been suggested to either induce or inhibit the onset of puberty [93] . Since NPY is a well-known regulator of energy homeostasis , NPY has been suggested to coordinate reproductive function with nutrient status [94] . Studies of Drosophila and Aplysia indicate similar roles for NPY-like peptides in processes related to nutrient homeostasis , such as feeding behavior [56] , [95] . However , functional analyses in vertebrate [96] and invertebrate models [56] have not described obvious reproductive deficits in animals deficient for NPY-like peptides . Given the fact that S . mediterranea possesses an expanded collection of NPY-like peptides relative to other animals , additional work will be required to determine whether the function of NPY-8 represents an ancestral or derived function for NPY-like peptides . Coordinated signaling between the hypothalamus , the pituitary , and the gonads controls vertebrate reproduction . Although our initial observation with pc2 RNAi implicated prohormone processing in planarian germ cell development , the site of action of this effect was difficult to interpret since pc2 expression was detected in both the nervous system and the testes . Our studies of npy-8 have clarified the role of the nervous system in planarian reproduction . npy-8 is expressed in both the central and peripheral nervous systems , and its transcripts are not detected in tissues affected by npy-8 RNAi , such as the testes . Therefore , peptides from NPY-8 are likely to act in a neuroendocrine fashion to influence reproductive development . Since amputation studies suggest that signals from the cephalic ganglia are essential for the maintenance of mature gonads in planarians [14] , [15] , one possible source of NPY-8 is from the cephalic ganglia . The function of pc2 within the testes is presently not known , but testes are likely to be a site of prohormone processing since we detect the expression of multiple peptide prohormones ( ilp-1 and spp-10 ) in this organ . Because peptide hormones can act as endocrine and paracrine signaling molecules in the vertebrate testes [12] , it is possible that peptides play similar roles in planarians . Therefore , we propose that peptides ( e . g . NPY-8 peptides ) from the nervous system promote events associated with reproductive maturation ( i . e . the production of differentiated germ cells ) and peptides produced in the testes may provide feedback to the CNS and other organ systems about the physiological state of the gonads . Additionally , peptides expressed within the testes may serve as paracrine factors that regulate germ cell maturation . This possibility of coordinated signaling between CNS and the gonads may explain why the effects of pc2 RNAi on the reproductive system are more severe than those of npy-8 RNAi . Due to a lack of sufficient markers our studies have not examined the effects of neuropeptide signaling on ovarian development; future efforts will be directed at examining this question . Although a chromosomal translocation distinguishes sexual and asexual S . mediterranea [42] , [97] , the strain-specific differences that account for their divergent modes of reproduction remain uncharacterized . With the exception of genes expressed in the reproductive system [98] , little is known about the transcriptional differences between these strains . Here we identify npy-8 as enriched in sexual animals and show an important role for this gene in sexual development . Interestingly , the regressed testes of mature sexual animals treated with either pc2 RNAi or npy-8 RNAi resemble the primordial germ cell clusters of asexual planarians that also label exclusively with gH4 and nanos [44] . These observations , together with the loss of somatic reproductive structures in npy-8 ( RNAi ) animals , suggest that lack of NPY-8 expression in asexual planarians may , in part , account for their inability to promote germ cell differentiation and initiate sexual maturation . However , because the phenotypes observed with pc2 ( RNAi ) were more severe than those observed with npy-8 ( RNAi ) , we anticipate future studies may uncover additional factors that act in concert with npy-8 to influence planarian reproductive maturation . According to one estimate , schistosomiasis ( infection by Schistosoma ) can be directly attributed to as many as 280 , 000 deaths per year in sub-Saharan Africa alone [99] . Despite the medical and economic impact of schistosomiasis , only a single chemotherapeutic agent ( praziquantel ) is currently used in treatment of this disease [100] . Therefore , identifying novel anthelmintic agents is an important goal of flatworm research . Schistosome eggs can become lodged in host tissues , such as the liver and bladder , forming granulomas that are the major cause of the pathology associated with schistosomiasis [100] . Thus , targeting reproductive function in adult animals represents a promising means by which to treat and control schistosome infection . The S . mansoni genome is predicted to encode two NPY-like prohormone genes: Sm-npp-20a and Sm-npp-20b [39] , [101] . Comparison of the predicted peptides from these prohormones with NPY-like peptides from S . mediterranea found that the NPY-like peptide encoded from Sm-npp-20a shares its closest similarity to NPY-8A ( ∼48% identity , ClustalW ) ( Figure 2C ) . Given this observation , and the similarities in the reproductive anatomy between planarians and trematodes [2] , it is possible that these animals employ similar mechanisms to control their reproductive output . Therefore , our results justify efforts aimed at understanding the role of peptide hormones in flatworm reproductive physiology and suggest that neuropeptide signaling may represent a viable target for the treatment and eradication of flatworm parasites .
Sexual and asexual S . mediterranea were maintained at 20°C in 0 . 75× and 1 . 0× Montjuïc salts , respectively [102] . To minimize non-specific background from gut contents after feeding , animals were starved at least 1 wk prior to use . For all experiments with sexual S . mediterranea , sexually mature animals ( ∼1 cm in length , unless otherwise specified ) with a well-developed gonopore were used , unless otherwise specified . All chemicals were obtained from Sigma-Aldrich ( St . Louis , MO ) unless otherwise stated . The peptide standards for Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry ( MALDI-TOF MS ) calibration were purchased from Bruker Daltonics ( Billerica , MA ) . For LC/MS analysis , peptide extracts were prepared from 80–100 sexual or asexual planarians . Whole animals were mechanically homogenized in 8–10 mL of acidified acetone ( 40∶6∶1 acetone/water/HCl ) or acidified methanol ( 90∶9∶1 methanol/acetic acid/water ) . After sonication , vortexing , and centrifugation of the homogenate , the supernatant was collected and the organic solvent was removed by evaporation in a SpeedVac concentrator ( Thermo Scientific , San Jose , CA ) . The supernatant was then filtered through a Microcon centrifugal filter with a 10 kDa cutoff ( Millipore , Billerica , MA ) , evaluated for peptide content by MALDI-TOF MS sampling of 0 . 5 µL and subjected to sequential separations by HPLC prior to tandem MS for peptide identification . Peptide extracts were fractionated using a microbore HPLC system Magic 2000 ( Michrom BioResources , Inc . , Aubum , CA ) with a C18 reverse phase column ( Dionex , 1 , 000 µm i . d . , particle size 3 µm , and pore size 100 Å ) at a 20 µL/min flow rate over a 70 min run . A four-step linear solvent gradient was generated by mixing mobile phases A ( 95% water and 5% acetonitrile ( ACN ) , 0 . 1% formic acid ( FA ) and 0 . 01% trifluoroacetic acid ( TFA ) , and B ( 95% ACN , 5% water , 0 . 1% FA , and 0 . 01% TFA ) as follows: 5%–10% B in 20 min , 10%–50% B in next 30 min , 50%–80% B in next 10 min , isocratic 80% B for 5 min , 80%–5% B in 4 min . Fractions were manually collected , evaluated for peptide content by MALDI-TOF MS , and subjected to 2nd stage separation using a Micromass HPLC system ( Manchester , U . K . ) equipped with a C18 reverse phase column ( Dionex , 300 µm i . d . , particle size 3 µm , and pore size 100 Å ) and coupled to a HCT Ultra ion-trap mass spectrometer via an electrospray ionization source ( ESI ) ( Bruker Daltonics , Bremen , Germany ) . Second stage separation parameters were optimized individually for each fraction using either the same water/ACN solvent system or water/methanol with 0 . 1% FA as a counter-ion . Mass spectrometric detection of eluting peptides was controlled by the Esquire software ( Bruker Daltonics , Bremen , Germany ) in a data-dependent manner . Tandem MS ion precursor selection was limited to 3 ions per min sorted by signal intensity , preferred charge state was set to +2 , and the active dynamic exclusion of previously fragmented precursor ions limited to 2 spectra per minute . The scan m/z ranges for MS and MS/MS were 300–1 , 800 and 50–3 , 000 , respectively . For peptide identification , tandem mass spectra were converted to the . mgf file format ( Mascot generic file ) and processed for sequencing automatically using the PEAKS Studio 4 . 5 software ( Bioinformatics Solutions , Inc . , Waterloo , CA ) . PEAKS generated data were manually inspected and verified . Automatic sequencing was performed against an in-house planarian prohormone database using the following search parameters: cleavage sites , variable Post-Translational Modifications ( PTMs ) ( including N-terminal pyro-Glu and pyro-Gln , C-terminal amidation , and disulfide bond; the maximum number of PTMs on a single peptide was set to four ) , mass tolerance equal 0 . 3 Da for the precursor ion , and 0 . 5 Da for fragments . Criteria for peptide assignments and prohormone confirmation were based on confidence scores generated by PEAKS for each sequenced peptide and detection mass error . A PEAKS confidence score is given as a percentage value from 1% to 99% and represents the statistical likelihood that an amino acid sequence matches a given MS fragmentation spectrum . The PEAKS statistical algorithm considers factors such as signal to noise , total intensity , and spectrum tagging ( PEAKS Studio Manual 4 . 5 http://www . bioinformaticssolutions . com/products/peaks/support/PEAKSStudioManual4 . 5 . pdf ) . Our results are based on the current database of 51 prohormones . Our criteria for the validation of a prohormone include the identification of at least one peptide from the prohormone with a PEAKS score >80% and a mass accuracy ≤300 ppm , or with a score of >50% and a mass accuracy within 150 ppm . In addition , we manually verified automatic sequencing results , examined prohormone cleavage sites , and evaluated the possible PTMs of the identified peptides . A match of at least three consecutive fragments in an ion series from manual sequencing to an automatically generated peptide sequence was considered sufficient to validate the peptide assignment . As prohormone identification increases with the number of detected encoded peptides , we employed high identification criteria for the first peptide but allowed lower standards for assignment of additional peptides from the same prohormone ( PEAKS score >20% , mass accuracy ≤500 ppm ) provided the fragmentation spectrum satisfied manual verification . In cases in which a prohormone had already been confirmed by tandem MS , occasionally we assigned peptides by mass match with MALDI-TOF-MS data . Such assignments were based on a mass-match within 200 ppm to protonated molecular ions of peptides predicted by NeuroPred ( http://neuroproteomics . scs . uiuc . edu/neuropred . html ) [52] . These assignments are tentative since they are not accompanied by sequencing data . Two distinct bioinformatic approaches were used to identify prohormone genes in the S . mediterranea genome . First , similarity searches were performed with collections of peptides or prohormones from invertebrate species such as Drosophila melanogaster , Aplysia californica , Apis mellifera [32] , Caenorhabditis elegans [73] , and various Platyhelminthes [39] with stand-alone BLAST ( BLOSSUM62 or PAM30 matrices and Expect values ≥10 ) . Peptides and prohormones were obtained from genome databases ( i . e . Wormbase , http://www . wormbase . org ) , from NCBI , or from an online catalog of bioactive peptides ( http://www . peptides . be , [103] ) . Additionally , sequence tags generated by de novo MS sequencing of unassigned peptides were also used as queries for genomic BLAST searches ( BLOSSUM62 or PAM30 matrices and Expect values ≥10 ) . As an alternative to similarity searching we analyzed translated S . mediterranea EST [98] , [104] and 454 ( Roche , Mannheim , Germany ) sequence data ( Y . Wang and P . A . Newmark , unpublished ) for sequences that possessed characteristics of prohormone genes including multiple dibasic cleavage sites and a signal sequence ( www . cbs . dtu . dk/services/SignalP ) . Translations of nucleotide sequences were performed with longorf . pl , a script that translates the longest open reading frame in a nucleotide sequence ( www . bioperl . org/wiki/Bioperl_scripts ) . Putative prohormone genes identified using these two approaches were used as queries to search the S . mediterranea genome to determine if additional related prohormones existed in the genome . The full-length coding sequences of prohormone genes were predicted using a variety of gene and splice-site prediction tools , including NetGene2 ( http://www . cbs . dtu . dk/services/NetGene2 ) , FSPLICE ( http://www . softberry . com ) , GENSCAN ( http://genes . mit . edu/GENESCAN . html ) , and GeneQuest ( v8 . 0 . 2 , DNASTAR , Madison , WI ) . Where full-length sequences could not be predicted in silico , 5′ and 3′ Rapid Amplification of cDNA Ends ( RACE ) ( FirstChoice RLM-Race Kit , Ambion , Austin , TX ) analyses were performed following the manufacturer's protocol . The predictions of all genes reported here were independently verified by cDNA analysis ( see below ) . Once verified , genes were considered to be genuine prohormone genes if they ( 1 ) possessed a signal sequence , ( 2 ) possessed basic cleavage sites that flanked predicted or MS-confirmed peptides , and ( 3 ) were less than 200 amino acids in length . Sequences were excluded if they shared similarity with genes previously annotated to be other than neuropeptide prohormones . All genes were named according to the S . mediterranea genome nomenclature guidelines [105] . Translated nucleotide sequences were downloaded either from the Schistosoma mansoni FTP server ( ftp . sanger . ac . ik/pub/pathogens/Schistosoma/mansoni ) or from the NCBI taxonomy browser ( http://www . ncbi . nlm . nih . gov/Taxonomy/ ) . These sequences were then compared to the sequences of MS-confirmed S . mediterranea prohormones using BLASTP . NPY-family members were not included in this analysis , although three NPY-like proteins have been previously described in Schistosoma [39] , [101] . Additionally we analyzed EST sequences in the NCBI database to identify schistosome prohormone genes . Newly annotated schistosome prohormones were analyzed further with SignalP and Neuropred to predict final gene products . These genes were named as described previously [39] . To facilitate efficient analyses of prohormone genes , we constructed a plasmid vector that permits TA-mediated cloning of PCR-amplified cDNAs . To generate a suitable vector backbone , oligonucleotide primers 5′-GATCACGCGTCGATTTCGGCCTATTGGTTA-3′ and 5′-GATCACGCGTGCTTCCTCGCTCACTGACTC-3′ were used to amplify the kanamycin and ampicillin resistance markers and the origin of replication of plasmid pCRII ( Invitrogen , Carlsbad , CA ) ; this PCR product was digested with MluI and ligated to generate a circular plasmid . Following circularization , an Eam1105I restriction site was removed from the β-lactamase gene of this plasmid by introduction of a silent mutation using site-directed mutagenesis ( Quickchange II , Statagene , La Jolla , CA ) . For the functional elements of the vector , two mini genes were synthesized ( Integrated DNA Technologies , Coralville , IA ) : T7TermSP6 and T7TermT3 . T7TermSP6 included ( 5′ to 3′ ) KpnI , MluI , T7-terminator , AscI , T7 Promoter , SP6 promoter , GACCTTAGGCT ( an Eam1105I site ) , and XhoI . T7TermT3 included ( 5′ to 3′ ) SacI , MluI , T7 terminator , T7 promoter , T3 promoter , GACCTTAGGCT ( an Eam1105I site ) , and NotI . T7TermSP6 and T7TermT3 were shuttled to pBluescript SK II+ using the KpnI and XhoI sites from T7TermSP6 or the SacI and NotI sites from T7TermT3 . These plasmids were digested with MluI and EcoRI and ligated with the MluI site of the vector backbone . A XhoI and NotI-digested PCR fragment including the ccdB and camR genes from plasmid pPR244 [47] were inserted to generate the final plasmid-pJC53 . 2 . Eam1105I ( Fermentas , Burlington , Ontario ) restriction digest of this plasmid generates 3′ T overhangs that can be ligated to an A-tailed Taq polymerase-amplified PCR product [106] . The ccdB gene prevents any undigested plasmid from giving rise to viable clones [107] . Once cDNAs have been inserted into pJC53 . 2 , riboprobes for in situ hybridization analysis can be generated by in vitro transcription with SP6 or T3 RNA polymerases and dsRNA for RNAi knockdowns can be generated by in vitro transcription with T7 RNA polymerase , or by transformation of E . coli ( HT115[DE3] ) [108] . To generate riboprobes for in situ hybridization , prohormone genes not represented by EST clones [98] were PCR amplified ( Platinum Taq , Invitrogen , Carlsbad , CA ) from cDNA generated from total RNA ( iScript cDNA Synthesis Kit , Bio-Rad , Hercules , CA ) or 3′ RACE cDNA ( RLM-RACE Kit , Ambion , Austin , TX ) generated from either total or poly- ( A ) + RNA ( Poly-A Purist , Ambion , Austin , TX ) . For cDNA preparations , RNA was extracted using Trizol Reagent ( Invitrogen , Carlsbad , CA ) . For cloning , 2–3 µL of PCR product was ligated with 70 ng of Eam1105I-digested pJC53 . 2 ( Rapid DNA Ligation Kit , Roche , Mannheim , Germany ) and used to transform DH5α . In vitro transcriptions with the appropriate RNA polymerase were performed using standard approaches with the addition of Digoxigenin-12-UTP ( Roche , Mannheim , Germany ) , Fluorescein-12-UTP ( Roche , Mannheim , Germany ) , or Dinitrophenol-11-UTP ( Perkin Elmer , Waltham , MA ) . In situ hybridizations were performed using the formaldehyde-based fixation procedure essentially as described previously [109] . However , due to their large size , sexual animals were killed in 10% N-Acetyl Cysteine , fixed for 20–30 min in 4% Formaldehyde in PBSTx ( PBS+0 . 3% Triton X-100 ) , permeabilized with 1% SDS ( 10 min at RT ) prior to reduction ( 10 min at RT ) , and treated with 10 µg/mL Proteinase K ( 10–20 min at RT ) after bleaching . Some samples were processed in either a BioLane HTI ( Hölle & Hüttner , Tübingen , Germany ) [98] or an Insitu Pro ( Intavis , Koeln , Germany ) hybridization robot [102] . Sexual animals were imaged with either a Microfire digital camera ( Optronics , Goleta , CA ) mounted on a Leica MZ12 . 5 stereomicroscope or a Leica DFC420 camera mounted on a Leica M205A stereomicroscope ( Leica , Wetzlar , Germany ) . Both microscopes were equipped with a Leica TL RC base . Asexual animals were imaged over a piece of white filter paper and illuminated from above with an LED light source . For FISH , following post-hybridization washes and blocking , animals were incubated in α-Digoxigenin-POD ( 1∶1000 , Roche , Mannheim , Germany ) , α-Fluorescein-POD ( 1∶1000 , Roche , Mannheim , Germany ) , or α-Dinitrophenol-HRP ( 1∶100 , Perkin Elmer , Waltham , MA ) overnight at 4°C , washed in MABT , equilibrated in TNT ( 100 mM Tris pH 7 . 5 , 150 mM NaCl , and 0 . 05% Tween-20 ) , and developed in Amplification Diluent containing a fluorescent-tyramide conjugate ( Cy3-tyramide , Cy5-tyramide , or Fluorescein-tyramide; TSA-Plus , Perkin Elmer , Waltham , MA ) . Following development , animals were washed in TNT and HRP activity was quenched by a 1 h incubation in 1 . 5%–2 . 0% H2O2 dissolved in TNT . Following HRP inactivation , animals were washed in MABT , incubated in a different α-hapten-HRP antibody , and the process was repeated with a different fluorescent-tyramide conjugate . Samples were mounted in Vectashield ( Vector Laboratories , Burlingame , CA ) and imaged on a Zeiss LSM 710 confocal microscope ( Carl Zeiss , Germany ) ( Plan-Apochromat 20×/0 . 8 , C-Apochromat 40×/1 . 2 W korr UV-VIS-IR , or Plan-Apochromat 63×/1 . 4 Oil DIC objectives ) . Fluorescein , Cy3 , and Cy5 were excited with 488 nm , 561 nm , and 633 nm lasers , respectively . Images were processed using either Zen 2008 ( Carl Zeiss , Germany ) or ImageJ [110] . Northern blot procedures were performed essentially as previously described [111] and hybridization signals were detected using an anti-digoxigenin alkaline phosphatase-conjugated antibody and chemiluminescence ( CDP-STAR , Roche , Mannheim , Germany ) . Chemiluminescent signals were detected using a FluorChem Q ( Alpha Innotech , San Leandro , CA ) . Sequences of EST clones corresponding to pc2 [43] , [98] were assembled with one another and the S . mediterranea genome ( Sequencher 4 . 7 , Gene Codes , Ann Arbor , MI ) to determine the full-length sequence and genomic structure of the pc2 gene . For RNAi analysis of pc2 , EST clone PL05006A1C09 [98] , which corresponds to pc2 , was shuttled to plasmid pPR244 using a Gateway reaction ( Invitrogen , Carlsbad , CA ) [47] . For npy-8 RNAi , a 3′ RACE product specific to npy-8 was cloned in pJC53 . 2 . RNAi feedings were performed essentially as described previously [112] , with some modifications . In pc2 RNAi experiments , ∼6 . 25 mL of IPTG-induced culture was pelleted , frozen at −80°C , and resupended in 30 µL of a mixture of homogenized beef liver and water . ∼5 mature sexual animals ( >1 cm in length ) received 1–2 feedings over the course of ∼48 h . npy-8 RNAi experiments were performed similarly to pc2 RNAi except feedings included 50% less bacteria and animals were fed every 5–7 d over the indicated time course; for some feedings , bacteria were omitted . On occasion , because of either refusal to feed or improper nutrition , some animals ( both controls and treatment groups ) decreased in size over the long time courses of the npy-8 RNAi experiments . Therefore , only animals >1 cm in length at the time of fixation were included in our analyses at time points greater than 4 wk . For all RNAi experiments with bacterially expressed dsRNA , control feedings were performed with bacteria containing empty plasmid pPR242 . For RNAi experiments conducted with juvenile planarians , dsRNA was generated by in vitro transcription [113] , [114] . To generate dsRNA , templates cloned in pJC53 . 2 were amplified with a modified T7 oligonucleotide ( GGATCCTAATACGACTCACTATAGGG ) , cleaned up using the DNA Clean & Concentrator kit ( Zymo Research , Orange , CA , D4003 ) , and eluted in 10 µL of water . 4 µL of each PCR product was used as template for in vitro transcription in a reaction containing 5 . 5 µL DEPC-treated water , 5 µL 100 mM mix of rNTPs ( Promega , E6000 ) , 2 µL high-yield transcription buffer ( 0 . 4 M Tris pH 8 . 0 , 0 . 1 M MgCl2 , 20 mM spermidine , 0 . 1 M DTT ) , 1 µL thermostable inorganic pyrophosphatase ( New England Biolabs , Madison , WI , M0296S ) , 0 . 5 µL Optizyme recombinant ribonuclease inhibitor ( Fisher Scientific , Pittsburg , PA , BP3222-5 ) , and 2 µL HIS-Tagged T7 RNA polymerase [115] . Samples were incubated at 37°C for 4–5 h and then treated with RNase-free DNase ( Fisher Scientific , Pittsburg , PA , FP2231 ) . Synthesized RNA was then melted by heating at 75°C , 50°C , and 37°C each for 3 min . 2 . 5–10 µg of each dsRNA solution was mixed with 45 µL of 3∶1 liver to water mix and used to feed up to 8 worms . For these experiments , animals without visible gonopores ( juveniles ) were fed every 4–5 d for the indicated time period and starved 1 wk before fixation . Unless otherwise specified , as a negative control , animals were fed dsRNA synthesized from the ccdB and camR-containing insert of pJC53 . 2 . To analyze the structure of the testes , animals were killed in 2% HCl for 3 min , fixed in either Methacarn ( 6 MeOH:3 Chloroform: 1 Glacial Acetic Acid ) or 4% formaldehyde for 1–2 h , dehydrated in MeOH , bleached in 6% H2O2 in MeOH , and stained with 4′ , 6-diamidino-2-phenylindole ( DAPI ) ( Sigma-Aldrich , St . Louis , MO ) . Alternatively , samples were processed for in situ hybridization , as described above . Following staining , animals were mounted in Vectashield , flattened , and imaged on either a Zeiss SteREO Lumar ( Carl Zeiss , Germany ) or a Zeiss LSM 710 confocal microscope ( DAPI was excited with a 405 nm laser ) . To examine if npy-8 ( RNAi ) affected overall growth , animals were immobilized on ice and imaged on a Leica M205A stereomicroscope . The area of each animal was determined using ImageJ . To examine transcript levels in npy-8 knockdowns , juvenile animals were fed either liver homogenate or 45 µL of liver homogenate mixed with 2 . 5 µg of in vitro synthesized npy-8 dsRNA . 7 d later RNA was extracted from individual planarians using Trizol Reagent ( Invitrogen , Carlsbad , CA ) . Following DNase treatment ( DNA-free RNA Kit , Zymo Research , Orange , CA ) , reverse transcription was performed ( iScript cDNA Synthesis Kit , Bio-Rad , Hercules , CA ) and quantitative PCR was conducted using Power SYBR Green PCR Master Mix ( Applied Biosystems , Warrington , UK ) and a 7900HT real-time PCR system ( Applied Biosystems ) . Standard curves were generated from serial dilutions of either plasmid DNA containing the gene of interest ( npy-8 and npy-1 ) or from genomic DNA ( β-tubulin GB: DN305397 ) . All samples were measured in triplicate to account for pipetting error . Absolute quantities of each transcript were determined from the standard curves and the levels of npy-8 or npy-1 were normalized to the level of β-tubulin in each sample . The mean value ( i . e . npy-8/β-tubulin or npy-1/β-tubulin ) for each treatment ( i . e . control or npy-8 ( RNAi ) ) was then compared using a Student's t test . The primers used for these studies were npy-8 Forward AATCAGAAAAGGCCGATGTTTG , Reverse CAAATAGTTCCGAAAGGCATCAG; npy-1 Forward GTCGACCAAGATTCGGTAAACG , Reverse CATTCTTTTATGAAAATCCCCTGT; β-tubulin F TGGCTGCTTGTGATCCAAGA R AAATTGCCGCAACAGTCAAATA . To investigate the effect of pc2 RNAi on the proteolytic processing of prohormones , peptide profiles were measured by MALDI-TOF MS and compared by principal component analysis followed by a t test in tissue extracts prepared from 7 individual control and 7 individual RNAi-treated animals . Extracts were prepared by homogenizing each specimen in 100 µL of acidified acetone ( see above ) . Following centrifugation at 14 , 000× g for 15 min , supernatant was collected , dried in SpeedVac concentrator ( Thermo Scientific , San Jose , CA ) , and reconstituted in 30 µL of 0 . 01% TFA . For MALDI-TOF MS analysis , 0 . 7 µL of each extract was spotted on a stainless steel sample holder and co-crystallized with 0 . 7 µL of freshly prepared concentrated DHB matrix ( DHB: 2 , 5-dihydroxybenzoic acid , 50 mg/mL 50% acetone ) . Three technical replicates were sampled for each biological sample , 42 spots total . Positive ion mass spectra were acquired manually in 600–6 , 000 m/z region using a Bruker Ultraflex II mass spectrometer in linear mode with external calibration . For each spot 700 laser shots in 7 acquisitions were accumulated into a sum spectrum representative of a replicate . For comparison of peptide profiles in control and pc2 ( RNAi ) animals , raw MALDI-TOF MS data were loaded into an evaluation version of ClinProTools software ( Bruker Daltonics , Bremen , Germany ) using the following processing parameters: convex hull baseline subtraction , baseline flatness 0 . 2 , mass range 1 , 000–6 , 000 m/z , Savizky-Golay smoothing over 1 m/z width with 11 cycles , data reduction factor of 10 , null spectra exclusion enabled , recalibration with maximum peak shift of 200 ppm . All spectra were normalized to the total ion count ( TIC ) prior to PCA calculations . Sum spectra from technical replicates were grouped into a representative sample spectrum in ClinProTools , thus representing a biological replicate for statistical calculations . From representative sample spectra a mean spectrum was generated by ClinProTools to reveal general peptide features for control and pc2 ( RNAi ) groups . Standard deviation of signal intensities among biological replicates was derived for each peak in the group profile . Unlimited peak picking on the base of maximal peak intensity and minimal signal-to-noise ratio of 6 was done on the mean spectrum representative of each sample group in order to take advantage of noise reduction effect due to spectra addition . Peptide profiles of mean spectra representative of biological replicates were compared by principal component analysis followed by Anderson-Darling ( AD ) normality test and paired Student's t test for peaks showing normal distribution . Peaks not showing a normal distribution ( pAD≤0 . 05 ) were evaluated by the Wilcoxon or Kruskal-Wallis tests , respectively [116]–[118] . To decrease the number of false positives while computing individual peak statistics on highly complex spectra , the Benjamini-Hochberg procedure incorporated into ClinProTool was automatically applied for p value adjustment during analysis [119] . | Flatworms cause diseases affecting hundreds of millions of people , so understanding what influences their reproductive activity is of fundamental importance . Neurally derived signals have been suggested to coordinate sexual reproduction in free-living flatworms , yet the neuroendocrine signaling repertoire has not been characterized comprehensively for any flatworm . Neuropeptides are a large diverse group of cell-cell signaling molecules and play many roles in vertebrate reproductive development; however , little is known about their function in reproductive development among invertebrates . Here we use biochemical and bioinformatic techniques to identify bioactive peptides in the genome of the planarian flatworm Schmidtea mediterranea and identify 51 genes encoding >200 peptides . Analysis of these genes in both sexual and asexual strains of S . mediterranea identified a neuropeptide Y superfamily member as important for the normal development and maintenance of the planarian reproductive system . We suggest that understanding peptide hormone function in planarian reproduction could have practical implications in the treatment of parasitic flatworms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/germ",
"cells",
"chemical",
"biology/protein",
"chemistry",
"and",
"proteomics",
"neuroscience",
"genetics",
"and",
"genomics/functional",
"genomics"
] | 2010 | Genome-Wide Analyses Reveal a Role for Peptide Hormones in Planarian Germline Development |
Assessment of the response to the 2014–15 Ebola outbreak indicates the need for innovations in data collection , sharing , and use to improve case detection and treatment . Here we introduce a Machine Learning pipeline for Ebola Virus Disease ( EVD ) prognosis prediction , which packages the best models into a mobile app to be available in clinical care settings . The pipeline was trained on a public EVD clinical dataset , from 106 patients in Sierra Leone . We used a new tool for exploratory analysis , Mirador , to identify the most informative clinical factors that correlate with EVD outcome . The small sample size and high prevalence of missing records were significant challenges . We applied multiple imputation and bootstrap sampling to address missing data and quantify overfitting . We trained several predictors over all combinations of covariates , which resulted in an ensemble of predictors , with and without viral load information , with an area under the receiver operator characteristic curve of 0 . 8 or more , after correcting for optimistic bias . We ranked the predictors by their F1-score , and those above a set threshold were compiled into a mobile app , Ebola CARE ( Computational Assignment of Risk Estimates ) . This method demonstrates how to address small sample sizes and missing data , while creating predictive models that can be readily deployed to assist treatment in future outbreaks of EVD and other infectious diseases . By generating an ensemble of predictors instead of relying on a single model , we are able to handle situations where patient data is partially available . The prognosis app can be updated as new data become available , and we made all the computational protocols fully documented and open-sourced to encourage timely data sharing , independent validation , and development of better prediction models in outbreak response .
The 2014–15 EVD outbreak in West Africa has eclipsed in magnitude all combined past EVD outbreaks since the disease was first identified in 1976 [1] . As of February 17 , 2016 ( http://www . cdc . gov/vhf/ebola/outbreaks/2014-west-africa/case-counts . html ) , a total of 28 , 639 cases have been reported ( 15 , 251 laboratory-confirmed ) and 11 , 316 total deaths . The outbreak constitutes one of the most serious worldwide health emergencies in modern times , with severe socioeconomic costs , particularly in the West African nations of Liberia , Sierra Leone , and Guinea . Although vaccine development is promising [2] , the prospect of future outbreaks looms . The report of the WHO Ebola Interim Assessment Panel also points to several shortcomings in the initial response [3] , noting that “better information was needed to understand best practices in clinical management” and that “innovations in data collection should be introduced , including geospatial mapping , mHealth communications , and platforms for self-monitoring and reporting” . Given these circumstances , the development of accurate and accessible computational methods to track the progression of the outbreak and model various aspects of the disease is beneficial not only for the research community , but also for health care personnel in the field . In particular , prognosis prediction models based on the available patient information would be of great utility . Such predictive models can identify the clinical symptoms and laboratory results that should be tracked most closely during the onset of EVD , and give health care workers the ability to more accurately assess patient risk and therefore manage treatment more efficiently [4] . This data-driven prioritization could lead to higher recovery rates through stratified treatment [5] , especially in resource-constrained areas , and would help doctors limit the evaluation of experimental EVD vaccines and treatments [6] with potentially harmful side effects only to highest-risk patients . These improvements in treatment , however , will only be achieved once larger datasets become available to overcome biases resulting from small samples . Schieffelin et al . [7] presented the only publicly accessible , at the time of publication , clinical dataset from the West African EVD outbreak ( available in various formats at http://fathom . info/mirador/ebola/datarelease ) to enable clinical investigations . Although a large amount of very useful case and resource data has been made public throughout the outbreak ( https://data . hdx . rwlabs . org/ebola ) , thanks to the efforts of numerous individuals and organizations , there is to our knowledge no other public source offering a similar level of clinical detail . The Schieffelin et al . dataset includes epidemiologic , clinical , and laboratory records of 106 patients treated at Kenema Government Hospital in Sierra Leone during the initial stages of the outbreak . The study also provides a simple heuristic to estimate mortality risk by defining an Ebola Prognostic Score ( EPS ) , which predicts patient outcome based on symptom counts . EPS offers statistically significant differences between surviving and deceased patients with p < 0 . 001 . While data from other published clinical studies are not available , their summary results suggest that more advanced prognostic prediction models could be potentially useful to the field . Levine et al . [8] developed a diagnostics model using data from the Bong County Ebola Treatment Unit in Liberia , which predicts laboratory-confirmed EVD cases using six clinical variables . Yan et al . [9] carried out a multivariate analysis of 154 EVD patients from the Jui Government Hospital in Sierra Leone , and reported that age , fever , and viral load are independent predictors of mortality , while Zhang et al . [10] recently reported that age , chest pain , coma , confusion , and viral load are associated with EVD prognosis using a set of 63 laboratory-confirmed cases also from the Jui Government Hospital . In this study , we employed the Schieffelin et al . EVD dataset to develop novel predictive models for patient prognosis , integrating a data-driven hypothesis making approach with a customizable Machine Learning ( ML ) pipeline , and incorporating rigorous imputation methods for missing data . We evaluated the predictors using a variety of performance metrics , identifying top predictors with and without viral load measurements , and packaged them into a mobile app for Android ad iOS devices ( http://fathom . info/mirador/ebola/prognosis ) . Our protocol exemplifies how data-driven computational methods can be useful in the context of an outbreak to extract predictive models from incomplete data , and to provide rapidly actionable knowledge to health workers in the field . Moreover , prognosis prediction software could complement ongoing efforts to develop rapid EVD diagnostics [11] and safe data-entry devices [12] . Given the availability of only one dataset from a single location , one Ebolavirus species ( Zaire ebolavirus ) , and very specific time span and laboratory protocols , these models need to be interpreted in an exploratory sense and require further validation with independent clinical data from other EVD treatment sites [8] [9] [13] [14] [15] . We have made all of these resources publicly available and fully documented with the hope to encourage further methods development , independent validation , and greater data sharing in outbreak response .
Our analysis and modeling is based on the EVD clinical and laboratory data initially described by Schieffelin et al [7] . The Sierra Leone Ethics and Scientific Review Committee and the ethics committee at Harvard University have approved the study and public release of this clinical data , which has been de-identified to protect patient privacy . As indicated by Schieffelin , “these committees waived the requirement to obtain informed consent during the West African Ebola outbreak” and “all clinical samples and data were collected for routine patient care and for public health interventions . ” The larger dataset comprises 213 suspected cases evaluated for Ebola virus infection at the Kenema Government Hospital ( KGH ) in Sierra Leone between May 25 and June 18 , 2014 . Outcome data was available for 87 of 106 Ebola-positive cases , giving a Case Fatality Rate ( CFR ) of 73% over the entire dataset . We considered 65 patients between 10 and 50 years of age . Within this group , not all individuals had complete clinical chart , metabolic panel , and virus load records available ( Fig 1 ) . Sign and symptom data were obtained at time of presentation on 34 patients that were admitted to KGH and had a clinical chart . Metabolic panels were performed on 47 patients with adequate sample volumes , with a Piccolo Blood Chemistry Analyzer and Comprehensive Metabolic Reagent Discs ( Abaxis ) , following the manufacturer’s guidelines . Virus load was determined in 58 cases with adequate sample volumes using the Power SYBR Green RNA-to-CT 1-Step quantitative RT-PCR assay ( Life Technologies ) at Harvard University . Both metabolic panel and PCR data used to develop our models was collected during triaging of the patients upon admission , and follow-up data , although available for some patients , was not included in our analyses . We compiled this data into a single file in CSV format , and made it available in a public repository ( http://dx . doi . org/10 . 5281/zenodo . 14565 ) , together with all original Excel spreadsheets and the cleaning and aggregation scripts ( http://fathom . info/mirador/ebola/datarelease ) , as well as a Dataverse hosted on the Harvard Dataverse Network ( http://dx . doi . org/10 . 7910/DVN/29296 ) . In a separate effort , we designed the tool Mirador ( http://fathom . info/mirador/ ) to allow users to identify statistical associations in complex datasets using an interactive visualization interface . This visual analysis is guided by an underlying statistical module that ranks the associations using pairwise Mutual Information [16] . Mirador automatically computes a sample estimate of the Mutual Information between each pair of variables inspected by the user , and performs a bootstrap significance test [17] to determine if the variables are independent within a confidence level set through the interface . This calculation relies on an optimal bin-width algorithm [18] , which finds the grid minimizing the Mean Integrated Squared Error between the estimates from the data and the underlying joint distributions . The user can then rigorously test the hypothesis of association suggested by Mirador using more specialized tools such as R or SPSS , and finally incorporate them into predictive models . We used the Maximal Information Coefficient ( MIC ) statistic developed by Reshef et al [19] , calculated with the MINE program ( http://www . exploredata . net/ ) , to rank the associations found with Mirador . Since only 21 patients in the dataset contain complete clinical , laboratory , and viral load information , we applied three Multiple Imputation ( MI ) programs to impute the missing values: Amelia II , which assumes the data follows a multivariate normal distribution and uses a bootstrapped expectation-maximization algorithm to impute the missing values [20]; MICE [21] Multivariate Imputation by Chained Equations , where missing values in each variable are iteratively imputed given the other variables in the data until convergence is attained; and Hmisc [22] , which is also based on the chained equations method . All MI methods require that the missing entries satisfy the Missing Completely At Random ( MCAR ) condition in order to generate unbiased results . Specifically , MCAR means that the distribution of the missing entries is entirely random and does not depend neither on the observed nor the missing values . Furthermore , Amelia requires the observed data to follow a multivariate normal distribution . We used Little’s MCAR chi-square test [23] and Jamshidian and Jalal's test for Homoscedasticity , Multivariate Normality , and MCAR [24] to rigorously test for these conditions . After testing for the MCAR condition , we run each MI program m times to generate m “completed” copies of the original dataset , which we aggregated into a single training set of larger size ( S4 Fig ) . We performed a detailed comparison of the performance of the predictor when using values imputed by each of the three MI programs , which is described in the results . The ML pipeline takes as inputs the source data and a list of covariates , and outputs a trained predictor that can be evaluated with several accuracy metrics . It includes the following classifiers: a single-layer Artificial Neural Network ( ANN ) [25] implemented from scratch , and Logistic Regression ( LR ) , Decision Tree ( DT ) , and Support Vector Machine ( SVM ) classifiers from scikit-learn [26] . Each classifier was trained on all possible combination of input covariates , from the subset of found with Mirador and MINE , to avoid issues with variable selection methods [27] , and to generate an ensemble of predictors that could be applied to different combinations of available clinical data . We applied multiple cross-validation in order to train the classifiers for each selection of covariates . We first split the records without missing values into two sets with identical CFR , then set one aside for model testing . We combined the second set with the remaining records that include missing values , and used this data as the input for the MI programs . Depending on the percentage of complete records reserved for testing and the number of MIs , we ended up with testing sets of 6–10 cases and training sets of 200–300 cases . This ensured having more than 10 samples per variable during predictor training , the accepted minimum in predictive modeling [28] . We generated 100 of such testing/training set pairs by randomly reshuffling complete records between test set and training set . Each model was initially ranked by its mean F1-score , which is the weighted average of the precision and sensitivity . The mean and standard deviation were calculated over the 100 cross-validation iterations for each combination of input covariates . We then used the bootstrap method originally introduced by Harrell [29] to quantify the optimistic bias [30] in the area under the receiver operator curve ( AUC or c-statistic ) . We generated 100 bootstrap samples with replacement for each model , and re-trained the model on these samples . We evaluated the AUC on the bootstrap sample and the original sample , and reported the mean of the AUCboot—AUCorig difference as the estimated optimism . Finally , we carried out standard logistic regression with variable selection , with the goal of evaluating the effect of our MI protocol on other model selection algorithms , and comparing the resulting standard model with the top-ranking models from our pipeline . We used the built-in step ( ) function in R to perform backward variable selection with the Akaike Information Criterion ( AIC ) , the ROCR package to compute AUC , and the Boot package to estimate of the optimistic bias with bootstrap sampling . The ML models generated by our pipeline are essentially Python scripts together with some parameter files . The Kivy framework ( http://www . kivy . org ) allowed us to package these scripts as mobile apps that can be deployed on tablets or smartphones through Google or Apple’s app stores . We created a prototype app including the models described in this paper , currently available as Ebola CARE ( Computational Assignment of Risk Estimates ) , shown in Fig 2 . We have only implemented the ANN classifier into the Ebola CARE app for the time being , because the scikit-learn classifiers could not be compiled to run on Android devices , which is a requirement for our prognosis app . Once installed , the app is entirely stand-alone , does not require Internet connectivity to run , and can be updated once better models are available .
We began by identifying the clinical and laboratory factors that provide the strongest association with EVD outcome . Earlier reports indicate that EVD mortality rates in this outbreak are found to be significantly different among children [31] and older adults [7] , and this pattern holds in our data: CFR is higher than 90% for the 18 patients older than 50 years of age , and 75% for the 14 patients under 10 years of age; we therefore restricted our analyses to patients between 10 and 50 years of age . Within this age range , exploratory analysis with Mirador ( http://fathom . info/mirador/ ) , led us to identify 24 clinical and laboratory factors that show plausible association with EVD outcome: virus load ( PCR ) , temperature ( temp ) , aspartate aminotransferase ( AST ) , Calcium ( Ca ) , Alkaline Phosphatase ( ALK ) , Chloride ( Cl ) , Alanine Aminotransferase ( ALT ) , Creatinine ( CRE ) , Total Carbon Dioxide ( tCO2 ) , Albumin ( Alb ) , Blood Urea Nitrogen ( BUN ) , Total Protein ( TP ) , weakness , vomit , edema , confusion , respiratory rate , back pain , dizziness , retrosternal pain , diarrhea , heart rate , diastolic pressure , and abdominal pain . Boxplots and histograms for all factors are depicted in S1 and S2 Figs , which also presents the P-values for the association between Outcome and each factor , for the Fisher exact and T-tests ( for nominal and numerical factors , respectively ) . We applied the Maximal Information Coefficient ( MIC ) statistic developed by Reshef et al . [19] , calculated with the MINE program ( http://www . exploredata . net/ ) , to rank these 24 factors . We used the ranking to select two informative subsets of 10 variables each ( shown in Fig 3 ) , one with PCR and the other without , by picking the top 5 laboratory results and top 5 clinical chart variables . The PCR set comprises PCR , temp , AST , ALK , CRE , tCO2 , heart rate , diarrhea , weakness , and vomit , while the non-PCR set includes temp , AST , ALK , CRE , tCO2 , BUN , heart rate , diarrhea , weakness , and vomit . None of these variables are capable of predicting outcome accurately in isolation . The performance of the univariate LR classifier is highest with PCR as input , with an F1-score of 0 . 67 , and below 0 . 5 for all other variables . This result is consistent with the recent report from Crowe et al . [32] , which highlights the importance of viral load in the prognosis of EVD . We evaluated the impact of the MI step on the predictors’ performance , and chose MI parameters accordingly . In all three MI modules , Amelia II , MICE and Hmisc , we can adjust the fraction of complete records to be included in the data to impute , as well as the number of imputed copies that are aggregated into a single training set . We considered all combinations of these two parameters , when allowing 20% , 35% , and 50% as the percentages of complete records used during imputation , and 1 , 5 and 10 for the number of imputed copies . We examined the resulting 9 combinations of parameters across the 4 predictors , LR , ANN , DT , and SVM . Accuracy , as measured by mean F1-score , in the PCR case does not seem to depend on the number of imputed copies , percentage of completed records , and MI algorithm ( S3A Fig ) . In contrast , both higher percentage of completed records and higher number of imputed copies do have a definite enhancing effect in the mean F1-score for the non-PCR case ( S3B Fig ) , while the choice of MI algorithm does not seem to have a significant impact . Counter intuitively , the standard deviation of the F1-score in the PCR case increases with larger percentage of completed records . However , this trend can be explained as follows: the complete records not included in the training set are used to construct the testing set , therefore higher percentages of complete records used during MI result in smaller testing sets . The effect of a single false positive or negative is proportionally larger in smaller testing sets than in larger ones , which results in higher variation of the F1-score in the latter . We then verified the validity of the MCAR condition in both the PCR and non-PCR sets , crucial to guarantee unbiased imputations , using Little’s chi-square test and Jamshidian and Jalal’s test . Since the all data used in our models was collected at presentation , there is lower risk of non-random missing patterns due to patient death and withdrawal . The tests for MCAR indeed confirm this: Little’s statistic takes a value of 45 . 28 with a P-value of 0 . 11 on the PCR set , while the non-PCR set gives a statistic value of 19 . 06 with a P-value of 0 . 32 , meaning that in both cases there is no evidence in the data against the MCAR hypothesis at the 0 . 05 significance level . Furthermore , Jamshidian and Jalal’s test for Homoscedasticity , Multivariate Normality , and MCAR does not reject the multivariate normality or MCAR hypothesis at the 0 . 05 significance level for both the PCR and non-PCR sets , with P-values of 0 . 79 and 0 . 06 , respectively . This last result in particular validates the use of the Amelia II package , which assumes that the data follows a normal distribution . Based on these findings as well as on a published review from Horton et al . [33] , which shows a marginal improvement with Amelia over the other MI methods , we chose Amelia II as the default MI method . One weakness of the Amelia II program is that combinations of variables that are highly collinear might cause the MI computation to fail to converge . We addressed this problem by re-running the MI using either MICE or Hmisc when Amelia is detected to fail converging more than 5 times . We generated out training sets with 50% of the complete records in the data to impute , and 5 imputed copies for aggregation into a single training set . The performance difference between 5 and 10 imputed copies did not seem large enough to justify the increased computing times . Having developed and carefully evaluated our models , we demonstrate that we are able to predict EVD prognosis with a mean F1-score of 0 . 9 or higher , for EVD patients aged 10 to 50 . We arrived at this by exhaustively generating two separate ensembles of predictors , one with PCR data and the other without . The predictors including PCR data are plotted on a scatter plot of the mean F1-score vs standard deviation ( Fig 4a ) computed over 100 rounds of cross-validation for each predictor . The ensemble consists of 4 × ( 29–1 ) = 2044 predictors ( LR , ANN , DT , SVM ) that were trained on all combinations of the PCR set ( 9 variables ) , having PCR as a fixed input variable . The LR and ANN classifiers are the best performers over all the four prediction methods , with 156 models ( 71 ANN , 64 LR , 21 SVM ) yielding an F1-score of 0 . 9 or higher . Similarly , we generated 4 × ( 210–11 ) = 4052 predictors without PCR data ( Fig 4c ) , which were trained on all combinations of the non-PCR set of variables ( 10 variables ) with at least two elements . We obtained 45 models ( 18 ANN , 24 LR , 3 SVM ) with a mean F1-score of 0 . 9 or higher . A number of the variables emerged as those most often included in the top-ranked models , both in the PCR and non-PCR cases respectively ( Fig 4b and 4d ) . Notably , in addition to temperature , CRE , ALK , and tCO2 levels are consistently present in the predictors including PCR , while the lack of PCR data makes AST levels and the onset of diarrhea more relevant for accurate prognosis . The optimistic bias of the AUC for the top predictors , both in the PCR and non-PCR cases , is below 0 . 01 for most of them , with a standard deviation of 0 . 03 ( Fig 5a and 5b ) . This analysis indicates that even though our models are over-fitted for the current data , the magnitude of bias is minor . S1 and S2 Tables detail all the top-performing predictors and their optimism-corrected AUC scores in the PCR and non-PCR cases , respectively . S5 Fig shows aggregated ROC curves over all the models for each predictor , for the PCR and non-PCR cases . The aggregated AUCs are 0 . 96 ( LR ) , 0 . 95 ( ANN ) , 0 . 94 ( SVM ) , and 0 . 84 ( DT ) in the PCR models , and 0 . 88 ( LR , ANN ) , 0 . 86 ( SVM ) , and 0 . 77 ( LR ) in the non-PCR models . The similar performance of our simple ANN predictor and scikit-learn’s LR classifier suggests that the dependency between the covariates and outcome can be modeled linearly , however larger datasets would enable us to train more complex ANNs with potentially better performance across different groups of patients . The comparison with variable selection shows an effect of the MI protocol similar to that observed in the top-ranked models . The optimistic bias of the AUC for the selected PCR and non-PCR models consistently decreases to less than 0 . 01 as the number of imputed copies increases from 1 to 5 ( Fig 5c and 5d ) . On the other hand , these models assign very small coefficients and odd ratios very close to 1 to the laboratory covariates ( Tables 1 and 2 ) . This suggests that most of the information in these models is captured by the clinical symptoms ( temperature , diarrhea , vomit ) , although weakness consistently presents an odd ratio less than 1 , contradicting the expected dependency with outcome . In general , the laboratory variables are the highest ranked according to MIC , and are also included in most of the top-ranked models , using either the LR or ANN classifiers . These results lead us think that the variable selection approach is discarding relevant information for outcome prediction , which we are able to capture in our ensemble of ML predictors . The Ebola CARE app packages a total of 82 ANN models , selected from those with a mean F1-score above 0 . 9 , but discarding the models with a standard deviation of 0 , in order to avoid potentially overfitted models . This set incorporates 64 PCR and 18 non-PCR models , so the app can still be used when viral load information is not available . We entered into Ebola CARE all the patients who had complete data for at least one model in the app , and recorded the risk prediction as presented after inputting the symptoms . Predictions for a total of 34 patients were obtained in this way . For this subgroup of patients , the mortality rate was 79% ( 7 survived , 27 died ) , and the app only misclassified two , one in each outcome group . In other words , the precision and sensitivity were both 0 . 96 . However , this number is likely overestimating the performance of the app , since some of these patients used in this test were also included in model training . The data used in this study is hosted at a Dataverse in the Harvard Dataverse Network ( http://dx . doi . org/10 . 7910/DVN/29296 ) , the source code of Mirador and the ML pipeline is available on Github ( https://github . com/mirador/mirador , https://github . com/broadinstitute/ebola-predictor ) , and the model files ( all training and testing sets ) are deposited on Zenodo ( http://dx . doi . org/10 . 5281/zenodo . 19831 ) .
This work represents the first known application of ML techniques to EVD prognosis prediction . The results suggest that a small set of clinical symptoms and laboratory tests could be sufficient to accurately prognosticate EVD outcome , and that these symptoms and tests should be given particular attention by health care professionals . By aggregating all the high-performing models obtained in our exhaustive analysis , we can construct a composite algorithm that runs the best predictor depending on the available data . We have developed a simple app , Ebola CARE , which can be installed on mobile tablet or phone devices , and would complement rapid EVD diagnostic kits and data-entry devices . Our Ebola CARE app is a proof-of-concept , only applicable to Ebola Zaire patients treated in similar conditions as those in KGH . New clinical data will enable us and other groups to independently validate the app , and to generate more generalizable models with higher statistical significance . Within the current constrains , the results also shed light on the most informative clinical predictors for adult patients -temperature , diarrhea , creatinine , alkaline phosphatase , aspartate aminotransferase , total carbon dioxide- and demonstrate that PCR provides critical additional information to quantify the seriousness of the Ebola virus infection and better estimate the risk of the patients . In general , these results are consistent with the findings from Schieffelin , Levine , Yan , and Zhang . Current discrepancies–for instance Zhang reports chest pain , coma , and confusion as significantly associated with EVD prognosis whereas we do not–could be attributed to the small sample sizes , missing data , and different clinical protocols at the various treatment sites . The prevalence of missing data in the dataset used in this study , and the lack of other publicly available datasets , are fundamental challenges in predictive modeling . By combining MI with four distinct ML predictors , we offer a direct approach for dealing with the first challenge . The use of ANN and LR classifiers in combination with a MI enrichment methodology shows promise as a way to accurately predict outcome of EVD patients given their initial clinical symptoms and laboratory results . New patient data is critical to validate and extend these results and protocols . Richer datasets incorporating more diverse samples from different locations will allow us and other researchers to train better ML classifiers and to incorporate population variability . The development of survival models could be another very important application of these techniques to assist not only in prognosis upon patient intake but also during treatment , as shown by Zhang . Our current data includes time courses that would be useful in this kind of models , but unfortunately only for a handful of patients . All these facts highlight the importance of immediate availability of clinical data in the context of epidemic outbreaks , so that accurate predictive tools can be quickly adopted in the field . In summary , we have made our protocol and mobile app publicly available , fully documented ( https://github . com/broadinstitute/ebola-predictor/wiki ) , and readily adaptable to facilitate and encourage open data sharing and further development . Our integration of Mirador , a tool for visual exploratory analysis of complex datasets , and an ML pipeline defines a complete framework for data-driven analysis of clinical records , which could enable researchers to quickly identify associations and build predictive models . Our app is similarly designed to be easily updated as new predictive models are developed with our pipeline , validated with better data , and packaged , to generate actionable diagnosis and help inform urgent clinical care in outbreak response . | We introduce a machine-learning framework and field-deployable app to predict outcome of Ebola patients from their initial clinical symptoms . Recent work from other authors also points out to the clinical factors that can be used to better understand patient prognosis , but there is currently no predictive model that can be deployed in the field to assist health care workers . Mobile apps for clinical diagnosis and prognosis allow using more complex models than the scoring protocols that have been traditionally favored by clinicians , such as Apgar and MTS . Furthermore , the WHO Ebola Interim Assessment Panel has recently concluded that innovative tools for data collection , reporting , and monitoring are needed for better response in future outbreaks . However , incomplete clinical data will continue to be a serious problem until more robust and standardized data collection systems are in place . Our app demonstrates how systematic data collection could lead to actionable knowledge , which in turn would trigger more and better collection , further improving the prognosis models and the app , essentially creating a virtuous cycle . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"clinical",
"laboratory",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"microbiology",
"neuroscience",
"artificial",
"neural",
"networks",
"viruses",
"diarrhea",
"filoviruses",
"mathematics",
"forecasting",
"statistics",
"(mathematics)",
"signs",
"and",
"symptoms",
"artificial",
"intelligence",
"gastroenterology",
"and",
"hepatology",
"computational",
"neuroscience",
"rna",
"viruses",
"viral",
"load",
"research",
"and",
"analysis",
"methods",
"cardiology",
"computer",
"and",
"information",
"sciences",
"medical",
"microbiology",
"clinical",
"laboratories",
"mathematical",
"and",
"statistical",
"techniques",
"microbial",
"pathogens",
"prognosis",
"heart",
"rate",
"diagnostic",
"medicine",
"ebola",
"virus",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"cognitive",
"science",
"statistical",
"methods",
"hemorrhagic",
"fever",
"viruses",
"organisms"
] | 2016 | Transforming Clinical Data into Actionable Prognosis Models: Machine-Learning Framework and Field-Deployable App to Predict Outcome of Ebola Patients |
Many bacteria have complex cell shapes , but the mechanisms producing their distinctive morphologies are still poorly understood . Caulobacter crescentus , for instance , exhibits a stalk-like extension that carries an adhesive holdfast mediating surface attachment . This structure forms through zonal peptidoglycan biosynthesis at the old cell pole and elongates extensively under phosphate-limiting conditions . We analyzed the composition of cell body and stalk peptidoglycan and identified significant differences in the nature and proportion of peptide crosslinks , indicating that the stalk represents a distinct subcellular domain with specific mechanical properties . To identify factors that participate in stalk formation , we systematically inactivated and localized predicted components of the cell wall biosynthetic machinery of C . crescentus . Our results show that the biosynthesis of stalk peptidoglycan involves a dedicated peptidoglycan biosynthetic complex that combines specific components of the divisome and elongasome , suggesting that the repurposing of preexisting machinery provides a straightforward means to evolve new morphological traits .
The shape of most bacteria is determined by a cell wall made of peptidoglycan ( PG ) , a mesh-like heteropolymer that surrounds the cytoplasmic membrane and provides resistance against the internal osmotic pressure [1 , 2] . The backbone of PG is formed by strands of alternating N-acetylglucosamine ( GlcNAc ) and N-acetylmuramic acid ( MurNAc ) subunits . These glycan chains are connected by short peptides that are attached to the MurNAc moieties , giving rise to a single elastic macromolecule known as the PG sacculus [3] . The PG meshwork needs to be continuously remodeled to allow for cell growth and division [4] . In Gram-negative bacteria , this task is achieved by a large and seemingly redundant set of PG synthesizing and degrading enzymes . Insertion of new cell wall material is initiated by the translocation of lipid-linked GlcNAc-MurNAc-pentapeptide precursors across the cytoplasmic membrane to the periplasm [5–7] . Glycosyltransferases ( GTases ) then incorporate the disaccharide units into preexisting glycan strands , while the L-Ala–D-Glu–L-Lys/meso-DAP–D-Ala–D-Ala pentapeptides of adjacent glycan strands are crosslinked by transpeptidases ( TPases ) [2 , 8] . Depending on their domain structure , PG synthases can be classified as bifunctional GTases/TPases ( class A PBPs ) , monofunctional TPases ( class B PBPs ) and monofunctional GTases [8] . The majority of TPases are DD-TPases , also known as penicillin-binding proteins ( PBPs ) [9] . These proteins catalyze the formation of D-Ala4–meso-DAP3 ( 4-3 ) crosslinks , in a reaction that releases the D-Ala5 moiety of the donor molecule [10] . Alternatively , crosslinks can also be formed between two meso-DAP3 residues ( 3–3 crosslinks ) , catalyzed by specific LD-TPases that use tetrapeptide side chains as donor moieties and release their terminal D-Ala4 residue to gain energy for the crosslinking reaction [11] . For PG to grow , cells require not only synthetic but also lytic enzymes that cleave bonds in the PG meshwork and thus make space for the insertion of new material [12] . Depending on their cleavage specificity , these so-called autolysins can be typically sorted into three main categories . Lytic transglycosylases act on the glycan strands and cleave the β-1 , 4-glycosidic bond between MurNAc and GlcNAc , leaving 1 , 6-anhydro-MurNAc as the terminal residue [13] . Amidases , by contrast , hydrolyze the amide bond between the peptide and the MurNAc moiety [12] , whereas endo- and carboxypeptidases hydrolyze specific amide bonds within the peptides [8 , 14] . The formation and degradation of PG need to be closely coordinated to prevent cell lysis [2 , 15] , a task that is presumably achieved by the assembly of synthetic and lytic enzymes into dynamic multi-protein complexes [16] . In the majority of rod-shaped bacteria , two of these complexes have been identified to date . The first one , called the elongasome ( or Rod complex ) , mediates the dispersed incorporation of new PG along the lateral walls of the cell during the elongation phase . Its positioning is controlled by the actin-like protein MreB [17–19] , which forms patch- or arc-like filaments that are attached to the inner face of the cytoplasmic membrane [20–24] . These structures move around the circumference of the cell and , thus , ensure even growth of the rod-shaped sacculus . In Gram-negative bacteria , their effect on the PG biosynthetic machinery is mediated by the transmembrane protein RodZ [25–27] , which links MreB to a periplasmic complex containing the elongation-specific monofunctional TPase PBP2 [2 , 28 , 29] . Towards the end of the elongation phase , PG synthesis is taken over by a second complex , called the divisome [2 , 30] , which mediates pre-septal elongation and subsequent constriction of the PG sacculus at midcell . Its positioning and activity are regulated by FtsZ , a tubulin homolog that assembles into a dynamic ring-like structure at the future division site . This so-called Z-ring then recruits , directly or indirectly , all other components of the cell division machinery . The divisome includes a variety of PG synthases and hydrolases , among them the division-specific monofunctional TPase PBP3 [31] , which act together to coordinately remodel the PG layer during the division process . Of note , in some species , MreB relocalizes to the division site before the onset of cell constriction but then moves back to the lateral walls as cytokinesis progresses , suggesting that the elongasome and divisome cooperate during certain stages of the division cycle [32 , 33] . While the function of the elongasome and divisome and their roles in establishment of generic rod and coccoid morphologies have been studied intensively [2 , 34] , the mechanisms generating more complex cell shapes are still poorly understood . A model organism known for its distinctive morphological features is the alphaproteobacterium Caulobacter crescentus ( henceforth Caulobacter ) [35] . This species is characterized by a biphasic life cycle that involves two morphologically and physiologically distinct cell types . One of them , the swarmer cell , possesses a single polar flagellum mediating swimming motility . The stalked cell , by contrast , displays a tubular extension ( stalk ) whose tip carries an adhesive holdfast mediating surface attachment . Whereas the stalked cell undergoes repeated cycles of chromosome replication and cell division , the swarmer cell is arrested in G1 phase , searching its environment for nutrients . However , at a defined point in the cell cycle , it sheds its flagellum , starts to establish a stalk at the previously flagellated pole , and enters S phase . The cell then elongates , forms a new flagellum at the pole opposite the stalk , and finally divides asymmetrically to produce a stalked cell and a new swarmer cell [36] . The biological role of the Caulobacter stalk is still controversial , but it may serve as a spacer to elevate the cell above the substratum and thus enhance its access to nutrients [37] . Consistent with this idea , its length increases up to 20-fold under conditions of phosphate limitation [38] . In Caulobacter species , the stalk consists almost exclusively of the three cell envelope layers ( inner membrane , cell wall and outer membrane ) and does not contain any cytoplasm [35 , 39] . Moreover , it is compartmentalized by large disc-like protein complexes , so-called crossbands , which are deposited at irregular intervals along its length , serving as non-selective diffusion barriers that physiologically separate the stalk envelope from the cell body [35 , 40] . Formation of the stalk is driven by zonal incorporation of new cell wall material at the stalk base , as detected by the labeling of newly synthesized PG with tritiated glucose [38] , radiolabeled D-cysteine [41] , or fluorescently labeled D-alanine derivatives [42] . To date , various mutants have been identified that lack stalks under standard growth conditions [43–46] . However , in all cases , cells regained the ability to form stalks after transfer into phosphate-limited media , indicating that they suffered from a block in the cell cycle-regulated initiation of stalk formation rather than a defect in the underlying biosynthetic machinery . Depletion of MreB or the elongasome-specific GTase RodA [47] , by contrast , impaired stalk elongation under all growth conditions [48] . Similar results were obtained after inhibition of the elongasome-specific TPase PBP2 [49] with the β-lactam antibiotic mecillinam . However , because of the severe global cell shape defects observed in these cases , it was difficult to conclude on a specific role of MreB , RodA and PBP2 in the stalk biosynthetic pathway . Finally , a moderate reduction in stalk length was observed for mutants lacking the cytoskeletal protein bactofilin A ( BacA ) or the BacA-associated class A PBP PbpC [50] . Together , these results suggest that components of the generic PG biosynthetic apparatus may be critical for stalk formation , but the precise composition of the machinery responsible for this process still remains elusive . In the present study , we comprehensively investigate the mechanism of stalk formation , focusing on phosphate-limiting conditions to obtain a sensitive readout of the contributions that individual factors make to this process . We show that phosphate starvation induces a G0-like resting state that is characterized by the absence of key cell cycle regulators , including FtsZ . Comparing the muropeptide profiles of isolated stalk and cell body PG , we then identify significant differences in the composition of cell walls from these two compartments , suggesting that stalks are formed by specialized machinery with distinct biosynthetic properties . Systematic deletion and localization studies of cytoskeletal and PG biosynthetic proteins then indeed reveal a distinct set of factors involved in stalk elongation , which we characterize in detail with respect to their impact on PG composition and the spatial regulation of PG biosynthesis . Morphometric analysis of the corresponding mutants shows that these factors make varying and , in part specific , contributions to stalk and cell body elongation , indicating that these two modes of growth a mechanistically distinct . Finally , we identify MreB as a key component of the stalk biosynthetic complex and pinpoint a region on its surface that appears to be required for stalk formation but largely dispensable for elongasome-mediated lateral growth . Collectively , our results show that stalk formation represents a specialized growth process that is mediated by a composite complex including components of both the elongasome and divisome , with distinctive properties that clearly differentiate it from other PG biosynthetic machineries .
Although the stimulatory effect of phosphate starvation on Caulobacter stalk elongation has been known for decades [38] , the underlying regulatory mechanisms are still poorly understood . Prompted by the fact that stalk formation is tightly linked to cell cycle progression , we set out to investigate the effects of phosphate deprivation on central cellular processes such as DNA replication and cell division . First , flow cytometry was used to assess the replicational state of cells after transfer from standard to phosphate-free ( M2G-P ) medium . To this end , replication initiation was blocked with rifampicin and ongoing rounds of replication were allowed to finish . Previous work has shown that Caulobacter cells contain a single chromosome that is replicated only once per division cycle [51 , 52] . Consistent with this finding , we observed that cells accumulated either one or two chromosome equivalents when grown in standard conditions , indicating that a large fraction of the population was in S-phase ( Fig 1A ) . However , upon phosphate deprivation , DNA replication gradually ceased , with most cells arrested in G1 phase after 24 h of incubation . These data suggest that the lack of phosphate leads to a block in the cell cycle prior to S-phase , thereby preventing new rounds of chromosome replication . To support this conclusion , we visualized the number and positions of the chromosomal replication origins . In doing so , we made use of a fluorescently ( GFP- ) tagged derivative of the chromosome partitioning protein ParB , which interacts with specific motifs ( parS ) in the origin region [53 , 54] . The expression of GFP-ParB thus typically results in the detection of either one or two foci , depending on the number of origin copies in the cell . Microscopic analysis revealed that most ( ~ 85% ) cells exhibited a single ParB focus at the stalked pole when subjected to 24 h of phosphate starvation , indicating that they are arrested in G1 phase ( Fig 1B ) . To clarify the reason for this G1 arrest , we analyzed the cellular levels of the replication initiator protein DnaA and the cell cycle master regulator CtrA , which act as positive and negative regulators of chromosome replication , respectively [52] . Interestingly , both proteins were rapidly depleted from the cells during phosphate starvation ( Fig 1C ) , indicating that key drivers of the Caulobacter cell cycle are absent under this condition . To correlate changes in cell cycle progression with the growth behavior of cells , we monitored changes in cell mass and number after a shift to phosphate-limiting conditions . Interestingly , the optical density of cultures kept increasing exponentially for more than 10 h and only leveled off after ~ 50 h of incubation ( Fig 1D ) , suggesting that cells made use of internal phosphate storage compounds to compensate for the lack of an external phosphate source . Consistent with the detection of DNA replication events ( Fig 1A ) , cells still multiplied during the initial exponential phase . However , after longer starvation periods ( > 24 h ) , the viable-cell count started to decline , whereas the cell mass still increased , likely due to continued elongation of the cell bodies and stalks in the absence of cell division events ( see Fig 2B ) . Western blot analysis indeed revealed that the essential cell division protein FtsZ was depleted from the cells upon phosphate starvation ( Fig 1C ) . The same was true for the cell division regulator MipZ , an inhibitor of FtsZ polymerization that limits Z-ring formation to the midcell region [54] . In line with these findings , an FtsZ-YFP fusion induced after prolonged phosphate starvation formed multiple foci in the vicinity of the stalk-distal pole instead of a defined midcell band ( Fig 1B ) , indicating the absence of a functional and properly localized Z-ring [54] . Notably , FtsZ was never observed at the stalk base , supporting the previous notion that it does not play any role in stalk formation [54] . Taken together , our results demonstrate that phosphate starvation arrests the Caulobacter cell cycle in a G1-like phase , thereby stalling DNA replication and cell division until phosphate becomes available again . Phosphate starvation induces Caulobacter to enter a non-replicative state in which cells continue to elongate their cell body and stalk . To investigate this atypical mode of growth , we set out to visualize sites of active PG biosynthesis using the fluorescent D-amino acid 7-hydroxy-coumarin-amino-D-alanine ( HADA ) [42 , 55] as a tracer . As a control , we initially analyzed the growth dynamics of cells growing in phosphate-replete medium . To this end , cells were synchronized and then pulse-labeled with HADA at different stages of the cell cycle . Consistent with previous results [41] , we observed disperse incorporation of new cell wall material before the onset of cell division , followed by zonal growth at midcell during the constriction phase ( S1 Fig ) . Moreover , concurrent with the switch from disperse to zonal growth , an additional intense focus of fluorescence appeared at one of the cell poles , reflecting the establishment and outgrowth of the stalk . This polar signal faded gradually as the cell cycle progressed and was no longer detectable in late pre-divisional cells . Thus , HADA reliably detected all known growth zones in Caulobacter cells . Next , we used HADA labeling to determine the pattern of PG synthesis under phosphate-limiting conditions ( Fig 2A ) . After 6 h of incubation in phosphate-free medium , most cells showed a bright fluorescent patch at the stalked pole as well as a faint disperse signal extending throughout the rest of the cell body . Cells longer than ~ 4 μm often displayed an additional bright focus at their center , which could reflect FtsZ-dependent zonal growth or cell division , consistent with the observation that the viable-cell counts still increased in the early phase of starvation ( Figs 1D and S2A ) . Interestingly , the intensity of the polar signal decreased considerably upon appearance of a midcell focus , suggesting that the machineries mediating stalk formation and cell division may compete with each other for at least some of their components ( Fig 2A ) . After longer starvation periods ( >18 h ) , midcell foci were almost undetectable , and HADA fluorescence was largely limited to the stalk base , which correlates with the lack of cell division events at this time point . Notably , the intensity of the polar signal decreased slightly during long-term incubation ( S2B Fig ) , although the rate of stalk elongation remained constant at all time points ( Fig 2B ) . The increase in cell body length , by contrast , was most pronounced during the early phases of starvation , when cells still showed midcell HADA foci , suggesting that it may , at least in part result , from FtsZ-mediated zonal growth at the cell center . Collectively , phosphate starvation induces a switch in the pattern of PG synthesis that ultimately limits cell growth to the stalked cell pole . Our and previous labeling studies suggest that stalk formation is driven by the insertion of new cell wall material at the stalk base [38 , 41 , 42] . To determine whether stalk PG is still subject to modification or turnover , phosphate-starved Caulobacter cells were incubated with HADA for an extended period of time ( 1 . 5 h ) . After this treatment , staining was observed throughout the entire cell envelope ( Fig 2C ) , including distal segments of the stalk that were clearly formed prior to the start of the labeling procedure ( considering a stalk elongation rate of 0 . 28 ± 0 . 03 μm/h; see Fig 2B ) . Given that HADA is likely incorporated by the action of periplasmic DD- or LD-TPases [42 , 56 , 57] , this finding indicates the presence of transpeptidase activity in the stalk compartment that mediated the addition of HADA to preexisting PG independently of the pole-associated biosynthetic complex . After transfer of the cells to HADA-free medium , fluorescence was rapidly lost in the cell bodies and in the basal region of the stalk , whereas it was stably retained in the distal stalk segments , indicating that stalk PG is not turned over at significant rates ( Fig 2C and 2D ) . Notably , the same behavior was observed for a strain lacking crossbands . The differences in the behavior of cell bodies and stalks may thus not result from restrictions in the diffusion of envelope-localized PG biosynthetic enzymes but rather from the retention of these enzymes at the stalk base , possibly due to the lack of cytoplasm in the stalk compartment . The distinct mode of growth involved in stalk formation opens the possibility that there may be compositional differences between the PG layers encompassing the cell body and stalk compartments . To address this issue , phosphate-starved cells were agitated vigorously to shear off stalks from the cell bodies . After separation of the two compartments by differential centrifugation ( S3 Fig ) , PG was isolated from each of the fractions and subjected to muropeptide analysis . Interestingly , stalk PG contained a high proportion of 3–3 crosslinked peptides and non-crosslinked tripeptides ( resulting from the cleavage of 3–3 bonds ) , whereas these muropeptide species were barely detectable in the cell body samples ( Fig 3 and S1 Table ) . Similarly , the total fraction of crosslinked peptide side chains was significantly higher in stalk PG , mostly because of a higher proportion of trimeric muropeptides . The glycan chain lengths , by contrast , did not vary between the two compartments . Collectively , these findings indicate that the PG layers of stalks and cell bodies differ in both the type and extent of peptide crosslinks . Previous work has shown that 3–3 crosslinks are generated by LD-TPases , which are characterized by a conserved YkuD domain [11] . The Caulobacter genome contains two so-far uncharacterized open reading frames , CC_1511 and CC_3744 , which encode proteins with this signature domain ( now referred to as LdtD and LdtX , respectively ) . To determine how these factors contribute to the distinctive composition of the stalk cell wall , we generated a strain carrying in-frame deletions in both the ldtD and ldtX gene and analyzed the composition of PG purified from its stalk and cell body compartments . In both samples , 3–3 crosslinked peptides and non-crosslinked tripeptides were virtually undetectable ( Fig 3 and S1 Table ) , indicating that the formation of these muropeptide species is linked to the activity of the two predicted LD-TPases . Notably , however , the total fraction of crosslinked peptides barely changed in either of the compartments , because the loss of 3–3 crosslinks was compensated by a proportional increase in the fraction of 4–3 crosslinks . Thus , LD-TPase activity is not the main factor responsible for the elevated degree of crosslinking detected in stalk PG ( see Fig 4 for a summary of the key results obtained in this study ) . The stalk is physiologically separated from the cell body , because it is devoid of cytoplasm and contains crossband complexes that block the exchange of periplasmic and membrane proteins [40] . It was conceivable that crossbands could help establish the differences in the PG composition observed for the two compartments , for instance by facilitating the establishment of distinct pools of PG biosynthetic enzymes or blocking the diffusion of lipid II into the stalk structure . To test this idea we determined the muropeptide profile of stalk and cell body PG isolated from a crossband-less strain ( ΔstpAB , SW51 ) . Notably , we still observed a higher content of 3–3 crosslinks and a higher total proportion of crosslinked peptides in stalk PG ( S1 Table ) . Similar to the differences in PG turnover ( Fig 2C ) , this characteristic thus appears to be independent of the presence of crossbands . Stalk formation involves a growth process that is distinct from the disperse and zonal incorporation of PG mediated by the elongasome or division complex , respectively . To determine the composition of the underlying machinery , we systematically analyzed all predicted PG biosynthetic proteins encoded in the Caulobacter genome for their contribution to stalk elongation under phosphate-limiting conditions ( see Fig 4 for a summary ) . In doing so , we initially focused on enzymes with PG synthase activity , including RodA , mono- and bifunctional PBPs and LD-TPases . Previous studies have shown that the depletion of RodA or the inhibition of the monofunctional DD-TPase PBP2 with mecillinam largely abolished the synthesis of stalks in phosphate-replete media , although these treatments concomitantly induced severe morphological defects in the cell body [48 , 49] . To verify these results , we determined the consequences of RodA depletion on stalk elongation during phosphate limitation ( S4 Fig ) . The absence of RodA indeed led to a severe decrease in both stalk and cell body length . Similar tendencies were observed for mecillinam-treated cells , although the effects were generally less pronounced . These results indicate a global morphogenetic function of RodA and PBP2 in phosphate-starved cells . To further investigate these proteins , we generated functional fluorescent protein fusions [58] and analyzed their localization in phosphate-limited media . Consistent with a global role in PG biosynthesis , both GFP-RodA ( S9 Fig ) and GFP-PBP2 formed large patches that were distributed throughout the cell bodies . Moreover , both proteins frequently formed a faint but distinct focus at the stalked pole , particularly in cells with clearly elongated ( > 4 μm ) bodies ( Fig 5A ) . Thus , RodA and PBP2 appear to be partly associated with the stalked pole , where they likely cooperate to mediate stalk formation . Previous work has also implicated bifunctional PBPs in stalk elongation , largely based on the analysis of mutants lacking one of these proteins [50 , 59] . To verify and extend these results , we analyzed strains carrying single or multiple mutations in the PBP-encoding pbpY , pbp1A , pbpC , pbpX , and pbpZ genes [59 , 60] . Our results confirm that deletion of pbpC led to a moderate reduction in stalk length , whereas the absence of any other PBP , either alone or in combination , did not have any effect ( Fig 5B ) . However , as observed under standard growth conditions [59 , 60] , at least one bifunctional PBP was required for viability during phosphate starvation ( S5A Fig ) . In line with the results of the deletion studies , localization analyses revealed that none of the bifunctional PBPs except for PbpC accumulated at the stalked pole , indicating that these proteins may not be specifically associated with the stalk biosynthetic machinery ( S5B Fig ) . Notably , however , PbpX appeared enriched in the stalk compartments , but the significance of this observation remains unclear . Finally , we analyzed the role of the two predicted LD-TPases LdtD and LdtX in stalk formation . Although these proteins make a significant contribution to PG crosslinking in the stalk compartment ( Fig 3 ) , their inactivation did not have any apparent phenotypic effect ( Fig 5B ) . LD-TPase activity may thus not contribute to the establishment of the stalk structure per se but rather have an accessory function that serves to modify the biophysical properties of the PG layer . Localization studies indicate that LdtD and LdtX do not accumulate at the stalk base , suggesting that they may act independently of the polar stalk biosynthetic machinery ( S5C Fig ) . Apart from PG synthases , stalk formation must also involve autolytic enzymes that cleave the PG sacculus and , thus , enable the insertion of new cell wall material at the stalk base . However , to this point , the nature of the factors involved has remained unknown . To address this issue , we systematically screened mutants lacking one or multiple predicted PG hydrolases for defects in stalk growth under phosphate-limiting conditions . The enzymes tested included all LytM-like and NlpC/P60-like endopeptidases , AmiC-like and CHAP domain-containing amidases , soluble and membrane-bound lytic transglycosylases , and carboxypeptidases identified in the Caulobacter genome ( S2 Table ) . In most cases , the lack of single factors and even the absence of whole enzyme families had no apparent effect on stalk length ( S6 Fig ) . Four strains , however , displayed obvious morphological defects ( Fig 6; see S7 Fig for the phenotypes in PYE medium ) . One of them was a mutant lacking the protein DipM , a catalytically inactive LytM-like endopeptidase homolog that was previously shown to be critical for proper PG remodeling during cell division [61–63] . The absence of DipM led to a severe reduction in stalk length , combined with the formation of branches within the stalk structure or the establishment of multiple stalks , often emanating from the same pole ( Figs 6B , 6C and S8 ) . Even shorter stalks were observed in the combined absence of the soluble lytic transglycosylases SdpA and SdpB ( Fig 6B and 6C ) , two proteins previously found to be associated with the divisome complex [64] . Apart from its aberrant morphology , the ΔsdpAB mutant frequently showed membrane blebs that were associated with the residual stalk structures , suggesting a defect in membrane attachment or homeostasis ( S8 Fig ) . Milder effects on stalk length were caused by inactivation of the divisome-associated carboxypeptidase CrbA ( Billini et al , unpublished ) or the LytM-like endopeptidase LdpA , a thus-far uncharacterized protein encoded in an operon with the polarly localized scaffolding protein bactofilin A ( BacA ) [50 , 65] ( Figs 6B , 6C and S8 ) . Importantly , despite their stalk elongation defects , none of the four deletion strains showed a significant reduction in cell length ( Fig 6C ) , indicating that stalk and cell body growth are mechanistically distinct processes that proceed independently of each other . To further investigate the functions of the five autolytic factors identified in the mutational screen , we generated fluorescently ( mCherry- ) tagged derivatives of these proteins and analyzed their localization patterns under conditions of phosphate starvation ( Fig 7 ) . Both the DipM and CrbA fusions accumulated at the stalk base and may , thus , be specifically associated with the polar stalk biosynthetic machinery . The SdpA and SdpB fusions , by contrast , were distributed throughout the cell envelope , suggesting that the two proteins may either act independently of the polar complex or associate with it in a very transient manner . Unlike the other proteins analyzed ( S9 Fig ) , LdpA-mCherry was quantitatively cleaved at the junction between the two fusion partners , preventing further analysis . In order to determine how the absence of the different autolytic factors influenced the pattern of PG biosynthesis , mutants lacking these proteins were grown in phosphate-limiting conditions and subjected to HADA staining ( Fig 8 ) . Consistent with their relatively mild stalk elongation defect , ΔldpA cells still displayed a pattern similar to that of the wild-type strain . In the ΔdipM and ΔcrbA strains , by contrast , the polar signals were much fainter and new cell wall material was often incorporated at non-polar sites . An even more pronounced effect was observed in the ΔsdpAB mutant , which virtually lacked polar foci and instead showed patchy or even HADA fluorescence throughout the cells . Thus , the severity of the stalk elongation defect scales with the loss in polar PG biosynthesis . To obtain more detailed insight into the effects of the different mutations on the structure of the PG layer , we isolated whole-cell sacculi from wild-type and mutant cells after prolonged ( 24 h ) phosphate starvation and subjected them to muropeptide analysis ( S3 Table ) . For the wild-type strain , whole-cell sacculi gave similar results as PG from isolated from a cell body fraction ( compare Fig 3 and S1 Table ) , indicating that the characteristic features of stalk PG are largely obscured by the excess of cell body PG in the whole-cell preparations . Interestingly , there were hardly any differences between the muropeptide profiles obtained under phosphate-limiting ( S3 Table ) and phosphate-replete [66] conditions . The average composition of cell body PG thus appears to be independent of the phosphate supply . Among the mutant strains , ΔldpA cells showed essentially the same average PG composition as the wild-type strain . The same was true for the ΔdipM mutant , with exception of a significant increase in the proportion of non-crosslinked tetrapeptides ( S3 Table ) , which could indicate an elevated level of endopeptidase and/or carboxypeptidase activity . The muropeptide profiles of the remaining strains , by contrast , showed marked global changes . In line with the notion that CrbA acts as a carboxypeptidase , removing the terminal D-Ala residue of pentapeptide side chains ( Billini et al , unpublished ) , the ΔcrbA mutant displayed a considerable decrease in the total content of tetrapeptides that was accompanied by a proportional increase in the content of pentapeptide-containing muropeptide species ( S3 Table ) . In ΔsdpAB cells , on the other hand , the average glycan chain length increased from 7 to 9 . 4 disaccharide units , consistent with the loss of lytic transglycosylase activity . Surprisingly , the mutant cells additionally showed a severe reduction in the degree of crosslinkage . At the same time , their total content of pentapeptide side chains was reduced , whereas the proportion of tripeptide side chains was considerably elevated ( S3 Table ) . These results suggest that the lack of SdpAB leads to reduced transpeptidation or , more likely , elevated endopeptidase activity . Collectively , our results show that several components of the autolytic machinery are critical for proper PG remodeling during stalk formation , with some of them localizing to the stalk base under phosphate-limiting conditions ( see Fig 4 for a summary ) . Notably , most of the proteins , including DipM , SdpA , SdpB and CrbA , are associated with the cell division apparatus under standard growth conditions [61–64] ( M . Billini , unpublished ) , suggesting parallels in the mechanisms that reshape the PG layer during cell constriction and stalk growth . Polymer-forming scaffolding proteins are critical for the regulation of many growth processes in bacteria [4 , 67] , suggesting that this group of proteins may also play a critical role in stalk formation . Previous work has indeed implicated the bactofilin homolog BacA in stalk biogenesis [50] . Re-analysis of a ΔbacA mutant revealed a significant reduction in both stalk and cell body length during phosphate starvation ( Figs 9 and S10 ) . Notably , deletion of the endopeptidase gene ldpA , which lies in a putative operon with bacA , had a very similar effect on stalk length , whereas it barely affected the cell body ( Fig 6 ) . These results suggest that LdpA and BacA may specifically cooperate in stalk formation , whereas BacA is additionally involved in a distinct pathway involved in cell body elongation . As another scaffolding protein , MreB was shown to be required for stalk formation in media containing moderate to high levels of phosphate [48 , 68] . To clarify the contribution of this protein to stalk biosynthesis under phosphate starvation , we employed strains producing MreB or the adapter protein RodZ under the control of an inducible promoter . When starved for phosphate in the absence of inducer , both mutants showed a drastic reduction in stalk length or occasionally even failed to form stalks at all ( Figs 9 and S10 ) . The effects on the cell bodies , by contrast , differed depending on the protein depleted . Cells lacking RodZ showed a length distribution indistinguishable from that of the wild-type strain . Depletion of MreB , by contrast , markedly decreased the median cell length . Similar effects were observed for wild-type cells treated with the MreB inhibitor A22 [69 , 70] ( Fig 9 ) . These findings indicate that , under phosphate-limiting conditions , RodZ appears to be specifically required for stalk biosynthesis , whereas MreB additionally contributes to cell body elongation , again supporting the idea that these two processes are driven by distinct mechanisms . Apart from MreB , the MreCD complex has been identified as a factor critical to lateral growth in many rod-shaped bacteria [4] . MreC is thought to serve as a scaffold that interacts with various PG biosynthetic enzymes , including the monofunctional TPase PBP2 [71 , 72] and the monofunctional transglycosylase RodA [73] . In E . coli , it is part of the elongasome complex [74] , whereas it was shown to establish an elongasome-independent structure in Caulobacter cells [68 , 75] . To test for a role of this protein in stalk formation , we analyzed the morphology of a conditional mreC mutant grown under phosphate-limiting conditions ( Fig 9 ) . In the absence of inducer , the cells started to elongate but eventually became amorphous and lyzed . In most cases , stalks were either absent or barely recognizable , indicating that the MreCD complex may be essential for both cell wall integrity and stalk biosynthesis during phosphate starvation . Given the role of MreC in the activation of PBP2 and RodA in E . coli , the differences in the phenotypes obtained after inactivation of these proteins are unexpected . Either Caulobacter MreC has functionally diverged from its E . coli homolog or its depletion is more complete than the depletion of MreB or the inhibition of PBP2 with mecillinam . Collectively , our results show that the bactofilin homolog BacA and the elongasome components MreB , RodZ and MreC are required for proper stalk biosynthesis in Caulobacter cells . To clarify whether the role of the different scaffolding proteins in stalk formation involves their recruitment to the stalked pole , we analyzed the localization patterns of fluorescently tagged derivatives in cells subjected to phosphate starvation ( Fig 10 ) . Both the MreB and RodZ fusion formed a distinct focus at the stalk base and , in rare cases , also a second focus at the pole opposite the stalk . Together with the polar localization of PBP2 ( Fig 5A ) , these findings indicate that key components of the elongasome complex relocate to the site of stalk biosynthesis in phosphate-limiting conditions . There , they colocalize with BacA , which retains its polar position irrespective of changes in the phosphate supply ( Fig 10 ) . The MreC fusion , by contrast , formed a broad band at midcell , whereas it was largely excluded from the polar regions ( Fig 10 ) . In line with the global morphological defects caused by its depletion , MreC may have a general role in cell wall biosynthesis , but it does not appear to be stably associated with the polar stalk biosynthetic machinery . To determine the role of the different scaffolds in polar PG biosynthesis , cells lacking these factors were subjected to HADA staining after phosphate deprivation ( Fig 11 ) . Interestingly , despite its severe stalk elongation defect ( Fig 9 ) the ΔbacA mutant still displayed intense polar foci , indicating that BacA is an accessory factor that is not critical for the global reorganization of PG biosynthesis induced under phosphate-limiting conditions . Consistent with this idea , muropeptide analysis showed that deletion of bacA did not have any appreciable effects on global PG composition ( S4 Table ) . Depletion of MreB or RodZ , by contrast , strongly decreased the intensity of the polar HADA signals , and frequently led to the insertion of cell wall material at pole-distal sites . In both cases , these defects were accompanied by significant changes in the whole-cell muropeptide profiles . Similar to the ΔsdpAB mutant ( compare S3 Table ) , the degree of crosslinkage was significantly reduced , mostly due to a decrease in the proportion of highly crosslinked ( trimeric and tetrameric ) muropeptide species . Moreover , there was a striking increase in the proportion of muropeptides with tripeptide side chains , indicative of high levels of LD-TPase activity . Thus , cell wall stress caused by reduced levels of PBP2-mediated DD-transpeptidation may trigger a fail-safe mechanism that stabilizes the PG meshwork through the formation of abundant 3–3 crosslinks . Collectively , these results demonstrate that MreB and its transmembrane adapter RodZ play a central role in the establishment of the polar PG biosynthetic zone that gives rise to the stalk structure ( see Fig 4 for a summary ) . Our data demonstrate that several components of the PG biosynthetic machinery localize to the stalked pole in phosphate-starved cells , suggesting that they assemble into a complex mediating the synthesis of stalk PG . To obtain more insight into the factors mediating the recruitment of these proteins , we reanalyzed the localization patterns of DipM-mCherry , CrbA-mCherry , Venus-MreB , CFP-RodZ , and BacA-CFP in all deletion strains that showed defects in stalk elongation ( ΔdipM , ΔsdpAB , ΔcrbA , ΔldpA , and ΔbacA ) . However , in all cases , the positioning of the fusion proteins remained unaffected , indicating that neither lytic factors nor the bactofilin cytoskeleton are required for complex assembly . Given the prevalence of elongasome components among the polarly localized proteins , we then tested the role of MreB in the recruitment process . Treatment of cells with the MreB inhibitor A22 not only led to the delocalization of the known MreB interactor RodZ but also abolished the polar foci of DipM and CrbA ( Fig 12 ) . Thus , MreB appears to be a key organizer of the stalk biosynthetic complex . Notably , A22 had no effect on the polar localization of BacA , indicating that the bactofilin scaffold acts independently of MreB . To analyze the dynamics of the polar MreB assembly , we constructed a sandwich fusion in which mCherry was inserted into a surface-exposed loop of the MreB protein , following the strategy previously used for its E . coli homolog [26] ( Fig 13A and 13B ) . A strain carrying the respective allele ( mreBsw ) in place of the endogenous mreB gene showed normal growth rates ( S11A Fig ) . However , in rich medium , the sandwich fusion failed to fully relocate to midcell during the early phases of cell division . Moreover , cells were slightly longer and more highly curved than the wild type and occasionally formed branches and/or filaments ( Figs 13C and S11B ) . Under phosphate-limiting conditions , by contrast , the distribution of cell lengths was similar to that of the wild-type strain ( S11B Fig ) . Strikingly , mreBsw cells failed to form stalks in both standard and phosphate-limited medium . Consistent with this observation , the fusion protein no longer condensed into polar foci during phosphate starvation but retained the patchy localization pattern typically observed in exponentially growing cells [76] ( Fig 13C and 13D ) . This unusual behavior went along with changes in the global muropeptide profile that were qualitatively similar to those observed for MreB- and RodZ-depleted cells but considerably less pronounced ( S4 Table ) . Consistent with the slightly aberrant morphology of the mutant cells , this finding suggests that the insertion of mCherry leads to a mild general defect in MreB function . Importantly , however , it appears to additionally block a specific set of interactions that are critical for the polar recruitment of MreB , thereby preventing stalk formation . To our knowledge this is the first report of a Caulobacter strain that is completely stalkless under all growth conditions . Collectively , these results demonstrate that MreB has a key role in the assembly and function of the polar stalk biosynthetic complex in Caulobacter ( see Fig 4 for a summary ) .
Bacterial cells come in a variety of different shapes , but in most cases the mechanisms generating this morphological diversity are poorly understood [77] . This study uses the Caulobacter stalk as a readily amenable model system to investigate the molecular principles underlying the development of species-specific morphological traits . Previous work has shown that stalk growth is driven by zonal PG incorporation at the old cell pole [38 , 41 , 42] . Initially , this process was thought to be mediated by FtsZ and mechanistically similar to pre-septal cell elongation [68 , 78 , 79] . However , localization studies revealed that FtsZ is not detectable at the stalked pole , neither during normal cell cycle progression [54] nor during phosphate starvation ( Fig 1B ) , excluding the divisome as a relevant player in stalk formation . Other reports implicated MreB and RodZ in stalk growth , suggesting that the elongasome could have a dual role in both cell body and stalk elongation [48 , 68] . Clarification of this issue is complicated by the fact that the inactivation of factors with a global role in PG biosynthesis leads to pleiotropic morphological defects . Exploiting the fact that phosphate starvation suppresses Caulobacter cell division while strongly promoting stalk elongation , we were able to disentangle cell body- and stalk-specific growth processes and specifically identify proteins involved in the synthesis of stalk PG . Our results indicate that stalk biogenesis is driven by a specialized biosynthetic complex whose composition and biosynthetic activities are clearly distinct from those of the generic cell elongation and division machineries . Interestingly , the stalk biosynthetic complex is a hybrid composed of factors typically associated with the elongasome ( MreB , RodZ , RodA , PBP2 ) or divisome ( DipM , SdpA , SdpB , CrbA ) ( Fig 14 ) . The recruitment of components from the cell elongation machinery may reflect the need to incorporate new cell wall material into an existing sacculus to drive the elongation of the stalk structure , a process that may be mechanistically similar to dispersed PG biosynthesis during lateral cell growth . Notably , HADA ( Fig 2A ) and D-Cysteine [41] labeling clearly indicate that , during stalk growth , newly synthesized PG is not primarily detected in the basal stalk segment but rather in the adjacent polar regions of the cell body . This observation suggests that stalk elongation does not occur simply by addition of new material to the existing stalk template . Instead , it appears to be mediated through expansion of the stalk-proximal polar cap and its simultaneous remodeling into a new stalk segment , a process reminiscent of the medial growth and constriction of the PG sacculus during Caulobacter cell division . The common requirement for extensive PG remodeling may explain why the cell division and stalk biosynthetic complexes show a considerable overlap in their autolytic machineries . Interestingly , the importance of some of these shared components varies substantially between the two complexes . For instance , combined inactivation of the lytic transglycosylases SdpA and SdpB has no obvious effect on cell division [64] , whereas it largely abolishes stalk formation , indicating that the functional context of these proteins varies depending on the process they mediate . It remains to be clarified to what extent the different cell wall biosynthetic complexes compete for their shared components . Interestingly , during the Caulobacter cell cycle , stalk growth occurs predominantly within a short time window at the transition from dispersed to medial peptidoglycan biosynthesis ( S1 Fig ) . It is therefore tempting to speculate that the elongasome and divisome have higher priority in the recruitment of shared factors , thereby restricting assembly of the stalk biosynthetic complex to phases in which they are not fully active . Overall , stalk formation clearly demonstrates how the reshuffling of preexisting machinery can serve as a straightforward means to generate novel morphological features in bacteria ( see also [80] ) . The striking diversity of cell shapes observed in certain lineages , such as the alphaproteobacteria , may therefore not be based on major new additions to the repertoire of cell wall biosynthetic proteins but rather on subtle changes in protein activities and localization patterns . A key finding of our work is the central role of MreB in the stalk biosynthetic complex . We show that this cytoskeletal protein condenses at the stalked pole during phosphate starvation and facilitates the polar recruitment of several other factors that are critical to stalk formation . Notably , our attempts to integrate mCherry into a surface-exposed loop of MreB led to the serendipitous identification of a Caulobacter strain that was completely devoid of stalks under both phosphate-limiting and -replete conditions . This is in stark contrast to other mutants described previously , which are stalk-less in rich medium but still elaborate stalks upon phosphate starvation [43–46] , suggesting that they have a defect in the regulation of stalk formation rather than in the biosynthetic machinery mediating this process . Importantly , apart from their failure to form stalks , cells producing the MreB sandwich fusion showed only mild general cell shape defects . The region surrounding the insertion site of mCherry may thus contain determinants that are specifically required for MreB’s function in stalk formation but largely dispensable for elongasome-mediated longitudinal growth of the cell body . Previous work has shown that the positioning of MreB filaments is strongly influenced by their intrinsic curvature [81 , 82] , a parameter controlled by the concentration of the membrane adapter RodZ [83] . The high enrichment of the MreB-RodZ complex at the stalked pole may thus be sufficient to change the architecture of MreB filaments such as to facilitate their interaction with the more highly curved stalked pole . However , the cues promoting the relocation of MreB from the lateral regions of the cell to the stalked pole still remain unknown . Although MreB clearly has a key role in stalk biogenesis , it is not the only scaffolding protein contributing to this process . Previous work has shown that the bactofilin BacA is required for proper stalk length [50] , and our analyses revealed an additional role for this protein in cell body elongation during phosphate starvation ( Fig 9 ) . Notably , the bacA gene lies in a putative operon with ldpA , a gene encoding a putative LytM-like endopeptidase that also functions in stalk formation . This genetic context is conserved in a variety of other species , suggesting a functional link between the two gene products [84 , 85] . Support for this notion comes from studies in the human pathogen Helicobacter pylori , which demonstrated that both genes in this conserved operon are required to establish the characteristic helical cell shape of this species [84] . Notably , apart from its putative interaction with LdpA , Caulobacter BacA was shown to recruit a class A PBP ( PbpC ) involved in stalk elongation and in the targeting of proteins to the stalk lumen [50 , 86] . Importantly , the polar localization of BacA was independent of the presence of MreB . The bactofilin cytoskeleton thus appears to constitute a functionally independent morphogenetic module that has been coopted by Caulobacter to modulate stalk formation . This module appears to act downstream of the MreB-dependent stalk biosynthetic complex , as it was not able to establish a stalk structure in the absence of a functional MreB cytoskeleton . The ultimate determinant mediating the polar recruitment of the stalk biosynthetic machinery in Caulobacter still remains unknown . In Asticcacaulis excentricus , a member of the Caulobacteraceae that is characterized by subpolar stalks , the site of stalk formation was shown to be defined by the polarity determinant SpmX [87] . However , despite its conservation , this protein is not required for proper stalk localization in Caulobacter cells [88] . Notably , deletion of SpmX or transfer of the cells to phosphate-limited media restores polar stalk growth in A . excentricus [87] , suggesting that the pathway observed in Caulobacter is still present in A . excentricus but normally obscured by the the action of the newly coopted localization factor SpmX . It will be interesting to see whether the A . excentricus SpmX homolog organizes an alternative stalk biosynthetic complex or simply recruits the polar machinery to a pole-distal position . Although the functionality and localization of the peptidoglycan biosynthetic machinery changes drastically upon transition of Caulobacter cells from phosphate-replete to phosphate-limiting media , the overall composition of their PG layer remains largely unaffected . This finding is unexpected because significant changes in both glycan chain lengths and the degree of cross-linking were observed in other species in response to changes in their growth conditions [89] . However , analyzing the muropeptide profiles of isolated stalk and cell body fractions , we identified clear differences between these two compartments that are likely obscured in whole-cell analyses due to the small contribution of stalks to the total cellular PG content . Most importantly , stalk PG showed a significantly higher degree of crosslinkage , which was mostly due to a higher frequency of 3–3 crosslinks , indicative of elevated LD-TPase activity . The precise reason for this difference remains unclear . It is conceivable that the LD-TPases LdtD and LdtX are part of the polar stalk biosynthetic complex and , thus , preferentially act on newly synthesized PG produced by this machinery . However , localization studies did not give any evidence for an enrichment of these proteins at the stalked pole . An alternative explanation may be provided by the observation that the turnover rate of PG is significantly lower in the stalk than in the cell body . Thus , LD-TPases may act uniformly throughout the entire cell envelope , but most of the 3–3 crosslinks formed in the cell body may be lost as a consequence of PG remodeling , whereas those in the stalk are retained over prolonged periods of time . Notably , peptides with 3–3 crosslinks are stiffer than those with 3–4 crosslinks and adopt a more extended conformation that is better suited to connect glycan strands in stressed PG [90] . Their increased frequency may therefore help to modulate the mechanical properties of the stalk and render it more resistant to bending or breakage under conditions of high laminar flow [37 , 91] Collectively , our study shows that , in Caulobacter , multiple cell-wall biosynthetic machineries act in concert to generate stalks of proper size and stability , thereby ensuring optimal performance of this cellular structure in the environmental context . It will be interesting to see how the nature and the regulation of these components have changed during evolution to bring about the large variety of morphologies found in other stalked members of the alphaproteobacterial lineage .
Caulobacter strains [92] were grown at 28°C in peptone-yeast-extract ( PYE ) medium [35] , supplemented with antibiotics at the following concentration when appropriate ( μg ml-1; liquid/solid medium ) : spectinomycin ( 25/50 ) , streptomycin ( -/5 ) , gentamicin ( 0 . 5/5 ) , kanamycin ( 5/25 ) , chloramphenicol ( 1/1 ) , mecillinam ( 15/- ) . Gene expression from the xylX promoter ( Pxyl ) or vanA promoter ( Pvan ) , was induced by supplementation of the media with 0 . 3% D-xylose and 0 . 5 mM sodium vanillate , respectively , prior to analysis of the cells . To induce phosphate starvation , stationary cells were diluted 1:20 in M2G-P medium [50] and incubated at 28°C for the indicated times . In case of the conditional mreB , rodZ , and mreC mutants , cells were grown to exponential phase ( OD600 ~ 0 . 5 ) in PYE medium supplemented with xylose , washed three times , and then resuspended to an OD600 of 0 . 05 in inducer-free medium . The cultures were then grown for 7 h to achieve protein depletion , diluted ( 1:20 ) in M2G-P medium , and cultivated for additional 24 h before analysis . The conditional amiC and dipM mutants were treated in a similar fashion , with 12 h of cultivation in PYE medium prior to transfer into M2G-P . The synchronization of Caulobacter was achieved by density gradient centrifugation using Percoll ( Sigma-Aldrich ) [93] . To determine the viable-cell count in cultures , various dilutions of the cell suspensions were spread on PYE plates , and the number of colony-forming units ( CFU ) was determined after three days of incubation at 28°C . E . coli strain TOP10 ( Invitrogen ) and its derivatives were cultivated at 37°C in LB broth ( Karl Roth , Germany ) . Antibiotics were added at the following concentrations ( μg/ml; liquid/solid medium ) : spectinomycin ( 50/100 ) , gentamicin ( 15/20 ) , kanamycin ( 30/50 ) , chloramphenicol ( 20/30 ) . The bacterial strains , plasmids , and oligonucleotides used in this study are listed in S5–S8 Tables . E . coli TOP10 ( Invitrogen ) was used as host for cloning purposes . All plasmids were verified by DNA sequencing . Caulobacter was transformed by electroporation . Non-replicating plasmids were integrated into the Caulobacter chromosome by single-homologous recombination at the xylX ( Pxyl ) or vanA ( Pvan ) locus [94] . Gene replacement was achieved by double-homologous recombination using the counter-selectable sacB marker ( M . R . K . Alley , unpublished ) [54] . Proper chromosomal integration or gene replacement was verified by colony PCR . Cells were grown to exponential phase in PYE medium , harvested by centrifugation , and resuspended in the same medium to an OD600 of 0 . 05 . The suspensions were then transferred to 24‐well polystyrene microtiter plates ( Becton Dickinson Labware ) , incubated at 32°C with double‐orbital shaking in an Epoch 2 microplate reader ( BioTek , Germany ) , and analyzed photometrically ( OD600 ) at 15 min intervals . For light microscopic analysis , cells were transferred onto pads made of 1% agarose . Images were taken with an Axio Observer . Z1 ( Zeiss , Germany ) microscope equipped with a Plan Apochromat 100x/1 . 45 Oil DIC and a Plan Apochromat 100x/1 . 4 Oil Ph3 phase contrast objective , an ET-mCherry filter set ( Chroma , USA ) , and a pco . edge sCMOS camera ( PCO , Germany ) . Images were recorded with VisiView 3 . 3 . 0 . 6 ( Visitron Systems , Germany ) and processed with Metamorph 7 . 7 . 5 ( Universal Imaging Group , USA ) and Illustrator CS6 ( Adobe Systems , USA ) . To generate demographs , fluorescence intensity profiles were measured with ImageJ 1 . 47v ( http://imagej . nih . gov/ij ) . The data were then processed in R version 3 . 5 . 0 [95] using the Cell Profiles script ( http://github . com/ta-cameron/Cell-Profiles ) [96] . Box and violin plots for the statistical analysis of imaging data were generated in R version 3 . 5 . 0 using the ggplot2 [97] and Reshape2 [98] packages , respectively . 10 μl cell suspension were applied to an electron microscopy grid ( Formvar/Carbon Film on 300 Mesh Copper; Plano GmbH , Germany ) and incubated for 1 min at room temperature . Excess liquid was removed with Whatman filter paper . Subsequently , the cells were negatively stained for 5 sec with 5 μl of 1% uracyl acetate . After three washes with H2O , the grids were dried , stored in an appropriate grid holder , and analyzed in a 100 kV JEM-1400 Plus transmission electron microscope ( JEOL , USA ) . Western blot analysis was performed as described [54] , using anti-CtrA [99] , anti-FtsZ [63] , anti-MipZ [54] , anti-DnaA [100] , or anti-SpmX [88] at dilutions of 1:10 , 000 ( anti-CtrA , anti-FtsZ , anti-MipZ , and anti-DnaA ) , and 1:50 , 000 ( anti-SpmX ) . Goat anti-rabbit immunoglobulin G conjugated to horseradish peroxidase ( Perkin Elmer , USA ) was used as secondary antibody . Immunocomplexes were detected using the Western Lightning Plus-ECL chemiluminescence reagent ( Perkin Elmer , USA ) . Signals were recorded with a ChemiDoc MP imaging system ( Bio-Rad ) and analyzed using the Image Lab 5 . 0 software ( Bio-Rad ) . HADA-staining experiments were conducted as described [42] . Briefly , 50 μl of a culture were incubated for 2 min with 0 . 5 mM HADA . The cells were then fixed by addition of ice-cold ethanol to a concentration of 70% and incubated at 4°C for 20 min . Subsequently , they were washed three times with PBS and subjected to fluorescence microscopic analysis . For chase experiments , phosphate-starved Caulobacter cells were grown for 90 min in the presence of 0 . 5 mM HADA . The cells were washed three times with M2G-P medium , resuspended in fresh M2G-P medium , and further cultivated for the indicated time intervals . Cells were fixed and washed as described above prior to imaging . Protein sequences containing the indicated domains were retrieved from the UniProt Knowledgebase [101] . Their overall domain composition was determined using the SMART server [102] . The prediction of protein localization and membrane topology was performed with Signal-BLAST [103] and TMHMM [104] , respectively . Cultures were grown in the indicated media and supplemented with 20 μg/ml rifampicin 3 h prior to analysis to block the re-initiation of chromosome replication . At the indicated time points , cells were diluted to an OD600 of 0 . 1–0 . 2 , incubated for 25 min under vigorous shaking with the DNA-specific fluorescent dye Hoechst 33342 ( 10 μM; Thermo Fisher Scientific , Germany ) , and fixed by addition of ethanol to a final concentration of 70% . Subsequently , the suspensions were analyzed by flow cytometry in a customized Fortessa Flow Cytometer ( BD Biosciences , Germany ) , using the UV 440/40 nm channel . Data were acquired with FACSdiva 8 . 0 ( BD Biosciences ) and processed with FlowJo v10 ( FlowJo LLC , USA ) . For whole-cell analyses , cultures were rapidly cooled to 4°C and harvested by centrifugation at 16 , 000 rpm for 30 min . The cells were resuspended in 6 ml of ice-cold H2O and added dropwise to 6 ml of a boiling solution of 8% sodium dodecylsulfate ( SDS ) that was stirred vigorously . After 30 min of boiling , the suspension was cooled to room temperature . Peptidoglycan was isolated from the cell lysates as described previously [105] and digested with the muramidase cellosyl ( kindly provided by Hoechst , Frankfurt , Germany ) . The resulting muropeptides were reduced with sodium borohydride and separated by HPLC following an established protocol [105 , 106] . The identity of eluted fragments was assigned based on the retention times of known muropeptides from Caulobacter [107] . To prepare stalk and cell body fractions , 100 ml cultures grown in M2G-P medium were rapidly cooled to 4°C and harvested by centrifugation at 16 , 000 rpm for 30 min . After resuspension in M2G-P medium , the cells were vigorously agitated for 2 min at maximum speed in a kitchen blender . The suspension was submitted to three rounds of centrifugation at 9 , 000 rpm and 4°C . The supernatants ( stalk fraction ) and the first pellet ( cell body fraction ) were collected separately and kept in ice . The stalk fraction was subjected to an additional centrifugation step at 10 , 000 rpm and 4°C to remove residual cell bodies and cell debris . Subsequently , stalks were collected by centrifugation at 20 , 000 rpm and 4°C for 30 min , resuspended in 3 ml ice-cold H2O , added dropwise to 3 ml of a boiling 8% SDS solution , and then further processed as described above to isolate stalk PG . The isolation of cell body PG was achieved as described for whole-cell samples . | Bacteria show a variety of different cell shapes that are critical for survival in the environmental niche they inhabit . While the mechanisms generating the prototypic rod-shaped and coccoid morphologies have been studied intensively , only little is known about the processes that underlie the formation of more complex morphological features . The model organism Caulobacter crescentus is characterized by a polar stalk , which carries an adhesive organelle mediating surface attachment at its tip . This structure forms through the insertion of new cell wall material at its base and elongates considerably in phosphate-limited conditions . Our work reveals significant differences in the architecture of cell walls isolated from stalks and cell bodies , respectively , hinting at the existence of a stalk-specific cell wall biosynthetic apparatus . To identify components of this machinery , we systematically inactivated and localized proteins with a predicted enzymatic or regulatory function in cell wall biosynthesis in C . crescentus . Our results show that stalk formation is mediated by a pole-associated complex composed of proteins that have previously been identified as components of the cell elongation and cell division machineries . The stalk biosynthetic apparatus may thus have evolved through the repurposing of preexisting machinery , indicating that even complex morphological traits can emerge without the need for extensive changes to the complement of morphogenetic factors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"physiology",
"cell",
"walls",
"chemical",
"compounds",
"phosphates",
"caulobacter",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"cell",
"polarity",
"materials",
"science",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"macromolecules",
"polymers",
"polymer",
"chemistry",
"chemistry",
"biochemistry",
"cell",
"biology",
"peptidoglycans",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"physical",
"sciences",
"materials",
"organisms",
"cell",
"fusion"
] | 2019 | A specialized MreB-dependent cell wall biosynthetic complex mediates the formation of stalk-specific peptidoglycan in Caulobacter crescentus |
Much effort is being devoted for developing new indicators to evaluate the human exposure to Aedes mosquito bites and the risk of arbovirus transmission . Human antibody ( Ab ) responses to mosquito salivary components could represent a promising tool for evaluating the human-vector contact . To develop a specific biomarker of human exposure to Aedes aegypti bites , we measured IgG Ab response to Ae . aegypti Nterm-34 kDa salivary peptide in exposed children in 7 villages of Southern Benin ( West Africa ) . Results showed that specific IgG response presented high inter-individual heterogeneity between villages . IgG response was associated with rainfall and IgG level increased from dry ( low exposure ) to rainy ( high exposure ) seasons . These findings indicate that IgG Ab to Nterm-34 kDa salivary peptide may represent a reliable biomarker to detect variation in human exposure to Ae . aegypti bites . This preliminary study highlights the potential use of Ab response to this salivary peptide for evaluating human exposure to Ae . aegypti . This biomarker could represent a new promising tool for assessing the risk of arbovirus transmission and for evaluating the efficacy of vector control interventions .
Numerous mosquito species of the genus Aedes ( Dipteria: Culicidae ) are vectors of major ( re ) -emerging human arboviruses , such as Dengue and Chikungunya . Aedes aegypti species is the primary vector of these diseases worldwide . No effective treatment and vaccine are currently available and the transmission can only be reduced or interrupted by controlling mosquito populations and by preventing the human-vector contact . Exposure to Aedes aegypti bites is currently evaluated by entomological methods , at immature stage ( eg: number of positive breeding habitats ) and/or adult stage ( collection of adult mosquitoes by traps , Pyrethrum Spray Catch and human landing catches ) . These methods present several limitations , such as poor capacity to predict epidemics [1] and for addressing the number of adults vectors produced over time [2] . These methods are labor-time consuming and costly regarding large-scale follow up of mosquito density required . Furthermore , larval and pupal indices target immature stages and do not measure the exposure to adult bites . The density of adult females could be closely associated with the disease incidence [3] , [4] , but adults collection of Ae . aegypti females is fastidious and hard work . These current entomological methods are mainly applicable at the community level and cannot be used to gauge the heterogeneity of individual exposure . They are not accurate to assess individual attractiveness to mosquitoes or other environmental and socioeconomic factors which could induce important variations in individual exposure to vector bites . In order to improve vector control and to predict the risk of arboviruses transmission , complementary methods and indicators are urgently need to evaluate the real human exposure to Ae . aegypti bites . One promising approach is to quantify the human antibody ( Ab ) response to arthropod salivary proteins used as a biomarker of human exposure to mosquito bites [5] . At the time of biting , the vector injects in the host skin , saliva containing bioactive molecules which facilitate blood feeding [6] . Some of these molecules induce specific Ab response in individuals exposed to bites [7] . Previous studies have shown that anti-saliva Ab response could be an useful indicator to measure the human exposure to arthropod vector bites such as ticks [8] , Triatoma [9] , Phlebotomus [10] , Glossina [11] and Anopheles species [12]–[14] . Concerning Aedes genus , studies on human allergic reactions have suggested that quantitative evaluation of anti-saliva Ab responses could give a measure of human exposure to Aedes bites [15] , [16] and increased during rainy season [17] . A significant increase in the anti-saliva Ab response was also observed according to seasonal and spatial Ae . caspius density in Southeast France [18] . Regarding Ae . aegypti species , it has been demonstrated that IgM and IgG responses to whole saliva could be promising indicator of Ae . aegypti exposure in temporarily exposed populations [19] . One study in tropical countries showed that IgE and IgG4 responses to Ae . aegypti saliva could be detected in young Senegalese children and that their level increased during the rainy season [20] . Interestingly , IgG response to Ae . aegypti saliva was positively associated with entomological indicators in a study conducted in urban area in Bolivia [21] . Recently , our team has shown that IgG Ab level to Ae . albopictus can evaluate the exposure to this species in adult individuals [22] . This study demonstrated also a low-level immune cross-reactivity between Ae . albopictus and Ae . aegypti saliva suggesting the potential to develop specific biomarker to each species . These results established that specific Ab response against arthropod saliva could evaluate human exposure to vectors bites . Nevertheless , the whole saliva could not be used as convenient biomarker because some families of salivary proteins are common to many bloodsucking Diptera [6] . This could induce potential cross-reactivity which potentially skew and/or overestimate the evaluation of exposure to a specific vector . In addition , the use of whole saliva presents other drawbacks such as: i ) lack of reproducibility between saliva batches and ii ) its adequate production needed for large-scale studies . An optimization of this indicator would be the identification of specific proteins and/or peptides . In this way , our team has validated one salivary peptide ( gSG6-P1 ) as pertinent specific biomarker of exposure to An . gambiae and An . funestus bites [23]–[25] . By immuno-proteomic approach , a recent study had identified 15 proteins in the sialome of female of Ae . aegypti to be potentially antigenic [26] . Among them , the putative 34 kDa family secreted salivary protein appeared specific to Aedes genus . Using similar approach than used for gSG6-P1 peptide , the N-terminal extremity peptide ( Nterm-34 kDa peptide ) of the 34 kDa protein appeared to be an interesting candidate for validation as a biomarker specific to Ae . aegypti bites . The present study aimed at determining whether the IgG Ab response to Nterm-34 kDa peptide could be a biomarker of exposure to Ae . aegypti bites in African children living in area of exposure to this vector species . The immunological follow up of a cohort of children was carried out for two years and the changes in specific IgG level were evaluated according to the rainfall quantity and the season of exposure .
This study followed the ethical principles according to the Helsinki Declaration and was approved by the National Ethical Committee of Benin ( IRB 00006860 ) and the IRD ethical committee ( April 2008 ) . Written informed consent was obtained for all children enrolled in the study and signed by one of their parents . The study was carried out in rural area of the Ouidah-Kpomassé-Tori Bossito ( OKT ) health district in southern Benin ( West Africa ) . This site is characterized by a sub-equatorial climate with two dry seasons ( from December to March and from August to September ) and two rainy seasons ( from April to July and from October to November ) . The annual average of rainfall is around 1 , 200 mm of which 700–800 mm during the major rainy season and 400–500 in the short rainy season . In this area , a previous study indicated that Ae . aegypti is the major Aedes species caught inside and around the households [27] . Data for the present study were collected during a longitudinal survey conducted between February 2008 and October 2009 in 7 villages of the OKT health district ( 1 = Aidjédo; 2 = Dokamé; 3 = Kindjitokpa; 4 = Guézohoué; 5 = Hékandji; 6 = Satré; 7 = Wanho ) . After census , 420 children ( 60 for each village ) aged from 0 to 60 months old were randomly selected as previously described [28] . Children were visited every 6 weeks and overall 14 visits were conducted during the studied period . At each visit , a dried blood spot was collected in filter-paper from each individual for immunological analysis . The immunological assays were performed on a sub-sample ( n = 205 ) of children for whom blood spots were available for , at least 12/14 visits ( 89 of 205 children missed one [n = 53] or two [n = 36] visits ) . No newborn during the study period was included in the present study . All filter papers were kept at +4°C before used . As previously described for An . gambiae [23] , a peptide design strategy using bio-informatic tools was conducted to identify the potential antigenic properties of the Nterm-34 kDa salivary peptide and to select it as candidate biomarker of exposure to Ae . aegypti bites . The antigenicity of this N-terminal extremity peptide ( 19 amino-acids ) of the putative 34 kDa family secreted salivary protein ( gi|94468336; [29]; figure 1 ) was computerizing predicted with the BcePred and the FIMM databases . In addition , sequence alignments with the Blast program in Vectorbase database demonstrated the specificity to Ae . aegypti by comparison with known genomes and EST libraries of other mosquitoes or organisms . The Nterm-34 kDa peptide was then synthesized , purified ( >95% ) by Genepep SA ( St-Jean de Vedas , France ) . The peptides were shipped in lyophilized form and then resuspended milliQ water and stored in aliquots at −20°C until their use . Enzyme-linked immunosorbent assay ( ELISA ) was carried out to evaluate the level of IgG Ab to Nterm-34 kDa peptide in eluates obtained from standardized dried blood spots ( 1 cm diameter ) . All ELISA conditions were determined after several preliminary experiments . Samples were eluted by incubation in 350 µl of phosphate buffer ( PBS+Tween 0 . 1% , Sigma-Aldrich , St . Louis , MO ) at 4°C for 24 hours . The peptide ( 10 µg/mL in 100 µl of PBS ) was coated at 37°C for 150 minutes onto 64 wells of a 96-well Maxisorp plates ( Nunc , Roskilde , Denmark ) . For each individual sample , one “no antigen” well will be performed to measure the individual non-specific ELISA reactivity by using only 100 µl of PBS for the coating as previously described [23] , [25] . Plates were blocked using 300 µl of Protein-Free Blocking-Buffer ( Pierce , Thermo Scientific , France ) for 45 minutes at 37°C . Each eluate was incubated at 4°C overnight at 1/20 dilution in PBS-Tween 1% in two wells containing peptide and in one “no antigen” well ( 100 µl for each well ) . Mouse biotinylated Ab to human IgG ( BD Bioscences , San Diego , CA ) was incubated at a 1/1000 dilution in PBS-Tween 1% ( 90 minutes at 37°C ) and peroxidase-conjugated streptavidin ( GE Healthcare , Orsay , France ) was added ( 1/1000 dilution in PBS-Tween 1%; 60 minutes at 37°C ) . Colorimetric development was carried out using 2 , 2′-azino-bis ( 3-ethylbenzthiazoline 6-sulfonic acid ) diammonium ( ABTS; Thermo Scientific , France ) in 50 mM citrate buffer ( pH 4 ) containing 0 . 003% H2O2 and absorbance ( OD ) was measured at 405 nm . In parallel , specific IgG Ab levels were also evaluated in individuals ( n = 10 ) living in the North of France and with no known exposure to Ae . aegypti mosquito and were used to calculate the specific immune response threshold ( TR ) . Individual results were expressed as the ΔOD value calculated according to the formula ΔOD = ODx−ODn , where ODx represented the mean of individual OD values in antigen wells and ODn the OD value in “no antigen” well . A subject was considered as an “immune responder” if his ΔOD was higher than the TR = mean ( ΔDOunexposed ) +3SD = 0 . 151 . All data were analyzed with GraphPad Prism5 software ( San Diego , CA ) . After verifying that ΔOD values were not normally distributed , the non-parametric tests were used to compare the ΔOD . Mann–Whitney test was used for comparison of Ab levels of two independent groups and the Wilcoxon matched-pairs test was used for comparison of two paired groups . The non-parametric Kruskal–Wallis test was used for comparison of more than two groups . All differences were considered significant at P<0 . 05 .
The evolution of specific IgG level from February 2008 to October 2009 were presented during studied period and compared to the accumulated monthly rainfall registered in the same studied area ( Figure 2 ) . For each visit , the median value of IgG Ab level was higher than the specific immune response threshold ( TR = 0 . 151 ) . Specific IgG level was more pronounced i ) in July 2008 compared to previous dry season and ii ) during all studied period in 2009 . High inter-individual heterogeneity in specific IgG Ab level was observed whatever the studied months . The specific IgG response showed significant seasonal variations from the start to the end of the study ( P<0 . 0001 , Kruskal–Wallis test ) . The lowest Ab levels were observed in 2008 during the dry season ( from February to May ) . For both years , IgG level increased significantly from February to March-April ( P<0 . 0001 , Wilcoxon matched-pairs test ) and from May to July ( P<0 . 0001 and P = 0 . 002 for 2008 and 2009 respectively , Wilcoxon matched-pairs test ) , whereas a decrease was observed from March-April to May ( P<0 . 0001 , Wilcoxon matched-pairs test ) . A considerable increase was thereafter observed from July . In contrast , it has been observed a different evolution of IgG level from July to August between 2008 ( non significant increase; P = 0 . 11 , Wilcoxon matched-pairs test ) and 2009 ( significant decrease; P = 0 . 023 , Wilcoxon matched-pairs test ) . Globally , the specific IgG response was globally higher in 2009 than in 2008 . In the same way , rainfall was more intense during the year 2009 ( total = 1 , 252 . 41 mm ) than in 2008 ( total = 962 . 23 mm ) . In 2008 , the curve of rainfall was closely associated with the specific IgG response . The rainfall started from February to April , highly increased from May to June , and then decreased from July to August . Interestingly , the peak of rainfall on June 2008 and 2009 was always followed by a peak of IgG Ab level in July . Similar results were observed for the percentage of immune responders ( ΔOD>TR; Table 1 ) . During the first dry season , 63 . 42% of children were responders ( 28 . 78% , 84 . 84% , 76 . 74% for February , March-April and May respectively ) , whereas this percentage reached to an average of 97 . 28% ( [95 . 95%–98 . 97%] ) from July 2008 to October 2009 . Altogether , these results suggest a possible association between specific IgG Ab response and the intensity of rainfall . This association appeared to be more pronounced in 2008 compared to 2009 . In the objective to highlight the potential association of specific IgG response with rainfall , IgG Ab level to Nterm-34 kDa peptide was compared between the peak of the dry ( February ) and rainy ( July ) seasons in 2008 ( Figure 3A ) and 2009 ( Figure 3B ) . For both years , specific IgG response increased significantly ( P<0 . 0001 in 2008 and P<0 . 001 in 2009 Wilcoxon matched-pairs test ) in the rainy season compared to dry season . The increase of IgG level was more pronounced in 2008 than 2009 . Interestingly , almost all individuals were immune responders ( >TR ) in rainy season 2008 , whereas the median value was closed to TR and only 28 . 78% of individuals presented positive IgG in dry season . The results of the percentage of immune responders confirmed these differences between 2008 and 2009 ( Table 1 ) . High increase was observed from dry season ( 28 . 78% ) to rainy season ( 98 . 98% ) in 2008 . In 2009 , these percentages did not differ between both seasons ( 96 . 51% and 96 . 90% , respectively ) . These results suggest that the intensity of IgG response to Nterm-34 kDa peptide increased with the rainy season . The evolution of specific IgG level in the different villages was compared between the peak dry ( February ) and the rainy ( July ) seasons in 2008 ( Figure 4A ) and 2009 ( Figure 4B ) . A significant variation of IgG Ab levels was observed between villages for both years . IgG level significantly increased for all villages from dry to rainy season in 2008 ( P<0 . 0001 for villages 1 , 2 , 3 , 4 , 5 , 7 and P = 0 , 003 for village 6 , Wilcoxon matched-pairs test ) , whereas different trends were observed in 2009 . Indeed , increase of IgG Ab levels with the 2009 rainy season was observed only in four villages ( 1 , 4 , 7; P<0 . 0001 and 2 , P = 0 . 89; Wilcoxon matched-pairs test ) , whereas during the same period , the specific IgG level decreased in village 6 ( P>0 . 05 ) , and appeared to be maintain in villages 3 and 5 ( P>0 . 05; Wilcoxon matched-pairs test ) .
This study described for the first time , the development of IgG Ab response to Ae . aegypti Nterm-34 kDa salivary peptide in human individuals exposed to Ae . aegypti bites . The immunological result seemed to confirm the bioinformatic predictions which suggested the potential antigenic properties of this salivary peptide . IgG response to Nterm-34 kDa varied according to the season and was positively associated with the intensity of rainfall . This observation appeared to be more pronounced in 2008 than in 2009 . Interestingly , high level of specific IgG response was observed only during rainy season for both years , when almost 100% of children are immune responders . Altogether , these results indicated that the IgG response to Nterm-34 kDa peptide could represent a promising candidate as biomarker of human exposure to Ae . aegypti bites . Evaluation of Ab responses to salivary components might represent an epidemiological marker of Aedes exposure . It has been previously reported an increase of Ab response to Aedes saliva according to the period of high exposure to mosquito bites [17] , [20] . However , the use of whole saliva as biomarker is hampered by its potential to cross-react with others arthropods [30] . To optimizing salivary biomarker for assessing human exposure to Aedes bites , this study reports the existence of Ab response to Aedes salivary peptide in individuals . The IgG response to Nterm-34 kDa salivary peptide in children was different between villages and between individuals within the same village . The heterogeneity of IgG level to mosquito saliva components had already been reported by previous studies [20] , [23] . It suggests the high heterogeneity of exposure to vector bites among villages and among individuals , as also known for vector-borne diseases transmission . Even if the influence of epidemiological factors ( history of exposure , human genetic background , pathogen infections , nutritional status , etc… ) on individual Ab response could not be excluded , this result suggests that specific Ab response to Nterm-34 kDa salivary peptide could be pertinent for evaluating the individual exposure to vector bites . This is in accordance with several previous studies showing that such biomarker could be individual indicator for evaluating the real human-vector contact [12] , [17] , [18] , [20] , [22] , [25] . In addition , variations in the levels of IgG to Nterm-34 kDa peptide appeared related to the intensity of rainfall . The level of specific IgG response globally increased with the increase of rainfall and decreased otherwise . The specific Ab response was higher in 2009 compared to 2008 , which was in accordance to the higher rainfall observed in 2009 than in 2008 . However , it could be noticed a slight drop of IgG Ab response between August 2008 and February 2009 while rainfall drops considerably . It could probably due to the persistence of considerable exposure to mosquito bites . Indeed , despite the decline of rainfall , it continued to rain during this period , even with weak intensity . It can be favorable for maintaining the proliferation of important densities of Ae . aegypti in persistent domestic breeding sites . This could also probably due to the production of mosquitoes in containers filled by people when rain scant as observed in others previous studies [31] , [32] . The present results highlight a probable influence of the rainfall on the increase of specific IgG Ab level during the studied period . This association was relevant when the evolution of the specific IgG response was compared between the dry and the rainy seasons for both years , taking into account the peak of the dry ( February ) and the rainy ( July ) seasons . Similar influence of rainfall had been previously noticed for Ab response to whole saliva in human populations exposed to Aedes [17] , [20] and to Anopheles bites [12] . Positive association between the levels of IgG Ab response to saliva components and the densities of adult mosquito was clearly reported in several sites and for different mosquito genus; i . e . Anopheles , Aedes and Culex [12] , [18] , [21] , [33] , [34] . We can thus hypothesize that the increase of anti-Nterm-34 kDa IgG response during the rainy season could reflect the increase of human exposure to high densities of Ae . aegypti mosquito . It is well known that greater proliferation of Ae . aegypti adult mosquito occurs during rainfall , especially in African rural context [35] . Additionally , previous investigations indicated that captured female density peaked during times of heavier rainfall in tropical regions [4] , [36] . It has been also previously developed a mathematical model which , applied to field data , showed that rainfall triggered the dynamics of Aedes mosquito aggressiveness [37] . Collectively , these results indicated that association between rainfall and the level of IgG Ab response to Nterm-34 kDa salivary peptide may reflect the real intensity of human exposure to Ae . aegypti bites . Regarding the 2009 season-dependent evolution of specific IgG level according to villages , an increase in the rainy season was observed only for villages 1 , 2 , 4 and 7 . In contrast , the level of specific Ab response appeared unchanged in villages 3 and 5 and decreased in village 6 . It could probably indicate that children in villages 3 , 5 and 6 could be more protected or less exposed to Aedes bites than those in the others village . However , we can't exclude that studied individuals could be exposed to bites of other Aedes species such as Ae . vittatus which its presence was reported in our study area [27] . Nevertheless , the lack of studies on the sialome of this Aedes species has not allowed a bio-informatic comparison with Ae . aegypti salivary proteins during the identification of the Nterm-34 kDa peptide . In this study , the percentages of immune responders were lower and significantly changed at the first three time points in 2008 . It could probably be explained by the progressive development of immune response due to cumulative exposure of Aedes bites in youngest children . Thereafter , this Ab response level could reach a baseline at determined age and from this age no difference of Ab level can be detected . This hypothesis may explain that the percentage of immune responders remained high and did not differ from July 2008 until the end of study . Nevertheless , in contrast to the proportions of immune responders , the level of specific IgG increased during the rainy season . It suggest , as previously observed for whole saliva [20] , that only the level of specific Ab increased during the season of high exposure to Aedes bites , but not the percentage of responders . Altogether , our results showed that individuals exposed to Aedes bites could develop IgG response to Nterm-34 kDa salivary peptide . The Ab response differed between individual and increased during season of high exposure to mosquito bites . These data represent a first step to validate the Nterm-34 kDa salivary peptide as a potential biomarker of human exposure to Aedes bites . Further studies are needed for final validation taking into account: ( i ) entomological indicators , even those present considerable limitations; ( ii ) arbovirus transmission and ( iii ) others exposed areas with different dynamics of Aedes populations . If validated , the level of specific Ab response to Nterm-34 kDa salivary peptide could be used for control and survey programs: ( i ) to assess the risk of arboviruses transmission and ( ii ) to evaluate the efficacy of vector control strategies . | Aedes aegypti mosquito is the primary vector of major ( re ) -emerging human arboviruses , such as Dengue and Chikungunya . In absence of effective treatment and vaccine , the evaluation of human exposure to vector bites is crucial to estimate the risk of the viruses' transmission . Currently , exposure to Aedes aegypti bites is mainly evaluated by entomological methods which are indirect and fastidious to apply on a large scale . Human antibody ( Ab ) responses to arthropod salivary proteins were shown as a useful indicator of exposure to arthropod vector bites . Nevertheless , the whole saliva could not be a specific tool because some families of salivary proteins are common between many arthropod vectors . To develop a specific biomarker of exposure to Aedes aegypti bites , we assessed the evolution of IgG Ab response to Ae . aegypti Nterm-34 kDa salivary peptide in exposed children . The results indicate that children exposed to the bites of Ae . aegypti could develop specific Ab response to Nterm-34 kDa salivary peptide . This specific IgG response presented high inter-individual heterogeneity and increased significantly during the Ae . aegypti exposure season . Taken together , these preliminary results suggest that Ab responses to Nterm-34 kDa salivary could constitute a relevant immuno-epidemiological indicator for evaluating human exposure to the Ae . aegypti vector and by consequence the risk of arbovirus transmission . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"antigen",
"processing",
"and",
"recognition",
"entomology",
"dengue",
"fever",
"neglected",
"tropical",
"diseases",
"immunology",
"biology",
"arboviral",
"infections",
"zoology",
"infectious",
"disease",
"control",
"immune",
"response",
"immunoglobulins"
] | 2012 | First Attempt To Validate Human IgG Antibody Response to Nterm-34kDa Salivary Peptide as Biomarker for Evaluating Exposure to Aedes aegypti Bites |
In checkpoint-deficient cells , DNA double-strand breaks ( DSBs ) are produced during replication by the structure-specific endonuclease MUS81 . The mechanism underlying MUS81-dependent cleavage , and the effect on chromosome integrity and viability of checkpoint deficient cells is only partly understood , especially in human cells . Here , we show that MUS81-induced DSBs are specifically triggered by CHK1 inhibition in a manner that is unrelated to the loss of RAD51 , and does not involve formation of a RAD51 substrate . Indeed , CHK1 deficiency results in the formation of a RAD52-dependent structure that is cleaved by MUS81 . Moreover , in CHK1-deficient cells depletion of RAD52 , but not of MUS81 , rescues chromosome instability observed after replication fork stalling . However , when RAD52 is down-regulated , recovery from replication stress requires MUS81 , and loss of both these proteins results in massive cell death that can be suppressed by RAD51 depletion . Our findings reveal a novel RAD52/MUS81-dependent mechanism that promotes cell viability and genome integrity in checkpoint-deficient cells , and disclose the involvement of MUS81 to multiple processes after replication stress .
Faithful completion of DNA replication and accurate transmission of the genetic information to daughter cells is of paramount importance . To ensure genome integrity , cells have evolved a sophisticated mechanism that supervises the replication process , the replication checkpoint [1] . Replication checkpoint is a system well conserved from lower to higher eukaryotes , and , in humans , is orchestrated by the ATR kinase [2] . ATR regulates directly or indirectly the function of several proteins involved in maintaining replisome stability , promoting restart of perturbed replication forks , and controlling cell cycle arrest [3] . The coordination of these activities is needed for completing replication , and avoiding accumulation of DNA damage or chromosomal rearrangements [4] . Consistently , replication checkpoint mutants fail to resume replication without accumulating DNA damage once the cause of the arrest is removed . These mutants also show chromosomal instability [1] . It has been suggested that inability of checkpoint mutants to resume replication at perturbed forks is directly related to their impaired capacity to stabilise them , eventually leading to accumulation of collapsed forks [1] , [3] . Studies in yeast demonstrated that collapsed forks can be processed by exonucleases or converted into unusual replication intermediates , i . e . reversed forks , which can be substrates for endonucleases [5] , [6] , [7] . MUS81 is a structure-specific endonuclease that shows a remarkable preference for cleaving branched DNA substrates , such as nicked Holliday's Junctions ( HJs ) , D-loops or three-way junctions [7] , [8] , [9] . MUS81 forms a heterodimeric complex with the non-catalytic EME1 subunit . Genetic studies in yeast have shown that this complex is involved in the resolution of HJs or in the processing of other replication intermediates generated at the perturbed forks [7] , [9] , [10] . In fission yeast , MUS81 is responsible for the formation of DNA double-strand breaks ( DSBs ) , which are frequently observed in replication checkpoint mutants [11] . In addition , MUS81-dependent cleavage may take place downstream of RAD51 or RAD52 [12] , [13] . In human cells , it has been shown that MUS81 is rapidly engaged at stalled replication forks to produce DSBs when fork collapse is triggered by loss of the Werner syndrome ( WRN ) RecQ helicase [14] , [15] , [16] . It remains unknown whether this function of MUS81 in human cells can be extended to other pathological conditions associated with replication checkpoint deficiency . Similarly , it is not known if cleavage by MUS81 in checkpoint-deficient cells occurs as a consequence of impaired , checkpoint-regulated RAD51 function [17] . Finally , the identity of the structure cleaved by MUS81 at stalled replication forks after checkpoint demise , as well as the mechanism underlying the fork collapse , remains undefined . Here , we report that down-regulation of several replication-checkpoint factors inevitably leads to MUS81-dependent DSBs , which is essential to allow cellular recovery from replication stress . We also provide insights into the underlying mechanism by demonstrating that MUS81 cleavage is correlated to the loss of CHK1 activity , but is independent from the loss of RAD51 function . Moreover , we demonstrated that in vitro MUS81 acts on a D-loop formed by RAD52 but not RAD51 , and that in vivo , after CHK1 inactivation , MUS81 functions downstream of RAD52 . Our findings also suggest that loss of RAD52 promotes enhanced RAD51 chromatin association that is toxic in the absence of MUS81 . Altogether , these results provide an insight into how multiple mechanisms can cooperate at the distressed replication forks to allow cellular recovery , and viability in human cells with defective checkpoint function , a condition that may characterize a subset of human tumours and that may be exploited in targeted therapy .
In human cells , inactivation of replication checkpoint factors results in formation of DSBs [4] . To determine whether these DSBs are derived from MUS81-mediated endonucleolytic cleavage at stalled forks , we performed a neutral Comet assay after depletion of different checkpoint proteins . To this end , hTERT-immortalised human primary fibroblasts were transfected with siRNAs directed against selected checkpoint factors , by themselves or in combination with siRNAs against MUS81 ( Figure 1A ) . In addition , we down-regulated expression of CHK2 , a checkpoint kinase that is not activated after perturbed replication , and used siRNAs as internal controls . The efficiency of RNAi was evaluated by Western blotting 48 h post-transfection , and protein level of the targets was reduced by at least 80% in comparison to cells transfected with GFP siRNAs ( siCtrl; Figure 1A ) . After transfection , cells were treated with 2 mM HU for 6 h , a condition sufficient to induce DSBs only under pathological condition [14] , and subjected to neutral Comet assay . Depletion of each of the selected checkpoint factors resulted in DSBs formation when replication was not perturbed , and further enhanced after HU treatment ( Figure 1B ) . The DSBs level was highest after CHK1 or ATR depletion , whereas RAD9 or TOPBP1 down-regulation resulted in a limited DSBs enhancement . No increase was observed in cells depleted of MUS81 , TIPIN or CHK2 . Interestingly , MUS81 depletion reduced DSBs levels , albeit to a different extent , in all the checkpoint-deficient cells , either untreated or treated with HU . The only exceptions were TIPIN or CHK2 knock-down cells ( Figure 1B ) . Consistent results were also obtained analysing the formation of pan-nuclear staining of the phosphorylated H2AX . In this case , checkpoint impairment after ATR or CHK1 depletion gave rise to a strong accumulation of nuclei showing intense γH2AX staining , which was reduced by concomitant MUS81 knock-down ( Figure S1A and B ) . To investigate whether MUS81-dependent DSBs occurred during S-phase , cells released from G0-phase were transfected with siRNAs targeting CHK1 , MUS81 or both , and then exposed to HU ( Figure S2A ) . Suppression of DSBs by MUS81 depletion was detected in S-phase cells ( Figure S2A and B ) . However , this was not simply due to an altered S-phase progression , since MUS81 knock-down did not induce premature G2 entry . Upon oncogene-induced replication stress or in the absence of the WRN RecQ helicase , MUS81 function is required for cell viability [14] , [18] . To examine whether MUS81-dependent DSBs were needed for cellular recovery in HU-treated replication checkpoint-deficient cells , we performed viability assays in cells in which ATR , CHK1 or TIPIN was down-regulated or chemically inhibited . As reported in Figure 1C , depletion of the selected checkpoint proteins enhanced cell death during both unperturbed and HU-perturbed proliferation . Interestingly , MUS81 down-regulation per se did not increase cell death , but hypersensitized ATR or CHK1 depleted cells to the replication arrest ( Figure 1C ) . Moreover , even a short-term inhibition of ATR by ETP-46464 [19] or CHK1 by UCN-01 was sufficient to induce a MUS81 requirement for survival upon replication perturbation ( Figure 1C ) . In contrast , MUS81 down-regulation did not synergize with loss of TIPIN ( Figure 1C ) . We next investigated whether enhancement of cell death by MUS81 depletion could reflect premature cell cycle progression . We analysed Bromo-deoxyuridine ( BrdU ) incorporation and arrest in G2-phase during recovery from HU in the cells in which MUS81 was down-regulated by RNAi , and the replication checkpoint was inhibited by UCN-01 or ETP-46464 . To analyse replication recovery , cells were treated with HU , recovered for 60 or 120 min in drug-free medium , and exposed to BrdU for 30 min immediately before sampling . Both ATR and CHK1 inhibition caused a strong reduction in BrdU incorporation after HU , which was not reverted by prevention of MUS81-dependent DNA breakage ( Figure S3A ) . Since inhibition of ATR or CHK1 results in accumulation of cells in G2-phase after over-night recovery from HU treatment [19] , we analysed whether MUS81-depletion affected G2-phase arrest . For this purpose , cells enriched in S-phase by serum deprivation were transfected with Ctrl or MUS81 siRNAs , exposed to HU and recovered for 18 h . Western blotting analysis confirmed efficient depletion of MUS81 by RNAi after synchronisation ( Figure S3B ) . As shown in Figure S3C , Ctrl RNAi cells showed a delayed S-phase upon HU treatment , and after an overnight recovery in the drug-free medium the cells accumulated in G2/M . Inhibition of CHK1 slowed further S-phase after HU , but MUS81 depletion did not rescue the phenotype ( Figure S3C ) . Interestingly , MUS81 depletion delayed cell cycle progression after HU in mock-inhibited cells ( Figure S3C ) . Altogether , our observations indicate that down-regulation of replication checkpoint factors , but not of the DNA damage checkpoint kinase CHK2 or the replication fork protection factor TIPIN , results in MUS81-dependent DSBs . In addition , MUS81-dependent cleavage of stalled forks is required to maintain cell viability of replication checkpoint-deficient cells . Disruption of replication checkpoint function can lead to loss of CHK1 phosphorylation [4] . We observed the highest levels of DNA breakage after ATR or CHK1 silencing , while depletion of TIPIN , which may or may not affect CHK1 activation , did not induce MUS81-dependent DSBs . Thus , we investigated whether formation of DSBs by MUS81 was directly related to a defective CHK1 phosphorylation in checkpoint-deficient cells . To this aim , we down-regulated ATR , RAD9 or TIPIN , then we analysed CHK1 phosphorylation by Western blotting using phospho-specific antibodies . Down-regulation of ATR or RAD9 , but not of TIPIN , impaired CHK1 phosphorylation at S345 ( Figure 2A ) . It is worth noting that the residual level of CHK1 phosphorylation seems to be inversely correlated with the amount of MUS81-dependent DSBs . In fact , residual CHK1 phosphorylation in RAD9 RNAi cells corresponded to less DSBs than in ATR knock-down cells ( see Figure 1B ) . CHK1 phosphorylation is a pre-requisite to kinase activation [20] , and CHK1-mediated phosphorylation of downstream targets may contribute to preventing replication fork collapse via MUS81-dependent cleavage . Thus , cells in which MUS81 was down-regulated by RNAi were treated with UCN-01 to inhibit CHK1 , alone or in combination with 2 mM HU , and then processed by neutral Comet assay . Inhibition of CHK1 by UCN-01 recapitulated the phenotype of CHK1 RNAi-treated cells , albeit with a reduced accumulation of DSBs ( Figure 2B ) . However , MUS81 down-regulation decreased the number of DSBs formed as a consequence of UCN-01 treatment as efficiently as observed after CHK1 RNAi ( Figure 2B and 1B ) . Chemical inhibition of CHK1 also allowed the analysis of time-dependent formation of DSBs at perturbed forks , as well as their genetic dependency on MUS81 . UCN-01 triggered DSBs already after 4 h of HU treatment , and increased substantially at 6 h , when DSBs are detectable also in cells treated with UCN-01 or HU alone ( Figure S4A ) . Similarly to what observed after 6 h of the combined UCN-01 and HU treatment , the DSBs detected at 4 h were MUS81-dependent ( Figure S4B ) . Generation of DSBs by MUS81 could be secondary to accumulation of single-stranded DNA ( ssDNA ) regions or gaps at the leading or lagging strand , caused by the checkpoint inhibition . Using alkaline Comet assay , we analysed the formation of ssDNA regions or gaps after CHK1 inhibition at forks stalled by HU . As expected , in the HU-treated cells , ssDNA or gaps start to accumulate already at 1 h after CHK1 inhibition , and greatly increased at 4–6 h ( Figure S4C ) , when DSBs ( see Figure S4A ) are also detected by alkaline Comet assay . Thus , formation of ssDNA regions or DNA gaps precedes MUS81-dependent cleavage at perturbed replication forks . RAD51 is a CHK1 substrate , and its phosphorylation at T309 is required for recovery after HU stress [17] . Since it is unclear whether this phosphorylation is required for stabilizing stalled forks , we investigated the relation between MUS81-dependent cleavage of perturbed replication forks and the assembly of active RAD51 nucleoprotein filaments at ssDNA regions . In agreement with previous reports [17] , we observed that depletion of ATR or chemical inhibition of CHK1 almost completely prevents RAD51 from nuclear foci assembly after HU treatment . In contrast , TIPIN down-regulation does not affect the ability of cells to relocalize RAD51 into foci ( Figure S5 ) . Next , we analyzed whether loss of RAD51 phosphorylation and/or CHK1 activity equally promote MUS81 function . To this end , we analysed DSBs formation in cells depleted of RAD51 ( Figure 2C ) . Consistently with other reports [21] , [22] , DSBs were detected after HU treatment . These DSBs , however , were not sensitive to MUS81 RNAi ( Figure 2D ) . Consistently , analysis of DSBs in BRCA2 mutant cells revealed that MUS81 knock-down did not affect the level of DNA breakage induced by CHK1 inhibition and HU treatment ( Figure S6 ) . To further substantiate this observation , we transfected the RAD51 RNAi-depleted cells with an RNAi-resistant , wild-type RAD51 or the phosphorylation-defective RAD51-T309A mutant ( Figure 2E ) . As shown in Figure 2F , the RNAi-resistant RAD51 proteins were expressed at levels comparable to the endogenous protein , and are not knock-downed by the siRNA oligo directed against the UTR of RAD51 . The expression of the RAD51-T309A mutant also increased the amount of DSBs at levels similar to those associated with RAD51 loss ( Figure 2G ) , and these breaks were abolished by the expression of a wild-type RAD51 ( data not shown ) . Most importantly , DSBs observed in cells expressing the RAD51-T309A mutant were not reduced by depletion of MUS81 ( Figure 2G ) . It is notable that loss of RAD51 function or phosphorylation results in DSBs levels substantially lower than in CHK1-deficient cells . Altogether , these findings show that formation of MUS81-dependent DSBs in replication checkpoint-deficient cells is a consequence of the reduced CHK1 activity on targets distinct from RAD51 . Furthermore , our results suggest that MUS81-dependent cleavage does not occur downstream of RAD51 . Since we showed that DSBs occurring in RAD51-depleted cells are independent from MUS81 function , we examined the possible involvement of RAD52 in generating a MUS81 substrate . Neutral Comet assays were performed in cells in which CHK1 was chemically inhibited and RAD52 depleted by RNAi alone or in combination with MUS81 . Comparable reduction in protein levels was verified by Western blotting ( Figure 3A ) . RAD52 down-regulation barely affected the level of DSBs after single treatments with HU or UCN-01 ( Figure 3B ) . Upon combined treatment , however , RAD52 down-regulation efficiently suppressed DSBs , and this reduction was comparable to that observed following MUS81 depletion . In contrast , MUS81/RAD52 co-depletion resulted in the reappearance of DSBs , at levels similar to that of treatment with UCN-01 alone ( Figure 3B ) . Even though either MUS81 or RAD52 down-regulation suppressed DSBs formation after CHK1 inhibition , MUS81 depletion strongly increased the mean tail moment observed by alkaline Comet assay , which are reversed by RAD52 but not RAD51 knock-down ( Figure S7 ) . Since the alkaline Comet assay detects both ssDNA or DNA gaps and DSBs , in the single depleted MUS81 or RAD52 cells , where DSBs are almost absent , only ssDNA regions are likely formed . In contrast , the reduction of the mean tail moment observed in the double MUS81/RAD52-depleted cells implies that only DSBs are formed . To analyse whether DSBs generated in the absence of MUS81 and RAD51 might depend upon SLX4 or GEN1 , two endonucleases that can substitute for MUS81 in processing DNA replication intermediates [23] , [24] , we analysed DSBs formation in the MUS81/RAD52 double-depleted cells in which each single endonuclease was down-regulated ( Figure 3C ) . We found that , after UCN-01 and HU treatment , GEN1 depletion suppressed DSBs accumulation in the MUS81/RAD52-depleted cells ( Figure 3D ) . Surprisingly , SLX4 down-regulation enhanced formation of DSBs in cells treated with UCN-01 alone ( Figure 3D ) . Interestingly , whereas the DSBs produced by MUS81 in a wild-type background after UCN-01 and HU treatment are neither RAD51-independent nor repaired by RAD51 , since RAD51 inhibition did not affect their appearance ( Figure S8 ) , DSBs formed by GEN1 depend on RAD51-mediated strand invasion ( Figure S8 ) . We next analysed the recruitment of RAD52 to chromatin in cells treated with UCN-01 or HU , with or without prior MUS81 down-regulation . To this end , cellular fractionation experiments followed by SDS-PAGE and Western blotting were performed . In wild-type cells , chromatin localization of RAD52 did not change overtly after CHK1 inhibition or HU-induced replication arrest , however , the combined treatment increased three-fold the amount of chromatin-associated RAD52 ( Figure 3E ) . This enhanced recruitment of RAD52 to chromatin was unaffected by MUS81 down-regulation ( Figure 3E ) . Together , analyses of DSBs and RAD52 chromatin association indicate that MUS81 and RAD52 cooperate in the replication checkpoint-deficient cells , and that RAD52 may work upstream of MUS81 . One of the putative MUS81 substrates that may be formed at the HU-stalled or collapsed forks is a D-loop [7] . To verify this possibility , we evaluated the ability of a purified MUS81/EME1 complex to cleave a model D-loop assembled in vitro by either RAD52 or RAD51 . The human MUS81/EME1 complex was immunopurified from 293T cells transiently transfected with plasmids expressing Myc-tagged MUS81 and GST-tagged EME1 ( Figure 4A ) . Purified RAD52 or RAD51 were pre-incubated with P32-labelled oligonucleotides complementary to a region of the φX174 plasmid . The resulting nucleoprotein complexes mediated formation of the D-loops comprised by φX174 RFI supercoiled dsDNA and the P32-labelled oligonucleotides . The resulting D-loops were incubated with increasing amounts of the MUS81/EME1 complex . Cleaving or nicking the D-loop should result in the loss of superhelicity , displacement of the oligonucleotide , and disappearance of the D-loop . As shown in Figure 4B , MUS81/EME1 cleaved the D-loop produced by RAD52 in a concentration-dependent manner , as demonstrated by the reduction in the amount of the substrate . In contrast , the D-loop produced by RAD51-mediated strand invasion appeared resistant to endonucleolytic cleavage ( Figure 4B ) . Notably , cleavage of this type of a D-loop required the presence of both MUS81/EME1 nuclease and RAD52 since MUS81/EME1 was extremely less efficient in cleaving the protein-free D-loops produced in the control experiment by heat-mediated annealing . To investigate if RAD52 stimulation of MUS81 activity was specific for the D-loop , we prepared a Cy5-labelled 3′-flap substrate , which represents one of the acknowledged and preferred MUS81 substrates . As shown in Figure 4C , MUS81 cleaved the 3′-flap substrates , giving rise in the generation of the nicked product . As expected , incubation of RAD52 alone did not result in any cleavage but , surprisingly , it prevented almost completely MUS81 from cutting the ssDNA flap . Next , we investigated whether RAD52 and MUS81 do physically interact , performing a pull-down assay using purified recombinant RAD52 as bait and HeLa nuclear extracts , which were treated with a nuclease to get rid of any DNA-bridged interactions . As Figure 4D shows , recombinant RAD52 pulled-down MUS81 from the nuclear extracts . As expected , RAD51 , which is known to associate with RAD52 , was also found in the RAD52 pull-down . Collectively , our results suggest that , after inhibition of CHK1 activity , loading of RAD52 in chromatin increased leading to the formation of an intermediate , likely a D-loop , which is cleaved by MUS81 . In the absence of both RAD52 and MUS81 , however , DSBs occur as a result of GEN1-dependent cleavage downstream of RAD51 . These results also suggest that RAD51 and RAD52 may compete at the fork , and that RAD52 may be recruited at collapsed forks independently of MUS81 . MUS81 has been involved in replication restart after prolonged replication inhibition [25] . Having shown that , in the absence of CHK1 function , RAD52 and MUS81 cooperate in the formation of DSBs at stalled forks , and that their function is required to ensure viability of replication-stressed checkpoint-deficient cells , we studied their relationship with restart of such collapsed forks . Using a double CldU/IdU labelling approach on interphase nuclei , we examined the ability of MUS81 or RAD52-depleted cells to restart replication forks [18] . MUS81 or RAD52 down-regulation did not reduce the number of cells that incorporate the first label ( CldU ) as compared to the wild-type cells ( siCtrl and data not shown ) , but differently affected incorporation of the second label ( IdU ) at active replication factories ( Figure 5A and B ) . As expected , CHK1 inhibition severely decreased the ability of HU-treated cells to restart DNA synthesis at stalled forks , as evidenced by the absence of nuclei with more than 60% of CldU/IdU colocalizing foci ( Figure 5A and B ) . A reduction of the ability to incorporate the second label was also observed in MUS81-depleted cells in both unperturbed and HU-exposure conditions ( Figure 5A and B ) . In cells treated with UCN-01 and HU , MUS81 down-regulation did not modify the extent of restart at active replication foci ( Figure 5A and B ) . Interestingly , in RAD52-depleted cells , the increased number of CldU-positive nuclei in which IdU is incorporated at active nuclear foci , demonstrates that HU-stalled forks were recovered in the presence of UCN-01 ( Figure 5A and B ) . Moreover , in UCN-01 and HU-treated cells , RAD52 down-regulation apparently enhanced also the number of nuclei showing only IdU-positive replication foci , which were barely detectable in MUS81-depleted cells ( data not shown ) . Altogether , these results indicate that MUS81 does not contribute to the reduction of fork restart caused by CHK1 inhibition , and that this impairment is mostly dependent on the activation of RAD52 . To address whether RAD52 might also play a MUS81-independent role in cells with a compromised CHK1 function , we evaluated cell death after recovery from replication stress in cells depleted of MUS81 , RAD52 or both , with or without persistent CHK1 inhibition . We found that , in wild-type cells , combined exposure to HU and UCN-01 resulted in a 20% cell death . When cells were allowed to recover in the absence of the CHK1 inhibitor , only a minimal reduction in toxicity was observed ( Figure 6A ) . After CHK1 inhibition , cell death of MUS81-depleted cells increased by two-fold , but decreased significantly when UCN-01 was left during recovery ( Figure 6A ) . Among other enzymes involved in the resolution of intermediates thought to accumulate at collapsed forks , i . e . SLX4 , GEN1 or BLM , only GEN1 depletion increased cell death in CHK1-deficient cells after HU treatment ( Figure S9A ) . RAD52 down-regulation also resulted in enhanced cell death during recovery from UCN-01-induced replication stress . This phenotype , however , was unaffected by persistent CHK1 inhibition during recovery ( Figure 6A ) . Interestingly , the simultaneous inactivation of RAD52/MUS81 was associated with extreme toxicity . Indeed , cells depleted of RAD52 and MUS81 showed about 60% cell death , independently of CHK1 activity during recovery ( Figure 6A ) . Increased cell death of MUS81-depleted cells was also observed following down-regulation of GEN1 or SLX4 , and at a lesser extent after BLM RNAi ( Figure S9B ) , suggesting their involvement in processing intermediates formed at stalled forks in the absence of MUS81 . These results show that MUS81-depleted cells are protected from replication stress induced by CHK1 inhibition only when CHK1 function is restored during recovery . To reinforce this conclusion , we verified whether CHK1 activation was enhanced in MUS81 knock-down cells after recovery from replication stress induced by checkpoint impairment . Following exposure to HU and UCN-01 , CHK1 phosphorylation at Ser345 , a diagnostic readout of its activation , was clearly enhanced ( Figure 6B ) . The levels of Ser345-phosphorylated CHK1 were , however , greatly reduced by MUS81 depletion , suggesting that breakage at collapsed forks is responsible for further checkpoint signalling ( Figure 6B ) . Recovery from replication stress also reduced CHK1 phosphorylation in wild-type cells , even though the residual level of phosphorylated CHK1 is unaffected by UCN-01 removal ( Figure 6B ) . In contrast , phosphorylation level of CHK1 , even though lower , did not decrease during recovery in MUS81 RNAi cells as compared with Ctrl RNAi cells ( Figure 6B ) . Since , in the absence of MUS81 , CHK1 phosphorylation was maintained during recovery from replication stress , we tested the possibility that this was due to prolonged checkpoint activation . However , we found that was not the case ( Figure S2C ) . Thus , we hypothesized that CHK1 activation was required to sustain RAD51 function . We reasoned that overexpression of the phosphomimetic RAD51-T309D mutant in cells depleted of MUS81 would ameliorate viability independently from the presence of UCN-01 during recovery . Consistently , by overexpressing the RAD51-T309D mutant , cell survival was strikingly increased during recovery from replication arrest , irrespectively of CHK1 inhibition ( Figure 6C ) . In contrast , overexpression of wild-type RAD51 protein did not modify the toxic response induced by MUS81 depletion , and its dependence on CHK1 inhibition ( Figure 6C ) . Finally , overexpression of the RAD51-T309D mutant did not rescue the elevated cell death observed in the RAD52 or in the RAD52/MUS81-depleted cells , but instead exacerbated the phenotype of the double knock-down cells ( data not shown ) . Since , in human cells , RAD52 may be required for RAD51 chromatin loading under perturbed replication [26] , we analysed whether RAD52 down-regulation affected RAD51 recruitment after CHK1 inhibition and HU treatment . Analysis of chromatin fractions from cells treated with HU or UCN-01 showed that the amount of RAD51 in chromatin did not increase , and only a small reduction was observed after a concomitant treatment ( Figure 6D ) . We observed that the amount of RAD51 chromatin-bound is higher in HU-treated MUS81 knock-down cells than in Ctrl RNAi cells , but decreases after a combined exposure of HU and UCN-01 ( Figure 6D ) . Interestingly , the loading of RAD51 was unaffected by RAD52 depletion in untreated cells , and was greatly increased after UCN-01 or HU treatment ( Figure 6D ) . Co-depletion of RAD52 and MUS81 determined also a strong increase in the level of chromatin-associated RAD51 after replication arrest by HU ( Figure 6D ) , which was maintained in the presence of the CHK1 inhibitor ( Figure 6D ) . Elevated chromatin accumulation of RAD51 and the hypersensitivity of the RAD52/MUS81 knock-down cells to replication stress induced by the a combined HU and UCN-01 exposure , prompted us to verify whether exacerbated cell death was related to the inability to properly process RAD51-dependent intermediates . To this aim , we analysed whether RAD51 depletion could improve viability of the double RAD52/MUS81 knock-down cells after replication stress . Interestingly , concomitant depletion of RAD51 , RAD52 and MUS81 ( Figure 6E ) reduced significantly cell death in cells treated with UCN-01 , alone or in combination with HU , irrespectively of the presence of CHK1 activity during recovery ( Figure 6F ) . The severe phenotype of the RAD52/MUS81 double-depleted cells was also ameliorated by GEN1 down-regulation , but not by SLX4 depletion ( Figure S10A ) . In contrast , GEN1 down-regulation did not rescue , but rather reduced , viability of the RAD52 single-depleted cells ( Figure S10B ) . Our results indicate that a CHK1-regulated RAD51 function can prevent the MUS81-dependent cell death derived from replication stress , induced by CHK1 inhibition . Moreover , our findings also suggest that loss of RAD52 engages a RAD51-dependent recovery in which MUS81 may play an important and additional function , together with BLM , to clear potentially toxic intermediates . We have previously shown that loss of MUS81 increases chromosomal damage in WRN-deficient cells , whereas it decreases chromosome abnormalities upon oncogene-induced replication stress [14] , [18] . In both cases , however , MUS81 down-regulation increases cell death as we observed in replication checkpoint-deficient cells . Thus , we investigated whether MUS81 down-regulation enhanced chromosomal instability in CHK1-inhibited cells . To this end , we induced replication stress by concomitant CHK1 inhibition and HU treatment , and analysed chromosomal damage in metaphase cells after recovery in the absence of UCN-01 to limit cell death . Given that double knock-down cells were extremely sick , we limited the analysis of chromosome damage in Ctrl , MUS81 or RAD52 RNAi-treated cells ( Figure 7A–C ) . A combined UCN-01 and HU treatment resulted in a significant increase of chromosome aberrations , mainly chromosome breaks , compared to HU-treated cells ( Figure 7A , C ) . Depletion of MUS81 enhanced chromosome damage in cells treated with HU alone , while barely affected genome instability caused by the combined treatments ( Figure 7B , C ) . RAD52 down-regulation resulted in increased levels of chromosomal damage in both unperturbed and HU-exposure conditions . In contrast , a significant reduction in chromosome aberrations was observed in cells treated with UCN-01 and HU ( Figure 7A , C ) . These results indicate that most of the chromosomal damage , resulting from checkpoint failure , is the result of the engagement of RAD52 and MUS81 at collapsed forks , which would grant viability , but at the expense of genome stability .
One of the essential functions of the replication checkpoint is to maintain integrity of replication forks when they undergo pausing or stalling . Consistently , studies from model organisms and human cells with impaired replication checkpoint activity have shown elevated levels of collapsed forks and DSBs accumulation after replication perturbation [27] , [28] , [29] , [30] , [31] , [32] . Our data indicate that MUS81 is responsible for the creation of DSBs after depletion of some crucial components of the replication checkpoint , extending recent findings indicating that MUS81 acts in the cells with mutant WRN protein , or following replication stress induced by oncogene activation or camptothecin treatment [14] , [18] , [25] , [33] . In fission yeast , an analogous MUS81-dependent formation of DSBs has been reported [11] , suggesting that the function of MUS81 at the collapsed forked is conserved across species . Despite the number of functions that the replication checkpoint fulfils upon replication stalling [1] , here we demonstrate that loss of CHK1 activity is sufficient to cause MUS81-mediated DSBs . This agrees with earlier observations showing that depletion of ATR , RAD9 or TOPBP1 results in a reduced CHK1 phosphorylation [20] , [34] , [35] , [36] , and DSBs suppression following MUS81 down-regulation in unperturbed CHK1-inhibited cells [37] . Although it has been shown that a proper CHK1 activation requires the presence of additional factors , such as TIPIN [38] , [39] , we observed no accumulation of MUS81-dependent DSBs in TIPIN knock-down cells after HU treatment . Given that , in our cell model system , TIPIN down-regulation does not reduce CHK1 phosphorylation , and since TIPIN/TIM-deficient cells retain the ability to sustain CHK1 activation [40] , it is possible that even a reduced amount of active CHK1 is sufficient to protect from MUS81-dependent DSBs . It has been recently reported that DSBs induced by MUS81 are detrimental to cell survival in CHK1 inhibited cells [37] . In contrast , we show that MUS81-dependent DSBs are essential to limit cell death upon replication stress induced by HU treatment and CHK1 inhibition . Moreover , another structure-specific endonuclease , GEN1 , is also required to prevent excessive cell death in CHK1-inhibited cells experiencing replication arrest . However , almost all the DSBs formed are MUS81-dependent under our experimental conditions . Thus , the observed pro-survival role of GEN1 might be related to resolution of late homologous recombination ( HR ) intermediates , rather than to cleavage at collapsed forks , as recently proposed [41] . These results are not necessarily in conflict with the observations of Forment and colleagues , since only untreated cells have been analysed . Moreover , our results are in agreement with previous reports indicating that DSBs produced by MUS81 are required to allow replication recovery , and viability , under different conditions resulting in fork collapse [11] , [14] , [25] , [33] . In yeast , replisome stabilization requires the functional homolog of the human CHK1 , Rad53 . In the absence of Rad53 , stalled replisomes collapse and replication intermediates become vulnerable to degradation by exonucleases and endonucleases [5] , [6] . Thus , CHK1 inactivation may be instrumental for replication fork collapse also in humans . How CHK1 may contribute to stalled fork stabilization remains enigmatic , however , the main human recombinase RAD51 might have a pivotal and early role in this process [22] , [42] . Since CHK1 phosphorylates RAD51 at T309 [17] , DSBs generated upon replication stress induced by UCN-01 treatment could stem from loss of RAD51 function . Our results indicate that , even though RAD51 depletion results in formation of DSBs in unperturbed cells [43] , these are MUS81-independent . This suggests that CHK1-dependent protection of perturbed forks from DSBs is unrelated to RAD51 phosphorylation . Indeed , neither RAD51 depletion nor the expression of an unphosphorylatable RAD51-T309A mutant , is sufficient to induce MUS81-dependent DSBs . Moreover , it is unlikely that cleavage by MUS81 is a consequence of combined loss of CHK1-dependent fork protection and CHK1-regulated RAD51 loading at distressed forks . Indeed , over-expression of a phosphomimetic RAD51-T309D mutant is not sufficient to revert MUS81-dependent DSBs after treatment with UCN-01 ( our unpublished results ) . Alternatively , CHK1 might actively prevents targeting of collapsed forks by MUS81 , as reported in fission yeast [44] . However , we detected no signs of a CHK1-dependent phosphorylation of MUS81 that can be reduced upon replication arrest by CHK1 inhibition . Further experiments are needed to clarify this point , which is outside the scope of this work . It is generally thought that MUS81 may cleave RAD51-dependent recombination intermediates , such as HJs , mainly outside DNA replication . The identity of the MUS81 substrate and how it is generated at perturbed forks , however , remains unresolved [10] , [12] , [14] , [45] , [46] , [47] , [48] . Our data suggest that , upon fork collapse , MUS81 does not target a RAD51-dependent recombination intermediate , as similarly reported in WRN-deficient cells [21] . In fact , RAD51 depletion does not stimulate or revert formation of MUS81-dependent DSBs upon CHK1 inhibition , suggesting that MUS81 either targets the stalled forks directly , or processes other intermediates that form independently of RAD51 . Since we observe that CHK1 inhibition and HU treatment stimulate RAD52 binding to chromatin , and that RAD52 depletion abrogates MUS81-dependent DSBs , we conclude that MUS81 does not cleave collapsed forks directly , but rather after the formation of a RAD52-dependent intermediate . It is worth noting that remodelling of collapsed forks prior to MUS81-dependent cleavage , might explain why DSBs are not formed immediately after replication arrest . Interestingly , MUS81-dependent DSBs accumulate after that CHK1 inhibition has induced a large amount of ssDNA , suggesting that MUS81 is cleaving an intermediate assembled from unreplicated leading or lagging strand . One of the substrate that could be generated by the ssDNA annealing activity of RAD52 , perhaps through assistance of an helicase , is a D-loop , which is an ideal substrate for MUS81 . Indeed , our in vitro studies support this hypothesis , and also demonstrate that MUS81 specifically targets D-loops assembled by RAD52 . The apparent inability of MUS81/EME1 to cleave D-loops produced by RAD51 , provides a mechanistic understanding of the RAD51 independency showed by DSBs formed by MUS81 in vivo . This conclusion is further reinforced by data in yeast , showing that MUS81 may act on RAD52-dependent D-loops produced at collapsed forks [12] . Such a D-loop might result from either pairing of the extruded leading or lagging strand after fork regression , or by the attempt to repair a ssDNA gap behind the replication fork ( see Figure 8 ) . Formation of D-loop by RAD52 requires its ssDNA annealing activity , probably associated with SUMO-conjugation [49] . Interestingly , we notice that CHK1 inhibition determines a striking accumulation of high-molecular-weight forms of RAD52 , which may correspond to SUMO-modification of RAD52 ( Figure 3C ) . Thus , it is likely that replication distresses , induced in CHK1-deficient cells , elevate DNA annealing activity of RAD52 , which would correlate with D-loop formation . Alternatively , enhanced RAD52 SUMOylation could be a consequence of impaired RAD51 loading , secondary to CHK1 inhibition . Indeed , in yeast rad51 mutants , increased SUMOylated Rad52 has been found [50] . However , our observations that chromatin loading of RAD51 is not reduced in CHK1-inhibited cells , and that RAD51 down-regulation does not result in MUS81-dependent DSBs , favour the first hypothesis . DSBs produced at distressed replication forks may be channelled to the BIR pathway , which can occur in a way dependent on RAD51 or RAD52 [51] , [52] . Interestingly , inhibition of RAD51-dependent strand-invasion , during recovery from the combined UCN-01 and HU treatment , does not affect the level of DSBs . Thus , it is likely that DSBs produced by MUS81 in response to CHK1 inhibition , triggers a RAD52-dependent BIR pathway . Since RAD51-dependent or independent BIR events are expected to produce different intermediates [51] , it is not surprising that BLM , which processes double HJs , does not seem particularly important for viability under our experimental conditions . Recent reports have evidenced that RAD52 might contribute to survival of BRCA2-deficient cells , promoting repair of DNA damage arising in cells with defective RAD51 loading [53] . Our data indicate that chemical inhibition of RAD51 , or its improper loading as it may occurs in BRCA2-deficient cells , minimally affects formation of MUS81-dependent DSBs or results in excessive MUS81-dependent DSBs . The most likely explanation is that DNA transaction induced at stalled forks when CHK1 is inhibited is peculiar , and does not occur when RAD51 is not functional but CHK1 is still active . Even though our observations support a cooperation between RAD52 and MUS81 in response to replication stress , their synergistic effect on viability suggests that these proteins have also independent functions , consistently with the observed milder phenotype of RAD52 knock-down cells , respect to MUS81 or RAD51-depleted cells . Absence of MUS81-dependent DSBs in BRCA2-defective cells may be also related to the function of BRCA2 in protecting the stalled forks from MRE11-dependent degradation [54] . Accordingly , in the absence of BRCA2 , MRE11-dependent degradation could prevent formation of the MUS81 substrate . Indeed , increased levels of ssDNA are detected in MUS81 knock-down cells , which may be related to further exonucleolytic cleavage at distressed forks to favour a RAD51-dependent pathway . We show that recovery from replication stress of MUS81-depleted cells requires a CHK1-regulated RAD51 , a phenotype that we do not observe after RAD52 depletion . Strikingly , elevated RAD51 foci was reported in MUS81-null MEFs upon spontaneous fork collapse [25] , and we show a reduced viability of MUS81-depleted cells after down-regulation of BLM , SLX4 or GEN1 , all targeting uninterrupted branched intermediates , most likely generated downstream RAD51 . From this point of view , it is possible that structures left unprocessed by MUS81 are then channelled back into a RAD51-dependent recombination to ensure viability , perhaps with the help of RAD52 and CHK1 ( see Figure 8 ) . Also depletion of RAD52 , preventing the formation of the MUS81 substrate , would channel collapsed forks to a RAD51 route , as suggested by the strong accumulation of RAD51 on chromatin . Most importantly , concomitant depletion of RAD52 and MUS81 gives a similar increase in the amount of chromatin-bound RAD51 , but also results in a strong reduction of viability . Interestingly , poor viability of RAD52/MUS81 depleted cells after checkpoint inactivation is ameliorated by RAD51-depletion . Similarly , viability of the RAD52/MUS81-depleted cells is improved by knock-down of GEN1 , which is responsible for the DSBs observed in this background . Thus , it is likely that loss of RAD52 , precluding the formation of the MUS81 substrate , determines the formation of a RAD51-dependent intermediate that should be normally processed by MUS81 , which is also absent , and becomes toxic after GEN1 cleavage . Since in yeast Yen1 ( GEN1 ) may substitute for Mus81 during repair of DSBs at perturbed forks [55] , our results might suggest that such relationship is not valid in human cells . However , if RAD52 is present , such “redundancy” between MUS81 and GEN1 can be observed also in our hands ( see Figure S9 ) . Thus , it is likely that intermediates accumulating downstream of RAD51 become toxic once they are cleaved by GEN1 , because of the absence of RAD52 or of loss of CHK1-mediated regulation . Given the well-known difference in substrate specificity between MUS81 and GEN1 , with the latter preferentially acting on the uninterrupted intermediates [56] , it is possible that these types of structures ( i . e . single HJ or reversed forks ) are accumulating in the RAD52/MUS81 double-depleted cells . Even though MUS81 is important to limit cell death , its depletion minimally affects chromosomal damage in checkpoint-deficient cells ( Figure 7 ) . However , depletion of MUS81 rescues the instability occurring at common fragile sites after oncogene expression or under unperturbed conditions [18] , [41] , [57] . In contrast , depletion of RAD52 reduces genome instability in the absence of CHK1 . Since RAD52 down-regulation is expected to prevent formation of the MUS81 substrate , it is tempting to speculate that genome instability occurring upon checkpoint failure mostly depends on MUS81 cleavage , or further processing of the uncleaved substrate ( see Figure 8 ) . Depletion of RAD52 also rescues the ability of stalled forks to restart under CHK1 inhibition . This might be related to a forced switch to the RAD51-dependent BIR pathway that is expected to be faster than that , RAD51-independent , engaged in wild-type cells , as proposed in yeast [58] , [59] . Interestingly , depletion of RAD52 , almost specifically , improves the ability of stalled forks to restart following CHK1 inhibition , i . e . after checkpoint inactivation . This observation , together with the evidence that , in our experimental condition , CHK1-inhibited cells are blocked in S-phase , might suggest that preferential engagement of a RAD52-dependent pathway is linked to an attempt of cells to activate a checkpoint response bypassing CHK1 inhibition , as speculated for BIR in yeast [58] . Our observation that RAD52 depletion also results in more de novo origin firing , is consistent with a “checkpoint-like” function of the RAD52-MUS81 pathway , however , additional studies are necessary to confirm such intriguing hypothesis . Altogether , our results show that MUS81 is responsible for the generation of DSBs after replication stress induced when CHK1 activity is impaired . They also demonstrate that the generation of MUS81-dependent DSBs is not the consequence of an altered function of RAD51 , but depends on the presence of RAD52 . Furthermore , integrity of this RAD52/MUS81-dependent mechanism is critical for cell viability under replication stress , and its loss engages toxic RAD51-dependent transactions . Given that replication stress has been associated with cancer progression , and since CHK1 inhibitors are considered in anticancer therapy , our findings may also improve an educated approach to replication checkpoint inhibition in cancer cells , by capitalizing on potential synergistic effects and reduced functionality of the recombination factors . It is clear , for example , that inhibition of RAD52 in CHK1-deficient cells should reduce the viability of cancerous cells , and should prevent genetic instability and thereby the risk of resistance emergence .
The GM01604 hTERT-immortalised normal human fibroblasts , the 293T cells and the WI-38 SV40-transformed normal human fibroblasts were obtained from Coriell Cell Repositories ( Camden , NJ , USA ) or American Type Culture Collection ( Manassas , VA , USA ) . Cells were cultured in Dulbecco's modified Eagle's medium ( DMEM; Life Technologies ) supplemented with 15% FBS ( Boehringer Mannheim ) for the GM01604 fibroblasts and 10% FBS for WI-38 fibroblasts and 293T cells . All the cells were incubated at 37°C in a 5% CO2 atmosphere . HU and BrdU were obtained from Sigma-Aldrich . HU was dissolved in sterile PBS as a stock solution ( 200 mM ) and stored at +4°C . BrdU was dissolved in sterile PBS as a stock solution ( 3 mg/ml ) and stored at −20°C . UCN-01 ( Alexis Biochemicals ) was used at 400 nM concentration to inhibit CHK1 activity , while to inhibit ATR activity the ETP-46464 compound ( a gift of Dr . Fernandez-Capetillo ) was used at 10 µM concentration . The specific RAD51 inhibitor B02 was from Merck chemicals , and was used at 27 µM according to Huang et al [60] . MUS81 , CHK1 , CHK2 , ATR , RAD9 , TOPBP1 , TIPIN , RAD51 , BLM , SLX4 , GEN1 and RAD52 expression were knocked down by transfection with SMARTpool siRNAs ( Dharmacon ) directed against proteins of interest at the final concentration of 10 nM . Transfection was performed using Interferin ( Polyplus ) according to the manufacturer's instructions . As a control , a siRNA duplex directed against GFP was used . For genetic complementation experiments , cells were first transfected with a mix of two distinct siRNA oligos targeting the UTR region of human RAD51 ( Qiagen Flexi tube; cat# SI00045010 and SI02629837 ) at a 10 nM concentration using HiPerfect reagent ( Qiagen ) and then nucleofected using the Amaxa device ( Kit #VACA-01 ) to express RNAi-resistant wild-type ( RAD51wt ) , phosphorylation-defective ( RAD51-T309A ) , or a phosphomimetic ( RAD51-T309D ) mutant form of RAD51 ( see Text S1 ) . Nucleofection of plasmids a was performed using 2 µg of supercoiled DNA according to the manufacturer's instructions . Western blot and chromatin fractionation were performed as described in Franchitto et al [14] . Blots were incubated with primary antibodies against: MUS81 ( Abcam ) , SLX4 ( Abcam ) , Phospho-Ser345-CHK1 ( Cell Signaling ) , CHK1 ( Santa Cruz Biotechnology ) , BLM ( Santa Cruz Biotechnology ) , CHK2 ( Calbiochem ) , ATR ( Calbiochem ) , RAD9 ( Calbiochem ) , TOPBP1 ( Bethyl ) , TIPIN ( Bethyl ) , RAD51 ( Abcam ) and RAD52 ( Santa Cruz Biotechnology ) , PCNA ( Santa Cruz Biotechnology ) , Tubulin β ( Sigma-Aldrich ) , phoshpo-H3 histone ( Santa Cruz Biotechnology ) and Lamin B1 ( Abcam ) . The anti-GEN1 antibody was a kind gift of Prof . Yungui Yang ( Beijing Genomics Institute ) . After incubations with horseradish peroxidase-linked secondary antibodies ( Vector Laboratories ) , the blots were developed using the chemiluminescence detection kit ECL-Plus ( Amersham ) according to the manufacturer's instructions . To immunopurify the human MUS81/EME1 complex , 293T cells were transiently transfected with a 1∶1 ratio of plasmids expressing the Myc-tagged MUS81 and the GST-tagged EME1 ORFs . Transfection was performed using the Dreamfect reagent ( Ozbiosciences ) according to the manufacturer's direction . Cells were collected 60 h after transfection and nuclear pellets stored frozen for subsequent immunopurification . For immunopurification , nuclei obtained from 5×107 cells were lysed in CSK buffer ( 200 mM NaCl , 300 mM sucrose , 3 mM MgCl2 , 1 mM DTT , 10 mM PIPES - pH 6 . 8 ) containing 0 . 5% Triton X-100 , protease inhibitors and benzonase . After removal of nuclear debris by centrifugation , cleared lysate was incubated with 0 . 3 ml of agarose beads conjugated with anti-Myc antibodies under rotation . After incubation , the beads were extensively washed with TNT buffer ( 50 mM Tris/Cl buffer pH 7 . 6 , containing 300 mM NaCl , 0 . 5% Triton X-100 , 2 mM DTT , 3 mM MgCl2 and protease inhibitors ) . After washing , beads were incubated under rotation with the Myc peptide to elute the MUS81/EME1 complex . One tenth of the eluate was analysed for the presence of the MUS81/EME1 complex using Coomassie staining and the remaining volume of eluate was further purified by incubation with GSH-agarose to capture the intact MUS81/EME1 heterodimer . Finally MUS81/EME1 heterodimer was retrieved by elution using PBS containing 25 mM Glutathione , concentrated and the amount and purity of the complex estimated by SDS-PAGE followed by Coomassie staining . The occurrence of DNA double-strand breaks was evaluated by neutral Comet assay as described [61] . Alternatively , cells were subjected to Comet assay under alkaline conditions to detect both DSBs and single-stranded DNA gaps or nicks . Cell DNA was stained with ethidium bromide ( Sigma ) and examined at 40× magnification with an Olympus fluorescence microscope . Slides were analyzed by a computerized image analysis system ( Comet IV , Perceptive UK ) . To assess the amount of DNA damage , computer-generated tail moment values ( tail length×fraction of total DNA in the tail ) were used . A minimum of 200 cells was analysed for each experimental point . Apoptotic cells ( smaller comet head and extremely larger comet tail ) were excluded from the analysis to avoid artificial enhancement of the tail moment . To determine whether RAD52 and RAD51 produce the MUS81/EME1 cleavable structures we generated the D-Loops produced by these proteins . Human RAD51 and RAD52 proteins were purified as described in [62] , [63] , [64] . The concentration of the proteins were determined using their molar extinction coefficients; 12 , 800M−1cm−1 ( RAD51 ) and 40 , 380 M−1cm−1 ( RAD52 ) . RAD51- or RAD52-mediated D-loops were produces essentially as described in [65] . Briefly , 20 nM ( molecules ) γ-P32-labeled ssDNA oligonucleotide ( 5′-ATT TTG TTC ATG GTA GAG ATT CTC TTG TTG ACA TTT TAA AAG AGC GTG G-3′ ) was incubated with 1 µM RAD52 or RAD51 protein at 37°C for 7 minutes in the reaction buffer ( 50 mM HEPES-NaOH ( pH 7 . 5 ) , 5 mM MgCl2 , 100 mM NaCl2 , 0 . 1 mg/ml BSA and 1 mM DTT ) . The RAD51 reaction also contained 1 mM ATP . D-loop formation was initiated by addition of 10 nM ( molecules ) of φX174 RFI supercoiled dsDNA followed by incubation at 37°C for 20 min . The protein-free D-loops were produced by mixing the oligonucleoted with the supercoiled dsDNA in the reaction buffer , heating the mixture to 95°C and then slowly cooling the mixture to room temperature . The MUS81/EME1 complex was then added to the D-loops at the indicated concentrations and the reactions were further incubated for 20 or 60 min at 37°C . The reaction was stopped by adding 1 µl of 10% SDS , followed by immediately adding 1 µl of 10 mg/ml proteinase K and incubation at 37°C for 30 min . 6X DNA Gel Loading Dye ( Thermo ) was added to the reaction and the samples were resolved on the 0 . 8% agarose gel in 1X TAE Buffer at 5 V/cm at room temperature . The reaction products were visualized using a phosphorimager system ( GE Healthcare ) . D-loops were quantified using ImageJ software [66] . Substrates representing 3′-flap ( JLBD20 ) and nicked product ( LMBD20 ) were produced by annealing the following oligonucleotides: ( GGATGGCTTAGAGCTT AATTCCGCTCATGGATGCTATCACGC ) , L ( CGTACTGCAATCTTGAACCG-Cy5-GGAA TTAAGCTCTAAGCCATCC ) , M ( GGATGGCTTAGAGCTTAATTCC ) and BD20 ( CGGTTC AAGATTGCAGTACG , by incubating the Cy5-labeled oligo L with 1 . 5-fold excess of the two unlabeled nucleotides in T50 buffer ( Tris-HCl pH 7 . 5 , 50 mM NaCl ) at 95°C and gradually cooling to 25°C in a dry bath over 4 hours . The substrate is one of those preferred by MUS81 as described in Ciccia and colleagues [8] . The reactions contained nuclease buffer ( 50 mM Tris-HCl pH 7 . 5 , 100 mM NaCl , 5 mM MgCl2 , 20 mM Glycine , 2 mM Dithiothreitol ) , 10 nM JLBD20 oligo in reaction buffer , and 100 nM of MUS81-EME1 , RAD52 or both proteins . The reaction mixtures were incubated at 37°C for 90 minutes and stopped by adding 0 . 5% SDS , followed by immediately adding 0 . 3 mg/ml proteinase K and incubation at 37°C for 30 minutes . The samples were then separated on a 15% non-denaturing TBE-PAGE gel and analyzed with Cy5 detection using the BioRad Chemidoc system . The percentage of nicked product was quantified using ImageJ software . To determine whether RAD52 associates with MUS81 we performed pull-down experiments using purified His-tagged RAD52 ( see above ) as bait and HeLa nuclear extract ( NE ) as source of the pray . Briefly , 5 µg of recombinant RAD52 was incubated overnight with 1 mg of NE in binding buffer ( Tris/Cl buffer pH 7 . 6 containing 150 mM NaCl and 0 . 5% Triton X-100 ) . One-fiftieth of the NE was put apart to be used as input . The pull-down material was then incubated for 1 h at RT with 4 µg of anti-His antibody-coupled Protein G to capture RAD52 complexes , and after extensive washing in binding buffer , proteins were released by incubation in 1× Laemmli sample buffer . DNA replication sites were visualized by incorporation of chlorodeoxyuridine ( CldU ) and iododeoxyuridine ( IdU ) into DNA . GM01604 cells were transfected with siRNAs directed against GFP ( control ) or against MUS81 or RAD52 , and 48 h thereafter treated for 6 h with 400 nM UCN-01 alone or in combination with 2 mM HU . The CldU label ( 25 µM ) was added 10 min before treatments and after 6 h cells were washed extensively and labelled with 200 µM IdU for 45 min . Cells were then washed with PBS , fixed with cold 70% ethanol , and stored at 4°C . Antibody staining was performed as previously reported [18] . Images were acquired as greyscale files using Metaview software ( MDS Analytical Technologies ) and processed using Adobe Photoshop CS3 ( Adobe ) . For each time point , at least 200 nuclei were examined by two independent investigators and foci were scored at 60× . GM01604 cells were transfected with siRNAs directed against GFP ( control ) , or against MUS81 , CHK1 , ATR , TIPIN , RAD51 , SLX4 , BLM , GEN1 and RAD52 ( Qiagen ) . Viability was evaluated by the LIVE/DEAD assay ( Sigma-Aldrich ) according to the manufacturer's instructions . Cell number was counted in randomly chosen fields and expressed as percent of dead cells ( number of red nuclear stained cells/total cell number ) . For each time point , at least 200 cells were counted . WI-38 SV40-transformed fibroblasts were transfected with siRNAs directed against GFP ( siCtrl ) , MUS81 ( siMUS81 ) or RAD52 ( siRAD52 ) . Forty-eight hours after interference , cells were treated for 6 h with 2 mM HU or pre-treated for 1 h with 400 nM UCN-01 and then 6 h together with HU . At the end of treatments , all the cells were recovered in drug-free medium for 21 h . Cell cultures were incubated with colcemid ( 0 . 2 µg/ml ) at 37°C for 3 h until harvesting . Cells for metaphase preparations were collected and prepared as previously reported [67] . The analysis of chromosomal aberrations was performed by scoring at least 100 Giemsa-stained metaphases per experimental point . | The replication checkpoint ensures a smooth duplication of the genome . It counteracts the replication stress , which can cause chromosome rearrangements as found in most tumours . Given the importance of dealing with perturbed replication , and since in tumours secondary mutations or epigenetic changes may hamper efficiency of the replication checkpoint , it is crucial to determine the mechanisms responding to replication perturbation upon checkpoint inactivation . Furthermore , it is highly relevant to understand how failure of these mechanisms correlates with chromosomal damage after replication perturbation . Here , we investigated pathways that , in checkpoint-deficient human cells , are involved in the handling of perturbed DNA replication forks , and we uncovered a previously unappreciated function of RAD52 and MUS81 in ensuring viability of cells , but at the expense of genome instability . We also demonstrated that checkpoint deficiency can trigger different mechanisms of recovery from replication arrest depending on the presence of RAD52 or MUS81 , resulting in a poor survival and reduced genome instability or increased survival and chromosomal damage . Our work provides new clues about how human cells deal with replication stress , and how genome instability may arise in cancer cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Survival of the Replication Checkpoint Deficient Cells Requires MUS81-RAD52 Function |
In the olfactory bulb , lateral inhibition mediated by granule cells has been suggested to modulate the timing of mitral cell firing , thereby shaping the representation of input odorants . Current experimental techniques , however , do not enable a clear study of how the mitral-granule cell network sculpts odor inputs to represent odor information spatially and temporally . To address this critical step in the neural basis of odor recognition , we built a biophysical network model of mitral and granule cells , corresponding to 1/100th of the real system in the rat , and used direct experimental imaging data of glomeruli activated by various odors . The model allows the systematic investigation and generation of testable hypotheses of the functional mechanisms underlying odor representation in the olfactory bulb circuit . Specifically , we demonstrate that lateral inhibition emerges within the olfactory bulb network through recurrent dendrodendritic synapses when constrained by a range of balanced excitatory and inhibitory conductances . We find that the spatio-temporal dynamics of lateral inhibition plays a critical role in building the glomerular-related cell clusters observed in experiments , through the modulation of synaptic weights during odor training . Lateral inhibition also mediates the development of sparse and synchronized spiking patterns of mitral cells related to odor inputs within the network , with the frequency of these synchronized spiking patterns also modulated by the sniff cycle .
Lateral inhibition is one of the critical mechanisms underlying responses to sensory neurons [1] , but the detailed mechanisms at the network level in the olfactory system are not clear [e . g . 2] . In the Limulus eye [1] and the cat retina [3] it mediates contrast enhancement between areas of differing illumination . It has also been found in the auditory pathway ( reviewed in [4] ) and the somatosensory system [5] . In the olfactory system , the clearest evidence for lateral inhibition is the interaction between mitral cells in the olfactory bulb , mediated through inhibitory granule cells [6]–[7] and periglomerular cells [8] . The possible underlying circuits and their computational properties have been widely investigated experimentally [9]–[11] especially in terms of odor selectivity and dynamics of mitral cell responses [12]–[14] . A major problem in interpreting the experimental findings in vivo is that they are usually obtained in single cells or in small randomly selected sets of cells , whereas a clear understanding of fundamental processes , such as the spatio-temporal organization of the mitral-granule cell network , requires simultaneous recording from a relevant subset of cells activated by any given odor . The functional effects of network-wide processes , in relation to the patterns of glomeruli activated by different odors , therefore remain relatively unknown and extremely difficult to explore experimentally . To gain insight into this problem we have focused on mitral-granule cell interactions , the best understood circuit in the olfactory bulb . For this purpose we have constructed a biophysical network model of multicompartment mitral and granule cells with connections similar to those in the real olfactory bulb . As input we have used the activation of individual glomeruli by a large set of odors identified by intrinsic imaging [15] . This model has allowed us to investigate several fundamental questions: 1 ) How does the network self-organize and modulate mitral cell responses to different odor molecules ? Lateral inhibition has been suggested to be the primary mechanism . However , experimentally the focus is almost exclusively on individual mitral and granule cell interactions , and whether lateral inhibition is able to shape network-wide connectivity is not known . 2 ) How does the precise timing of mitral cell action potentials subject to lateral inhibition relate to the sniff cycle ( as shown in [16] ) ? This appears to be one of the critical processes affecting responses in awake mice [2] , but the mechanisms responsible for the firing behavior at the network level are not understood . 3 ) Why is the olfactory bulb network connectivity sparse and distributed ( as shown in [10] and [17] ) ? There are no experimental data for describing the underlying mechanism . We have previously proposed a physiologically plausible process using a small network with simulated odors and all-to-all possible connectivity [18] . Its validity for more realistic odor inputs , network size , and intrinsically sparse connectivity required testing , which was carried out in the present study . The results show that: 1 ) Lateral inhibition mediated through broadly tuned granule cells results in mitral cell output that reflects the spatial distribution of the glomerular input and the temporal structure of the inhibitory processing . Together , the mitral-granule cell circuit operates to generate a unique spatiotemporal representation for each odor . 2 ) Odor identity can emerge from a single sniff as a specific distribution of spikes in which each mitral cell makes its own contribution according to the specific type of input it receives and the network of granule cells it activates . 3 ) The mitral<->granule cell interactions through dendrodendritic synapses can account for the distributed network connectivity observed experimentally . We discuss how these specific predictions from the model provide new hypotheses for experimental testing .
In the experiments , 72 different odor molecules were used for stimulation . Individual glomeruli in each of the clusters illustrated in Fig . 1B were differentially activated by these odors , as shown in the table of Fig . 2 . Each odor induced activity in a group of glomeruli , usually belonging to the same cluster but often with outliers . Mori et al . ( 2006 ) [15] classified these responses into 4 different intensity levels: very strong , strong , moderate and weak , represented by the different size circles in their table . Since it can be assumed that the intensities reflected the postsynaptic dendritic depolarization of the mitral cell glomerular tufts [29] , in our model we used the response levels to set 4 different levels for the peak synaptic excitatory conductance activated in the glomerular tufts . The aggregate activation of many OSN inputs onto a given tuft with a single EPSP was implemented using a double exponential conductance change with 20 and 200 ms for rise and decay time , respectively [18] , [30] . To represent the range of intensities with adequate sensitivity down to the weakest concentration without saturating the network at the highest concentration , we set the peak conductance sensitivity to give suprathreshold responses to levels 3 and 4 . We simulated each of the 72 odor responses [15] in Fig . 2 in order to analyze and compare the network responses . Key aspects of the network properties are illustrated by three examples , highlighted in the table in Fig . 2: a relatively strong glomerular response ( to octanal , in cluster B ) , a relatively widely distributed odor response ( to k3-3 , in clusters B and C-D ) , and a relatively weak response ( to ( + ) -Cvn , in clusters B and D ) . Histograms in Fig . 2B show the relative strengths and distribution of the glomerular responses , which are the input magnitudes used to activate the mitral cells . The network self-organization in these representative cases is illustrated in Fig . 3 , where we show the raster plots for mitral and granule cell spike discharges during the first seven seconds of odor presentation for three odors . During a moderate glomerular input ( e . g . Fig . 3A , odor ( + ) -Cvn ) , the spiking response dynamics of the most active mitral cells ( around site 240 ) showed the progressive appearance of a bursting pattern , accompanied by weak inhibition of surrounding mitral cells to an extent of approximately 1 . 5 mm on either side of site 240 ( note the lighter area in the mitral cells raster plot after the first 2 sec of simulation ) , reflecting the extent of the lateral dendrites of the activated mitral cells . Bursts were aligned with odor input activation ( sniffs ) , with most spikes at the onset of the sniff and fewer spikes later in the sniff cycle . The mitral cell inhibition was correlated with activation of the granule to mitral cell inhibitory synapses , as shown by the firing of granule cells in Fig . 3A ( right panel ) , again reflecting the 1 . 5 mm extent of the mitral cell lateral dendrites . This granule cell spiking produced the mitral cell inhibition , as evidenced by the similarity between the granule cell spiking population and the extent of the mitral cell inhibition in Fig . 3A . These results demonstrate several basic properties of the network response to a glomerular activity pattern . As expected , the odor drives the mitral cells receiving direct input from the activated glomeruli . This defines the mitral cell cluster related to the activated glomerular cluster . A new property shown by this realistic simulation is that the action potentials in the lateral dendrites of the activated mitral cells bring about synaptic excitation of the connected granule cells , which elicits spiking in these cells . The sharp cutoff of the granule cell spiking at 1 . 5 mm on either side of site 240 provides a novel indication of precisely the extents of the lateral dendrites of the driven mitral cells . This spatial pattern may be considered a 1-dimensional representation of 2-dimensional lateral , or surround , inhibition . The surround inhibition in this case is relatively weak because of the relatively weak input and the consequent weak granule cells activity . These results thus indicate an unexpectedly extensive engagement of the granule cells in the dorsal olfactory bulb area by activation of just a few glomeruli . As we will discuss later , this is not in contrast with the discontinuous mosaic of clusters of mitral-granule cell “glomerular units” shown by tracing experiments [10] but , rather , it suggests possible constraints on the transsynaptic virus transport mechanisms . Very strong localized odor activation produced much stronger network responses . As shown for octanal , in Fig . 3B , the activated mitral cells ( sites 430–490 ) developed an intense intermittent bursting pattern . In comparison with ( + ) -Cvn , the lateral inhibition was much stronger , as shown by the complete lack of spikes in the surrounding mitral cells . This strong bursting and surround inhibitory activity is explained , respectively , by the intense activation of the granule cell population at the site of the activated mitral cells ( through feedback inhibition ) and on either side ( through lateral inhibition ) . In this example the same properties evidenced in the moderately activated network are seen intensified . The glomerular cluster is larger and more strongly activated , leading to a larger and more intense mitral cell response . The granule cell response is correspondingly more intense and widespread , and is associated with virtually complete inhibition of surrounding mitral cells . The discharge patterns in both cases include bursting , occasionally overwhelmed by the strength of the activation . To contrast with these examples of localized glomerular input , in Fig . 3C we illustrate a third example , odor k3-3 ( an aliphatic ketone ) , where a less localized , more distributed pattern of glomerular activation involved the activation of two main groups of mitral cells , stronger at the 460–490 site than at the 240–270 site . This provided the opportunity to analyze the interactions of the lateral inhibition elicited by the two sites . The mitral cell discharges ( Fig . 3C , left ) at both of these sites showed patterns of oscillatory bursts . Because of the separation of the two sites , the lateral dendrites of the two mitral cell populations spanned the entire network , thus activating the entire granule cell population , developing relatively uniform bursting spike discharges ( Fig . 3C , right ) . This led to strong lateral inhibition of nearly all the mitral cells not driven by the input . Note however that this inhibition was slightly weaker over the network from sites 0 to 240 , reflecting the slightly weaker input to the mitral cells at the 240 site . This indicates that the inhibition of the mitral cells reflects a balance between the amount of excitation of the mitral cell lateral dendrites and the corresponding excitation of the granule cells . Comparison of these three responses to odor stimulation thus indicates several basic properties underlying odor coding in the olfactory bulb network . The three different odors are initially represented by the three different , spatially restricted , distributions of activated glomeruli . These spatial patterns are processed by the more extensive spatial distributions of lateral inhibition brought about by the interactions with broadly tuned granule cells . The result is a mitral cell output reflecting the spatial distribution of the glomerular input and the temporal structure of the inhibitory processing , which together represent a unique spatiotemporal representation for each odor . The spiking activity shown in Fig . 3 provides the driving force for the formation of the synaptic conductances that represent the odor training of the network in laying down the neural substrate for an odor perception [18] . The spiking activity in turn is driven by the weights of the tuft and dendrodendritic synapses . In order to visualize and analyze the distribution of the synaptic weights , we arranged plots of the mitral to granule cell excitatory weights and granule to mitral inhibitory weights as shown in Fig . 4 . Results for all 72 odors are reported in Fig . S1 . In each panel , the top histogram shows the relative strengths and distribution of the glomerular responses . The middle graph shows the final weight configuration , after a 10 sec simulation of odor input , of the excitatory synapses from mitral to granule cells , with the 500 mitral cells on the abscissa and the 10 , 000 granule cells on the ordinate . The background control is illustrated in Fig . 4A , in which there is no odor input; weights of mitral to granule cell ( excitatory ) synapses show only random low values of peak conductance , generated by the background input ( Fig . 4A , middle ) ; this activity is not enough to generate any potentiation of the inhibitory synapses ( Fig . 4A , bottom ) , which stay at their initial value of zero . The moderately intense focal glomerular activation by ( + ) -Cvn ( Fig . 4B , top ) at mitral cell sites 240–250 elicited action potentials that propagated through their secondary dendrites . This activated the excitatory conductances of the mitral to granule cell synapses between granule cell sites 2 , 000–8 , 000 ( Fig . 4B , middle , yellow areas ) . Note the slight increase in excitatory weights at mitral cell sites 400–590; this reflects the summation of the very weak glomerular input with the background activity ) . This excitatory activity in turn activated granule cells and , thus , potentiation of granule-to-mitral cell inhibitory synaptic conductances ( Fig . 4B , bottom ) . Yellow to white pseudocolors represent synaptic weights fully potentiated , and in this case they are distributed throughout the extent of the most active mitral cell lateral dendrites . Comparison with the spiking data of Fig . 3A shows a close correlation between the location and strength of the synaptic weights and the patterns of mitral and granule cell spiking . For example , the thin horizontal bands of potentiated inhibitory weights ( at mitral cell sites 240–250 in the bottom graph of Fig . 4B ) correspond to granule cells connected to the most active mitral cells; the strong feedback inhibition generated by their activation is responsible for the emergence of the bursting behavior observed in Fig . 3A . The analysis of synaptic weight distribution in the network for the case of a much stronger odor is shown in Fig . 4C ( octanal ) . In this case , the wider range of potentiated excitatory synapses ( Fig . 4C , middle ) generated a strong and widespread potentiation of inhibitory weights ( Fig . 4C , bottom ) , with an evident inhibition of the mitral cells surrounding the most active ones , as indicated by the darker purple area in the excitatory weights map in the range of mitral cells 280–400 and 490–120 . Note that this correlates with the strong lateral inhibition shown in the left panels of Fig . 3B . For a more distributed input , such as odor k3-3 ( Fig . 4D , top ) , the excitatory weights reflected the glomerular activation at the two sites ( Fig . 4D , middle ) , and the consequent strong inhibitory weights on the most active mitral cells ( Fig . 4D , bottom ) . In this case , the distributed input involved more or less synaptic plasticity of the entire network of inhibitory weights , with the flanking weights reflecting the overlap of mitral cell lateral dendrites , for example granule cells in the range around sites 2300 , 6200 , and 8000 . The overall effect on network activity was a widespread bursting behavior that involved the entire granule cell network ( see Fig . 3C , right ) . The formation of different excitatory and inhibitory clusters in response to all odors is shown in Fig . S1 . In summary , these results demonstrated how the learning of different odors can generate , through the differential activation of distributed glomeruli , widely different network behavior with: i ) distinct firing properties involving a variable population of granule cells , ii ) an emergent oscillatory bursting behavior that can span a large portion of the olfactory bulb , and iii ) a powerful lateral inhibition surrounding the most active glomeruli . The relation between lateral inhibition and glomerular activation is critical to odor representation and processing . To understand this property in a more quantitative way , for each odor we calculated the average inhibitory conductance on any given mitral cell as a function of the input strength . We were particularly interested in mitral cells receiving a weak input ( i . e . an odor strength of 1 or 2 ) . For any given odor , these cells correspond to flanking components . Typical cases can be identified in the histograms representing odor input for each odor , for example mitral cell 292 for Eug ( Fig . 5A , left ) , mitral cell 422 for octanal ( Fig . 4C , top ) , and mitral cell 235 for Gua ( Fig . 5A , right ) . They all receive a weak input , but the input of their respective neighbors is quite different , as can be seen in the different histograms for these odors: weak neighbors for Eug , very strong for octanal , and medium for Gua . Given the low connection probability ( 10% ) and the spatially distributed glomerular activation , the emergence of a significant lateral inhibition cannot be taken for granted , and it would be difficult to explore experimentally . In order to analyze the relation between input strength and the effect of the lateral inhibition it generates , mitral cells activated by any given odor were grouped in two input classes , low ( strength 1–2 ) and strong ( strength 3–4 ) . For each odor and each group , the average overall inhibitory conductance was then calculated from the final weight configuration at the end of the 10 sec simulation , including both feedback and lateral actions . The results are shown in Fig . 5B , as a function of the proportion of cells receiving a strong input . Note that without any lateral inhibition the peak inhibitory conductance on mitral cells receiving a weak input ( Fig . 5B , red circles ) would be 0 , since their firing rate would be too low to generate any feedback inhibition . However , the results show that lateral inhibition is developed as an odor activates a small proportion of mitral cells with a strong input . The differential effect on the weaker flanking components can be clearly seen in Fig . 5C , where we show the somatic membrane potential during the first few seconds of a simulation for three mitral cells . As the weights develop , a strong odor , such as octanal , will completely silence a flanking component , such as cell 422 ( Fig . 5C , top ) , whereas a progressively lower effect can be seen for cell 235 during presentation of the less strong Gua ( Fig . 5C , middle ) , and for cell 292 during presentation of the weak Eug ( Fig . 5C , bottom ) . These results demonstrate that significant lateral inhibition can be developed by any given odor that is able to generate strong activity in a relatively small proportion of mitral cells , independently of their spatial location ( but within the reach of their lateral dendrites ) . The olfactory code relayed from mitral cells to the cortex for odor recognition is sculpted by the activity of granule cells , which are ideally positioned for this role in the olfactory bulb circuit . The granule-mitral cell connection probability can thus be expected to have a paramount role in modulating mitral cell firing . Experimentally , the average probability with which a granule cell forms synapses with mitral cells is unknown , although there are findings suggesting that , in general , it is sparse and spatially distributed ( see Discussion ) . To test to what extent a sparse connectivity can significantly modulate mitral cell firing we carried out additional simulations for odor k3-3 using different connection probabilities between granule and mitral cells ( Fig . 6 ) . The configuration of inhibitory weights after a 10 sec simulation using an average maximum potential connectivity of 2 , 5 , and 15% is shown in Fig . 6A . In comparing the results obtained with the control value of 10% ( Fig . 4D ) , with those obtained using a higher connectivity ( Fig . 6A , 15% ) , we observed a sharp difference in the clustering of high synaptic weights around the two regions with active mitral cells with respect to those in other regions . This difference tended to be smaller with 5% connection probabilities and almost disappeared with 2% connectivity . We hypothesized that these differences may result in a significant change of mitral cell firing properties . We tested this in the model by analyzing the average instantaneous firing rate , which is the most relevant parameter to characterize and understand mitral cell responses to an odor sniff . Without GC all mitral cells were more or less active , depending on background activity and/or odor input ( Fig . 6B , top ) . The background activity of mitral cells not receiving any odor input ( e . g . cells in the 325–425 range ) was much reduced already with only 2% connectivity , and almost completely suppressed with connection probabilities above 5% . A higher connection probability ( 10 and 15% ) also suppressed firing of flanking components , a typical contrast enhancement effect . Most interestingly , the firing rate of strongly activated mitral cells was little affected by connection probability . This was more clearly evident from the analysis of their ISI distribution in the range of 10–50 ms , as shown in Fig . 6C ( left ) for 10% connection probability . The two distributions ( with or without granule cells in the network ) were statistically indistinguishable ( Mann-Whitney Rank Sum Test , p = 0 . 88 ) , in contrast with the distributions of 50–200 ms ISIs ( Fig . 6C , right , Mann-Whitney Rank Sum Test p = 0 . 021 ) . This effect was very robust with changes in connection probability ( Fig . 6D ) . The distributions of 50–200 ms ISIs with or without GC in the network were significantly different in all cases except 2% connectivity ( Mann-Whitney Rank Sum Test p = 0 . 115 ) . These model results predict that a sparse granule-mitral connectivity will be able to significantly affect mitral cell firing in two different ways: 1 ) suppressing flanking or weaker components , and 2 ) changing the firing pattern structure of the stronger components . The overall computational effects for odor coding of these mechanism will be investigated in a future work . The relative importance of the feedback and lateral inhibition generated by a sparsely connected network of granule cells is poorly understood . We have shown how they can modulate the overall mitral cell firing structure by different odors ( Fig . 3 ) and under different connectivity ( Fig . 6 ) . To test their effects on different odor concentrations we carried out , without granule cell synapses in the network , a 50 sec long simulation of odor k3-3 . Odor strength was progressively increased every 5 sec , from 0 ( no odor ) to 1 ( the maximum strength used in all simulations ) in steps of 0 . 1 . The results , shown in Fig . 7 ( top ) for mitral cells 420–500 , show a rather high and diffuse activity of all mitral cells , with those receiving relatively lower inputs ( <0 . 4 ) practically indistinguishable from the background activity ( average S/N = 0 . 25 dB ) . We then repeated the simulation including granule cells and starting from the weights configuration obtained after odor learning ( Fig . 4D ) . The results , illustrated in Fig . 7 , showed a surprising effect of feedback inhibition for low odor concentrations , which selectively suppressed the instantaneous firing activity of the most active mitral cells ( e . g . cells 475–484 ) , making them clearly distinguishable ( average S/N = −13 . 1 dB ) from neighbor cells activated by the background noise ( e . g . cells 430–439 ) or by weaker components ( 486–500 ) . These results suggest that lateral and feedback inhibition work together , in a complementary way , to enhance the contrast between an odor signal and the background noise over a wide range of odor concentrations , including subthreshold inputs . We wished to test the network model against physiological recordings of mitral cells responding to odor stimuli . We have previously shown that a reduced network model gives generic mitral cell responses consistent with those reported for different odors in a homologous series [7] . A more detailed , although qualitative , comparison with the present model can be carried out with the data reported in [16] . These authors found that the temporal firing structure of mitral cells in response to a sniff cycle is very precise , with different mitral cells responding with different onset times and firing rate to the same odor ( Fig . 8A ) . We were particularly interested in understanding the reasons for the temporal firing distribution of the response with respect to a sniff cycle . The experimental findings ( Fig . 6A , see [16] ) are quite clear from this point of view: Different cells respond to the same odor with different timing . This is increasingly recognized as an important computational property for odor coding and discrimination ( e . g . [1] , [31] , [32] ) , but the underlying processes are unknown . In order to investigate this issue , which is otherwise experimentally limited within a network framework , we calculated the average distribution of spike times of all the mitral cells activated during 70 sniffs by a strong , a medium , and a weak odor . As shown in Fig . 8B , the distributions were different , and all were clearly distinguishable from the “no odor” condition ( Fig . 8B , gray line ) . A more specific analysis of mitral cells activated by k3-3 , obtained by grouping the mitral cells according to the strength of their input ( Fig . 8C ) , suggested that the different distributions are correlated with the interaction between the odor input strength and granule cell activity . Mitral cells receiving a weak or very weak input ( Fig . 8C , top two panels ) are mostly modulated by weak lateral inhibition generated later in the sniff cycle , whereas cells receiving medium and strong inputs ( Fig . 8C , bottom two panels ) generate earlier strong inhibition that impacts the response latency . Taken together these results indicate how odor identity could emerge from a single sniff as a specific distribution of spikes composed of spatially and temporally positive or negative contributions ( with respect to the “no odor” condition ) from all the mitral cells activated by the odor , each mitral cell making its own contribution according to the specific type of input it receives and the underlying network of granule cells it activates . It is also of interest to compare the model properties with recent experimental studies which revealed sparse and segregated lateral connectivity between mitral and granule cells [10] , [17] , [33] , as illustrated in Fig . 9 ( left panel ) . It showed the connectivity in terms of widely distributed columns of granule cells labeled by widely distributed clusters of mitral cells , formed during the lifetime of the animal . It therefore represents the “maximum” average connectivity that can be obtained between any mitral and any granule cell in the network . During odor presentation , each synaptic weight will independently follow the synaptic plasticity rule to increase/decrease its value , according to the local spiking activity , shaping the actual network connectivity . We have previously shown [18] that this process , through the interaction among odor inputs , action potential backpropagation , and dendrodendritic synapses can generate the kind of distributed interconnectivity observed experimentally [10] . We tested whether our dynamic circuit model could produce similar patterns . As a typical example , we considered all mitral cells activated by odor k3-3 and calculated the peak inhibitory synaptic conductance along their lateral dendrites after odor learning . This is equivalent to the experimental protocol used in the PRV injection experiments , where a few nearby glomeruli are injected with the virus to give labeling of the widely distributed granule cell clusters . Although it contained no inherent cluster connectivity , as shown in Fig . 9 ( right ) , the circuit model generated narrow labeled granule cell patterns similar to those observed experimentally . This included the strong clustering of inhibitory synapses below the most active mitral cells in the model ( Fig . 9 , right ) , corresponding to a region of intense staining of mitral and granule cells in the experiments ( Fig . 9 , left ) , and the weaker clustering of granule cells in a region with no active mitral cells ( Fig . 9 , indicated in orange ) . In the top traces in Fig . 9 , right , it can be seen that , associated with this pattern , the soma of strongly activated mitral cells showed strong action potential firing , whereas the invasion of distal lateral dendrites was weaker . This suggests that the granule cell inhibitory synapses , once established , can gate the backpropagation of a train of action potentials in a precise location along a mitral cell lateral dendrite as suggested experimentally ( e . g . [9] , [25] , [34] ) . These results confirm our previous suggestion [18] , [26] that the sparse and distributed patterns observed experimentally can emerge from the interaction between mitral and granule cell activity during an odor presentation .
Lateral inhibition of mitral and tufted cells has been firmly established by many experiments at the single cell level over the past nearly 50 years [6]–[7] , [35]–[41] , providing evidence that excitation of mitral cells leads to dendrodendritic activation of the granule cells , which then brings about feedback and lateral inhibition of the mitral cells . The interpretation of these studies has been supported by a number of realistic ( e . g . [24] , [26] ) , simplified ( e . g . [39] , [42] ) , or artificial ( e . g . [43]–[44] ) computational models . The present study has enabled us to extend the characterization of lateral inhibition from the single cell to a realistic network level , and provided a clearer representation of the sharpening by the lateral inhibition of the neural response to different odor inputs . We have shown that activity-dependent mechanisms are capable of sculpting the network , leading to the formation of dendrodendritic synaptic clusters in a large , sparsely-connected network . The independent evolution of each synaptic weight , according to the local dendritic spiking activity , will shape the actual network by forming the widespread mosaic of clustered connectivity observed experimentally [10] . This mechanism may be in effect , for example , during the important process of merging newborn granule cells , with their facilitated synaptic plasticity [45] , into the existing bulb network to drive stimulus response decorrelation [44] . The spatial extent of the lateral inhibition was correlated with the strength of mitral cell activation; stronger glomerular activation results in more extensive lateral inhibition . Stronger antidromic activation of mitral cells has been known to be associated with more extensive lateral inhibition of surrounding mitral cells [46] . The model indicates how orthodromic mitral cell activation from the glomeruli produces the same result in the mitral cell ensemble . In relation to recent experimental findings [16] , the model explains in terms of different input strengths the results showing different timing among the mitral cells activated by an odor , and supports the suggestion that the time-to-first-spike can be a critical property for odor identification as the mitral cells project their ensemble response to olfactory cortex [31] . With the use of single odor stimulation , as in experiments discussed in [15] and the present model , the spatial pattern of inhibition reached the full extents of the lateral dendrites of the activated mitral cells , since the action potentials can propagate , in the absence of active inhibitory synapses , to the ends of the lateral dendrites . This has been experimentally demonstrated [25] , although there is evidence that it may not occur under some conditions [47] . However , our model was also able to demonstrate intermittent clusters ( Fig . 9 ) . This was surprising because the initial specific connectivity rules between mitral and granule cell dendrites formed during neurogenesis are not known; we hypothesize that when those rules will be identified the clustering will be even clearer and more widespread . Full propagation is most likely to occur under conditions of single odor stimuli , in which the odor activates specific isolated clusters , with associated mitral cells spanning regions of reduced connectivity with granule cells , as in the present study . In nature , however , odors are usually smelled as combinations of many components that may activate glomeruli in much more complicated and dense patterns [48] . This experience will thus drive mitral cells to set up correspondingly more or less complicated patterns of inhibition which will modulate and gate the backpropagating action potentials , as demonstrated experimentally [9] , [25] and computationally [26] . This forms the foundation for the emergence of lateral inhibition in such a large-scale network ( see Figs . 3 and 4 ) and provides potential functions for modulating precise spatio-temporal patterns of action potentials during exposure to different odorants . There are two major components of information contained in an odorous molecule: odor molecule identity and odor concentration . It is suggested that mitral cells might employ both firing rate of individual mitral cells and spatial patterns of spike timings of particular combination of active mitral cells in encoding odor molecule concentration and identity information separately [49]–[50] . Recent experiments showed that mitral cells sharing the same glomerulus have highly correlated firing rates in response to odors , but their spike timings are relatively different with respect to the phase of the sniff cycle [50] . These studies suggest that mitral cells may use different coding channels to represent information , but cannot extract more specific information on the underlying process . Our study is in full agreement with those studies , and suggests that the spatial patterns of precise timing locked to sniff cycle may represent odor identity , by generating a spatio-temporal representation of odor molecule information . Lateral inhibition by recurrent granule cells not only plays a role in decreasing firing rates of olfactory bulb activities resulting in “spiking packets” [51] , but can also enhance synchrony [30] and precise timing of both mitral cells and granule cells , suggesting a temporally sparse code . In awake animals , the act of sniffing increases the air velocity and changes the duration of airflow in the nose , which improves olfactory detection [52] . It has also been shown that the waking state [53] enhances the level of inhibition in the network , which increases the sparseness of the mitral cell responses . Our model is consistent with this result ( Fig . 5 C; see also Fig . 6B ) , which will be explored in more depth in a future study . Behavioral studies have suggested that there is an active process modulating neuronal responses during sniffing [52] . This may produce optimal temporal sequences in the olfactory bulb [54] . Recordings in vivo and in vitro have shown evidence that the frequencies of response oscillations are modulated by breathing and sniff rates [1] , [16] , [51]–[52] , [54] . Our model provides one explanation by showing how lateral inhibition can modulate this process by strongly suppressing the background noise while synchronizing the mitral cell responses . A more comprehensive analysis of mitral cell responses in the presence of mixtures at different concentrations in relation to the sniff cycle will be presented in a future study . Recent studies have suggested several possible mechanisms that could result in sparse odor coding in the olfactory bulb . Interestingly , there are also recent experimental findings in vivo [48] showing that natural odorants are represented by dense ( as opposed to sparse ) glomerular activation . In this paper we demonstrate that a sparse representation can emerge naturally from the mitral-granule cell interactions , realistically implemented in our model with self-organizing dendrodendritic synapses driven by mitral cell activity . Feedback and lateral inhibition cooperate to maintain a sparse representation complementarily acting over different odor concentrations . This differs from more theoretical and speculative artificial network models suggesting that sparse coding in the mammalian olfactory bulb can arise from an external cortical signal generating an incomplete feedforward inhibition [49] or from feedforward inhibition in the mushroom body of insects [55]–[57] . In regard to downstream processing , in modeling the analogous antennal lobe of the fly it has been recognized [55] that it is necessary to know how the output is processed in the downstream mushroom body and lateral horn . Those downstream mechanisms are not well known even in the fly , and that applies as well to the olfactory cortex; in particular , the precise cortical targets of the mitral cells in the mammalian olfactory bulb are not known . We have therefore focused on the processing inherent in the olfactory bulb network . It remains to note that synaptic plasticity is fundamental to any dynamic network . The clustered activity generated by odor stimuli is dependent on action potential backpropagation along the mitral cell lateral dendrites ( according to the sequence of events discussed in Results ) , rather than the specific plasticity rule used to evolve the synaptic weights during odor presentation [18] , [26] . Synaptic plasticity in the mitral-granule circuit has not been observed directly . We consider this lack of information as a shortcoming of the experimental techniques rather than a demonstration that there is no plasticity in the olfactory bulb . Indeed , recent studies have shown more or less direct evidence for plasticity of olfactory input in mitral cells [58]–[59] , and in granule cells [60]–[62] . Although further experimental investigation is required to have a more detailed picture , one of the reasons that can explain the problems encountered in investigating plasticity in the olfactory bulb can be easily predicted by our model , and is related to network connectivity . The granule cell inhibitory synapses , once formed , can prevent any further activity-dependent plasticity of synapses far from the soma of the most active mitral cells ( see Fig . 7 ) . The model thus predicts that plasticity could be more easily characterized by recording from mitral and granule cells forming reciprocal perisomatic connections . In conclusion , the present model suggests a physiologically based mechanistic explanation of how dendrodendritic excitation and inhibition , generated by experimentally measured odor inputs , can drive self-organization of the evolving dynamics in a large-scale olfactory bulb network . The process promotes the emergence of clustered organization in granule cell synaptic weight structure and mitral cell responses found in experimental studies [10] , [16] , [33] . The results demonstrate how odors can be represented in the olfactory bulb by a combination of temporal and spatial patterns , with both feedforward excitation and lateral inhibition via dendrodendritic synapses as the underlying mechanisms necessary and sufficient to maintain a sparse representation of odor identity .
The network was composed of multi-compartment canonical models of 500 mitral and 10000 granule cells , implemented as described in our previous studies [18] , [24] , and connected through dendrodendritic synapses [63] . The canonical model for mitral cells was implemented with 312 compartments representing an axon , the soma , the apical dendrite , and 2 lateral dendrites each 1 . 5 mm in length , in the range indicated by anatomical measurements [40] . In the real case the 1 . 5 mm is the maximum extent of the mitral cell lateral dendrites . Cell stains show that some dendrites will be less , but a uniform extent enabled us to assess more clearly the extent of the lateral inhibition under different conditions in the model . Uniform passive properties were used , with Ra = 150 Ω·cm , τm = 20 ms , and Rm and Cm adjusted to obtain an input resistance of about 100 MΩ [40] . Resting potential was set at −65 mV and temperature at 35°C . Cells were modeled as regular firing cells ( see Fig . 1d in Ref . [64] ) , with Na , KA , and KDR conductances uniformly distributed over the entire dendritic tree [65] . Kinetics for the Na conductance were from hippocampal pyramidal neurons [66] , whereas those for KA and KDR were from mitral cell data [67] . This resulted in the mitral cells firing within the range of experimentally observed firing rates and , in further agreement with experimental findings , somatic action potentials backpropagated at full amplitude up to the tuft [68] , and an AP could initiate in the tuft or in the primary dendrite for moderate to strong odor inputs [69] ( see also Fig . 1e in Ref . [64] . Granule cells ( GC ) were modeled with a soma and a 20 segment radial dendrite ( 250 µm of total length ) representing the dendritic tree . Na+ and KA channels were distributed throughout [70]–[72] whereas KDR was present only in the soma [70] . In agreement with experimental findings [70] , the dendritic KA resulted in a significant effect on the spike latency of these cells ( see Fig . 1Bb of Ref . [24] ) , and adaptation under strong inputs [73]–[74] . It should be noted that a number of additional ion channels and mechanisms were not included in our model cells . Virtually all of them , such as additional K+ conductances , Ih current , persistent Na+ current , Ca2+ and Ca2+-dependent currents , but also activity-dependent changes in channels density or kinetic , non-uniform channels distribution , intracellular Ca2+ dynamics , intercellular variability , additional external inputs etc , may result in some modulation of the results . This is precisely why we did not include them in the model at this stage . Our focus has been on understanding the processes underlying lateral and feedback inhibition and their main consequences . Future works will eventually investigate how and to what extent additional cell types , mechanisms , and external inputs can affect the basic findings shown in this paper . There are a number of different types of cells in the olfactory bulb network , each of them carrying out or involved in functions that in most cases are not understood . Rather than include a priori in the network all cell types for which there was some experimental suggestion on their electrophysiological properties or function , we decided to keep the model simple enough to allow a clear identification of the key mechanisms underlying the effects of lateral and feedback inhibition , and constrained enough by experimental findings to allow not only a direct comparison of the results with specific experimentally observable but also to make experimentally testable predictions . From this point of view , as shown and discussed in our previous papers [e . g . 18] , [24] , as long as there are action potentials propagating along the mitral cell lateral dendrites and granule cells making local plastic synaptic dendrodendritic connections with them , the network will self-organize following an odor presentation . The level of details included here are thus those necessary and sufficient to have these mechanisms in place . With respect to our previous reduced model network , in this paper we added a new level of realism in network size and connectivity and odor input , obtaining a more realistic network dynamics that has not been achieved by other modeling approaches . Network connectivity is presented and discussed in Results . Effective dendrodendritic coupling between granule cell synapses and mitral cell secondary dendrites was implemented by connecting a GC synapse , containing the same proportion of AMPA and NMDA channels , with the appropriate compartment of mitral cells secondary dendrites containing GABA channels . The AMPA conductance was modeled as an alpha-function with a time constant of 3 ms and a reversal potential of 0 mV . The NMDA conductance was based on a NEURON model [75] of experimental findings [76]–[77] , assuming an external magnesium concentration of 1 mM and a reversal potential of 0 mV . The model parameters were adjusted to obtain a time-to-peak and decay time constant of 10 and 50 ms , respectively . A double exponential function was used to model the GABA conductance , with a reversal potential of −80 mV . Rise and decay time constants were 1 and 200 ms , respectively , implicitly taking into account the mechanisms underlying the overall time course observed experimentally for the inhibitory potential elicited by a single AP in a mitral cell [78] . Unless otherwise noted , the peak excitatory and inhibitory conductances of each synapse were 0 . 5 nS and 3 nS , respectively , equivalent to about 1 and 5 real individual synapses , respectively [78] . Synapses ( excitatory or inhibitory ) were activated whenever the corresponding presynaptic compartment reached the threshold of −40 mV , in agreement with experimental findings [79] suggesting that recurrent inhibition of the secondary dendrite of a mitral cell does not necessarily require the generation/propagation of an action potential to the soma of the GC . Synaptic weights of dendrodendritic synapses started at zero and , in response to odor input , followed the same plasticity rule previously used [18] , [26] . Briefly , each component ( inhibitory or excitatory ) of each dendrodendritic synapse was independently modified according to the local membrane potential , of the lateral dendrite of the mitral cell or the granule cell synapse , to calculate the instantaneous presynaptic ISI . After each spike , the peak conductance , w , of any given synapse was updated from its current value w{exc , inh} , p = gsyn , {exc , inh}•S ( p ) to w{exc , inh} , p+Δ = gsyn , {exc , inh}•S ( p+Δ ) , where the function Δ = {0 , +1 , −1} followed a classical scheme [80]–[81] in which Δ = 0 for ISI>250 ms , Δ = −1 for 30<ISI<250 ms , and Δ = 1 for ISI<30 ms . The typical sigmoidal activation function S ( p ) [82] was defined as S ( p ) = 1/ ( 1+exp ( ( p-25 ) /3 ) ) . In this way , the weight ( i . e . the peak synaptic conductance ) of any given synapse could go from a fully depressed ( for p≈0 ) to a fully potentiated state ( for p≈50 ) , or vice-versa , in about 50 consecutive spikes of the appropriate frequency . Unless explicitly noted otherwise , p = 0 at the beginning of a simulation . This plasticity rule is non-hebbian , since it changes a synaptic weight ignoring any postsynaptic activity . However , we have previously noted , shown , and discussed [18] , [26] that the formation of synaptic clusters consistent with those observed experimentally [10] is robust and does not depend on the specific choice for the functional form used to update the synaptic weights . Odor stimulation of mitral cells was modeled using a synchronous activation , in all tuft compartments , of synaptic inputs with a double exponential conductance change ( 20 and 200 ms rise and decay time , respectively ) , and an individual peak conductance in the range 0–0 . 5 nS . This corresponds to a total aggregate input conductance of up to 10 nS . It elicits 6 somatic spikes during a single activation , within the range observed experimentally for the number of APs generated during a respiratory cycle [81] . To simulate the learning process of a given odor , unless explicitly noted otherwise , the synaptic inputs to the mitral cells belonging to the active glomeruli were activated at a random frequency in the range of 2–10 Hz , corresponding to the range of natural sniffing frequency during explorative behavior in rats [55] . An independent random background synaptic activity ( eliciting spikes at around 2 Hz ) , was added to the soma of all cells to model in vivo activity . During preliminary tests , we found that a 10 sec odor presentation was sufficient to stabilize synaptic weights and cell activity . All simulations were thus carried out for 10 sec . Experimentally , the instantaneous firing rate of individual mitral cells during an odor presentation is highly variable ( e . g . [39] ) , depending on many factors such as input strength , stimulation/sniffing frequency , and network interactions . In our model , the instantaneous firing rate of mitral cells activated by an odor was up to 75 Hz , within the physiological range observed in vivo . All simulations were carried out with the NEURON simulation program ( v7 . 3 , [83] ) on a BlueGene/Q IBM supercomputer ( CINECA , Bologna , Italy ) or a Cray XE6 system ( INCF , Lindgren , Sweden ) . Under control conditions , the model network was composed by a system of 12 , 877 , 500 non-linear differential equations , and a typical 10 sec simulation required about 2 hours using 400 processors using a fixed time step of 25 µs . To test the robustness of the results , we ran additional test simulations of odor k3-3 using a smaller time step ( 10 µs ) , cells modeled with a larger number of compartments ( 5 µm membrane segments ) , or a different random number sequence . In all cases ( see Fig . S2 ) , the final inhibitory weights distribution was statistically indistinguishable from control ( Mann-Whitney rank sum test p>0 . 913 ) . The model and simulation files are available for public download under the ModelDB section of the Senselab database suite ( http://senselab . med . yale . edu , acc . n . 144570 ) . Movies from each simulation were created offline , with a custom developed command-line environment that generates a movie from the spike times . We used the open-source tool Octave ( www . gnu . org/software/octave ) to create all frames , updating each cell's visual aspect in each frame , and pipeline them to concurrent instances of POVRay ( www . povray . org ) , a multi platform , free tool for creating high quality three dimensional photo-realistic scenes . All frames were finally joined together into a compressed MPEG-4 movie using FFmpeg ( www . ffmpeg . org ) . A typical hi-resolution ( 1280*720 pixels ) 10 sec simulation movie is composed by 600 frames . On 16 quadcore CPUs interconnected via InfiniBand , the rendering of all frames takes less than 4 minutes . | In the paper we address the role of lateral inhibition in a neuronal network . It is an essential and widespread mechanism of neural processing that has been demonstrated in many brain systems . A key finding that would reveal how and to what extent it can modulate input signals and give rise to some form of perception would involve network-wide recording of individual cells during in vivo behavioral experiments . While this problem has been intensely investigated , it is beyond current methods to record from a reasonable set of cells experimentally to decipher the emergent properties and behavior of the network , leaving the underlying computational and functional roles of lateral inhibition still poorly understood . We addressed this problem using a large-scale model of the olfactory bulb . The model demonstrates how lateral inhibition modulates the evolving dynamics of the olfactory bulb network , generating mitral and granule cell responses that account for critical experimental findings . It also suggests how odor identity can be represented by a combination of temporal and spatial patterns of mitral cell activity , with both feedforward excitation and lateral inhibition via dendrodendritic synapses as the underlying mechanisms facilitating network self-organization and the emergence of synchronized oscillations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"circuit",
"models",
"computational",
"neuroscience",
"olfactory",
"system",
"biology",
"sensory",
"systems",
"neuroscience",
"coding",
"mechanisms"
] | 2013 | Sparse Distributed Representation of Odors in a Large-scale Olfactory Bulb Circuit |
Salmonella enterica serotype Typhimurium ( S . Typhimurium ) is one of the most frequent causes of food-borne illness in humans and usually associated with acute self-limiting gastroenteritis . However , in immunocompromised patients , the pathogen can disseminate and lead to severe systemic diseases . S . Typhimurium are facultative intracellular bacteria . For uptake and intracellular life , Salmonella translocate numerous effector proteins into host cells using two type-III secretion systems ( T3SS ) , which are encoded within Salmonella pathogenicity islands 1 ( SPI-1 ) and 2 ( SPI-2 ) . While SPI-1 effectors mainly promote initial invasion , SPI-2 effectors control intracellular survival and proliferation . Here , we elucidate the mode of action of Salmonella SPI-2 effector SseI , which is involved in control of systemic dissemination of S . Typhimurium . SseI deamidates a specific glutamine residue of heterotrimeric G proteins of the Gαi family , resulting in persistent activation of the G protein . Gi activation inhibits cAMP production and stimulates PI3-kinase γ by Gαi-released Gβγ subunits , resulting in activation of survival pathways by phosphorylation of Akt and mTOR . Moreover , SseI-induced deamidation leads to non-polarized activation of Gαi and , thereby , to loss of directed migration of dendritic cells .
Salmonella enterica serovars are pathogenic bacteria that cause severe diseases ranging from enteric fever ( e . g . by Salmonella Typhi ) to gastroenteritis and bacteraemia caused by non-typhoidal Salmonella ( NTS ) . Salmonella Typhimurium , the model organism of NTS infection , has a broad host spectrum and is one of the most frequent causes of food-borne illness in humans and other vertebrates including food-producing animals . S . Typhimurium infection is usually associated with acute self-limiting gastroenteritis in immunocompetent individuals . However , in immunocompromised patients , S . Typhimurium can disseminate and lead to severe systemic diseases [1–4] . S . Typhimurium are facultative intracellular bacteria , which exploit uptake by phagocytic intestinal cells but are also able to force their uptake into non-phagocytic epithelial cells [5] . Inside host cells , Salmonella reside and proliferate in a specific membrane compartment defined as Salmonella-containing vacuole ( SCV ) . Uptake and intracellular life of Salmonella depends on two type-III secretion systems ( T3SS ) that are encoded within Salmonella pathogenicity islands 1 ( SPI-1 ) and 2 ( SPI-2 ) . These T3SSs act as molecular syringes that translocate > 40 Salmonella effector proteins into the host cell cytosol . While initial invasion is mainly promoted by SPI-1 T3SS , intracellular survival and proliferation largely depends on SPI-2 T3SS effectors [6–9] . At least 28 effectors are secreted by the SPI-2 T3SS into host cells . A “core” subset of effectors ( e . g . , SseF , SseG , SifA , and PipB2 ) appear to be involved in organization and maturation of Salmonella containing vacuoles ( SCV ) [9] . Other effectors play major roles in suppression of innate immune signaling pathways or modulate adaptive immune responses [9–12] . Recently , the SPI-2 effector SseI ( also known as SrfH ) has attracted increased attention , because it inhibits directed migration of dendritic cells and is involved in long-term systemic infection [13] . Moreover , pseudogenization of the effector gene sseI confers rapid systemic hyperdissemination of S . Typhimurium ( sequence type ) ST313 , which commonly causes systemic bacteremia in children and immunocompromised adults in sub-Saharan Africa [14] . SseI consists of 322 amino acids and its N-terminal part is similar to other SPI-2 effectors , suggesting a role in translocation and membrane localization . In fact , cysteine-9 of SseI has been shown to be palmitoylated in host cells to achieve membrane binding [15] . Until now , however , the molecular mechanism of SseI has remained unknown . Because crystallographic studies suggested that the 37 kDa SseI effector protein exhibits structural similarity with the catalytic domain of the deamidating Pasteurella multocida toxin ( PMT ) [16] , we studied whether SseI possesses deamidase activity . Deamidation is a post-translational modification , which is exploited by various bacterial exotoxins and effectors [17 , 18] . A prototype of these exotoxins is PMT [17 , 19 , 20] . This exotoxin is a 145 kDa protein that is responsible for atrophic rhinitis in pigs . The toxin activates osteoclast differentiation , while differentiation of osteoblasts is blocked [21 , 22] . The underlying molecular mechanism of the action of PMT is the activation of heterotrimeric G proteins by deamidation [23] . PMT deamidates a specific glutamine residue in the α-subunits of Gq/11 , Gi/o and G12/13 proteins , which plays a crucial role in hydrolysis of GTP and in inactivation of heterotrimeric G proteins [24] . Thus , deamidation of the glutamine residue by PMT freezes the G protein in its active state . Here , we elucidated the molecular mode of action of SseI . We report that the Salmonella SPI-2 T3SS effector deamidates heterotrimeric G proteins of the Gi family in vitro and in vivo . Thereby , SseI is responsible for the activation of Akt kinase after target cell invasion and increases cell survival during Salmonella infection . Moreover , our studies reveal a pivotal role of SseI-induced deamidation of Gαi in the inhibition of dendritic cell migration .
The comparison of the amino acid sequence of the deamidase domain of PMT ( residues 1144–1240 ) with SseI revealed high sequence similarity ( S1A Fig ) . More importantly , amino acids known to be crucial for deamidase activity of PMT are conserved in SseI ( e . g . , C178 , H216 , and D231 ) [19] . Therefore , we utilized a previously described method to detect deamidation of G proteins by bacterial effectors [23 , 24] . To this end , SseI was coexpressed in E . coli with the α-subunit of Gi2 . Deamidation of Gαi2 was determined by immunoblot analysis , utilizing a monoclonal antibody ( GαQE ) that specifically recognizes Gα after deamidation of a specific glutamine residue in the switch II region [25] . The GαQE antibody detected Gαi2 coexpressed with SseI in E . coli but not the solely expressed Gαi2 , indicating a deamidation activity of SseI ( Fig 1A ) . Purified Gαi2 was subjected to mass spectrometric ( MS ) analysis . MS analysis identified a tryptic peptide of Gαi2 ( m/z 455 . 21702+ ) corresponding to amino acid residues 199–206 . This peptide includes the glutamine residue ( Gln-205 ) essential for the hydrolysis of GTP . Additionally , a second peptide was identified with a mass shift of 1 Da ( m/z 455 . 71022+ ) . Tandem MS analysis revealed that Gln-205 was deamidated , resulting in a glutamic acid residue at this position . No relevant deamidation occurred when Gαi2 was coexpressed with mutant SseI ( SseI-C178A ) , which lacks deamidation activity ( Fig 1B and S1B and S1C Fig ) . Similarly , recombinantly expressed wild type ( wt ) SseI ( amino acids 137–322 , SseIC ) , but not the C178A , H216A , or D231A mutants , caused deamidation of purified Gαi2 ( Fig 1C ) . The effector SseI is secreted by a type III secretion system of S . Typhimurium . Therefore , SseI is not taken up by eukaryotic cells as compared to AB-type bacterial exotoxins like PMT . To enable cellular uptake of SseI , we utilized the receptor binding and translocation domain of PMT [26] . The deamidation domain of SseI ( SseIC ) was fused C-terminally to the N-terminal part of PMT ( amino acids 1–505 ) ( Fig 2A ) . This chimera , PMT-SseIC , was recombinantly expressed , purified and tested for cellular activity . Treatment of HEK-293 cells with increasing concentrations of PMT-SseIC led to a deamidation of Gα proteins as determined by immunoblot analysis using the GαQE antibody . Treatment of cells with the C178A mutant of PMT-SseIC exhibited no deamidation ( Fig 2B ) . In a next step , we compared the activity of PMT-SseIC with that of PMT . HEK-293 cells were intoxicated either with PMT or with PMT-SseIC and cell lysates were tested for deamidated Gα proteins . PMT led to two deamidation signals migrating at the molecular mass of the Gαq and Gαi proteins ( Fig 2C ) . However , PMT-SseIC only induced one deamidation signal migrating at the molecular mass of Gαi . Deamidation of the essential glutamine residue in the switch II region impairs the GTP hydrolysis by the α-subunit resulting in a permanent active phenotype of the G protein [23] . Activated Gαi family members inhibit adenylyl cyclase leading to decreased cAMP levels . We utilized a FRET-based approach to determine how cAMP levels respond to SseI treatment . The FRET sensor Epac2-camps [27] exhibits a decreased FRET ratio , when cAMP is increased . When HeLa cells , expressing the FRET sensor , were treated with the adenylyl cyclase activator forskolin , the FRET ratio declined , indicating increased cAMP levels . In line with an activation of Gαi , pretreatment with PMT-SseIC or PMT attenuated the effect of forskolin , indicating an inhibitory effect on adenylyl cyclase ( Fig 2D ) . Similar results were obtained by direct measurements of cAMP levels ( see ELISA assay S2A Fig ) . Activation of heterotrimeric G proteins leads to dissociation into Gα and Gβγ-subunits [28] . Both subunits can interact with their specific effectors and stimulate specific signaling cascades . Therefore , we studied the impact of SseI on phosphoinositol-3-kinase ( PI3K ) γ , a prototypical Gβγ effector [29–31] . Activation of Gβγ was monitored by translocation of the phosphatidylinositol-3 , 4 , 5-triphosphate ( PIP3 ) sensor protein GFP-Grp1PH from the cytosol to the membrane after additional ectopic expression of p110γ without and with p101 ( Fig 2E and S2B Fig ) . As a control , we used N-formylmethionine-leucyl-phenylalanine ( fMLP ) , which stimulates Gi-coupled GPCRs ( G protein-coupled receptors ) . fMLP induced translocation of Grp1PH to the membrane in the presence of complete PI3Kγ ( p101 with p110γ ) ( S2Bb Fig ) . However , in the absence of complete PI3Kγ ( p101 without p110γ ) , fMLP did not redistribute Grp1PH , indicating the insensitivity of endogenously expressed PI3Ks ( S2Ba Fig ) . Treatment of transfected cells with PMT-SseIC , but not with the inactive mutant , strongly stimulated the redistribution of Grp1PH to the plasma membrane ( Fig 2E and S2Bc/d Fig ) . This effect was not further stimulated by fMLP incubation , indicating an activation of the PI3K independent of the stimulation by GPCRs . Various recent studies suggested that Gαi/o proteins and PI3-kinases play crucial roles in regulation of immune cell signaling and cell survival in infection [32–36] . Therefore , we studied the downstream signaling of Gβγ subunits and of PI3Kγ in more detail . To this end , we transfected bone marrow-derived macrophages ( BMDMs ) or RAW264 . 7 macrophages with a GFP expression vector either harboring SseI or the catalytically inactive mutant ( SseI-C178A ) . After starvation of cells for 4 h , phosphorylation of Akt , mTOR and the mTOR-effector S6 ribosomal protein was analyzed by immunoblotting . Transfection with SseI strongly increased phosphorylation of the PI3K downstream effectors Akt , mTOR , and S6 ribosomal protein in BMDMs ( Fig 2F and S2C Fig ) . Similar results were obtained in RAW264 . 7 macrophages ( S2D and S2E Fig ) . To study the role of SseI in infection of RAW264 . 7 macrophages and BMDMs , we employed wt- or ΔsseI-Salmonella Typhimurium strains and used immunofluorescence microscopy . As depicted in Fig 3A , S . Typhimurium were identified by Salmonella O-antigen antiserum inside and outside of macrophages . Notably , we observed deamidated G proteins , determined by the GαQE antibody , only in macrophages infected with wt-Salmonella . Moreover , deamidation occurred strictly after internalization of bacteria ( Fig 3A , S3A and S3B Fig ) . Next , we studied the activation of the PI3K/Akt pathway by SseI during infection of RAW264 . 7 macrophages with wt- or ΔsseI-Salmonella strains . Immunoblot analysis revealed G protein deamidation as early as ~2 h post infection ( p . i . ) with subsequent increase over time ( Fig 3B ) . Infection with wt- and ΔsseI-Salmonella strains led to phosphorylation of Akt at 1 h p . i . . Hereafter , phosphorylation of Akt decreased over time in macrophages infected with ΔsseI-Salmonella . In wt-Salmonella-infected macrophages , phosphorylation of Akt remained longer than in the ΔsseI-Salmonella-infected cells ( Fig 3B ) . Immunofluorescence microscopic studies revealed that 5 h p . i . an increase in Akt-phosphorylation was seen in wt-Salmonella-infected cells , but not in non-infected or ΔsseI-infected cells ( Fig 3C and S3C Fig ) . In line with an essential role of PI3K , LY294002 ( LY29 ) , a specific PI3K inhibitor , decreased phosphorylation of Akt in a concentration-dependent manner ( Fig 3D and 3E ) . Several publications have shown that activation of Akt leads to anti-apoptotic and pro-survival outcomes in various mammalian cell types [33 , 36] . Therefore , we were prompted to investigate , whether the infection with wt- or ΔsseI-Salmonella has an effect on the survival of host cells . To this end , we measured cell death of infected cells by determination of LDH release from infected RAW264 . 7 macrophages . This study revealed that infection with ΔsseI-Salmonella caused higher amounts of LDH release than with wt-Salmonella ( Fig 4A ) . Accordingly , cell viability measured by the metabolic capacity of the cells , was increased after infection with wt-Salmonella as compared to ΔsseI-infected cells ( Fig 4B ) . Inhibition of PI3K with LY29 diminished the effects of wt-Salmonella ( Fig 4A and 4B ) . Recent studies from the Monack laboratory showed that SseI inhibits directed cell migration [10 , 13] . Therefore , we were prompted to study the role of G proteins in the action of SseI in more detail . To this end , we investigated the migration of dendritic cells in a 3D collagen model . We infected bone-marrow derived mature dendritic cells ( DCs ) with wt- or ΔsseI-Salmonella , seeded the cells into collagen gels and applied a CCL19 chemokine gradient . ΔsseI-infected cells migrated with comparable speed and directness towards the chemokine compared to non-infected cells . In contrast , wt-infected DCs completely lost their ability to migrate in a directed manner , confirming previous results [10 , 13] ( Fig 5A–5C; S1 Video ) . As found for macrophages , time-resolved Western blot analysis revealed deamidation of Gα-subunits by wt-Salmonella in DCs ( Fig 5D ) . Moreover , very similar as observed in macrophages , downstream signaling of the PI3K pathway was elevated in wt-Salmonella-infected DCs , exhibiting increase in Akt phosphorylation ( Fig 5D ) . To investigate whether the inhibition of migration observed with wt-Salmonella was due to the enzymatic activity of SseI , we ectopically expressed wt SseI or the inactive C178A-SseI mutant in DCs . As shown in Fig 5E and 5F ( see also S2 Video ) , transfection with wt sseI led to a similar loss of chemotaxis as compared to infection with wt-Salmonella , while cells transfected with SseI-C178A mutant migrated normally . Interestingly , we did not detect a difference in the speed of migration between wt- and inactive sseI-transfected cells ( Fig 5G ) . To gain further evidence , whether deamidation of Gαi-subunits is the cause for impaired chemotaxis , we investigated the directed migration of DCs derived from Gnai deleted mice . The 3 Gαi isoforms ( Gαi1 , 2 , 3 ) are encoded by Gnai1 , 2 and 3 . Relevant for DCs are Gαi2 , 3 ( see Discussion section ) . Gαi1 is only poorly or not at all expressed in dendritic cells [37] ( http://www . immgen . org/databrowser/index . html ) . We purified Gαi2 and Gαi3 as recombinant proteins and performed in vitro deamidation assays . Both isoforms of the Gαi-subunit are deamidated in a comparable manner ( S4A Fig ) . Because double Gnai2-/- / Gnai3-/- mice are not viable [38] , we obtained DCs from Gnai2-/- or Gnai3-/- C57BL/6 mice . In uninfected cells ( controls ) , depletion of Gαi2 ( Gnai2-/- ) impaired chemotaxis as compared to wild type DCs , while Gαi3 depletion ( Gnai3-/- ) had no major effect on migration ( Fig 6A ) . Thus , as reported for macrophages [39] , our data indicated a crucial role of Gαi2 but not of Gαi3 in directed migration of DCs . Infection with wt-Salmonella impaired chemotaxis of control , Gαi2- and Gαi3-depleted DCs ( Fig 6A ) , while migration speed was not significantly altered ( S4B Fig ) . As expected , immunoblot analysis with anti-Gαi1 , 2 antibody revealed deletion of Gαi2 in Gnai2-/- DCs ( Fig 6B ) . While infection of control Gnai2+/+ cells with wt-Salmonella resulted in strong labeling of deamidated G proteins after 6 h , almost no labeling was observed in Gnai2-/- DCs ( Fig 6B ) . In Gnai3-/- DCs deamidation was observed in a similar intensity as in Gnai3+/+ cells ( Fig 6B ) . This indicates that Gαi2 is the predominant SseI substrate in mouse DCs . To mimic SseI-induced deamidation , we expressed mutant Gαi2Q205E or wt Gαi2 in Gnai2-/- DCs . Expression was monitored by immunoblotting ( S4C Fig ) . Ectopic expression of wt Gαi2 partially rescued the impaired chemotaxis of Gnai2-/- DCs , while expression of mutant Gαi2Q205E strongly inhibited directed migration of DCs ( Fig 6C and 6D , and S3 Video ) . Again , the migration speed was not changed by expression of Gαi2 , or Gαi2Q205E ( Fig 6E ) .
Our studies elucidate the molecular mechanism of the S . Typhimurium SPI-2 T3SS effector SseI . This effector protein plays a crucial role in S . Typhimurium infection after invasion of the pathogen and modulates the immune responses of the host [9] . Here , we show that SseI deamidates a specific glutamine residue ( Gln205 in Gαi2 ) in the α-subunits of heterotrimeric G proteins , which is involved in GTP hydrolysis , thereby the G protein is persistently activated . Thus , T3SS effector SseI exhibits the same mode of action as exotoxin PMT from Pasteurella multocida . This finding is in line with the structural similarity of both toxins , although the primary sequence identity of the catalytic domains is only ~ 20% [16] . Exchange of the conserved catalytic amino acid residues C178 , H216 and D231 inhibited the deamidase activity of SseI . Because SseI is per se not able to enter target cells , we constructed a chimeric toxin ( PMT-SseIC ) , consisting of the N-terminal binding and translocation domain of PMT and the deamidating domain of SseI . Additionally , we transfected mammalian cells with a GFP-SseI construct . These experiments confirmed our in vitro results in intact cells , showing that SseI acts as a deamidase on heterotrimeric G proteins . While PMT deamidates several G proteins , including Gi/o , Gq/11 and G12/13 [23 , 24] , our data indicate that in intact cells , SseI is specific for Gi proteins . In line with activation of Gi , SseI decreased cAMP levels in cells . Besides inhibition of adenylyl cyclase , Gi is crucially involved in activation of PI3 kinase via release of Gβγ subunits [29 , 40] . Accordingly , we observed the translocation of a PIP3 sensor protein ( GFP-GrpPH ) to the cell membrane after treatment with PMT-SseIC . Many reports describe increase in Akt/mTOR signaling in Salmonella-host interactions [33 , 36 , 41] . We found that SseI increased phosphorylation and activation of Akt , its target mTOR and S6 ribosomal protein . These effects depended on the active SseI protein . The inactive mutant SseI-C178A had no effect . To study the role of SseI in infection , we employed S . Typhimurium , in which SseI was deleted . These studies confirmed that after infection of macrophages with wt-S . Typhimurium , Gi proteins were deamidated , while this was not observed with ΔsseI-Salmonella . Importantly , deamidation of G proteins was detected only after uptake of Salmonella into host cells , indicating the importance of invasion for activation of SPI-2 T3SS and translocation of the effector SseI . wt-Salmonella strongly activated the pro-survival kinase Akt in macrophages and DCs . Notably , also infection with ΔsseI-Salmonella resulted in Akt phosphorylation . However , the phosphorylation of Akt , observed in the absence of SseI , diminished with time , whereas wt-Salmonella caused longer lasting Akt phosphorylation . Short-term Akt activation by ΔsseI-Salmonella is probably caused by Salmonella outer protein B ( SopB ) , which is an effector of SPI-1 T3SS and a strong activator of Akt [42 , 43] . Activation of Akt by SopB differs from SseI and might be independent of class I PI3 kinase [44 , 45] . In contrast , we observed that activation of Akt by SseI is inhibited by the class I PI3K inhibitor LY29 . In line with the activation of the PI3K-Akt signaling pathway , we observed increased survival of host cells after infection with wt-Salmonella , as compared to ΔsseI-infection . Previously , it was reported that SseI inhibits the directed motility of macrophages and DCs towards a chemokine gradient [13] . Using an infection model , we confirmed that wt-Salmonella but not ΔsseI-Salmonella inhibited directed migration of DCs . Until recently , the molecular mechanism by which SseI affects directed migration was largely enigmatic . Our data indicate that the SseI-catalyzed deamidation of Gi proteins is responsible for this effect . Concomitantly with inhibition of directed migration , infection with wt-Salmonella but not with ΔsseI-Salmonella caused deamidation of Gi in DCs . It is widely accepted that Gi proteins play a pivotal role in chemotaxis and most , if not all , chemokine receptors are coupled to Gi proteins [37 , 46] . Three Gi proteins ( Gi1-3 ) exist , which share 85–95% amino acid sequence identity in their Gα subunits . While Gi1 is only poorly expressed in leukocytes , Gi2 and Gi3 are highly expressed [37] ( http://www . immgen . org/databrowser/index . html ) . Especially , Gi2 appears to be essential for immune cell signaling . Several studies showed that Gi2 plays an important role in directed migration and homing of B and T-cells [47–50] and in macrophage migration [39] . Similarly , we show that deletion of Gαi2 inhibits the migration of DCs towards a CCL19 gradient . In line with recent publications , deletion of Gαi3 had minor effects on directed DC migration [39 , 51] . However , it is notable that wt-Salmonella infection was still able to inhibit directed migration even in Gnai2-/- DCs , possibly indicating that Gαi3 is able to substitute in part for Gαi2 . When wild type Gαi2 was re-expressed in Gnai2-/- DCs , directed migration was partially rescued . Importantly , expression of the mutant Gαi2-Q205E that mimics the action of SseI , inhibited DC migration . Moreover , our studies reveal that SseI-induced deamidation of Gi proteins inhibits directed migration but does not affect the migration speed . Also pertussis toxin was shown to inhibit directed migration of DCs without affecting migration speed [52] . However , in contrast to SseI , pertussis toxin inactivates Gi proteins by blocking the interaction with GPCRs . Thus , how might SseI , which activates Gi proteins , inhibit directed cell migration ? McLaughlin and coworkers demonstrated that SseI binds to the scaffold protein IQGAP1 ( Ras GTPase-activating-like protein ) [13] , which interacts with numerous proteins ( e . g . Rho proteins ) involved in regulation of actin dynamics [53] . Because IQGAP1 also binds to the inactive SseI mutant , this cannot explain the action of SseI [13] . Earlier studies , which were mainly based on two-hybrid screen analyses , found interactions of SseI with the actin-binding protein filamin and the LIM domain fragment of TRIP6 [54 , 55] . However , these studies did not consider an enzyme activity of SseI . We propose that inhibition of directed migration by SseI is caused by blockade of the GTPase cycle and the non-polarized persistent activation of Gi . SseI blocks rapid deactivation and activation of Gi , which is probably essential to follow a chemoattractant gradient . In line with this hypothesis are recent findings about RGS ( regulators of G protein signaling ) proteins in immune cells . RGS proteins decrease the activity of G proteins by accelerating GTP hydrolysis and , thereby , accelerate G protein inactivation but also its activation [56–58] , which appears to be essential for chemotaxis [8 , 37] . It was shown that immune cells , harboring the Gαi2 G184S mutant , which causes resistance to RGS proteins , exhibit major alteration in chemotaxis [8] . The active state of the G184S Gαi2 mutant is prolonged , a condition mimicking the effects of SseI that persistently activates Gi proteins . Thus , a similar scenario as seen with the G184S Gαi2 mutant but even stronger , might be induced by SseI , resulting in insufficient downstream signaling from GPCRs-involved in chemotaxis and loss of polarization necessary for directed migration . In line with this hypothesis are also previous findings obtained with constitutively activated PI3Kγ , which was fused with the CAAX box of K-Ras resulting in persistent cell membrane localization and non-polarized enhancement of PIP3 production [59] . Leukocytes with constitutively activated PI3Kγ lost cell polarity and failed to follow a chemotactic gradient but they did not show differences in mean migration velocity . Surve and coworkers reported that selective activation of Gβγ by compound 12155 causes directed migration of neutrophils [60] . This effect was inhibited by Wortmannin , underlining the essential involvement of PI3K . PI3K is able to activate Rac , which is crucially involved in polarity control during neutrophil chemotaxis [61] . A recent study performed with optogenetic activation of PI3K sheds some light on the complexity of signaling pathways involved in directed migration downstream of PIP3 [62] . The authors showed that PI3K activates Rac mainly via the guanine nucleotide exchange factor P-Rex-1 . However , Rac activation by chemoattractants ( involving GPCRs and not only PI3K ) depended on multiple Rac regulators indicating involvement of a signaling network . Taken together our data elucidate the molecular mode of action of the SPI-2 effector SseI by deamidation of a crucial glutamine residue in the α-subunits of Gi proteins of the host , which results in inhibition of the turn-off mechanism of the G protein . The finding that Gi proteins are persistently activated by SseI will offer a new perspective in studies on the biological effects of this SPI-2 T3SS effector and may help to identify further signal factors involved in chemotaxis , acting downstream of Gi proteins .
PCR primers were from Apara ( Denzlingen , Germany ) . All other reagents were of analytical grade and purchased from commercial sources . All Antibodies used in this study are listed in S1 Table . S . enterica serovar Typhimurium NCTC12023 ( ATCC 14028 ) was used as parent strain for the ΔsseI strain , which was created by deletion of the SseI coding sequence [63] . Murine BMDMs and DCs were isolated and cultivated as described before [64 , 65] with adjustments . 6–12 weeks old wt ( Charles River ) , Gnai2-/- , and Gnai3-/- ( Dr . B . Nürnberg , maintained at animal facility , Medizinische Fakultät und ICePhA , University of Tübingen ) mice ( C57BL/6N , female , [66] ) were euthanized and femurs and tibiae were extracted . Bones were cut open and bone marrows were flushed with sterile , ice-cold PBS . Cells and matrix were resuspended and pipetted through a 70 μm cell strainer . Cell suspension was centrifuged ( 300 x g , 5 min ) . Supernatants were discarded , pellets were resuspended in RPMI-1640 with 10% fetal calf serum ( FCS ) and 1% penicillin/streptomycin ( P/S ) . Dendritic cells: Cell concentration was adjusted to 2 . 5 x 105 cells/ml and 10 ml suspension was added into 10 cm bacterial petri dishes . GM-CSF was added for a final concentration of 20 ng/ml . At day 3 , 10 ml medium with 20 ng/ml GM-CSF was added . At day 6 , 10 ml medium were carefully removed and 10 ml fresh medium with 20 ng/ml GM-CSF were added . At day 8–9 , non-adherent cells were collected and centrifuged ( 300 x g , 5 min ) . Pellets were resuspended in 10 ml fresh medium with 20 ng/ml GM-CSF without P/S and transferred into 6 cm cell culture dishes . LPS ( 200 ng/ml ) were added for 24 h for further maturation of the DCs , when needed . Only non-adherent cells were used for experiments . Bone marrow-derived macrophages: Cell concentration was adjusted to 5–7 x 105 cells/ml and 10 ml suspension was added into 10 cm tissue culture dishes . M-CSF was added for a final concentration of 20 ng/ml . At day 2 , 10 ml fresh medium with 20 ng/ml M-CSF were added . At day 4 , 10 ml medium were removed and exchanged by 10 ml fresh medium with 20 ng/ml M-CSF . At day 6–8 , medium and non-adherent cells were discarded , adherent cells were detached with trypsin/EDTA and plated in medium with 20 ng/ml M-CSF without P/S at desired concentrations for experiments . HEK-293 ( female; ACC 305 from DSMZ , Germany ) and HeLa ( female , ATCC CCL-2 ) cells were cultured in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% FCS and 1% P/S in an atmosphere of 5% CO2 at 37°C . RAW264 . 7 cells ( male , ATCC TIB-71 ) were cultured in DMEM supplemented with 5% FCS and 1% P/S . SseI encoding DNA was amplified from S . enterica serovar Typhimurium NCTC12023 and cloned into the bacterial expression vector pCOLDII and mammalian expression vectors pEGFP-C1 and pcDNA3 . 1 by standard cloning techniques . Additionally , a C-terminal fragment of SseI harboring the homology domain to PMT-C3 , starting with amino acid 137 was cloned into pCOLDII . Chimeric PMT-SseIC consists of the N-terminal portion ( amino acid 1–505 ) of PMT and the PMT-C3 homology domain of SseI ( amino acids 137–322 ) . PMT-SseIC was cloned in accordance to the method described before for other PMT fusion proteins into the pGEX2T vector [26] . The biological inactive mutant of SseI ( C178A ) was constructed by site-directed mutagenesis . Gαi2 DNA ( with optimized codon usage for prokaryotic expression ) was ordered as gBlock ( Integrated DNA Technology , IDT ) and cloned into pCOLDII . All oligonucleotides used in this study are listed in S2 Table . HeLa cells were transfected using PEI as described previously [67] . Transfection of RAW264 . 7 cells was performed utilizing lipofectamin2000 ( Thermo Fisher Scientific ) according to the manufacturer’s instructions . Transfection of BMDMs and BMDCs was done using amaxa nucleofector kits ( Lonza ) according to manufacturer’s instructions . For in vitro infection studies S . enterica serovar Typhimurium NCTC12023 and the isogenic variant MvP393ΔsseI were used . BMDMs or RAW 264 . 7 cells were infected with Salmonella with indicated MOIs for 30 min . Then cells were washed 3 times with warm PBS and incubated with DMEM ( RAW264 . 7 ) or RPMI-1640 ( BMDMs ) supplemented with 100 μg/ml gentamicin for 60 min . For the remaining time of the experiment , the medium was changed to DMEM or RPMI supplemented with 10 μg/ml gentamicin . For immunofluorescence detection of deamidation in RAW264 . 7 macrophages , these cells were infected with Salmonella for 5 h ( MOI = 1 ) . Dendritic cells were infected with a MOI = 30 for 30 min . Due to non-adherence of stimulated DCs , cells were resuspended in infection-medium and centrifuged ( 300 x g / 5 min ) . Pellets were resuspended in RPMI-1640 containing 100 μg/ml gentamicin and further treatment of the cells was similar to treatment of BMDMs . For lysates cells were treated with RIPA buffer ( 1 mM EDTA , 25 mM Tris , 150 mM NaCl , 1% ( v/v ) Triton X-100 , and 1% ( m/v ) sodium deoxycholate , pH 7 . 4 ) , containing complete protease inhibitor ( Roche ) and phosphatase inhibitor ( phosphatase inhibitor cocktail 2/3 , Sigma ) for 20 min on ice with occasional vortexing . Lysates were then centrifuged at 4°C ( 10 , 000 rpm for 10 min ) . cAMP was measured by fluorescence resonance energy transfer ( FRET ) using a single chain cAMP sensor as described before [27] . The construct EPAC2-camps was kindly provided by Dr . Viacheslav O . Nikolaev ( University of Göttingen , Germany ) . HEK-293 cells were incubated with wildtype PMT-SseI or inactive mutant ( PMT-SseI-C178A ) for 8 h . Then forskolin ( 10 μM ) and 3-isobutyl-1-methylxanthin ( IBMX 100 μM ) were added to the medium to induce adenylyl-cyclase activity and cells were incubated for further 45 min . Medium was discarded and cells were lysed with lysis buffer from the cAMP Parameter Kit ( Biotechne ) . Further treatment of the samples was performed according to the manufacturer’s instructions ( cAMP Parameter Kit/R&D Systems Biotechne ) . The subcellular distribution of GFP-Grp1PH reflects the activity of PI3Ks , i . e . enhancement of membrane-associated fluorescence was taken as an indicator of PI3K activation . Assay was performed as previously described [31] . For fluorescence microscopy , cells were prepared as described before [26] . Fixed samples were analyzed with an inverted Axiovert 200 M microscope ( Carl Zeiss ) with a Yokogawa ( Tokyo , Japan ) CSU-X1 spinning disc equipped with an emission filter wheel and 405 nm / 488 nm / 561 nm laser lines . SseI and Gαi2 ( His ) were expressed as N-terminal His6-tagged proteins and purified by affinity chromatography via a Ni-NTA column . Expressions of PMT or chimeric constructs of PMT and SseI were performed as glutathione-S-transferase fusion proteins in accordance with the method previously described [67] . Coexpression of SseI or the inactive mutant ( SseI-C178A ) with GST-Gαi2 was done as previously described [23] . G proteins used for the in vitro deamidation assay in S4A Fig were purified as GST-fusion proteins . The GST-tag was cleaved with Thrombin . Purified Gαi2 or Gαi3 ( 1 μM ) was preincubated in deamidase buffer ( 50 mM sodium HEPES ( pH 7 . 5 ) , 50 mM NaCl , 1 mM EDTA , 5 mM MgCl2 and 0 . 1% NP-40 ) together with Guanosine diphosphate ( GDP , 5 μM ) for 15 min at 30°C . SseI-C deamidase domain ( 100 nM ) was added and reaction was performed at 16°C for 16 h . Reaction was stopped by the addition of 5x Lämmli buffer and shock freezing in liquid N2 , followed by immunoblot analysis using the deamidation specific GαQE antibody . For immunoblotting , samples were subjected to SDS-PAGE and transferred onto polyvinylidene difluoride-membrane . Deamidation specific antibody anti-Gαq Q209E ( 3G3 ) was used as described before [25] . Binding of the appropriate horseradish peroxidase-coupled secondary antibody was detected by using enhanced chemiluminescent detection reagent ( New England Biolabs ) and the imaging system LAS-3000 ( Fujifilm ) . Gel bands were destained for 10 min using 30% acetonitrile / 70% 0 . 1 M NH4HCO3 , followed by 10 min washing with 0 . 1 M NH4HCO3; this procedure was repeated twice . Prior to drying in a Vacuum Concentrator ( Concentrator 5301 , Eppendorf ) , gel bands were shrunk with 100% acetonitrile . By applying 0 . 1 μg trypsin per gel band , proteolytic digest was performed overnight at 37°C in 0 . 1 M NH4HCO3 . Peptides were extracted from the gel matrix with 5% formic acid . Peptides were analyzed by LC/MS on a Q-TOF mass spectrometer ( Agilent 6520 , Agilent Technologies ) coupled to a 1200 Agilent nanoflow system via a HPLC-Chip cube electrospray ionization interface carrying a ProtID-II-Chip ( Agilent Technologies ) . For peptide separation , a linear gradient ranging from 3% acetonitrile ( ACN ) to 30% ACN running for 30 min at a flow rate of 300 nl/min was used . MS spectra were acquired from m/z 50 to m/z 3000 while operating the Q-TOF in the 2 GHz extended dynamic range mode . Internal mass calibration was enabled . Three multiple charged peptides were selected in a data-dependent manner from each survey MS-scan for acquisition of fragment mass spectra . For peak list generation , raw data files were processed using Mascot Daemon2 . 4 . 0 with default settings for the Agilent Q-TOF . Using Mascot Server 2 . 4 . 1 database searches were performed with tryptic specificity and two missed cleavages against a small database comprised of known contaminants and the target sequence . Mass tolerances of 50 ppm and 0 . 05 Da were allowed for peptide and fragment masses , respectively . As variable modifications deamidation of glutamine and phosphorylation of serine and threonine were selected in addition to Carbamidomethyl ( C ) , Gln-> pyroGlu ( N-term . Q ) and oxidation ( M ) . Extracted Ion Chromatograms were generated using the Qualitative Analysis module within the MassHunter B04 . 00 software . RAW264 . 7 macrophages were infected with a MOI of 30 . 1 . 5 h p . i . medium was exchanged by Earle’s balanced salt solution ( EBSS ) containing 10 μg/ml gentamicin and LY29 ( 30 μM ) , where indicated . Further treatment of the cells was performed according to the manufacturer’s instructions ( Promega CytoTox-ONE/CellTiter-Blue ) . In-plate fluorescence was acquired with the Tecan Infinite M200 plate reader . Chemotactic migration assays of dendritic cells were performed as described before [64] . Briefly , pre-treated dendritic cells were mixed with bovine collagen ( 1 . 5 mg/ml ) and incubated in self-built migration chambers at 37°C , 5% CO2 until the collagen network was polymerized . The collagen-cell mixture was covered with medium , containing CCL19 ( 500 ng/ml ) , which diffuses into the gel matrix and thus forming the chemokine gradient . Cell migration was monitored with a Lionheart FX automated microscope . Cell tracks were analyzed with ImageJ and the Manual Tracking plugin , followed by evaluation of the data with Ibidi chemotaxis and migration tool . Calculation of the chemotactic indices: Chemotacticindex=1n∑i=1nyi ( distancemigratedparalleltochemokinegradient ) di ( accumulatedmigrationdistance ) Results are presented as means ±SEM , unless otherwise stated . Statistical analyses were performed using GraphPad PRISM Version 5 . 04 . Significance was assessed by Student`s t-test and Mann-Whitney test , depending on whether the data was normally distributed . p values < 0 . 05 were considered statistically significant ( * = p < 0 . 05; ** = p < 0 . 01;*** = p < 0 , 001; ns , not significant ) . Multiple group comparisons were analyzed by ANOVA and Bonferroni post-tests . All animal experiments were performed in compliance with the German animal protection law ( TierSchG ) . The animals were housed and handled in accordance with good animal practice as defined by FELASA ( www . felasa . eu ) and the national animal welfare body GV-SOLAS ( www . gv-solas . de ) . The animal welfare committees of the University of Freiburg as well as the local authorities ( Regierungspräsidium Freiburg , licenses X-13/03A and X-17/01F ) approved all animal experiments . | Salmonella Typhimurium is one of the most common causes of gastroenteritis in humans . In immunocompromised patients , the pathogen can cause systemic infections . Crucial virulence factors are encoded on two Salmonella pathogenicity islands SPI-1 and SPI-2 . While SPI-1 encodes virulence factors essential for host cell invasion , intracellular proliferation of the pathogen depends mainly on SPI-2 effectors . Here , we elucidate the mode of action of Salmonella SPI-2 effector SseI . SseI activates heterotrimeric G proteins of the Gαi family by deamidation of a specific glutamine residue . Deamidation blocks GTP hydrolysis by Gαi , resulting in a persistently active G protein . Gi activation inhibits cAMP production and stimulates PI3Kγ by Gαi-released Gβγ subunits , resulting in activation of survival pathways by phosphorylation of Akt and mTOR . Moreover , deamidation of Gαi leads to a loss of directed migration in dendritic cells . The data offers a new perspective in the understanding of the actions of SseI . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"cell",
"motility",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"molecular",
"probe",
"techniques",
"pathogens",
"immunology",
"microbiology",
"directed",
"cell",
"migration",
"salmonellosis",
"bacterial",
"diseases",
"developmental",
"biology",
"enterobacteriaceae",
"molecular",
"biology",
"techniques",
"bacteria",
"bacterial",
"pathogens",
"salmonella",
"typhimurium",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"white",
"blood",
"cells",
"immunoblot",
"analysis",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"salmonella",
"chemotaxis",
"biochemistry",
"cell",
"biology",
"post-translational",
"modification",
"deamidation",
"cell",
"migration",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
] | 2018 | Salmonella Typhimurium effector SseI inhibits chemotaxis and increases host cell survival by deamidation of heterotrimeric Gi proteins |
The type I interferon ( IFN ) response is a crucial innate immune signalling pathway required for defense against viral infection . Accordingly , the great majority of mammalian viruses possess means to inhibit this important host immune response . Here we show that vaccinia virus ( VACV ) strain Western Reserve protein C6 , is a dual function protein that inhibits the cellular response to type I IFNs in addition to its published function as an inhibitor of IRF-3 activation , thereby restricting type I IFN production from infected cells . Ectopic expression of C6 inhibits the induction of interferon stimulated genes ( ISGs ) in response to IFNα treatment at both the mRNA and protein level . C6 inhibits the IFNα-induced Janus kinase/signal transducer and activator of transcription ( JAK/STAT ) signalling pathway at a late stage , downstream of STAT1 and STAT2 phosphorylation , nuclear translocation and binding of the interferon stimulated gene factor 3 ( ISGF3 ) complex to the interferon stimulated response element ( ISRE ) . Mechanistically , C6 associates with the transactivation domain of STAT2 and this might explain how C6 inhibits the type I IFN signalling very late in the pathway . During virus infection C6 reduces ISRE-dependent gene expression despite the presence of the viral protein phosphatase VH1 that dephosphorylates STAT1 and STAT2 . The ability of a cytoplasmic replicating virus to dampen the immune response within the nucleus , and the ability of viral immunomodulators such as C6 to inhibit multiple stages of the innate immune response by distinct mechanisms , emphasizes the intricacies of host-pathogen interactions and viral immune evasion .
The innate immune response is the first line of defense against invading pathogens . Interferons ( IFNs ) are one of the key players in this early response to infection and are particularly important to protect against viruses , as can be seen by the increased susceptibility of IFNα/β receptor ( IFNAR ) knock out mice to viral infections [1] . There are two main branches to the IFN response; their production and the signalling initiated in response to the binding of secreted IFNs to their receptors at the cell surface . Type I IFNs , which include IFNβ , several IFNα variants and other tissue or species-specific members , are produced directly in response to virus detection by cellular pattern recognition receptors ( PRRs ) . Upon recognition of pathogen associated molecular patterns ( PAMPs ) such as viral DNA or RNA , PRRs activate several signalling pathways many of which converge on the kinases TANK-binding kinase ( TBK1 ) and IκB kinase-ε ( IKKε ) . These kinases , in complex with adaptor proteins such as TANK , NAK-associated protein 1 ( NAP1 ) or similar to NAP1 TBK1 adaptor ( SINTBAD ) , phosphorylate the transcription factor IFN regulatory factor 3 ( IRF-3 ) . Once phosphorylated , IRF-3 dimerises and translocates into the nucleus and , in combination with other transcription factors , drives transcription from promoters containing cognate binding sites , including the IFNβ promoter [2] . Once produced and secreted from cells , type I IFNs can act in a paracrine or autocrine fashion by binding to the IFNAR , which is composed of the two subunits IFNAR1 and IFNAR2 . The binding of type I IFN to the receptor complex leads to the cross activation of the two Janus protein kinases , Tyk2 and Jak1 that are bound to the cytoplasmic domains of the IFNAR1 and IFNAR2 , respectively . Once activated these kinases phosphorylate the transcription factors signal transducer and activator of transcription 1 ( STAT1 ) and STAT2 . These phosphorylated proteins then heterodimerise and bind to IRF-9 to form the IFN stimulated gene factor 3 ( ISGF3 ) transcriptional activator complex . This tripartite complex translocates into the nucleus where it binds to IFN stimulated response elements ( ISREs ) found in the promoter of IFN stimulated genes ( ISGs ) and induces their transcription . The type I IFN signalling pathway and its regulation is reviewed in [3] . The importance of the IFN response for protection against viral infections is illustrated by the array of mechanisms and proteins used by viruses to evade and inhibit these cellular pathways , reviewed in [4] . Vaccinia virus ( VACV ) is a well-studied member of the Poxviridae and was the vaccine used in the eradication of smallpox [5] . It is a large DNA virus , with approximately 200 genes , that replicates exclusively in the cytoplasm of infected cells [6] . Between one third and one half of these 200 genes have been shown to have immunomodulatory or immunoevasive roles [7 , 8] . Many of these immunomodulatory proteins are able to inhibit type I IFN production , either through inhibition of the NF-κB pathway , for example VACV proteins B14 [9] and A49 [10] , or through inhibition of the IRF-3 signalling pathway , as with VACV proteins A46 [11] and K7 [12] . In contrast , very few inhibitors have been identified that act post-IFN production . To date two VACV proteins are known to inhibit IFN signalling after type I IFN has been secreted from cells . B18 is a secreted VACV protein that binds type I IFN in solution and on the surface of cells and prevents its interaction with the IFNAR [13–15] and VH1 is a virally-encoded phosphatase that is packaged within virions [16] and dephosphorylates both STAT1 and STAT2 , therefore acting as an intracellular inhibitor of JAK/STAT signalling [17 , 18] . C6 is a predicted member of the VACV Bcl2-like protein family , a family of 10 proteins whose previously studied members have various innate immune inhibitory functions [19–25] . It is expressed early during infection and its deletion attenuates the virus in both intranasal and intradermal models of infection in the mouse [26] . Despite being attenuated , VACV strains engineered to lack C6 showed enhanced immunogenicity in vivo [27 , 28] . Previously , C6 was shown to bind to the TBK1/IKKε adaptor proteins , SINTBAD , NAP1 and TANK , and prevents the TBK1/IKKε-dependent activation of IRF-3 and therefore inhibits the induction of type I IFNs [26] . Given many VACV proteins have been shown to have multiple functions , for example N1 that inhibits both NF-κB signalling [29] and apoptosis [23 , 30] , and the observation that the hitherto only known function of C6 occurs in the cytoplasm of infected cells despite C6 being present in the nucleus and cytoplasm , we investigated whether C6 may have additional immunomodulatory functions . In this study , VACV protein C6 is shown to be a dual function protein that inhibits type I IFN signalling as well as type I IFN production . Data presented show that the inhibition of IFNα-induced JAK/STAT signalling occurs at a late stage in the pathway , downstream of STAT phosphorylation , heterodimerisation , and nuclear translocation and downstream of ISGF3 binding to the ISRE , thus indicating an inhibitory function of C6 at or after formation of the transcriptional complex . Furthermore , C6 is shown to associate with the transcriptional activating domain ( TAD ) of STAT2 , providing a plausible mechanism by which this viral protein could disrupt transcriptional complex formation . Interestingly , C6 was able to inhibit the transcriptional induction of all but one of the IFNα-dependent genes tested . This indicates that the step ( s ) inhibited by C6 downstream of ISGF3-ISRE interaction is likely to be conserved for a large number of ISGs , rather than a gene-specific transcriptional requirement . To our knowledge , this additional function of C6 makes C6 the first nuclear inhibitor of the IFN response encoded by the cytoplasmic-replicating DNA virus , VACV . The ability of VACV , a virus whose replication cycle takes place entirely within the cytoplasm of infected cells , to extend its influence into the host cell nucleus to inhibit crucial innate immune signalling pathways , highlights the complexity of virus-host interactions .
Previously , C6 was characterised as an inhibitor of IFNβ production through inhibition of the IRF-3/7 signalling pathway [26] . However , many viral proteins are known to have multiple functions . To determine whether C6 was able to also inhibit signalling downstream of type I IFNs , its effect on the expression of an ISRE-dependent luciferase reporter gene ( ISRE-luciferase ) was assessed . HeLa and HEK293T cells were co-transfected with expression plasmids for ISRE-luciferase and V5- or TAP- tagged C6 or control proteins ( S1 Fig ) . Cells were then stimulated with IFNα for 8 h and the expression of luciferase was measured by luminescence . Treatment of HEK293T and HeLa cells with IFNα led to an induction of luciferase expression and in both cell types this induction was significantly inhibited by the co-expression of either TAP-C6 ( p<0 . 0001 and p<0 . 01 respectively ) or C6-V5 ( p<0 . 001 and p<0 . 001 respectively , Fig 1A and 1B ) . As expected the positive controls Nipah Virus V protein ( NiV-V ) and Parainfluenza virus 5 V protein ( PiV5-V ) that are known to inhibit IFN signalling [31 , 32] also inhibited IFNα-induced luciferase expression , whilst expression of GFP or VACV protein B14 had no inhibitory effect . To confirm the ability of C6 to inhibit type I IFN signalling , the effect of C6 on the induced transcription of endogenous ISGs was assessed next . HeLa cells stably expressing GFP alone ( EV ) or GFP in combination with V5 tagged- C6 , PiV5-V or B14 were stimulated with IFNα for 8 h . RNA was extracted from these cells and used for qPCR analysis of ISG mRNA expression . IFNα treatment of cells resulted in induction of the well-characterised ISGs tested , including interferon-induced protein with tetratricopeptide repeats 1 ( IFIT1 ) , IFIT3 and MxA . The presence of C6 significantly inhibited the induction of gene expression when compared to both the EV and B14 controls ( Fig 1C ) . Once again PiV5-V expression also inhibited the induction of ISG expression as expected ( Fig 1C ) . Finally , these stably transduced cells were used to confirm the ability of C6 to inhibit IFNα-induced gene expression at the protein level using flow cytometry analysis of IFIT1 expression . Cells were stimulated with IFNα for 8 h and then fixed , permeabilised and stained with an anti-IFIT1 antibody . Stained cells were analysed by flow cytometry to assess for IFIT1 protein expression . Once again treatment of EV transduced cells with IFNα resulted in an induction of IFIT1 expression , which was significantly inhibited by expression of C6 or the positive control , PiV5-V ( p<0 . 01 , Fig 1D ) . In contrast , expression of B14 had no effect on IFIT1 expression . Together these data show that C6 inhibits IFNα mediated gene expression at both the mRNA and protein level . The IRF-3 signalling pathway is initiated in response to detection of viral RNA , DNA or proteins by PRRs in the cell and leads to the activation of the kinases TBK1 and IKKε . These kinases then phosphorylate IRF-3 causing its translocation into the nucleus from where it drives transcriptional activation of a number of target genes including IFNβ . C6 inhibits IRF-3 signalling at the level of TBK1 and IKKε , preventing the nuclear translocation of IRF-3 [26] . Several studies have described potential crosstalk between the IRF-3 and JAK/STAT signalling pathways , which together constitute the type I IFN response . In addition , IRF-3 is known to cause the transcriptional activation of a subset of ISRE containing gene promoters directly [33] and an additional phosphorylation event on STAT1 by IKKε is required for the full induction of approximately 30% of IFNα-responsive ISGs [34] . To rule out that the inhibitory effect of C6 on type I IFN signalling was an indirect consequence of its ability to interfere with TBK1 and IKKε function , and that this inhibitory activity was instead a novel function for C6 , the effect of a TBK1/IKKε specific inhibitor , BX795 , on the ISRE-dependent reporter gene assay was assessed . Cells transfected with the ISRE-luciferase reporter were treated for 3 h with BX795 before stimulation with IFNα for a further 6 h in the continued presence of BX795 . BX795 treatment had no inhibitory effect on the IFNα-dependent induction of firefly luciferase in the cells tested , when compared to a carrier ( dimethyl sulphoxide , DMSO ) -treated control ( S2 Fig ) . To confirm the effectiveness and specificity of BX795 in this assay , BX795 treatment was also used in conjunction with firefly luciferase under the control of promoters responsive to the IRF-3 and NF-κB signalling pathways . Cells transfected with these luciferase reporters were stimulated with polyI:C and IL-1β respectively . Whilst treatment with BX795 had no effect on IL-1β-induced activation of the NFκB reporter gene , the induction of the IRF-3-responsive promoter by poly I:C was significantly reduced ( p<0 . 001 ) , indicating that the inhibitor was effective at the doses used ( S2 Fig ) . These data indicate that inhibition of TBK1 and IKKε does not cause the inhibition of ISRE-dependent gene expression that is seen in the presence of C6 , implying that this is instead a novel function of C6 . To determine how C6 inhibits the cellular response to IFNα , the phosphorylation status of STAT1 and STAT2 in cells expressing C6 following IFNα treatment was examined . HeLa cells stably expressing C6 or control proteins were stimulated with IFNα for 45 mins . Cell lysates were used for immunoblot analysis of levels of phosphorylated STAT1 and STAT2 , including quantification of proteins using Odyssey software ( LICOR ) . Neither C6 nor the negative control protein B14 inhibited the IFNα-induced phosphorylation of STAT1 or STAT2 , whereas PiV5-V protein inhibited the phosphorylation of both proteins as expected ( Fig 2 ) . Following phosphorylation , STAT1 and STAT2 heterodimerise and bind to IRF9 to form the ISGF3 complex . This complex then translocates into the nucleus where it drives expression of genes with an ISRE in their promoter . To determine whether C6 prevents the dimerisation of STAT1 and STAT2 , an immunoprecipitation of STAT1 from IFNα-stimulated cells was performed in cells transiently transfected with C6 or N1 , a VACV Bcl-2-like protein that does not inhibit IFNα signalling . Immunoblot analysis showed that following IFNα stimulation an increased amount of STAT2 was found associated with STAT1 . However , C6 did not alter the amount of STAT2 bound to STAT1 either before or after stimulation ( Fig 3 ) . Following its formation , the ISGF3 complex translocates into the nucleus . To assess whether this nuclear translocation was inhibited by C6 the localization of STAT1 and STAT2 before and after IFNα stimulation was assessed by confocal microscopy . HeLa cells stably expressing GFP alone ( EV ) or in combination with V5-tagged C6 or PiV5-V were stimulated with 500 U/ml IFNα for 1 h . Cells were then fixed , permeabilised and stained for endogenous STAT1 ( Fig 4A ) or STAT2 ( Fig 5A ) . Following IFNα stimulation the percentage of cells showing a nuclear stain for STAT1 and STAT2 increased to approximately 50% ( Fig 4B ) and 70% ( Fig 5B ) , respectively . These localization patterns were not altered by C6 or the negative control , EV . The localisation of STAT1 could not be assessed in PiV5-V expressing cells due to its degradation by this viral protein , however , PiV5-V expression inhibited the translocation of STAT2 completely ( Fig 5A ) . Together these data indicate that C6 does not inhibit the pathway prior to ISGF3 complex formation and nuclear translocation . To obtain additional evidence that the inhibitory effect of C6 on IFNα signalling is downstream of ISGF3 complex formation , a plasmid encoding IRF-9 fused to the C-terminal region ( amino acids 747–851 ) of the transcriptional activation domain of STAT2 ( referred to as IRF9-S2C ) was utilised . Previously , this fusion protein , which overcomes the need for ISGF3 complex formation , has been shown to act as a constitutively active ISGF3-like transcriptional activator in the absence of IFN stimulation [35] . When a plasmid encoding IRF9-S2C was co-transfected into HEK293T cells along with the ISRE-luciferase reporter gene , a large increase in firefly luciferase expression was observed in cells expressing IRF9-S2C relative to those transfected with empty vector ( EV ) only ( Fig 6 , columns 1 and 2 ) . Interestingly , when co-transfected into cells , C6 inhibited IRF9-S2C-driven ISRE reporter activity significantly ( p<0 . 0001 ) , whereas , neither PiV5-V protein nor NiV-V protein showed any inhibitory activity ( Fig 6 ) . This is in keeping with the known ability of these two viral proteins to inhibit the IFNα signalling pathway upstream of ISGF3 complex formation [31 , 32] . These data confirm that C6 inhibits IFNα signalling at a late stage following ISGF3 complex formation . To establish whether C6 inhibited IFNα signalling by preventing the binding of the ISGF3 complex to the ISRE in promoters of ISGs , the ability of the IRF9-S2C fusion protein to bind the ISRE was assessed . HEK293T cells were transfected with plasmids expressing IRF9-S2C and V5-tagged C6 or control proteins and cell lysates were harvested 16 h later . A biotin-labelled ISRE probe optimised previously for ISGF3 binding ( ISREcore ) [36] , or a control biotin-labelled ISRE sequence that was shown to lack ISGF3 binding ( ISRErandom ) [36] were incubated with cell lysates and streptavidin beads were used to immunoprecipitate the biotinylated DNA probe and associated proteins . C6 or GFP expression did not inhibit the binding of either IRF9-S2C or endogenous STAT2 to the biotin-labelled ISRE ( Fig 7 ) . In contrast , NiV-V protein was able to inhibit endogenous STAT2 binding to the ISRE but not binding of the IRF9-S2C construct as expected ( Fig 7 ) . Neither IRF9-S2C nor endogenous STAT2 bound to the ISRErandom control sequence as expected . Ku70 , a known DNA binding protein , was used here as a control for DNA input and gel loading and was found to bind to both the ISREcore and ISRErandom DNA probes . To investigate whether C6 interacts with any of the components of the ISGF3 complex , an immunoprecipitation assay using FLAG-tagged STAT1 , STAT2 and IRF-9 was performed . Plasmids expressing these proteins were co-transfected into HEK293T cells along with a V5-tagged C6 expression vector . Immunoprecipitation with anti-FLAG beads co-precipitated C6 with STAT2 but not with STAT1 or IRF-9 ( Fig 8A ) . To determine if the interaction between C6 and STAT2 was affected by IFN stimulation , HeLa cells were transfected with plasmids expressing HA-tagged STAT2 and either TAP-tagged C6 or N1 then mock-stimulated or stimulated with IFNα prior to immunoprecipitation ( Fig 8B ) . This showed that the interaction between C6 and STAT2 did not require prior stimulation with IFN . To confirm the interaction between C6 and STAT2 and to ascertain whether it occurs at endogenous protein levels and during viral infection , HEK293T cells were infected with VACVs expressing TAP-tagged C6 or N1 under the natural promoters for these genes . Cells were then lysed and immunoprecipitations were performed against the FLAG epitope in the TAP tag of these viral proteins . Once again C6 associated with STAT2 and not STAT1 , whilst N1 did not associate with either protein ( Fig 8C ) , confirming the specific interaction between STAT2 and C6 at endogenous protein levels during viral infection . As C6 inhibits IFNα signalling initiated by IRF9-S2C expression , its ability to interact with this fusion protein , which contains only the C-terminal 104 aa of the STAT2 transactivation domain , was assessed . To this end , HEK293T cells were co-transfected with either IRF9-S2C or HA-IRF-9 and C6-TAP or N1-TAP . Immunoprecipitation with anti-FLAG beads showed association between IRF9-S2C and C6 but not between C6 and HA-IRF-9 ( Fig 8D ) . This indicates that C6 associates with the final 104 aa of the STAT2 transactivation domain , a region known to be important for recruitment of downstream chromatin modifying enzymes and transcriptional machinery . Finally , the biological importance of C6 for inhibition of the JAK-STAT pathway leading to activation of the ISRE promoter was assessed during VACV infection . The activity of C6 in blocking this pathway was likely to be masked to some degree during infection by the presence of the virus phosphatase VH1 , which dephosphorylates STAT1 and STAT2 and is delivered into cells by the invading virion immediately after infection . Therefore , two methods were used to assess if C6 contributed to the inhibition of the JAK-STAT pathway during infection . One method was simply to transfect the ISRE-luciferase reporter plasmid into cells 16 h before the cells were infected with either wt VACV or the vΔC6 mutant and then measure ISRE-luciferase at different times p . i . Preliminary experiments established that infection of cells at 5 PFU/cell for 5 h was optimal before luciferase activity was measured in cell lysates . Under these conditions , infection by vΔC6 induced significantly greater luciferase activity than did wt VACV . Immunoblotting for VACV protein D8 showed that the virus infections were equivalent and immunoblotting for GAPDH showed equal loading of samples . This experiment was conducted using either crude or purified virus preparations ( n = 4 ) and in each case a significant difference between the viruses was observed ( Fig 9A ) . The second method exploited the ability of the IRF9-S2C protein to activate ISRE promoter within the nucleus and downstream of the position at which the VH1 phosphatase mediates inhibition of the JAK-STAT pathway . Optimisation experiments to determine the amount of the IRF9-S2C plasmid to transfect and the length of time after transfection prior to virus infection showed that this potent inducer of ISRE-dependent gene expression was best only transfected a few hours before infection because overnight transfection induced very high levels of luciferase activity . Transfection of the IRF9-S2C plasmid for 4 h induced a modest 2-fold induction in luciferase activity and this was increased further by virus infection ( Fig 9B ) . However , following virus infection for 5 h , wt VACV induced lower levels of luciferase than vΔC6 , as in Fig 9A . These significant differences were seen reproducibly in multiple experiments ( n = 4 ) . Immunoblotting confirmed equal infection and protein loading . Collectively , these data show that during VACV infection protein C6 is able to diminish expression from the ISRE promoter over and above the effect of the VH1 phosphatase .
Previously , C6 was identified as a VACV immunomodulator and virulence factor and was shown to inhibit the induction of type I IFNs through inhibition of the IRF-3/7 signalling pathway at the level of the TBK1/IKKε kinase complex [26] . This study identifies a second function for VACV protein C6 as an inhibitor of the cellular response to type I IFN . Data presented demonstrate that C6 inhibits IFNα-induced expression of ISGs at both the mRNA and protein level ( Fig 1 ) . The inability of a pharmacological inhibitor of TBK1/IKKε to reduce IFNα-induced reporter gene expression ( S2 Fig ) indicates that inhibition of this kinase complex by C6 is unlikely to explain the ability of C6 to also inhibit the cellular response to IFNα . Therefore , to elucidate the mechanism by which C6 has its inhibitory effect on this second pathway , the IFNα-induced phosphorylation and nuclear translocation of STAT1 and STAT2 were examined and C6 was found to have no effect on these early events of this signalling pathway ( Figs 2 , 4 and 5 ) . Furthermore , both endogenous STAT2 and a constitutively active ISGF3 mimic , IRF-9-S2C , were still able to bind to the ISRE in the presence of C6 ( Fig 7 ) , indicating C6 exerts its inhibitory effect after ISGF3 binding to the ISRE . Interestingly , C6 interacts with STAT2 ( Fig 8A–8C ) and the transactivation domain ( aa 747–851 ) of STAT2 fused to IRF-9 ( Fig 8D ) but not with STAT1 or IRF-9 ( Fig 8A ) . The STAT2 transactivation domain is known to be required for the recruitment of chromatin modifiers and transcriptional machinery [37] . Therefore , the ability of C6 to interact specifically with this domain gives insight into how this viral protein may inhibit this crucial signalling pathway at such a late stage . C6 is one of many VACV proteins that inhibit the IFN response , however , it is the first such protein known to inhibit both branches of the IFN response , inhibiting both IFNβ production and the cellular responses to type I IFN . C6 is also the first VACV protein identified to inhibit the response to type I IFN in the nucleus of infected cells; VACV inhibitors to date act early in the JAK/STAT signalling pathway , either extracellularly to prevent binding of secreted type I IFN to their receptor ( B18 ) [13–15] , or in the cytoplasm of infected cells to dephosphorylate activated STAT1 and STAT2 ( VH1 ) [17 , 18] . C6 instead acts at a very late stage in the pathway , after the ISGF3 complex has formed , translocated into the nucleus and bound to the ISRE . Despite this , its inhibitory action was evident on 6 out of 7 of the ISGs examined , suggesting that the protein or step it targets is required for the induction of many ISGs . The requirement for many protein inhibitors of a single , albeit important , signalling pathway is not well understood . However , VACV shows a similar ‘belt and braces’ approach to other signalling pathways , for example the NF-κB pathway , for which it is currently known to possess 10 inhibitors [9–12 , 23 , 29 , 30 , 38–42] . These proteins are not completely redundant however , as they cause virus attenuation in vivo when deleted individually [10 , 11 , 39 , 43–47] . Similarly , previous work has shown that deletion of C6 leads to an attenuated phenotype in both intradermal and intranasal models of VACV infection in mice [26] , despite the presence of other IFN-signalling inhibitors . The identification of a second function of C6 means that the observed attenuation of the C6 deletion virus cannot be attributed to a single function as yet , but highlights the importance of this immunomodulatory protein . Structure-based mutagenesis of C6 may enable the dissection of these different activities as was done for the related VACV protein N1 [30] . The use of multiple proteins to inhibit a single pathway may be explained in many ways , such as the possibility of incomplete inhibition by any one protein , or the requirement for different immunomodulators in different cell types or infection stages . It could also be explained by possible crosstalk between innate immune signalling pathways meaning that inhibition of a pathway at a certain point could be overcome by activation of another communicating immune signalling pathway . However , the late stage at which C6 inhibits the type I IFN response would suggest that cross talk from other pathways would still be unable to activate ISGF3-driven ISG transcription in the presence of C6 . The role of C6 in inhibiting the JAK-STAT pathway during virus infection may be masked to some degree by the effect of VH1 that dephosphorylates STAT1 and STAT2 rapidly after infection . VH1 is an essential gene for VACV replication [16] , preventing its elimination by genetic manipulation . Nonetheless analysis of ISRE-driven gene expression early after infection with either wild type virus or a mutant virus lacking C6 showed a functional role for the C6 protein in diminishing ISRE-dependent gene expression ( Fig 9 ) . The roles of VH1 and C6 are therefore complementary . VH1 is expressed late in infection and packaged into the virion , whereas C6 is an early VACV protein , expressed from approximately 2 h post infection . These differential expression patterns might explain the requirement of both inhibitors , perhaps with their relative importance differing over the life cycle of the virus . The principal transcriptional activating complex responsible for type I IFN-induced gene expression , ISGF3 , consists of three components , STAT1 , STAT2 and IRF-9 . In this complex , STAT2 provides a potent and essential transcriptional activation domain [48] . The mechanistic process by which ISGF3 , likely through this C-terminal domain of STAT2 , signals to and promotes transcription of ISGs by RNA polymerase II remains unclear . However , a number of cellular proteins known to have roles in transcription , such as components of the Mediator complex [49] , or chromatin modification , including histone deacetylases ( HDACs ) [50] , histone acetyltransferases ( HATs ) [51] and chromatin remodeling complexes [52–55] , have been identified as being essential for ISGF3-driven transcription . The precise mechanism by which C6 inhibits JAK/STAT signalling remains to be determined , however it is possible that the association of C6 with the STAT2-transactivation domain could prevent or alter the interactions of STAT2 with these or other cellular proteins required for ISG transcriptional induction . Indeed some such proteins have been shown to bind directly to the STAT2 transactivation domain , for example the HATs p300/CBP [51] and GCN5 [56] . In the future , further work is needed to assess the exact consequence of the C6-STAT2 TAD interaction . The importance of the IFN-signalling pathway in preventing viral replication and spread dictates that the majority of mammalian viruses have one or multiple mechanisms of inhibiting the response of infected cells to type I IFN . There is an array of different mechanisms by which viral proteins inhibit JAK/STAT signalling , some of which are reviewed in [4] . Many such mechanisms focus on inhibiting the early steps in the JAK/STAT signalling pathway , by degradation of either STAT1 or STAT2 as with PiV5 [32] and PiV2 V proteins [57] , respiratory syncytial virus NS1 and NS2 proteins [58] and dengue NS5 protein [59] , by inhibition of STAT phosphorylation as with Sendai virus C protein [60] , or by cytoplasmic sequestration of the ISGF3 complex as with NiV [31] and Hendra virus [61] V proteins . Fewer viral proteins have been identified that inhibit the later stages of this signalling pathway , once the ISGF3 complex has reached the nucleus . The human cytomegalovirus ( HCMV ) IE1 protein interacts with STAT2 and inhibits the binding of STAT2 and promyelocytic leukemia protein ( PML ) , a protein that associates with STAT1 , STAT2 , and HDACs , to ISG promoters [62] . Therefore , HCMV IE1 delivers its inhibitory action within the nucleus and via an interaction with STAT2 , but again upstream of the inhibitory action of C6 . Conversely , adenovirus E1A protein has its inhibitory effect , like C6 , downstream of ISGF3- promoter binding but does so through interacting with and preventing the functioning of HATs and histone ubiquitylating complexes required for full ISG transcriptional activation and not through a direct interaction with STAT2 [63] . Similarly , influenza A virus nonstructural protein 1 ( NS1 ) inhibits the cellular response to type I IFNs in the nucleus through interaction with a complex involved in transcriptional elongation , hPAF1C and once again not by a direct interaction with STAT2 [64] . Interestingly , NS1 inhibits type I IFN production through binding dsRNA produced by the virus and thus preventing its detection by PRRs [65–67] . NS1 and C6 have therefore both evolved to inhibit both type I IFN production and type I IFN-induced signalling but by distinct mechanisms in each pathway . To our knowledge VACV protein C6 is the first viral protein shown to both associate with the STAT2 transactivation domain and inhibit IFNα-dependent ISG induction after ISGF3 binding to the ISRE . The sequence of events that occur following ISGF3 binding to the ISRE is poorly understood and further elucidating the mechanism by which C6 inhibits this signalling pathway in the nucleus may enhance our knowledge of the late stages of the type I IFN signalling pathway . Lastly , it is notable that although VACV is a cytoplasmic DNA virus , it has evolved mechanisms to inhibit IFN production or activity within both the cytoplasm and the nucleus and indeed outside the infected cell by the expression of soluble type I and type II IFN binding proteins .
HEK293T ( ATCC CRL-11268 ) cells were maintained in Dulbecco’s Modified Eagle’s Medium ( DMEM , Invitrogen ) with 10% heat treated ( 56°C , 1 h ) foetal bovine serum ( FBS , Seralab ) and penicillin/streptomycin ( P/S , 50 μg/ml , PAA laboratories ) . HeLa ( ATCC CCL-2 ) cells were maintained in Minimum Essential Medium ( MEM , Invitrogen ) supplemented with 10% FBS , P/S and 1:100 non-essential amino acids ( Gibco ) . HeLa cells stably expressing GFP only or in combination with V5-tagged C6 ( V5-C6 ) , V5-Parainfluenza virus 5 V protein ( V5-PiV5-V ) or V5-tagged B14 ( V5-B14 ) , were obtained after transduction of cells with lentiviruses ( see below ) and sorting to obtain GFP-positive cells in a MoFlo MLS high-speed cell sorter ( Beckman Coulter ) . For each protein two populations were sorted based on GFP expression level; the top 30% of GFP-expressing cells ( high expressers ) and the next 30% ( middle expressers ) . Immunoblot analysis was used to determine the expression levels of the protein of interest in these two populations . Based on the observed expression of these proteins , the high GFP-expressing populations were chosen for , EV ( GFP only ) , V5-C6 , and V5-PiV5-V and the middle GFP-expressing V5-B14 population . For protein expression level of cell lines see S1C Fig . Lentivirus particles for transduction were generated after transient co-transfection of HEK293T cells with entry and packaging vectors and the bicistronic genomic vector encoding GFP and the appropriate V5 tagged protein using PEI ( CellnTec ) . The recombinant VACV Western Reserve strain C6-TAP and N1-TAP viruses were described [68] . The tandem-affinity purification ( TAP ) tag used here and in plasmids described below contains 2 copies of the streptavidin-binding sequence and 1 copy of the FLAG epitope [69] . Wild type ( wt ) VACV strain WR and the deletion mutant lacking the C6L gene were described [26] . Antibodies used were from the following sources; Rabbit ( Rb ) anti-FLAG ( Sigma-Aldrich , F7425 , diluted 1:5000 ) , Mouse ( Ms ) anti-V5 ( AbD Serotec Ltd , MCA1360 , diluted 1:5000 ) , Rb anti-HA ( Sigma Aldrich , H6908 , diluted 1:1000 ) , Ms anti-α-tubulin ( Millipore , 05–829 , diluted 1:5000 ) , Rb anti-actin ( Sigma , A2066 , diluted 1:1000 ) , Ms anti-Phospho-STAT1 ( Invitrogen , 333400 , diluted 1:750 ) , Rb anti-phospho-STAT2 ( Millipore , 07–224 , diluted 1:1000 ) , Rb anti-STAT1 for immunofluorescence ( Millipore , 06–501 , diluted 1:300 ) , Rb anti-STAT1 for western blotting ( Cell signalling , 9172S , diluted 1:1000 ) , Rb anti-STAT1 for immunoprecipitation ( Santa Cruz , sc-345 , diluted 1:100 ) , Rb anti-STAT2 ( Santa Cruz , sc-476 , diluted 1:100 for immunofluorescence and 1:500 for western blotting ) , Ms anti-Ku70 ( Abcam , ab3114 , diluted 1:1000 ) , Ms anti-IFIT1 for flow cytometry ( Abcam , ab70023 , 1:500 ) , and Alexa Fluor 546 goat anti-Rb IgG ( H+L ) ( Invitrogen , A-11010 , diluted 1:750 for immunoblotting ) or Alexa Fluor 647 donkey anti-Ms IgG ( H+L ) ( Invivogen , A- 31571 , diluted 1:2000 for flow cytometry ) . Reagents used in this study were BX795 ( Tocris ) , poly ( I:C ) ( InvivoGen ) , Protein G Sepharose 4 Fast Flow ( GE Healthcare ) , High Capacity Streptavidin Agarose Resin ( Thermo Scientific ) , human IFNα and human IL-1β were from Peprotech , poly ( dI:dC ) and ANTI-FLAG M2 Affinity Gel were from Sigma Aldrich . Biotinylated DNA for immunoprecipitations were synthesised by Integrated DNA technologies . The sequences were ISREcore; TGCCTCGGGAAACCGAAACTGAAGCCA and ISRErandom ACTGATCGGAAACCGAAACGATCTATG . These sequences were taken from [36] . Codon-optimised TAP-C6 and N1-TAP , were described previously [68] and [30] . B14-TAP was kindly provided by Dr . Brian Ferguson ( Department of Pathology , University of Cambridge , UK ) . The sequence of PiV5-V and NiV-V were amplified by PCR from plasmids kindly provided by Prof . Richard Randall ( University of St Andrews , UK ) and then subcloned into mammalian expression vectors pcDNA3 . 1 ( Invitrogen ) with an N-terminal V5 tag . V5-C6 was produced by PCR amplification of C6 from VACV WR DNA and cloned into pcDNA3 . 1 . GFP-V5 was provided by Dr . Christian Ku ( Department of Pathology , University of Cambridge ) . The sequences of IRF-9 , STAT1 and STAT2 were amplified by PCR from HeLa cDNA and subcloned into mammalian expression vector pcDNA4/TO with a C-terminal TAP tag and/or vector pcDNA3 . 1 with a N-terminal HA tag . pcDNA3 IRF9-STAT2C was a gift from Prof . Curt Horvath ( Addgene plasmid 37544 ) [35] . pcDNA4/TO ( Invitrogen ) was used in luciferase reporter assays as EV . ISRE-luciferase , NF-κB-Luciferase , and TK renilla were obtained from Dr . Andrew Bowie ( Trinity College , Dublin , Ireland ) , and ISG56 . 1-Luciferase was from Ganeth Sen ( Lerner Research Institute , Ohio , USA ) . Reporter gene assays were performed in HeLa or HEK293T cells seeded in 96-well plates . Cells were transfected with 100 ng firefly luciferase reporter plasmid , 10 ng GL3-renilla luciferase plasmid and 100 ng of expression plasmid for the protein of interest . For the BX795 reporter gene assay only the firefly report plasmid and GL3-Renilla plasmids were transfected . For the IRF9-S2C reporter gene assays 100 ng firefly reporter plasmid and 10 ng GL3-Renilla plasmid were transfected along with 50 ng IRF9-S2C and 50 ng C6 expression vector or control plasmids , except in the empty vector only control where 100 ng pcDNA4 was transfected only . Transit-LT1 ( Mirus , 2 μl per 1 μg DNA ) was used for transfection of HeLa cells and PEI ( CellnTec , 2 μl per 1 μg DNA ) for HEK-293T cells . Sixteen hours post transfection cells were stimulated as indicated in the figure legends . Cells were harvested in passive lysis buffer ( Promega , 100 μl/well ) . The firefly-luciferase readings of each sample were normalised to the renilla-luciferase readings and fold inductions were calculated relative to the non-stimulated controls for each plasmid . Experiments were performed in triplicate and conducted at least 3 times . HeLa cell lines stably expressing the proteins of interest were grown in 12-well plates and RNA was extracted using the RNeasy kit ( QIAGEN ) . One μg of each RNA sample was used to synthesise cDNA using Superscript III reverse transcriptase according to the manufacturer’s protocol ( Invitrogen ) . ISG mRNA was quantified by real-time PCR using a ViiA 7 Real-Time PCR System ( Life Technologies ) , fast SYBR Green Master Mix ( Applied Biosystems ) and the following primers , IFIT1 ( Fwd: CCTGAAAGGCCAGAATGA GG , Rev: TCCACCTTGTCCAGGTAAGT ) IFIT3 ( Fwd: ACACAGAGGGCAGTCATGAGTG , Rev: TGAATAAGTTCCAGGTGAAATGGC ) MxA ( Fwd: ATCCTGGGATTT TGGGGCTT , Rev: CCGCTTGTCGCTGGTGTCG ) GAPDH Fwd: ACCCAGAAGACTGTGGATGG , Rev: TTCTAGACGGCAGGTCAGGT ) . Amplification of ISGs was normalised to glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) amplification from the same sample , and the fold induction of genes in response to IFNα was calculated relative to the unstimulated control of the cell line . Experiments were performed in biological triplicate and conducted three times . HeLa cells stably expressing the proteins of interest were grown in 6-cm dishes and stimulated with 500 U/ml IFNα or mock-stimulated for 8 h . Cells were removed from the dishes by addition of trypsin ( GIBCO ) , transferred to Eppendorf tubes and washed twice with ice-cold PBS . Cells were fixed in 4% paraformaldehyde and permeabilised with 0 . 1% Triton X . Cells were then incubated with anti-IFIT1 ( Abcam , ab70023 ) in Triton buffer ( 0 . 5% BSA , 0 . 02% sodium azide , 0 . 1% Triton X-100 in PBS ) for 1 h at 4°C . Cells were washed twice with Triton buffer and then incubated with Alexa 647 Donkey anti-Mouse ( Invivogen ) in Triton buffer for 1 h in the dark at room temperature . Cells were washed twice in Triton buffer , once in 0 . 5% BSA in PBS and then analysed on a CyAn ADP Analyser ( Beckman coulter ) . Collected data were analysed using Summit ( Beckman Coulter ) . HeLa cells stably expressing the proteins of interest were grown on glass coverslips in 6-well plates . Cells were stimulated with 1000 U/ml IFNα for 1 h . Cells were fixed with 4% paraformaldehyde . Auto-fluorescence was quenched in 150 mM ammonium chloride in PBS and the cells were then permeabilised in 0 . 1% Triton X-100 in PBS . Cells were incubated in blocking buffer ( 0 . 5% BSA in PBS ) for 30 min , stained with primary antibody for 1 h ( STAT1: 1:300 , STAT2 1:100 in blocking buffer ) and for 1 h with secondary antibody ( Alexa 546 , Invitrogen ) . Coverslips were mounted onto microscope slides in Mowiol 4–88 containing 4' , 6-diamidino-2-phenylindole ( DAPI ) . Slides were visualised and imaged using a Zeiss LSM 780 Confocal microscope . Images were viewed using LSM Image Browser ( Zeiss ) . HEK293T cells were grown in 10-cm dishes and transfected with the constructs outlined in the figure legends using either Transit LT1 ( Mirus ) or calcium phosphate transfection . Sixteen hours later cells were stimulated with IFNα or mock-treated as described in figure legends , then lysed in lysis buffer ( 150 mM NaCl , 20 mM Tris-HCl pH 7 . 4 , 10 mM CaCl2 , 0 . 1% ( v/v ) Triton-X , 10% ( v/v ) glycerol and protease ( cOmplete Mini , Roche ) and phosphatase inhibitors ( PhosSTOP , Roche ) ) and cleared by centrifugation . Samples were then incubated with 30 μl Protein G Sepharose 4 Fast Flow ( GE Healthcare ) and anti-STAT1 ( Santa Cruz , sc-345 ) for 6 h , or ANTI-FLAG M2 Affinity Gel ( Sigma Aldrich ) or Anti-HA Agarose ( Sigma Aldrich ) and 2 h . Immunoprecipitations were washed 3 times in lysis buffer and bound proteins were eluted by boiling in buffer containing SDS . Samples were then analysed by SDS-PAGE ( polyacrylamide gel electrophoresis ) and immunoblotting with the stated antibodies . HEK293T cells were grown in 10-cm dishes and co-transfected with expression plasmids for IRF9-S2C , and V5-NiV-V , V5-C6 or V5-GFP using calcium phosphate transfection in triplicate for each condition . Sixteen hours later cells were lysed in lysis buffer ( 25 mM Tris-HCl pH 7 . 4 , 150 mM NaCl , 1 mM EDTA , 1% NP40 , 5% glycerol and protease ( cOmplete Mini , Roche ) and phosphatase ( PhosSTOP , Roche ) inhibitors ) . Lysates were incubated firstly with 10 ng/ml poly ( dI:dC ) for 30 min , then with 100 pmol biotin-labelled ISREcore or biotin-labelled control DNA for 1 . 5 h and finally with 30 μl High Capacity Streptavidin Agarose Resin ( Thermo Scientific ) for 3 . 5 h . Immunoprecipitations were washed four times with lysis buffer and proteins were eluted by boiling in buffer containing SDS . Samples were then analysed by SDS-PAGE and immunoblotting with the stated antibodies . Un-paired student’s T-tests were used to analyse data , with Welch’s correction applied when variances differed significantly between samples . Statistical significance is expressed as: *P<0 . 05 , **P<0 . 01 , ***P<0 . 001 and ****P<0 . 0001 . | In response to a viral infection , infected host cells mount an early , innate immune response to limit viral replication and spread . Type I interferons ( IFNs ) are produced by a cell when a viral infection is detected and are a crucial aspect of this early immune response . IFNs are released from the infected cell and can act on the infected cell itself or neighbouring cells to initiate a signalling pathway that results in the production of hundreds of anti-viral proteins . In this work we investigated a vaccinia virus protein called C6 , a known inhibitor of type I IFN production . Here we show that C6 also inhibits signalling initiated in response to type I IFNs , therefore providing a dual defence against this essential immune response . The results show that , unlike the majority of viral inhibitors of IFN signalling , C6 inhibits the signalling pathway at a late stage once the proteins required for IFN-stimulated gene transcription have reached the nucleus and bound to the DNA . This work illustrates the complex relationship between infecting viruses and the host immune response and further investigation of the mechanism by which C6 inhibits this important immune pathway will likely increase our knowledge of the pathway itself . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"phosphorylation",
"luciferase",
"molecular",
"probe",
"techniques",
"enzymes",
"enzymology",
"immunoblotting",
"plasmid",
"construction",
"signal",
"inhibition",
"immunoprecipitation",
"dna",
"construction",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"proteins",
"oxidoreductases",
"stat",
"signaling",
"molecular",
"biology",
"precipitation",
"techniques",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"post-translational",
"modification",
"interferons",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling"
] | 2016 | Vaccinia Virus Protein C6 Inhibits Type I IFN Signalling in the Nucleus and Binds to the Transactivation Domain of STAT2 |
Soil-transmitted helminth ( STH ) infections are endemic in Honduras and efforts are underway to decrease their transmission . However , current evidence is lacking in regards to their prevalence , intensity and their impact on children's health . To evaluate the prevalence and intensity of STH infections and their association with nutritional status in a sample of Honduran children . A cross-sectional study was done among school-age children residing in rural communities in Honduras , in 2011 . Demographic data was obtained , hemoglobin and protein concentrations were determined in blood samples and STH infections investigated in single-stool samples by Kato-Katz . Anthropometric measurements were taken to calculate height-for-age ( HAZ ) , BMI-for-age ( BAZ ) and weight-for-age ( WAZ ) to determine stunting , thinness and underweight , respectively . Among 320 children studied ( 48% girls , aged 7–14 years , mean 9 . 76±1 . 4 ) an overall STH prevalence of 72 . 5% was found . Children >10 years of age were generally more infected than 7–10 year-olds ( p = 0 . 015 ) . Prevalence was 30% , 67% and 16% for Ascaris , Trichuris and hookworms , respectively . Moderate-to-heavy infections as well as polyparasitism were common among the infected children ( 36% and 44% , respectively ) . Polyparasitism was four times more likely to occur in children attending schools with absent or annual deworming schedules than in pupils attending schools deworming twice a year ( p<0 . 001 ) . Stunting was observed in 5 . 6% of children and it was associated with increasing age . Also , 2 . 2% of studied children were thin , 1 . 3% underweight and 2 . 2% had anemia . Moderate-to-heavy infections and polyparasitism were significantly associated with decreased values in WAZ and marginally associated with decreased values in HAZ . STH infections remain a public health concern in Honduras and despite current efforts were highly prevalent in the studied community . The role of multiparasite STH infections in undermining children's nutritional status warrants more research .
Honduras is among 30 countries in the Americas that are endemic for soil-transmitted helminth ( STH ) infections , which are caused by four species of intestinal nematodes: the common roundworm , Ascaris lumbricoides; the whipworm , Trichuris trichiura; and the hookworms , Necator americanus and Ancylostoma duodenale [1] . The health impact of these infections is more dramatic in children , for whom STH show a particular predilection [2] partly due to their differential exposure to contaminated soil . Health adverse effects such as anemia , growth stunting , protein-calorie malnutrition , fatigue , and poor cognitive development tend to occur and persist in populations affected by STH [3] , and all too often , helminth infections are seen as normal and unavoidable part of life in endemic populations [4] . According to the World Health Organization ( WHO ) , two thirds of Honduran children aged 1–14 years require preventive chemotherapy ( PC ) for STH [1] . In fact , the Preventive Chemotherapy and Transmission Control ( PCT ) databank of the WHO estimates that 2 . 6 million Honduran children ( 769 , 405 pre-school and 1 , 832 , 476 school-age children ) are at risk for STH transmission therefore requiring regular administration of anthelminthic drugs [5] . Organized STH control activities in the country began in 1998 with the establishment of the Healthy Schools Program , a collaborative effort between the ministry of health , the ministry of education and the World Food Program [6] . By 2001 , Honduras had started subnational control activities [1] , [7] and these soon evolved into a national program guided by the recommendations outlined in the World Health Assembly resolution 54 . 19 [6] , [8] . Although the goal of providing anthelminthic medication in a regular manner to at least 75% of all school-age children at risk has yet to be attained [1] , [5] , [6] , [7] , Honduras' efforts of tackling STH transmission are , nevertheless , commendable . An important complement to these efforts would be to undertake a complete situation analysis at subnational levels; one that would establish baseline data in terms of prevalence as well as transmission risk factors so priority areas can be identified and intervention efforts tailored to specific populations [9] . Additionally , to effectively monitor the success of PC and other interventions , studies assessing STH-associated morbidity in infected children are needed in Honduras . Based on scientific evidence linking STH morbidity with worm burden ( i . e . , the number of adult parasites inhabiting the intestine [10] ) , the elimination of moderate and heavy infections is the target of PC programs [1] . In addition to worm burden , polyparasitism -the concurrent infection with multiple parasite species- has also been associated with children's malnutrition [11] , [12] . As well , even when recent data is scarce , some studies have reported that even light infections may impose a threat to children's health [13] , especially if living in endemic communities with poor nutritional status [10] , [11] , [14] . Hence , the overlap of poverty , malnutrition and STH endemicity in some populations may obscure the true effect of these helminthiases in childhood health and accordingly , more research is needed to fully appreciate the burden of these infections on people living in these areas [15] . With this in mind , the aim of this study was to investigate the prevalence and intensity of STH infections in a sample of Honduran school-aged children and examine whether STH are negatively associated with malnutrition and anemia .
The present study was nested within a parent study entitled ‘Gender and parasitic diseases: Integrating gender analysis in epidemiological research on parasitic diseases to optimize the impact of prevention and control measures’ ( principal investigator , T . W . Gyorkos , McGill University , Canada ) and both received ethics clearance from McGill University Health Centre , Montreal , QC ( file number MUHC 10 -175 – PED Nov . 23rd 2010 ) and Brock University , St . Catharines , ON ( file number - BU 10–161 – Sanchez/Gyorkos Jan 13th 2011 ) . In the absence of an institutional ethics board in the participating academic unit of the Honduran university , the Ethics Officer of the Masters Program in Infectious and Zoonotic Diseases ( MEIZ ) of the School of Microbiology , National Autonomous University of Honduras , reviewed the protocol and provided clearance ( file number OF-MEIZ- 001-2011 ) . As the study population comprised minors , both parental consent and children's assent were required prior to enrolment of children . Parents and guardians of children in grades 3–5 were invited to an information session in which the study's objectives , benefits and risks were fully explained . Parents and guardians who gave oral consent were presented with an information package containing a detailed lay description of the study , an invitation to participate and a consent form for their signature . All parents or guardians consenting for their children to participate signed the informed consent form . Children whose parents consented were invited to participate in the study during sessions held at the schools and those who expressed assent in responding to a questionnaire , providing a stool and blood sample and allowing the collection of anthropometric measurements were then enrolled in the study . Children assents were obtained verbally and documented through a child assent form . Also , since the study was undertaken during class time at participating schools , authorizations from schools' Principals were sought in advance and only schools with such authorizations were approached for enrolment . Laboratory reports were issued with accompanying lay interpretations and recommendations . Also , parents of children with STH infections were offered anti-parasitic treatment for their child . If agreed , albendazole tablets ( 400 mg ) were administered to the child by the school teacher or parent . A “deworming tracking card” was issued for each child . Parents and teachers were encouraged to keep track of the children's deworming treatment in order to either avoid missing the school's annual or bi-annual treatment or prevent excessive treatments in case deworming was offered by third parties ( e . g . , international or national medical brigades , faith-based missions , etc . ) . Both the parent study and present study were school-based , cross-sectional studies , designed as explorative and hypothesis generating studies . For the parent study , power and sample size determination were performed utilizing the PS software ( version 3 . 0 , January 2009 , by William D . DuPont and Walton D . Plummer Jr . ) . This was based on a two-sided chi-square test to compare STH infection between boys and girls . Using previous studies in Peru as a reference [16] it was assumed that half of the children in this school-age group will be male and that the prevalence of any STH would be 50% in males ( a conservative estimate ) . An estimated design effect of 2 . 7 was used with a significance level of 0 . 05 . A total of 314 participants were therefore needed to detect a minimum risk ratio of 1 . 5 with 80% power . The present study was bound by this sample size determination . This study was implemented during February and March 2011 with the collaboration of the National University of Agriculture ( UNA ) located in the city Catacamas , in the municipality of the same name ( 14°51′35 . 46″; N 85°53′58 . 19″W ) in the Department of Olancho , about 210 km north-east of the capital of Honduras , Tegucigalpa . Geographically , Catacamas is the largest municipality in the country and is nested in a fertile valley at approximately 450 m above sea level . Catacamas municipality consists of the urban core ( Catacamas proper ) and 14 main villages which in turn are comprised of smaller 339 hamlets . Catacamas human development index ( HDI ) value for 2009 was 0 . 675 [17] , slightly over to that of Honduras ( 0 . 625 for 2011 ) [18] . However , 60% of Catacamas' population resides in rural areas , the majority lacking public services such as electricity , potable water and indoor plumbing . As means of livelihood inhabitants engage in mixed agricultural farming , rearing animals such as cattle , pigs and poultry and growing crops such as corn , beans , coffee and vegetables . Others work as traders or labourers while a few work in public or private service [19] . The following nine surrounding rural communities ( most between 2–3 hour driving distance from the city ) were visited as potential study-sites: Colonia de Poncaya , Las Lomas de Poncaya , Las Parcelas , Corosito de Poncaya , El Cerro del Vigía , El Hormiguero , Santa Clara , Los Lirios and Campamento Viejo . The combined eligible school population was 445 children . Schools located in those communities were identified and principals contacted by UNA's personnel to inform them about the study and obtain authorization to approach the school and potentially enrol their pupils . As well , information was obtained in regards to school enrolment and status of their deworming program , if any . Schools which had provided deworming treatment within the last three months were not eligible for the study . The target participants for the study were children in grades 3–5 ( aged 9–11 ) since STH infections , especially A . lumbricoides and T . trichiura tend to peak at this age [1] . Also , at this age children are old enough to understand survey questions and provide basic information . Using a pre-tested , 30-minute , face-to-face standardized questionnaire in Spanish , the Gender Study collected demographic and epidemiological data as well as children's living conditions and knowledge regarding STH infections . From these data , the present study extracted children's general demographics ( name , date of birth , age , and sex of the child ) , STH and deworming history , self-reported living conditions ( household's type of floor , water access and type and use of sanitary facilities ) , and the possession of major home appliances . Body weight and height measurements were taken for each child to calculate anthropometric indicators . Weights were taken using a digital electronic balance to the nearest 0 . 1 kg while children were wearing school uniforms and without shoes . Height was taken to the nearest 0 . 1 cm using a height pole mounted on the wall . In order to minimize intra-individual errors , all measurements were taken twice by different researchers and the average value calculated and used thereof . Age , height and weight were then used to calculate the following indicators: a ) height-for-age Z-score ( HAZ ) to assess stunting; b ) weight-for-age Z-score ( WAZ ) to assess underweight; and c ) body-mass-index-for-age Z-score ( BAZ ) to assess thinness . Calculations were done with the WHO AnthroPlus software version 1 . 04 ( WHO , Geneva , Switzerland ) using the WHO international reference values ( available at http://www . who . int/growthref/tools/en/ ) . Because of its inability to differentiate between relative height and body mass , WAZ is not recommended for the assessment of growth beyond childhood ( >10 years of age ) [19] . Therefore , BAZ was used as a complement to HAZ . These indicators are recommended by the WHO as they provide an assessment of the child's nutritional status in comparison with a healthy reference population [20] , [21] . According to the 2007 WHO growth reference for school-aged children and adolescents , stunting , underweight and thinness are defined as <−2 standard deviations ( SD ) HAZ , WAZ and BAZ , respectively [20] . A single fecal sample was collected from each child and samples were taken to UNA's laboratory for analysis on the same day using the Kato-Katz technique [22] with a template of 41 . 7 mg , as recommended by the WHO [23] . Kato-Katz templates were obtained from Vestergaard-Frandsen Disease Control Textiles ( Vestergaard Frandsen SA , Aarhus , Denmark ) . Kato–Katz slides were examined microscopically in a systematic manner within 30–60 min of preparation; helminth eggs counted for each parasite species and the number thus obtained multiplied by a factor of 24 in order to get the number of eggs per gram of feces ( epg ) . Egg counts were utilized to classify infection intensities into light , moderate , or heavy infections as follows , respectively: for A . lumbricoides , 1–4 , 999 epg , 5 , 000 – 49 , 999 epg and ≥50 , 000 epg; for T . trichiura , 1–999 epg , 1 , 000–9 , 999 epg and ≥10 , 000 epg; and for hookworms , 1–1 , 999 epg , 2 , 000–3 , 999 epg and ≥4 , 000 epg [1] , [24] . Haematological analyses were done using the BC – 3000Plus AutoHematology Analyzer ( Mindray Medical Instrumentation , USA ) in a private medical laboratory contracted in Catacamas . Anemia was determined when children aged 6–11 years had hemoglobin ( Hb ) values <11 . 5 g/dL or hematocrit ( Hct ) <34% . For children aged 12–14 years , these values were Hb <12 g/dL or Hct <36% [25] . Total serum protein concentrations were measured by the Biuret method and children were considered within the reference values if concentrations were within 6–8 . 5 g/dL [26] . Data were entered by a researcher into Microsoft office Excel spreadsheet 2007 ( Microsoft ) and verified for accuracy ( compared with data in questionnaires ) by a different researcher . Data were cleaned by checking for errors and missing values . Statistical analyses were done using IBM , SPSS Statistics ver . 20 ( IBM . Somers , NY ) . Descriptive statistics for continuous variables and frequency ( proportions ) for categorical variables were used to describe the characteristics of the study population . Weight and height measurements were subjected to a reliability test and the inter observer technical error of measurements was assessed using the Mueller and Martorell method [27] . Differences in proportions for categorical variables ( e . g . , age group , sex of the child , stunting , thinness , underweight and anemia ) were calculated using Chi square test of independence . Differences in mean values for continuous variables ( e . g . , HAZ , WAZ , BAZ , total proteins , Hb and Hct ) were assessed using the student t-test analysis . Since STH clinical importance is generally associated with increased worm burden , infections of moderate and heavy intensity were merged into one category “moderate-to-heavy” . This was also useful for computational reasons since those infections were in minority among studied children . Also , to assess polyparasitism , a category termed “infection status” was created to denote conditions of non-infected , monoparasitism or polyparasitism ( co-infections with 2 or 3 STH ) . One-way ANOVA was used to analyze differences in anthropometric mean Z-scores of the study population by infection status and by infection intensity ( negative , light and moderate-to-heavy ) of each parasite species . A generalized estimating equations ( GEE ) approach was used to construct both multivariable linear and logistic regression models to account for possible within-school data correlation ( clustering at the school level ) . For these models , intensity of infection was not analyzed by parasite species . Rather , infection categories “negative-to-light” and “moderate-to-heavy” were created to denote individuals with such infections by any of the three parasites under study . Linear regression models to test for associations between anthropometric indicators and intensity of infection categories were constructed adjusting for age , sex , socio-economic status ( SES ) and anemia . Similar models were done to test for association between those indicators and infection status . Using principal component analysis ( PCA ) , the SES variable was constructed from five factors: type of floor , access to tap water , having a toilet , having a television set and having a fridge . Separate logistic regression models were constructed to assess associations between stunting and thinness odds and STH intensity of infection and infection status adjusting for age stratum ( 7–10 or >10 years of age ) , sex and SES . Odds ratios ( OR ) were determined with 95% confidence intervals ( CI = 95% ) .
Of the nine visited , seven schools were enrolled in the study: Colonia de Poncaya , Las Lomas de Poncaya , Las Parcelas , Corosito de Poncaya , El Cerro del Vigía , El Hormiguero , and Campamento Viejo . The reasons for not including the two remaining were: recent deworming treatment ( Santa Clara n = 26 ) and time-constraints to complete questionnaires and measurements ( Los Lirios n = 19 ) . ( Los Lirios' children , however , were examined for STH and treatment provided if needed ) . Thus , the number of eligible participants in grades 3 to 5 among participating schools was 400 children . The parents of 368 ( 92% ) children provided written informed consent for their children to participate and almost all ( 357 , 97% ) children assented to be enrolled . After enrolment , 37 participants were dropped from the study due to insufficient or no stool sample ( n = 20 ) , or unreliable Kato-Katz results that could not be repeated ( n = 17 ) . Also , five children declined blood collection but they were kept in the study since their remaining data was complete . The final study sample was 320 children aged 7–14 years ( mean 9 . 76±1 . 4 ) and 154 ( 48% ) were girls . Demographic , household and nutritional characteristics of the study sample are shown in Table 1 . Additionally , habitual or occasional open defecation was reported by 15 . 6% and 12 . 8% of the children , respectively . As for STH history , 58 . 1% of the children reported having expelled ‘worms’ in the past and 85 . 9% recalled having received deworming treatment sometime in the past but not recently . Five of the seven schools enrolled in the study had ongoing deworming programs , some starting as far back as 2007 . Frequency of deworming was twice a year for two schools and once a year for three schools . The last deworming treatment had been within the last 4–6 months for four schools and two years for the remaining one . There was no statistical difference between overall infection with any STH and schools' deworming schedule ( p = 0 . 767 ) . Replicate weight and height measurements showed high reliability when tested for the inter-observer technical error of measurements . The reliability coefficient ( R ) was 0 . 962 for weight and 0 . 973 for height . Nutritional indicators of the study population are presented in Table 1 . The nutritional status of most children was within healthy parameters but a few cases of stunting ( n = 18 , 5 . 6% ) , thinness ( n = 7 , 2 . 2% ) and underweight ( n = 3 , 1 . 3% ) were observed . Of the children who were stunted , thin or underweight , girls accounted for 50% , 43% and 67% of the cases , respectively . No child had a total protein value below the normal range and of 315 children examined , 7 ( 57% girls ) were anemic . Overall , of 320 children , 33 ( 10 . 3% ) had at least a form of nutritional deficit . Five of these children ( 15 . 2% ) were negative for any STH , while 28 ( 84 . 8% ) were infected with one or more STH . Among the latter , 15 children were monoparasitized , while 13 were polyparasitized . Results of the one-way ANOVA analysis revealed that mean values for WAZ scores were significantly lower in children with moderate-to-heavy infections by either T . trichiura ( p = 0 . 020 ) or A . lumbricoides ( p = 0 . 015 ) compared to children with no or light infections . This was not observed in the case of hookworm infections , likely due to the fact that the vast majority ( 94% ) of such infections were light . On the other hand , the scores for the other two indicators ( HAZ and BAZ ) did not differ significantly across the various infection intensities of any of the helminth species . However , as depicted in Figure 1 , a negative trend –although not always significant- between infection intensity and the mean values of all anthropometric indicators was identified . In other words , the heavier the intensity , the lower the HAZ , BAZ and WAZ mean values ( Figure 1 , plots A , B and C ) . A similar trend was observed in terms of infection status: as polyparasitism increased , the mean values of all anthropometric indicators decreased ( Figure 1 , plot D ) . As data in Table 1 show , this trend was significant in terms of WAZ scores ( p = 0 . 012 ) , marginally significant for HAZ scores ( p = 0 . 071 ) but not significant for BAZ scores ( p = 0 . 202 ) . Estimated coefficients ( β ) from multivariable GEE linear models are shown in Table 3 . Compared to no or light infections , moderate-to-heavy infections with any STH were significantly correlated with a decrease in WAZ scores ( β = −0 . 34 , 95% CI = −0 . 62 to −0 . 06 , p = 0 . 018 ) . This correlation was only marginally significant for HAZ scores ( β = −0 . 20 , 95% CI = −0 . 44 to 0 . 04 , p = 0 . 108 ) but not significant for BAZ ( p = 0 . 622 ) . Polyparasitism was found inversely correlated with both WAZ and HAZ scores . For WAZ , this correlation was significant ( β = −0 . 37 , 95% CI = −0 . 66 to −0 . 08 , p = 0 . 012 ) whereas for HAZ , it was only marginally significant ( β = −0 . 24 , 95% CI = −0 . 50 to 0 . 02 , p = 0 . 074 ) . However , no evidence for association between polyparasitism and BAZ scores was found ( p = 0 . 446 ) . With respect to age , there was a strong negative correlation between age and HAZ and BAZ scores ( β = −0 . 16 , 95% CI = −0 . 24 to −0 . 09 , p<0 . 001 and β = −0 . 12 , 95% CI = −0 . 21 to −0 . 03 , p = 0 . 008 , respectively ) . Conversely , WAZ scores were not correlated with age ( p = 0 . 428 ) . Multivariable GEE logistic models revealed that age of the studied population was significantly associated with stunting . Children >10 years old were three times more likely to be stunted ( OR = 3 . 31; 95% CI = 1 . 23–8 . 90 , p = 0 . 018 ) than younger children . Age , on the other hand , was only marginally significantly associated with thinness ( p<0 . 15 ) . Neither infection intensity nor infection status ( polyparasitism ) was found associated with stunting or thinness . Finally , since only three children were underweight ( WAZ <−2SD ) no statistical model was produced for this nutritional indicator .
In a little more than a decade , moderate economic progress alongside dedicated efforts for STH control –mainly through national deworming campaigns- have contributed to the decrease of Honduras' national STH prevalence [1] . Yet , as the data from our study show , some rural communities have persistently high STH transmission and perhaps they face greater struggles in overcoming poverty and inequities [17] . Indeed , considering that five of the seven participating schools reported some form of mass-deworming during the past year , a prevalence of 72 . 5% for any STH among these children is remarkably high . According to the work of Hall and colleagues , a prevalence of 70% or greater carries a high probability of disease [12] . Moreover , 36% of infected children were harbouring moderate-to-heavy infections . The new vision for a world free of childhood morbidity due to these helminths , according to the WHO , is reducing the prevalence of STH infection of moderate and heavy intensity to ≤1% [1] . Therefore , these data alone underscore the need for Honduras to continue and sustain its deworming program , and more importantly , to implement and monitor integrated control efforts [28] , [29] . The predominance of T . trichiura over A . lumbricoides ( 66 . 9% versus 30 . 3% ) may indicate that the single-dose albendazole schedule currently used for deworming has been less effective for reducing trichuriasis as this parasite is less susceptible to this drug [30] , [31] . Even though it might not be feasible to implement a different PC regimen in Honduras at the moment , it is important to be vigilant of the different patterns of transmission of individual STH species . At the same time , it would be useful to conduct baseline studies to measure reinfection rates [32] and drug efficacy [33] , as well as to make efforts to detect potential emergence of resistance to benzimidazoles [34] , [35] , [36] . In terms of prevalence , our study shows that children older than 10 years of age were more likely to be infected with any STH than younger children , underscoring the importance of deworming children throughout their primary school years [1] . The fact that in our study boys were more likely than girls to harbour hookworm infection warrants further investigation as there might be gender-related factors playing an important role in exposure to hookworms , as previously suggested [37] , [38] . Along with high prevalence , we also found a high proportion of polyparasitism with almost half of those infected ( 44% ) harbouring 2 or 3 helminths . This finding is consistent with the epidemiological profile of endemic countries [16] , [39] , [40] , [41] and although already observed in Honduras [6] , [42] , [43] , it has not received sufficient attention in the country . The impact of infections by multiple parasite species has been subject to some attention in the last decade [44] and studies show that concurrent infections may have additive or synergistic detrimental effects , especially in childhood [11] , [45] . Given that regular PC interventions will eventually result in reduced infection intensity , light polyparasitic infections will become more relevant [45] . Therefore , addressing polyparasitism in future WHO recommendations merits consideration . In terms of nutritional status , the majority of studied children were within the WHO reference values for growth and nutrition . This is uncommon for a Honduran rural population [46] . In fact , the proportion of children suffering chronic undernutrition ( as measured by stunting ) identified in this study was 5 . 6% well below current national urban ( 13 . 7% ) or rural ( 32% ) figures . On the other hand , the proportion of global undernutrition ( as measured by thinness ) was 2 . 2% , below the national urban ( 6 . 2% ) or rural ( 14 . 8% ) averages [47] and very close to the expected value ( in a healthy population , approximately 2 . 1% of individuals will fall either above or below 2SD of the HAZ , BAZ and WAZ reference values ) . Although assessing food-security was beyond the scope of this study , a possible explanation for this finding is that Catacamas valley and surrounding areas are situated in fertile lands and food insecurity is not as dramatic as in other parts of the country [19] , [48] . We found that the risk of stunting increased with age , a phenomenon also found in similar studies conducted in Peru in both school-age children [16] and in pre-schoolers [49] , as well as in Malaysia [50] , Colombia [51] and Guatemala [39] . It appears , therefore , that once stunted , children continue to be so , even when more acute indicators such as WAZ and BAZ fall within healthy parameters . Longitudinal studies could help elucidate the most favourable moment for children to prevent growth faltering that may lead to stunting . As mentioned , we aimed to ascertain potential associations of STH infections with the children's nutritional status ( stunting , underweight and thinness ) but our data did not support such associations . It is recognized that studying the impact of intestinal helminths on child growth and nutrition in endemic populations is not an easy endeavour as it is difficult to control for other environmental or socio-economic factors or seasonal changes in the food supply [4] , [12] . It is worth mentioning , however , that SES status ( as we measured it ) was not found associated with either increased odds of stunting or thinness or with a decrease in anthropometric indicators Z-scores . Notwithstanding this lack of association in our study , when looking at the distribution of the actual Z-scores for these indicators in the multivariable analyses , we found that both moderate-to-heavy infections with any STH and polyparasitism were significantly associated with lower WAZ scores . Additionally , at the species level , negative trends ( albeit not all significant ) were observed between STH infection and WAZ , BAZ and HAZ mean scores; namely , as intensity of infection or the number of species parasitizing increased , Z-score mean values for the three measured anthropometric indicators decreased . A similar finding was reported by Ordoñez and Angulo ( 2002 ) in a cross-sectional study in which polyparasitized children had lower HAZ and WAZ scores [51] . Thus , examining anthropometric Z-scores values might be useful in providing additional insight into the impact of STH on children's nutritional status as they may reveal subtle changes missed when focusing only on end outcomes ( i . e . , stunting , thinness and underweight ) . By the same token , Z-score values may be able to pin-point improvements in children's nutrition after anthelminthic treatment even if significant changes in end outcomes cannot be demonstrated . Limitations of this study arise from its cross-sectional nature as direct causal relationship between STH and nutritional status cannot be determined . This is why large-scale prospective studies with rigorous design and the corresponding funding are necessary to investigate neglected tropical diseases including helminthiases . Another potential limitation stems from the fact that our investigation was part of a parent study with a sample size calculation based on being able to detect an important difference in sex-specific STH prevalences instead of nutritional indicators and this may limit the precision of our results . We trust that our findings will shed light into the design of future studies in Honduras . In terms of our parasitological findings , the analysis of a single stool sample may have underestimated STH prevalence in our study but by the high prevalence obtained , this underestimation might be minimal . Further , recent work suggests that Kato-Katz is reasonably accurate for A . lumbricoides and T . trichiura although less so for hookworms [52] . Likewise , intensity of infection may have been underestimated although recent publications suggest that Kato-Katz results are fairly reliable for the three STH investigated in the present study [52] , [53] . Malaria , other intestinal parasites or gastrointestinal infections were not determined and a role for these on children's nutritional status cannot be ruled out . Finally , an important limitation in identifying undernutrition factors is that this study did not entail an exhaustive investigation of underlying causes of malnutrition ( e . g . , social determinants , food security , dietary intake and expenditure , etc ) . Future research should address this gap although cross-sectional studies might not be able to reveal concrete answers , as shown by Gray et al . ( 2006 ) [46] . Strengths of this study are: including a design effect in the sample size estimation to take into account clustering by school , obtaining a high participation rate , and that our sample is likely representative of the communities' school children as in Honduras 95% children attend primary school [47] . Also , by utilizing laboratory protocols and anthropometric measurements recommended by the WHO , our results permit comparisons with other studies both nationally and internationally . In conclusion , the prevalence data obtained in this study contribute with accurate and updated information to map out the situation of STH infections in Honduras . Further , this study provides a unique insight into the nutritional status of a cohort of school children living in rural Honduras and , very importantly , it is to our knowledge , the first in the country to explore the potential impact of STH infections on children's nutritional status . Our results also underscore the need for more research into the health effects of polyparasitism in children . It is hoped that the findings presented here will be useful in informing current STH control efforts in Honduras and encourage future investigations that take into account the social and geographical differences across the country . | Soil-transmitted helminth ( STH ) infections are endemic in Honduras but their impact on children's health is not well studied . With the purpose of determining the prevalence and intensity of STH infections and their association with nutritional status in a sample of Honduran children , a cross-sectional study was undertaken in 2011 . School-age children were enrolled , and in addition to demographic data , blood and stool samples and anthropometric measurements were obtained to determine nutritional status and STH infection . The overall STH prevalence among 320 studied children was 72 . 5% and almost half of the infected children harboured multiple parasites . Polyparasitism was more likely to occur in children attending schools with absent or annual deworming schedules than in pupils attending schools deworming twice a year . Prevalence by species was 30% , 67% and 16% for Ascaris , Trichuris and hookworms , respectively . Infections of moderate to heavy intensity as well as multiparasite infections were significant predictors of decreased weight-for-age scores in children ages 7–10 years after controlling for key confounders . Sustainable efforts to control STH infections in Honduras are required . Future research providing more insight on the nutritional impact of polyparasitic STH infections in childhood is necessary . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"epidemiology",
"biology",
"microbiology",
"parasitology"
] | 2013 | Soil-Transmitted Helminth Infections and Nutritional Status in School-age Children from Rural Communities in Honduras |
Cryptococcus neoformans is a leading cause of invasive fungal infections among immunocompromised patients . However , the cellular constituents of the innate immune response that promote clearance versus progression of infection upon respiratory acquisition of C . neoformans remain poorly defined . In this study , we found that during acute C . neoformans infection , CCR2+ Ly6Chi inflammatory monocytes ( IM ) rapidly infiltrate the lungs and mediate fungal trafficking to lung-draining lymph nodes . Interestingly , this influx of IM is detrimental to the host , since ablating IM or impairing their recruitment to the lungs improves murine survival and reduces fungal proliferation and dissemination . Using a novel conditional gene deletion strategy , we determined that MHC class II expression by IM did not mediate their deleterious impact on the host . Furthermore , although ablation of IM reduced the number of lymphocytes , innate lymphoid cells , and eosinophils in the lungs , the effects of IM were not dependent on these cells . We ascertained that IM in the lungs upregulated transcripts associated with alternatively activated ( M2 ) macrophages in response to C . neoformans , consistent with the model that IM assume a cellular phenotype that is permissive for fungal growth . We also determined that conditional knockout of the prototypical M2 marker arginase 1 in IM and deletion of the M2-associated transcription factor STAT6 were not sufficient to reverse the harmful effects of IM . Overall , our findings indicate that C . neoformans can subvert the fungicidal potential of IM to enable the progression of infection through a mechanism that is not dependent on lymphocyte priming , eosinophil recruitment , or downstream M2 macrophage polarization pathways . These results give us new insight into the plasticity of IM function during fungal infections and the level of control that C . neoformans can exert on host immune responses .
The ubiquitous encapsulated yeast Cryptococcus neoformans causes invasive fungal infections in immunocompromised patients , particularly those with AIDS , solid organ transplants , and cancer [1] . Even with optimal combination antifungal therapy , cryptococcosis has a high morbidity and mortality rate [2 , 3] . Since C . neoformans enters the respiratory tract before disseminating to the central nervous system [1] , defining the cellular and molecular mechanisms of the pulmonary innate immune response is critical for the development of novel treatment options that can promote fungal sterilization in the lungs . C-C chemokine receptor 2 ( CCR2 ) - and Ly6Chi-expressing inflammatory monocytes ( IM ) and their derivatives , including macrophages and dendritic cells ( DCs ) , exhibit beneficial roles in innate host defense against many fungal pathogens , including Aspergillus fumigatus [4 , 5] , Blastomyces dermatitidis [6] , Candida albicans [7] , and Histoplasma capsulatum [6 , 8] . For example , during pulmonary aspergillosis , IM directly engage and kill fungal cells , regulate innate immune activation of neutrophils , and facilitate adaptive CD4+ T cell responses [4 , 5] . The role of IM in innate immunity to acute infection with C . neoformans has not been systematically examined . In models of subacute and chronic pulmonary cryptococcosis , there is an initial fungal expansion phase that is followed by prolonged , but largely progressive , fungal clearance through CD4+ and CD8+ T cell-dependent mechanisms [9 , 10] . During the expansion phase , CCR2-dependent signals mediate the accumulation of DCs and macrophages in the lungs [11–13] . The latter express inducible nitric oxide synthase ( NOS2 ) and tumor necrosis factor ( TNF ) and appear to act in a fungicidal manner in vitro [13] . Defective CCR2 signaling impairs fungal clearance and correlates with the development of T helper 2 ( Th2 ) cytokine-dominated responses [14 , 15] . More recent work indicates that IM and macrophages may play important roles in protective immune responses generated by candidate vaccine strains of C . neoformans [16–19] . These data suggest that IM and their derivatives may play beneficial roles in innate immunity during subacute and chronic cryptococcal infections . On the other hand , in a lethal model of acute cryptococcosis , it has been observed that enhanced IM accumulation in the lungs correlates with decreased survival [20] . It is not known if IM may promote progressive infection by specific effects on fungal growth or T helper responses or if IM influx is part of a general inflammatory response in the lungs during acute cryptococcal infection . C . neoformans is also a facultative intracellular pathogen that has been shown in vitro to replicate within monocytes and macrophages and exit via non-lytic exocytosis [21–24] . Thus , it has been proposed that these cells can function as “Trojan horses” that facilitate fungal proliferation and dissemination [25–27] . Together , these data support an alternative model in which IM could be detrimental in the host response to acute C . neoformans infection . In this study , we sought to clarify the role of IM and their derivatives in a murine model of acute cryptococcosis using a highly virulent serotype A strain of C . neoformans . We utilized CCR2-DTR depleter mice [5] and a new constitutive CCR2-Cre mouse model to probe the functional role of IM in C . neoformans control and host survival . Interestingly , we found that in the absence of IM , murine survival is improved and there is decreased fungal burden in the lungs and disseminated sites , indicating that IM are harmful for host anti-cryptococcal immunity . We did not find experimental evidence that immunopathology or cellular crosstalk between IM and lymphocytes or eosinophils influence these infectious outcomes . However , we observed that IM in the lungs exhibit an alternatively activated ( M2 ) -like macrophage transcriptional profile in response to C . neoformans . In particular , IM significantly upregulated expression of the gene encoding arginase 1 ( ARG1 ) , an M2 marker that may modulate immune responses due to its competition for L-arginine substrate with NOS2 , a marker for classically activated ( M1 ) macrophages [28–30] . M2 macrophages have previously been demonstrated to have decreased anti-cryptococcal activity against C . neoformans in vitro compared to classically activated ( M1 ) macrophages [31] . Interestingly , the conditional knockout of Arg1 in IM and the deletion of STAT6 , a transcriptional regulator of Arg1 [32 , 33] , in hematopoietic cells could not reverse the impact of IM on host outcomes during acute cryptococcosis . In summary , our study defines a novel cell-intrinsic role for IM as mediators of detrimental host immune responses to a respiratory fungal pathogen and indicates that the subversion of these potential antifungal effector cells by C . neoformans occurs early in the IM response .
Our studies utilized an acute infection model in which C57BL/6 mice were administered 103 yeast cells of C . neoformans serotype A strain H99 intratracheally ( i . t . ) . This infection model was uniformly fatal with a range of inocula from 102 to 105 yeast cells ( S1A Fig ) . We found that IM accumulated in the lungs of mice within the first week after respiratory challenge and persisted during the course of progressive infection ( Fig 1A ) . There was also an increase in pulmonary CD11b+ DCs and macrophages during infection ( S1B Fig ) , consistent with the known developmental relationships between lung-infiltrating IM and these immune cell subsets during fungal infections [5 , 12 , 13] . To determine the role of IM at the onset of infection , these cells were transiently ablated in CCR2-DTR mice [5] by administering intraperitoneal ( i . p . ) diphtheria toxin ( DT ) on days -1 , +1 and +3 relative to infection with H99 ( Fig 1B ) . DT treatment of CCR2-DTR mice also resulted in significant decreases in DCs and macrophages in the lungs on day 7 post-infection ( p . i . ) ( S2 Fig ) . DCs in CCR2-DTR mice returned in numbers comparable to non-transgenic littermate controls ( WT ) by day 14 p . i . , while macrophages exhibited a slower return toward WT levels ( S2 Fig ) . Compared to WT mice , CCR2-DTR mice had significantly prolonged survival ( median survival 35 . 5 days for CCR2-DTR versus 27 . 5 days for WT ) ( Fig 1C ) and decreased fungal burden in the lungs on days 7 , 14 , 21 and 26 p . i . ( Fig 1D ) . CCR2-DTR mice had a similar ~1 log reduction in lung fungal burden compared to WT littermates when challenged with a 10-fold higher inoculum of 104 H99 yeast cells ( S3A Fig ) or with 104 yeast cells of a less virulent C . neoformans serotype D strain 52D ( S3B Fig ) . The lungs of WT mice were grossly enlarged compared to those of CCR2-DTR mice ( S4A Fig ) , which may be due to the observed differences in fungal burden as well as a trend toward increased pulmonary infiltrates in WT mice ( S4B–S4D Fig ) . Histopathology analysis found that multinucleated giant cells ( MGC ) contain phagocytized fungal organisms multifocally in the lungs of both WT and CCR2-DTR mice ( S4C–S4D Fig ) . However , there were more extracellular fungal organisms and fewer macrophages and MGC in the CCR2-DTR lungs compared to the WT lungs ( S4C–S4D Fig ) . CCR2-DTR mice also had a decreased fungal burden in the mediastinal lymph node ( Fig 1E ) and a trend toward a decreased fungal burden in the brain ( Fig 1F ) on days 21 and 26 p . i . compared to WT mice . These results suggest that IM and their derivatives are harmful to the host during acute cryptococcosis by promoting fungal proliferation and dissemination from the lungs . To examine whether the recruitment of IM to the lungs promotes detrimental responses to C . neoformans , we utilized CCR2-/- mice [34] , in which the ability of monocytes to migrate out of the bone marrow ( BM ) is significantly impaired , resulting in an ~75% reduction in circulating monocytes under homeostatic conditions [35] . We confirmed that CCR2-/- mice have an ~79% decrease in the number of IM in the lungs compared to WT mice 14 days after C . neoformans challenge ( Fig 2A ) . CCR2-/- mice demonstrated a significant survival benefit ( median survival 31 . 5 days for CCR2-/- versus 26 days for WT ) ( Fig 2B ) and a lower lung fungal burden on day 14 p . i . ( Fig 2C ) compared to WT mice . Collectively , these data indicate that the initial recruitment of IM to the lungs plays a key role in mediating harmful outcomes during acute cryptococcosis . Consistent with this model , IM depletion later in the course of infection , by administering DT on days +6 , +8 , and +10 p . i . to CCR2-DTR mice , did not give rise to any differences in survival between CCR2-DTR and WT mice ( S3C Fig ) . To establish a strategy to investigate the role of IM functions in our model , we generated a CCR2-Cre mouse that allows us to conditionally knockout genes of interest in IM . We used a bacterial artificial chromosome ( BAC ) transgenic approach to introduce the Cre recombinase gene downstream of the CCR2 promoter ( Fig 3A and S5A Fig ) . We identified four potential CCR2-Cre founder mice and evaluated the efficiency and specificity of Cre expression in these mice by crossing them to Rosa26flSTOP-tdRFP mice [36] . The founder mouse selected to establish the CCR2-Cre colony had excellent expression of tdRFP in monocytes in all tissues analyzed , including ~90% of monocytes in the blood and lungs ( Fig 3B and S5B–S5G Fig ) . Ablation of IM led to a reduction in the number of pulmonary lymphocytes , including natural killer ( NK ) cells , innate lymphoid cell subsets ( ILCs ) , and CD4+ T cells on day 7 p . i . ( Fig 4 ) , suggesting that IM may regulate lymphocyte activity against C . neoformans . We also found a marked reduction in the levels of the Th2 cytokines IL-5 and RANTES/CCL5 along with a slight decrease in TNF production at the same timepoint in IM-ablated CCR2-DTR mice ( S6 Fig ) . These data indicated that IM could be deleterious because they facilitate the generation of harmful Th2 responses during acute cryptococcosis . To examine the possible link between MHC class II ( MHCII ) antigen presentation by IM and host Th2 responses , we generated CCR2-Cre MHCIIfl/fl mice to conditionally knockout MHCII expression in IM and their derivative cells . We confirmed the loss of MHCII expression in a subset of CD11b+CD11c+ cells , consistent with IM-derived DCs ( S7 Fig ) . However , conditional knockout of MHCII in these cells had no effect on survival ( Fig 5A ) or lung fungal burden on day 14 p . i . ( Fig 5B ) compared to MHCIIfl/fl littermate controls . Therefore , MHCII antigen presentation by IM and their derivatives does not mediate harmful immune responses to C . neoformans . To determine if lymphocytes are essential for IM-mediated outcomes , we next crossed the CCR2-DTR mice with RAG-/-γc-/- mice , in which lymphocytes , including T- and B- cells , NK cells , and ILCs , are absent . These CCR2-DTR RAG-/-γc-/- mice enabled DT-mediated ablation of IM in a lymphocyte-deficient background . We found that CCR2-DTR RAG-/-γc-/- mice had improved survival compared to non-transgenic RAG-/-γc-/- littermates ( median survival of 28 days for CCR2-DTR RAG-/-γc-/- versus 23 days for RAG-/-γc-/- ) ( Fig 6A ) and a decreased lung fungal burden on day 7 p . i . ( Fig 6B ) . Therefore , lymphocytes were not required to mediate the detrimental effects of IM in our model . We note that although IM are the most prevalent CCR2-expressing cells , subsets of T cells , NK cells , and ILC precursors can express variable levels of CCR2 [4 , 37 , 38] . Thus , these results confirm that IM , not other CCR2-expressing lymphocytes , were responsible for the phenotypes we observed . Previous studies have suggested that eosinophils are associated with cryptococcal disease in humans and mice [39–48] and positively correlate with murine susceptibility to cryptococcosis [45 , 49] . We observed a reduction in eosinophils on days 7 and 14 p . i . in the lungs of IM-ablated CCR2-DTR mice compared to WT littermates ( Fig 4 ) . Since the eosinophil-active cytokines IL-5 and RANTES/CCL5 were also diminished in CCR2-DTR mice ( S6 Fig ) , we investigated whether eosinophils may mediate the downstream effects of IM . After respiratory challenge with C . neoformans , eosinophil-deficient ΔdblGATA mice [50] did not exhibit any differences in survival ( Fig 7A ) or lung fungal burden on days 7 and 14 p . i . ( Fig 7B ) compared to WT control mice . These findings indicate that pulmonary eosinophilia likely represents a byproduct of an ineffective , Th2-skewed immune response rather than a functional immune response to C . neoformans . Therefore , eosinophils are unlikely to mediate the harmful effects of IM . Since our data suggested that cellular crosstalk between IM and lymphocytes or eosinophils is not a primary mechanism by which IM regulate infectious outcomes , we next examined the role of direct intrinsic functions of IM in mediating detrimental immune responses to C . neoformans . RNASeq analysis of IM sorted from the lungs of naive and infected CCR2-GFP reporter mice [5] on days 5 and 10 p . i . was performed ( Fig 8A ) . We found that , compared to naive IM , infected IM demonstrated significant increases ( P adj < 0 . 05 ) in the transcription of genes commonly associated with an alternatively activated ( M2 ) macrophage phenotype [51] , including Arg1 , mannose receptor C-type 1 ( Mrc1/Cd206 ) , transglutaminase 2 ( Tgm2 ) , resistin like alpha/found in inflammatory zone 1 ( Retnla/Fizz1 ) , and the chemokines Ccl17 and Ccl24 . Compared to naive IM , infected IM on day 10 p . i . also exhibited a high suppressor of cytokine signaling ( Socs ) 1:Soc3 ratio , that has been associated with the M2 phenotype [52 , 53] . There was no differential expression of the M2 marker chitinase-like 3 ( Chil3/Ym1 ) . Among transcripts associated with classically activated ( M1 ) macrophages , Nos2 was not detected by RNASeq , and , except for a slight increase in the chemokine Cxcl9 on day 10 p . i . ( P adj < 0 . 05 ) , there was no differential expression of the remaining transcripts . Among DC-associated transcripts , there was a slight increase in Jak2 on day 10 p . i . ( P adj < 0 . 05 ) but no changes in other transcripts , including those of the zinc finger and BTB domain containing 46 transcription factor ( Zbtb46 ) and MHC class II molecules ( H2-Eb1 , H2-Eb2 , H2-DMa , H2-Ab1 ) . The RNASeq findings were validated by qRT-PCR in naive and infected mice on day 10 p . i . ( Fig 8B ) . We confirmed there was no significant change in the M1 marker Nos2 , but we did detect a slight decrease in expression of Tnf in infected IM . We again saw significant increases in the M2 markers Arg1 , Mrc1 and Retnla/Fizz1 and no change in Chil3/Ym1 transcription , but we also detected an increase in the V-maf musculoaponeurotic fibrosarcoma oncogene homolog B transcription factor ( Mafb ) , that promotes macrophage differentiation [54] . Finally , we confirmed no change in the DC marker Zbtb46 . Together , these results suggest that C . neoformans subverts IM to assume an M2 macrophage-like phenotype that may be more permissive for fungal proliferation . Arg1 , an archetypal M2 macrophage marker , was one of the most highly expressed transcripts in infected IM ( Fig 8 ) . Previous studies indicate that ARG1 may compete with the M1 macrophage marker NOS2 for L-arginine substrate that it metabolizes into urea and L-ornithine , thereby reducing nitric oxide production by NOS2 [28–30] . Since M2 macrophages are less fungicidal than M1 macrophages against C . neoformans in vitro [31] , we investigated the potential role of IM-intrinsic ARG1 activity in the response to C . neoformans in the lungs by generating CCR2-Cre Arg1fl/fl mice . Our results showed that conditional knockout of Arg1 in IM decreased total arginase activity in the lungs by approximately 50% ( S8A Fig ) but did not have any effect on survival ( Fig 9A ) or lung fungal burden on days 7 and 14 p . i . ( Fig 9B ) . Additionally , we observed that Vav1-Cre Arg1fl/fl mice , that lack Arg1 in all hematopoietic cells , had no difference in survival compared to Arg1fl/fl or fl/+ littermate controls ( S8B Fig ) . Therefore , although infected IM highly express Arg1 , it does not appear that Arg1 mediates the detrimental effects of IM or other hematopoietic cells in response to C . neoformans challenge . STAT6 is a transcription factor that mediates M2 macrophage polarization in the context of IL-4 and IL-13 signaling by regulating expression of M2 macrophage markers , including Arg1 , Mrc1 , and Retnla/Fizz1 [55 , 56] . To determine if more global blockade of M2 macrophage polarization pathways would reverse the detrimental host outcomes in our model of cryptococcosis , we generated STAT6-/- BM chimeras to knockout STAT6 in hematopoietic cells , since STAT6fl/fl mice are not commercially available . There was no difference in overall survival ( Fig 9C ) and lung fungal burden on days 7 and 14 p . i . ( Fig 9D ) in STAT6-/- BM chimeras compared to control mice . Therefore , blocking STAT6-mediated pathways in hematopoietic cells , including IM , does not appear to be sufficient to counteract the harmful effects of IM during acute cryptococcosis . These results suggest that STAT6-regulated M2 markers , while associated with this macrophage phenotype , do not play an active role in macrophage function during acute cryptococcosis and that the subversion of IM function by C . neoformans more likely occurs at an earlier stage in the host-pathogen interaction .
In this study , we establish a critical role for IM in mediating detrimental host outcomes in a model of fatal respiratory infection with C . neoformans . These findings contrast with the beneficial functions of IM described in murine models of subacute and chronic pulmonary cryptococcosis and other fungal infections [4–8 , 11 , 14 , 15] but align with studies that suggest IM and their derivatives are associated with progression of infection by C . neoformans [20–27] . The studies on IM during subacute and chronic cryptococcosis utilized CCR2-/- mice on a BALB/c or mixed C57BL/6 and 129 background with the less virulent C . neoformans serotype D strain 52D [11 , 14 , 15] . We observed an improvement in lung fungal burden with ablation of IM in CCR2-DTR mice on the C57BL/6 background infected with 52D . However , Murdock et al . have shown that infection of C57BL/6 mice with the same inoculum of 52D i . t . is nonfatal up to 8 weeks p . i . , suggesting this combination represents a more chronic model of infection [57] . Therefore , mouse genetic background may be an important factor in determining the role of IM in the immune response to C . neoformans , aside from the acuity of the infection , though we cannot yet exclude any contribution from differences in the fungal strains themselves based on available data . In any case , these disparate results indicate that IM possess a plasticity of function that can regulate the outcomes of infection and , thus , would make them an important target for immunomodulatory therapies against C . neoformans . Accordingly , we sought to identify the mechanisms by which IM regulate infectious outcomes during cryptococcosis . To aid our investigation , we generated a CCR2-Cre transgenic mouse that demonstrates excellent Cre activity in IM , especially in the lungs . Along with the inducible Ccr2-creERT2 mouse previously generated by Becker and colleagues [58] , the constitutive CCR2-Cre mouse expands the tools available to dissect IM function in a variety of immunologic processes . One of our first observations was that IM contribute to Th2 cytokine production and regulate the presence of NK cells , ILCs , and CD4+ T cells in the lungs after C . neoformans challenge . IM and their macrophage and DC derivatives often play important roles in regulating lymphocyte responses to pulmonary infections through antigen presentation and chemokine secretion [5 , 59 , 60] . Interestingly , we found that neither conditional deletion of MHCII in IM nor lymphocyte deficiency affected IM-mediated infectious outcomes , suggesting that IM-lymphocyte crosstalk is not relevant for host immune responses during acute C . neoformans infection . The cellular source of IL-5 and RANTES/CCL5 in our model remains unclear . We did detect some transcription of RANTES/CCL5 , but not IL-5 , by IM in our RNASeq data . RANTES/CCL5 has been reported to be expressed by many cell types ( immgen . org [61] ) while the sources of IL-5 are typically Th2 cells and type 2 ILCs [62–64] . Thus , although the observed changes in these cytokines upon IM ablation may be related to direct secretion by IM , an alternative possibility is that differences in fungal burden influenced cytokine secretion by other immune cells , or a combination thereof . We also investigated whether IM regulate the immune response to C . neoformans through the recruitment of eosinophils to the lungs . Some studies have suggested that eosinophils may play beneficial roles in rat models of cryptococcosis [46 , 65 , 66] or other fungal infections like aspergillosis [67–69] . However , eosinophilia has been correlated with poor outcomes in mouse and human cryptococcosis [39–49] , and in the presence of IM in our model , we observed an upregulation of pulmonary IL-5 and RANTES/CCL5 , cytokines that play important roles in eosinophil maturation and trafficking [70] . Ultimately , we found that eosinophil-deficient ΔdblGATA mice have no change in infectious outcomes compared to WT mice . It has also been previously reported that IL-5-/- and eosinophil-deficient PHIL mice have little to no change in infectious outcomes during cryptococcosis , although the data was not formally shown [49] . Thus , our results confirm that eosinophils are a hallmark of the progression of cryptococcal infection , but do not appear , in and of themselves , to play an active role in mediating the outcomes of infection . Since cellular crosstalk between IM and lymphocytes or eosinophils did not seem to play an important role in our model , we next investigated the intrinsic functions of IM that may facilitate progression of C . neoformans infection . Based on our transcriptional profiling of infected IM , it appears that these cells preferentially express markers of M2 macrophages in response to pulmonary challenge with C . neoformans . M2 macrophages are generally anti-inflammatory cells and are involved in tissue homeostasis or repair , though they can be phenotypically heterogenous [71] . It has previously been reported that C . neoformans may use monocytes and macrophages as protected reservoirs or “Trojan horses” that aid in fungal dissemination [21–25 , 27] and that M2 macrophages are less fungicidal against C . neoformans than M1 macrophages [31] . Additionally , our histopathology analysis demonstrated that the lack of IM decreases the number of macrophages and multinucleated giant cells in the lung parenchyma and increases the incidence of extracellular fungal organisms in the lungs . Therefore , these results support the idea that exposure to C . neoformans renders IM and their macrophage derivatives permissive for fungal proliferation , so that their physical absence from the lungs can actually ameliorate infectious outcomes . It remains unclear whether IM directly facilitate dissemination of C . neoformans in this acute infection model or whether the higher lymph node and brain fungal burdens in WT mice are simply a reflection of the higher lung fungal burden in these mice compared to IM-ablated CCR2-DTR mice . However , studies on other fungal and bacterial infections have previously demonstrated the ability of IM to participate in direct transport of microbes from the lung to disseminated sites [5 , 72 , 73] . Next , we examined if the harmful effects of IM during acute cryptococcosis could be reversed by targeting M2 macrophage polarization pathways . Arg1 is a known M2 marker that was markedly upregulated in infected IM . Previous studies indicate that the fungal pathogen C . albicans can suppress NOS2 activity and induce ARG1 activity in human macrophages , resulting in decreased NO production and improved fungal survival [74–77] . Our studies show that conditional knockout of Arg1 in IM does not improve survival or lung fungal burden after C . neoformans challenge . Although it has been suggested that ARG1 and NOS2 can compete for the same pool of L-arginine substrate [29 , 30 , 78 , 79] , it may be that disrupting ARG1 activity and increasing L-arginine availability is not sufficient . A second , positive signal may also be required to induce NOS2 transcription and activity , although it would be important to avoid triggering an uncontrolled inflammatory reaction that could lead to harmful immunopathology . Other studies have also indicated that despite the association of Arg1 with M2 macrophages , ARG1 may not always be a functional component of the immune response [80] . We subsequently considered whether the functional outcomes of M2 polarization may be controlled further upstream . Previous studies indicate that global deletion of IL-13 , IL-4 , and IL-4Rα , as well as conditional knockout of IL-4Rα in myeloid cells , improves murine outcomes after respiratory challenge with C . neoformans [81–83] . Therefore , we investigated the role of STAT6 , a transcription factor stimulated by IL-4 and IL-13 that regulates the expression of several M2 markers including Arg1 [55 , 56] . Knockout of STAT6 in hematopoietic cells using BM chimeras resulted in similar survival and lung fungal burden as control mice . These results contrast with the worse survival of global STAT6-/- mice infected with C . neoformans strain KN99α , an H99-derived strain , observed by Wiesner et al [64] . Since we used BM chimeras , it is possible that a radioresistant cell population is contributing to the phenotype observed by Wiesner et al or that differences between the parental H99 strain and KN99α could be in play . Nevertheless , our results indicate that STAT6 signaling in IM does not play an important role in our infection model and suggest that C . neoformans may suppress pro-inflammatory signals further upstream that would otherwise direct the differentiation of IM into fungicidal M1 macrophages . For example , we previously observed that mice deficient in the DAP12 signaling adapter have improved infectious outcomes after C . neoformans challenge and that DAP12-deficient macrophages have improved uptake and killing of C . neoformans in vitro [84] . Thus , DAP12-mediated pathways may be important targets for promoting a more beneficial , classically activated immune response to C . neoformans . Additional unanswered questions about the role of IM during cryptococcosis remain . We do not yet know if IM may influence the function of other myeloid cells like neutrophils , the role of which remains unclear during C . neoformans infection ( reviewed in [85] ) , or of non-myeloid cells , e . g . , lung epithelial cells , that can coordinate innate immune responses to other fungal infections [86] . Given the potential influence of mouse genetic background and fungal strain on IM function , it may be important to evaluate not only host immune differences , but also fungal-specific factors like capsule composition and Titan cell formation that may alter host-pathogen interactions [87 , 88] . In summary , our study establishes a novel role for IM as crucial arbiters of infectious outcomes during acute cryptococcosis . Unlike in other pulmonary and disseminated fungal infections [4–8] , IM do not aid in host defense but rather are subverted by C . neoformans to maintain a passive state that can be harnessed by the fungus for replication and dissemination . Our work was assisted by the generation of a CCR2-Cre mouse that will facilitate continued mechanistic evaluation of IM function in cryptococcosis . Using genetic reprogramming to target pathways that result in classical activation of IM and aid in fungal clearance would validate the concept that immunomodulation can be developed as a new therapeutic approach to manage cryptococcal infections .
Chemicals were from Sigma-Aldrich , cell culture reagents were from Life Technologies/Gibco , and microbiological culture media were from BD Biosciences unless otherwise noted . Arginase activity was measured using an Arginase Activity Assay Kit ( Sigma ) . Antibodies for flow cytometry were purchased from BD Biosciences , eBioscience or Tonbo unless otherwise indicated . Restriction enzymes were from New England Biolabs . C57BL/6J ( stock #000664 ) , MHCIIfl/fl ( stock #013181 ) [89] , and Arg1fl/fl ( stock #008817 ) [90] mice were purchased from the Jackson Laboratory ( JAX ) . CCR2-/- mice ( JAX stock #004999 ) [34] were generously provided by Dr . E Pamer ( MSKCC ) . Rosa26flSTOP-tdRFP mice [36] were generously provided by Dr . J . Sun ( MSKCC ) [91] . The ΔdblGATA mice [50] on a C57BL/6 background were generously provided by Dr . H . Rosenberg ( NIH ) . Vav1-Cre mice ( JAX stock #008610 ) [92] were generously provided by Dr . F Geissmann ( MSKCC ) . RAG2-/-γc-/- mice ( stock #4111 ) [93 , 94] were purchased from Taconic . CD45 . 1+ mice ( stock #564 ) were purchased from Charles River Laboratories . The CCR2-DTR depleter mice and CCR2-GFP reporter mice were generated as previously described [5 , 95] . All mouse strains were bred and housed in the Memorial Sloan Kettering Cancer Center’s ( MSKCC ) Research Animal Resource Center under specific pathogen-free conditions . Mice on the RAG-/-γc-/- background were maintained on amoxicillin- and Vitamin E-containing chow . To ablate monocytes , CCR2-DTR or CCR2-DTR RAG-/-γc-/- mice and their control littermates were injected intraperitoneally with 200 ng ( 10 ng/g body weight ) of diphtheria toxin ( List Biological Laboratories ) every other day for three doses , starting the day before infection ( Fig 1B ) , unless otherwise noted . All experiments were conducted using male and female mice at age 6–8 weeks with sex- and age-matched mice in experimental and control groups . Experiments with CCR2-DTR , CCR2-DTR RAG-/-γc-/- , CCR2-Cre and Vav1-Cre mice used littermate control mice that were weaned from the same litters and co-housed . All animal studies were performed with approval from the MSKCC Institutional Animal Care and Use Committee under protocol 13-07-008 and were compliant with all applicable provisions established by the Animal Welfare Act and the Public Health Services ( PHS ) Policy on the Humane Care and Use of Laboratory Animals . A bacterial artificial chromosome ( BAC ) containing the CCR2-Cre transgene was generated using the recombineering strategy developed by Heintz and colleagues [96 , 97] . Briefly , the BAC clone RP23-182D4 containing the endogenous CCR2 locus [5] was obtained from the BACPAC Resource Center at the Children’s Hospital Oakland Research Institute ( CHORI ) . A 970 bp fragment upstream and a 711 bp fragment downstream of the CCR2 start codon were amplified ( see all primers in Table 1 ) . The 987 bp Cre gene was amplified from the CreERt2 frt Kan ( R ) frt plasmid [98] , kindly provided by Dr . T . Buch ( Univ . of Zurich ) . Overlap PCR was performed with these three fragments to generate a 2668 bp CCR2-Cre recombination cassette , in which the Cre gene is flanked by two homology boxes from the CCR2 gene and the first nucleotide of the CCR2 endogenous locus after the Cre stop codon is deleted ( Fig 3A ) . The recombination cassette was cloned into the AscI and NotI sites of the pLD53 . SC-AB shuttle vector [96] , previously provided by Dr . D . Littman ( NYU ) [5] . The cassette was sequenced with overlapping primer sets to ensure the absence of mutations . For homologous recombination , the modified shuttle vector was electroporated into the BAC clone RP23-182D4 . Clones that underwent BAC cointegration and resolution were isolated by chloramphenicol and ampicillin selection followed by sucrose negative selection , as previously described [5 , 96] . Proper integration of the CCR2-Cre construct into the BAC was confirmed by Southern blot ( S5A Fig ) and sequencing of the modified regions . The CCR2-Cre BAC was then purified and injected into fertilized C57BL/6J oocytes by the University of Michigan Transgenic Animal Core . Four potential founder mice were identified out of 30 pups screened by PCR . These founders were bred to Rosa26flSTOP-tdRFP mice and immune cells in the BM , blood , lungs and spleen were evaluated by flow cytometry . The progeny of two founders exhibited comparable tdRFP expression in a high percentage of monocytes ( Fig 3B ) , and the lineage of one of these founders ( #584 ) was chosen to establish the CCR2-Cre colony . Bone marrow from STAT6-/- mice ( JAX stock #005977 ) [99] to generate chimeras was kindly supplied by Dr . P . Loke ( NYU ) . Recipient CD45 . 1+ mice were exposed to 900 cGy in a cesium irradiator and then given 3–5 x 106 donor STAT6-/- or C57BL/6 ( CD45 . 2+ ) BM cells by tail vein injection . Baytril 100 ( enrofloxacin ) was provided in the drinking water at a concentration of 0 . 4 mg/mL for the first 2 weeks after irradiation . Mice were used in experiments 6–8 weeks after irradiation . C . neoformans serotype A strain H99 #4413 was kindly provided by Dr J . Heitman ( Duke ) . C . neoformans serotype D strain 52D ( 24067 ) was obtained from ATCC . All C . neoformans strains were maintained and grown as previously described [84] . Briefly , fungal strains were grown on Sabouraud dextrose agar ( SAB ) plates from frozen glycerol stocks and then cultured overnight at 37°C in liquid YPD medium ( 1% yeast extract , 2% peptone , 2% dextrose ) . Fungal cells were then washed and resuspended in phosphate-buffered saline ( PBS ) at a concentration of 103 cells per 50 μl volume . Mice were anesthesized with inhaled isoflurane and given 50 μl of the fungal cell suspension intratracheally using a blunt-ended 20-gauge needle , as previously described [100] . To assess fungal burden , murine lung and brain tissue were mechanically homogenized in PBS using a PowerGen 125 homogenizer ( Fisher ) . Lymph nodes were dissociated using ground glass slides . CFU in all tissues were counted after plating serial dilutions of the homogenates on SAB plates . To analyze cytokine levels , whole lungs were mechanically homogenized in 2 mL PBS containing cOmplete Protease Inhibitor Cocktail ( Roche ) . ELISA assays were then performed on the supernatants from the homogenates using Ready-SET-Go ELISA kits ( eBiosciences ) , except DuoSet ELISA Development Systems kits ( R&D ) were used to measure RANTES/CCL5 and IL-5 . For flow cytometry analysis , single cell lung suspensions were prepared as previously described [5] by digestion with DNase I ( Roche ) and collagenase type 4 ( Worthington Biochemical ) and mechanical disruption using a gentleMACS Dissociator ( Miltenyi Biotec ) . Total lung cells were counted using a Coulter counter and stained with fluorescent antibodies . Flow cytometry data was collected on a BD LSR II flow cytometer and analyzed with FlowJo ( v9 . 7 . 6 ) . Monocyte progenitors are defined as Lin- ( CD3ε -CD19-NKp46-Sca-1-Ly6G- ) CD45+CD117 ( c-Kit ) +CD115+ with MDPs being Ly6Clo and cMoPs being Ly6Chi . Inflammatory monocytes are CD45+Ly6G-MHCIIloCD11b+Ly6Chi . Ly6Clo monocytes are CD45+Ly6G-MHCIIloCD11b+Ly6Clo . Macrophages are CD45+Ly6G-SiglecF+CD11chi . DCs are CD45+Ly6G-CD11c+MHCIIhi and either CD11b+ or CD103+ . Neutrophils are CD45+CD11b+Ly6G+ . Eosinophils are CD45+Ly6G-SiglecF+CD11clo . NK cells are Lin- ( CD3ε-CD5-CD19-CD11c- ) CD45+TcRβ+NK1 . 1+ . ILCs are Lin- ( CD3ε-CD5-CD8α-CD19-CD11c-CD11b-NK1 . 1- ) CD45+CD90 . 2+CD127+ . CD4+ T cells are CD45+CD3ε+CD90+CD4+ . CD8+ T cells are CD45+CD4-CD19-CD8+ . CD19+ B cells are CD45+CD4-CD8-CD19+ . For additional gating information , see reference [84] . The lungs of euthanized mice were perfused with 4% paraformaldehyde ( PFA ) in situ via a catheter inserted through an incision in the trachea . The lungs were then harvested and fixed by immersion in 4% PFA . Lungs were then processed by the MSKCC Molecular Cytology Core Facility to generate 4 μm sections of paraffin-embedded lungs stained with hematoxylin & eosin ( H&E ) . Slides were scanned using a Zeiss Mirax Midi slide scanner with 20x/0 . 8NA objective . Slides were reviewed and scored by a pathologist . Morphometry analysis was carried out on scanned slide images using Pannoramic Viewer ( v1 . 15 . 3 , 3DHISTECH ) . Areas of lung inflammation were measured and expressed as a percentage of total lung area in each histologic section . Monocytes were pre-enriched from single cell lung suspensions pooled from 6–7 CCR2-GFP mice per group using the negative selection EasySep Mouse Monocyte Isolation Kit ( STEMCELL Technologies ) . The enriched cells were then analyzed with a BD FACSAria cell sorter to obtain DAPI-Lin-CD11b+Ly6ChiCCR2GFP cells ( Lin- = Ter119-CD3-CD19-NK1 . 1-CD11c-Ly6G- ) . RNA was extracted from sorted cells using TRIzol LS ( Thermo Fisher Scientific ) according to the manufacturer’s instructions . RNASeq was performed by the MSKCC Integrated Genomics Operation . After RiboGreen quantification and quality control by an Agilent BioAnalyzer , 500ng of total RNA underwent polyA selection and TruSeq library preparation according to instructions provided by Illumina ( TruSeq Stranded mRNA LT Kit , catalog # RS-122-2102 ) , with 8 cycles of PCR . Samples were barcoded and run on a HiSeq 2500 in a 50bp/50bp paired end run , using the TruSeq SBS Kit v4 ( Illumina ) . An average of 44 . 6 million paired reads was generated per sample . At the most the ribosomal reads represented 0 . 01% of the total reads generated and the percent of mRNA bases averaged 73 . 5% . The RNASeq data have been deposited in NCBI’s Gene Expression Omnibus ( GEO ) [101] and are accessible through GEO Series accession number GSE122765 ( https://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE122765 ) . Statistical analysis of RNASeq data was performed by the MSKCC Bioinformatics Core . The output data ( FASTQ files ) were mapped to the target genome using the rnaStar aligner [102] that maps reads genomically and resolves reads across splice junctions . The 2 pass mapping method [103] was used , in which the first mapping pass uses a list of known annotated junctions from Ensembl . Novel junctions found in the first pass were then added to the known junctions and a second mapping pass was performed in which the RemoveNoncanoncial flag was used . After mapping , the output SAM files were post processed using the PICARD tool AddOrReplaceReadGroups to add read groups , sort the files , and covert them to the compressed BAM format . The expression count matrix was then computed from the mapped reads using HTSeq ( www-huber . embl . de/users/anders/HTSeq ) and one of several possible gene model databases . The raw count matrix generated by HTSeq was then processed using the R/Bioconductor package DESeq ( www-huber . embl . de/users/anders/DESeq ) which was used to both normalize the full dataset and analyze differential expression between sample groups . For quantitative RT-PCR , cDNA was generated from RNA using a QuantiTect Reverse Transcription Kit ( Qiagen ) , and qRT-PCR was performed on a StepOnePlus Real Time PCR System ( Applied Biosystems ) using TaqMan Fast Advanced Master Mix and TaqMan Gene Expression Assays ( ThermoFisher Scientific ) including Arg1 ( Mm00475988_m1 ) , Mrc1 ( Mm01329362_m1 ) , Retnla/Fizz1 ( Mm00445109_m1 ) , Hprt ( Mm03024075_m1 ) , Nos2 ( Mm00440502_m1 ) , Tnf ( Mm00443258_m1 ) , Chil3/Ym1 ( Mm00657889_mH ) Zbtb46 ( Mm00511327_m1 ) , and Mafb ( Mm00627481_s1 ) . All results are expressed as mean ± SEM . A Mann-Whitney U test was used for statistical analysis of two group comparisons , and one-way ANOVA was used for three groups or more , unless otherwise noted . Survival data was analyzed by Mantel-Cox test . All statistical analyses were performed with GraphPad Prism software , v6 . 0f . A P value < 0 . 05 was considered significant and indicated with an asterisk . | Cryptococcus neoformans is a fungus that is prevalent throughout the environment and can cause a fatal infection of the central nervous system when inhaled into the lungs by patients with impaired immune systems . Our understanding of the immune responses that either help clear C . neoformans from the lungs or permit development of disease remains limited . In this study , we used a mouse model of lethal C . neoformans infection to determine that inflammatory monocytes , immune cells that are often among the first responders to infections , actually facilitate the progression of infection rather than clearance . These findings establish a foundation for future work to target the immune response of inflammatory monocytes as a strategy to improve the outcomes of patients that develop C . neoformans infections . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"cryptococcosis",
"cryptococcus",
"neoformans",
"medicine",
"and",
"health",
"sciences",
"cryptococcus",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"pathogens",
"immunology",
"microbiology",
"pulmonology",
"animal",
"models",
"fungi",
"model",
"organisms",
"signs",
"and",
"symptoms",
"experimental",
"organism",
"systems",
"fungal",
"diseases",
"fungal",
"pathogens",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"mycology",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"animal",
"studies",
"medical",
"microbiology",
"microbial",
"pathogens",
"mouse",
"models",
"immune",
"response",
"eukaryota",
"diagnostic",
"medicine",
"cell",
"biology",
"monocytes",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms"
] | 2019 | Inflammatory monocytes are detrimental to the host immune response during acute infection with Cryptococcus neoformans |
Membrane proteins are frequently present in crowded environments , which favour lateral association and , on occasions , two-dimensional crystallization . To better understand the non-specific lateral association of a membrane protein we have characterized the free energy landscape for the dimerization of a bacterial outer membrane protein , NanC , in a phospholipid bilayer membrane . NanC is a member of the KdgM-family of bacterial outer membrane proteins and is responsible for sialic acid transport in E . coli . Umbrella sampling and coarse-grained molecular dynamics were employed to calculate the potentials of mean force ( PMF ) for a variety of restrained relative orientations of two NanC proteins as the separation of their centres of mass was varied . We found the free energy of dimerization for NanC to be in the range of to . Differences in the depths of the PMFs for the various orientations are related to the shape of the proteins . This was quantified by calculating the lipid-inaccessible buried surface area of the proteins in the region around the minimum of each PMF . The depth of the potential well of the PMF was shown to depend approximately linearly on the buried surface area . We were able to resolve local minima in the restrained PMFs that would not be revealed using conventional umbrella sampling . In particular , these features reflected the local organization of the intervening lipids between the two interacting proteins . Through a comparison with the distribution of lipids around a single freely-diffusing NanC , we were able to predict the location of these restrained local minima for the orientational configuration in which they were most pronounced . Our ability to make this prediction highlights the important role that lipid organization plays in the association of two NanCs in a bilayer .
Cellular membranes not only separate the contents of a cell from its surroundings , they also play a key role in cell regulation and metabolism . Accounting for approximately a quarter of the coding regions of an organism's genome [1] , membrane proteins control the transport of solutes between a cell and its surroundings , facilitate cellular movement , and regulate many aspects of cellular behaviour . Gram-negative bacteria are surrounded by two membranes separated by a periplasmic layer . The outer membrane lipid bilayer is composed of phospholipids in the inner ( i . e . periplasmic ) leaflet , and of lipopolysaccharides in the outer leaflet . Within this membrane are many species of outer membrane proteins ( OMPs ) , a class of integral membrane proteins whose secondary structures are almost exclusively [2] . Many of these are porins ( OmpC , OmpF , LamB , NanC , for example ) , through which small ( approximately ) molecules can diffuse across the membrane . Porins provide a route for many antibiotics into bacterial cells and are potential vaccine targets [3] . Both in vivo and in vitro , membrane proteins are often present in a crowded environment . Thus , cell membranes generally have a high membrane area fraction ( approximately 25% or greater ) occupied by proteins [4] . A similar degree of crowding may be found in membranes studied in vitro [5] , [6] . Such crowding may result in the clustering of proteins [7] . Whilst the majority of discussion as to the nature of membrane protein cluster formation has focussed on lipid rafts [8] , it should be noted that lateral interactions of crowded membrane proteins are a more general property [9]–[12] . In vitro , control of lateral association of a single membrane species in a highly crowded system may be used to induce two-dimensional crystallization [13] . Interactions within a crowded environment may lead to dynamic lateral interactions of membrane proteins , for example , those seen in recent time-resolved AFM studies of OmpF-containing membranes [6] . In studying such interactions , one wishes to distinguish between specific oligomerization of membrane proteins ( for example dimerization of transmembrane in glycophorin [14] and of in OMPLA [15] ) and less specific interactions . Specific protein interactions are those in which the distributions of orientations of the oligomerized proteins are grouped almost exclusively into very few states ( often only a single state ) . Less specific ( or non-specific ) interactions are those determined by other effects , such as ( local ) crowding , rather than purely due the specific interactions between residues on each of the proteins . In less specific oligomerization there may be some orientational dependence , but a more broad distribution of orientations is generally found among the oligomers . Benjamini and Smit found that it was important to determine the effect that non-specific interactions had on the crossing angle for pairs of before investigating the role of any specific interactions between the helices [16] . It is therefore of interest to explore the energy landscape of lateral interactions of a relative ‘featureless’ OMP . NanC ( Figure 1A ) provides a good example of such a protein , as it is both structurally [17] and functionally [18] monomeric , whilst forming two dimensional crystals in DMPC lipid bilayers [19] . NanC is member of the KdgM-family of bacterial outer membrane proteins , responsible for sialic acid transport in E . coli . Experimental determination of the free energy of membrane protein dimerization in vitro has been used to characterize their properties in a membrane or membrane-like environment [20]–[25] . Characterization of the free energy landscape for membrane proteins gives us an insight into how the proteins will move and interact within the membrane and allows us to make predictions about their behaviour . There are many published examples of experimentally determined dimerization energies for membrane proteins and peptides [20]–[22] , [24] , [25] , but relatively few for proteins: one important example being the dimerization free energy of the phospholipase OMPLA , which was found to be in the region to [23] . Computer simulations provide a complement to both in vitro and in vivo experiments [26] , enabling us to probe the microscopic interaction underlying membrane protein association . Molecular dynamics ( MD ) simulations have been used to explore a range of membrane proteins [26] , in addition to related approaches such as Monte Carlo [27] and Brownian dynamics [28] simulations . In particular , simulations using a coarse-grained approximation [29] have been used to study dimerization of transmembrane domains [30] and of rhodopsin [31] . In the latter case the simulations were also used to characterize large-scale organization of rhodopsin dimers into rows-of-dimers , as seen experimentally in disk membranes . Many computational studies that explore free energy landscapes use the potential of mean force ( PMF ) [32] as a convenient description because it enables us to characterize a given reaction or transition as a function of a specific reaction coordinate ( or set of coordinates ) . Not only does this enable us to characterize the free energy landscape as a function of the reaction coordinate ( or coordinates ) , it also provides an opportunity subsequently to parameterize reduced models of complex systems using different simulation paradigms [33] . Calculations of PMFs for the association of membrane proteins have largely focussed on proteins . Thus , dimerization free energy landscapes for transmembrane have been calculated using MD with umbrella sampling [30] , [34] and with adaptive biasing force methods [35]; and also using Monte Carlo [27] , and dissipative particle dynamics [36] . These have yielded free energies of dimerization in the region of to . To date there has only been one computational study to calculate the association free energy of a membrane protein: the free energy of association for two OmpF trimers was estimated to be in the region of [6] . It has long been suggested that lipids play an important role in the interaction between proteins in a membrane [37] , [38] . For example , simulation studies have shown that hydrophobic mismatch may modulate the aggregation of proteins in the membrane [39] , [40] . It is therefore important that we capture the effects that lipids have on free energy landscapes if we are to understand membrane protein association in different bilayer environments . In this paper we develop and apply a method for calculating the free energy of association for rotationally restrained proteins in a lipid bilayer . This allows us to resolve detailed structure in the ( one-dimensional ) PMFs , which reflect protein-lipid-protein interactions . We apply this method to characterize the association free energy of a coarse-grained model of NanC .
The PMFs calculated for each of the four rotational combinations are shown in Figure 2 . The PMFs were set to zero at an inter-protein separation of 8 nm , where the potentials have become approximately constant . The sampling methods , biasing potentials , rotational restraints and simulation details are given in the Methods section . We categorized the PMFs in Figure 2 by the depths of their potential well , which resulted in three categories of well depth . The first category contains the PMFs in Figures 2A and B , which both have depths of approximately occurring at inter-protein separations of approximately 3 . 2 nm . This first category corresponds to the orientational configurations of maximal contact , and , where wide faces of both proteins are brought into contact ( shown in Figures 1D and E , respectively ) . It is interesting to note that the depths of the PMFs for these two parallel and anti-parallel orientational configurations are approximately the same . They are also similar in depth to the orientationally-unrestrained PMF calculated for this coarse-grained NanC system ( shown in Figure S2 ) , which has a depth of . This is much greater than the to calculated for the dimerization of OMPLA [23] , the only experimental free energy for dimerization of an OMP , but as that was for a protein exhibiting specific oligomerization measured in detergent micelles , we would not expect a good agreement . The next category contains the PMF in Figure 2C , with a potential well depth of approximately occurring at a separation of approximately 3 . 5 nm . This corresponds to the intermediate orientational configuration in Figure 1F , where a wide face of one protein is brought into contact with a narrow face of the other . The decrease in the depth of the PMF indicates that the configurations with two wide faces in contact are more stable than this intermediate contact configuration , where . The third category contains the PMF shown in Figure 2D , which is the shallowest of the four PMFs with a potential well depth of approximately , occurring at an inter-protein separation of 3 . 5 nm . This PMF corresponds to the orientational configuration with minimal protein contact , where narrow faces of both proteins are brought into contact ( shown in Figure 1G ) . This configuration is the least stable of the four configurations considered here . The correlation between the depth of the PMFs and the orientational configuration of the proteins suggests that the strength of the interaction may correlate with the overall extent of the resultant protein-protein interface . As well as the restrained global minima ( the global minima for the specific restrained orientations ) of the potential wells in the PMFs of Figure 2 , there are also multiple local minima , which occur at a variety of centre of mass separations . For example , the PMF in Figure 2B has a restrained global minimum ( labelled ) and two higher-energy local minima ( labelled and ) , which we refer to as restrained metastable states . By fitting quadratic curves to the minima in Figure 2B we calculated their locations as 3 . 26 nm , 3 . 62 nm and 4 . 07 nm for , and , respectively . The nature of the restrained global ( ) and local ( and ) minima is illustrated by the simulation snapshots shown in Figures 3A–C . These snapshots were taken from the simulation windows used to calculate the PMF in Figure 2B for an orientational configuration of . The snapshot shown in Figure 3A is from the umbrella sampling window in which the proteins were restrained with a separation of 3 . 3 nm , which is closest to the minimum at 3 . 26 nm in Figure 2B , labeled . We see that there is one lipid molecule between the two proteins at this restrained global minimum . It should be noted that this is the only lipid in between the two proteins; there is no equivalent lipid on the extracellular side of the membrane ( the view from the other side of the membrane is shown in Supporting Information Figure S1 ) , so the restrained global minimum configuration for this orientation has space for one lipid on the periplasmic side of the membrane . A snapshot from the umbrella sampling window with the proteins restrained with a separation of 3 . 6 nm is shown in Figure 3B , which is the window closest to the minimum at 3 . 62 nm , labelled in Figure 2B . We can see that there are two lipid molecules between the two proteins in this snapshot . The snapshot in Figure 3C is taken from the umbrella sampling window in which the proteins are restrained with a separation of 4 . 1 nm , which is the window closest to the minimum at 4 . 07 nm , labelled in Figure 2B , in which we see that three lipid molecules can occupy the space between the two proteins . These observations suggest that the existence of these restrained metastable states is a result of protein-lipid-protein interactions in this orientationally-restrained system . To investigate the suggestion that these restrained metastable states were the result of the lipid ordering between the proteins , we calculated the lipid distribution around a freely diffusing NanC in a POPE bilayer . The distribution for a specific coarse-grained particle in the tail of all of the lipid molecules is shown in Figure 3D , where distinct annuli are visible , indicating regions of preferred occupation . We calculated the lipid distribution in a direction that corresponds to the direction of the other protein for the orientational configuration , indicated by the region between the dashed lines in Figure 3D . The average lipid distribution across both leaflets and all coarse-grained lipid particles in this direction is shown in Figure 3E , where again we can see there are preferred distances from the protein at which the lipids are observed . Further details of the averaging calculation are given in the Methods section . We can use this directional lipid distribution to predict the separations at which the region between two proteins would be optimally packed by the lipid molecules . The alignment process is illustrated in Figures 3F–G and explained in the Methods section . For the minimum labelled in the PMF in Figure 2B , which occurs at a separation of 3 . 26 nm , we predict an optimal separation of 3 . 24 nm with one intervening lipid . For the first restrained metastable state labelled in Figure 2B , which occurs at a separation of 3 . 62 nm , we predict a separation of 3 . 63 nm with two intervening lipids . For the second restrained metastable state labelled in Figure 2B , which occurs at a separation of 4 . 07 nm , we predict a separation of 4 . 02 nm with three intervening lipids . Our predictions for the locations of the restrained metastable states are in close agreement with their location in the PMF . This supports our suggestion that the restrained metastable states observed in the PMFs are due to the protein-lipid-protein effects caused by the distribution of lipids between the two NanC proteins . For the other orientational configurations , the proteins have different faces facing the other protein and will therefore have a different optimal lipid distribution for each face . This may be one reason why the restrained local minima are better defined for and occur at regular intervals . Such features are not usually observed in PMFs calculated with proteins that are free to rotate ( for example , see Figure S2 for an orientationally-unrestrained PMF calculated for this coarse-grained NanC system ) . In the orientationally-unrestrained case the proteins would be able to rotate to alter the distance between their surfaces , provided they are not perfectly rotationally symmetric , so that the intervening region could be optimally packed with lipids without leaving any voids . However , for a system with rotationally restrained proteins , there is an optimal separation at which multiple lipid molecules can occupy the intervening space between the proteins . Also observed in each of the PMFs is an energetic barrier , which occurs at an inter-protein separation of approximately 5 . 5 nm . Extending the arguments made above about the interaction of the two proteins individual lipid distributions , we can see that at distances greater than 2 . 5 nm in Figure 3E the fluctuations in lipid distribution have decayed to small oscillations around some constant average value , which indicates that these lipids are not as strongly influenced by the protein . From this argument we can think of this barrier as the point at which the lipids whose positions are strongly dependent on each protein begin to interact with one another , that is , there are lipids between the proteins that are affected by both of the proteins . We can think of this as the separation at which the annuli of lipids around each protein overlap with each other . We wished to formally characterize the dependence of the PMF depth on the orientation of the proteins , which we suggested was related to the extent of protein contact . To do this we calculated the solvent accessible surface area ( SASA ) of the two proteins as a function of the separation of their centres of mass . For proteins with an approximately elliptical cross-section , we would expect the orientations with greater contact between the proteins to have a larger buried surface area . However , given the significance of the lipid effects that we identified above , it is important to corroborate this . Any features of the combined surfaces of the two proteins that would allow room for a lipid could have a large effect on the free energy . The SASA is calculated using a spherical probe whose size determines the level of detail in the surface calculated for a specific set of atoms/particles . We used a probe with a radius of 0 . 47 nm , which is twice the radius of the coarse grained particles ( 0 . 235 nm ) and should be a reasonable measure for the size of a lipid . We chose this size probe because it is the lipids that are the ‘solvent’ of interest when we bring two proteins together in a membrane . Further details are given in the Methods section . For each of the four orientational configurations , Figure 4 shows the buried surface area as a function of distance from the minimum of their respective PMFs . We chose this measure since we wanted to remove the effect that the difference in protein radius has on the location of the minimum . In Figure 4 we see that there is a stratification of the buried surface areas in the region around the minima of the PMFs . As with the PMF depths in Figure 2 , the buried surface areas can be divided into three categories . The buried surface area is largest for the orientations and , where the two wide protein faces are brought together . The next largest buried surface area around the minimum of the PMF is for the orientation , where one narrow face is brought into contact with one wide face . The smallest buried surface area around the minimum of the PMF is for the orientation , where two narrow faces are brought into contact . The correlation between the depth of the PMF and the buried surface area can be seen in the inset plot in Figure 4 , in which these two quantities are plotted . We see that there is a negative correlation between the two quantities . For a protein orientation with a larger buried surface area , the minimum of the PMF is deeper .
We have characterized the free energy landscape of a pair of NanC proteins in a phospholipid bilayer . An interesting feature of these restrained free energy calculations is that certain restrained metastable states , which would usually not be seen , are now resolved . These local minima are associated with the ability of lipids to occupy the space between the two proteins at a given separation . Niemelä et al . [42] found that close to proteins , the lipids in a bilayer have reduced mobility , diffusing with the protein , and it is interactions involving these surrounding lipids upon protein association that we are observing here . Proteins' interactions with lipids have been shown to modulate local lipid formation [43] , further demonstrating the important role of interactions involving both proteins and lipids in determining structures observed in the bilayer . PMFs for the dimerization of TM helices [27] , [30] and for other more complex proteins , including rhodopsin [44] and OmpF [6] have revealed similar features , suggesting that a role for lipids in the energetics of their interactions may be a general feature of membrane proteins . These features may not affect the kinetics of association , as we do not know if they present metastable barriers to association , but they will affect the dynamics of the system; the NanC proteins will need to negotiate the complex free energy landscape created by these protein-lipid-protein interactions if they are to reach an energetically stable state through oligomerization . This result may also be compared with studies of membrane protein interactions using more approximate ( and hence more general ) models and DPD simulations [36] . For example , such studies have suggested that changes in lipids may result in the modulation of mismatch-driven interactions of membrane proteins [45] . We identified a correlation between of the depth of the well in the free energy of association with the buried surface area at the interface of the two proteins . More generally , it has been suggested that oligomer stability of membrane proteins such as glycophorin A [22] and bacteriorhodopsin [46] may be correlated with the buried surface area at the interface . However , studies of the dimeric outer membrane protein OMPLA [15] failed to reveal such a correlation . This may reflect the role of lipids in OMPLA dimerization , confirming the need for detailed energy landscape calculations such as those presented herein . Features of the methodology used in this work mean that care should be taken when interpreting the results . The treatment of solvents in the coarse-grained model is only approximate , so entropic contributions to solvation and lipidation/delipidation may not be captured as reliably as with a fully atomistic model . The nature of the coarse-grained model also does not enable us to separate out the contributions to the PMF due to energy and entropy , as is sometimes done using atomistic calculations . However , the observations we make here relating to the behaviour of lipids is mostly phenomenological and any quantitative observations are limited to relative comparisons between simulations of the same system . Furthermore , in choosing to look at a highly restrained system where the relative positions and orientations are restrained , we are also looking at the change in free energy along a narrow slice through configuration space . Although this path may be tightly defined , it is only by using such a highly restrained system that we are able to identify some previously unobserved behaviour , specifically the effect of protein-lipid-protein interactions on the free energy of protein dimerization . Such effects would usually be lost when averaging over a larger range of configurations . The results presented here highlight some of the effects that contribute to the free energy of association for a bacterial outer membrane protein that undergoes non-specific oligomerization in a POPE bilayer . These processes will play a role in many protein-protein interactions , even those with some specific oligomerization modes , although in the latter case the non-specific interactions will likely be masked at close range by the specific interactions . We would expect the protein-lipid-protein interactions to be present in many membrane protein systems , as they seem to be determined by the underlying lipid-protein interactions . The PMF for the association of two OmpF trimers calculated by Casuso et al . [6] had a potential well that was approximately twice as deep as the ones we present here for NanC . However , the OmpF protein is much larger than NanC and the oligomerized proteins would therefore have a correspondingly larger buried surface compared to our NanC system . Given the conclusions of this study , it will be of great interest to apply similar methods to those presented to calculate orientationally-dependent PMFs for a variety of other membrane proteins . Information obtained from PMFs , such as the orientational dependence of the free energy of association , are necessary for parameterizing yet coarser ( i . e . more approximate ) models ( for example those of Yiannourakou et al . [33] ) , in order to enable simulation studies of the emergent properties of large , crowded and complex membrane models [47] .
Umbrella sampling was used to obtain the PMF for each of the four orientational configurations while varying the inter-protein separation [32] . The umbrella sampling was performed using simulation windows in which one protein was restrained at relative positions with the desired inter-protein separation . We chose this measure as our reaction coordinate because it is a natural choice for characterizing the separation of two proteins and it would also enable the PMFs to be used to parameterize larger scale models , as done by Yiannourakou et al . [33] . To calculate the PMF from these individual biased simulation windows we employed the weighed histogram analysis method ( WHAM ) [48] . For the WHAM method to produce a converged PMF , we need to ensure that all points along the reaction coordinate are sufficiently sampled . This means that the histograms from the sampling of the reaction coordinate in each simulation window need to overlap with adjacent simulation windows and that these histograms are smooth . The effect of enforcing these requirements is that all points along the reaction coordinate are thoroughly sampled in multiple simulation windows . It is these considerations that determined our positioning of the umbrella sampling windows and the strength of the biasing potentials used in each . For each orientational configuration , umbrella sampling windows were distributed at positions with varying inter-protein separation along a line connecting the centres of mass of the two proteins . The umbrella potential was applied to the centre of mass of each protein's particles , restraining them at relative positions with the desired centre of mass separation . The sampling windows were distributed at 0 . 1 nm intervals from an inter-protein separation of 2 . 8 nm to 8 nm . In each of these simulation windows we applied a harmonic umbrella potential with a force constant of . To improve the overlap of the histograms from adjacent simulation windows in the region of the local minima , which improved the resolution , additional simulation windows were used . These additional simulation windows were distributed at 0 . 05 nm intervals from a inter-protein separation of 2 . 8 nm to 4 . 5 nm , where the proteins were in close proximity . These closely separated windows had a stronger harmonic force constant of , in order that we could better resolve the barriers surrounding the restrained local minima . The lipidation/delipidation of the protein-protein complex at close range is a slow process . It is important that we adequately sample both the lipidated and delipidated state . To do so , we also performed simulations in which that interface was manually delipidated , with the intervening lipids returned to the bulk of the bilayer , and the system re-equilibrated . Manually delipidated simulations were carried out for the orientational configurations , but were not required for as no persistent lipidated state was observed . Delipidated simulation windows were distributed from an inter-protein separation of 2 . 8 nm to 3 . 7 nm for , and to 4 . 0 nm for , separated by 0 . 5 nm in all cases and using the stronger position restraint of . To analyse the free energy of association for specific relative orientations of NanC we had to ensure that we restrained their orientations in each simulation window , as well as their relative positions . This was achieved by applying a rotational potential to the particles of each protein . The rotational potential for each protein acted around a vector in the direction ( approximately perpendicular to the plane of the membrane ) through the protein's centre of mass . By applying a suitable rotational potential to the proteins , we were able to restrain their rotation without influencing the positional umbrella potential . Kutzner et al . [49] showed that we can virtually eliminate both the radial forces and forces parallel to the axis of rotation by using a restraining potential of the form ( 1 ) where is a unit vector parallel to the rotation axis; and are the current and reference positions of the particle , respectively; and are the current and reference positions of the centre of mass of the particles in each protein , respectively; is a rotation matrix , which describes the motion of the potential; is the force constant for the rotational potential , and is a small constant required to avoid a singularity at the axis of rotation . The application of this potential to the particles of each protein results in a purely rotational force ( a torque ) about the proteins' centres of mass that acts in an approximately perpendicular direction to the plane of the membrane . This rotational potential is implemented by the enforced rotation feature of the GROMACS MD simulation package , which at each time-step applies an appropriate translational force to each of the restrained particles in order to create the desired torque [49] , [50] . The proteins were inserted into the membrane in the desired relative orientations for each configuration and so we wished to restrain their orientations , which was achieved by setting the rotation matrix , , equal to the identity matrix . The reference positions for the particles were taken from the initial protein structures at maximum separation . The force constant was set to , which kept rotational drift below , and the constant was set to . Kutzner et al . showed that values for greater than were shown to give good results for rotating the subdomain of ATPase using the same value of that we have used here [49] . To calculate the PMF of association for two NanC proteins we employed coarse-grained molecular dynamics simulations of a bilayer system . The system consisted of the two coarse-grained proteins embedded in a symmetric bilayer formed from 424 coarse-grained POPE lipid molecules . The bilayer was solvated and counter ions were added to neutralize the system . The coarse-grained force field we implemented was a modified version of the MARTINI forcefield [29] , [41] , in which approximately four heavy atoms were mapped to each coarse-grained particle . This mapping can be seen between Figure 1A and Figure 1B . All simulations were run using GROMACS v4 . 6 ScalaLife 2012 ( available from http://www . scalalife . eu ) [50] . The simulations were performed under conditions of constant temperature ( 310 K ) and pressure ( 1 bar ) using a timestep of 40 fs . We have provided a GROMACS simulation configuration ( mdp ) file in the Supporting Information ( Data S1 ) for a simulation window in which the protein restrained at relative positions corresponding to a centre of mass separation of 4 nm with a force constant of . Each of the simulation windows were equilibrated for between and . The production simulations consisted of at least of simulation for each of the 0 . 1 nm separated simulation windows , where the applied umbrella potential force constant was . The length of simulation was increased if the PMF had not converged sufficiently . The convergence of each PMF was evaluated by comparing the PMFs obtained using non-intersecting subsets of production simulation data ( see Figure S3 ) . For each of the 0 . 05 nm separated simulation windows , where the larger force constant of was applied , of production simulations were performed . We also performed of production simulation for the manually delipidated simulation windows , which were separated by 0 . 05 nm . To combine the simulation data to obtain the PMFs we used the g_wham program , distributed with GROMACS [51] , using a tolerance of . In an attempt to predict the location of the restrained metastable states we performed a simulation of a single NanC protein freely diffusing in a POPE bilayer . This extended simulation consisted of a single coarse-grained NanC protein model ( the same model used for the PMF calculations ) embedded in a 25 nm square membrane constructed from coarse-grained POPE molecules . To analyse the lipid distribution around the single protein , we rotated each frame of the trajectory so that the NanC protein was aligned with its position at the start of the simulation . From this aligned trajectory we were able to calculate the position of the lipid particles in relation to the protein for the entire simulation . The particle density was calculated for a 6 nm square region around the NanC protein for each of the particles in the coarse-grained lipid molecules . To calculate the protein density in a given direction , we calculated a linear projection of this two-dimensional density . For the case of the protein orientation configuration , the direction we are interested in is the same for both of the proteins , as they have the same face oriented toward the other protein . This direction is marked by the dashed lines in Figure 3D . In order to characterize the lipid particle density in this direction , we projected the two-dimensional density onto a series of 4 nm lines emanating from the protein's centre of mass , at regular angular intervals , within the region marked by the dashed lines . The dashed lines represent an angular window of and the individual projection lines were separated by . To predict the location of the minimum and the local minima of Figure 2B , assuming the lipid behaviour corresponds to that shown in Figure 3D , we aligned the peaks of the mean lipid species plot ( where the mean was taken across the linear projections for all coarse-grained lipid particles in both leaflets and is shown in Figure 3E ) with those of the same plot overlaid with the x-axis reversed . For the case of a single intervening lipid we aligned the first peak with the first peak of the reversed plot ( see Figure 3F ) . For the case of two intervening lipids we aligned the first peak with the second peak of the overlaid plot ( see Figure 3G ) . Finally for the case of three intervening lipids , we aligned the first peak with the overlaid third peak and the second peak with the overlaid second peak ( see Figure 3H ) . To obtain the buried surface area of the proteins at various positions along the reaction coordinate , we analysed the surface area of the simulation windows with the higher translational restraining potential , , to enable the analysis of the surface area on a finer scale , using window separations of 0 . 05 nm instead of 0 . 1 nm . Using the higher force constant also ensured that the surface area was measured for a conformation that was sampled closer to the centre of the window; with the weaker force constant we would be measuring the surface area for conformations with separations that could differ significantly from the position of the window centre . All of the surface area calculations were carried out using the g_sas tool in GROMACS using of production simulation trajectory . | Cells are surrounded by selectively-permeable bilayer membranes , enabling the cell to control its internal environment . Embedded within these membranes are a variety of membrane proteins , many of which facilitate this environmental control and are integral to numerous metabolic processes . Their location within the membrane and their mutual association are controlled by many factors . We use molecular dynamics simulations to investigate the free energy of association for a pair of relatively simple membrane proteins . By doing so , we are able to characterize the effect that the geometrical properties of the protein have on their mutual association in a bilayer environment , showing that there is a correlation between the buried surface area of two proteins when in contact and the strength of their interaction . We also observe the effect of protein-lipid-protein interactions in this free energy characterization . Such interactions are related to the preferential distribution of lipids around proteins in the membrane . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"chemical",
"properties",
"physics",
"biochemistry",
"lipids",
"computational",
"chemistry",
"transmembrane",
"proteins",
"protein",
"interactions",
"molecular",
"dynamics",
"biochemistry",
"simulations",
"proteins",
"biophysic",
"al",
"simulations",
"dimerization",
"statistical",
"mechanics",
"chemistry",
"transmembrane",
"transport",
"proteins",
"biology",
"computational",
"biology",
"physical",
"chemistry"
] | 2014 | The Free Energy Landscape of Dimerization of a Membrane Protein, NanC |
Thioester-containing protein 1 ( TEP1 ) is a central component in the innate immune response of Anopheles gambiae to Plasmodium infection . Two classes of TEP1 alleles , TEP1*S and TEP1*R , are found in both laboratory strains and wild isolates , related by a greater or lesser susceptibility , respectively to both P . berghei and P . falciparum infection . We report the crystal structure of the full-length TEP1*S1 allele which , while similar to the previously determined structure of full-length TEP1*R1 , displays flexibility in the N-terminal fragment comprising domains MG1-MG6 . Amino acid differences between TEP1*R1 and TEP1*S1 are localized to the TED-MG8 domain interface that protects the thioester bond from hydrolysis and structural changes are apparent at this interface . As a consequence cleaved TEP1*S1 ( TEP1*S1cut ) is significantly more susceptible to hydrolysis of its intramolecular thioester bond than TEP1*R1cut . TEP1*S1cut is stabilized in solution by the heterodimeric LRIM1/APL1C complex , which preserves the thioester bond within TEP1*S1cut . These results suggest a mechanism by which selective pressure on the TEP1 gene results in functional variation that may influence the vector competence of A . gambiae towards Plasmodium infection .
Thioester-containing proteins ( TEPs ) are a major component of the innate immune response of insects to invasion by bacteria and protozoa [1] , [2] . Anopheles gambiae thioester-containing protein 1 ( TEP1 ) is a complement-like protein that plays a central role in the opsonization of gram-negative bacteria in the hemolymph [3] . TEP1 also binds to the surface of Plasmodium ookinetes that traverse the midgut epithelium following ingestion of an infectious blood meal , targeting those ookinetes for lysis and , in certain mosquito strains , melanization [4] . TEP1 activity has been demonstrated against both P . berghei [4] and P . falciparum [5] , [6] . Deciphering the molecular basis of the TEP1-mediated immune response is relevant to understanding the determinants of vector competence and a potential source of novel vector-based malaria control strategies . The crystal structure of full-length TEP1 revealed significant structural homology to complement factor C3 [7] . TEP1 is composed of a series of eight macroglobulin ( MG ) domains , the β-sheet CUB domain and α-helical thioester domain ( TED ) . The TED contains an intramolecular β-cysteinyl-γ-glutamyl thioester bond that is protected from inadvertent hydrolysis by sequestration within a protein-protein interface formed by the TED and MG8 domains . Based upon the known mechanism of complement factors [8] , activation of the thioester in TEP1 is presumed to involve a large conformational change causing dissociation of the TED-MG8 interface in the direct proximity of a pathogen , whereupon the thioester may react with nucleophilic groups on , and covalently attach TEP1 to , the surface of the pathogen . TEP1 lacks two additional domains that are present in complement factors , the anaphylatoxin ( ANA ) and C345C domains . The ANA domain in particular plays a key role in the activation of complement factor C3 , which is cleaved intracellularly in a protease-sensitive region between the MG6 and ANA domains prior to secretion . The ANA domain contacts both the MG3 and MG8 domains in the structure of mature , circulating C3 [9] . Activation of C3 occurs after regulated proteolysis immediately following the ANA domain whereby the anaphylatoxin C3a is released . Dissociation of C3a destabilizes the remaining C3b fragment and leads to a large-scale conformational change and rapid activation of the thioester bond [10] , [11] . In contrast , TEP1 is secreted as a full-length protein into the mosquito hemolymph where it is cleaved by as yet unknown protease ( s ) . Cleavage of TEP1 , producing TEP1cut , does not instantly lead to activation of the thioester [12] , suggesting that full-length TEP1 is a pro-form [13] that must undergo conversion to an active species following cleavage within the protease-sensitive region . TEP1cut is meta-stable in solution and precipitates over time . This precipitation is concomitant with hydrolysis of the thioester bond and is prevented in vivo by formation of a ternary complex between TEP1cut and a heterodimer of two leucine-rich repeat proteins , LRIM1 and APL1C [12] , [14]–[16] . The ternary complex TEP1cut/LRIM1/APL1C was formed in vitro only after chemical inactivation of the thioester bond of TEP1cut by treatment with methylamine ( MeNH2 ) [15] . This raised the question as to whether LRIM1/APL1C stabilizes a conformation of TEP1cut that contains an active thioester , or a distinct conformation in which the thioester has either reacted with substrate or been hydrolyzed by water . The TEP1 gene is highly polymorphic , with distinct alleles conferring variable levels of protection from pathogens . Two alleles were originally identified in laboratory mosquito strains ( indicated in brackets ) as being susceptible ( G3 ) and refractory ( L3–5 ) to infection with P . berghei [4] . Recently , additional alleles were identified from laboratory strains conforming to two major classes S and R: TEP1*S1 ( PEST ) , TEP1*S2 ( 4Arr ) , TEP1*S3 ( G3 ) , TEP1*R1 ( L3–5 ) and TEP1*R2 ( 4Arr ) with TEP1*S2 and TEP1*R2 alleles displaying intermediate phenotypes with respect to P . berghei infection [17] . The refractory allele , TEP1*R1 , has been expressed in vitro and utilized in structural and functional studies [7] , [12] , [15] . The TEP1*S1 and TEP1*R1 proteins share 93% sequence identity with the majority of amino acid differences being confined to three hypervariable loops within the TED domain [3] , [4] . Two of these loops , the pre-α4 loop and the catalytic loop , are situated in close proximity to the thioester itself at the TED-MG8 domain interface [7] and are complemented by amino acid differences within the MG8 domain that interact with the pre-α4 and catalytic loops and also conform to the TEP1*S/R division of alleles . A recent study of wild mosquito populations from five locations in West , Central , and East Africa detected three similar sets of TEP1*S/R alleles as observed in the laboratory strains; s ( TEP1*S ) , rA ( TEP1*R2 ) and rB ( TEP1*R1 ) [6] . Furthermore , specific geographical variation in allelic frequencies and a statistically significant decrease in P . falciparum oocysts within s/rB heterozygous vs . s/s homozygous mosquitoes were observed . The concordance of laboratory and field studies prompted us to further investigate the structure and properties of TEP1*S1 in comparison to TEP1*R1 . Here we report the crystal structure of full-length TEP1*S1 . We also report the relative reactivity of the thioester bond in TEP1*S1 and TEP1*R1 to hydrolysis and the association of TEP1*S1cut with LRIM1/APL1C . These results suggest a potential mechanism by which allelic variation in TEP1 , particularly in the pre-α4 and catalytic loops , may translate to functional variation towards distinct pathogens .
Full-length TEP1*S1 crystallized in space group P43 and the structure was determined at 3 . 7 Å resolution ( PDB 4D93 ) , ( see Materials and Methods , Table S1 and Figure S1 ) . To facilitate comparison with recent studies of TEP1 alleles the structure of TEP1*S1 is numbered according to the complete protein sequence . The previously determined structure of TEP1*R1 [7] has been re-refined ( PDB 4D94 ) to correct some errors in the original model and was used as a reference structure for refinement of TEP1*S1 . Residues in the new TEP1*R1 and TEP1*S1 models are numbered according to the complete peptide chain ( including signal peptide ) for comparison with other reports . The refined model of TEP1*S1 has three molecules , two of which ( chains A , C ) comprise residues 22–1338 with the exception of four gaps; residues 561–562 , 575–582 in the linker ( LNK ) domain , 606–628 within the protease-sensitive region , and 822–829 in the CUB domain . A third molecule ( chain B ) with higher B-factors has poor or absent density for much of domains MG1 , MG2 , MG4 , MG5 and MG6 , but is otherwise complete for residues 629–1338 with the exception of 822–829 in the CUB domain . No significant difference in conformation is apparent between the three molecules . Further description of the structure is based upon molecule A . The overall structure of TEP1*S1 is very similar to that of TEP1*R1 ( Figure 1A ) . The first six MG domains form a super-helical quaternary structure , with MG6 split by the insertion of the linker and protease-sensitive regions ( 585–607 ) . Thus the TEP1cut N-terminal fragment ( β chain ) and C-terminal fragment ( α chain ) are interleaved within the MG6 domain . Following the MG6 domain two additional domains , MG7 and MG8 , are divided by the nested insertions of the CUB and TED domains . SDS-PAGE of redissolved crystals confirms TEP1*S1 within the crystal to be full-length protein ( Figure 1B ) . Some protein domain motion is evident between TEP1*S1 and TEP1*R1 ( Figure 1C ) , confirmed by analysis using the program DYNDOM [18] ( Table S2 ) . The TEP1*S1 MG3 , MG7 , CUB , TED , and MG8 domains are superimposable as a rigid body upon TEP1*R1 . The MG1 , MG2 , MG5 and MG6 domains also form a rigid body but are rotated 11° relative to TEP1*R1 . One hinge for this movement is the MG2–3 linker ( 217–222 ) a short sequence identically conserved with human complement factor C3 . The second hinge is the MG4 domain itself which is rotated 26° relative to the two other rigid domains . As there is no sequence variation between TEP1*S1 and TEP1*R1 at the interface of the MG2 , MG6 and TED domains or the MG3–MG4 domain interface , these rearrangements likely reflect inherent flexibility and different packing constraints within the TEP1*S1 crystal vs . TEP1*R1 . Amino acid variation between TEP1*R1 and TEP1*S1 was previously noted to be largely confined to domains surrounding the TED [4] , [7] . Analysis of alleles TEP1*R1–2 and TEP1*S1–3 [17] confirms and extends this observation . No amino acid substitutions that separate TEP1*R and TEP1*S alleles occur within domains MG1–MG6 , and except for five substitutions to similar residues , all variation between TEP1*S and TEP1*R alleles are confined to the TED , CUB and MG8 domains . These polymorphisms are hereafter described as mutations to the TEP1*R1 allele , i . e . R{res#}S . We focused on amino acid differences between TEP1*S1 and TEP1*R1 that are preserved in all S and R laboratory alleles [17] and wild mosquito populations [6] . Of 42 such polymorphisms within the TED ( Table S3 ) , 18 occur within three previously identified hypervariable loops termed the pre-α4 loop ( 914–920 ) , the catalytic loop ( 966–974 ) and the β-hairpin ( 1054–1069 ) . An additional polymorphism S1108R was noted from the crystal structure of TEP1*R1 as potentially significant [7] . The remaining 23 polymorphisms are generally localized in short loops between the TED α-helices and introduce no significant alteration to the structure ( Table S3 ) , with the possible exception of four ( F960S , E1005V , K1009V , T1012N ) on the face of helix α7 and adjacent to the post-α5 turn that form a crystal contact within the TEP1*R1 structure . The pre-α4 and catalytic loops form part of the TED-MG8 domain interface ( Figure 2A ) that protects the thioester from premature activation or hydrolysis . Both loops are ordered in the TEP1*S1 structure . The TEP1*R1 catalytic loop contains a sequence of five residues including Lys 966 and Glu 970 ( 966KAGAE970 ) . These charged residues also occur in the TEP1*S1 catalytic loop but their positions are switched ( 966ETGKV970 ) . TEP1*S1 Glu 966 adopts a different conformation than Lys 966 in TEP1*R1 ( Figure 2B ) , directed into a pocket occupied by a Cl− ion in TEP1*R1 , within hydrogen bonding distance of Ser 921 Oγ and Phe 923 N ( the same conformation observed for Glu 1098 in complement factor C3 ) . In contrast Ser 921 is within hydrogen bonding distance of TEP1*R1 Tyr 971 but not TEP1*S1 Trp 971 . These differences impart a ∼1 . 8 Å displacement of residues 966–968 in TEP1*S1 relative to TEP1*R1 . In the pre-α4 loop the substitution N919G permits a different backbone conformation for TEP1*S1 with Gly 919 O within hydrogen bonding distance of Val 914 N ( Figure 2C ) . The L914V and S1108R substitutions were previously noted as potentially affecting the environment of the thioester [7] . The displacement of the catalytic loop in TEP1*S1 leads L914V to introduce a small cavity within the TED-MG8 interface between Trp 971 and the thioester bond . The S1108R substitution does not cause any perturbation in the interface however , the Arg side chain adopts a conformation within hydrogen bonding distance of the carbonyl oxygen of Tyr 1307 instead of a water molecule as seen in TEP1*R1 . Substitutions in the TED domain at the TED-MG8 interface are complemented by substitutions within the MG8 domain ( Table S4 ) . Three pairs of substitutions noted in the TEP1*R1 structure [7] are preserved between the TEP1*S and TEP1*R alleles , two of which produce significant differences in the TEP1*S1 structure ( Figure 2D ) . The K1260N substitution preserves the hydrogen bonding distance to Gly 858 N in the thioester motif but not to Tyr 884 . The N1275Y substitution is no longer compatible with hydrogen bonding to Trp 915 in the pre-α4 loop , and the conformation adopted by TEP1*S1 Tyr 1275 forms is neither favorable for alternative hydrogen bond formation nor π-stacking interactions with nearby aromatic residues . Though the substitution N1276K appears to introduce a repulsive electrostatic interaction with Lys 970 in the TEP1*S1 catalytic loop we note that ( i ) the density for this side chain is poor , ( ii ) the nearby substitutions R1227S/R1228Q compensate for the introduction of this charge and ( iii ) Asn is conserved at this position in TEP1*S2–S3 [17] ( the corresponding region was not sequenced for wild alleles reported by White et al . [6] ) . An additional 11 polymorphisms conserved between S and R alleles occur in the MG8 domain but introduce no discernible alterations ( Table S4 ) . Amino acid variation within the CUB domain is localized to peripheral residues , none of the S/R-conserved polymorphisms are observed in the central β-strands β5 , β6 , β7 or β10 ( Table S5 ) . Two pairs of substitutions , V797A/R800K and V1183I/N1187D , are located on adjacent strands linking the CUB domain to the MG7 and MG8 domains , respectively , with the substitution T831K adjacent at the end of the β4–β5 turn . This is a site of large structural changes in the conversion of complement factor C3 to C3b [10] , [11] , and the site of C3b cleavage by factor I [19] . To assess the functional role of TEP1 polymorphisms in vitro , we sought to determine the relative stability of the TEP1*S1 compared to TEP1*R1 . In addition to wild-type alleles we generated the following TEP1 variants: ( i ) TEP1*R1 with thioester cysteine mutation C859A , ( ii ) TEP1*R1-sTED2 , in which residues 878–1108 in the TED and 1227–28 , 1260–61 and 1275–76 in MG8 were replaced with TEP1*S1 , and ( iii ) TEP1*R1 with MG3 glycosylation mutant N312D ( Figure 3A ) . We previously observed that limited proteolysis of TEP1 in the protease-sensitive region leads to slow hydrolysis of the thioester bond and , in the absence of the LRIM1/APL1C complex , hydrolysis of the thioester leads to precipitation [15] . We purified TEP1*R1-C859A and observed that , while it was a stable full-length protein , the protein precipitated rapidly following proteolysis ( Figure S2 ) , suggesting that thioester hydrolysis is the rate-limiting step in the precipitation of TEP1cut . We therefore measured the rate of precipitation of TEP1*R1cut and TEP1*S1cut to determine the rate of thioester hydrolysis in TEP1cut at 20°C , the same temperature used for in vivo studies of P . berghei infection . The half-life of TEP1*S1cut is 8 . 5 h ( Figure 3B ) , significantly shorter than the half-life of TEP1*R1cut ( 6 . 5 days ) , suggesting that TEP1*S1cut is more susceptible to hydrolysis of the thioester bond than TEP1*R1cut ( Figure 3C ) . The soluble fractions of the TEP1cut proteins analyzed with silver-stained SDS-PAGE also reflects the shorter half-life of TEP1*S1cut ( Figure S3 ) . The half-life of TEP1*R1-sTED2cut is 12 h ( Figure 3B ) , confirming that the increased reactivity of TEP1*S1 towards hydrolysis is largely due to variation within the TED domain and the TED-MG8 interface in particular . The glycosylation site Asn 312 was previously noted to form a significant fraction of the interface between the MG3 and MG8 domains [7] . The half-life of the glycosylation mutant TEP1*R1cut-N312D is 8 days however ( Figure 3C ) , indicating that removal of this glycosylation site does not affect the stability of the thioester in TEP1*R1cut . Precipitation of TEP1cut is an indirect effect of thioester hydrolysis and may not correlate quantitatively with the rate of reaction of the thioester . We therefore measured the fraction of TEP1*S1cut containing an intact thioester by treatment of samples with MeNH2 as a function of time . Methylated and hydrolyzed TEP1cut were simultaneously quantified by monitoring the modification of Gln 862 with quantitative mass spectrometry ( see Materials and Methods ) . The fraction of methylated TEP1*S1 decreased with time with an estimated half-life of 9 h ( Figure 3D ) , in close agreement with the rate of precipitation of the protein . This supports the conclusion that hydrolysis of the thioester bond is the rate-limiting step in precipitation of TEP1cut . We previously observed the ternary complex TEP1cut/LRIM1/APL1C was formed only after chemical inactivation of the thioester bond of TEP1*R1cut by MeNH2 [15] , demonstrating that LRIM1/APL1C interacted with a reacted form of TEP1*R1cut without an intact thioester . To test whether this was also the case for TEP1*S1 , we prepared TEP1*S1cut and incubated for 36 h at 20°C in the absence or presence of LRIM1/APL1C ( Figure 4A ) . TEP1*S1cut incubated without LRIM1/APL1C precipitated ( Figure 4A , lanes 1–2 ) , while TEP1*S1cut mixed with LRIM1/APL1C remains soluble ( Figure 4A , lanes 5–6 ) . The presence of an intact thioester bond in thioester-containing proteins can be determined by heating under denaturing conditions in the absence of reducing agent , promoting autolytic cleavage of the peptide chain at the site of the thioester bond [20] . The thioester bond is hydrolyzed in precipitated TEP1*S1cut ( Figure 4B , lane 1 ) . In contrast , soluble TEP1*S1cut in complex with LRIM1/APL1C possesses an intact thioester , as shown by heat-induced fragmentation of the C-terminal fragment ( Figure 4B , lane 6 ) . We conclude that the conformational changes in TEP1 required for the binding of LRIM1/APL1C is distinct from that involving reaction of the thioester . Hence the complex between LRIM1/APL1C and TEP1*S1cut is a distinct species from the complex of LRIM1/APL1C and TEP1*R1cut ( MeNH2 ) [15] . The preceding experiments suggest that formation of the ternary complex between TEP1*S1cut and LRIM1/APL1C is due to a conformational change with a similar half-life as the measured rate of thioester hydrolysis . Thus previous attempts to produce a ternary complex between TEP1*R1cut and LRIM1/APL1C were unsuccessful simply because the period of incubation was too short . Accordingly FLAG immunoprecipitation assays were performed with 6×His-tagged TEP1cut proteins and FLAG-tagged LRIM1/APL1C . TEP1*S1cut co-immunoprecipitated with LRIM1/APL1C within 24 h ( ∼3× t1/2 for thioester hydrolysis ) whereas TEP1*R1cut remained in the supernatant after 48 h ( Figure 4C ) . However , after incubation at 20°C for 24 days ( ∼4× t1/2 for thioester hydrolysis ) TEP1*R1cut remained soluble and was co-immunoprecipitated with LRIM1/APL1C ( Figure 4D ) . Thus the conformational change following limited proteolysis in vitro that allows TEP1*S1cut and TEP1*R1cut to bind LRIM1/APL1C is comparable to their respective rates of thioester hydrolysis and precipitation in the absence of LRIM1/APL1C .
As a central component of humoral immunity in A . gambiae , the TEP1 gene is under selective pressure . Significant variation within two major allelic forms , TEP1*S and TEP1*R , are found in both laboratory and wild mosquito populations . Comparison of the structures of TEP1*S1 and TEP1*R1 reveals the consequences of this variation on the pro-form of TEP1 and stabilization of the intramolecular thioester bond . We observe distinct side chain and backbone conformations of two hypervariable loops within the thioester domain and two complementary substitutions within the MG8 domain that directly influence the TED-MG8 interface and the surrounding environment of the thioester bond . An important caveat in analysis of the present structures is that the role of specific polymorphisms may be relevant to another conformation of TEP1 than is represented in the full-length protein . At present three soluble forms of TEP1cut have been identified in vitro ( [12] , [15] and this study ) . The first form contains an intact thioester and does not bind LRIM1/APL1C , ( e . g . TEP1*R1cut 0–48 h post-cleavage ) . The second form contains a thioester but requires LRIM1/APL1C for stability in solution ( e . g . TEP1*S1cut 24–36 h post-cleavage ) . The third form does not contain a thioester and also requires binding of LRIM1/APL1C for stability in solution ( e . g . TEP1*R1cut ( MeNH2 ) 12 h post-cleavage ) . Distinct phenotypes for TEP1*S and TEP1*R alleles are observed for the response to both P . berghei [4] and to P . falciparum [6] . Our results provide the first evidence for a distinct chemical property of TEP1*S and TEP1*R proteins; the rate of thioester hydrolysis and precipitation in the absence of LRIM1/APL1C . Furthermore this difference affects the relative amount of the three in vitro soluble TEP1cut forms arising from cleavage in the protease-sensitive region . Within 24–36 h post-cleavage at 20°C the major soluble form of TEP1*S1cut has an intact thioester and binds LRIM1/APL1C , whereas TEP1*R1cut has an intact thioester but does not bind LRIM1/APL1C . In the absence of LRIM1/APL1C ∼90% of TEP1*S1cut has undergone hydrolysis of the thioester and precipitated from solution within 24 h at 20°C , whereas ∼90% TEP1*R1cut has an intact thioester bond and is soluble [12] . Hence , our results suggest that phenotypic variation in TEP1 alleles can result not only by activity in a single pathway but by distinct mechanisms arising from different forms present in the hemolymph . Our in vitro studies may directly pertain to in vivo studies of P . berghei infection that are also conducted at ∼20°C [21] with microscopic analysis of TEP1 binding at 24–48 hours post-infection [4] , [12] , [22] . Our results are consistent with a model for activation of TEP1*S as proposed by Fraiture et al . ( 2009 ) [12] ( Figure 5 ) . Full-length TEP1*S represents a pro-form . Cleavage within the protease-sensitive region produces a meta-stable species similar to the pro-form that does not interact with LRIM1/APL1C . A slow ( 8 h ) spontaneous conformational change generates a mature form of TEP1*S and exposes a cryptic binding site for LRIM1/APL1C . In the absence of LRIM1/APL1C however , the thioester bond in the mature form is susceptible to hydrolysis , presumably coupled to a large conformational change , producing a reacted form that rapidly aggregates and precipitates from solution . This model is consistent with the roles of TEP1*S3 , LRIM1 and APL1C in the immune response of A . gambiae G3 to P . berghei ookinetes [4] , [12] , including the concept of basal immunity [22] , as spontaneous formation of the active immune complex TEP1*S3cut/LRIM1/APL1C at 20°C is slow relative to the residence of ookinetes beneath the basal lamina . TEP1*R1cut also forms a complex with LRIM1/APL1C that presumably contains an active thioester by a spontaneous conformational change with a half-life of 6 . 5 days at 20°C ( Figure 5 ) . We previously observed that TEP1*R1cut retained an active thioester after 60 h incubation with LRIM1/APL1C and a small component of a high molecular weight complex with LRIM1/APL1C [15] . Originally interpreted as hydrolysis of TEP1*R1cut , this may now be considered to be slow formation of the same complex formed by TEP1*S1cut . Such a slow rate of complex formation however , cannot account for the LRIM1/APL1C-dependent activity of TEP*R1 against P . berghei ookinetes [14] that traverse the midgut epithelium 18–24 h post blood-meal . Hence additional factor ( s ) must exist that accelerate the conformational change of TEP1*R1cut in vivo . In the context of the present model , such factor ( s ) could act in three ways . First , protease ( s ) that cleave TEP1 may comprise or recruit chaperone ( s ) that accelerate the maturation of TEP1 , revealing the binding site for LRIM1/APL1C; TEP1 activation would remain LRIM1/APL1C-dependent . Second , factors could accelerate both maturation and activation of TEP1 directly in an LRIM1/APL1C-independent manner . Third , factors may interact with a complex of reacted TEP1cut and LRIM1/APL1C to activate other TEP1 molecules , i . e . a “TEP1 convertase” as proposed previously to explain the interaction of TEP1*R1cut ( MeNH2 ) with LRIM1/APL1C [15] ( Figure 5 ) . Distinct TEP1*S/R phenotypes are observed for both the LRIM1/APL1C-dependent response to P . berghei [12] , [14] and the response to P . falciparum that is LRIM1/APL1C-independent in Yaoundé and Ngousso strains [23] , [24] that carry TEP1*S alleles [6] . This suggests a functional role of TEP1*S/R polymorphisms in the active form of TEP1 , i . e . direct interaction of the pre-α4 and catalytic loops with the thioester and pathogen surfaces at the point of covalent attachment . The selective pressure that has given rise to these polymorphisms is not only ( even unlikely ) Plasmodium , but environmental pathogens such as bacteria encountered in both adult and pre-adult stages . Our results suggest a possible trade-off between selection for reactivity of the thioester upon activation and steady-state stability of the thioester in circulating TEP1cut . This may be relevant to immune responses based upon basal immunity , as is indicated in the case of Plasmodium [22] , compared to responses based upon infection-induced upregulation of TEP1 expression . Many outstanding questions remain regarding the mechanism of TEP1-mediated immune responses . The structure of the TEP1cut/LRIM1/APL1C ternary complex and the interaction of LRIM1/APL1C with reacted TEP1*R1cut [15] , the source of phenotypic differences between different TEP1*S and TEP1*R alleles , and the role of polymorphism in the TED β-hairpin remains unknown . The interaction of LRIM1/APL1C with MeNH2-treated TEP1*R1cut [15] and with distinct TEP proteins TEP3 and TEP4 [16] suggest additional roles for LRIM1/APL1C in TEP1-mediated immunity besides stabilization of a re-circulating active immune complex . Further structural and functional studies of TEP1 , LRIM1/APL1C and the identification of additional factors are required to address these questions .
TEP1*S1 was generated by total gene synthesis ( Genscript ) and subcloned into pFastbac1 with a C-terminal 6×His tag . TEP1*R1-sTED2 was constructed as follows using QuickChange site-directed mutagenesis ( Stratagene ) . An SphI site was inserted into TEP1*R1-pFastbac1 corresponding to TEP1*S1 H878Y and removed from the pFastbac1 MCS . Digestion of both vectors with SphI/AfeI allowed replacement of TEP1*R1 residues 878–1108 with the corresponding sequence from TEP1*S1 . Finally ( i ) 1227–8 , 1260–1 and 1275–6 in the MG8 domain ( TED-MG8 interface ) were mutated to the corresponding residues in TEP1*S1 , and ( ii ) residues 960 , 1005 , 1009 , 1012 in the TED were mutated back to the corresponding residues of TEP1*R1 . All TEP1 and LRIM1/APL1C constructs were expressed using the baculovirus expression system . Purification , limited proteolysis , thioester autolytic cleavage assay and immunoprecipitation experiments were performed as previously described [7] , [12] , [15] . Following limited proteolysis and re-purification TEP1 samples were concentrated to an OD280 of 0 . 5–1 . 0 and stored at 20°C . To measure rate of precipitation as a result of thioester hydrolysis upon proteolysis , samples and matching blank ( filtrate from concentration ) were centrifuged at 17 , 000×g , 20°C for 10 min and A280-A330 recorded in a standard UV spectrophotometer ( Shimadzu UV1800 ) . Separate time points are all derived from the same protein batch and purification and qualitatively similar results derived from 2–3 independent biological replicates . Half-lives were calculated from samples with a decay to <25% initial value and fit to log-linear plot with R2>0 . 99 ( except TEP1*R1-N312D , final value 30% initial , R2 = 0 . 95 ) . To determine the rate of thioester hydrolysis by quantitative mass spectrometry , TEP1*S1 was cleaved as before and MeNH2 was added at specific time points to react with intact thioester bonds , methylating Gln 862 . Samples were TCA precipitated and redissolved in 0 . 4 M NH4HCO3 containing 8 M urea followed by reduction and alkylation with DTT and iodoacetamide , respectively . Trypsin digestion was performed for 16 h at 37°C at a 10-fold molar excess of protein to trypsin . TFA and acetonitrile was added to final concentrations of 0 . 5% and 5% , respectively , followed by purification with C18 spin columns ( Pierce ) and elution in 80% acetonitrile . Tryptic peptides corresponding to hydrolysis ( deaminated , [Dea] = +1 m/z ) or methylation ( [Me] = +14 m/z ) of TEP1*S1 Gln 862 were characterized by time-of-flight LC-MS . Three specific fragments selected for quantitative analysis on an AB SCIEX 5500 Q-TRAP instrument coupled to an online Waters nanoACQUITY Ultra High Pressure Liquid Chromatography system and analysis with Multiquant 2 . 0 software . Assuming the deaminated and methylated peptides have similar ionization efficiency , the fraction of intact thioester is equal to IMe/ ( IMe+IDea ) . Reported data is the average of three fragments , two instrument replicates . TEP1*S1 ( PDB 4D93 ) , TEP1*R1 ( PDB 4D94 ) . | Anopheles mosquitoes transmit malaria , the world's most devastating parasitic disease , of which Anopheles gambiae is the principal vector for malaria in Sub-Saharan Africa . Different populations of mosquitoes vary widely in how readily they become infected with malaria parasites , while some strains do not transmit malaria at all . The mosquitoes' innate immune system is a significant factor that may influence the level of malaria infection; in particular the thioester-containing protein 1 ( TEP1 ) targets malaria parasites for destruction during their initial invasion of the body cavity . The TEP1 gene varies significantly across mosquito populations with two major classes of alleles , TEP1*S and TEP1*R . We report the three-dimensional molecular structure of the TEP1*S1 protein and compare it to the previously determined TEP1*R1 structure . Differences between the structures are localized around the active site and thioester bond , and correlate with a difference in stability of this bond within the two proteins and their interaction with a heterodimer of two other immune genes , LRIM1 and APL1C . These results shed light on the mechanism of mosquitoes' natural immunity to malaria infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"immune",
"system",
"proteins",
"proteins",
"immunity",
"innate",
"immunity",
"protein",
"structure",
"vector",
"biology",
"immunology",
"biology",
"microbiology",
"anopheles",
"host-pathogen",
"interaction"
] | 2012 | Molecular Basis for Genetic Resistance of Anopheles gambiae to Plasmodium: Structural Analysis of TEP1 Susceptible and Resistant Alleles |
Many types of large cells have multiple nuclei . In skeletal muscle fibers , the nuclei are distributed along the cell to maximize their internuclear distances . This myonuclear positioning is crucial for cell function . Although microtubules , microtubule associated proteins , and motors have been implicated , mechanisms responsible for myonuclear positioning remain unclear . We used a combination of rough interacting particle and detailed agent-based modeling to examine computationally the hypothesis that a force balance generated by microtubules positions the muscle nuclei . Rather than assuming the nature and identity of the forces , we simulated various types of forces between the pairs of nuclei and between the nuclei and cell boundary to position the myonuclei according to the laws of mechanics . We started with a large number of potential interacting particle models and computationally screened these models for their ability to fit biological data on nuclear positions in hundreds of Drosophila larval muscle cells . This reverse engineering approach resulted in a small number of feasible models , the one with the best fit suggests that the nuclei repel each other and the cell boundary with forces that decrease with distance . The model makes nontrivial predictions about the increased nuclear density near the cell poles , the zigzag patterns of the nuclear positions in wider cells , and about correlations between the cell width and elongated nuclear shapes , all of which we confirm by image analysis of the biological data . We support the predictions of the interacting particle model with simulations of an agent-based mechanical model . Taken together , our data suggest that microtubules growing from nuclear envelopes push on the neighboring nuclei and the cell boundaries , which is sufficient to establish the nearly-uniform nuclear spreading observed in muscle fibers .
One of the fundamental challenges of cell biology is to define principles of spatial organization of the cell [1] , and , in particular , to unravel the mechanisms that control the position , size , and shape of organelles . The nucleus is the principal organelle and organizational center of eukaryotic cells . In textbooks , it is typically depicted in the middle of the cell; however , the nucleus’ actual position , ( as in the apical/basal position in developing neuroepithelia [2] ) , depends on the cell’s migratory state , cell cycle stage , and differentiation status [3] . Proper nuclear position is vital for many cell functions , including spatially correct cell division and the direction of cell migration [3] . Multinucleation is one mechanism adopted by cells to generate and sustain large cell sizes . Muscle cells are one of the largest cell types , which are formed by fusion of mononucleated myoblasts and contain up to several tens ( invertebrates ) to several hundred ( vertebrates ) nuclei . Myonuclei are typically positioned at the cell’s periphery , and are distributed to maximize internuclear distance . However , in muscles undergoing repair , they are found towards the cell center , and in muscle diseases known as Centronuclear Myopathies , myonuclei are also found to be mispositioned [4 , 5] . It has been argued [6] , that correct positioning of myonuclei is not only an indicator , but also a cause of muscle diseases . A possible mechanism is provided by the Myonuclear Domain Hypothesis [7 , 8] , which suggests that each nucleus caters for a particular domain of the cell by making the gene products locally needed . Mispositioned nuclei would consequently not be able to guarantee the correct supply of products to their cytoplasmic domains , affecting muscle function . In this work , we focus on nuclear positioning mechanisms in multinucleated muscle fibers . Drosophila is a good in vivo model system for investigating muscle development , growth , and homeostasis [9–11] , due to the simplicity of its muscle pattern , the ease of genetic manipulation , and the homology of relevant genes and processes to mammalian muscle . Nuclei in newly fused Drosophila embryonic muscle cells undergo an orchestrated series of movements , best described in lateral transverse muscles: after fusion of the myoblasts , the resulting muscle cell is thought to disassemble its centrosomes and redistribute γ-tubulin around each nuclear envelope . The myonuclei initially form a cluster close to the cell center . The cluster splits into two subclusters that then migrate towards the opposing cell poles . Subsequently both clusters break apart , and the nuclei spread out evenly along the cell long axis [12 , 13] . As the nuclei spread in the muscle cell , sarcomeres , the fundamental contractile units in muscle , form into myofibrils within each cell , and , at the end of embryogenesis , the nuclei become positioned along the long axis of the cell at its periphery , thereby maximizing internuclear distance . During the subsequent larval stages of development , the muscle cells grow 20-40 fold over the course of 5 days without the addition of new myonuclei [14] . Nevertheless , the myonuclei remain appropriately positioned along the cell , although the mechanisms that are responsible for this are not clear . While the actomyosin network may be involved in nuclear positioning [15] , microtubules ( MTs ) , MAPs ( MT Associated Proteins ) , and MT-based motors , such as kinesin and dynein , have been shown to play a major role [12 , 16–18] . As examples , in embryos in which MTs are severed in the muscle cell , the central cluster does not split; in many motor mutants , nuclear spreading in the muscle cell is perturbed [19] . However , the precise mechanisms controlling myonuclear positioning remain poorly understood . Modeling has proven to be very useful in complementing cell biological methods in problems of positioning with , for example , the mitotic spindle [20 , 21] . Mathematical modeling focused on multinucleated cells and nuclear positioning is in its infancy . Simple conceptual models of nuclei repelling each other were used in [22] and [23] to show that such models can explain regular distribution of nuclei in muscle cells and in the Drosophila blastoderm syncytium . Detailed mechanical simulations were done in [24] to understand multiple nuclear movements in multinucleate fungus Ashbya gossypii . Here , we use computational modeling to understand the mechanisms regulating nuclear positioning in Drosophila larva muscle cells . We hypothesize that nuclear positioning is a result of a MT-motor based force balance . Rather than assuming the nature of this force balance , we screened multiple computer-generated forces by comparing the spatial nuclear patterns that they predict to quantitative microscopy data from biological specimens . We then simulated a detailed agent-based model to confirm the predictions of the screen . One model explains all biological data , including many subtle patterns of multi-nuclear positioning . Based on this model we propose that myonuclei are positioned by establishing a force balance via MT-mediated repulsion .
We have focused on data obtained from Ventral Longitudinal ( VL ) muscles VL3 and VL4 of Drosophila 3rd instar larvae ( Fig 1A–1C ) . The 200 analyzed muscle cells ( 103 VL3 and 97 VL4 cells ) contained between 6 and 23 nuclei ( VL3: 15 . 3 ± 2 . 9 , VL4: 9 . 3 ± 1 . 6 ) . Representative confocal microscopy images of fixed samples with fluorescently labeled muscle cells ( actin ) , myonuclei ( lamin , Hoechst ) and microtubules ( alpha-tubulin ) are shown in Fig 1A–1C . We used cells from three control genetic backgrounds ( w1118 , Dmef2-GAL4;UAS-2xEGFP , Dmef2-GAL4;UAS-GFP RNAi ) . VL muscles are flat rectangular cells with nuclei located on one cell side . We employed the following terminology: cell length is the dimension along the long axis of the cell ( y-direction ) , cell width is the dimension along the short axis of the cellular rectangle ( x-direction ) ( Fig 1B ) . We referred to the edge of the z-projection of the cell as the cell boundary , and distinguished between cell sides ( long segments of the rectangular shape ) and cell poles ( short segments of the rectangle ) . We defined the subcellular localization of nuclei and measured several geometric parameters relevant to nuclear positioning , including nuclear numbers , shapes , nearest neighbor distances , and distances to the cell boundary , and used statistical tools to analyze the data . The following features of VL muscle cells and nuclei informed our modeling: 1 ) Both cell types share a similar length ( VL3: 499 . 2 ± 57 . 7μm , VL4: 491 . 8 ± 57 . 1μm ) , but VL4 cells’ width is significantly smaller ( VL3: 80 . 4 ± 17 . 5μm , VL4: 51 . 7 ± 11 . 3μm ) . 2 ) Nuclei are not randomly positioned . We simulated random nuclear positioning by generating random and independently uniformly distributed x- and y-coordinates of the nuclear centers in rectangular domains . Analyzing the experimentally measured nearest-neighbor distances between the nuclei showed that the arrangement is not random ( Fig 1E ) ; rather , there is a characteristic distance between the neighboring nuclei . 3 ) Along the long axis of the cell , nuclei are relatively evenly spread , with a slight increase of the nuclear density near the cell poles ( Fig 1F , lower row ) . 4 ) Along the short axis of the cell , nuclei tend to form a single file ( SF ) near the cell center in narrower VL4 cells , whereas in wider VL3 cells , nuclei are typically found in a double file ( DF ) arrangement ( Fig 1F , upper row ) . In fact , the average nuclear x-position within each cell is a function of the cell width ( Fig 1G ) : the nuclear x-position increases with increasing cell width , irrespective of the cell type . Thus , nuclei in a wide VL4 cell have the same position as those of a VL3 cell of equal width . 5 ) VL3 and VL4 nuclei have the shape of flat ellipsoidal discs with their long axis oriented along the cell’s long axis ( Fig 1B and 1D ) . 6 ) Microtubules ( MTs ) anchored at the nuclear envelopes form asters around each nuceus ( Fig 1C ) . When MT organization is disrupted , nuclei have been shown to be mispositioned [25] , suggesting that MTs exert forces on the nuclei , and that the resulting force balance is the key to the nuclear positioning . One general approach to modeling movements of cell organelles generated by MT-motor forces is “interacting particle modeling” . After assuming or calculating a mean MT-motor force between material objects in the cell as a function of distance between the objects , one can solve equations of motion for the interacting objects via position-dependent force laws . This leads to solving a system of ordinary differential equations ( ODEs ) , the number of which is equal to the number of the objects [22 , 23 , 26 , 27] . Such models can be simulated rapidly such that many different internuclear force types can be screened . Another approach is to avoid assumptions and approximations and to formulate a “detailed agent-based model” , which involves solving numerically equations of elasticity theory for each MT , together with equations of motions for motors and objects to which MTs and/or motors are anchored [24] . This leads to solving a large system of partial differential equations ( PDEs ) , or a gigantic ODE system , which is much more difficult and computationally expensive than the first approach . While this approach provides more detailed predictions , many motors are involved in the force balance [19] , and it is not clear which combination of the motors generates the force in this context . Further , mechanical characteristics of most of these motors are unknown . In principle , one could use all possible motor combinations [28] , but the long computation times and high dimensionality of the model parameter space makes parameter scans of the detailed stochastic model impossible [29] . In this work , we systematically combine the two approaches to suggest mechanisms of nuclear positioning . We begin by screening various forces using interacting particle modeling and determine which model can be responsible for not only qualitative features of the observed spatial patterns , but also explain quantitatively all the subtle geometry of the nuclear positioning . Such an unbiased and systematic computational screen was needed for a few reasons . First , there are multiple simple and intuitive models that predict roughly uniform nuclear distribution , and choosing between them by a traditional thinking process is vulnerable to psychological biases . Second , spatial patterns generated by simple forces may be highly complex , counterintuitive , and non-robust [30] . After this screen , we used learned force characteristics to inform a detailed agent-based stochastic model with explicit simulations of individual MTs to confirm the lessons from the screening . The screen of the interacting particle models resulted in a single model that fits the data best . According to this model , the nuclei repell each other and the cell boundaries with the long-range forces that decrease with distance . This strongly suggested that pushing of MTs , either by polymerization , or by kinesin-generated force on the MT plus ends , establishes nuclear positioning: the number of growing MT ends decreases with the distance from the nucleus , and we expect a repulsive force decreasing with distance in this simple scenario . To test this model in a concrete molecular context and confirm the assumptions A3-5 that underline the usage of deterministic , isotropic , and distance-dependent forces in the interacting particle model ( Sec Force-balance and force-screening ) , we turned to a detailed agent-based simulation of microtubule-generated mechanics of the multi-nuclear cell . Such a model allowed examination of whether stochastic effects are negligible , and whether MT bending and resulting elastic forces result in unforeseen effects . We choose to use the microscopic , stochastic simulation tool Cytosim [24 , 33] , which has been successfully applied to a wide range of cell biological problems [34–36] . Using Cytosim , we simulated hundreds of MTs per nucleus , which are distributed uniformly around the nuclear circumference , cantilevered in the nuclear envelope and grow in a radially symmetric way . Individual MTs are treated as elastic rods , with the length of each MT characterized by the stochastic dynamic instability process [29] , whereby each MT undergoes repeated stochastic cycles of growth , catastrophe , shortening and rescue . Contact of a growing MT end with a neighboring nucleus or cell boundary results in MT bending , which generates an elastic pushing force . While pushing forces do not decrease the growth rate of MTs , catastrophy rates can increase to a maximum of 0 . 02 per second ( see Tab 2 ) , resulting in relatively brief , but typically more than 50s long , force generation events . Similar force events would be observed in stochastic simulations of forces exerted by a kinesin plus-end motor . Thus we did not include explicit molecular descriptions of kinesins in these simulations . The sum of pushing forces from all MTs constitutes the net force .
In this study , we used a “big” imaging data set generated from 2500 myonuclei from Drosophila larvae to computationally examine potential mechanisms of positioning of multiple nuclei in the cells . Rather than employing a traditional reductionist modeling approach , we used a different modeling philosophy , sometimes called “ensemble modeling” or “reverse engineering” . The idea is to start with multiple possible models and to use the ( predicted ) data to eliminate as many models as possible , dependent on their ability to recapitulate biological systems . This approach was successfully applied to cell signaling dynamics , metabolic networks , cell cycle , and spindle geometry [28 , 39–41] . In some instances , a small number of models can be analyzed one by one , as in a recent study on chemotaxis model inference [42] . In other instances , the number of model variants is so great that an unsupervised or semi-supervised computer screen of the models is necessary [28] . We searched computationally for the types of forces that could occur between pairs of nuclei and between nuclei and the cell boundary and could lead to positioning of the myonuclei . A similar problem , mitotic spindle positioning , has a long history [21] and reductionist modeling proved helpful in that case . However , an approach philosophically similar to ours was recently applied successfully to search for forces positioning the sperm MT aster in sea urchin eggs [43] . We started with a large number of potential forces and formulated a few hundred potential models , each characterized by a few mechanical parameters . We then used 1 . 5 million simulations to solve the differential equations describing movements of the nuclei predicted by each of these models at various parameters . We filtered out the vast majority of the models that were not able to predict the uniform spread of the nuclei along the cell long axis or the tendency of the nuclei to self-organize into the single file in narrow cells and double file in wide cells . The remaining 12 model classes were further tested by their ability to predict correct average nuclear position along the short axis of the cell and on robustness , the relative insensitivity of the models’ behavior to parameter values . These tests left us with two possible models , the parameters of which were fully determined by requiring the models to quantitatively fit the data in 200 imaged cells . The remaining two models made three non-trivial predictions: 1 ) the double-file pattern in wide cells is a zig-zag; 2 ) the average nuclear position along the cell short axis has the forked bifurcated dependence on the cell width , and 3 ) nuclear density is higher near the cell poles . Note that both models closely fit the experimental data very well , even though these data was not used to find the models’ parameters . Remarkably , these two models make opposite predictions about the nuclear shapes . One of the models predicts that the nuclei have ellipsoidal shapes with the long axes oriented perpendicular to the cell long axis , which is contrary to the experimental data . Incidentally , this model is also less robust than the other , ultimate , model , which not only predicts correctly that the ellipsoidal nuclei have long axes oriented along the cell long axis , as observed , but also fits very well the measured dependence of the nuclear aspect ratio as function of the cell width . Ultimately , only one model recapitulates all characteristics of nuclear positioning in VL muscle cells . It suggests that , nuclei repel each other and the cell boundary with forces decreasing with the distance . Our data suggest a simple molecular mechanism , which generates MT pushing forces , either by MT polymerization , or by MT interactions via kinesin motors on the nuclear envelopes and cell cortex . We support the computational screen of the simple models , in which the nuclei interact as particles by isotropic and deterministic forces , with simulations of a detailed agent-based mechanical model , in which we simulate hundreds of MTs undergoing dynamic instability , bending and pushing on the nuclei and boundary with elastic forces . These simulations support the types of forces hypothesized in the simple models , confirming that subtle stochastic , elastic and geometric effects do not invalidate the simple models’ assumptions . More importantly , the agent-based simulations generate the single- and double-file nuclear patterns in narrow and wide cells , respectively , as observed and as predicted by the simple models . Note that each simulation of the microscopic model took hours up to many days on an Linux machine with a Intel Core i7-7700 processor . As such , parameter exploration of the detailed models , or testing whether they reproduce subtle observed data features , is prohibitive . In the future , we plan to use more sophisticated mathematical methods [44] of solving the inverse problems—inferring the models from the data . While the involvement of MTs and molecular motors in the nuclear positioning is firmly established , we do not provide direct proof that a mechanical force balance is the main mechanism of nuclear positioning . Another possibility is that there is a preexistent , perhaps morphogen-governed , pattern in the cell , and that MTs simply tether the nuclei to special locations in this pattern . Relevant to this thought is the fact that small nuclear clusters aggregate at neuromuscular junction in mammalian cells . Our model makes detail predictions about nuclear positions in resting/fixed muscle cells . However , functioning muscle cells contract , and it is likely that the actomyosin contraction forces are orders of magnitude greater than the MT-based forces . Thus , it is hard to imagine that MT asters are sufficient to resist nuclear displacement during muscle contractions , and additional nuclear tethers might be involved in maintainaing an established pattern [15] . Future in vivo experiments , including genetic and biophysical manipulations and live cell imaging , will be required to investigate nuclear positioning in contracting muscle cells . However , we note that our model generates specific , testable predictions about the nuclear pattern in cases where the cells acquire unusual shapes and sizes or contain variable numbers of myonuclei . Another intersting aspect of muscle biology that could benefit from our modeling approch is the initial positioning of nuclei in developing embryonic muscle fibers . In the early embryonic muscle cells in Drosophila , after myoblast fusion , the nuclei initially cluster together , then split into two clusters that segregate to the cell poles , and finally spread along the cell length [16] . It remains to be tested if a force balance model can explain these dynamics . Even more challenging is the problem of coupling of the cell growth , shape change , and protein synthesis with the dynamics of nuclear numbers , positions , sizes and transcriptional activity . Last , but not least , the majority of the muscle cell types are cylindrical , with nuclei positioned at the cell periphery on all the cell’s surfaces . These essentially 3D nuclear patterns require special studies . Active , non-random nuclear positioning has been attracting increasing attention lately [45] . In a number of recent studies , force generated by MTs and motors were shown to be crucial for nuclear positioning and movement [24 , 46 , 47] . We suggest that the approach that we describe here—sequential computational screen of particle interaction models followed by detailed agent based simulation of the force balance model—is the optimal way to incorporate modeling as part of the experiments directed at understanding not only multi-nuclear positioning in mammalian muscle cells , but also in syncytium , giant cells , granulomas and osteosarcomas , as well as mechanisms of other organelles’ positioning .
The following Drosophila stocks were maintained at standard conditions on cornmeal medium: w1118 ( Bloomington 3605 ) , Dmef2-GAL4 [48] , UAS-2xEGFP ( Blomington 6874 ) , UAS-GFP RNAi ( J . Zallen , SKI ) . Crosses ( UASxGAL4 ) were performed at 25°C; embryos hatched within a 2h period were selected and raised to third instar larval stage . Wandering third instar larvae were dissected and fixed in 10% formalin and labeled as previously described [16] . Muscle cells were labeled using Alexa Flour-conjugated phalloidin ( Life Technologies ) ; nuclear DNA was visualized with Hoechst 33342 ( Invitogen ) . Anti-Lamin ( DHSB , ADL 67 . 10 ) and anti-α-Tubulin ( Sigma , T9026 ) primary antibodies , and Alexa Flour-conjugated secondary antibodies ( Life Technologies ) were used to label the nuclei and microtubules , respectively . Whole larvae were mounted in ProLong Gold antifade reagent ( Invitrogen ) . VL3 and VL4 muscles in abdominal hemisegments 2-6 were imaged on a LSM 700 confocal microscope ( Zeiss ) . Quantification of confocal z-projections was performed using standard ImageJ and Matlab measurement tools . VL3 and VL4 cells were traced by hand , based on phalloidin labeling . Automated thresholding of fluorescence intensities of anti-Lamin and/or Hoechst labeling was used to generate binary images of VL nuclei . Nuclear centroids were used to calculate nearest neighbor distances . Cell widths and lengths heights were defined as average widths and lengths of the measured boundary . Nuclear ( x , y ) positions were transformed onto positions in a rectangle using a mapping that preserves relative distances from the boundary . ImageJ’s ellipse fitting tool was used to define nuclear shapes . Nuclear pattern of both experimental and simulated origins were categorized into single file ( SF ) , double file ( DF ) or neither using a histogram of the relative nuclear x-positions with 7 equally spaced bins . If the histogram had only one peak containing at least 60% of all nuclei , the cell was classified as SF . If the histogram had two peaks which together contained at least 60% of all nuclei , and the number of nuclei they contained differed by less than 50% , the cell was classified as DF , in all other cases as neither . Parameters are shown in Table 1 . Simulation details are given below , r = 7μm is the radius of a nucleus . The criteria K1-4 for valid patterns are as follows ( referring to Fig 3 △ = K1 , ◻ = K2 , ▼ = K3 , ○ = K4 , ◼ = K5 ) . K1: All nuclei centroids have to be at least r away from all cell sides and poles . K2: All nuclei centroids have to be at least 2r apart . K3: The nuclear pattern is classified as DF or SF for the wide and thin cell respectively ( see DF/SF classification details ) . K4: max y − min y > 2/3×cell height . K5: The mean nearest neighbor distance is larger than 30μm and 45μm for the wide and thin cell respectively . The last criterion avoids counting random patterns as false-positive ( compare Fig 1E ) . Candidate models have to lead to valid patterns in both cell geometries , but not for the same parameters ( this avoids missing good models ) . Parameters are shown in Table 1 . Simulation details are given below . The criteria for valid patterns are similar to Filter 1 , however since cells of variable width and number of nuclei were used , the SF/DF criteria is dropped and criteria K5 is replaced by K5⋆: We required that the mean NND of all nuclei ( in all 14 cell geometries ) is 40 ± 5μm . For each model , we take only parameter combinations leading to valid patterns ( K1-K4 , K5⋆ ) and minimize the mean error of the predicted average x-position as a function of cell width ( compared to the measured behavior ) . The curves shown in Fig 4A correspond to the parameters that minimize that error . Robustness with respect to the force range ( Fig 4B ) is obtained as follows: For each model and each combination of force ranges cN , cS , for each value of MS the fraction of the 14 cells fulfilling criteria K1-K4 , K5⋆ was determined , then added for all four criteria and finally maximized over the values of MS . This yielded a score between 0 and 4 for each model and pair ( cN , cS ) , the color in the Fig 4B represents this score . Only SF and DF cells ( see above ) were used and nuclei within 10% cell height of the poles were disregarded . The remaining N ^ y-positions in each cell were normalized via y norm = y max y - min y ( N ^ - 1 ) ( if the y-positions were equally spaced , this would yield a y-spacing of exactly 1 ) and all positioned were shifted , such that the middle-most nucleus has y-position zero . Now the normalized y-positions of all cells were collected ( using all SF y-positions for the SF auto-correlation analysis , and separating y-positions of nuclei right , and left of the middle of the cell for the DF correlation analysis ) . For the final histograms a bin spacing of 0 . 25 was used . To determine equilibrium positions , Eq ( 1 ) was solved on a rectangular domain using Matlabs ode solver ode15 , a variable-step , variable-order solver . To model finite size effects of nuclei , a size exclusion term was added in Eq ( 1 ) . For two nuclei whose centroids are a distance d apart , it takes the form f SE = Q SE ( 1 d 2 - 1 ( 2 r ) 2 ) H ( 2 r - d ) , where r = 7μm is the nuclear radius , H is the Heaviside function and QSE = 2000 describes the strength of the size exclusion . For size exclusion effects between nuclei and the cell boundary , 2r was replaced by r . Simulations were run until all of the right-hand-sides in Eq ( 1 ) had a 2-norm of less than 10−4 . Codes are available upon request . The simulation software Cytosim ( Ver . 3 , 2007 ) [26 , 33] was used . The configuration files are available upon request . To calculate a ( distance dependent ) force from a measured nuclear speed v we used the effective viscosity of the aster ηeff consisting of the sum of the nuclear and MT drag as implemented in Cytosim . The force is then given by f = v ηeff . | How the cell organizes its interior is one of the fundamental biological questions , but the principles of organelles’ positioning remains largely unclear . In this study we use computational modeling and image analysis to elucidate mechanisms of positioning of multiple nuclei in muscle cells . We start with the general hypothesis , supported by published data , that a force balance generated by microtubule asters growing from the nuclei envelopes are responsible for pushing or pulling neighboring nuclei and cell boundaries , and that these forces position the nuclei . Instead of assuming what these forces are , we computationally screen all possible forces by comparing predictions of hundreds simple mechanical models to experimentally measured nuclear positions and shapes in hundreds of Drosophila muscle cells . This screening results in the model , according to which microtubules from one nucleus push away both neighboring nuclei and cell boundaries . We also perform detailed stochastic simulations of the only surviving model with individual growing , pushing and bending microtubules . This model predicts subtle features of nuclear patterns , all of which we confirm experimentally . Our study sheds light on general principles of organelle positioning . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"kinesins",
"microtubules",
"muscle",
"tissue",
"geometry",
"animals",
"aspect",
"ratio",
"simulation",
"and",
"modeling",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"systems",
"science",
"mathematics",
"molecular",
"motors",
"experimental",
"organism",
"systems",
"cellular",
"structures",
"and",
"organelles",
"drosophila",
"cytoskeleton",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"proteins",
"agent-based",
"modeling",
"biological",
"tissue",
"biophysics",
"insects",
"muscle",
"cells",
"arthropoda",
"physics",
"biochemistry",
"cytoskeletal",
"proteins",
"eukaryota",
"cell",
"biology",
"anatomy",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"computational",
"biology",
"microtubule",
"motors",
"organisms",
"biophysical",
"simulations"
] | 2018 | Mechanical positioning of multiple nuclei in muscle cells |
Notch receptors act as ligand-dependent membrane-tethered transcription factors with a prominent role in binary cell fate decisions during development , which is conserved across species . In addition there is increasing evidence for other functions of Notch , particularly in connection with Wnt signalling: Notch is able to modulate the activity of Armadillo/ß-catenin , the effector of Wnt signalling , in a manner that is independent of its transcriptional activity . Here we explore the mechanism of this interaction in the epithelium of the Drosophila imaginal discs and find that it is mediated by the ligand-independent endocytosis and traffic of the Notch receptor . Our results show that Notch associates with Armadillo near the adherens junctions and that it is rapidly endocytosed promoting the traffic of an activated form of Armadillo into endosomal compartments , where it may be degraded . As Notch has the ability to interact with and downregulate activated forms of Armadillo , it is possible that in vivo Notch regulates the transcriptionally competent pool of Armadillo . These interactions reveal a previously unknown activity of Notch , which serves to buffer the function of activated Armadillo and might underlie some of its transcription-independent effects .
The Notch gene of Drosophila encodes a member of a family of conserved single transmembrane receptors with key tasks in the information processing activity of animal cells [1]–[4] . They are involved in a wide variety of processes during development but their best characterized function is in the process of lateral inhibition and related events , in which Notch signalling is used to choose between two alternative cell fates in a context dependent manner [4]–[6] . There are two prominent structural features that define the family: a tandem array of EGF repeats in the extracellular domain that act as docking sites for ligands to trigger and modulate the activity of Notch , and seven ankyrin ( ANK ) repeats in the intracellular domain that provide a major face for interactions with effectors [7]–[14] . It is well established that Notch acts as membrane-tethered transcription factor ( reviewed in [1] , [4] , [15] ) . Binding of members of the DSL ( Delta , Serrate , Lag1 ) family of Notch ligands to specific EGF-like repeats leads to the shedding of most of the extracellular domain and triggers a sequence of proteolytic cleavages in the membrane proximal region , which result in the release of the Notch intracellular domain ( Nintra ) from the membrane [1] , [15]–[19] . Nintra accesses the nucleus where it modulates transcription through interactions with a member of the CSL ( CBF in mammals , Su ( H ) in Drosophila , and Lag2 in C . elegans ) family of transcription factors , and Mastermind ( MAML in vertebrates ) [20]–[22] . The interactions between Notch and CSL are mediated by the ANK repeats [11] , [23] and result in the activation of specific target genes . Recently , a number of studies , particularly in Drosophila , suggest that endocytosis and traffic of Notch are required for the generation and activity of Nintra [24]–[33] . Inappropriate activation of Notch signalling has been associated with a number of tumours in humans; in particular with T-cell acute lymphoblastic/lymphoma ( ALL ) leukemias , where activating mutations in Notch have been found to be linked to the disease [34]–[36] . However , there is also evidence that Notch can act as a tumour suppressor [37]–[40] . In one instance , this tumour suppressor function has been associated with signalling by ß-catenin , the effector of Wnt signalling [38] , [39] . Functional interactions between Wnt and Notch signalling have been reported frequently ( reviewed in [3] ) and are underpinned by biochemical studies that identify Dishevelled , GSK3ß , and ß-catenin , all key elements of the canonical Wnt signalling pathway , as Notch interacting proteins [3] , [41]–[47] . Although in many instances these interactions probably reflect the convergence of the two signalling pathways onto common target genes , studies in Drosophila have shown that Notch can modulate Wnt signalling in an Su ( H ) -independent manner by targeting Armadillo , the Drosophila homologue of ß-catenin [43] , [45] , [48]–[50] . Here we explore the mechanism of the interaction between Notch and Wnt signalling in Drosophila . We find that in the absence of Notch , an activated form of Armadillo promotes changes in the proliferative and adhesive properties of epithelial cells in Drosophila . This observation reveals an effect of Notch on Wnt signalling that is independent of its ligands and the activity of Su ( H ) , but requires the endocytosis and traffic of Notch . Our results provide a mechanism for the interaction between Notch and Wnt signalling that has implications for the homeostasis of the cell and , perhaps for the development of tumours .
Experiments in Drosophila have suggested that Notch can modulate the activity of Armadillo in an Su ( H ) -independent manner [43] , [48]–[50] . This observation is reminiscent of the situation in the skin of the mouse where loss of Notch1 function leads to elevated levels of ß-catenin and sensitizes the tissue to the development of basal cell carcinomas [38] , [39] . For this reason and to explore further the mechanism of the interaction between Notch and Armadillo , we expressed in the imaginal discs an activated form of Armadillo , ArmS10 ( a GSK3ß insensitive form of Armadillo that promotes constitutive Wnt signalling [51] in cells mutant for Notch ) . Loss of Notch function during the development of the wing results in stage-dependent altered growth rates and patterning defects , with little evidence of an increased activity of Armadillo ( Figures 1A and S1A ) [43] , [52] , [53] . This observation could be due to the loss of the Su ( H ) -dependent activity , which might mask additional consequences of the loss of function of Notch in this system . In contrast to the effects of loss of Notch function , gain or loss of Wnt signalling has only subtle effects on the growth of the wing primordium [54]–[57] , and expression of ArmS10 ( along the anterior-posterior [AP] boundary using dpp-Gal4 driver ) results in changes in gross morphology and alterations in cell fate in a region-specific manner with little or no effect on the overall size of the wing pouch or cell proliferation ( Figures 1B and S1B ) [43] , [53] , [55] , [57]–[59] . However in the absence of Notch , expression of ArmS10 produces outgrowths in the wing discs ( Figures 1C and S1C ) , which are reminiscent of the effects of mutations in lgd , exp , and mer , which have been linked with tumour suppression in Drosophila [26] , [60] . In addition , there are some effects on cell fate , e . g . , in the notum neural development is observed in regions outside the proneural clusters where Armadillo gain of function or Notch loss of function on their own have little or no effect ( Figure S1D–S1F ) . In these experiments the clones are generated continuously , using the FRT/FLP system , and therefore the effect is a cumulative average of clones generated at different times and different places . In order to explore the origin and fate of these outgrowths in more detail , we overexpressed ArmS10 in clones of Notch mutant cells generated at defined times in development using the MARCM method [61] . Clones of Notch mutant cells generated early in larval development are never recovered , probably because of competition by surrounding wild-type cells , and with later inductions the number and size of clones of Notch mutant cells observed increases , though it never reaches the figures of wild-type clones [52] ( Figures 2A , S1A , S4A , and S7C for Notch clones and S5A for wild-type clones ) . In general the clones of Notch mutant cells are not frequent and do not grow well . Expression of ArmS10 in Notch mutant cells rescues the viability of the early generated clones ( 24–48 h after egg laying [AEL] , Figure S2 and Video S1 ) and leads to tightly packed spheres with large numbers of cells and abnormal polarity . There is usually a single large sphere per disc , which tends to be positioned on the edges of the disc suggesting a tendency of the cells to sort from the surrounding ones . Clones of cells mutant for Notch expressing ArmS10 induced between 48–72 h also give rise to sphere-like structures with large numbers of cells similar to the early ones ( Figures 2B and S3B; Video S2 ) , but those induced after 72 h appear scattered through the tissue , lose basal contact , and exhibit a variety of organizations ( Figure 2B1 , and for details and discussion , Figures S2 and S3 ) . It is reasonable to think that the later-induced clones represent the early events in the process of formation of the spheres of cells , and this suggests that coalescence of different clones is a component of the phenotype . These overgrown aggregates are not restricted to the wing pouch as they can also be found in the notum as well as in other discs and , interestingly , in the peripodial membrane where cells lose their characteristic flat epithelial appearance and can fuse with the cells of the wing epithelium ( Figure S3B and unpublished data ) . Altogether , these observations suggest that loss of function of Notch unlocks a potential for ArmS10 to regulate cell proliferation , polarity , and adhesion . Some of this activity might be mediated by canonical Wnt signalling , and the clones of Notch display elevated levels of ArmS10 and , most significantly , elevated levels in the nuclei ( Figure S6 ) . There are suggestions that the activity of Notch that regulates Wnt signalling does not require the biochemical events associated with ligand-dependent cleavage and transcriptional activity of Notch [43] , [45] , [49] , [50] . To test this further we assessed the effects of expressing ArmS10 in clones of cells lacking the Notch ligands , Delta and Serrate , as well as its transcriptional effector Su ( H ) . In order to study the effect of ligand-dependent signalling on the activity of Arm , we chose to express ArmS10 in cells that simultaneously lack Delta ( Dl ) and Serrate ( Ser ) . Clones of cells doubly mutant for Dl and Ser are more frequent than Notch mutant clones at any stage of development , and do not exhibit obvious phenotypic alterations ( Figure 3A ) . The differences between the two mutant conditions are further emphasized by their differential behaviour in the presence of ArmS10: in contrast to Notch mutant cells , Dl/Ser double mutant cells expressing ArmS10 remain integrated in the epithelium and do not exhibit growth defects relative to the Dl/Ser double mutants alone ( Figure 3B ) . Surprisingly , clones of cells lacking Su ( H ) exhibit phenotypes that are different from both Notch and Dl/Ser double mutants: many small clones scattered over the disc , with very rugged edges and associated with a large number of dead cells in the basal side ( Figure 3C and Video S4 ) . Expression of ArmS10 in these clones increases their size , reduces the number of apoptotic cells , and makes the clones more rounded in appearance but cells do not lose their polarity ( Figure 3D and Video S5; for details and comparisons see Figure S7 ) . One possible interpretation for these changes is that the expression of ArmS10 is able to rescue some of the apoptosis caused by the loss of Su ( H ) and give rise to bigger and more organized clones . These results are surprising as the different phenotypes caused by the loss of function of Dl , Ser , Notch , and Su ( H ) challenge the simple linear interaction between Notch , its ligands , and its effector . These differences and the singular phenotypes of Notch in the presence of Arm S10 also emphasize that the effects of Notch on the activity of Arm are unlikely to be mediated by its Su ( H ) -dependent transcriptional activity . The effects of loss of Notch function on the activity of Armadillo provide a clear cut experimental test for the possibility that Notch encodes more than the Su ( H ) -mediated activity . To do this , we tested the ability of different forms of the Notch receptor to complement the effects of loss of Notch function on Armadillo activity in the wing disc: a full length Notch molecule ( FLN ) , and two membrane-tethered versions of Nintra , a CD8eGFPNotch ( CeN ) and CD8Notch ( CN ) , with the extracellular and TM domain of CD8 , which cannot be cleaved as they lack the intramembrane cleavage sites , and do not act through Su ( H ) ( see Figure 4F ) [48] , [49] . To gain a better understanding of the activity of these molecules , we first analyzed their subcellular localization in wing imaginal discs ( Figure 4 ) . The CeN molecule , assessed using the fluorescence of enhanced green fluorescent protein ( eGFP ) ( or antibodies against CD8 , see below ) , localizes to a domain both apical to and overlapping with the adherens junctions , and to large intracellular puncta located throughout the cell ( Figure 4A and 4C ) . The CN molecule can only be visualized by immunostaining , but displays a similar distribution to CeN , ruling out an effect of eGFP on the distribution and activity of the molecules ( Figure 4D and unpublished data ) . These localizations show an overlap with that of endogenous Notch and are determined by the intracellular domain of Notch , as a control of CD8 fused to eGFP ( Ce ) is distributed to all membranes of the cells indiscriminately ( Figure 4B ) . We concentrated the rest of the studies on CeN , testing for ligand- and Su ( H ) -independent activities of Notch and in particular for its ability to regulate the Armadillo activity in the absence of Notch , though CN has similar activities in vivo . Expression of CeN in cells of the wing imaginal discs results in small clones with rounded edges ( Figure S5C ) , suggesting that it has an ability to reduce growth . The resulting adults exhibit gain of function phenotypes: loss of PNS precursors and veins ( unpublished data ) . We also find that CeN can provide the growth suppressor activity of Notch on the activity of ArmS10 . Clones of Notch mutant cells that express ArmS10 together with CeN are smaller than those that express ArmS10 alone ( Figures 2C , 2C1 , 5B , and 5D ) , and cells recover their polarity and adhesive properties , spreading over the disc ( see also Figure S3C and S3D ) . To eliminate the possibility that the effects are due to a “neomorphic” activity of the CeN molecule , we repeated the experiment using FLN and observed a similar reduction in the size of the clones as observed with CeN ( Figure 5C and 5D ) . We complemented these experiments by expressing Nintra , a form of Notch that promotes mainly the transcriptional activity , in cells that lack Notch and express ArmS10 . The result is a combination of two phenotypes: a larger disc and , additionally , a suppression of the activity of ArmS10 in the clones ( Figure S8 ) . While this could be construed to suggest that the effects of Notch on Arm are mediated by Nintra , this interpretation should be considered carefully . In cultured Drosophila cells , Nintra can reduce the activity of Armadillo on a Wnt reporter in a manner that is independent of Su ( H ) [49] , and this is likely to be also the case here . It is well established that in the wing , Arm and Nintra synergize [53] , [58] , and while this interaction can explain the large size of the discs observed in this experiment , it cannot explain the suppression of the activity of ArmS10 in the clones , which are now reduced in size . We surmise that this suppression is mediated by the excess of Nintra binding to proteins that interact with it , particularly Arm , and thereby neutralizes their activity . Altogether these results argue that the tumour suppressor activity of Notch is an intrinsic function of Notch itself , very likely mediated by the full length receptor , and that it does not require an interaction with its ligand nor its transcriptional function . The pattern and distribution of the CeN protein in the epithelial cells suggests that it is actively trafficking , as it is localized to the apical membrane of cells and to more basal puncta , reminiscent of vesicles , in a pattern that overlaps with that of the endogenous Notch ( Figures 4A and 6A ) . It has been shown before that Notch can be found in endosomes [26] , [33] . Therefore , for a preliminary characterization of these puncta , we investigated , whether they colocalized with several endosomal markers . We have detected a small degree of colocalization of Notch with either Rab5 or Rab7 and some more substantial localization with Rab11 ( early , late , and recycling endosomal markers , respectively; Figure S9A and unpublished data ) . However , many of the Notch puncta colocalized with SARA ( Figure S9C ) , an endosomal protein identified as an element of transforming growth factor-β ( TGF-β ) signal transduction pathway , which has been shown to regulate Notch signalling during asymmetric cell divisions [62] , and with Spinster and Carnation , proteins characteristic of late endosomes ( Figure S9B and unpublished data ) [63]–[66] . We have confirmed that CeN trafficks by checking that it colocalizes with endocytosed Dextran ( unpublished data ) and , most significantly , by uptake experiments using anti-CD8 and anti-Notch antibodies to label and chase cell surface bound Notch and CeN molecules in third instar wing discs ( Figures 6B–6D and S10; see Materials and Methods for details ) . In these experiments we observe internalization and change of subcellular localization of labelled Notch and CeN molecules within 10 min of labelling , suggesting that this traffic is likely to be an active process . Altogether these observations suggest that the CeN and endogenous Notch proteins are actively endocytosed . There are no clear endocytic motifs in the intracellular domain of Notch . However , deletion of the RAM-ANK domain ( CeN-ΔRANK ) leads to the accumulation of the protein in the apical region of the cell ( Figure S11B ) , either in the cell surface or in early endosomes , indicating that this domain is not necessary for the export of the molecule to the cell surface but that it is important for its endocytosis and traffic . This finding is highlighted by specific mutations in the ANK domain: receptors with point mutations in the ANK repeats , CeN-DM1 , and CeN-DM2 ( see Materials and Methods ) , also accumulate in the apical region of the cell near the adherens junctions and are not properly internalized ( Figure S11C and S11D ) . An uptake experiment in wing discs expressing CeN-DM1 clearly shows that this mutant form has an impaired traffic , probably slower ( Figure S12 ) . In contrast to CeN , which causes gain of function phenotypes , these mutant proteins have no activity when overexpressed in the imaginal discs on their own , i . e . , they produce no visible phenotype ( unpublished data ) . These results suggest that the activity of CeN requires its traffic and not cleavage or nuclear translocation . Altogether these results suggest that Notch undergoes very effective traffic in a ligand-independent manner and that endocytosis and traffic depends on structural motifs located in the intracellular domain . The phenotypes caused by the expression of ArmS10 in the absence of Notch lend support to the observation that both proteins interact and that in normal conditions Notch can downregulate both the amount and the activity of Arm [43] , [48]–[50] . Our experiments further suggest that this downregulation is mediated through the traffic of Notch . If this is indeed the case , we should observe Notch and Armadillo associated in endosomal vesicles and we might expect that the overexpression of Notch should affect the distribution of ArmS10 as well as of endogenous Arm . In fixed tissue we observe a high degree of colocalization between Notch and Armadillo in puncta that probably correspond to endocytic vesicles ( Figure 7A and 7B ) . This finding is confirmed by the observation that in antibody uptake and chase experiments , it is possible to observe some of the endocytosed Notch vesicles associated with Armadillo ( Figure 7C and 7D ) . The ability of Notch to interact with and possibly to recruit Armadillo is further demonstrated by the observation that overexpression of a full length Notch molecule in wild-type cells leads to an expansion of the domain of Armadillo localization to a broader apical domain with a subapical vesicular pool within the domain of Notch overexpression ( Figures 8 and S13 ) . In the case of ArmS10 , analysis of the effect of Notch on Arm must take into consideration the effects of ArmS10 on endogenous Arm , which is displaced from the adherens junctions into a cytoplasmic pool [43] . Full length Notch reduces cell surface ArmS10 , which can now be found in a large pool of subapical vesicles , and increases the number of apical vesicles of the endogenous Arm within its domain of expression . The alterations induced by Notch ( summarized in Figure 9 ) correlate with a decrease in the concentration and activity of Arm observed before [43] . The effects of Notch on Arm are mirrored by the effects of Arm on Notch: expression of ArmS10 induces a delocalization of Notch and CeN from the cell surface into a diffuse subapical domain and a general reduction in the amount of Notch or CeN in the cell ( Figure S13B , S13C , S13B1 , and S13C1 ) . We interpret these observations as resulting from the titration of molecules involved in the regulation of Wnt signalling by the very stable ArmS10 [67] , [68] , which in our case leads to a concomitant alteration in the localization and traffic of Notch . Expression of a different form of activated Arm , ( ArmΔNMyr ) , an N-terminally deleted myristylated form [69] emphasizes these interactions: this form of Arm distributes itself throughout the membranes of the cells [70] and induces a relocalization and concentration of Notch to the sites of ArmΔNMyr expression ( Figure 10A ) . The accumulation that we observe is likely to result from the removal of Notch from its normal sites of traffic and degradation . A similar form of Arm lacking the membrane association ( ArmΔN ) exhibits a much weaker interaction with Notch ( Figure 10B ) , underscoring that the pool of Notch that Arm interacts with is membrane bound . Altogether these observations support an interaction between Notch and Armadillo , and show that Notch downregulates , in a ligand- and Su ( H ) -independent manner , the activity of Arm by changing its localization to vesicles where it may be sequestered or targeted for degradation . This finding is corroborated by the observation that CeN can suppress the ability of ArmS10 to activate Wnt signalling in a dominant manner ( unpublished data ) as we had shown before for the related molecule TN [43] .
The relationship we have uncovered between Notch and Armadillo in Drosophila is reminiscent of that described in the skin of mice where targeted removal of Notch1 results in high levels of activated ß-catenin that prime the cells for the development of tumours [38] , [39] . In both , the wing disc and the skin , the defects ensue from two sequential steps: loss of a tumour suppressor ( Notch ) followed by activation of an oncogene ( Armadillo/ß-catenin ) , which sensitizes the system for the development of tumours . This sequence is well characterized in human cancers and our results suggest that Drosophila could be a good experimental system to study its causes and possible therapies . In the imaginal discs , this activity of Notch , which has been proposed to set a threshold for Wnt signalling [3] , [43] , can modulate growth and patterning in the rapidly dividing epithelium and might provide a paradigm for similar interactions and function in other systems . The large aggregates of cells that result from the activation of Armadillo in the absence of Notch could be construed as tumours , as they exhibit overgrowth and defects in polarity and adhesion . However , this correspondence awaits further experiments , and they might correspond to cells with compromised differentiation . Whatever the nature of these aggregates , this activity of Notch is not restricted to the developing wing as the same effect is observed in all imaginal discs ( unpublished data ) . As on their own neither increased Armadillo activity nor loss of Notch function elicit a similar effect , these cells represent a synthetic phenotype that reveals the physiological potential of these pathways as well as their close interactions . The effects of loss of Notch function in the mouse skin and the imaginal discs show that Notch performs an important function as a buffer against fluctuations in the activity of Arm/ß-catenin , and that as such it plays a role in the homeostasis of the cell . It is likely that the Axin/APC/GSK3-based complex that degrades Armadillo and ß-catenin is not totally effective and that , even in the absence of Wnt ligands , there are leakages of active Arm/ß-catenin that result in small bursts of signalling . We surmise that the role of Notch is to interact with the leaked activated Arm/ß-catenin and to degrade it in order to maintain the levels of spontaneous signalling low , thus providing its buffering function . The existence of complexes between Notch and Armadillo has been reported before [43]–[47] and is supported by our observation of their reciprocal change of localization in the overexpression experiments . There is little question that some of these interactions are likely to be associated with the transcriptional activity of both molecules , but our observations that Notch is able to recruit Armadillo to an apical domain , that endocytosed Notch can be found associated with Armadillo and that ArmS10 is stabilized in the absence of Notch , provide evidence for another level of interaction that is likely to be the basis for its buffering activity . This function might be associated with features of tumour suppressors as it would provide the mechanism to cope with transient high fluctuations in the amount or activity of oncogenes . It also raises the possibility of an association between the levels of Notch and the oncogenicity of ß-catenin , i . e . , there might be a tissue specific traffic of Notch that determines its tumour sensitivity . Our observations also have implications for the mechanism of activation of Arm/ß-catenin . There is evidence for distinct pools of Arm/ß-catenin involved in signalling and adhesion and , although it is generally accepted that the signalling pool is associated with a cytoplasmic soluble pool , a number of experiments cast doubts on this simple relationship [74]–[79] . Our results support the existence of an active pool of Arm/ß-catenin that , in epithelial cells , exists at or near the adherens junctions , and that it is this pool , rather than the general GSK3-sensitive pool , that is the target of Notch ( see also [43] ) . A GSK3-insensitive pool subject to further regulation by Axin has been suggested as the source of active Armadillo [67] , [79] , and interactions have been described between Axin and Notch in the regulation of Armadillo [48] . It might be that this pool corresponds to the membrane-associated pool that we identify here and that rather than a putative cytosolic pool , it is this pool that contains the transcriptionally competent Arm/ß-catenin . On the basis of these observations we surmise that , in the absence of a Wnt signal , Notch sequesters a cell surface-located pool of Armadillo , probably active Armadillo , traffics with it , and induces its degradation . This possibility is consistent with the effects of overexpression of Notch on the amount and localization of ArmS10 ( Figures 8 , 9 , and S13 ) and with the observation that suppression of endocytosis and traffic by mutations in the Drosophila Dynamin encoded by shibire result in a substantial increase in the amount of Armadillo ( [80] and unpublished data ) . Our results argue for a function of the traffic of Notch in the regulation of the activity and the amounts of Armadillo . However , in the mouse skin , Nintra can suppress some of the tumorous phenotypes caused by loss of Notch1 function by modulating the activity and the amount of ß-catenin [38] , [81] , [82] , and similar interactions have been observed in other systems [81] , [82] . Although it is possible that this reflects a contribution of the transcriptional activity of Notch to the regulation of ß-catenin , we think this is unlikely to provide the major component as the suppression of ß-catenin is not Su ( H ) dependent [49] , and in some cases the effect is not only on the activity of ß-catenin but also its amount [38] . Furthermore , there is increasing evidence that Nintra can perform activities and interactions that are not easy to reconcile with its function as a transcription factor [83]–[86] . One possibility is that cleaved Nintra has interactions and activities that do not involve Su ( H ) and its vertebrate counterparts , but it might also be the case that under experimental conditions in which there is an excess of this form of Notch , some of it engages in abnormal complexes with proteins that the intracellular domain of Notch normally interacts with , triggering a squelching effect [87] with functional consequences . Taking into account all the evidence presented here , we believe that squelching might be the cause of many of the interactions reported between Nintra and Armadillo/ß-catenin . It will be important to bear this in mind when interpreting the effects of overexpressing Nintra , particularly in cultured systems . It could be argued that the effects of CeN and related molecules are due to “neomorphic” effects . We believe that this is not the case . In fixed-tissue and antibody uptake and chase experiments , a fraction of CeN colocalizes with Notch , suggesting parallel activities of the two molecules . It is likely that the effects of CeN reveal the strong dominant effect on Su ( H ) -independent activities of Notch , much like Nintra reflects the transcriptional activity of the receptor . CeN also points to the existence of a pool of Notch that is usually in limiting amounts but active in specific trafficking functions . Thus we believe that the activity of CeN reveals the ligand-independent activity of the Notch receptor that targets the activity of Armadillo , and which is mediated by a pool of receptor that is not engaged in Delta , Serrate , Lag1 ( DSL ) –dependent signalling . Interestingly , Wnt signalling itself affects the traffic of Notch and promotes the degradation of the ligand-independent forms we use in our experiments ( unpublished data , manuscript in preparation ) . This observation is consistent with the notion that Wnt signalling promotes the degradation of molecules that oppose its activity , e . g . , Axin [79] , [88] , and this includes Notch ( unpublished data , manuscript in preparation ) . Furthermore , it is likely that the interaction between Dishevelled and Notch that has been described [41] , [43] , [45] is part of this regulatory interaction . Altogether these and the increasing number of reports of structural and functional interactions between elements of these two pathways lend support to the notion that both act in an integrated manner as a single functional module , which we have dubbed Wntch ( for Wnt and Notch signalling ) [3] . Our observations and conclusions could account for the frequent appearance of defects in Notch traffic and signalling in screens geared to uncover tumour suppressors in Drosophila [24] , [32] , [89] , [90] . We would like to suggest that Notch might be used to link the endocytic pathway and traffic apparatus to integrate and modulate signalling events , a function that might play crucial roles in the development of organisms and particularly in tissue homeostasis . A corollary of this suggestion is that the strong requirement for endocytosis and traffic in the transcriptional activity of Notch might be associated with its role in trafficking , which might be evolutionarily ancestral to its role as a transcription factor and perhaps extend to elements of signalling pathways other than Wnt .
The following Drosophila UAS and Gal4 stocks were used: ( w;;dpp-Gal4/TM6B ) ; ( w;UAS-CeN/CyOftz;MKRS/TM6B ) ; ( w;UAS-CeN-DM1/CyOftz;MKRS/TM6B ) ; ( w;UAS-CeN-DM2/CyO;MKRS/TM6B ) ; ( w;UAS-CeN-ΔANK/CyO;MKRS/TM6B ) ; ( UAS-ArmS10 on the 2nd ) ; ( UAS-ArmS10 on the X ) ; ( UAS-Ce ) ; ( UAS-CN ) ; ( w; If/CyOwg;UAS-FLN ) ; ( UAS-ArmΔNMyr ) ; ( UAS-ArmΔN a gift from G . Struhl ) ; ( Arm-GFP ) . To generate the clones using the FRT/FLP system , ( Df ( 1 ) N81k1 [FRT101w+]/FM6; ; dpp Gal4 , UAS FLP/TM2 ) females were crossed to ( ywGFP [FRT101w+]/Y; UAS-ArmS10 ) , ( ywGFP [FRT101w+]/Y; UAS-ArmS10/CyO; UAS-FLN/TM6B ) , ( ywGFP [FRT101w+]/Y; UAS CeN , UAS-ArmS10/CyOftz ) , or ( ywGFP [FRT101w+]; UAS ArmS10; UAS Nintra/SM6a-TM6B ) males . To generate Notch clones using the MARCM system , N55e11FRT19A/FM7-GFP , N55e11FRT19A/FM6; UAS-ArmS10 or N55e11FRT19A/FM6; UAS-ArmS10 , CeN females were crossed to P{ry[+] neoFRT19A}19A , P{w[+] tubP-GAL80} L1 , P{ry[+] hsFLP}1 , w; CyO/P{w[+] UAS-nucZ}20b , P{w[+] UAS-CD8:GFP} LL5; TM6 , Tb , Hu/P{w[+] tubP-GAL4} LL7 males ( FRT19 MARCM stock ) . Clones were induced in larvae 24–48 h , 48–72 h , or 72–96 h AEL by applying a 1-h heat shock ( 37°C ) . To generate the control clones ( FRT19A;UAS-CeN/+ ) , ( FRT19A;UAS-ArmS10/+ ) , or FRT19 males were crossed to females from the FRT19 MARCM stock and induced similarly in 48–72-h and 72–96-h AEL larvae . To generate the ligand mutant clones , ( UAS-ArmS10/+; FRT82 , Dlrev10 , SerRx82/+ ) or ( FRT82 , Dlrev10 , SerRx82/TM6B ) males were crossed to ( hsFLP , tub-Gal4 , UAS-GFP/FM6;;FRT82 , Tub-Gal80/TM6 ) females induced similarly in 48–72-h AEL larvae . For the Su ( H ) clones ( UAS-ArmS10;FRT40 , Su ( H ) Δ47/+ ) or ( w;FRT40 , Su ( H ) Δ47/CyO ) males were crossed to ( hsFLP;FRT40 , Tub-Gal80;tub-Gal4 , UAS-GFP/SM6-TM6B ) females induced similarly in 48–72-h AEL larvae . The clones expressing ArmS10 were recognized by α-Myc staining . CeN-DM1 and CeN-DM2 were generated by PCR-based mutagenesis of the sequence encoding the intracellular domain of Notch ( amino acids 1767–2703 ) and subcloning of the resulting constructs into the pUAST-DEST12 vector . UAS-CN: The sequence-encoding extracellular and transmembrane domains of CD8 ( obtained from CeN-DM1 construct ) was cloned into pUAST using the KpnI and XbaI sites to generate UAS-CD8 . The reverse primer used for amplification of the CD8 fragment contained a MluI site in addition to the XbaI site . The NICD sequence was amplified from pENTR-NICD and cloned into UAS-CD8 using the MluI and XbaI sites . UAS-Ce: The CD8-eGFP sequence was amplified from UAS-CeN-DM1 and cloned into the KpnI and XbaI sites of pUAST . UAS-CeN was generated from the UAS-CeN-DM1 mutant construct . UAS-CeN-DM1 was digested with BsiWI to remove the fragment of DNA containing the DM1 mutation . pENTR-NICD was also digested with BsiWI to generate the equivalent wild-type NICD fragment . This wild-type fragment was ligated into the remainder of the BsiWI-digested UAS-CeN-DM1 plasmid to replace the mutated version . Correct insert orientation was ascertained by digestion with MfeI and BsiWI . UAS-CeN-ΔRANK: The DNA sequence encoding residues 2142–2703 of Drosophila Notch was amplified from UAS-FLN and cloned into the XbaI site of UAS-Ce . Correct insert orientation of the resulting clones was assessed using MfeI and BsiWI . The antibodies used in this study were: mouse monoclonal antibody against the extracellular domain of Notch , α-NECD , ( C458 . 2H , 1∶5 , DSHB ) ; rat monoclonal against E-Cadherin ( DCAD2 , 1∶20 , DSHB ) ; antibody against the intracellular domain of Notch , α-NICD ( mouse monoclonal C17 . 9C6 , 1∶10 , DSHB; and sheep antisera generated in the lab , 1∶1 , 000 ) ; α-senseless ( from H . Bellen ) ; α-Armadillo ( N27A1 , 1∶20 , DSHB; and rabbit antisera 1∶50 , from A . Muller ) ; α-Scribble ( 1∶1 , 000; from C . Doe ) ; α-Myc ( 1∶1 , 000; from AbCam ) ; Rab7 and Sara ( 1∶100 , from M . Gonzalez-Gaitan ) ; Carnation ( 1∶750 , from H . Krämer ) ; Alexa-conjugated secondary Ab ( 1∶200 ) from Molecular probes . Imaginal wing disc were dissected from third instar larvae and fixed for 30 min ( 4% paraformaldehyde in balanced salt solution [BBS] with 1 mM CaCl2 ) . Then they were immunostained with the indicated antibodies in BBS ( 50 mM BES , 280 mM NaCl , 1 . 5 mM Na2HPO4 . 2H2O ) +0 . 1% Triton X-100 , 0 . 5% BSA 1 mM CaCl2 ) using standard antibody staining protocols . Discs were mounted in Vectashield and viewed using a confocal microscope ( see below ) . Imaginal wing disc were dissected from third instar larvae in cold BBS . Discs were pulse labelled with mouse α-NECD ( a 1∶2 mix of C458 . 2H DSHB supernatant in BBS ) and/or α-CD8 ( 1∶15 from Caltag Laboratories ) for 1 h at 4°C . After washing six times for 2 min each in cold BBS at 4°C , the discs were either fixed immediately ( 0-min chase ) or transferred to Clone 8 medium at 19°C for 10 , 30 , or 60 min . Fixation was done in 4% paraformaldehyde ( in BBS ) at room temperature for 30 min . Afterwards standard antibody staining protocols were used . Comparing the results of both protocols , we got the impression that the antibodies used on fixed tissue reveal the most stable pool of protein , while the pulse-chases reveal a specific pool that shows where the protein is located in that moment . Wing discs were examined under a Nikon Eclipse E800 microscope coupled to a BioRad MRC1024 or Zeiss LSM 510-Meta confocal units . The images of the pulsed-chased wing discs of the different time points were acquired in the same conditions of laser intensity , gain , and pinhole , and processed exactly the same way . Adobe Photoshop and Excel were used to assemble the figures and to quantify the clone areas in pixels . For the analysis of the relative size of the clones in different genetic backgrounds , images from third instar imaginal discs were assembled at the same resolution and magnification . Clones in chosen regions were then highlighted with a lasso and their areas calculated in pixels using Photoshop toolkit and Excel . The fluorescent intensity profiles were performed with the software package ImageJ ( RGB Profiler plugin ) . | Establishment of the correct shape and pattern of tissues within an organism requires the integration of molecular information present in signalling and transcriptional networks and demands delicate exchanges and balances of their activities . A large body of experimental work has revealed close correlations in the activities of two pathways: Notch and Wnt , which suggest the existence of multiple links between them . Notch signalling relies in part upon the activity of the Notch protein , a membrane-bound receptor with a transcription factor domain that can be released from the membrane by proteolytic cleavage . On the other hand Wnt proteins are ligands that trigger changes in the activity of ß-catenin , which is called Armadillo in the fruit fly Drosophila melanogaster . In this study we uncover a previously unknown activity for Notch: endocytosis and trafficking of full length Notch , which targets Armadillo for degradation . This activity of Notch is independent of its ligands , Delta and Serrate , and of its downstream effector , the transcription factor Suppressor of Hairless . We further show that in the absence of Notch , which has been shown to act as a tumor suppressor in mammals , expression of an activated form of Armadillo causes tissue overgrowth and changes in the polarity of cells . Our results suggest that Drosophila Notch can promote the degradation of activated forms of Armadillo and may buffer cells against fluctuations in Wnt signalling activity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/cell",
"growth",
"and",
"division",
"cell",
"biology",
"developmental",
"biology/morphogenesis",
"and",
"cell",
"biology",
"cell",
"biology/cell",
"signaling"
] | 2009 | Ligand-Independent Traffic of Notch Buffers Activated Armadillo in Drosophila |
Accurate diagnosis of infection with the parasite Strongyloides stercoralis is hampered by the low concentration of larvae in stool , rendering parasitological diagnosis insensitive . Even if the more sensitive agar plate culture method is used repeated stool sampling is necessary to achieve satisfactory sensitivity . In this manuscript we describe the development of a coproantigen ELISA for diagnosis of infection . Polyclonal rabbit antiserum was raised against Strongyloides ratti excretory/secretory ( E/S ) antigen and utilized to develop an antigen capture ELISA . The assay enabled detection of subpatent rodent S . ratti and human S . stercoralis infection . No cross-reactivity was observed with purified E/S from Schistosoma japonicum , the hookworms Ancylostoma caninum , A . ceylanicum , nor with fecal samples collected from rodents harboring Trichuris muris or S . mansoni infection . Strongyloides coproantigens that appear stable when frozen as formalin-extracted fecal supernatants stored at −20°C remained positive up to 270 days of storage , whereas supernatants stored at 4°C tested negative . These results indicate that diagnosis of human strongyloidiasis by detection of coproantigen is an approach worthy of further development .
The diagnosis of gastrointestinal infections typically relies on either empirical clinical diagnosis , or demonstration of the pathogen using standard microbiological techniques . However , a number of important gastrointestinal pathogens are difficult to detect using such techniques , and in some of these infections such as amoebiasis and strongyloidiasis , a potentially life-threatening pathogen may be present despite few or no clinical symptoms and negative diagnostic tests . Serological diagnosis by ELISA [1] , [2] and agar plate coproculture [1] , [2] are currently the standard techniques used for parasitalogical diagnosis of infection with Strongyloides stercoralis . Agar plate coproculture has been reported to show high sensitivity and specificity . However , it requires a 24–48 hours incubation time , an experienced technician to correctly identify the larvae , and can represent a biohazard to the laboratory scientist [1] , [2] . While serologic diagnosis , usually by ELISA is highly sensitive [3] , specificity can be low due to the persistence of antibodies from previous infection or cross-reactive antibodies [4] . Coproantigen assays have been developed for the diagnosis of a range of human and animal intestinal infections . In general antibodies raised against whole parasite extracts are coated on microtiter plates , and subsequently fecal antigen is captured and detected with the same or second parasite-specific antibody in a capture assay [5] , [6] . The use of monoclonal antibodies is generally believed to increase both the sensitivity and the specificity of such assays [7]–[10] . Examples of coproantigen ELISAs developed for the detection of a range of intestinal infections include amoebiasis [11]–[13] , bacterial and viral gastroenteritis [14] , trematodes [15] and cestodes [16] , [17] . Of note , the more successful coproantigen assays have been developed against intestinal infections with either large parasites or pathogens with a likely high fecal antigen load . However , coproantigen detection is not limited to diagnosis of intestinal infections , as it has also been applied to the detection of gastric adenocarcinoma [18] and the detection of fecal occult blood [19] . Another advantage of such coproantigen assays is the ability to detect cryptic or pre-patent infections [6] , [20] . In the case of nematode infections , a number of assays have been reported for the detection of animal host parasites , including Necator americanus [21] , Ancylostoma ceylanicum [20] , Teladorsagia circumcincta [22] and Heligmosomoides polygyrus , [6] . There have also been reports where assay sensitivity and/or specificity was unsatisfactory e . g . detection of Strongyloies ratti [23] , Haemonchus contortus [24] and Ostertagia ostertagi [25] . With the exception of the O . Ostertagia assay , assays of satisfactory performance have employed specific antiserum raised against excretory/secretory ( E/S ) antigens of the respective parasites , rather than utilizing antiserum raised against total somatic antigen . However , the usefulness of these assays in the detection of heterologous antigen in human infections , or in the development of an assay to detect human nematode infection has not been reported . In this study , polyclonal antiserum was raised against S . Ratti E/S antigens and used to develop an assay capable of detecting Strongyloides antigen present in the feces of rodents harbouring S . Ratti . This assay was then tested in a pilot proof of principle study in a human patient with patent S . Stercoralis infection . The sensitivity of the antibody for detection of S . Ratti antigen as well as heterologous S . Stercoralis coproantigen was investigated . Techniques to reduce cross-reactivity with fecal components were also explored , and the analytical specificity of the coproELISA was determined by testing E/S and fecal supernatants collected from other animals or humans with helminth infections . Finally , the effect of preservation and storage conditions of fecal samples and coproELISA was investigated .
The S . ratti life cycle was maintained in 4–8 week old male Wistar rats and infection was established as previously described [26] . Parasites to establish the life cycle were kindly provided by Prof . Mark Viney ( University of Bristol , UK ) . E/S antigens for immunization was collected from parasitic adult worms harvested from infected rats 10–14 days post-infection and rinsed extensively in RPMI ( Invitrogen , Carisbad , CA ) containing 200 µg/ml ceftriaxone ( Roche , Basel , Switzerland ) , 2 . 5 µg/ml Amphotericin B ( Sigma-Aldrich , St . Louis , MO ) and 400 µg/ml Gentamicin . Cleaned parasitic adult worms were incubated at 37°C in 5% CO2 for 24 hours in 3–4 ml of RPMI containing 20 µg/ml ceftriaxone , 0 . 25 µg/ml Amphotericin B and 40 µg/ml Gentamicin in 6 well tissue culture plates ( Falcon ) . Immediately after incubation , a pre-mixed cocktail of protease inhibitors ( complete , mini-protease inhibitor cocktail , Roche ) was added to a 1X final concentration , and the E/S products stored at −20°C . Frozen E/S products were centrifuged at 3 , 000 g for 5 min and the supernatant collected . Concentration and dialysis of E/S was undertaken in Centricon YM-10 devices ( Millipore , Billerica , MA ) according to the manufacturer's instructions . Protein concentration was determined using a BCA kit ( Thermo Scientific , Waltham , MA ) . S . stercoralis E/S products were kindly donated by Prof Gerhard Schad ( School of Veterinary Medicine , University of Pennsylvania ) . Adult worms were collected from an experimentally infected hamster harboring ∼390 parasitic adult worms . E/S products were collected from these parasitic adult worms as previously described , and lyophilized . Lyophilized E/S antigens were reconstituted in distilled water , concentrated and quantitated as previously described for S . ratti E/S antigens . Schistosoma japonicum E/S products were collected from 100 pairs of S . japonicum adult worms that were cultured for 24 hours in RPMI . A . caninum E/S antigens were kindly provided by Tegan Don ( Queensland Institute of Medical Research , Brisbane , Australia ) . Adult A . caninum worms were harvested from the small intestine of necropsied pound dogs . A . ceylanicum E/S products were donated by Prof Jerzy Benkhe ( University of Nottingham , UK ) . Adult worms were harvested from experimentally infected hamsters . All protein preparations were concentrated , dialysed and quantitated as previously described for S . ratti E/S antigens . Uninfected rat feces and fecal samples from S . ratti-infected rats were collected from laboratory rats housed at the Queensland Institute of Medical Research . Pooled feces collected from 6 mice without parasitic infection and 6 mice infected with S . mansoni was supplied by Mary Duke ( Queensland Institute of Medical Research , Brisbane , Australia ) . Lyophilized T . muris-infected mouse fecal supernatant was supplied by Prof . Jerzy Benkhe ( University of Nottingham , UK ) . Queensland Medical Laboratories ( Brisbane , Australia ) supplied anonymous uninfected human feces and three consecutive S . stercoralis-infected stool samples . These had been collected from a single patient with S . stercoralis infection proven by agar plate coproculture . Fecal supernatants were prepared at a ratio of 1∶3 ( v/v ) in a solution of PBS-T ( PBS containing 0 . 3% Tween-20 ) , 4% formalin or 10% formalin , and vortexed to homogenise in a 50 ml conical tube . Samples were centrifuged at 3 , 200 g for 15 min at 4°C , after which supernatant was collected and centrifuged to remove fecal debris . The cleared supernatants were then aliquoted and stored at −20°C unless otherwise specified . A New Zealand white rabbit was immunized with lyophilized S . ratti E/S products at the Institute for Medical and Veterinary Sciences ( Adelaide , Australia ) . The rabbit was bled prior to immunization to provide a negative control . The primary immunization consisted of 400 µg E/S antigen emulsified in Freund's complete adjuvant , injected subcutaneously . The antibody response was boosted with 200 µg E/S antigen emulsified in Freund's incomplete adjuvant and injected sub-cutaneously . Terminal bleed serum was shipped overnight at 4°C to the laboratory prior to aliquotting and storage at −80°C . Protein A sepharose ( GE Healthcare , Chalfont St . Giles , United Kingdom ) was used to purify α-E/S immunoglobulin from serum according to the manufacturer's instructions . IgG was eluted from the Protein A column by the addition of 1 ml 0 . 1 M glycine ( pH 3 . 0 ) . The eluate was immediately neutralized in 60 µl 1 M Tris ( pH 9 . 0 ) and separated into ten separate fractions . Optical density was monitored by spectrophotometry at a wavelength of 320 nm to identify fractions containing IgG . These were pooled and dialysed by four changes of PBS in Centriprep YM-50 concentrators ( Millipore ) . Purified α-E/S Ig was quantitated by spectrophotometry where 1 mg/ml Ig = A320 of 1 . 43 [27] . In a 1 . 5 ml tube , 1 . 63 mg of α-E/S Ig was diluted in 100 µl NHS-LC biotin working solution ( 0 . 1 mg NHS-LC biotin ( Thermo Scientific ) in 100 µl dimethylformamide ) . The volume brought up to 1 . 1 ml with PBS and biotinylated according to the manufacturer's instructions . Unbound biotin was removed , and the biotinylated anti-E/S antibody was dialysed by centrifugation in a 30 kDa nanosep spin column ( Pall Life Sciences , USA ) at 10 , 000 g at 4°C . Biotinylated α-E/S antibody was recovered from the spin column by washing the membrane with 500 µl PBS and collecting the retained immunoglobulins . Biotinylated α-E/S antibody was aliquotted and stored at −80°C . Biotinylation was confirmed by dot blot analysis of labelled immunogobulin . The assay components were titrated by doubling dilution to determine optimal signal:noise ratios . 96 well flat bottom microtiter plates ( Nunc , Thermo Scientific ) were coated with 50 µl ( 5 µg/ml ) α-E/S Ig diluted in coating buffer ( 15 mM Na2CO3 , 35 mM NaHCO3 , pH 9 . 6 ) and incubated overnight at 4°C or at room temperature for 90 min . After incubation , wells were washed 4–5 times in PBS-T ( PBS containing 0 . 05% Tween20 ) . Wells were then blocked with 150 µl of PBS containing 2% casein ( Merck , San Diego , CA ) for 60 minutes at room temperature . Uninfected rat fecal supernatant ( nRFS ) and infected fecal supernatants ( iRFS ) were diluted 1∶4 in PBS unless otherwise specified , and 50 µl added to each well and incubated for 90 min at room temperature . Known positive and negative formalin-fixed rat fecal supernatants , E/S products diluted in PBS and uninfected formalin-fixed fecal supernatant were assayed in duplicate . Wells containing only PBS or nRFS were included on each plate for standardization . Plates were then washed 4 times in PBS-T , after which 50 µl of 1∶500 α-E/S-B was added , and incubated at room temperature for 60 min . The plates were again washed 4 times in PBS-T and 50 µl of NeutrAvidin-HRP ( Thermo Scientific ) diluted 1∶10 , 000 was added , and the plates incubated for a further 60 min at room temperature . Plates were then washed 4 times in PBS-T prior to the addition of 100 µl per well of the developing substrate , ABTS ( 2 , 2′-azino-bis ( 3-ethylbenzthiazoline-6-sulphonic acid ) ) . After incubation at room temperature for 20–30 min plates were read on a Versamax microplate reader ( Molecular Devices ) at 405 nm , and analysed with Prism 4 ( Graphpad software , La Jolla , CA ) . Approval for maintenance of the life cycle of S . ratti in laboratory rats was obtained from the Animal Ethics Committee of the Queensland Institute of Medical Research . Ethical approval to test anonymised , non-reidentifiable fecal samples from patients with known parasitologic status for strongyloides was granted by the Royal Brisbane and Womens Hospital Human Research Ethics Committee .
Polyclonal rabbit antiserum was initially raised against S . ratti E/S products ( α-E/S Ab ) . This antiserum had an end titer of 1∶512 , 000 when measured by ELISA and was shown to recognise 14 protein bands on western blot ( Figure 1A ) . Reactivity to adult worm antigen was lower , with only 4 protein bands recognised by western blot . Even less immunoreactivity was observed with infective 3rd stage larvae where no reaction by immunoblot was observed ( Figure 1A ) . We next characterised α-E/S Ab using immunohistochemistry to identify the target organs . This resulted in specific immuno-staining of the ovaries and intestine of the parasitic adult worm ( Figure 1B ) . It was also apparent that during the production of E/S , that some contamination with rat host proteins had occurred , resulting in the generation of an antibody response to contaminating rodent host antigen ( Figure 1A ) . An antigen capture assay was developed using purified IgG of α-E/S Ab ( α-E/S IgG ) for the capture antibody; the same antibody now biotinylated was used for detection . To determine the lower limit of detection of this assay , Strongyloides E/S products were diluted in PBS from ≥2 µg/ml to ∼10 ng/ml by doubling dilution , and tested in the antigen-capture ELISA . The lower limit of E/S detection was determined to be 80 ng/ml for S . ratti; the limit of detection for and heterologous S . stercoralis antigen was 500 ng/ml . We next tested whether fecal components interfered with assay performance . PBS-T-extracted uninfected human fecal supernatant ( nHFS ) , uninfected rat fecal supernatant ( nRFS ) and infected rat fecal supernatant ( iRFS ) were evaluated at concentrations ranging from undiluted to 1∶8 in PBS ( data not shown ) . High background was observed in nRFS across all dilutions , making it impossible to discriminate between uninfected and infected samples . At the starting dilution , nHFS exhibited moderate assay background but became negative at a dilutions ≥1∶4 . Various techniques were evaluated in an attempt to reduce coproantigen ELISA cross-reactivity . These included a range of blocking agents , sample diluents , the addition of protease inhibitors , adsorption against host proteins and heat treatment . However , the majority of approaches had no effect on the background when tested with PBS-T-extracted feces . Likewise , changing the blocking agent from casein to bovine serum albumin , skim milk or fetal calf serum did not improve the assay performance . The addition of a commercial cocktail of protease inhibitors to fecal extracts did not decrease the background , suggesting proteases were not affecting the assay performance by desorbing protein coating the plate ( data not shown ) . Pre-adsorption of cross-reactive antibodies using a cocktail of rat gut antigen and nRFS resulted in some reduction of cross-reactivity with host proteins but did not enable reliable distinction between infected and uininfected samples ( data not shown ) . It was observed that changing the fecal extraction diluent from PBS to formalin resulted in a significant beneficial effect on assay background ( Figure 2A ) , such that when fecal samples were extracted in formalin , a clear discrimination between uninfected and infected samples was apparent , with only infected rat feces testing positive . Again , the addition of FCS and heat treatment resulted in no significant improvement in assay performance . Using formalin extraction , the assay was sufficiently sensitive to detect infection in samples diluted at least eight-fold . However , assay sensitivity was reduced approximately 4 fold when E/S was diluted in formalin-extracted nRFS ( Figure 2B ) . We next assessed analytical specificity of the assay . Known amounts of E/S collected from Ancylostoma caninum , A . ceylanicum and Schistosoma japonicum adult worms were tested at concentrations ranging from 100 µg/ml to 12 . 5 µg/ml . No cross-reactivity was observed from the E/S of species of helminth tested , even at concentrations as high as 100 µg/ml ( Figure 3A ) . Likewise , fecal supernatants from mice with monoparasitic infections Trichuris muris and S . mansoni which are more likely to co-infect humans and Formalin-fixed feces collected from uninfected human , tested negative ( Figure 3B ) . To explore the stability of coproantigen for subsequent detection in S . ratti-infected rat feces , the effect of storage on fecal supernatants was investigated . Stability was investigated by incubating unprocessed feces stored in formalin and processed feces as formalin-extracted fecal supernatant at 4°C for various lengths of time . Other formalin-extracted fecal supernatants were stored in aliquots at −20°C . Frozen formalin-extracted nRFS pooled from 6 rats consistently tested negative . In contrast , frozen formalin-extracted iRFS consistently tested positive by coproELISA ( Figure 4A ) . The effect of prolonged storage at 4°C of formalin-extracted fecal supernatant , and unprocessed formalin-preserved S . ratti rat feces for testing by coproantigen ELISA is shown in Figure 5 . When formalin-extracted fecal supernatants were tested , both uninfected and infected rat feces extracted in 10% formalin tested negative . Feces collected simultaneously from both uninfected rats and S . ratti-infected rats , tested positive when extracted in 4% formalin , indicating that some assay cross-reactivity occurred at sub-optimal concentration of formalin . In contrast , S . ratti-infected rat feces stored unprocessed in formalin tested positive when extracted as fecal supernatant , whereas uninfected rat feces stored unprocessed in formalin consistently tested negative even after 270 days at 4°C . Fecal samples collected from a human subject with proven S . stercoralis infection diagnosed by the agar plate method were tested next . Three separate samples were processed within 24 hours of collection . One gram of each stool was subject to concentration by sodium nitrate flotation for parasitological diagnosis of infection [28] . In addition , approximately 20% of the first and second stool samples were tested by Baermann coproculture . All 3 samples tested negative in 8 coverslips of sodium nitrate-floated feces for the presence of Strongyloides and all samples tested negative using the Baermann migration test . Figure 4B shows the results of the assay in which the three formalin-extracted human samples were tested with coproantigen ELISA from dilutions of 1∶2 to 1∶8 . In this experiment , a 1∶4 dilution of formalin-fixed nHFS was used as the negative control and the cut-off threshold for the assay determined . Uninfected human feces extracted in 10% formalin yielded a positive result at the assay cut-off when assayed at a 1∶2 dilution , but remained negative at dilutions of 1∶4 and below . The first and second samples tested positive at dilutions of 1∶2 and 1∶4 , while sample the third tested positive at all dilutions . When extracted in 10% formalin , all test samples tested positive for coproantigen , even when diluted 1∶8 .
Prior to this study , there had been one report describing coproantigen detection for S . ratti . In this work , antiserum was raised against whole worm antigen of S . ratti larvae and parasitic adult worms [23] . However , the authors reported low signal:noise ratios for positive fecal samples and minor cross-reactivity with the nematodes Syphacia muris and Necator americanus . The observation of high assay background is likely due to contamination of the antigen preparation with host gut and fecal material . Thus , the assay was not considered sufficiently sensitive for use as a diagnostic test [23] . More recently coproantigen tests have used antiserum raised to E/S antigens for the diagnosis of animal nematode infections . In general this approach has proved to be more useful [6] , [20] , [22] . An objective of this study was to determine whether heterologous S . stercoralis coproantigen could be detected in an infected human with antiserum raised to rodent Strongyloides E/S antigens . Data supporting this approach included studies documenting cross-reactivity between species with various published serodiagnostic and coproantigen assays [4] , [21] , [29] , [30] . Although , there have been numerous reports of the detection of coproantigen in animal nematode infections [6] , [20] , [22] , to our knowledge this is the first report that heterologous human infection has been detected by coproELISA with antiserum raised to antigens from an animal nematode . The anti-Strongyloides E/S polyclonal antibody was observed to cross-react with contaminating rat gut proteins present in rat feces . This cross-reactivity resulted in false positive tests when fecal supernatants collected from control rats were tested by coproELISA . Cross-reactivity with fecal components is a common occurrence in coproantigen assays and many techniques have been published describing how to reduce cross-reactivity [6] , [22] , [31] , [32] . Published methods for reducing background in coproELISAs are aimed at improving specificity by adsorbing cross-reactive antibodies , optimising method for extraction of the target antigen including the diluent used for fecal supernatant , or the choice of assay blocking agent . A range of these methods was tested with the Strongyloides α-E/S indirect coproELISA . However , the only method that reliably reduced assay background while preserving a positive signal in infected feces was formalin extraction of fecal supernatants . A possible explanation for this is that formalin treatment resulted in cross-linking the epitopes of the cross-reactive antigens in such a way that the non-specific antibodies in α-E/S Ab no longer bound . An alternative hypothesis is that the host antigens are fixed to host debris that are removed by centrifugation . Coproantigen detection in formalin-extracted fecal supernatant has also been successfully applied for diagnosis of Giardia [33] , Echinococcus [8] , [29] , Taenia [34] and Cryptosporidium [35] . An important consideration in the Strongyloides coproantigen ELISA is the use of formalin-extracted fecal supernatant for the discrimination between uninfected and infected samples as formalin exerts its effect by protein cross-linking . Thus , the sensitivity of E/S product detection in formalin-extracted feces could be impaired . This hypothesis was confirmed in an experiment where purified S . ratti E/S was diluted in PBS and formalin fixed uninfected faecal supernatant . An additional advantage of this preservation method is that it removes potential biohazards . Dilution of E/S in unfixed faecal supernatant has also been demonstrated to decrease assay signal:noise ratio and may be a contributing factor in the decreased sensitivity [6] , [20] , [33] . Even though the sensitivity for detection of S . stercoralis E/S ( 0 . 5 µg/ml ) was approximately 6 fold less than the detection limit for S . ratti E/S ( 0 . 08 µg/ml ) . The sensitivity of nematode coproELISAs currently range from 0 . 01 µg/ml to 0 . 5 µg/ml E/S [6] , [20] , [22] , which is in agreement with the observed sensitivity of the Strongyloides coproELISA . An important hurdle in diagnostic assays for parasitic infections is cross-reactivity with antibodies or antigens from heterologous parasites , a major issue described in many nematode coproantigen and serodiagnostic assays including current Strongyloides antibody tests [22] , [36] . No cross reactivity was observed in experiments with E/S antigens from S . japonicum , murine S . mansoni fecal supernatants or samples from the more closely related nematodes A . caninum and A . ceylanicum , or fecal supernatants from mice harboring T . muris infection . Together , these experiments suggest that the assay is specifically detecting Strongyloides antigens , as no cross-reactivity was noted with the other helminths tested . The last and most important investigation of the Strongyloides coproantigen ELISA involved testing S . stercoralis-infected and -uninfected human fecal samples . Importantly all S . stercoralis-infected samples extracted in 4% or 10% formalin tested positive when assayed at the standard sample dilution of 1∶4 . Two fecal supernatants extracted in 4% formalin tested negative when diluted further to 1∶8 . However , when simultaneously extracted in 10% formalin the samples tested positive . This suggests that this assay is more robust when fecal supernatant is extracted at the common fixative concentration , 10% formalin . This finding is similar to that reported by [33]where fixation in 10% formalin increased ELISA values in comparison with ELISA values obtained with distilled water extracted supernatants . Finally , the stability of coproantigen after prolonged storage was addressed . A few reports have examined the apparent stability of coproantigen after prolonged storage at 4°C and −80°C , as well as following environmental desiccation of fecal samples [6] , [20] , [37] . The results of the data from this study suggest that antigens present in formalin-extracted fecal supernatants are not stable when left at 4°C , but were so when stored at −20°C . Interestingly , when infected feces were stored in formalin at 4°C for at least 6 months prior to fecal supernatant extraction they tested positive . This suggests that the coproantigens are stable when feces are preserved in formalin and stored at 4°C , but are not stable when fecal supernatant was extracted prior to refrigerated storage . These results contrast to a similar study where formalin extracted and water extracted faecal supernatants were both stable at 4°C for the detection of Giardia cyst antigens [33] . The Strongyloides antigens detected by the coproELISA are currently undefined and future work in identifying the actual antigen ( s ) and raising monoclonal antibodies should enhance assay sensitivity and eliminate host cross-reactivity . The efficacy of the coproELISA is likely to be greatly enhanced if host cross-reactivity is eliminated and unfixed samples could also be assessed , broadening the utility . Another advantage of developing the Strongyloides coproantigen monoclonal antibodies would be in the conversion of the coproELISA into a rapid immunochromatographic dipstick test , a number of which are commercially available [38] . Further studies with a wider population of patients harbouring a variety of monoparasitic infections will also be required to accurately determine the specificity and sensitivity of the coproELISA . In conclusion , we have developed a coproELISA for the detection of strongyloidiasis , and have demonstrated that coproELISA using antibodies raised against antigens of a related animal parasite can be used to diagnose human infection in an antigen capture ELISA . Furthermore , this is the first report of an antigen capture assay using antibodies raised against native Strongyloides E/S antigens . Unlike many published assays where fecal supernatant is extracted in PBS containing a non-ionic detergent [6] , [16] , [20] , [22] , [29] , the current assay was only effective in discriminating uninfected from Strongyloides-infected fecal supernatant when extracted from formalin-treated feces . Future work with this assay will lead to a better diagnosis of human infection and enhanced understanding of the biochemistry of Strongyloides infections . | Strongyloides stercoralis is almost unique among human nematode infections in its ability to replicate within a patient's body , potentially leading to life-long infections if left untreated . Given the potential for severe life threatening Strongyloides infections and the unsatisfactory results of current parasitologic and antibody tests , there is a need for more efficient diagnostic tools . In this study we generated an assay to specifically detect proteins expelled by Strongyloides . Initially this assay for Strongyloides detection was not specific for the parasite; however , after developing a methodology using formaldehyde preservation of feces we specifically detected Strongyloides antigens in rodent and human stool . This methodology was then tested for cross-reactivity with purified proteins from closely related parasites and furthermore for cross-reactivity against faeces collected from animals harbouring single parasitic infections . Using this approach we found no non-specific reactivity with host or to various parasite antigens , suggesting that this assay is truly specific for Strongyloides detection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/helminth",
"infections",
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases",
"infectious",
"diseases/neglected",
"tropical",
"diseases"
] | 2011 | A Coproantigen Diagnostic Test for Strongyloides Infection |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.